The Basic Works of Aristotle (Modern Library Classics) Read online

  An introduction to a philosopher which did no more than confirm the student in established opinions, or an edition whose apparatus did no more than supply the reader with instruments by which to find what he had conceived to be useful prior to his reading of the philosopher and prior to philosophic analysis of his standards of utility, would aid the reader to find what he was looking for but at the expense of its subject, for the philosophy would almost certainly not be understood, and misconceived philosophic doctrines, however ingeniously contrived, are of doubtful ultimate utility. The third objective of an introduction to the works of a philosopher, to which the preceding two must be subordinate since there is no adequate reason for reading the works of a philosopher other than the philosophy they express, is more easily obscured than achieved by aids to reading or to philosophy. Some aid is needed, however, and therefore a method of reading Aristotle’s works is suggested in the Introduction by a brief statement of the interrelations and continuity of his doctrines. The reader is advised to treat this interpretation skeptically until and unless he can find it confirmed in his own reading of the text, for it is useful only as a device by which to permit Aristotle to speak for himself. The achievement of Aristotle can be discovered only by reading and rereading his works, and the appreciation of that achievement depends quite as much on the deepened sense of value and the precision of criteria which he inculcates as on the materials he treats. The Middle Ages may seem to have exaggerated in calling him the Philosopher, but the understanding of what he said is still an unparalleled introduction to philosophy.

  It is as difficult to reconstruct some notion of the appearance of Aristotle as to determine the lineaments and characteristics of his thought. The representation of him which was most familiar a generation ago, the statue in the Palazzo Spada in Rome, is almost certainly not a portrait of Aristotle. It was long supposed to be Aristotle because of its fragmentary inscription which should in all probability be restored more correctly as “Aristippos,” and in any case the head does not belong to the statue. The portrait reproduced as the frontispiece, a bust in the Kunsthistorisches Museum in Vienna, has rather better claim to rank as a genuine portrait of Aristotle, although the identification rests on a tortuous argument. As proposed by Studniczka (Das Bildnis des Aristoteles; Leipzig, 1908), the identification goes back to a bust which was found in Rome about 1590 and which was bought by the learned antiquary Fulvio Orsini. It was identified by an inscription on its base. This bust is lost, but two drawings, one of them by Rubens, have survived. A family of twelve busts, varying in quality, preservation, and probable date, has been assembled, which seem, from their close correspondence, not only to represent one man but to imitate one original portrait, and which further, from their similarity to two drawings of the lost bust, may be portraits of Aristotle. The identification is plausible, though by no means certain. The style places the original portrait approximately in the time of Aristotle, and of the twelve extant busts the Vienna head probably gives the best idea of the original. The nose is almost entirely modern, but there is little other restoration. Several features ascribed to Aristotle by ancient tradition may be seen in these portraits: small eyes, short beard, and thinning hair.

  Grateful acknowledgment is hereby extended to the Oxford University Press for permission to reprint the translation of the works of Aristotle prepared under the editorship of W. D. Ross. The arduous task of reading proof, checking quotations, and preparing the enormous materials of this volume for publication was rendered manageable by the assistance of Dr. Herbert Lamm and Dr. Meyer W. Isenberg, while the actual consummation of the task was not only facilitated by the co-operation of the staff of Random House, but is largely due to the jogging encouragement and reproaches of Mr. Bennett A. Cerf and Mr. Saxe Commins.

  RICHARD MCKEON

  INTRODUCTION

  C.D.C. Reeve

  Aristotle’s Writings

  A list of Aristotle’s papers, probably made in the third century B.C., seems to describe most of his extant writings, as well as a number of works—some in dialogue form—that are now lost. When Sulla captured Athens in 87 B.C., these papers were brought to Rome, where they were edited, organized into different treatises, and arranged in a logical sequence by Andronicus of Rhodes in around 30 B.C. Most of the writings he thought to be genuinely Aristotelian have been transmitted to us via manuscript copies produced between the ninth and the sixteenth centuries.

  These writings, of which the present volume include a rich selection, may be classified as follows: logic, dialectic, metaphysics: Categories, On Interpretation, Prior Analytics, Topics, On Sophistical Refutations, Metaphysics; science and philosophy of science: Posterior Analytics, Physics, On the Heavens, On Generation and Corruption, Meteorology, History of Animals, On the Parts of Animals, On the Motion of Animals, On the Progression of Animals, On the Generation of Animals; psychology and philosophy of mind: On the Soul, Sense and Sensibilia, On Memory and Reminiscence, On Sleep, On Dreams, On Prophesying by Dreams, On Length and Shortness of Life, On Youth, Old Age, Life and Death, Respiration; ethics and politics: Nicomachean Ethics, Magna Moralia, Eudemian Ethics, Politics, Rhetoric, Constitution of Athens; aesthetics: Poetics.

  The most credible view of these writings is that they are lecture notes written or dictated by Aristotle himself and not intended for publication. Their organization into treatises and the internal organization of the treatises into books and chapters may, however, not be his. No doubt this accounts for some, though not all, of their legendary and manifest difficulty.

