IT IS ONLY PROPER THAT before we start on our historical study of rationalism in classical antiquity, I explain what I mean by rationalism. Roughly speaking, very roughly, there are two sources of knowledge: perception and inference. Perception is supposed to give us the facts, the qualities of things and the spatiotemporal relationships between things and events. No extended discussion is needed to show that if we want to know what the color, sound, taste, or texture of something is, we look and see; and if we want to know where something is and when it occurs, we also look and see. (The names of other senses can be substituted for looking and seeing, if my statement appears to be obscure.) But since all perception is of particular, localized, and dated things and events, no act of perception can give one a general law. And general laws are what is wanted in science and philosophy. Regardless of the problem of how we reach our general laws, whether it is simply by abstraction from perception of characteristics observed to be shared in common, or by intuition, or otherwise, once they are established, we use them to serve as premises for systematic study. By systematic study I mean the sort of thing one is supposed to have in geometry, where from the premises we deduce by logical means alone certain inferences, or theorems. It makes little difference here whether the deductions are made by the substitution of equivalent terms or by syllogisms or by the procedures of relational logic, for that is a problem for specialists. But, regardless of many-valued logics, which are of the greatest theoretical importance, the logic which is used in classical geometry and in traditional science is two-valued, and the two values are truth and falsity. In short, the Law of Contradiction is the principal rule which we try to follow. There are of course other rules as well, but I think no one will deny that a rational system relies above all on consistency as its aim.
But, as Kant pointed out in different terms, if the general premises of a system are to mean anything, they must have some content, and that content eventually comes from perceptual experience. And if perceptual experience is to have any rational significance, it must be organized into classes all of whose members are similar in some respect. The child who learns to call things by common nouns has already begun to classify. His classifications are not his own, to be sure, but inherited with his language and taught him by his associates, parents, brothers and sisters, and schoolteachers in the main. There are, however, hundreds of possible ways of classifying anything, since even if we had only five senses, we would have at least five ways of grouping our experiences together, and there are, besides the five, combinations of them and combinations of the combinations. If everything we speak of could be perceived by all the five senses, there would be many hundreds of combinations and permutations as a start. But when we say that something, for instance, is a dog, we are not classifying exclusively on the basis of its sensory qualities; we have other information, furnished by zoologists, to guide us and that is only slightly sensory. For the internal anatomy of the animal and its relationship in a phylogenetic series also determine its genus and species.
Moreover, though a child might call his dog his playmate and friend, and a host of other endearing things, such names would be of little scientific interest. For the way we group things is determined by their relation to other things which are irrelevant to our feelings and our daily life. We enter a world which has already been systematized by others and our task is to locate our own personal experiences in that larger system of ideas. We have to translate something which is not an idea at all, but a thing, here and now, bathed in emotions of love and hate, fear and aspiration, associated with our past and perhaps indicative of our future, into an idea which in turn can be situated in a larger group of ideas. This is accomplished very early in life in a rudimentary way, but as life goes on, complications and refinements enter, and by the time we begin studying any one of the sciences, we wake up to the fact that things are not as they seem. The moment at which table salt becomes sodium chloride, the world takes on a new and sometimes bewildering aspect, for though you might say that it is only a question of terminology, in fact the terminology orients one in a vastly different direction from that of crude perception.
The rationalist will attempt to make his classifications in accordance with whatever scientific methods are generally accepted at the time at which he is working. These methods have all turned out to be open to question, but one cannot question an idea of which one knows nothing. That is obvious. What is important is that when a method of study is accepted as right, a body of information compiled according to the method will exist and any new information will be expected to be consistent with it. That is an essential trait of what I mean by rationalism. To take but a single example, in Aristotle we find that the elements are classified according to their being hot or cold, wet or dry. For reasons which no one knows any longer, it was established before his time that there were four elements, Earth, Water, Air, and Fire. They were the cold-dry, the cold-moist, the hot-moist, and the hot-dry.1 It will be noted that each of the qualities can be directly perceived by our senses. Moreover, each of the elements, in an impure form, can also be perceived directly. And what is more, the typical behavior of the elements, Earth always falling downward, Fire always rising upward, is verifiable by perception. And since the elements were reproduced in the human body in the four humors, and each humor when predominant determined a temperament, though this elaboration is not attributable to Aristotle so far as we know, one had a thoroughly systematic and verifiable account of an important set of facts. That is what rational science stood for in ancient Greece and what it stands for now. Involved in it, as a method, is the ability of every man to check its results to see if its truths are interpersonal. And the method is indifferent to the individual’s feelings, his moral life, his religious ideas.
