by Julia Annas
Epicurus’ theory of knowledge was not regarded as particularly impressive; indeed it was widely regarded as hideously crude. Later Epicureans, however, did develop interesting analogues of what we think of as issues of induction – how, from a number of particular observations, we can come to make justifiable generalizations about all occurrences of this kind of thing. Are we justified, for example (‘we’ being Epicurean philosophers living in Italy) in inferring that because all the humans we have observed are mortal, so are humans in hitherto undiscovered countries, such as Britain? (If there are any humans there, adds Philodemus, the philosopher whose example this is.)
Empiricist theories of knowledge, like an emphasis on the knower’s relation to particular facts, are the minority stream in ancient epistemology. What emerges from even a cursory survey of ancient concerns with knowledge, however, is the width and diversity of approaches. A student of epistemology in the ancient world would find a number of challenging theories and an extensive tradition of debate. She would find several ways of understanding knowledge: theories about wisdom and theories about apprehension of particular facts, theories privileging abstract reasoning and theories privileging the basic reports of the senses. She would also find extensive engagement not just with knowledge, but more generally with problems of belief and the powers of reasoning, both positive and negative.
Chapter 5
Logic and reality
The syllabus
If you specialize in philosophy at university, you discover that there are some skills you have to acquire, and topics you have to cover, in order to become competent in the subject. You will have to do some courses in logic and critical thinking, and cover topics in metaphysics, epistemology (and possibly philosophy of science), and in ethics, political philosophy (and possibly aesthetics). You may also have to do some history of philosophy, which will almost certainly be done in a way critical of philosophers, past and present, in what are seen as ‘other’ traditions, although philosophers in what is seen as ‘your’ tradition will be treated more respectfully.
In the ancient world things were not so different. After the establishment of Plato’s Academy, philosophical schools devoted to different philosophical traditions were the major places where philosophy was learnt, taught and passed on. Wealthy individuals might have philosophy tutors in their homes, but these would typically have been trained in some philosophical school. Each school would belong to a definite tradition, within which certain texts (typically Aristotle’s, or the Stoics’) were privileged. And from fairly early on the philosophical curriculum consisted of three parts: logic, physics and ethics. This happened early enough for it to be ascribed (unconvincingly) to Plato, though it is clear that neither Plato nor Aristotle wrote with such a curriculum in mind; it fits the interests of later schools, like the Stoics and Epicureans, far better. So far we have looked at an important topic in the ethics part of the curriculum, and also at theory of knowledge, which was considered part of logic, since logic was construed broadly, so as to cover what we call epistemology and philosophy of language. But there was also logic as we generally understand the term, more narrowly. And there is the topic that sounds oddest to us, ‘physics’.
Logic
Why is logic needed as part of philosophy? This topic was controversial then as now, some holding that logic was a part of philosophy in its own right, others that it was only a ‘tool’ that we use in order to improve our study of philosophy proper. Either way, we need logic to ensure that our arguments are sound ones, with no lurking fallacies for opponents to exploit, and also to enable us to detect weaknesses in the ways our opponents argue. In ancient philosophy logic has the function of sustaining philosophical truths and demolishing philosophical mistakes. Logic developed for its own sake was often regarded as a potential distraction from the central concerns of philosophy.
Logic is one of the more impressive achievements of Aristotle. Finding no given systematic techniques for classifying and distinguishing arguments that just persuade people from arguments which lead to true conclusions by valid inferences (and also finding, as today, many influential people glorying in conflating the two), Aristotle systematized the notion of valid argument and constructed an extensive logical system.
The centre of Aristotle’s logic is the idea of a deduction, in Greek sullogismos. He defines it quite generally: a deduction is an argument in which, some things having been laid down, something other than the things laid down comes about by necessity, because these things are so. More formally, the conclusion of a deduction follows necessarily from the premisses. Aristotle adds that the conclusion must be something different from the premisses; hence he is not trying to capture what modern logicians are after when they hold that ‘If p then p’ is a valid argument. He also holds that the truth of the conclusion must come about in a way that is through the truth of the premisses, thus excluding redundant premisses making no contribution in establishing the truth of the conclusion. Here too he diverges from modern notions of purely formal inference. There has been a large amount of (unsettled) modern discussion as to what Aristotelian deductions are, in terms of modern formal logic, and hence as to how his logic should be classified.
In modern terms Aristotle’s is only a fragment of logic, since, despite the wide scope of his definition of a deduction, he systematically studies only a much narrower range of deductions, those that have come to be known as Aristotelian syllogisms. He considers statements, positive and negative, that have the form of claiming that a predicate P, ‘belongs to’ or does not ‘belong to’ a subject S, in all, some or no cases. (As developed since the Middle Ages, these statements take the more familiar form of ‘All Ss are P,’ Some Ss are P,’Some Ss are not P, and ‘No Ss are P’.) Aristotle’s greatest contribution is the use of schematic letters, which enables him to study the form of an argument regardless of its particular content. He systematizes the ways in which two statements in one of these forms, which share a common term (the ‘middle’ term) lead to a conclusion. Some of these combinations will give valid arguments, others not. Aristotle devotes great ingenuity to showing which forms are valid, and which are not. (He also begins to develop a system of ‘modal logic’, that is, a logic of statements modified by ‘necessarily’, ‘possibly’ and so on, but less successfully.)
