The Complete Works of Aristotle: The Revised Oxford Translation 
edited by Jonathan Barnes.
Princeton, 1250 pp., £53, August 1984, 0 691 09950 2
Show More
Show More

‘Aristotle and Plato’, ‘Plato and Aristotle’ – the coupling of names is something we take for granted. They are the two giants of ancient philosophy, are they not, and who but Kant among later philosophers deserves to rank as high as they? Yet Aristotle’s greatness was not always so visible.

A character in Cicero invokes him thus: ‘Aristotle, whom I think I may with justice call the chief of all philosophers, with the exception of Plato’. Why so defensive? The implication is that even to rank Aristotle next after Plato is contentious, as if the thought was new and daring.

Indeed it was. Aristotle died in 322 BC. Not until the first century BC (the century in which Cicero is writing) was the ancient original of The Complete Works of Aristotle put together and made available in a scholarly edition. It took another two centuries or so for the philosophical community to get used to the dense and difficult argumentation of these treatises and acquire a real understanding of their content. And it was only in the Middle Ages that Aristotle achieved the ascendancy that made him ‘the teacher of those who know’, honoured and admired (Dante continues) by Plato and the whole family of philosophers.

Any attempt to explain this long delay in recognition must start from the point that the treatises which comprise The Complete Works of Aristotle were not written for publication. Not only will they not have been on sale in the ancient equivalent of your local bookshop, but it is unlikely that many copies were made for private circulation. A quite well authenticated tradition has it that the originals ended up in a cellar in Turkey, where they mouldered for years until they were rediscovered and brought to Rome in the first century BC to form the basis for that first scholarly edition which I mentioned earlier. The story may sound too romantic to be true, but modern scholars are still debating the extent to which professional philosophers outside the small circle of Aristotle’s own pupils and associates knew and responded to the treatises.

Epicurus, for example, seems to show intimate knowledge of some arguments about motion in one section of the work we now know as Aristotle’s Physics (the Physics, like the Metaphysics, is itself an editorial creation, put together out of connected but originally distinct Aristotelian essays), but staggering ignorance of the sophisticated views of teleology in another section. When we get to Cicero’s day, he reports that Aristotle is ignored by all but a few philosophers. And when we look beyond professional philosophers to the wider intellectual world, there is little doubt that, during his lifetime and for some three hundred years thereafter, Aristotle was known mainly as the author of a series of dialogues of which only excerpts are given in The Complete Works of Aristotle.

Only excerpts are given because only excerpts have come down to us. It seems that the dialogues, the works which Aristotle actually published, were driven off the market when the treatises became known and were seen at once to be more ‘accurate’ and ‘finished’ from a philosophical point of view. To judge by the excerpts which survive as a result of being quoted by other writers, the dialogues were indeed poor stuff and deserved their extinction. They would give Cicero’s readers every reason to be surprised at the suggestion that Aristotle is second only to Plato.

But Cicero did more for Aristotle than give the treatises a boost in, so to speak, the Rome Review of Books. He furnished some of the vocabulary with which learned men in later centuries would discuss and try to make sense of the central concepts of Aristotle’s philosophy. One important example of Ciceronian translation will serve both to signal the excellence of Jonathan Barnes’s Revised Oxford Translation and to highlight a feature of Aristotle’s philosophy which will have made it hard for the ancients to see his greatness.

Aristotle is unique among ancient philosophers in his respect for people’s opinions: both the opinions of other philosophers and the opinions of the ordinary man. He does not defend a ‘common sense’ philosophy in the manner of G.E. Moore, but if something is believed by absolutely everyone, then, he holds, it must be true. Aristotle also does something that a 20th-century philosopher like Moore could never have dared. He establishes science on the basis of the opinions of ‘the majority’ and of ‘the wise’.

Thus he sets out in the Physics to derive the fundamental concepts and principles needed for the understanding of nature from a probing examination of the ideas of the ordinary man and a handful of previous seekers after truth. The ideas have to be clarified, made consistent and generalised, but no amount of dialectical manipulation can alter the fact that it is from people’s opinions – both their explicit opinions and the opinions implicit in their modes of speech – that Aristotle learns how to conceive matter and form, time and teleology, and many other aspects of nature. Even his astronomy, with the stars and planets revolving around the Earth on spheres which maintain uniform speeds of their own accord, owes as much to the popular belief that there are gods in the heavens as to the sophisticated mathematical models developed in Plato’s Academy.

In according this high status to opinion Aristotle was making a radical break with earlier philosophy. Both Plato and the Pre-Socratics are contemptuous of opinion and claim to have found a better path to truth. Much the same holds for the leading philosophers after Aristotle – Epicurus and the Stoics. Even though they insist that philosophical theories must be tested against people’s naturally formed preconceptions, in practice they look for the truth first and then tell people what their preconceptions must be. People’s actual opinions, about their own preconceptions as about anything else, count as evidence for little but the corruptness of their minds.

