- Epicureanism Stephen Everson It is tempting to portray Epicureanism as the most straightforward, perhaps even simplistic, of the major dogmatic philosophical schools of the Hellenistic age. Starting from an atomic physics, according to which ‘the totality of things is bodies and void’ (Hdt 39 (LS 5A)),<sup>1</sup> Epicurus proposes a resolutely empiricist epistemology, secured on the claim that every appearance (and not merely every perception) is true, maintains a materialist psychology and espouses hedonism in ethics. Indeed, it is perhaps not too far-fetched to see in Epicurus’ work an attempt to return to the natural philosophy of the pre-Socratics, and especially that of his atomist predecessor Democritus. However, even if there is some truth in this, the natural philosophy we find in him is much more sophisticated than any produced before the work of Plato and Aristotle. Epicurus certainly eschews dialectic and rejects the central role given to definition in the acquisition of epistêmê, understanding, but he nevertheless builds on the sophisticated empiricism we find in Aristotle. Again, whilst he returns to an earlier tradition of natural philosophy in denying the place accorded to teleological explanation by Plato and Aristotle, unlike his predecessors he is duly aware of the need to meet the challenge posed by those who deny that natural change and the development of natural substances can be properly explained without the use of such explanation. Moreover, whilst Epicurus is at pains to reject natural teleology, he seems not to renounce formal as well as final causes: we find no attack on Aristotle’s contention that one must distinguish a substance from its material constitution. Most importantly, perhaps, Epicurus is concerned to provide the kind of systematic ethical theory which was simply unknown before the Republic and the ethical writings of Aristotle. The temptation to render Epicurus more simple than he actually is is perhaps made more intense by the fact that his philosophical ambitions are congenial to a scientifically minded contemporary taste. Not least, of course, Epicurus seeks to explain all natural phenomena as the result of the motion of atoms through space. Furthermore, his system is a firmly naturalistic one. What he attempts is precisely to explain the behaviour of material substances (including those material substances which are human beings) in a way which is consistent with his atomistic materialism. Abstract objects, such as Platonic Forms or the objects of Aristotelian nous play no role in his system. His theories are moreover radically constrained by the available perceptual evidence and he does not seek to crown his enquiry by the acquisition of the sort of necessary universal knowledge required by Plato and Aristotle. In contrast to Plato, who mistrusted the evidence of the senses, and Aristotle, who, whilst renouncing this Platonic mistrust nevertheless denied the ability of the senses to provide genuine knowledge, Epicurus places perception (together with the related capacity for prolêpsis) right at the centre of his scientific method and is very cautious about forming beliefs that go beyond what is given in perception. In his espousal of materialism and empiricism, Epicurus seems a very modern ancient philosopher, someone who rejects precisely those parts of Platonism and Aristotelianism which can make them appear alien to the contemporary reader. Materialism and empiricism can take many forms, however, and, as we shall see, we must be careful not to assimilate Epicurus too quickly to their popular contemporary versions. LIFE AND WORKS OF EPICURUS Although an Athenian citizen, Epicurus was born on the island of Samos in 341 BC, where his father had gone from Athens as a settler ten years before. It is possible that he was first introduced to philosophical enquiry by Pamphilus, a Platonist who also lived on the island, and possible too that, when still young, he came under the influence of Nausiphanes, a follower of Democritus. Certainly he acquired a knowledge of early atomism from Nausiphanes, although this may have been after he visited Athens. When Epicurus was 18, he went to Athens to do his military service, and it is reported that he went to the Academy to hear Xenocrates lecture. After his two-year stint in the army, Epicurus joined his father in Colophon, where the latter had gone after the Athenian settlers had been ordered out of Samos by Perdiccas. Little detail is known of the next fifteen years of his life. He probably worked as a school teacher in Colophon, before moving to Mytilene in 311 to teach philosophy. A lost polemical work, Against the Philosophers at Mytilene, suggests that he did not fit in to the philosophical scene there very happily, and he seems to have left quite soon for Lampsacus, forming there a philosophical circle around himself. In 307 he returned to Athens, and, in order to set up a philosophical school, he bought a house which came to be known as the ‘Garden’. Epicurus lived there until his death in 271 BC. Although there is strong secondary evidence that Epicurus was a keen and vitriolic literary polemicist, there is also compelling evidence that he was a highly goodnatured man in person, and inspired great loyalty amongst his students. He wrote a great deal. Forty-one works are cited in Diogenes Laertius’ biography of him (in book X of his Lives of the Philosophers), but this list is of Epicurus’ ‘best works’: according to Diogenes, his complete works ran to around three hundred rolls, so that he surpassed all previous writers in the number of his books (X.26). Unfortunately, very little of this has come down to us. Diogenes reproduces three philosophical letters written to Epicurus’ followers: the Letter to Herodotus (Hdt.), in which he provides an epitome of his natural science; the Letter to Pythocles (Pyth.), an outline of his theories about celestial phenomena; and the Letter to Menoeceus (Men.), which gives the basics of his ethics. Diogenes also cites forty ‘Principal Beliefs’ (Kuriai Doxai, KD), and a further collection of maxims survives in a manuscript in the Vatican (Vaticanae sententiae, VS). In addition, some of the papyri found at Herculaneum have contained fragments of perhaps his principal work, the De Natura, which, according to Diogenes, ran to thirty-seven books. In addition to these few works which have survived, we have various other sources for Epicurean doctrine. Most important is the De Rerum Natura of the first-century BC Roman poet Lucretius, in which he sets out Epicurean teaching in helpful detail. The papyri discovered at Herculaneum have also contained works by Philodemus, another Epicurean of the first-century BC.More bizarrely, many fragments have been found in central Turkey of a wall erected by one Diogenes of Oenoanda to set out the principles of Epicureanism. In addition to these Epicurean sources, there is quite a lot of evidence for Epicurean philosophy in the work of Cicero and Plutarch, two opponents of Epicureanism. Given the state of our evidence, we do not have much reason to find developments either in Epicurus’ own work, or even in that of his followers. Thus, later writers are generally taken to provide pretty straightforward evidence for Epicurean claims and arguments. I shall follow that practice here, but it is worth noting at least the possibility that later Epicureans may manifest doctrinal shifts from Epicurus’ own claims. One reason perhaps why Epicurus’ philosophical system has seemed more simple than those of Plato and Aristotle and of his Hellenistic opponents is that the most accessible of his own writings to have survived are the letters, which are precisely intended to present introductory outlines of his views and arguments. It is clear from the works which survive on papyrus, both those of Epicurus himself and of Philodemus, that there was a proper place for detailed and technical argumentation within Epicureanism, but this has so far played a minor role in forming our general sense of the nature of Epicurus’ work, not least perhaps because deciphering the remains of the papyri is a tremendously difficult and uncertain process. PERCEPTION AND COGNITION Throughout his work, Epicurus shows himself to be epistemically cautious. He is keenly aware of the danger of holding false beliefs and is duly anxious to provide a method which, if followed, will allow one to believe only truths. Central to this method is a reliance on perception—which, together with ‘prolêpseis’ and the ‘primary affections’ of pleasure and pain, provide the ‘criteria of truth’ (DL X.31). Epicurus provides an reasonably elaborate account of what happens in perception, according to which we perceive when we are struck by the films of atoms (eidôla) which are constantly emitted from the solid bodies around us. These preserve the shape of the objects from which they emanate (Hdt. 46 (LS 15A1)) and it is by coming into contact with these eidôla that we see and think of shapes (Hdt. 49 (LS 15A6)), since these delineations penetrate us ‘from objects, sharing their colour and shape, of a size to fit into our vision or thought, and travelling at high speed, with the result that their unity and continuity then results in the impression’ (Hdt. 49–50 (LS 15A8)). Hearing, too, involves the reception of atoms: it ‘results from a sort of wind travelling from the object which speaks, rings, bangs or produces an auditory perception in whatever way it may be’ (Hdt. 52 (LS 15A14)). Smell, too, ‘just like hearing, would never cause any affection if there were not certain particles travelling away from an object and with the right dimensions to stimulate this sense, some kinds being disharmonious and unwelcome, others harmonious and welcome’ (Hdt. 53 (LS 15A18)). Thus, we are able to perceive because we are receptive to the various kinds of atoms emitted by the solid objects around us. Indeed, perception just is the conscious reception of these atoms, its content entirely determined by their nature and properties: ‘all perception, says Epicurus, is irrational and does not accommodate memory. For neither is it moved by itself, nor when moved by something else is it able to add or subtract anything’ (DL X.31 (LS 16B1–2)). The content of a perception is thus not to be explained by reference to anything other than what produces that perception, although the objects of perception are distinct from the direct causes of the perception. In Hdt. 46, we are told that the eidôla have the same shape as the solid objects from which they emanate, but are much finer than the things which are apparent—what are apparent in perception, then, are not the eidôla themselves but the solid objects (cf. Lucretius IV. 256 ff.). (It is for this reason, of course, that Epicurus has to argue for his account of how we perceive: if we perceived the eidôla which cause the perceptions, then the truth of that account would be given in perception and not stand in need of argument.) Since the eidôla in fact preserve the relevant properties of the objects from which they are emitted, the perceptual affection reports correctly the nature of the solid object: And whatever impression we get by focusing our thought or senses, whether of shape or of properties, that is the shape of the solid body, produced through the eidôlon’s concentrated succession or after- effect. But falsehood and error are always located in the belief which we add. (Hdt. 50 (LS 15A9–10)) It is the passivity of perception—its inability to add or to subtract anything from the stimulus—which secures its utter epistemic reliability, and it is not until the mind begins to work with the perceptual reports that the possibility of error arises. Whilst the content of perception is entirely determined by the nature of the stimulus which produces it (and so by how things are), the content of belief is not so constrained and our beliefs can thus mis-report how things are. The claim that all perceptions are true is, of course, an extremely strong one. The occurrence of perceptual conflict was, after all, something which had been the subject of epistemological scrutiny since Protagoras’ move to global subjectivism in order, if we are to believe Plato’s Theaetetus, to preserve the reliability of perception despite the occurrence of prima-facie conflicting