Epicurianismo

Epicureanism

Epicureanism is one of the three dominant philosophies of the Hellenistic age. The school was founded by Epicurus (341-271 bc) (see Prolēpsis ). Only small samples and indirect testimonia of his writings now survive, supplemented by the poem of the Roman Epicurean Lucretius, along with a mass of further fragmentary texts and secondary evidence. Its main features are an anti-teleological physics, an empiricist epistemology and a hedonistic ethics.

Epicurean physics developed out of the fifth-century atomist system of Democritus. The only per se existents are bodies and space, each of them infinite in quantity. Space includes absolute void, which makes motion possible, while body is constituted out of physically indissoluble particles, 'atoms'. Atoms are themselves further measurable into sets of absolute 'minima', the ultimate units of magnitude. Atoms are in constant rapid motion, at equal speed (since in the pure void there is nothing to slow them down). Stability emerges as an overall property of compounds, which large groups of atoms form by settling into regular patterns of complex motion. Motion is governed by the three principles of weight, collisions and a minimal random movement, the 'swerve', which initiates new patterns of motion and obviates the danger of determinism. Atoms themselves have only the primary properties of shape, size and weight. All secondary properties, for example, colour, are generated out of atomic compounds; given their dependent status, they cannot be added to the list of per se existents, but it does not follow that they are not real. Our world, like the countless other worlds, is an accidentally generated compound, of finite duration. There is no divine mind behind it. The gods are to be viewed as ideal beings, models of the Epicurean good life, and therefore blissfully detached from our affairs.

The foundation of the Epicurean theory of knowledge ('Canonic') is that 'all sensations are true' - that is, representationally (not propositionally) true. In the paradigm case of sight, thin films of atoms ('images') constantly flood off bodies, and our eyes mechanically register those which reach them, neither embroidering nor interpreting. These primary visual data (like photographs, which 'cannot lie') have unassailable evidential value. But inferences from them to the nature of external objects themselves involves judgment, and it is there that error can occur. Sensations thus serve as one of the three 'criteria of truth', along with feelings, a criterion of values and psychological data, and prolēpseis, naturally acquired generic conceptions. On the basis of sense evidence, we are entitled to infer the nature of microscopic or remote phenomena. Celestial phenomena, for example, cannot be regarded as divinely engineered (which would conflict with the prolēpsis of god as tranquil), and experience supplies plenty of models adequate to explain them naturalistically. Such grounds amount to consistency with directly observed phenomena, and are called ouk antimarturēsis, 'lack of counterevidence'. Paradoxically, when several alternative explanations of the same phenomenon pass this test, all must be accepted as true. Fortunately, when it comes to the foundational tenets of physics, it is held that only one theory passes the test.

In ethics, pleasure is the one good and our innately sought goal, to which all other values are subordinated. Pain is the only bad, and there is no intermediate state. Bodily pleasure becomes more secure if we adopt a simple lifestyle which satisfies only our natural and necessary desires, with the support of like-minded friends. Bodily pain, when inevitable, can be outweighed by mental pleasure, which exceeds it because it can range over past, present and future enjoyments. The highest pleasure, whether of soul or of body, is a satisfied state, 'static pleasure'. The short-term ('kinetic') pleasures of stimulation can vary this state, but cannot make it more pleasant. In striving to accumulate such pleasures, you run the risk of becoming dependent on them and thus needlessly vulnerable to fortune. The primary aim should instead be the minimization of pain. This is achieved for the body through a simple lifestyle, and for the soul through the study of physics, which offers the most prized 'static' pleasure, 'freedom from disturbance' (ataraxia), by eliminating the two main sources of human anguish, the fears of god and of death. It teaches us that cosmic phenomena do not convey divine threats, and that death is mere disintegration of the soul, with hell an illusion. Being dead will be no worse than not having yet been born. Physics also teaches us how to evade determinism, which would turn moral agents into mindless fatalists: the indeterministic 'swerve' doctrine (see above), along with the logical doctrine that future-tensed propositions may be neither true nor false, leaves the will free.

Although Epicurean groups sought to opt out of public life, they respected civic justice, which they analysed not as an absolute value but as one perpetually subject to revision in the light of changing circumstances, a contract between humans to refrain from harmful activity in their own mutual interest.

1 The Epicureans

For the foundation and nature of the Epicurean school, and for Epicurus' own writings, see Epicurus . After Epicurus' death in 271 bc the school continued to flourish, at Athens and in other centres around the Mediterranean, for at least five centuries. Although there were numerous developments and schisms, these were always moderated by appeals to the scriptural authority of Epicurus' own writings and those of his close collaborators Metrodorus, Polyaenus and Hermarchus.

The surviving writings of Epicurus include Letter to Herodotus, Letter to Pythocles, Letter to Menoeceus, Kyriai doxai (Key Doctrines) - all preserved in Book X of Diogenes Laertius - and parts of On Nature preserved in papyri excavated from a villa at Herculaneum. In addition, we have from the first century bc the Epicurean poem of Lucretius, dealing mainly with physics, and numerous fragmentary papyri of the Epicurean Philodemus, excavated from the same Herculaneum villa. Finally, there is the massive second-century ad philosophical inscription which the Epicurean Diogenes erected in a public colonnade of Oenoanda (in modern Turkey) to advertise his philosophy, preserving for us much supplementary information.