But a rationalistic technique is also supposed to produce results which are of the nature of fertile theorems, that is, laws from which the future can be predicted, other things being equal, and effects can be inferred from the presence of causes. This is a minimum requirement. The child is again a rudimentary rationalist when he learns that whining and wheedling will produce the results he wants from his mother, or that in order to make his tricycle move he must push against the pedals. He would be irrational if he imagined that by talking to his tricycle or slapping it he could induce it to get under way, just as he would be irrational if he did not whine and wheedle his mother once he had established the opposite causal law. This is why rationalism is always an opponent of superstition, magic, sacrifice, and prayer. The battle between rationalism and religion has always been a fierce one and, though religion may produce moral, aesthetic, political, and in general emotional satisfactions which science seems unable to produce, and may be high, deeper, more spiritual, and indeed more valuable to human life than science, that does not make it identical with science. On the contrary, when science is rational, it is bound to be an adversary of religion if the two are concerned with the same problem. It may be religious to ask people to pray for peace or rain or plentiful crops; prayer is not the rational way to get such results. I am not saying that it is not the more successful way, though as far as crops are concerned—I say nothing of peace and rainfall—scientific procedures of agronomy seem to work better. It may also be a good thing for people to have years of famine and drought and war; the rationalist is not necessarily interested in what is good or bad if what he is studying is causal relations. I can think of no scientific discovery, even in medicine, which has not been put to questionable, if not downright evil, uses. But that is irrelevant to the difference between rationalism and irrationalism or nonrationalism. To condemn painting because some painters have painted obscene paintings, or Catholicism because of Alexander Borgia, would be no more reasonable than to condemn physics because the release of nuclear energy has slaughtered 150,000 innocent Japanese or because biochemistry has been used to poison people.
This book then is a historical study of rationalism, as I have described it, in classical philosophy. It has resulted in a story of degeneration. I have not been able to discover all the causes of the change, nor have I attempted to study rationalism in all fields, for both would demand knowledge which I cannot claim to possess and years of further study which I cannot hope to have. I have simply taken on the whole four subjects: the distinction between appearance and reality, the method used to establish the distinction, the appraisal of life made by the men studied, and something about their ethical theories. I have not attempted to write another history of Greek and Roman philosophy as a whole; that has been done by many others. I have thought of philosophy, as of science, as a sheaf of problems bound together by a common name. It would therefore be futile to look in this book for the complete philosophy of anyone mentioned in it. It is not an encyclopedia or dictionary. I have, moreover, dealt only with the pagans, except in the case of Philo Judaeus whose influence on his pagan successors was too important to be omitted. In every case I have reread the texts used in the original and have interpreted them according to my own lights and not according to what some other historian has had to say on the subject. There is an element of hybris, I suppose, in this but if I had read everything that has been written on any one of the men treated in this study, not only would the book never have been written, which might not have been deplorable, but I should never have been able to reconcile the results of my reading. I have tried to reduce the number of footnotes to a minimum and have inserted textual references in the body of the text.
I cannot send this book to press without some words of acknowledgment to individuals and institutions who have helped me write it. First, to the American Council of Learned Societies which gave me a grant for travel and consequently for leisure; second, to the University of Pittsburgh which, in giving me the chair of Andrew Mellon Professor of Philosophy for a year, also obviated the necessity of spending most of my time in teaching; third, to the graciousness of their library staff who did their utmost to get me the books I needed. But I should also like to express my appreciation of the help which I received from my friend Harold Cherniss of the Institute for Advanced Study, as well as of that which I have received over the years from Arthur O. Lovejoy, whose historical writings have been to me a model of what such studies should be.
Paris 1959, Pittsburgh 1961
1 For further details, see Aristotle De generatione et corruptione ii. 4.