Various suggestions have been made as to why Aristotle should have limited himself in this way. Fairly plausible is the idea that, although he is interested in arguments as such, Aristotle is most concerned to formalize the type of argument which finds its home in his model of a completed science or body of knowledge, one in which what is at stake is the relations of kinds of thing, and claims about what holds universally are particularly important. Arguments involving individuals find no place in this logical system (though they appear in fleeting thoughts on Aristotle’s part about a ‘practical’ logic of arguments that lead to action).
Aristotle hints at ideas, developed further by his pupil Theophrastus, of systematizing arguments where what is studied are the relations between the statements, rather than the terms which form part of them. Real progress here, however, was left to the Stoics, in particular Chrysippus. Stoic logic concerns statements or axiomata, which assert or deny something. Compound statements are produced by joining simple statements by various connectives, such as ‘and’, ‘or’ and ‘if’. Stoic logic studies arguments which are made up of premisses and conclusion, where these are all statements; much of it overlaps with modern ‘propositional logic’, though there are differences. Five argument-schemata are taken as basic (the schematic letters P and Q stand in for statements). These are: (1) If P, then Q, P; therefore, Q (still familiar, as ‘modus ponens’), (2) If P, then Q, not-Q; therefore not-P (‘modus tollens’), (3) Not both P and Q, P; therefore not-Q, (4) Either P or Q, P; therefore, not-Q, (5) Either P or Q, not-P, therefore Q. From this basis Stoic logic developed in sophisticated and powerful ways.
As with Aristotle, the Stoics were not merely interested i
n argument for its own sake. They were concerned to produce arguments which were also ‘proofs’ – arguments which, as they put it, ‘by way of agreed premisses, reveal by deduction an unclear conclusion’. Logical form is studied in the service of representing our claims to knowledge, in this case the way we claim to reach knowledge of ‘unclear’ or theoretical matters by way of what we can agree on in our experience.
Epicurus and his school affected to despise formal logic as a trivial waste of time. But they also spent energy on studying what were called ‘signs’ on the basis of which we make inferences from what we experience to matters that are beyond our own experience; so they engaged other schools in discussion about logic to some extent.
Students of philosophy in the ancient world could (unless they were Epicureans) expect to study both Aristotelian and Stoic logic, which were seen as complementary, although there could be disputes as to which was the more important. By historical accident, Stoic logic was lost, along with much early Stoicism, at the end of antiquity, whereas Aristotle’s logic not only survived but became regarded as all there was to logic. It was elaborated in the Middle Ages, regarded as complete by Kant and dislodged from its place in the syllabus only by the rediscovery of propositional logic by Frege and Russell at the beginning of the twentieth century.
Nature and science
The third part of the philosophical curriculum, ‘physics’, no longer sounds as though it even belongs to philosophy. This is partly because of our narrowing of a term which originally meant the study of nature or phusis. Nature is just everything that there is, or the world (including humans, who are part of the world). Hence the study of nature can cover a number of very different things, and ‘physics’ covers a range of enquiries which for us have got segregated into different subject-matters and taught in very different ways.
One type of enquiry seeks explanations for the puzzling things we see around us and are exposed to. What explains the regularities of the sun and the moon? What brings about the seasons, so crucial for farmers? Why are there hurricanes, earthquakes, eclipses? In the ancient world these were regarded as issues which were part of the study of nature as philosophers undertook that. As philosophy developed, however, and especially after Aristotle, these questions lost much of their interest, since there were numbers of theories about them, but no decisive ways of deciding between these, and so no convincing way of showing any given answer to be correct. They became regarded as suitable material for dinner-party discussion rather than live philosophical questions. In the modern world, of course, the advances of science, whatever their other drawbacks, have provided us with firm answers to questions like these. They no longer seem remotely philosophical, and ancient discussions of them are often put into the history of science.
The study of nature narrowed in another way also, especially in the period after Aristotle, with the development of bodies of scientific knowledge separate from philosophy. Medicine, though crude by modern standards, developed as a specialized science, with differing schools. It was the mathematical sciences, however, which made the greatest strides, with Euclid’s Elements a high point. Archimedes was not only a great mathematician, but developed astronomy and also applied branches like engineering. Historians of science sometimes lament the fact that sophisticated technical ideas were applied in trivial ways; Heron of Alexandria describes a machine for making figures mechanically pour libations on an altar. But basic facts about the ancient economies precluded anything like the development of our industrial technology. Whether we are obviously the winners here is another matter.