  The Aristotelian World

  Of the various things that exist in the world described in Aristotle’s writings, “some exist by nature, some from other causes” (Physics 192b8–9). Those that exist by nature have a nature of their own, an internal source of movement, growth, and alteration (192b13–15). Thus, for example, a feline embryo has within it a source that explains why it grows into a cat, why that cat moves and alters in the ways it does, and why it eventually decays and dies. A house or any other artifact, by contrast, has no such source within it; instead, the source is “in something else external to the thing,” namely, the craftsman who manufactures it (Physics 192b30–31; also Metaphysics 1032a32–b10).

  A thing’s nature is the same as its essence or function, which is the same as its end, or that for the sake of which it exists. For its end just is to actualize its nature by performing its function (Nicomachean Ethics 1168a6–9), and something that cannot perform its function ceases to be what it is except in name (On the Parts of Animals 640b33–641a6, Politics 1253a23–25). Aristotle’s view of natural beings is therefore teleological: He sees them as being defined by an end (telos) for which they are striving, and as needing to have their behavior explained by reference to it. It is this end, essence, or function that fixes what the good for that being consists in, and what its virtues or excellences are (Nicomachean Ethics 1098a7–20, Physics 195a23–25).

  Most natural things, as well as the products of art or craft, are hylomorphic compounds, compounds of matter (hulě) and form (morphě). Statues are examples: Their matter is the stone or metal from which they are made; their form is their shape. Human beings are also examples: Their matter is (roughly speaking) their body; their soul is their form. Thus a person’s soul is not something separable from his body, but is more like the structural organization responsible for his body’s being alive and functioning appropriately.

  While the natures of such compounds owe something to their matter and something to their form, what they owe to form is more important (Metaphysics 1025b26–1026a6, Physics 193b6–7). For example, a human being can survive through change in his matter (we are constantly metabolizing), but if his form is changed, he ceases to exist (Politics 1276b1–13). That is why the sort of investigation into human beings we find in De Anima and in ethical and political treatises focuses on souls rather than bodies.

  These souls consist of distinct, hierarchically organized constituents (Nicomachean Ethics
, bk. I, ch. 13). The lowest rung in the hierarchy is the vegetative soul, which is responsible for nutrition and growth, and which is also found in plants and other animals. At the next rung up, we find appetitive soul, which is responsible for perception, imagination, and movement, and so is present in other animals too, but not in plants. This sort of soul lacks reason but, unlike the vegetative, can be influenced by it. The third element in the human soul is reason. It is divided into the scientific element, which enables us to contemplate or engage in theoretical activity, and the calculative or deliberative element, which enables us to engage in practical and political activity (Nicomachean Ethics 1097b33–1098a8, 1139a3–b5).

  Because the human soul contains these different elements, the human good might be defined by properties exemplified by all three of them or by properties exemplified by only some of them. In the famous function argument from the Nicomachean Ethics, bk. I, ch. 7, Aristotle argues for the latter alternative: The human good is happiness, which is “an active life of the element that has a rational principle” (1098a3–4). The problem is that the scientific and the deliberative element both fit this description. Human happiness might, therefore, consist in practical political activity, or in contemplative theorizing, or in a mixture of both. Even a brief glance at Nicomachean Ethics, bk. X, chs. 6–8 will reveal how hard it is to determine which of these Aristotle has in mind.

  Aristotelian Sciences

  The Aristotelian sciences provide us with knowledge of the world, how to live successfully in it, and how to produce what we need to do so. Hence they fall into three distinct types:

  I. Theoretical sciences: theology, philosophy, mathematics, natural sciences.

  II. Practical sciences: ethics, household management, statesmanship, which is divided into legislation and politics, with politics being further divided into deliberative science and judicial science (Nicomachean Ethics 1141b29–32).

  III. Productive sciences (crafts, arts): medicine, building, etc.

  Of these, the theoretical ones are the Aristotelian paradigm, since they provide us with knowledge of universal necessary truths. The extent to which ethics or statesmanship fit the paradigm, however, is less clear. One reason for this is that a huge part of these sciences has to do not with universal principles of the sort one finds in physics, but with particular cases, whose near infinite variety cannot easily be summed up in a formula (Nicomachean Ethics 1109b21, Rhetoric 1374a18–b23). The knowledge of what justice is may well be scientific knowledge, but to know what justice requires in a particular case one also needs equity, which is a combination of virtue and a trained eye (Nicomachean Ethics, bk. V, ch. 10). Perhaps, then, we should think of practical sciences as having something like a theoretically scientific core, but as not being reducible to it.

  Theoretical Science

  Each Aristotelian theoretical science deals with a genus—a natural class of beings that have forms or essences (Posterior Analytics 87a38–39, Metaphysics 1003b19–21). When appropriately regimented, it may be set out as a structure of demonstrations, the indemonstrable first principles of which are definitions of those essences. More precisely, the first principles special to biology, or to some other science that applies to only a part of reality, are like this. Others that are common to all sciences—such as the principle of non-contradiction and other logical principles—have a somewhat different character. Since all these first principles are necessary truths, and demonstration is a type of deductive inference, scientific theorems are also necessary.