In this climate of hostility to opinion those of Cicero’s readers who were able to get hold of Aristotle’s treatises will have been in for a shock. Here is a philosopher whose characteristic way of setting about a problem, be it in physics, the philosophy of mathematics, or ethics, is to collect the relevant opinions of ‘the majority’ and of ‘the wise’ and start puzzling over the conflicts and unclarities they present him with. It takes a lot of hard dialectical work for the Aristotelian solution to emerge from this process, but still, why take seriously a philosopher who deliberately limits the dialectical discussion to ideas which have won the allegiance of the ordinary man or another theorist? Is that not to proclaim that truth is bounded by opinion?

Even six centuries later, when Aristotle’s place as second only to Plato was established beyond dispute, the great Neo-Platonist commentator on the Physics, Simplicius, is still embarrassed by Aristotle’s reliance on opinion. He claims that Aristotle is just making his proofs more evident by deriving the truth from observation and from ideas with which his audience are familiar. This cannot be right, because it implies that Aristotle’s methods of inquiry are incidental to his conclusions, that the same conclusions could just as well have been presented and proved in a less ‘popular’ way. No, any reckoning with Aristotle’s thought must accept that, unlike other ancient philosophies, it is to a very considerable extent a web woven from opinion.

Now Aristotle has a technical term to mark the views which his inquiries will take into account: endoxon. A proposition is endoxon, he says, if it is the opinion of everyone or of the majority or of the wise – either all the wise or the majority or the most notable of them. Cicero translated endoxon into the Latin word probabile, and, with the endorsement of Boethius, probabile became the standard way to represent the type of proposition which Aristotelian philosophy takes off from. It was an unilluminating depiction at best, and came to be positively misleading.

First, Cicero also used probabile to translate several other, quite distinct technical terms of Greek philosophy, with the result that the Latin word stood for a heterogeneous jumble of ideas. Secondly, from the 17th century onwards ‘probable’ and its equivalents in other modern languages became more and more closely associated with inductive reasoning and statistical computation. One cannot, in modern English, sensibly say that a proposition which is believed by a majority of people, or by the most notable of the experts, is thereby shown to be probable. It may be believed because it is, in fact, probable, but it is not probable just because it is believed. And it would be very misleading to speak of Aristotle’s philosophy as built on probabilities.

So long as scholars wrote their commentaries on Aristotle in Latin – a practice which continued well into the 19th century – the difficulty could be covered up in the respectable murk of probabile. But when Benjamin Jowett’s will made financial provision for a project of translating the whole of Aristotle into English, decisions were taken which would exert a deep and lasting influence on Anglophone philosophers’ understanding of a central feature of Aristotle’s thought. Jowett’s will brought about the Oxford Translation of Aristotle – a massive collaborative effort in 11 volumes published between 1908 and 1930 (a 12th volume of fragments and excerpts from the dialogues was added in 1950) under the general editorship of J.A. Smith and W.D. Ross. It was, and still is, the most authoritative modern translation of Aristotle in existence, a milestone in the centuries-long endeavour, from Cicero’s day to our own, to understand his thought. It kept the best Aristotelian scholars of the time busy in Oxford during the years when, in Cambridge, G.E. Moore was formulating and defending his philosophy of common sense. Can it be accidental that, to render endoxon, the Oxford translators chose phrases like ‘generally accepted’ and ‘common opinion’? Occasionally they regress to ‘probable’, but they never bring to the reader’s attention the fact that Aristotle’s criteria for a proposition’s being endoxon give equal weight to the wise as to the majority. A proposition need not be common opinion or generally accepted to be endoxon for Aristotle. It can be radically at variance with common sense and still be endoxon if it has won the allegiance of just one notable thinker – e.g. Plato.

The Oxford translators’ picture of Aristotle as a philosopher who, as Ross himself once put it, ‘by virtue of his strong common sense ... rarely writes what anyone would regard as obviously untrue’ took a more linguistic turn under the leadership of Ryle and Austin during the heyday of ‘Oxford philosophy’. One of the most worthwhile fruits of that period of philosophy was that it opened scholars’ eyes to Aristotle’s predilection for, and marvellous skill at, conceptual analysis in the Oxford style. Not only in ethics, but in physics too, his endoxa are often the opinions implicit in ordinary language, and his dialectic is a statement and resolution of the conceptual puzzles they give rise to.

Often but not always. Just as Aristotle is no philosopher of common sense, because he has such regard for the speculative views of other philosophers, so he is not to be pinned down as a conceptual analyst: for he is equally attentive to the ordinary man’s explicit opinions about substantive matters of fact – that there are gods in the heavens, for instance.

These onesidednesses encouraged another: the idea that Aristotle is doing philosophy (as conceived in the Oxford of Ross or Ryle) instead of science. Oxford-trained scholars have frequently characterised the Physics as philosophy of science rather than science. But Aristotle sees the modern philosophical methods he uses as the proper means to arrive at an understanding of nature. The point is not just that for him, as for other ancient philosophers, the modern divide between philosophy and science does not exist. It is that for Aristotle, unlike other ancient philosophers, what the man on the Clapham omnibus thinks and says about, for example, time and teleology is as much a source of truth as are the expert opinions of a Newton or a Darwin.