appearances. Thus, if, for instance, the same wind seems cold to one person and warm to another, the mistake, according to Protagoras, would be to think of these perceptions as both seeking to represent the same state of affairs—in this case, the temperature of the wind—when, of course, they could not both be true. Rather, each correctly reports a distinct state of affairs; the wind’s temperature relative to the individual perceiver. The wind is in fact warm for the one perceiver and cold for the other. Perceptions report truly how things are for the perceiver (and, importantly, not merely how they seem to the perceiver). Protagoras’ wholesale subjectivism—the account is intended to apply not just to temperatures and colours but to all properties universally—was an extreme, and not obviously coherent, reaction to the possibility of perceptual conflict, and it did not find favour with either Plato or Aristotle, who had to find other ways to deal with the problem. Plato did so by denying that perceptions can be true or false at all: he treats them as mere sensations which provide the materials for beliefs. Aristotle, who did allow perceptions themselves to have propositional content, and so to be capable of being true and false, avoided the difficulties of perceptual conflict by denying that all perceptions are true: thus, in a case of perceptual conflict, at least one of the conflicting perceptions will be false and will be the result of a defect on the part of the perceiver. Against this background, Epicurus’ re-affirmation of the truth of all perceptions, without a move to any kind of subjectivism, can be seen to be very bold indeed—so bold, indeed, as to seem like hopeless epistemic optimism. Moreover, the argument which Diogenes Laertius cites as supporting this claim seems to be clearly insufficient to do this: All perception, he says, is irrational and does not accommodate memory. For neither is it moved by itself, nor when moved by something else is it able to add or subtract anything. Nor does there exist anything which can refute perceptions: neither can like sense refute like, because of their equal validity; nor unlike since they are not discriminatory of the same things; nor can reason, since all reason depends on the senses; nor can one individual perception, since they all command our attention. (DL X.31–2 (LS 16B1–7)) Thus, there is nothing which can convict any particular perception of error, since in any case where some state seems to cast doubt on the truth of a perception, that state can itself have no greater epistemic security than the perception which it calls into question. Even if this were right, however, it would not give Epicurus the conclusion he needs, since it would be consistent with this that there are indeed false perceptions, even though we can never have sufficient reason to believe of any particular perception that it is false.<sup>2</sup> Since there is nothing here to block the possibility of conflicting perception, the most sensible response when that possibility is realised would seem to be that of the sceptic; suspension of judgement. Indeed, in a case of two conflicting perceptions of the same sense, it would seem to be impossible to assent to both, since this would be to believe a contradiction. We are given a different argument by Sextus: For just as the primary affections, that is pleasure and pain, come about from certain agents and in accordance with those agents— pleasure from pleasant things and pain from painful things, and it is impossible for what is productive of pleasure not to be pleasant and what is productive of pain not to be painful but that which produces pleasure must necessarily be naturally pleasant and that which produces pain naturally painful—so also with perceptions which are affections of ours, that which produces each of them is always perceived entirely and, as perceived, cannot bring about the perception unless it is in truth such as it appears. (Math. VII.203) Here it is the passivity of the senses which secures their veridicality: they are such as to present the cause of the perception just as it is. This provides a much better route to Epicurus’ conclusion—and, indeed, it accords with the very start of Diogenes’ report of what Epicurus has to say about perception (and also with the remark at Cicero De Finibus 1.64 that one will not be able to defend the judgement of the senses without knowing the nature of things). It also provides a rather different kind of epistemological strategy from what we would have if we took the burden of the argument for the claim that all perceptions are true to be carried by the argument for the irrefutability of perceptions. The latter might look very much like an a priori epistemological argument, but the argument from the passivity of the senses is part of a theory about the way in which we are related perceptually to the world—and not itself given in experience. Given this, it is better to take the irrefutability argument as a subsidiary argument, an attempt to show that the universal conclusion—that all perceptions are true —is consistent with the available perceptual evidence, and thus does itself not fall foul of Epicurus’ scientific methodology. Eidôla are theoretical entities which we have reason to believe in because they explain things which are apparent. It is not part of the content of perception that we perceive because we are struck by atoms from the solid objects around us. His argument for the role of eidôla in perception, then, is an instance of his general method for establishing the truth of claims which are not directly supported by the evidence of perception itself. Unfortunately, our evidence for the details of this method is sketchy. In his brief report, Diogenes simply uses Epicurus’ technical terms without explicating their meaning: a belief will be true if it is ‘attested or uncontested’ and false if it is ‘unattested or contested’ (DL X.34 (LS 18B)). There is a much fuller account in Sextus, although doubt has been cast on its reliability. According to this, attestation is ‘apprehension through what is evident of the fact that the object of belief is such as it was believed to be’, and noncontestation ‘is the following from that which is evident of the non-evident thing posited and believed’. Contestation, alternatively, conflicts with noncontestation, being ‘the elimination of that which is evident by the positing of the non-evident thing’, whilst non-attestation ‘is opposed to attestation, being confrontation through what is evident of the fact that the object of belief is not such as it was believed to be’. Sextus concludes his report: ‘Hence attestation and non-contestation are the criterion of something’s being true, while non-attestation and contestation are the criterion of its being false. And the evident is the foundation and basis of everything’ (Math. VII.211–6 (LS 18A)). One obvious question here is why Epicurus needed two modes of assessment for beliefs rather than just one. The answer to this would seem to be that different types of belief will be assessed in different ways. Thus, if a non-perceptual belief (i.e. one which does not derive directly from perception) is nevertheless about something which can be perceived, then it should be assessed for whether it is attested or non-attested by perception. Sextus’ example of this is that of seeing someone from a distance: if I believe that the far-off figure is Plato, then that belief, which is not currently given by the perception itself, can be attested or non-attested by a later perception when the person is closer. In contrast, there are theoretical beliefs which can never be directly verified by reference to perception—such as Epicurus’ beliefs about eidôla—and it is these which are contested or non-contested by the perceptual evidence. Now, whilst it is reasonably straightforward to understand what it is for a belief to be either attested or contested by perceptual experience, matters are more difficult when it comes to construing the other two epistemic relations, those of non-attestation and non-contestation. The trouble is that whilst the words themselves, and perhaps also Epicurus’ own practice, suggest that nothing more or less than simple consistency with the perceptual evidence is sufficient for it to be either non-contested or nonattested (depending on what type of belief it is), Sextus’ account clearly places much stricter constraints on what the relations between beliefs and perceptions can be if the former are to be non-contested or non-attested by the latter. Furthermore, there would seem to be very good reason for this more restrictive view, since to accept a theoretical belief just on the grounds that it is consistent with the evidence seems extraordinarily lax, and to reject (rather than merely to hold in doubt) a perceptual belief just on the grounds that it is not attested by perception seems extraordinarily strict. What we should rather expect are three categories in each case: beliefs which our perceptions provide reason to accept, beliefs which they provide reason to reject, and those for which there is not perceptual evidence either way. Diogenes indeed does report a category of beliefs as those which ‘await’—‘for example waiting and getting near the tower and learning how it appears from near by’—(DL X.34 (LS 18B)), and this suggests that Epicurus did allow that there could be beliefs which were neither contested nor un-contested, nor attested nor un-attested. There is much room for interpretative manoeuvre on this point, but no space to effect that manoeuvring here. What can be said is that consistency with the perceptual evidence must certainly be a minimal condition for a theoretical belief to stand as non-contested—and a condition Epicurus is indeed concerned to show is satisfied when arguing for such beliefs—but that it may be that more is required than this. Indeed, the more minimal the constraints on what it is for a belief to be non-contested, the less important will be that notion for Epicurean science. Thus, whilst Epicurus introduces the eidôla at Hdt. 46 (LS 15A1), merely by saying that it is not impossible that there are such delineations of atoms, he continues by claiming that eidôla provide the most effective way of producing perception (Hdt. 49 (LS 15A6–8)). The theory is thus secured by something stronger than mere consistency with the perceptual evidence—what recommends it is not just that it provides a possible explanation of the evidence, but that it provides the best explanation of it. Again, at Hdt. 55–6, when he argues that we should not think that there are atoms of all sizes, he supports this by saying that ‘the existence of every size is not useful with respect to the differences of qualities’, where this seems to mean that we do not need to posit all sizes of atoms in order to explain the various qualities of visible bodies. If the only constraint imposed on theoretical claims was that they should not conflict with the perceptual evidence, however, then there would be no good reason to restrict the range of properties one attributed to the atoms. In the light of this, and of both Sextus’ report and the presence of the category of ‘that which awaits’ in Diogenes Laertius, there is some reason to think that non-contestation of a tbeoretical belief requires that it should be needed for an adequate explanation of the perceptual evidence. Even if we resist this, however, we can accept that in practice the Epicurean justification of theoretical beliefs did not stop at showing the mere consistency of those beliefs with the evidence of perception, and that if this is indeed all that is required to demonstrate that they are non-contested, then Epicurus seems to have required more for theoretical justification than non-contestation. In any case, the irrefutability argument in Diogenes can now be seen not to be a piece of a priori epistemological reasoning, but rather an attempt to show that the theoretical claim that all perceptions are true meets the minimal condition for acceptability—that it does not conflict with the perceptual evidence. The best explanatory account of perception gives us reason to think that our perceptions are always true and this claim is consistent with our perceptual experience itself, despite the fact that one might think