2 Foundations of physics

Whatever the universe consists of must be permanent. Nothing comes into being out of nothing or perishes into nothing - two fundamental principles widely regarded as indubitable, but defended by the Epicureans on empirical grounds (Lucretius I 159- 264): literal generation would be incompatible with the observed regularities of natural processes, such as the dependence of organic growth on the right material conditions; and literal annihilation would, given the infinity of past time, by now have ensured that nothing at all was left. Nor can the universe (the 'all') be changed by addition or subtraction, since by definition it has nothing outside it that could enter or into which bits could escape.

Next comes a statement of what the universe does consist of: bodies and space (Letter to Herodotus 39-40). This is taken to be self-evident, that is, underivatively known and unchallengeable. The reasoning is no doubt that bodies are the things which most obviously have independent existence; and that since that independence is most evident in their ability to move in space, the bits of space which they vacate as they move must exist independently of them.

Space is at this stage simply presented as what the bodies are in, and what they move through - a three-dimensional extension which persists whether occupied or unoccupied. When occupied it gets called 'place', when unoccupied 'void' (kenon, literally the 'empty'), and when things move through it 'room' (chōra, etymologically linked by Epicurus with chōrein, 'go'). Epicurus prefers to use these familiar terms rather than a generic one for 'space', for which Greek has no precise word (the concept itself being one which had emerged only gradually, with Epicurus perhaps the first to isolate it clearly), and which he manages to capture only by his invented phrase 'intangible nature' (anaphēs physis). That there is void, that is, that contrary to the majority ancient view some space is altogether empty, is argued by appeal to such phenomena as motion and permeation. Motion cannot be accounted for, it is argued ( Lucretius I 370-83 ), by the alternative hypothesis of complementary redistribution of matter on the model of a fish exchanging places with the water when it swims. Even the fish would be stuck fast unless there were some void to break the deadlock. Otherwise it would be unable to move until the water had cleared a path for it, or the water until the fish had cleared a path for it.

As for body, it remains for now largely unanalysed, beyond a set of arguments to show that it must exist microscopically as well as macroscopically: its underlying atomic structure cannot be demonstrated until it has been shown that body and space are the sole constituents of the universe. And the next move is to show just that. First, body and space are analysed as contradictory opposites: only three-dimensional things exist per se, and if these are resistant they must be body, if non-resistant void ( Lucretius I 430-9). This is the positive proof that body and space are not only irreducibly distinct but also jointly exhaustive. There then follows a supplementary argument ( Lucretius I 449-82 ), in which all other contenders for per se existence - including properties and time - are written off as secondary attributes, parasitic on body and/or space. None of these could exist independently of bodies and/or space. Time is dependent on change, that is, on moving bodies. And even facts about the past (for example, that there was a Trojan War), which might seem to outlive the bodies (Agamemnon and others) of which they are true and therefore to acquire independent existence, are truths about places which still persist (Troy, Mycenae and so on), or, if you prefer, about the universe.

Only now that it is fully established can the body-space dualism be deployed to show that at the lowest level of analysis there will be not only portions of empty space uninterrupted by body but also portions of body uninterrupted by empty space - and therefore, since there is no third thing, totally uninterrupted. Being perfectly solid, these are 'atoms', literally 'uncuttables'. The capacity of a body to disintegrate is directly correlated to the amount of void within it; therefore a body containing no void at all is altogether indestructible ( Lucretius I 511-39 ). That there are such bodies is confirmed both by the evident fact that matter is not completely annihilated by fragmentation, and by the observed regularities of nature, which imply that something altogether unchanging underlies them.

Having mapped the universe out into space occupied by discrete portions of body, Epicurus adds that both space and body are infinite in extent (Letter to Herodotus 41- 2, Lucretius I 958-97 ). First, the infinity of the universe itself (a highly controversial thesis in antiquity, rejected by both the Platonist and the Aristotelian tradition) is argued by appeal to the notion of a limit: it could only be limited if there were something beyond it to limit it, and the notion of the universe (the 'all') precludes that. (Lucretius adds the time-honoured argument: what if I go to the supposed edge of the universe and throw something?) Second, it is argued that each of the two constituents of the universe must also, taken on its own, be infinite: finite body in infinite space would be too dissipated to form compounds, while infinite body in finite space would simply not fit.

3 Minima

The fifth-century atomists (see Atomism, ancient; Leucippus; Democritus §2) had first introduced atoms at least partly, it is generally believed, to circumvent the puzzles that Zeno of Elea (§§4-6) had derived from the supposition of infinite divisibility: any magnitude will, as the sum of infinitely many parts, be infinitely large; and motion will be impossible, since it will require the traversal of infinitely many sub-distances. But if atoms, as seems inevitable given their varying shapes and sizes, are not the smallest possible magnitudes but simply indissoluble lumps, it is hard to see that this kind of indivisibility can do anything to thwart Zeno: each atom will consist of infinitely many parts, and threaten to be infinitely large and/or untraversable.

