Owen (1986) Tithenai ta phainomen

Owen G. E. L., (1986). Tithenai ta phainomen. In Nussbaum ed Martha, In Logic, Science and Dialectic.

1.

There seems to be a sharp discrepancy between the methods of scientific reasoning recommended in the Analytics and those actually followed in the Physics. The difference is sometimes taken to lie in the fact that the Postervor Analytics pictures a science as a formal deductive system based on necessary truths whereas the Physics is more tentative and hospitable both in its premisses and in its methods. But this is too simple a contrast. It is true that for much of the Physics Aristotle is not arguing from the definitions of his basic terms but constructing those definitions. He sets out to clarify and harden such common ideas as change and motion, place and time, infinity and continuity, and in doing so he claims to be defining his subject-matter (Phys. III 1, 200b12-21). But after all the Analytics shows interest not only in the finished state of a science but in its essential preliminaries; it describes not only the rigorous/ deduction of theorems but the setting up of the archai, the set of special hypotheses and definitions, from which the deductions proceed. And the Physics, for its part, not only establishes the definitions of its basic concepts but uses them to deduce further theorems, notably in books VI and VIII. The discrepancy between the two works lies rather in the fact that, whereas the Analytics tries (though not without confusion and inconsistency) to distinguish the two processes of finding and then applying the principles, the Physics takes no pains to hold them apart. But there seems to be a more striking disagreement than this. It concerns the means by which the principles of the science are reached.

In the Prior Analytics Aristotle says: ‘It falls to experience to provide the principles of any subject. In astronomy, for instance, it was astronomical experience that provided the principles of the science, for it was only when the phainomena were adequately grasped that the proofs in astronomy were discovered. And the same is true of any art or science whatever’ CAR Ta aU, 46a17-22). Elsewhere he draws the same Baconian picture: the phainomena must be collected as a prelude to finding the theory which explains them. The method is expressly associated with natural science (phusiké) and the natural stock and 85 from the example in these contexts — astronomy — it seems clear that the phainomena in question are empirical observations.” Now such a method is plainly at home in the biological works and the meteorology ;’ equally plainly it is not at home in the Physics, where as Mansion observes ‘in general everything comes down to/ more or less than founded detailed conceptual analyses — analyses often guided and illustrated by, rather upon, empirical data’ (Introduction a la physique aristotélicienne? (Louvain 1946), p.211). In this sense of ‘phainomena’ it would be grossly misleading for Aristotle to claim that he is establishing the principles of his physics upon a survey of the phainomena. And there his critics are often content to leave the matter.

But in other contexts similarly concerned with methods of inquiry ‘phainomena’ has another sense.‘ In the Nicomachean Ethics Aristotle prefaces his discussion of incontinence with the words: ‘Here as in other cases we must set down the phainomena and begin by considering the difficulties, and so go on to vindicate if possible all the common conceptions about these states of mind, or at any rate most of them and the most important’ (EN VII 1, 1145b2-6). Here Sir David Ross translates phainomena by ‘observed facts’, a translation evidently designed to bring Aristotle’s programme into conformity with such passages as those already cited. But this can hardly be its sense here. For, in the first place, what Aristotle proceeds to set out are not the observed facts but the endoxa, the common conceptions on the subject (as the collocation of phainomena and endoxa in his preface would lead us to expect). He concludes his survey with the words ‘these are the legomena (things said)’ (VII 1, 1145b8-20), and the legomena turn out as so often to be partly matters of linguistic usage or, if you prefer, of the conceptual structure revealed by language (especially VII 1, 1145b10-15, 19-20). And, secondly, after this preliminary survey Aristotle turns to Socrates’ claim that those who act against their own conviction/ or what is best do so in ignorance, and says that this is plainly in conflict with the phainomena (EN VII 2, 1145b27-8). But he does not mean that, as Ross translates it, ‘the view plainly contradicts the observed facts’. For he remarks later that his own conclusion about incontinence seems to coincide with what Socrates wanted to maintain (VII 3, 1147b14-15), and in reaching it he takes care to answer the question that he had named as a difficulty for Socrates, namely what kind of ignorance must be ascribed to the incontinent man (VII 2, 1145b28-9; 3, 1147b15-17). So Socrates’ claim conflicts not with the facts but with what would commonly be said on the subject, and Aristotle does not undertake to save everything that is commonly said. He is anxious, unlike Socrates, to leave a use for the expression ‘knowing what is right but doing what is wrong’, but he is ready to show a priori that there is no use for the expression ‘doing what is wrong in the full knowledge of what is right in the given circumstances’.° It is in the same sense of the word that all dialectical argument can be said to start from the phainomena.