that there are perceptual conflicts. I have so far been contrasting beliefs with perceptual evidence, but this is slightly misleading, since Epicurus’ criteria of truth are not limited to perception itself. The proper contrast is, as we have seen, between belief and what is evident, and things can be evident to us not merely through perception, but also through prolêpsis (plural: prolêpseis). If we are to understand Epicurus’ epistemology—and hence his natural science—we need to have some sense of the cognitive role played by prolêpsis. Perception is an entirely passive process, and the nature of the perceptual affection is determined entirely by the nature of the stimulus which produces it. In bringing prolêpseis into his account of cognition, Epicurus is able to extend the range of information the subject is able to receive. According to Diogenes’ brief exposition of prolêpsis (DL X.33), it is, for instance, in virtue of having the prolêpsis of man or horse that one can think and talk about men or horses and such prolêpseis are both selfevident and acquired through perception. Thus, what prolêpseis someone has depends upon his previous perceptual experience, and so these will differ between subjects. Thus, someone can talk and think about, say, cows and horses because he has prolêpseis of cows and horses—and he will have these if he has had sufficient previous perceptual experience of cows and horses. This is not, however, to say that he has had previous perceptions whose content is about the condition of cows and horses. Thus, the eidôla flowing from a horse will preserve the shape and colour of the horse, and so, in receiving them, one will have the perception that an object of a certain shape has a certain colour. This is guaranteed by the mechanism of the reception of the eidôla, but is not yet sufficient for the subject to have an experience with the content that that horse is that colour. In order to have an experience with that content, the subject must have some concept of a horse and it is here that prolêpseis come into play. In acquiring a prolêpsis of something, the subject acquires a recognitional ability for things of that kind. This requires repeated perceptual exposure to such things, after which one will be able to recognise newly perceived examples as similar to the ones he has already seen. This does not require any articulated theorising about what it is to be that kind of thing: it is important for Epicurus’ cognitive theory that prolêpseis operate prior to the level of belief, since this is what secures their ability to stand as criteria of truth for beliefs. People will differ in what prolêpseis they possess—in virtue of differing in their perceptual histories—but the acquisition of a prolêpsis is just as non-rational as having a perception. If someone has sufficiently many perceptions caused by horse-eidôla, he will come to have the prolêpsis of a horse and so will be able to distinguish horses from other types of thing both when he perceives them directly and when he thinks about them. ATOMS AND VOID Our direct experience is of solid objects in the world around us. In virtue of perception proper we can know that these objects have certain properties —such as size, shape and colour—and in virtue of prolêpsis we can come to recognise what sorts of objects they are (although this will always be in terms of properties which are perceptually apparent). Theory is required, however, if we are to come to know how these solid objects are materially constituted and how this explains their behaviour. In this section, I shall provide a brief outline of Epicurus’ theory of matter, generally citing his claims and arguments rather than discussing them. This is not because those arguments are uninteresting—indeed Epicurus’ arguments for the nature of the atoms are some of his more sophisticated—but because to discuss them seriously would require more in the way of historical context (in particular, Aristotle’s arguments for the continuous nature of matter and against the existence of void) than is possible here. As it is, this section should be seen just as an exposition of Epicurus’ basic physical theory. Epicurus begins the exposition of his physical theory in the Letter to Herodotus by affirming the temporal infinity of the basic constituents of the universe. ‘Nothing’, he claims, ‘comes into being out of what is not—for in that case everything would come into being out of everything with no need for seeds’ (Hdt. 38 (LS 4A1)). Epicurus thus secures his claim on something we observe, which is that when things are generated, they are generated from the relevant kind of seed (cf. Lucretius I.169–73 (LS 4B4– 5)). Thus, in order to grow an oak tree, we need to start with an acorn— for it is only an acorn which has the potential to generate an oak tree. If things could be generated ex nihilo, however, then there would be no necessary determinate conditions for their generation, and so no need for seeds. (Of course, it might be objected that Epicurus moves too quickly here from the generation of things with which we are acquainted to the claim that nothing can be generated ex nihilo. After all, we have no perceptual evidence that atoms do not come into existence spontaneously, merely that composite bodies do not, and this might be a respect in which the non-evident is dissimilar from the evident.) In any case, Epicurus holds to the analogy between the observed and the unobserved in this respect, and takes the universal need for composite things to be generated from bodies which possess the potential to generate those things to confirm the general thesis that nothing can be generated ex nihilo. He also maintains that nothing can pass away into nothing: if it could, then everything would already have perished. Given these two claims—that things cannot be created from nothing and they cannot perish into nothing, we can accept that the basic constituents of the universe persist for ever, since they will not have come into existence and cannot go out of existence. Epicurus’ next move is to establish the nature of those basic constituents: Moreover, the totality of things is bodies and void. That bodies exist is universally witnessed by perception itself, in accordance with which it is necessary to judge by reason that which is non-evident, as I said before; and if place, which we call ‘void’, ‘room’, and ‘intangible substance’ did not exist, bodies would not have anywhere to be or to move through in the way they are observed to move. Beyond these nothing can even be thought of, either by imagination or by analogy with what is imagined as completely substantial things and not as the things we call accidents and properties of these. (Hdt. 39–40 (LS 5A)) According to Epicurus, there are two basic kinds of substance (existing thing): bodies and void. Now, this is not yet a statement of atomism, since Epicurus tells us that the fact that bodies exist is given in perception, and we do not perceive atoms. Thus, the bodies which we know through perception to exist are the solid bodies which, he will argue, are composed of atoms. For Epicurus’ present purpose, however, this is quite sufficient, since we do perceive that there are solid bodies and this is enough to show that there are bodies which are extended and tangible. In itself, however, this does not show that void exists—what does show that is not that we perceive material objects, but that we see that they move. What is distinctive of void is that it is not solid: it offers no resistance. If all space were occupied by things which were solid, then solid objects would not be able to occupy a different space from the one they occupy at any time, and so could not move. Since we know from perception that bodies do move, we can infer that there is space which offers no resistance to the impact of bodies, and thus that void exists. Having established the existence of both bodies and void, Epicurus moves to discuss the nature of bodies. All bodies are either compounds or the basic constituents of compounds, and the latter are incapable of either alteration or dissolution: ‘the primary entities, then, must be atomic kinds of bodies’ (Hdt. 40–1 (LS 8A)). Epicurus’ atoms are uncuttable: it is physically impossible to split them. This is because, unlike those bodies which have atoms as constituents, they contain no void, and it is the presence of void in a body which renders it vulnerable to alteration (cp. Lucretius I.528–39 (LS 8 B2)). In expounding the existence of bodies which could not be divided, Epicurus was returning to the physical theory of Democritus and Leucippus in opposition to Aristotle, who had argued that matter must be continuous, that is, infinitely divisible. Epicurean atomism was more radical than that of his predecessors, however, since not only did he maintain that there are bodies which are physically indivisible, but he argued that there are minima which, whilst extended, have no parts at all— that is, they cannot be divided even conceptually. Each atom is perpetually in motion and if it were able to travel through the void without interference from other atoms, it would be carried downwards by its weight, and all atoms would travel at the same speed.<sup>3</sup> However, the trajectory of atoms, although not their velocity, can be affected by collisions with other atoms, so that one can have, for instance a system of atoms constituting some solid object. Although each atom will indeed be constantly moving, the trajectories of the constituent atoms will be such that the object which is constituted by them remains stationary. The basics of Epicurus’ atomic theory, then, are that matter is not continuous, but atomic, and that the physical atoms—the bodies which cannot be further divided physically—are constituted by minimal parts which are not even conceptually divisible. Every body—that is, every entity which is extended and solid—is either an atom or constituted by atoms. There are infinitely many atoms, and an infinite space for them to occupy and move about in. Each atom, because of its weight, has a natural tendency to move downwards (at a speed ‘as quick as thought’), but the locomotive history of many atoms is limited by the fact that they collide with other atoms. A collection of atoms can constitute a stable solid object when the atoms mutually deflect each other’s motion so as to maintain each other in a pattern. Even then, the atoms will not be at rest but will oscillate at their natural speed. SOUL, BODY AND PROPERTIES Atoms and void are the primary entities of Epicurean physics, and bodies and void are the only things which exist ‘per se’. Although Epicurus does not want to deny, for instance, that there are properties, he takes these to be parasitic on the existence of per se existents. Indeed, he is careful on this point: Now as for the shapes, colours, sizes, weights, and other things predicated of body as permanent attributes—belonging either to all bodies or those which are visible, and knowable in themselves through perception—we must not hold that they are per se substances: that is inconceivable. Nor, at all, that they are nonexistent. Nor that they are some distinct incorporeal things accruing to the body. Nor that they are parts of it; but that the whole body cannot have its own permanent nature consisting entirely of the sum total of them, in an amalgamation like that when a larger aggregate is composed directly of particles, either primary ones or magnitudes smaller than such-and-such a whole, but that it is only in the way I am describing that it has its own permanent nature consisting of the sum total of them. And these things have their own individual ways of being focused on and distinguished, yet with the whole complex accompanying them and at no point separated from them, but with the body receiving its predication according to the complex conception. (Hdt. 68–9 (LS7B1–2)) There are some properties which all bodies must have (shape, size, weight) and some which all visible bodies must have (colour). These are ‘permanent’ attributes of bodies: as Lucretius reports, they are those properties which ‘can at no point be separated and removed without fatal destruction resulting—as weight is to stones, heat to fire, liquidity to water, tangibility to all bodies, and intangibility to void’ (I.451–4 (LS 7A3)). Such properties are thus not merely permanent, but necessary. This necessity is not merely physical, but conceptual: one cannot conceive of the body without that property. They are not, however, per se substances like bodies themselves—they exist only as the properties of bodies, and so their existence is in that way derivative. In addition to these permanent properties, bodies can also have accidental properties: ‘by contrast slavery, poverty, wealth, freedom, war, peace, and all other things whose arrival and departure a thing’s nature survives intact, these it is our practice to call, quite properly, accidents’ (I.455–8 (LS 7A4)). Epicurus’ distinction here between visible and invisible bodies makes it clear that he does not think that atoms possess all the properties possessed by complex bodies. The only properties which atoms possess are those of shape, weight, size and the necessary concomitants of shape (Hdt. 54 (LS 12D1)), and so one cannot in general explain the fact that a complex body has some property by appealing to the possession of that very property by its constituent atoms. This is again made clear by Lucretius, who says that ‘you should not suppose those white objects which you see before your eyes as white to consist of white primary particles or those which are black to be the product of black seeds’ (II.731–33 (LS 12E1)). This, he points out, actually allows for a more satisfying explanation of the behaviour of coloured objects: Besides, if primary particles are colourless, and possess a variety of shapes from which they generate every kind of thing and thus make colours vary—since it makes a great difference with what things and in what sort of position the individual seeds are combined and what motions they impart to each other and receive from each other— it at once becomes very easy to explain why things which a little earlier were black in colour can suddenly take on the whiteness of marble, as the sea when its surface has been churned up by great winds, is turned into waves whose whiteness is like that of gleaming marble. All you need to say is that what we regularly see as black comes to appear gleaming white as soon as its matter is mixed up, as soon as the ordering of its primary particles is changed, as soon as some particles are added and some subtracted. But if the sea’s surface consisted of blue seeds, there is no way in which they could turn white. For things that are blue could never change to the colour of marble, no matter how you were to jumble them up. (Lucretius De rerum natura, II.499–514 (LS 12A3)) Given that Lucretius allows here that atomic change to a complex body can involve not merely the re-arrangement of atoms but also their loss and addition, the argument here doesn’t quite work—since one could accept that individual atoms cannot change, but maintain that when the sea changes colour it is indeed because there are blue atoms on the surface which are displaced by white atoms. Nevertheless, the passage is important because it suggests strongly that Epicurus accepts that the properties of a complex substance (a substance which has atoms as constituents) are determined by the properties—including the arrangement and motion—of its constituent atoms. For Lucretius infers the claim that the different atomic shapes and arrangements make the colours of a substance change from the general claim that the primary particles ‘possess a variety of shapes from which they generate every kind of thing’. That Lucretius feels entitled to infer from this that they are responsible for colours, and changes in colour, shows that what are ‘generated’ by the atoms are not just objects, but their properties as well—that is, that there is a particular arrangement of atoms of particular shapes that will determine not just that there is a certain kind of complex substance, but that that substance has the properties it does. The Epicurean treatment of the relation between macroscopic and microscopic properties can perhaps be best illustrated by considering his account of the psuchê—the ‘soul’—as this represents his most sustained attempt to explain the nature and behaviour of complex substances by reference to the nature and arrangement of their constituent atoms. For Epicurus, an animal body, like all solid objects, is a compound of atoms, and the psuchê is itself a material part of a living body: it is a ‘finestructured body’ diffused throughout the whole (Hdt. 63 (LS 14A1)). Thus, the psuchê is itself a body—that is, it is an individuated entity with its own distinctive atomic constitution. According to a report in Aetius, Epicurus took the material constitution of the psuchê to be specific to it: ‘it is a blend (krama) consisting of four things, of which one kind is fire-like, one air-like, one wind-like, while the fourth is something which lacks a name’ (Aetius 4.3.11 (LS 14C)). Although it is thus possible to specify the atomic constituents of the matter of the psuchê, what is important for the explanation of psychic functioning is that they form a ‘blend’. This is emphasised by Lucretius: The primary particles of the elements so interpenetrate each other in their motions that no one element can be distinguished and no capacity spatially separated, but they exist as multiple powers of a single body…. Heat, air and the unseen force of wind when mixed form a single nature, along with that mobile power which transmits the beginning of motion from itself to them, the origin of sensebearing motions through the flesh. (III.262–5; 269–72 (LS 14D1)) Because the elemental atoms are blended, they constitute a body which has particular powers lacked by things which are not so constituted—when contained within a larger body, it is, for instance, capable of sensation and thought. This last qualification is important for Epicurus, who enthusiastically denies that the psuchê can survive the death of the body, and emphasises the mutual dependency of psuche and the body which contains it. So, whilst it is indeed the psuchê which is responsible for perception, it is only able to produce that capacity in virtue of being contained within the body. Once the body disintegrates, the atoms of the psuchê are dispersed and so it loses its own capacities (Hdt. 63–4 (LS 14A3)). Neither the psuchê nor the body can survive the demise of the other, and it is the combination of body and psuchê which constitutes the living animal, not the psuchê by itself: ‘since conjunction is necessary to their existence, so also theirs must be a joint nature’ (Lucretius, III.347–8). Thus, it is not just the psuchê, but the whole body, which enjoys perception, which is, as Lucretius says, an affection which is common to the mind and the body (III.335–6). From the mere fact that Epicurus takes the psuchê to be itself a material body, one can tell very little about what relation he postulates between the psychological properties of the living animal and the movement of the atoms which constitute the psuchê. However, this becomes clearer if one reflects on his arguments for this materialist thesis.<sup>4</sup> The psuche must be material since, if it were not material it would be void, and ‘void can neither act nor be acted upon, but merely provides bodies with motion through itself’ (Hdt. 67 (LS 14A7)). Since it is evident that the psuchê does act on things and is, in turn, acted on, the idea that it is incorporeal is incoherent. That is, it is evident that there are psychological causes and effects, and if this is so, then what is changed and produces change must be something material, since immaterial things cannot be the agents or patients of change. The claim that only material things can bring about or undergo change will be well-motivated if it is assumed that all changes are either themselves atomic events, or are determined by those. That Epicurus accepts that psychological events require the occurrence of atomic events is clear from his arguments for the nature of the psuchê’s atomic constitution. So, according to Lucretius, the mind is ‘exceedingly delicate and is constituted by exceedingly minute particles’ (III.179–80): Nothing is seen to be done so swiftly as the mind determines it to be done and initiates; therefore the mind rouses itself more quickly than any of the things whose nature is seen plain before our eyes. But that which is so readily moved must consist of seeds exceedingly rounded and exceedingly minute, that they may be moved when touched by a small moving power. (III.182–8) In accordance with Epicurean scientific method, Lucretius starts off from something evident—that the mind produces its effects more rapidly than anything else does—and infers from this that the atoms of the psuchê are smaller and rounder than any other atoms. For this inference to work, however, psychological changes must require atomic changes, otherwise there would be no necessity that the atoms should be able to move as rapidly as the mind works. Again, when Lucretius comes to explain the occurrence of emotions, he does so by reference to the atoms which constitute the mind. When one is angry this is because of the heat in the psuchê and when one is frightened, this is the result of its coldness, ‘the companion of fear, which excites fright in the limbs and rouses the frame’ (III.288–93). Here there is a material explanation for the effects of the emotion. When one is afraid and one’s limbs shake, this can be explained by reference to the cold, the ‘companion’ of fear. For the emotion to have the effects it does—and that psychological states have causes and effects is the datum from which Epicurean theorising about the psuchê begins—there must be atomic events which determine those effects. In trying to understand Epicurus’ natural science, it is tempting to think that he must be a reductionist just because he espouses atomism—which can strike the contemporary reader as somehow an intrinsically ‘scientific’ theory of matter. This temptation should be resisted, however. There is no sign that Epicurus attempted to identify, say, the mental properties or events of people with their atomic properties or events. The cold is, after all, only the companion of fear and not the emotion itself. In this respect, Epicurus is perhaps more Aristotelian than he is sometimes given credit for being—for Aristotle too accepted a genuine role for material explanation within his natural science and psychology. Aristotle distinguished efficient causation from material causation: changes involving material substances are to be explained both by reference to the capacities of the agent and patient of the change and to the underlying material events on which the changes supervene. That Epicurus maintains an atomic theory of matter rather than one according to which matter is continuous, and that he renounces a teleological explanation of natural phenomena, puts no pressure on him to give up the distinction between efficient and material causes—and there is good reason to think that he does not give it up (even if he does not continue with the terminology). For if it is the properties of the psuchê which have causes and effects, and if Epicurus does not identify those properties with the arrangements of atoms which generate those properties but still thinks that the operations of the psuchê can be explained by reference to the arrangements of its constituent atoms, then, like Aristotle, he must distinguish antecedent causes from material causes and allow both a role in the determination of changes. ACTION AND RESPONSIBILITY Lucretius gives the following account of the causation of action: Now I shall tell you…how it comes about that we can take steps forward when we want to, how we have the power to move our limbs, and what it is that habitually thrusts forward this great bulk that is our body. First, let me say, images (simulacra) of walking impinge on our mind and strike it, as I explained earlier.