Epicurus (Letter to Herodotus 56-9) starts by making the first clear distinction in ancient thought between (1) things which are physically indivisible, and (2) those which modern scholarship sometimes calls theoretically, conceptually or mathematically indivisible, but which in ancient usage are called either 'minimal' (elachista) or 'partless' (amerē). (For Epicurus' main predecessor in a theory of type (2), see Diodorus Cronus §2.) Not only, Epicurus says (Letter to Herodotus 56), (1) can things not be 'cut' to infinity (a reference to atoms, which are physically 'uncuttable'), as proved earlier, 'but also (2) we must not consider that in finite bodies there is traversal [moving along something part by part] to infinity, not even through smaller and smaller parts [that is, not even in a convergent series such as '. Otherwise, he warns, Zenonian consequences will ensue.

By showing that a finite body cannot contain an infinite number of parts, Epicurus considers that he has established the existence of an absolutely smallest portion of body, which henceforward he calls 'the minimum (elachiston) in the atom'. Clearly it cannot be larger than an atom, or it would not be a minimum, so it must be either an entire atom or part of one. (For reasons which emerge later, it must in fact be the latter.) The idea that an extended magnitude like an atom consists of minimal or partless units is designed to avoid Zenonian paradoxes, but, as Aristotle had pointed out in Physics VI, it generates paradoxes of its own. In particular, it is hard to see how two partless items can be adjacent, since they cannot touch either part to part (they do not have parts) or whole to whole (or they would be co-extensive, not adjacent). Epicurus answers brilliantly with the analogy of the 'minimum in sensation' - the smallest magnitude you can see. Any larger visible magnitude consists of a precise number of these minima, which are seen as adjacent without touching in either of the two ways offered by Aristotle, but, as Epicurus puts it, 'in their own special way'. This provides a model which is readily transferable to the 'minimum in the atom'.

But it also follows, of course, that any extended magnitude will consist of an exact number of minima, since after analysis into minima there could not be a fraction of a minimum left over. This makes all magnitudes commensurable, and conflicts with the geometers' recognition of incommensurable lengths, for example, the side and diagonal of a square. So much the worse for geometry, was the conclusion of Epicurus and his collaborator Polyaenus, himself formerly a distinguished geometer. They concluded, at least provisionally, that geometry is false.

In one respect the analogy between the sensible and actual minimum fails. Visible minima are capable of independent motion (for example, a falling object viewed at a suitable distance might be a sensible minimum), but actual minima are not, says Epicurus. This perhaps reflects Aristotle's argument in Physics VI 10 that a partless entity (he is thinking especially of a geometric point) could not move except incidentally to the motion of a larger body. At all events, it follows that, since atoms move, no atom consists of just one minimum.

Atoms must vary widely in shape and size in order to account for the large range of phenomena ( Lucretius II 333-477 ). But they cannot vary indefinitely, since out of a finite set of minima only a finite number of shapes can be formed. Being partless, minima cannot be partly adjacent, so two minima can be arranged in only one shape, three minima in only two shapes, and so on (Lucretius II 478-531).

4 Motion

Surprisingly, atoms never stop moving, even within a compound object, since the medium through which they move is void, which can offer them no resistance. More surprisingly, for the same reason they move at a vastly greater speed than any familiar motion through an obstructive medium such as air; even than sunlight, which is seen to spread from horizon to horizon virtually instantaneously ( Lucretius II 142-64). More surprisingly still, they all move at equal speed, since in a vacuum, unlike air, there is no resistance from the medium to slow down the lighter ones more than the heavier ones (Letter to Herodotus 61). In stating all these claims, Epicurus is accepting paradoxical consequences of the hypothesis that void exists, consequences which Aristotle had drawn (Physics IV) in the belief that they were sufficiently absurd to discredit the hypothesis. Moreover, the equal speed of atoms was confirmed by another objection Aristotle thought he had found to atomism (Physics VI 2): if there is a minimal magnitude, there can be no differences of speed, because then in the time the faster object took to travel one minimum the slower one would, impossibly, have to travel less than one minimum. Epicurus welcomed this argument, along with the conclusion Aristotle thought absurd, because his theories of void and minima now offered two independent grounds for the same conclusion, that atoms move at equal speed.

The apparent lack of fit between these findings about atoms and the variable speed of macroscopic motions is explained as follows (Letter to Herodotus 62). Even in a compound object the individual atoms are perpetually moving, but in tight and regular cyclical patterns which make the complex as a whole stable. Phenomenal differences of speed, say between two runners, represent merely the aggregate motions of the atoms in each over an observed period of time.

There are three causes of an atom's motion. The first is its own weight, interpreted as an inherent tendency to move downwards (see §8). The second is collisions with other atoms, which can deflect an originally downward motion along any number of new rectilinear trajectories, thus generating the patterns of motion of which compounds are born.