This ambiguity in phainomena, which was seen by Alexander (Meteor. 33.6-9), carries with it a corresponding distinction in the use of various connected expressions. Induction (epagégé) can be said to establish the principles of science by starting from the data of perception (A.Pst. II 19, 100b3-5; I 18, 81a38-b9)./ Yet epagégé is named as one of the two cardinal methods of dialectic (Top. I 12, 105a10-19) and as such must begin from the endoxa, what is accepted by all or most men or by the wise (Top. I 1, 100b21-3); and in this form too it can be used to find the principles of the sciences (Top. I 2, 101a36-b4). Similarly with the puzzles (aporiaz). When the phainomena are empirical data such as those collected in the biology and meteorology, the aporiai associated with them will tend to be questions of empirical fact (Meteor. I 3, 357b26-30) or of the explanation of such facts,’ or the problem of squaring a recalcitrant fact with an empirical hypothesis (Meteor. I 2, 355b20-32). In the discussion of incontinence, on the other hand, where the phainomena are things that men are inclined or accustomed to say on the subject, the aporiai that Aristotle sets out are not unexplained or recalcitrant data of observation but logical or philosophical puzzles generated, as - - such puzzles have been at all times, by exploiting some of the things commonly said. Two of the paradoxes are veterans, due to Socrates and the sophists (EN VII 2, 1145b23-27, 1146a21-31). The first of the set ends with the words: ‘If so, we shall have to say that the man of practical wisdom is incontinent, but no one would say this’ (not that it happens to be false, but that given the established use of the words it is absurd) (ibid. 1146a5-7). The last ends: ‘But we say [i.e. it is a common form of words] that some men are incontinent, without further qualification’ (ibid. 1146b4-5).

Now if the Physics is to be described as setting out from a survey of the phainomena it is plainly this second sense of the word that is more appropriate. Take as an example the analysis of place. It opens with four arguments for the existence of place of which the first states what doke or seems to be the case (it appeals to established/ ways of talking about physical replacement) (Phys. IV 1, 208b1,5), the third states what certain theorists say (legoust: ibid. 208b26), the fourth quotes what Hesiod and the majority think (nomizousz: ibid. 208b32-3), and the remaining one relies on the doctrine of natural places which is later taken as an endoxon.* Of the aporiai which follow, one is due to Zeno, one is due to an equally rich source of logical paradoxes of which I shall say more in a later section, and all ultimately depend on the convictions or usage of the many or the wise. Nor are these arguments merely accessory to the main analysis: those of the dokounta which survive the preliminary difficulties are taken over as premisses for what follows.°

‘For if the difficulties are resolved and the endoxa are left standing’, as Aristotle says in both the Physics and the Ethics, ‘this in itself is a sufficient proof’.'° As for epagégé, when it is used in the argument it proves to be not a review of observed cases but a dialectical survey of the senses of the word ‘in’."

Can we appeal to this ambiguity in Aristotle’s terminology in order to explain how such a generalisation as that quoted from the Prior Analytics could be taken to cover the methods of the Physics? By now the ambiguity seems too radical for our purpose. Even within the second sense of phainomena, the sense in which it is equated with endoxa and legomena, some essential distinctions lie concealed. For an appeal to a legomenon may be an appeal either to common belief about matters of fact (e.g. EN 1 11, 1101a22-24) or to established forms of language (e.g. VIT 1, 1145b19-20: 2, 1146b4-5) or to a philosophical thesis claiming the factual virtues of the first and the analytic certainty of the second (e.g. I 8, 1098b12-18). And the broader ambiguity between the two senses of the word was one which Aristotle himself had the means to expose. For when he wishes to restrict phainomenon to its first sense he calls it expressly a perceptual phainomenon and distinguishes it from ‘an endoxon (Cael. III 4, 303a22-23). And in the De Caelo it is this more precise form of words that he uses to describe the criterion by which the correctness of our principles in physics must ultimately be assessed (7, 306a16-17).

I think such considerations show that it is a mistake to ask, in the hope of some quite general answer, what function Aristotle assigns to phainomena, or to aporiai, or to epagogé; for they show how the function can vary with the context and style of inquiry. But we have pressed them too hard if they prevent/ us from understanding how Aristotle could have taken the formula in the Analytics to apply to the Physics as well as to the Historia Animalium. If there is more than one use for the expression phainomena, the uses have a great deal in common. Thus for example it is not a peculiarity of phainomena in the second sense that they may fail to stand up to examination; for so may the phainomena of perception,” and within this latter class Aristotle is careful to specify only the reliable members as a touchstone for the correctness of physical principles."* As for his favourite example, astronomy, Aristotle knew (or came to realise) how inadequate were the observations of the astronomers (PA 15, 644b24-28). And of the biological ‘observations’ many were bound to be hearsay, /egomena to be treated with caution (e.g. HA II 1, 501a25-b1). Such phainomena must be ‘properly established’, ascertained to be ‘true data’ (A.Pr. I 30, 46a20, 25). In the same fashion the endoxa must pass the appropriate scrutiny, but in doing so they too become firm data.'* Nor, if Aristotle associates the phainomena with experience (empezria), as he does in the text from the Analytics, must it be supposed that his words are meant to apply only to phainomena in the first sense. Endoxa also rest on experience, even if they misrepresent it (e.g. Div. 1, 462b14-18). If they did not Aristotle could find no place for them in his epistemology; as it is, an endoxon that is shared by all men is zpso facto beyond challenge.'°

2.

There is no need to go on. It might indeed be objected that the evidence does not necessarily show that Aristotle was indebted to the Parmenides; both Plato and Aristotle may have been drawing on a lost source. These problems were surely discussed in the Academy,” and the Academy in turn must surely have drawn on earlier arguments, in particular those of Zeno and Gorgias. The general purposes of this paper would be as well served by such a theory, but it cannot account for the intricate correspondence that we have seen in our two texts. 102 Gorgias’ part in the matter is guesswork: the evidence for his sole adventure into abstract thought has been contaminated, probably beyond cure, by traditions to which both the Parmemides and the Physics contributed. Of Zeno luckily we know more; we know that Plato does echo some arguments of Zeno,/ but that he transforms them radically for his own ends.** The Parmenides was not a historical anthology, and when Aristotle’s words and ideas coincide closely with those of the dialogue he is under the spell of a work of astonishing brilliance and originality. A work, moreover, of logic or dialectic, not in the least a piece of empirical science; and the Physics is in great parts its successor.

This is not to say, of course, that Aristotle would call his methods in the Physics wholly dialectical. He, and his commentators on his behalf, have insisted on the distinction between ‘physical’ and ‘dialectical’, or ‘logical’, or ‘universal’, arguments; and no doubt some of the reasoning in the Physics falls within the first class. Yet even if the distinction were (as it seldom is) sharp and fundamental in sciences where a knowledge of particular empirical fact is in question (e.g. GA I1 8, 747b27-748a16), we need not expect it to be soin such an inquiry as the Physics. This is clear from the one major example of the contrast that is offered in the work, the dialectical and physical proofs that there can be no infinite physical body.** The dialectical proof is evidently distinguished by the fact that it proves too much:/starting from a definition that applies to mathematical as well as to physical solids, it reaches conclusions that apply to both sciences.*> Yet immediately after his promise to turn to physical arguments Aristotle produces a proof that no complex body can be infinite, and this proof shares the characteristics of its predecessor. It relies partly on quite general definitions of ‘body’ and ‘infinite’ (204b20-21), partly ona treatment of the ratio between finite and infinite terms which could be formulated quite generally,’ and which in fact is later given a different application to speed and resistance (IV 8, 215b10-216a11);and partly, perhaps, onthe argument against an infinite number of elements which occurs in the first book and relies largely on quite general premisses (III 5, 204b12-13; 16, 189a12-20). Certainly there are other arguments in the context which seem to depend on special empirical claims, such as the unfortunate hypothesis of natural places.*” But the impulse throughout the work is logical, and the restriction of its subject-matter to movable bodies and their characteristics does not entail a radical difference of method from other logical inquiries. It makes for better understanding to recall that in Aristotle’s classification of the sciences the discussions of time and movement in the Parmenides are also physics.