<sup>5</sup> It is after this that volition occurs. For no one ever embarks upon any action before the mind first previews what it wishes to do, and for whatever it is that it previews there exists an image of that thing. So when the mind stirs itself to want to go forwards, it immediately strikes all the power of the spirit distributed all over the body throughout all the limbs and frame: it is easily done because the spirit is firmly interlinked with it. Then the spirit in turn strikes the body, and thus gradually the whole bulk is pushed forward and moved. (IV.877–91) Here we have, as we should now expect, an account of what happens when we act which makes use of a mixture of both psychological and material causation. In order to walk, for instance, the person needs to form the intention to walk, and so needs to think about walking. For this to happen, he must have an image, or images, of walking and these come in the form of eidôla from outside.<sup>6</sup> These images have both a psychological and a material aspect: they are constituted by atoms whose impact on the mind will have mechanical effects, but they are pictorial in that they present an image to the mind. As the mind decides to walk, it transmits an impulse to the spirit which in its turn strikes the relevant parts of the body so that they move. We thus have a story which can be told at two levels. Mechanically, the atoms of the image strike those of the mind which impact those of the spirit which impact those of the body, whilst, psychologically, the mind responds to the image of walking by deciding to walk, thus causing the person to walk. The person walks because he decides to (and, in order to decide to, he must think about walking). This causal explanation is taken to be consistent with the determination of these psychological events by the material events which underlie them. Allowing that psychological events are determined by the movements of the atoms which constitute the person’s psuchê and body, however, was taken by some to raise the threat of a determinism inconsistent with moral responsibility. Epicurus deals with this threat in the remnants of Book 25 of his De Natura. There he distinguishes between a person’s atomic constitution and what he calls ‘developments’—and it is in virtue of the latter that we are responsible for actions (XXXIV.21–2 Ar2 (LS 20B)). The passage is notoriously obscure, but it is most happily read as providing a response to someone who seeks to excuse bad behaviour as the result of material causation, that is, as brought about by the motions of one’s constituent atoms. Epicurus is thus not concerned with the sort of determinist argument against moral responsibility which has become more familiar—that if our actions are caused by our mental states, which are themselves caused, then we are cannot be held responsible for how we act. Epicurus, like Aristotle, does not think that the fact that our actions are the effects of our practical deliberations provides any reason at all to deny that we are responsible for them. His response to his opponent here is to point out that our actions are not, or not only, determined by the motions of our constituent atoms, but by the ‘developments’, which, presumably, are our psychological states. The determinist’s mistake is to seek to explain our actions only by reference to their material causes, and so to leave out of account the psychological states which are the antecedent causes of our actions. Some have seen in this an Epicurean rejection of physicalism—a denial that our psychological states are in fact determined by the motions of our constituent atoms.<sup>7</sup> This is not required by the text, however, and would, as we have seen, go against the position we find implied elsewhere. Moreover, when Epicurus does move to deny determinism as such, he does so by positing indeterminacy at the atomic level. So, Cicero reports that, in order to avoid ‘the necessity of fate’, Epicurus posits an atomic swerve, fearing that ‘if the atom’s motion was always the result of natural and necessary weight, we would have no freedom, since the mind would be moved in whatever way it was compelled by the motion of atoms’ (De Fato 22–3 (LS 20 E2–3)). Thus Epicurus, it seems, modified his atomic theory so that not only could atomic motion result from the atom’s own weight, and from the impacts of other atoms, but it could also occur spontaneously as a minimal deviation from its existing trajectory—a swerve. This was introduced in order to preserve the ascription of moral responsibility. Epicurus seems to have accepted that if all atomic events were determined by previous atomic events, and if psychological events were determined by atomic events, then we could not properly ascribe responsibility to people for their actions. In order to preserve the prolêpsis that we are so responsible, he modifies the atomic theory so as to introduce indeterminacy at certain points, so that the chains of causation do not stretch back infinitely. That he was driven to this, however, confirms rather than casts doubt on the thesis that atomic events determine psychological events, since, if this were not so, there would be no need to deny that all atomic events are determined. It is difficult to regard Epicurus’ doctrine of the swerve as a great success. For, even if it does introduce indeterminacy into his system, it would not seem to do so in the right way. For, whilst it will serve to deny that there are infinite chains of atomic causes, it does nothing in itself to make these relevant to the determination of mental events and of actions, and it is difficult to see how Epicurus thought the mere denial of infinite causal chains of atomic events could make a relevant and constant difference to the determination of actions.<sup>8</sup> It is hard here not to support Carneades’ judgement, reported by Cicero, that in fact Epicurus did not need his swerve, but, having accepted that ‘a certain voluntary motion of the mind was possible’, this in itself provided what was needed against those who would deny that we are responsible for our actions: ‘a defence of that doctrine was preferable to introducing the swerve, especially as they could not discover its cause’ (Cicero, De Fato, 23 (LS 20E4)). PLEASURE AND THE GOOD LIFE Epicurus thus sees no conflict between the thought that we are material substances whose behaviour can be explained by reference to the movements of our constituent atoms, and the fact that we are capable of intentional action and practical deliberation. Such deliberation, according to Epicurus, is always conducted by reference to pleasure: ‘we recognise pleasure as the good which is primary and congenital; from it we begin every choice and avoidance, and we come back to it, using the affection as the yardstick for judging every good thing’ (Men. 129). Whenever we act, we do so to gain some pleasure, and our actions will be successful in so far as they achieve this. Pleasure and pain, the ‘primary affections’; are, we remember, Epicurus’ third criterion of truth, along with perceptions and prolêpseis—they have the same kind of epistemic reliability as these other states.<sup>9</sup> If something seems pleasurable to someone, then it is pleasant, and if it seems painful, it is indeed painful. That pleasant things are to be pursued and painful things avoided is evident to anything which is capable of experiencing pleasure and pain: ‘as soon as it is born, every animal seeks after pleasure and rejoices in it as the greatest good, while it rejects pain as the greatest bad and, as far as possible, avoids it; and it does this when it is not yet corrupted, on the innocent and sound judgement of nature itself’ (Cicero De Finibus 1.30 (LS 21A2)). This ‘cradle argument’ should not be taken simply to express an unhappy prejudice in favour of untrained, infantile or animal tastes: the point is that the badness of pain, and the goodness of pleasure, are evident simply in their perception. The judgement is ‘nature’s’ because it is delivered by the causal interaction with the world around us: it is not something whose truth needs to be established by theorising. ‘[Epicurus] thinks these matters are sensed just like the heat of fire, the whiteness of snow and the sweetness of honey, none of which needs confirmation by elaborate arguments’ (I.29). This provides the foundation for Epicurus’ account of the good life, which he identifies with a life of pleasure (properly conceived). He maintains, that is, not just that pleasure is a good but that it is the highest good, the final end of action. This is stated clearly by Torquatus, the spokesman for the Epicurean school in Cicero’s De Finibus: We are investigating what is the final and ultimate good, which as all philosophers agree must be of such and such a kind that it is the end to which everything is the means, but is not in itself the means to anything. Epicurus situates this in pleasure, which he wants to be the greatest good, with pain the greatest bad. (I.29 (LS 21A1)) The terms here are Aristotelian, and it was indeed Aristotle’s discussion of happiness (eudaimonia) which set the terms for Hellenistic ethical discussions. According to Aristotle, happiness is formally the final end of action: it is something which cannot be chosen for the sake of anything else, whereas other things are chosen for its sake. Thus, ‘we call that which is never desirable for the sake of something else more final than the things which are desirable both in themselves and for the sake of that other thing, and therefore we call final without qualification that which is always desirable in itself and never for the sake of something else’ (Ethica Nicomachea 1.7, 1097a31–5). That happiness is final without qualification is a formal condition on any substantive account of happiness, and Aristotle’s own substantive account satisfies this by distinguishing between those things which are desirable both for themselves and for the sake of happiness—such as virtue, intellectual activity and pleasure—and happiness, which is only valuable for itself. One achieves happiness precisely by engaging in those activities and having those things which are intrinsically valuable and are thus the components of happiness. The difference between the Aristotelian and Epicurean conceptions of happiness will immediately be apparent. Aristotle, like Epicurus, takes pleasure to be something intrinsically valuable (although he would not accept the claim that our perceptions of what is pleasant are incorrigible), but whilst he places pleasure as a constituent of happiness, along with other goods, Epicurus moves actually to identify it with happiness. A further formal condition which Aristotle set down for any account of happiness was that a happy life should be ‘self-sufficient’—that is, it must be such that it lacks nothing of value (otherwise there would a better good which would consist of happiness together with whatever it lacks, and this further good would then be more final than happiness itself). The danger for Epicurus’ identification of happiness with pleasure is that it will fail to meet this condition, because it will leave out of account those things other than pleasure which are intrinsically valuable, thus allowing a life which included these as well to be better than a life of pleasure. To make good his identification of the final end with pleasure, Epicurus will need to show either that other things are not in fact intrinsic goods or that, even if some are, we can, and perhaps always do, also desire these things for the sake of pleasure. There is no reason in principle why Epicurus’ conception of happiness should be radically less complex than that offered by Aristotle. Whether it is will depend, in part, on how he understands the relation between pleasure and what affords it. So, if he were to think of pleasure as a feeling or sensation which is produced in one by doing things, then his account of happiness would certainly be more simple than Aristotle’s: a happy life would just be one in which the subject enjoyed a great deal of that feeling, and enjoying that feeling would be the only thing worth pursuing. Other things will only be instrumentally valuable—valuable just in so far as they give rise to this feeling. If, alternatively, he were to identify pleasure with pleasurable activity, or make the degree and quality of the pleasure dependent on the type of activity which produces it, then his idea of what happiness would be like need not be substantially different from Aristotle’s. The happy life could just be one which involved the enjoyment of valuable activities, where the activities are pleasurable precisely because they are themselves valuable. This would