The third cause of atomic motion is the 'swerve' (parenklisis), whereby an atom may shift from its rectilinear trajectory onto an adjacent one - a displacement sideways by a distance of one minimum (there being no smaller distance). This happens 'at no fixed place or time', meaning that the occurrence of a swerve is causally undetermined. The theory, derided by Epicurus' opponents but now recognized as comparable in its implications to modern quantum indeterminism, looks like a drastic solution requiring a drastic problem. Two such problems are recorded ( Lucretius II 216-93 ). First, since all atomic motion starts out as vertical and equal in speed, without a swerve no collisions would ever have started, and hence no world could have been formed. It may be doubted whether this was a sufficiently pressing problem to motivate an abandonment of universal causality: given the infinite past history of the universe, Epicurus had no need to posit a very first collision; in which case every collision could have been explained as the effect of previous ones, stretching back into the infinite past. The second problem seems to have been the real motivation of the swerve: if all atomic motion is causally determined, free will becomes impossible (see §12).

5 Qualities

Atoms themselves have only the primary or ineliminable features of body: size, shape and weight. They lack the secondary properties of colour, flavour and so on. The ground for this parsimony is (Letter to Herodotus 54-5) that secondary properties are in their nature changeable, whereas atoms have been posited as the enduring entities which underlie change. Atomism instead treats colour and the like as purely macroscopic properties, caused by the individual shapes and overall arrangement of the constituent atoms in a compound body, and changed if that arrangement changes. Lucretius frequently compares the explanatory economy of this system to that of the alphabet, capable of generating endless different words simply by rearranging a modest stock of letters. In practice the crudely sketched atomistic explanations of macroscopic properties owe more to the shapes than to the arrangement of a thing's constituent atoms (fluids consist of smaller and rounder atoms, sweet things of smoother ones...), although the degree of separation between atoms plays some part.

There are two kinds of properties (symbebēkota). 'Permanent accompaniments' (Greek ta aidion parakolouthounta, Latin coniuncta) are essential to a thing's very existence, for example, tangibility for body and heat for fire, while accidents (symptōmata) are non-essential, for example, slavery or poverty for a human being. It is crucial to note that no ontological priority is implied for the properties of atoms over those of phenomenal bodies. The earlier atomist tradition had tended to treat atoms and void alone as real, with phenomenal properties mere arbitrary constructions placed upon them by the human mind and sense organs. This had led atomism in the direction of epistemological scepticism, a tendency which Epicurus himself strenuously resists. In discussing the status of properties (Letter to Herodotus 68-71), he emphasizes that although undoubtedly different from the per se existents, body and void, they are no less real for that. (The second-generation Epicurean Polystratus adds that even relative properties such as 'beneficial' and 'harmful', often decried as unreal by sceptics, have clear causal effects.) Accidents, in fact, have no existence at the microscopic level (Letter to Herodotus 70), and yet are real. This is a clear indication that Epicurus is consciously opposed to the atomist reductionism of his predecessors: colours and other accidents are real, yet irreducibly different from atomic structures. Although atomic structures are causally prior to the phenomena which they generate, they are in no way ontologically privileged over them.

6 The criteria of truth

Since, then, Epicurus' version of atomism allows the reality of sensible properties, he does not inherit Democritus' motive for casting doubt on the veracity of sense-perception (see Democritus §3 ). And indeed for Epicurus the primary 'criterion of truth' is the senses: 'All sensations are true.'

The soul is an atomic structure spread throughout the body, but with a command centre which houses rational thought (these can be considered functionally, although not anatomically, equivalent to the nervous system and brain, respectively). A sensation is the soul's mechanical but conscious registering of the phenomenal properties of an external body. In the paradigm case of sight, this occurs because the outer layers of bodies are constantly streaming off them in all directions, taking the form of 'images' (Greek eidōla, Latin simulacra), atom-thin films of matter which more or less preserve their colour and shape in transit (Letter to Herodotus 49-52, Lucretius IV 26-268). Streams of these can enter the eye, producing vision; isolated ones can also directly enter the mind, asleep or awake, and enable it to visualize objects ( Lucretius IV 722-822, 962-1036). Importantly, any act of picturing, by the eyes or the mind, is the registration of one or more images arriving from outside. The infallibility of the senses consists in their mechanical registering of these images, without adding, subtracting, embroidering or interpreting. A useful analogy is photography, which 'cannot lie' because it merely records mechanically the patterns of light arriving at the camera lens, and leaves it to us to interpret what they represent. All sensations are bona fide evidence about the external world. All error lies in the 'added opinion' (to prosdoxazomenon) by which the mind interprets these data. Note that truth here is representational, not propositional: the sensation is true because it accurately represents the physical data reaching the sense organ. The opinion based on it may be regarded as true or false according to whether or not it succeeds in representing accurately the state of affairs which caused the sensation. Even in a case of outright delusion it is the mind which, because deranged, misinterprets the perfectly accurate impressions which reach it. As for the more important counterexamples to the theory, those of optical illusions, the paradigmatic case is the square tower which looks round at a distance. The images start out square, but because of the distance which they must travel through obstructive air they are rounded on arrival at the eyes (if this is too crude to be credible, substitute the image of a coin, oval due to perspective). Therefore vision is accurately registering the images as they are on arrival. This is a correct, not a misleading, view of a distant tower, and to call it a case of mis-seeing, says Epicurus (Sextus Empiricus, Against the Professors VII 206-10), is like saying that you are mishearing someone across the room just because you are not hearing their voice as it would sound inside their mouth.