  Though we cannot grasp a first principle by demonstrating it from yet more primitive principles, it must—if we are to have any unqualified scientific knowledge at all—be “better known” to us than any of the science’s theorems (Nicomachean Ethics 1139b33–34). This better knowledge is provided by intuition (nous), and the process by which principles come within intuition’s ken is induction (1139b28–29, 1141a7–8).

  Induction begins with perception of particulars, which gives rise to retention of perceptual contents, or memories (Posterior Analytics 100a1–3). From a unified set of such memories experience arises (100a3–6), “when, from many notions gained by experience, one universal supposition about similar objects is produced” (Metaphysics 981a1–7). Getting from particulars to universals, therefore, is a largely noninferential process. If we simply attend to particular cases—perhaps to all, perhaps to just one—and have some acumen, we will get there (Prior Analytics 68b15–29, Posterior Analytics 88a12–17, 89b10–13). When these universals are appropriately analyzed into their “elements (stoicheia) and first principles,” they become intrinsically clear and unqualifiedly known (Physics 184a16–21).

  A universal essence is something out there in the world. Its analogue in a scientific theory, however, is a definition similar in structure to it (Metaphysics 1034b20–22). That is why the first principles of the sciences are not essences, but definitions of them.

  The inductive path to first principles and scientific knowledge begins with perception of particulars and of perceptually accessible, unanalyzed universals, and leads eventually to analyzed universal essences (first principles) and definitions of them. At this point, induction gives way to deduction, as we descend from these essences to other principles. Perception alone cannot reach the end of this journey, but without perception it cannot so much as begin. Perception, elaborated in theory, is the soul’s window on the Aristotelian world (Prior Analytics 46a17–18, On the Soul 432a7–9).

  Dialectic

  The first principles proper to a science cannot be demonstrated within that science. If they could, they would not be genuine first principles. They can, however, be defended by dialectic. For, since it “examines,” and does so by appeal not to scientific principles but to common or generally accepted opinions (endoxa), “dialectic is a process of criticism wherein lies the path to the [first] principles of all inquiries” (Topics 101a36–b4).

  Now opinions are endoxa when they are accepted without demurral “by every one or by the majority or by the wise, either by all of them, or by most or by the most notable and illustrious of them” (Topics 100b21–23), so that the majority do not disagree with the wise about them, nor do either group disagree among themselves (104a8–11). Generally accepted opinions, therefore, are beliefs to which there is simply no worthwhile opposition. Apparent endoxa, by contrast, are beliefs that mistakenly appear to have this uncontested status (100b23–25, 104a15–33).

  Defending first principles on the basis of endoxa is a matter of going through the difficulties (aporiai) “on both sides of a subject” until they are solved (Topics 101a35). Suppose, then, that the topic to be dialectically investigated is this: Is being a single unchanging thing, or not? A competent dialectician will, first, follow out the consequences of each alternative to see what difficulties they face. Second, he will go through the difficulties he has uncovered to determine which can be solved and which cannot. As a result, he will be well placed to attack or defend either alternative in the strongest possible way.

  Aporematic, which is the part of philosophy that deals with such difficulties, is like dialectic in its methods, but differs from it in important respects. In a dialectical argument, for example, the opponent may refuse to accept a proposition that a philosopher would accept: “The premises of the philosopher’s deductions or those of the one investigating by himself, though true and familiar, may be refused by … [an opponent] because they lie too near to the original proposition, and so he sees what will happen if he grants them. But the philosopher is unconcerned about this. Indeed, he will presumably be eager that his axioms should be as familiar and as near to the question at hand as possible, since it is from premises of this sort that scientific deductions proceed” (Topics 155b10–16). Since the truth may well hinge on propositions whose status is just like these premises, there is no guarantee that what a dialectician considers most defensible will be true.

  Drawing on this new class of endoxa, then, the philosopher examines both the cl
aim that being is a single unchanging thing, and the claim that it is not, in just the way that the dialectician does. As a result, he determines, let us suppose, that the most defensible, or least problematic, conclusion is that in some senses of the terms, being is one and unchanging, in others, not. To reach this conclusion, however, he will have to disambiguate and reformulate endoxa on both sides, partly accepting and partly rejecting them. Others, he may well have to reject outright, so that beliefs that initially seemed to be endoxa—that seemed to be unproblematic—will have emerged as only apparently such (Topics 100b23–25). These he will have to explain away: “We should state not only the truth, but also the cause of error—for this contributes towards producing conviction, since when a reasonable explanation is given of why the false view appears true, this tends to produce belief in the true view” (Nicomachean Ethics 1154a22–25). If, at the end of this process, the difficulties are solved and most of the most-authoritative endoxa are left, that, Aristotle claims, will be a sufficient proof of the philosopher’s conclusion (1145b6–7).