allow Epicurus, for instance, to treat virtuous activity as Aristotle does—something which is desirable in itself and, for that reason, something which can be chosen for the sake of happiness. These different conceptions of pleasure will result in accounts of happiness which differ in another respect as well. Treating pleasure as a feeling leads naturally to a subjective account of happiness, for if one takes pleasure to be a feeling which can be produced indifferently by various things, then those activities will be pleasurable for someone just if they happen to produce that feeling in him, and there is no reason to require that the same activities will be pleasant for everyone. Although it will be the case that for everyone to lead a happy life is to enjoy (a great deal of) the feeling, how one acts to achieve that can differ between people. If, alternatively, one takes pleasure to be dependent on the value of the activities and experiences which give rise to it, then one might be able to specify those activities and experiences which will be part of a happy for life for anyone. When we turn to consider what Epicurus has to say about pleasures, it would seem that he is no subjectivist: So when we say that pleasure is the end, we do not mean the pleasures of the dissipated and those that consist in having a good time, as some out of ignorance and disagreement or refusal to understand suppose we do, but freedom from pain in the body and from disturbance in the soul. For what produces the pleasant life is not continuous drinking and parties or pederasty or womanising or the enjoyment of fish and the other dishes of an expensive table, but sober reasoning which tracks down the causes of every choice and avoidance, and which banishes the beliefs that beset souls with the greatest confusion. (Men. 131–2 (LS 21A5)) Someone whose life was focused on what are sometimes called the pleasures ‘of the flesh’ would not, according to Epicurus, achieve happiness through these. His conceptions of happiness and of pleasure, then, are not sympathetic to the idea that it does not matter how one lives if one is to be happy, so long as what one does produces pleasure. This is not because he thinks, as some have, that such pleasures are not genuine pleasures or denies that they are good. Rather, as this passage suggests, they turn out to be the wrong kind of pleasure to be identified with the final end. They are, that is, kinetic pleasures, whereas the states he identifies with being happy, aponia and ataraxia, freedom from bodily and mental pains, are what he calls katastematic, or static, pleasures. In the De Finibus the distinction is illustrated by the difference between the pleasure one gets from quenching thirst and the pleasure of having had one’s thirst quenched (II.9). The first is an active pleasure—a pleasure of doing something or, perhaps, having something happen to one—whilst the second is static and results from the absence of pain or distress. It is not clear from our sources whether Epicurus thinks that whenever we have satisfied a desire there is a corresponding static pleasure; this would perhaps be a somewhat odd thing to think (how long would such a pleasure last?). Instead of taking the pleasure to be that, for instance, of having quenched one’s thirst (a different static pleasure from that of, say, having satisfied one’s hunger), one could rather take it to be the condition one is in when one has no unsatisfied desires and is not in pain or distress. If one were thirsty, then one would need to drink to achieve this, and if one were hungry, one would need to eat. The static pleasure which would result in the two cases would be the same (and its achievement would be contingent on the absence of other causes of bodily distress). The importance of the distinguishing between kinetic and katastematic pleasure is that it makes more plausible the identification of happiness, objectively conceived, with pleasure. According to Torquatus in the De Finibus, the greatest pleasure we experience is not any kind of gratification, but what is perceived once all pain has been removed: ‘For when we are freed from pain, we rejoice in the actual freedom and absence of all distress’ (De Finibus 1.37). Pleasure is the necessary consequence of the removal of pain, since there are no states which are neither painful nor pleasurable. This thesis is central to Epicurus’ hedonism. Whenever one is not suffering from pain or distress, one will be in a state of pleasure, and, further, this condition is not one which can be made more pleasurable: ‘Epicurus, moreover, supposes that complete absence of pain marks the limit of the greatest pleasure, so that thereafter pleasure can be varied and differentiated but not increased and expanded’ (De Finibus 1.38).<sup>10</sup> Thus, the combination of aponia, the absence of bodily pain, and ataraxia, the absence of mental distress, places one in a condition which one cannot rationally wish to improve. Once one has achieved these, life cannot get any better. At first sight this looks very odd. Epicurus accepts that every pleasure is something good (Men. 129), and this must include kinetic as well as static pleasures, but seems to deny that one’s life can be made better by pursuing more kinetic pleasures. Whilst pleasure is a good, it is not the case that more pleasures are better. That one should find this paradoxical is a sign, for Epicurus, that one has misunderstood the nature of pleasure, and so will not be able to organise one’s life to achieve what he takes everyone to aim at, i.e. the most pleasant life. For to pursue different kinetic pleasures as a means to achieving a more pleasurable life assumes that one can be in a state which is intermediate between pleasure and pain—and this, of course, is just what Epicurus denies. As long as one is not in pain or distress, then one is in a state of pleasure, and since there are not degrees— but merely varieties of—pleasure, one’s state cannot be improved by adding particular kinetic pleasures. Aponia and ataraxia thus together constitute happiness, which is the final end of action—that for which everything else is desired. Now, one could grant happiness this status without having to claim that whenever one acts one does so in order to achieve it. The claim could merely be that whilst all others goods can intelligibly be chosen for the sake of happiness, it cannot be chosen for the sake of anything else. Epicurus, however, seems to maintain the stronger thesis. Having identified happiness with aponia and ataraxia, he claims that achieving these is the goal of every action: ‘this is what we aim at in all our actions—to be free from pain and anxiety’ (Men. 127 (LS 21B1)). Again, this seems an absurdly strong thesis to hold. At least generally, one’s desires are for more specific things, such as eating or sleeping or listening to music or playing soccer—and even if one were to accept that in pursuing such things one was thereby aiming at achieving a good life, it is vastly implausible that their role in achieving this higher end is because they free one from pain. Epicurus, however, does not need to maintain that we always do, in fact, act in order to achieve aponia or ataraxia: his claim need be only that when we act, we always do so in order to get some pleasure or other. However, once we understand the nature of the static pleasures, we will see that these can, and should, provide the goal for our practical reasoning. Thus, he is clear that whilst all pleasures are good, not all are choiceworthy, so that whilst one always has some reason to choose something which will afford pleasure, there can be stronger reason not to choose it. ‘No pleasure is something bad per se: but what produces some pleasures produces stresses many times greater than the pleasures’ (KD 8 (LS 21D1)). Thus, the rational agent will resist some pleasures because satisfying the desire for them will lead to greater overall distress than leaving it unsatisfied. If such calculations are to be properly made, the choice must be referred to the goal of achieving aponia and ataraxia. To help with the successful pursuit of that goal, Epicurus classifies desires into three classes: ‘Some desires are natural and necessary, some natural but not necessary, whilst others are neither natural nor necessary but arise from empty belief’ (KD 29). This is explicated by a scholion which has survived in our manuscripts of Diogenes Laertius’ text, which reports that the first class of desires are for things which bring relief from pain, the second for things which will vary pleasure rather than remove pain, and the third are for such things as crowns and the erection of statues (DL X.149 (LS 21I)). In the Letter to Menoeceus, Epicurus himself expands on what it is for a desire to be necessary: ‘of the necessary, some are necessary for happiness, others for the body’s freedom from stress, and others for life itself’ (Men. 127 (LS 21B1). This classification of desires is not immediately obvious—in particular, it is not obvious how a desire can be natural without being necessary. If what it is for a desire to be natural is for it to be such that, given our nature, we cannot avoid having it, then how could such a desire not be necessary? Taking our cue from Epicurus’ own explication of necessary desires, we should think of necessary desires here as desires whose satisfaction is necessary for happiness, aponia or survival. These desires will not be as specific as, for instance, is the desire for some expensive food. What is necessary for survival is just the desire to eat. Nevertheless, the desire to eat an expensive food is clearly not unrelated to that necessary desire but is rather a specific instance or version of it.<sup>11</sup> Epicurus can thus intelligibly take it to inherit its naturalness from the more general desire, although its satisfaction is not necessary for the person’s survival. Indeed it is not important for its being non-necessary that it should be a desire for some expensive food: all particular types of food are such that a desire for them is not necessary. This is why Epicurus recommends that we stick to the more general desires, since the more general the desire, the less likely it is to go unsatisfied.<sup>12</sup> (Of course, desires for expensive things are, in the normal course of things, more likely to go unsatisfied than desires for things which are cheap and readily available.) All desires are either natural or empty, and empty desires according to KD 29 are empty because they are based on ‘empty’ belief. We know from Epicurus’ methodological discussions that empty beliefs are false beliefs which are not secured by reference to the criteria of truth. Thus, to reject a perception will be to confound perceptions with ‘empty belief’ and so actually to lose the criterion of perception altogether (KD 24 (LS 17B1)), and if one does not grasp the relevant prolêpsis, the words one uses will also be ‘empty’ (Hdt. 37 (LS 17C1)). Emptiness in one’s beliefs and language is the consequence of not securing them on the criteria of truth. As empty words are words which do not succeed in picking anything out, and empty beliefs are those which do not correspond to how things are, so empty desires will be those which are not for things which are genuinely pleasant. Thus, Epicurus can allow that people can have desires which arise from bad evaluative theories of the world—so that, for instance, they are persuaded, in whatever way, that crowns and public renown are pleasurable things to have—and they will not in fact gain pleasure from the satisfaction of these desires. Thus, although Epicurus accepts the primary affections of pleasure and pain as criteria of truth, this does not force him to accept that anything anyone believes to be pleasant is so, since such beliefs can be, and no doubt often are, unsecured by the appearances.