The veracity of the senses is not, as many hold, impugned by their ability to conflict with each other. Strictly speaking, no two sensations are commensurable. If sight and touch differ about the shape of an object, that is because sight reports the shape of a colour patch, touch the shape of a body. If two visual sensations of the same object conflict, as in the case of the tower, that is because they are reporting two different kinds of colour patch, mediated by different quantities of intervening air.

The photographic reliability of the senses is what makes them not only true but a criterion of truth. However, when Epicurus appeals to the evidence of the senses as his ultimate criterion, it is not usually to these raw sensations, still awaiting interpretation by the mind, but to facts directly attested by them, for example, that things move, or that round objects move more readily than jagged ones. He regularly talks as if these interpreted sensations acquire the self-evident veracity which initially belongs only to the raw impressions.

The second criterion of truth is prolēpsis, or 'preconception' - a key term in ancient epistemology, first introduced by Epicurus (see Prolēpsis). The prolēpsis of a thing is an instinctively acquired generic grasp of its nature, which enables you to recognize instances of it and is available for analysis in conceptual inquiries. Your prolēpsis of a human being, acquired unreflectively by accumulated past experience, enables you both to recognize humans when you meet them, and in the course of an inquiry to establish their essential features, for example, rationality and mortality. (When prolēpseis are said to be built up from past sensory experiences, these must again be interpreted sense-impressions, not raw ones.) Other prolēpseis appealed to in Epicurean arguments include those of god, body, utility and responsibility. Prolēpseis are taken to be common to all human beings, and therefore to function like a set of shared intuitions which we can hope to rediscover beneath our acquired false beliefs and to use as common ground for joint philosophical inquiry

The third criterion of truth is feelings (pathē), which generically divide up into just two kinds: pleasure and pain. As such, feelings certainly function as criteria of value, since Epicurus equates good with pleasant, bad with painful. But they also provide the introspective data on which Epicurus founds his psychological theory (Letter to Herodotus 63).

7 Scientific method

Starting from the direct data of the senses, broader and less accessible truths can be established. Epicurus apparently divided the accessing process into two classes, 'attestation' (or 'witnessing', 'confirmation', and so on: epimarturēsis) and 'non-contestation' (or 'lack of counterevidence', 'non-confirmation', and so on: ouk antimarturēsis). Attestation plays a relatively minor part in scientific inference, since it seems confined, if not to truths directly supplied by the senses, at most to inductive generalizations derived from these. The main concern of Epicurean science is the extension of such knowledge beyond the realm of direct experience, into inaccessibly remote regions of the universe such as the heaven or other worlds, and into the microscopic realm of atoms and void. Here theories cannot be directly confirmed, but are tested instead by their consistency with empirical data: hence 'non-contestation' by phenomena.

Not that mere consistency with experience is enough to establish truth. A theory is only entertained in the first place if it has some explanatory power or recommends itself in some comparable way. But there may be several alternative theories of equal merit in this regard, and if so each must be tested for 'non-contestation' by phenomena, to see which survives. There are many intrinsically credible general theories of matter, for example, including the four-element theory and the monistic fire theory of Heraclitus, and likewise many intrinsically plausible theories of vision, and so on. But all except the Epicurean theory fail somehow to be consistent with the entire range of phenomena against which they are tested.

In some other cases, however, especially astronomical ones, Epicurus admits that two or more rival theories survive the test. This might be considered an indication that the consistency test can at best prove a theory possible, not true. But Epicurus insists that in such cases all the successful theories are indeed true. Sometimes this means that several different explanations of the same phenomenon - for example, thunder, or the generation of perceptual 'images' (see §6) - operate concurrently in our world. In other cases it means no more than that, given the intrinsic possibility of the hypothesized causal process, it must operate somewhere in the infinite universe, even if not here.

All such 'sign-inferences' (sēmeiōseis) are from something evident to something non-evident. Often this is from the macroscopic to the microscopic: either causal inference from a macroscopic effect to a microscopic cause (for example, from the observed regularities of nature to the existence of unchangeable elements; see §2), or analogical inference (for example, from the mobility of observed spheres to that of spherical atoms, see §5). Others are from the macroscopic to the macroscopic: either from accessible to inaccessible entities or events (for example, from the mechanism of the water-wheel to the rotation of the stars), or from the present to the past or future (for example, from the current structure of a social institution like law or language to its historical origin, or from the impermanence of familiar compounds to the future destruction of the world).

The nature and validity of such sign-inferences is debated between the Epicureans and the Stoics in Philodemus, On Signs , including an important controversy about the justification of induction.

8 Cosmology

Epicurus argues that there can be no creating or controlling divinity (see §9), and that our world, one of infinitely many, is an accidental and temporary product of large-scale atomic collisions (for the role of the 'swerve' in this, see §4). Apparent evidence of divine creation can be explained mechanistically by appeal to the power of accident on a sufficiently large scale. In view of (see end of §3) the existence of no more than a finite number of types of atom, which can in turn be arranged relative to each other in only a finite number of ways, the number of possible structures of worlds is also - although unimaginably huge - a finite one. It seems to follow that the infinite stock of atoms must at sufficiently large intervals throughout infinite time and space, by sheer accident, actualize every such permutation, including even such well-ordered structures as our own world.