Physics and metaphysics
The study of nature, or ‘physics’ in the ancient sense, however, covered more than what became narrow scientific enquiries. From the beginning, ‘nature’ could be used for ‘what there is’, everything that there is to be studied. Hence much of ancient ‘physics’ is so broad as to correspond to what we think of as metaphysics. Is change a necessary feature of our world? What is change, anyway? In the world around us, what are the real entities, the things that are basic to a true view of the way the world really is? Are living things, like animals and humans, such basic entities? They seem to be the subjects of changes, the things changes happen to. But if what is real is the subject of change, then perhaps in looking for what is real we should not stop with the living things, but look for whatever it is which in them is the subject of change. Perhaps this is the material they are made out of. Issues like this are central to the philosophical enquiries of many of the so-called Presocratics and of Aristotle, who engaged with their ideas and is our major source for many of them. They are not part of modern science, but of more abstract philosophical enquiries, generally called metaphysical. Often the dividing-line between Aristotle’s ‘physics’ and his ‘metaphysics’ is a thin one.
Such questions were thought to arise naturally in the context of a general view of the world as a whole. Given the less ambitious scope of modern metaphysics, they are often studied in relative isolation. Thus we tend to see Plato’s ‘theory of Forms’, for example, as a metaphysical theory that has nothing to do with what we think of as physics or the study of nature. In the ancient world, however, it was mostly seen as one aspect of Plato’s ‘physics’ or theory of the world, which was primarily studied in the Timaeus, a dialogue not very popular today which contains Plato’s cosmology or account of the universe and its structure. (See box, p.82).
What you studied as ‘physics’ or metaphysics in the ancient world would depend a great deal on what tradition of philosophy you primarily belonged to. Epicureans, for example, held that physical and metaphysical questions mattered only to the extent that getting the answers to them wrong led us to be disturbed and unhappy; getting interested in them for their own sake was a misuse of time that would be better employed learning more directly how to live well. The Stoics thought it important to get right the major metaphysical points about the world: it is governed by providence, and a rational appreciation of it will discern how everything in it is for the best. But they were not much more interested than the Epicureans in getting the details right for their own sake.
Aristotle on nature
Among ancient philosophies it is the Aristotelian tradition which has the broadest and most generous conception of what the study of nature is. Aristotle had the reputation of being the philosopher most interested in causes and explanations. And, although his account of nature is not one that we, with our modern scientific knowledge, can still accept, we can still appreciate the main lines of it as embodying a response to our world which is highly worthy of respect.
* * *
Plato’s ‘theory of Forms’
Plato has no explicit theory of Forms. In some dialogues, especially Phaedo, Republic, Symposium, Phaedrus, and Timaeus, there are passages, some with argument and others more expressive and metaphorical, which introduce in various ways items we usually call Forms, but for which Plato never develops a standard terminology.
In contrast to the things in our experience which are beautiful, Plato introduces the idea of the ‘beautiful itself’, which is beautiful in a way not relative to context or time or perspective. Unlike all the beautiful things and people in our experience, the beautiful itself is never not beautiful. This idea is developed with value terms like beautiful, just, and good, and with mathematical terms like double and half. It is notoriously unclear how Plato’s arguments could be extended beyond terms with opposites. Despite a widely misinterpreted passage in the Republic, Plato does not think that there is a Form for every general term; Forms are not what came to be called universals. There is a Form only where there is an objective nature that can be known by being intellectually grasped; Forms are always associated with using your mind to reason, as opposed to relying uncritically on your sense experience. The most famous passages about Forms stress this contrast between the mindless assumption that what experience impresses on you is just what there is, and the critical use of reason to grasp realities, the Fo
rms, that are accessible only to the enquiring mind.
In the dialogue Parmenides Plato shows that he is aware of apparent inconsistencies in what he has said about Forms. The right response, however, he holds, is not to abandon Forms but to continue to argue on both sides until a defensible position is reached. He continued in this spirit, never successfully producing a definitive theory of Forms. Later philosophers have often simplified the issues, but Aristotle and late twentieth century philosophers have explored the different arguments Plato employs for and against the existence of Forms.
* * *
For Aristotle, nature is the world made up of things that have natures. What is it to have a nature? It is to be a thing which has a source internal to itself of changing and being changed. We can understand what a lion is only by looking at lions themselves and at the way they interact with their environment and other species. To understand an artefact like a shield, by contrast, we have to appeal to something external to the shield itself – the designs of humans that made it. Things with natures are primarily living things, such as plants and animals, including humans. For Aristotle, then, nature is, right from the start, not just whatever happens, the undifferentiated totality of what there is (as it is for Mill and others from the nineteenth century on). Nature is already a world of things that organize themselves and live characteristic lives, and to understand nature is to come to understand what kind of lives these are. Nature is active, a system of living and changing things. There is no hint in Aristotle of the view, notorious in many scientists since the early modern period, that nature is passive, lying out there to be mastered by the scientific mind.