<sup>13</sup> THE GOOD LIFE AND OTHERS For Epicurus, then, as for Aristotle, happiness is the central notion for practical reasoning. One worry for a theory of this kind is that it can seem to provide a necessarily selfish account of practical reasoning, since it looks as if all actions are ultimately to be judged by reference to whether they contribute to the agent’s own well-being. Aristotle escapes this—as do the Stoics after him, who identify happiness with virtue—because he takes the constituents of happiness to be desirable in themselves: they are constituents of happiness just because their value is autonomous. Thus in order for virtuous activity to contribute to the agent’s happiness, it must be chosen for its own sake, and not merely as something instrumental to his happiness. The virtuous person will indeed take pleasure in acting virtuously, but this pleasure comes from his awareness that he is acting well, that he is doing what he has reason to do anyway, and does not motivate his action. It is less clear that Epicurus, in identifying happiness with pleasure, even the static pleasures of ataraxia and aponia, can similarly escape the charge that he renders all practical reasoning ultimately selfish, concerned only with the good of the agent himself. The difference in principle between the two accounts can perhaps be helpfully illustrated by a non-ethical component of Aristotelian happiness: intellectual activity. Aristotle takes such activity to be the highest activity of which we are capable, and so the most valuable. Because of this, gaining a scientific understanding of the world is the most pleasurable activity and a component of the good life. For Epicurus, in contrast, understanding the world is not something autonomously valuable: if we were not alarmed by celestial phenomena and the prospect of death, there would be no need to study natural science (KD 11). Such study is necessary, since ‘one would not be able to banish fear about the most important things, if one did not know the nature of the whole universe’ (KD 12). People have a fear of the divine and of death, and they need to come to understand the nature of the gods and of the psuchê to see that neither of these fears is justified. If the gods exist at all, they lead a completely happy life, unconcerned with human lives and so of no threat to human happiness. Similarly, once one recognises that the pshchê perishes at the death of the person, one will see that one cannot be harmed after death and so death is ‘nothing to us’.<sup>14</sup> However, if we were not inclined to take cosmic events as signs of divine wrath or to think of death as a grave harm, we should have no need to understand the nature of the universe. The study of natural science is useful just because it dispels mental distress and so helps to achieve ataraxia. Its value is merely instrumental to the achievement of pleasure and thus happiness. Of course, this difference between Aristotle and Epicurus might have arisen just because Epicurus took a more philistine attitude to intellectual activity than did Aristotle, but it certainly exemplifies a general concern with his account of happiness. So, he says that ‘if you fail to refer each of your actions on every occasion to nature’s end, and stop short at something else in choosing or avoiding, your actions will not be consequential on your theories’ (KD 25 (LS 21E)). From this it looks very much as if Epicurus sets up as the over-arching principle of practical reasoning that whenever one acts one should do so in order to achieve pleasure for oneself. This was indeed the view of practical deliberation we find attributed to the Cyrenaics, who denied that happiness was the final end of action and who thought that one should act towards other people just so as to gain the most pleasure and least pain for oneself.<sup>15</sup> Epicurus, however, was no Cyrenaic and precisely seems to have wanted to allow that one can rationally be concerned with the good of others. The difficulty is seeing how this might be so, given his hedonism. Thus, Epicurus placed great store in the importance of friendship, saying that even though it will at least initially be motivated by utility, it is nevertheless something intrinsically valuable (Vatican Sayings 23). However, if one’s relationships with other people are motivated and controlled by a concern for one’s own pleasure, then whatever relationships they are, they won’t be much like friendships. It is clear from a passage in De Finibus I that the Epicureans were themselves worried by this, since Torquatus there reports different Epicurean accounts of the relationship between pleasure and friendship (without, unfortunately, ascribing any to Epicurus himself). Some, it seems, bit the bullet and allowed that ‘the pleasures which belong to friends are not as desirable per se as those we desire as our own’ (De Finibus 1.66 (LS 2201)). Even according to these Epicureans, however, we come to care about our friends as much as we do for ourselves, even though our concern is mediated by our own pleasures: Without friendship we are quite unable to secure a joy in life which is steady and lasting, nor can we preserve friendship itself unless we love friends as much as ourselves. Therefore friendship involves both this latter and the link with pleasure. For we rejoice in our friends’ joy as much as in our own and are equally pained by their distress. The wise man, therefore, will have just the same feelings towards his friend that he has for himself, and he will work as much for his friend’s pleasure as he would for his own. (De Finibus I.66–7 (LS 22 O)) This, however, seems to restate and preserve the problem rather than to resolve it, maintaining both that one has to care about one’s friend for his own sake and that one’s own pleasures are more desirable per se than those of one’s friend. Of course, the fact that one treats something as having value in itself does not commit one to thinking that it has as much value as other things one values: one could accept that one’s friend’s pleasure is per se desirable whilst denying that it is as per se desirable as one’s own pleasure. However, this would not provide a satisfactory reconciliation. One would hardly think someone a proper friend if he were willing to promote one’s interests just so long as they never conflicted with his own. In fact two different strategies are suggested in the text for reconciling hedonism and the demands of friendship. The first is to maintain both that a friendship is something which is in one’s own overall interests, but also that it cannot be conducted unless one does take one’s friend’s interests to have equal value to one’s own. Thus, as a matter of practical rationality, one would decide, in respect of one’s friend, to put into abeyance the general principle of referring every action to the criterion of one’s own pleasure, allowing the friend’s interests equal weight with one’s own. The second way would be to appeal to the psychological fact that one comes to be co-affected with the friend: one comes to rejoice in one’s friend’s joy as much as in one’s own and to be equally pained by his distress. Once this has happened, one can in fact appeal to the principle of pursuing one’s own pleasure in order to act in the interests of one’s friend, since, for instance, knowing that he is hungry or thirsty will disturb one’s own pleasure as much as if one were hungry or thirsty oneself. Whilst this is consistent with maintaining that one’s own pleasure is more desirable per se than that of other people, one can see why other Epicureans might have felt it unstable. These Epicureans, according to Torquatus, ‘though intelligent enough, are a little more timid in facing the criticisms from you Academics: they are afraid that if we regard friendship as desirable just for own pleasure, it will seem to be completely crippled’ (De Finibus 1.69 (LS 22 O)). It would be at least slightly odd to maintain that one does have as much reason to promote one’s friend’s interests as one’s own, but only because one will be pained as much as he will if one does not. Thus, the second Epicurean response is to allow that whilst one does first make contact with people and form relationships with them for the sake of one’s own pleasure, once ‘advancing familiarity has produced intimacy, affection blossoms to such an extent that friends come to be loved just for their own sake even if no advantage arises from the friendship’ (De Finibus 1.69 (LS 22 O)). This is a more interesting, if perhaps less subtle, position than the first, and accords with what Epicurus himself seems to have said in VS 23. It is more interesting, because it seems to allow the extension of the goodness of pleasure from oneself to other people. That is, in coming to love the friend for himself, his interests in themselves will provide one with reasons for action. One can still refer one’s actions to the criterion of whether they produce pleasure, but the range of relevant pleasures will have been extended to include those of one’s friend. It is not, of course, that one will not take pleasure in his wellbeing, but, in contrast to the first view, this is no longer the motivation for particular acts of friendship. There is nothing in Epicurus’ account of pleasure to provide an obstacle to such an extension: certainly we need first to experience our own pleasure in order to understand its nature, but, having grasped that, we can then understand what it is for someone else to gain pleasure and recognise this as a good. We do not, however, find in Epicurus any wholesale move in this direction as we do, say, in Mill: there is no attempt to argue that having come to recognise the goodness of pleasure, we should recognise that it is equally good whoever’s pleasure it is. Our ability to love the friend for himself comes about because we are close to him, and there is no argument to the effect that we should seek to extend this sort of concern beyond our friends. If Epicurus allowed this move to secure his account of friendship, it was not available to him when he came to place the virtues within the good life. This is a particular difficulty in the case of justice, since this requires that one take into account the interests of other people even when one has no affection for them. However, just as Epicurus was concerned to reconcile his official hedonism with the practices of friendship, so he assiduously maintained that it was compatible with virtue. Indeed, he maintained that the happy life was not possible unless the agent was virtuous. So, prudence ‘teaches’ that one cannot live pleasurably without living prudently, honourably and justly, and if one lives prudently, honourably and justly, one must live happily: ‘for the virtues are naturally linked with living pleasurably, and living pleasurably is inseparable from them’ (Men. 132 (LS 21B6)). One’s first reaction to this, however, is that it is just too blithe: one wants to know how it is that prudence—practical rationality—teaches this, and Epicurus does not go on to provide an argument here. As we have seen, Aristotle could allow that acting virtuously can contribute to one’s happiness just because such activity is intrinsically valuable. The virtuous agent will indeed take pleasure in acting virtuously. This is where Epicurus’ identification of happiness with the static pleasures causes difficulty—for the pleasure of acting well, even if he recognised it, would be a kinetic pleasure, and thus not something which could be a component of happiness. Although he avoids the dangers of subjective hedonism by identifying happiness with aponia and ataraxia, the effect of this identification is to make difficult any attempt to bring acting in other people’s interests within the sphere of the agent’s own happiness (except in the case of friendship). So, although the Epicurean wise man will, we are told, act in accordance with virtue, this has to be just because he is himself better off by doing so, and not because he recognises any reason to do so which is independent of his own well-being. In De Finibus I (42–54), Torquatus is duly at pains to show that the Epicurean will act virtuously. So, we are told that vices such as rashness, lust, cowardice and injustice trouble the mind by their very presence (50). Moreover, if one acts unjustly, one can never know that this will not be discovered, and so one will be troubled by the possibility of punishment. As Torquatus points out, reasonably enough, someone who has attenuated his desires in line with the Epicurean injunction to follow only those which are natural and necessary will in fact have little reason to act unjustly (52–3). Nevertheless, properly speaking, justice is not to be chosen for itself, but because it provides pleasure (53). It does so because if one treats other people properly, one will gain their affection, which is pleasant in itself, and one’s life will be made more secure. Thus, although Epicurus recognises that there are requirements of justice—requirements which he seems to have taken to be generated by social contracts—he does not allow that these do not provide reasons for action because justice is in itself a good thing (or injustice a bad thing, KD 34 (LS 22A4)) but rather because acting unjustly will produce more distress than acting justly: ‘The just life is most free from disturbance, but the unjust life is full of the greatest disturbance’ (KD 17 (LS 22B3)). Epicurus’ theory of the good life is thus a strange mixture of the revisionary and the conservative. It is perhaps in its account of pleasure that it is most revisionary: the states of aponia and ataraxia, the static pleasures, look very unlike the sort of things which had ben taken to be pleasures. Having set these up as the pleasures which are constitutive of happiness, however, Epicurus is then able to provide hedonistic arguments for restricting the rational agent’s pursuit of kinetic pleasures: once one has achieved them, one’s well-being cannot be increased through the addition of more of the latter. This prevents the Epicurean from espousing a view according to which one will be happier the greater number of pleasures one can experience. Given this, it is indeed plausible enough to think that the Epicurean wise man will in fact lead a life which does not violate the norms of virtue (although it will no doubt be easy enough to imagine situations where he might). However, Epicurus does not manage to show that his account of the good can accommodate the idea that the virtues present reasons for action which are autonomous: when the Epicurean acts virtuously this is because he regards this as the most effective means of achieving ataraxia. In the case of friendship, Epicurus is able to allow genuinely altruistic action because of the fact that one can come to care as much about the friend’s well-being as one does about one’s own. In the case of justice, however, his motivational concerns are not ultimately displaced from his own well-being. ABBREVIATIONS Sources frequently quoted are abbreviated as follows: DL Diogenes Laertius, Lives of the Philosophers. X=book 10 Hdt. Epicurus, Letter to Herodotus KD Epicurus, Kyriai Doxai (Key Doctrines) Math. Sextus Empiricus, Adversus mathematicos Men. Epicurus, Letter to Menoeceus Pyth. Epicurus, Letter to Pythocles VS Vaticanae sententiae (words of Epicurus found only in a Vatican MS) NOTES 1 Where possible—and, fortunately, this is frequently—I cite the translation given by Long and Sedley [6.3] in their The Hellenistic Philosophers, although I have very occasionally adapted their translations. Thus ‘LS 5A’ refers to passage A in section 5 of that work. I have done this not merely out of laziness, but because it seems to me helpful if the texts one cites, even in translation, have an existence which is independent of their employment in a particular context, so that the reader can more readily check up on how properly they are employed. Of course, one still needs to remember that these are translations of the real thing and not the thing itself. 2 In fact the argument as it stands is not a good one, even for the weaker conclusion, since it might well be that a series of perceptions can cumulatively provide compelling evidence against a single deviant perception, particularly if one has a theory of how that perception was produced. The argument would seem to miss the point, for instance, of Aristotelian warnings against accepting the perceptions of sick people rather than providing a rebuttal of such a strategy. 3 For the claim that all would travel downwards see Lucretius I.984–991 (LS 1064), and for the claim that all atoms travel at the same speed, see Hdt. 61 (LS 11 E1). 4 Perhaps it is helpful here just to clarify a distinction between what I shall call ‘materialism’ and what I shall call ‘physicalism’. I take a materialist thesis to concern substances: materialism about a certain kind of substance requires that one accept that substances of that kind have a material constitution. Physicalism, in contrast, is concerned with the relation between events (and perhaps states of affairs), and will give some sort of priority to physical events. Although Epicurus is straightforwardly a materialist about the psuchê, it is not yet obvious whether he thinks that, for instance, psychological events are determined by physical (in context, atomic) events. 5 Lucretius De rerum natura IV.722 ff. 6 It is not just perception which requires the impact of external eidôla, but all appearances and thoughts. 7 This is argued by David Sedley in his ‘Epicurean anti-reductionism’, in J.Barnes and M.Mignucci (eds) [6.8]. 8 Long and Sedley argue that volition itself can cause an atomic swerve, but there is no direct evidence for this, and if it were correct then all Epicurus’ opponents on this matter would be guilty of ignoratio elenchi. 9 See Sextus Math. VII.203, cited above p. 193. 10 Cf. KD 3: ‘The removal of all pain is the limit of the magnitude of pleasures. Wherever pleasure is present, as long as it is there, pain or distress or their combination is absent.’ 11 This elucidation here is the same as that offered by Julia Annas [6.18], Chapter 11. 12 See Men. 130–2(LS 2164–6). 13 Even non-necessary natural desires can arise from empty beliefs (when, however strongly felt, their frustration would not lead to pain—KD 30 (LS 21E3)): presumably the thought is that one might form a strong desire, say, to listen to minimalist music because it was fashionable, so that this desire, although a specific version of a natural and necessary desire, would not give rise to pleasure when satisfied. 14 ‘Accustom yourself to the belief that death is nothing to us. For all good and evil lie in perception, whereas death is the absence of perception. Hence a correct understanding that death is nothing to us makes the mortality of life enjoyable, not by adding infinite time, but by ridding us of the desire for immortality. For there is nothing fearful in living for one who genuinely grasps there is nothing fearful in not living’ (Men. 124 (LS 24A1)). 15 Pleasures for the Cyrenaics were firmly of the kinetic kind—they had no truck with taking such things as aponia and ataraxia to be genuine pleasures. BIBLIOGRAPHY ITEMS RELEVANT TO CHAPTERS 6–8 Texts and translations 6.1 Diogenes Laertius, Lives of Eminent Philosophers, edited and trans. by R.D. Hicks, Cambridge, Mass., Harvard University Press, 1925, reprinted with new introductory material 1972; especially book 4 (Sceptics), book 7 (Stoics), book 10 (Epicurus). 6.2 Sextus Empiricus, Outlines of Pyrrhonism and Adversus Mathematicos (= Against the Logicians, Against the Physicists, Against the Ethicists, Against the Professors), trans. by R.G.Bury, Loeb Classical Library, Harvard University Press, 1933–49 (repr. 1976–87). 6.3 Long, A.A. and Sedley, D.N., The Hellenistic Philosophers, 2 vols, Cambridge University Press, 1987. Selected texts, with English translation, commentary, and bibliography. 6.4 Hellenistic Philosophy: Introductory Readings trans. by Inwood, B. and Gerson, L.R., Indianapolis, Hackett, 1988. Proceedings of the Symposium Hellenisticum 6.5 Schofield, M., Burnyeat, M., Barnes, J., eds, Doubt and Dogmatism: Studies in Hellenistic Epistemology, Oxford, 1980. 6.6 Barnes, J., Brunschwig, J., Burnyeat, M., eds, Science and Speculation: Studies in Hellenistic Theory and Practice, Cambridge/Paris, 1982. 6.7 Schofield, M. and Striker, G., eds, The Norms of Nature: Studies in Hellenistic Ethics, Cambridge/Paris, 1986. 6.8 Barnes, J., and Mignucci, M., eds, Matter and Metaphysics, Naples, Bibliopolis, 1988. 6.9 Brunschwig, J. and Nussbaum, M., eds, Passions and Perceptions: Studies in Hellenistic Philosophy of Mind, Cambridge, 1993. 6.10 Laks, A. and Schofield, M., Justice and Generosity: Studies in Hellenistic Social and Political Philosophy, Cambridge, 1995. Other collections of essays 6. 11 Brunschwig, J., Papers in Hellenistic Philosophy, Cambridge, 1994. 6.12 Flashar, H., Gigon, O. and Kidd, I.G., Aspects de la philosophie Hellénistique, Geneva, Fondation Hardt, 1986. 6.13 Dillon, J.M. and Long, A.A., The Question of ‘Eclecticism’: Studies in Later Greek Philosophy, University of California, 1988. 6.14 Griffin, M. and Barnes, J., eds, Philosophia Togata: Essays on Philosophy and Roman Society, Oxford, 1989. Vol. 2, 1997. 6.15 Striker, G., Essays on Hellenistic Epistemology and Ethics, Cambridge, 1996. Books on Hellenistic philosophy 6.16 Algra, K. and others, eds, The Cambridge History of Hellenistic Philosophy, Cambridge, Cambridge University Press, 1998. 6.17 Annas, J., Hellenistic Philosophy of Mind, Berkeley and Los Angeles, University of California Press, 1992. 6.18 Annas, J., The Morality of Happiness, New York, Oxford University Press, 1993. 6.19 Flashar, H., ed., Die Philosophie der Antike, Band 4: Die Hellenistische Philosophie, Basel, 1994 (with detailed bibliographies). 6.20 Hicks, R.D., Stoic and Epicurean, London, 1910. 6.21 Long, A.A., Hellenistic Philosophy, 2nd edn, London/Berkeley/Los Angeles, University of Calilfornia Press, 1986. 6.22 Nussbaum, M.C., The Therapy of Desire: Theory and Practice in Hellenistic Ethics, Princeton, Princeton University Press, 1994. 6.23 Sharples, R.W., Stoics, Epicureans and Sceptics: An Introduction to Hellenistic Philosophy, London and New York, Routledge, 1994. 6.24 Zeller, E., Stoics, Epicureans and Sceptics, London, 1880 (translation, by O.Reichel, of Zeller’s Die Philosophie der Griechen in ihre historischen Entwicklung, vol. 3.1, 1852, rev. E.Wellmann, 1923). EPICUREAN BIBLIOGRAPHY Texts 6.25 Arrighetti, G., Epicuro: Opere, Torino, Einaudi, 1st edn, 1960; 2nd edn revised, 1973. Complete, with Italian translation. 6.26 Bailey, Cyril, Epicurus: The Extant Remains, Oxford, Clarendon Press, 1926. Main texts, with English translation. 6.27 Bailey, Cyril, Lucretius: De Rerum Natura, Oxford, Clarendon Press, 1947. Latin text with English translation and commentary. 6.28 Smith, Martin Ferguson, ed., Diogenes of Oenoanda. The Epicurean Inscription, Naples, Bibliopolis, 1993. Text with English traslation and commentary. Bibliography Full and recent bibliography in [6.16] Cambridge History of Hellenistic Philosophy. See also Flashar [6.19]. General studies 6.29 Bailey, Cyril, The Greek Atomists and Epicurus, Oxford, Clarendon Press, 1928. 6.30 Boyancé, P., Lucrèce et l’Épicurisme, Paris, 1963. 6.31 Rist, J.M., Epicurus: An Introduction, Cambridge, 1972. Collected papers See above, [6.5–6.15]. Special topics 6.32 Asmis, E., Epicurus’ Scientific Method, Ithaca, NY, Cornell University Press, 1984. 6.33 Clay, D., Lucretius and Epicurus, Ithaca, NY and London, Cornell University Press, 1983. 6.34 Englert, W.G., Epicurus on the Swerve and Voluntary Action, Atlanta, 1987. 6.35 Everson, S., ‘Epicurus on mind and language’, in Companions to Ancient Thought 3: Language, ed. S.Everson, Cambridge, 1994. 6.36 Festugière, A.J., Epicurus and his Gods, Oxford, Blackwell, 1955. 6.37 Furley, D.J., Two Studies in the Greek Atomists: (1) Minimal Parts; (2) The Swerve, Princeton, Princeton University Press, 1967. 6.38 Furley, D.J., ‘Democritus and Epicurus on Sensible Qualities’, in Brunschwig and Nussbaum [6.9], 72–94. 6.39 Furley, D.J., ‘Nothing to us?’ (on death), in Schofield and Striker [6.7], 75– 92. 6.40 Mitsis, P., Epicurus’ Ethical Theory: The Pleasures of Invulnerability, Ithaca, NY and London, Cornell University Press, 1988. 6.41 Nussbaum, M.C., ‘Therapeutic arguments: Epicurus and Aristotle’, in Schofield and Striker [6.7], 31–74. 6.42 Sedley, D., ‘Two conceptions of vacuum’, Phronesis 27 (1982), 175–93. 6.43 Sedley, D., ‘Epicurus’ refutation of determinism’, Syzetesis (Fest. Gigante), Naples, Bibliopolis, 1983, 11–51. 6.44 Sedley, D., ‘Epicurean anti-reductionism’, in Barnes and Mignucci [6.8], 295– 328. 6.45 Striker, Gisela, ‘Epicurus on the truth of sense impressions’, Archiv für Geschichte der Philosophie 59 (1977), 125–42. 6.46 Taylor, C.C.W., ‘All perceptions are true’, in Schofield et al. [6.5], 105–24.
Routledge History of Philosophy. Taylor & Francis e-Library. 2005.