Within worlds too, accident can account for specific instances of apparent purpose Animal parts, for instance, however well suited to their functions, came into existence accidentally before those functions were (or even could have been) conceived. In the early days of life on earth many non-viable creatures were generated, but did not survive ( Lucretius V 837-77 ; a widely admired anticipation of Darwinian survival of the fittest). Even human institutions such as language and law, often attributed to divine benefactors, are formalized versions of modes of behaviour with purely natural origins in human need and instinct.

Epicurus' account of the world's structure is largely Presocratic in inspiration. Since atoms have no inherent attractive powers, he cannot accept the geocentric cosmologies of Plato, Aristotle and others, in which heavy stuffs tend not strictly downwards but inwards towards the centre. The direction 'down' is itself taken to be an absolute one, so that throughout the universe objects fall parallel to each other, rather than towards the centre of a spherical earth. For him the earth's stability depends not on its being at the centre (in infinite space there is no centre) but on the cushioning effects of the air beneath it. In a notorious passage (I 1052-82 ) Lucretius dismisses as ill-conceived the geocentrists' impressively accurate description of the antipodes.

9 God

That our world cannot be a product of divine craftsmanship is argued on several grounds (especially Lucretius V 156-234 ). Quite apart from the world's obvious imperfection, and the difficulty of finding a motive for its creation by already blissfully happy beings, the true conception of a god is incompatible with the role of cosmic administrator. A god is a supremely tranquil being, whereas the burdens of government include attitudes of anger, favour and worry.

How do we know that god is like this? In Epicurus' view there is a natural conception (prolēpsis: see §6) of god as a blessed and immortal anthropomorphic being, a conception shared by all human beings, even though in most it has been obscured by a veneer of false beliefs, for example, that the gods are vengeful, or that they govern our lives, turn the heavens and so on. People tend to endow god with their own moral values, especially the competitive values of political society, and by the same token the Epicurean reversion to the true conception of divinity as tranquil and detached is also a rediscovery of the natural human goal, tranquillity (see §10). Epicurus is insistent that 'there are gods', and even that they should be worshipped, but as an act of veneration for a life to which we ourselves aspire, not in the hope of appeasement.

But how can there be gods, if that means literally (that is, biologically) immortal beings? If according to Epicureanism nothing exists independently except bodies and void, and a god can hardly be either void or a single atom, a god must be an atomic complex. But it is a cardinal tenet that no compound body can be everlasting ( Lucretius III 806-18 ).

Here scholarly opinion divides. The majority seek special ways in which an Epicurean god can nevertheless be literally immortal, by living in sheltered regions beyond our world and by consisting of constantly replenished streams of visual 'images' (on which see §6). Others (following such sources as Sextus Empiricus, Against the Professors IX 43-7) favour an idealist interpretation of Epicurean theology, whereby god, properly understood, just is our own idealization of a happy human life. On this view, when god is said by the sources to consist of visual 'images', this is simply because according to Epicurean cognitive psychology (§6) all imagination consists in the apprehension of images which enter us from outside. The images which provide the raw material for our conception of god need not flow to us from any objectively real divine being, but may be ordinary locally generated human images. Epicurus' main point in identifying god with a stream of images was apparently the negative one of denying that god is a 'solid body' at all (Cicero, On the Nature of the Gods I 49).

Epicurus seems to have spoken repeatedly about how we should think of god as being, behaving and so on, which in itself is equally compatible with the realist and the idealist interpretation. Equally powerless to settle the dispute is the insistence of Epicurus and his followers that their position is theistic, or that of his critics that it is atheistic. But on either reading it should be clear that the importance of god in Epicurus' system is not cosmological but ethical. Having a correct conception of god is identical with moral enlightenment.

10 Hedonism

Another 'innate' human attitude, along with the prolēpsis of god, is the pursuit of pleasure as the only positive value, or 'end' (see Telos). Epicurus argues that the behaviour of the newborn, and even of non-human animals, confirms that to maximize pleasure and minimize pain is the natural and primal drive (Cicero, On Ends I 30). Just as physics began with the crude mapping out of per se existents into bodies and space (§2), so ethics starts with the mapping out of all intrinsic values into pleasure and pain. And again as in physics, the next move consists in showing these two items to be jointly exhaustive: the absence of pain is itself pleasure, and there is therefore no intermediate state (Cicero, On Ends I 37-9).

This controversial thesis goes to the heart of Epicurus' ethics. In his view, most human misery results from ignorance of how to quantify pleasure. Where some hedonists, especially the contemporary Cyrenaic School (see Cyrenaics §4 ), had recommended the constant renewal of pleasure through self-indulgence, Epicurus observes that this accumulation does not increase the total of pleasure beyond that achieved when all pain has gone, but only 'varies' it. Freedom from pain is itself already a supremely pleasant state. The pursuit of luxury, far from increasing pleasure, enlarges your desires and leaves you needlessly vulnerable to the whims of fortune.

In his physics, Epicurus had completed the division of per se existents into bodies and void by showing that other claimants to per se existence, such as properties and time, are in fact parasitic on bodies for their existence. Likewise in ethics, he now examines the non-hedonic values which others assert, such as virtue, and argues that they are in fact valued not for their own sake but as instrumental means to pleasure (Cicero, On Ends I 42-54).