Now, however, Oxford has made amends. In 1976 the Jowett Trustees commissioned Jonathan Barnes (a fellow of Jowett’s own college, Balliol) to revise the Oxford Translation of Aristotle in the light of recent scholarship. In 1980 Barnes published an important article in the Revue Internationale de Philosophie which demonstrated that endoxon is best translated ‘reputable’: what Aristotle starts from are propositions of good repute, propositions which enjoy good standing with the majority or with the experts. Accordingly, in the Revised Oxford Translation endoxon is everywhere translated ‘reputable’.

Needless to say, there is more to the revision than that. There are replacement translations for three treatises which were badly served in the original; numerous minor corrections elsewhere; the inclusion of two portions of the corpus previously omitted; a much larger selection of fragments from the dialogues and other lost works; the whole encompassed in two large but portable volumes instead of 12. But if ever a change of translation deserved to make intellectual history, it is Barnes’s new rendering of endoxon as ‘reputable’. If the Revised Oxford Translation achieves the same influential status as its predecessor, future generations will find Aristotle a rather less comfortable philosopher than he has been for most of the 20th century. As in Cicero’s day, the question will be: what are we to think of a philosopher who announces time and again that the way to the truth is through the study of ‘reputable’ opinions?

Consider some of the conclusions for which Aristotle argues with the help of some of the opinions which were accepted and esteemed in his day. Slavery is just; the world is eternal and the same things have always existed in it; the atomic theory of matter is defective because it fails to explain the difference between dinner and breakfast. I have of course chosen examples which tell against Ross’s conviction that Aristotle ‘rarely writes what anyone would regard as obviously untrue’. But the point of choosing them is to raise the question whether reputable opinion can fulfil the heuristic and evidential role that Aristotelian philosophy assigns to it.

Suppose Greek society, Greek belief and the Greek language had developed differently from the way they did. Suppose there had been no Democritus or no Plato supplying Aristotle with their opinions to puzzle over, to clarify and correct. Aristotle’s official methodology implies that if the stock of available endoxa were significantly different, the conclusions of his dialectic would be different too. How then can he claim that his conclusions are true?

This was what worried Simplicius. If Aristotle’s conclusions are true, it should be possible to display their truth without reference to who believes what. Presumably, what counts in favour of a proposition is – or should be – not the bare fact that it is believed but the grounds or evidence for believing it. In which case why not bypass the dialectical manipulation of opinions and talk directly of the grounds and evidence for your conclusions?

But Aristotle does not make the clean separation between evidence and opinion on which this argument presumes. The fact that a proposition is believed by the majority or by experts is not for Aristotle just a sign that, if we asked them, they could cite evidence for the proposition. Their belief, as he treats it, is already some evidence in favour of what they believe; even if the opinion is not correct, it is likely to contain an element of truth which the dialectic can sift out and formulate clearly.

Aristotle can also reply to the objection that if the stock of available endoxa were significantly different, the conclusions of his dialectic would be different. He propounds a cyclical theory of history according to which each viable social arrangement, each art and each science is rediscovered innumerable times after periodic cataclysms. In particular, he twice states that ‘the same opinions recur in rotation among men, not once or twice or occasionally, but infinitely often.’ This blocks the counterfactual supposition of the objection. The stock of endoxa available to Aristotle to philosophise with in Athens in the fourth century BC is seen by him as the outcome of a history of social and cognitive development which is as natural and as repetitive as the growth of a lettuce or the maturation of a human being. To say it is natural and repetitive is not to say that it is predetermined in every detail: even lettuces vary, though not as much as human beings. But it is to say that when a society achieves a sufficient surplus to permit some of its members to engage in intellectual theorising, you can expect something like the materialistic tendency of Democritean Atomism, and something like the opposite tendency of a Plato, and finally an Aristotle to preserve the merits and correct the demerits of the two extremes.

I say ‘finally an Aristotle’ because the reply will not allow Aristotle to claim that his own conclusions are true (as opposed to: conclusions natural for a philosopher in a particular society at a particular period of history) unless he sees himself as the most mature philosopher since the last cataclysm and his own philosophical system as the goal and culmination of the growth of knowledge. Cicero reports that he did: ‘Aristotle, upbraiding the philosophers of old for thinking that philosophy had been perfected by their talents, says that this was either very silly of them or very conceited, but that he can see that, because of the great advance made in a few years, in a short time philosophy will be quite complete.’ Not all scholars trust Cicero to be reporting one of Aristotle’s lost works accurately here. Some are unwilling to believe that Aristotle could be like other philosophers – Kant, for example – who have similarly claimed to have brought philosophy to a state of completion. But my argument has been that the claim of finality is one to which Aristotle’s methodology commits him on many of the pages he writes.

It is also a claim which a 20th-century reader cannot possibly accept. The appeal to ‘our’ beliefs, ‘our’ intuitions, ‘our’ way of speaking, is as common in philosophy today as it was earlier in the century, and Aristotle’s name is still frequently invoked to show that opinion of one sort or another is a reputable basis for philosophy to start from. The irony is that if Aristotle’s professed methodology was sound, there would be little to add to The Complete Works of Aristotle.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences