by Tim O'Keefe
Democritus, however, famously denies that these sorts of properties exist in objects. Orange juice may appear to be sweet, but in reality it is no more sweet than it is sour. So our senses systemically mislead us, representing objects as having properties they do not really have, and this makes knowledge difficult, if not impossible, to attain. Epicurus believes that these conclusions are unacceptable, in part because this scepticism would have devastating practical consequences. He needs to reaffirm the reality of these sensible qualities, even while accepting Democritus’ basic atomist ontology and theory of perception.
Democritus and the unreality of sensible qualities
After establishing that atoms and void exist, Democritus gives a detailed account of how the perceptual qualities associated with each sense arise as a result of the causal interaction of atoms with the sense-organs.2 For instance, the taste “bitter” is explained as a result of sharp atoms tearing the tissue of the tongue, “sweet” as the soothing action of round and fairly large atoms on the tongue.3 The Epicureans largely accept Democritus’ account.4
However, the same object can affect different percipients differently. The same orange juice that seems pleasingly sweet to most of us will taste unspeakably vile to a person who has just brushed his teeth. A can of soda that looks red to me will have no colour in the dark, and will seem yellow to a dog, as it lacks the retinal cones needed to see red. On the basis of this sort of perceptual relativity, Democritus infers that objects are “no more” (ou mallon) sweet than disgusting, no more red than yellow, because they are neither sweet nor disgusting, neither red nor yellow (Sext. Emp. Pyr. I 213).
Some properties of bodies can be identified with properties of atoms, such as weight and hardness, and these are thought of as real properties of bodies (Theophr. Sens. 62). But when we look at the bodies themselves, we discover that, in themselves, the bodies have nothing like “sweetness” or “bitterness” or “redness” in them. Instead, they are simply atomic aggregates that can appear differently to different percipients. Democritus concludes that none of the sensible qualities exist in nature; instead, they are simply affections of the senses (Theophr. Sens. 63–4). Sweetness is not out there in the world; it exists only in my mind. This leads to one of Democritus’ most famous sayings: “Sweet exists by convention, bitter by convention, colour by convention; atoms and void [alone] exist in reality” and this in turn leads to scepticism: “We know nothing accurately in reality, but only as it changes according to the bodily condition, and the constitution of those things that flow upon (the body) and impinge upon it” (Sext. Emp. Math. VII, 135 [DK 68 B9]).
There is a radical discontinuity between the properties to which we have access, and which must form the basis of all our knowledge, and the properties that exist in reality. Democritus says we are severed from reality (Sext. Emp. Math. VII, 137 [DK 68 B6]), and he probably has in mind that the sensible qualities, which must form the basis of all knowledge, are found not to be a part of reality, that is, they are merely subjective. In fact, almost all of the reports furnished by the senses (for instance, “The honey is sweet”) turn out on inspection to be false. We think that the honey itself is really sweet, but the sweetness is not in the honey at all: it is simply a change in our sense-organ. Democritus’ epistemological pessimism is widely reported; for instance, he says that we know nothing, because truth is in an abyss (DL VII 72 [DK 68 B117] ), and that we do not grasp how each thing is or is not.
The complaint of the senses against reason shows that Democritus is well aware of the possibly self-stultifying nature of his philosophy: “Wretched mind, do you take your evidence from us and then try to overthrow us? Our overthrow is your downfall” (reported in Gal. On Medical Experience XV 8 [DK 68 B125], trans. Hankinson 1995). There is a painful irony in Democritus’ philosophy: his atomism is a response to the Eleatic philosophers, such as Parmenides and Melissus, who deny the reality of change and the phenomenal world. Atomism is supposed to provide an answer to the Eleatic challenge, as well as provide economical and comprehensive causal explanations for the features of the world. Democritus’ atomism, however, undercuts the authority of the senses as a source of information about the world, which in turn leads to the collapse of reason, including the theories arrived at by the use of reason, such as atomism.5
Epicurus and the reality of the relational
The Epicureans believe that Democritus’ doctrines lead to full-blown scepticism and make life impossible: saying that every single thing is “no more this than that”, for example no more sweet than bitter, throws our life into chaos (Plut. Adv. Col. 1108f). Epicurus tries to avoid this scepticism, and its devastating practical consequences, by staunchly defending the reality of sensible qualities. He denies the validity of Democritus’ inference of the unreality of sensible qualities on the basis of perceptual relativity, and he does so as part of a wide-ranging defence of the reality of the relative.
Democritus is far from the only figure to deploy an ou mallon (no more) argument. For instance, the sophist Protagoras says that the wind is, in itself, no more hot than cold, because it is both hot and cold. He avoids contradiction by saying that it is hot for one person, and it is cold for another (Pl. Tht. 152b). Plato is no sceptic, but he often uses ou mallon arguments to show that some property cannot be truly instantiated in the phenomenal world. Plato frequently employs the principle that for any thing to be truly F, it must be F without qualification. So, for instance, for something to be truly just, it must be always just, not just at some times and unjust at others. Or, to use another example, Socrates says that a beautiful maiden is not truly beautiful because, although she is beautiful in comparison to monkeys, she is not beautiful in comparison to the gods (Pl. Hp. mai. 289b–d). Later sceptics have an epistemological reading of the ou mallon principle. They start from the fact of the relativity of perceptual qualities, or of value predicates, and argue that we can no more say that the thing is F than not-F, because we have no criterion by which to judge between the reports and decide which property the object itself has (Sext. Emp. Pyr. I 188).