It remains to fill out the prescription for the maximization of pleasure, that is, to sketch the ideal Epicurean life. This involves calculating the relative roles of bodily and mental pleasures, and of static and 'kinetic' pleasures. Bodily feeling is in a way focal, since mental pleasure and pain consist ultimately in satisfaction and dissatisfaction, respectively, about bodily feeling. For instance, the greatest mental pain, namely fear, is primarily the expectation of future bodily pain (which is the main ground, and a mistaken one, for the fear of death). But although mental feelings ultimately depend on bodily ones, and not vice versa, mental feelings are a more powerful ingredient in an overall good life. Someone in bodily pain - which may be unavoidable - can outweigh this by the mental act of reliving past pleasures and looking forward to future ones. It is this ability to range over past and future that gives mental feeling its greater power. But misused, especially when people fear everlasting torture after death, it can equally well become a greater evil than its bodily counterpart.

Static pleasure is the absence of pain. The bodily version of it is called 'painlessness' (aponia), the mental version 'tranquillity' (ataraxia, literally 'non-disturbance'). Tranquillity depends above all on an understanding of the universe, which will show that contrary to the beliefs of the ignorant it is unthreatening. (This is, strictly speaking, the sole justification for the study of physics.) Kinetic pleasure is the process of stimulation by which you either arrive at static pleasure (for example, drinking when thirsty) or 'vary' it (for example, drinking when not thirsty). There are mental as well as bodily kinetic pleasures, for example, (perhaps) the 'joy' of resolving a philosophical doubt or holding a fruitful discussion with friends. Although kinetic pleasures have no incremental value, Epicurus does apparently consider them an essential part of the good life. This is particularly because the mental pleasure which serves to outweigh present pain will inevitably consist in reliving past kinetic pleasures and anticipating future ones. So a successful Epicurean life cannot be monotonous, but must be textured by regular kinetic pleasures. In the letter written on his deathbed, Epicurus claimed that despite the intense bodily pains this was the happiest day of his life, because of all the past joys of philosophical discussion that he could relive.

At the same time, these kinetic pleasures must be carefully managed. Some desires are natural, others empty. The latter - for example, thirst for honours - should not be indulged, because their satisfaction will bring either no pleasure or a preponderance of pain over pleasure. Even of the natural ones, some are non-necessary. For instance, the desire for food is necessary, but the desire for luxurious food is not. In order to be maximally independent of fortune, it is important to stick primarily to the satisfaction of natural and necessary desires. But occasional indulgence in those kinetic pleasures which are natural but non-necessary has a part to play, so long as you do not become dependent on them. True to this principle, Epicurean communities lived on simple fare, and even trained themselves in asceticism, but held occasional banquets (see Epicurus §3 ).

11 Social values

Ancient ethics does not problematize altruism as such, but it does seek the moral foundations of two specific forms of altruism: justice, that is, respecting the interests of your fellow citizens; and friendship. Given that Epicurean hedonism is egoistic - that all your choices as an agent aim at your own pleasure - is it possible to put someone else's pleasure before your own?

Epicurus analyses justice not as an absolute value but as a contractual relation between fellow citizens, its precise character engendered by current social circumstances (Key Doctrines 31-7). Sometimes it proves mutually advantageous to abstain from forms of behaviour which harm others, in return for a like undertaking from them. So long as such a contract proves socially advantageous, it is correctly called 'justice'. It imposes no moral obligation as such, and the ground for respecting it is egoistic - that even if you commit an injustice with impunity the lingering fear of being found out will disrupt your tranquillity (Key Doctrines 17, 35).

With regard to his own philosophical community, Epicurus attached positive value to justice and to the specific laws which enforced it not because philosophers need any restraint from wrongdoing but because they need protection from the harm that others might inflict. 'Do not take part in politics' was an Epicurean injunction, political ambition being a misguided and self-defeating quest for personal security. But the school nevertheless upheld the need for legal and political institutions, and sought to work within their framework.

Where the political life fails to deliver personal security, friendship can succeed. The very foundation of the Epicurean philosophical community was friendship, of which the mutual dealings of Epicurus and his contemporaries were held up as an ideal model by their successors. Unlike justice, friendship is held to have intrinsic value - meaning not that it is valuable independently of pleasure, but that it is intrinsically pleasant, not merely instrumentally pleasant like justice. Moreover, the pleasure lies in altruistic acts of friendship, not merely in the benefits received by way of reciprocation.

Later Epicureans were pressed by their critics for a more precise reconciliation of friendship with egoism, and developed the position as follows (Cicero, On Ends I 66- 70). According to one group, it is indeed for our own pleasure that we form friendships, and it is as a means to this, not ultimately for our friends' sake, that we share their pleasure and place it on a par with our own. A second group veered away from egoism: although friendship starts out as described by the first group, the outcome is something irreducibly altruistic, whereby we come to desire our friends' pleasure purely for their sake. A third group sought to restore egoism: the second group is right, but with the addition that friendship is a symmetrical contract, analogous to justice: each friend is committed to loving the other for the other's own sake.

12 Free will

Epicurus was arguably the first to make free will a central philosophical issue (see Free will). He takes it that determinism must be false. It is incompatible with basic moral attitudes; the data of self-awareness falsify it; and it is a position which cannot even be argued coherently, since to enter an argument is to assume that both parties to the debate are responsible for their beliefs and could adopt others (On Nature XXV).

Determinism is arrived at by two routes, each of which must therefore contain a false supposition. The first route is via a logical law, that of bivalence: since all propositions, including those about the future, are either true or false, if it is already true that a given event will occur, what makes it already true can only be causes already operative in the world and sufficient to bring it about. Therefore, everything that occurs, occurs of necessity. Epicurus' response is (like Aristotle's in De interpretatione 9, as widely interpreted) to reject the law of bivalence as regards future-tensed propositions. Predictions whose accuracy depends on human decisions yet to be made are, at the time of utterance, neither true nor false.

The second route is more directly physical. Determinism is arrived at by two routes, each of which must therefore contain a false supposition. The first route is via a logical law, that of bivalence: since all propositions, including those about the future, are either true or false, if it is already false that a given event will occur, it cannot occur, and if true, it cannot fail to occur. Therefore everything that occurs occurs of necessity. Epicurus' response is (like Aristotle's in De interpretatione 9, as widely interpreted) to reject the law of bivalence as regards future-tensed propositions. Predictions whose accuracy depends on human decisions yet to be made are, at the time of utterance, neither true nor false.

The second route is physical. We - our souls as well as our bodies - consist entirely of atoms, which move according to mechanical laws. How then can anything be genuinely 'up to us'? Are we not automata, our actions the outcome of infinite atomic causal chains? This time the Epicurean response is that the laws of atomic motion are not after all entirely deterministic. Atoms' motion through weight and blows is mechanical and invariable, but there must be a third, indeterministic, aspect of their motion, the minimal 'swerve' (see §4).

But how do swerves help explain free will? The question is much debated by scholars. Epicurus may appear to be merely substituting an unpredictable mechanism for a predictable one, not putting us in charge. Perhaps, as with the denial of bivalence, he is merely attempting to remove an obstacle to free will, not to explain it. Whether the swerve enters more directly than this into his account of volition will depend partly on whether he thinks that mental events such as volition are reducible to atomic changes in the soul, in which swerving atoms could play a part. There is in fact evidence (On Nature XXV) that he regarded mental events as irreducibly different from the soul atoms underlying them, and even as having their own causal efficacy on the atoms. (For his rejection of atomic reductionism, see §5 above.) If so, it may be safest to conclude that volition is already by its own nature autonomous - not part of an antecedent causal chain at all - and that this is the primary explanation of free will. Unfortunately its autonomy would be rendered impotent if either the laws of logic or those of physics had already determined our future actions independently of it. Therefore both sets of laws must be so rewritten as to circumvent this danger, and to keep alternative possibilities genuinely open. In the case of physics, the indeterministic swerve just is the most economical realization of that requirement.

13 Death

That the soul must itself be composed of atoms has, in Epicurus' judgment, one very cheering implication. It can easily be shown to perish with the body. Lucretius (III 417-829) presents a whole battery of arguments for this, based not only on the evidence for the soul's dissolubility but also on other indications, such as the continuous parallelism between its development and decay and those of the body. From the finding that there is no conscious survival after death Epicurus concludes that 'death is nothing to us'. To fear your own future non-existence is as groundless as to regret the time when you had not yet been born, and involves the absurdity of imagining yourself being there to witness and lament your own non-existence. To regret that your pleasures will be cut short is to make the computational mistake of supposing that a finite lifespan is ipso facto less pleasant than an infinite one. In a brilliant diatribe against the fear of death, Lucretius (III 830-1094 ) interprets the myths of punishment in the afterlife as allegories for moral malaise in this life, and portrays much human unhappiness, manifested in the vain search for security through wealth and power, as subconsciously nourished by the fear of death.

14 Influence

Epicureanism enjoyed exceptionally widespread popularity, but unlike its great rival Stoicism it never entered the intellectual bloodstream of the ancient world. Its stances were dismissed by many as philistine, especially its official rejection of all cultural and intellectual activities not geared to the Epicurean good life. It was also increasingly viewed as atheistic, and its ascetic hedonism misrepresented as crude sensualism (hence the modern use of 'epicure'). The school nevertheless continued to flourish down to and well beyond the end of the Hellenistic age. The poets Virgil and Horace had Epicurean backgrounds, and other prominent Romans such as Cassius, the assassin of Julius Caesar, called themselves Epicureans. In the first three centuries of the Roman Empire many writers show some debt to Epicurean thought, including not only the novelist Petronius but even the Stoic Seneca and the Platonist Porphyry. When Marcus Aurelius (§1) , Roman emperor ad 161-80, established four official chairs of philosophy at Athens, a chair of Epicureanism was among them. In later antiquity Epicureanism's influence declined, although it continued to provide a target for thinkers, both Christian and pagan, in search of a godless philosophy to attack. Serious interest in it was revived by Renaissance humanists, and its atomism was an important influence on early modern physics, especially through Gassendi.