Despite this great diversity among the different uses of the ou mallon argument, all of those who use the argument have a common interest in what a thing is “by nature” (phusis). And what this seems to mean, generally, is what a thing is in and of itself, that is, what it is intrinsically. The theme that is consistent throughout the various ou mallon arguments is the move from the observation that some property of an object differs relative to different observers, times or conditions (a appears F to me but not-F to you, or is F under certain circumstances but not-F under others) to the conclusion either that the object does not, in itself, have that property, or that we cannot know whether the object has that property or not.
The Epicureans, however, admit relational and dispositional properties into their ontology. Lucretius includes servitude, liberty and poverty among his list of accidental properties (DRN I 455–6). If “servitude” is a real property in one’s ontology, it cannot be an intrinsic property, because a person is not enslaved per se, but only because of certain very complex relationships that hold between him and other people.
The Epicureans believe that a correct understanding of the nature of relational properties defeats sceptical ou mallon arguments. Polystratus, the third head of Epicurus’ Garden, makes a sustained and convincing defence of the reality of relational and dispositional predicates, in the course of defending the reality of value properties “fair” and “foul” against the charge that these things are falsely believed in, because what is fair and foul is not the same everywhere and under all circumstances (Polystratus On Irrational Contempt 23.26–26.23, LS 7D). He says that relative predicates do not have the same status as things said not relatively, so we should not expect them to behave in the same way. For example, “bigger” is a relational predicate, so something cannot be bigger per se; it can only be bigger than something else. So to say that Simmias is not “really” bigger than Socrates because he is
both bigger than Socrates but smaller than Phaedo would be naive.
Polystratus writes that powers are the most evident case of such relational properties. It is not the case that the same things are nourishing or deadly for all creatures; instead, the same thing can be nourishing for some but deadly for others. But we do not conclude on this basis that properties such as deadliness are nonexistent. Peanuts may be deadly for somebody with an allergy but nourishing for other people, and cyanide may be deadly poison for me and not poisonous for some race of aliens. But that does not make cyanide “no more deadly than not deadly”, so that I become sceptical about the deadliness of cyanide for me. If I really think so, I should swallow some and see what happens, thus incrementally reducing the number of sceptics in the world. The deadliness of the cyanide for me is a real property, albeit a relational one, of the cyanide, not something that is merely conventional or subjective. And this is exactly what we should expect, if we understand the sort of property deadliness should be.
The Epicureans seem to think that the sensible qualities of bodies are such dispositional qualities. In discussing the properties of wine, Epicurus says that it would be a mistake to think that wine generally has heating or cooling properties. Instead, it has a mixture of powers, such that a certain quantity of it would be heating for certain individuals with a certain bodily condition and cooling for others under different circumstances. And he says that the same sort of thing applies to colours: colours are not intrinsic properties but are produced by the ordering and positioning of the atoms in relation to our sight (Plut. Adv. Col. 1109e–1110d).
So, it seems, Epicurus thinks that sensible qualities are complicated dispositional properties that cause certain sensory affections within the percipient’s mind when interacting with the percipient’s sense-organs under certain conditions. This theory would allow Epicurus to admit the phenomena of sensory variability and retain the basic Democritean account of how sensations arise as a result of the interaction of atoms emitted from objects with our sense-organs, while still holding that sensible qualities are real properties of bodies. The can really is red, because it has an atomic make-up such as to cause us to see red under ordinary circumstances. It might not appear red to a dog, but then there is no reason to expect a red object to appear the same way to a dog as it does to us.
Plutarch contends that the Epicurean theory falls prey to the same sort of sceptical difficulties that afflict Democritus. Plutarch notes that Colotes’ main objection against Democritus is that Democritus says bodies are in reality “no more this than that”. Plutarch says that Epicurus himself, however, admits the relativity of perceptual properties in a way that undermines his claim that they are real properties of objects. Plutarch gives several examples, such as the heating and cooling powers of wine discussed above, and also the way that an object in a dim room may appear coloured to one person and not coloured to another, owing to a difference in the strength of their vision. It follows, thinks Plutarch, that for Epicurus too the can is “no more this way than that”, but Epicurus, unlike Democritus, brazenly refuses to admit this consequence of his theory.
Plutarch’s objection to Epicurus, however, has little force. It rests on the fact, which Epicurus happily concedes, that the same object, in the same conditions, can appear differently to different percipients. Plutarch concludes from this that sensible qualities are not real properties of objects for Epicurus. This would follow only if, for something to be a real property of an object, it must be an intrinsic property. But Polystratus explicitly denies this. It is important not to conflate two very different pairs of distinctions: intrinsic versus relative, on the one hand, and objective versus subjective, on the other. The fact that some property is relative does not make it thereby subjective. Cyanide is deadly to me, although maybe not to all organisms. Similarly, the object in the room really has the property of causing certain sensations in certain people under certain conditions.
Regarding sensible qualities as dispositional properties does not resolve all possible sceptical difficulties that might arise from the relativity of perception. The same atomic state can cause differing perceptual states, depending on the condition of the percipient. More importantly, the same perceptual state can be caused by differing atomic states. Since this is so, there may be a problem with drawing inferences from perceptual experience about the extra-mental properties of bodies.
This problem may seem to be made even more pressing once we add to the mix the Epicurean doctrine that “all sensations are true”. The redness of the can, according to Epicurus, is a real property that the can itself has, in virtue of which it causes people like me to have visual experiences of red. Fair enough. And then, when I see the red can and say “The can is red”, I am saying something true. But the perception of the can as yellow by somebody with jaundice is equally true, as is the perception of it as grey by a person with complete colour blindness.
Consideration of these problems will be postponed until Chapter 10. The Epicurean understanding of sensible qualities as dispositional properties does not solve all possible sceptical difficulties. However, it does get around Democritus’ worry that because sensible qualities are relative they are somehow not real properties of bodies, but merely subjective.
Furthermore, experiences of bitterness, redness and so on are not caused merely by bodies having the dispositional properties to cause such states, although it is true that they do. They have these dispositional properties in virtue of other complicated structural properties of the atoms and groups of atoms; for example, the taste bitter is caused by rough and hooked atoms tearing up the tongue; the various colours we see are caused by the arrangement and shape of the atoms on the surfaces of bodies (scholion to Ep. Hdt. 44). Thus, these dispositional properties are tied systematically to complex structural properties of the atoms themselves and the bodies that are constituted by the atoms. This is probably why the atomists often identify having a sensible quality straightforwardly with some atomic property: if the sensible quality is a dispositional quality, and the dispositional property can be explained entirely in terms of some set of atomic properties, it is not difficult to see why the sensible quality would be identified with the atomic property. So the information we receive via sensible qualities is not trivial, because the different sensible qualities are tied to the underlying atomic structures of bodies.
FIVE
Cosmology
The Epicureans try to account for the formation of the earth and of the other heavenly bodies in terms of atoms blindly colliding in the void. Against most other cosmologists, they maintain that our world is only one of an infinite number of worlds, coming to be and falling apart in a spatially infinite universe that has existed and will exist forever. For the Epicureans, however, the most important feature of their cosmology is not that the universe is infinite; it is that it is purposeless. Explanations of the formation of the world and of phenomena such as earthquakes and lightning in terms of the interactions of atoms are supposed to displace retrograde and superstitious explanations that appeal to the gods. Not only that: the Epicureans are among the first philosophers to raise the problem of evil in order to argue that the world is not under the control of beneficent gods, as it is too flawed.
The infinite universe
The cosmos is our particular ordered world-system: the earth, sun, moon, planets, and stars. Leaving aside the stars – which are thought by the ancient Greeks to be in approximately the same area as the other celestial bodies – the cosmos is thus something like the solar system. But there is a sharp divide in antiquity between “two pictures of the world”.1 The first picture, defended by Plato, Aristotle and the Stoics, identifies the cosmos and to pan, “the all”, that is, the physical universe. So on this picture, the universe is finite and unique. Epicurus defends the alternative picture of the world, first developed by Leucippus and Democritus, of an infinite universe. Our particular limited cosmos is only one of an infinite number of cosmoi (the plural of cosmos)
, each of which comes into existence and will eventually fall apart. But the universe as a whole has no beginning and no end; it has always existed and will always exist. And spatially, the universe stretches infinitely in all directions.
According to the Epicureans, cosmoi form when there happens to be a great concentration of matter in one region of space. The cosmos starts as a turbulent mass, and the elements of earth, water, air and fiery ether are all mixed together. But over time they begin to separate out, with like element starting to unite with like. But this is not because of any mysterious affinity of like for like. Instead, the particles of earth, as the heaviest element, settled towards the middle of the cosmos, squeezing out the lighter elements. So eventually we get layers of earth, then water, then air, then ether. The sun and moon are spherical bodies midway between earth and ether in their density, and they thus float in the air. The settling process does not result in perfect uniformity, however, so some areas of the earth still stick out above the water, and there are high mountains and low plains (DRN V 416–508; see also Aëtius 1.4.1–4, IG I-92).
All atoms naturally fall “downwards” at uniform speed, as we have seen (Chapter 3), so “heavier” and “lighter” here do not mean differences in natural rate of fall; instead, “heavier” elements, owing to their size and shape, are better suited to push their way down through “lighter” elements. The “lighter” elements, in turn, are forced upwards by being squeezed that way, even though – contra Aristotle – the natural motion of elements such as fire is still down, not up (DRN II 184–215).
