by Tim O'Keefe
Epicurus’ metaphysics is resolutely materialistic: the only things that exist per se are atoms and void. Atoms are uncuttable bits of solid body, moving through void, which is simply empty space. Atoms have a limited stock of properties: size, shape, weight and resistance to blows (Chapter 2). These atoms fall through space because of their weight, and in order to explain how they collide, rebound and become entangled with one another to form macroscopic bodies, Epicurus posits a random atomic swerve (Chapter 3). There are many properties, such as being weary, being red or being a stimulant, that can be possessed only by conglomerates of atoms, not individual atoms, and are thus “emergent” in some sense. Yet Epicurus wants to say that only atoms and void exist per se, and that the possession of these emergent properties by macroscopic objects can be exhaustively explained in terms of the properties of the atoms that make up these objects, along with their relations to other atoms. In particular, Epicurus wants to account for the reality of sensible qualities, such as redness and bitterness, within his reductionist programme. Democritus denied that such properties really exist, saying that in truth there is only atoms and void, and this leads him to doubt the reliability of the senses; Epicurus needs to combat Democritus on this question (Chapter 4).
One important result of Epicureans physics is that a satisfying explanation for the formation of the world, and for phenomena such as earthquakes and rain, can be given entirely in terms of atomic motions, which he thinks excludes explanations that appeal to divine will (Chapter 5). In the biological realm, too, we can account for the functioning of organisms from the “bottom up” in terms of atoms and their properties, without any reference to purposes or functions within nature (Chapter 6). We are among those organisms, and the functioning of our minds is included in this programme: the mind is simply a bodily organ responsible for mental functions such as perception, as the heart is the organ responsible for pumping our blood. An important upshot of this analysis is that death is annihilation (Chapter 7). Such a materialistic view of the world might seem incompatible with human beings having freedom of action, but Epicurus tries to accommodate the possibility of freedom in the world, in part by using the indeterministic atomic “swerve” that somehow allows us to act as we wish (Chapter 8).
TWO
Atoms and void
The existence of atoms and void
There are bodies in motion. No argument is needed to establish this; we simply see bodies in motion. Epicurus may have tossed a rock in the air and pointed at it if asked to demonstrate this point, and if someone pressed him further even after he done this, he may have tossed a rock at the person. (In Chapter 9 we shall explore further Epicurus’ arguments against the sceptic, and in Chapter 10 his account of the role sensation plays in gaining knowledge of the world.) From this observation, it follows trivially that there are bodies. But establishing the existence of void – where “void” is simply empty space in which there are no bodies – requires some argument. The basic argument, however, is simple (Ep. Hdt. 40, DRN I 329–45):
1.
If there is motion, there is void.
2.
There is motion.
* * *
3.
Therefore, there is void.
Premise 2, as indicated above, is supposed to be a datum of experience. As for premise 1, if the universe were a plenum – that is, if it were packed totally full of body, with no empty space – there would be nowhere for bodies to move into, and so they would not move at all. That is because part of what makes bodies what they are is that they resist blows: they do not allow things to move through them; they get in the way. Once there is empty space a body in motion can run into a second body and sometimes push it out of the way as the second body in turn moves into empty space.
Lucretius gives a second reductio argument for the existence of void: if there were no void, all objects of equal size should have equal weight, since, being equally full of body, they would have equal quantities of matter. But this conclusion is obviously false. To account for the fact that a ball of wool weighs much less than a ball of lead of equal diameter, we must suppose that the ball of wool has more void space within it than does the ball of lead (DRN I 360–69).
The Epicureans were in a minority in believing in the existence of absolutely empty space, and plenum theorists would not be impressed with either of these arguments. As to the first, plenum theorists had their own explanation of how motion could happen in a plenum, via “reciprocal replacement”: the place that a moving body formerly occupied allows the bodies that it is pushing aside somewhere to go. The simplest example of this would be a rotating sphere; each piece of the sphere may push aside an adjacent piece, and each piece will have somewhere to go, without need for any absolutely empty space. Lucretius gives the slightly different example of a fish nosing through water, with the water it displaces going around its sides and filling in the space behind it where it used to be (DRN I 370–86). He inadequately objects to this theory by saying that the fish’s motion could not start unless there were some space already there for the water to move into, and before the fish begins moving that space is not available.
As to the second, Aristotle thought that heaviness and lightness were irreducible qualities of different types of matter, as opposed to being a function of the quantity of “full” space. Also, Lucretius’ argument presupposes that space is either absolutely empty or absolutely full, with “full” space being equal in density. However, the Presocratic Anaximenes (for instance) thought that the fundamental element, air, could exist in various states of density or rarefaction. Dense air forms stones and earth, less dense air water and clouds, and the most rarefied air becomes fire (Simpl. in Phys. 24, 26ff. [DK 13 A5]). If one thinks that matter is “squashable” in this way, one could account for differences in weight without positing void. Indeed, squashable matter would allow one to account for motion without void: even if the matter in front of a moving body would have no place to go, the moving body could compact it, while the matter behind the moving body expands to fill the space it formerly occupied.
Others went on the offensive against the intelligibility of void. In fact, the notion of void was not developed originally by Leucippus or Democritus, the inventors of atomism, but by Melissus, an Eleatic. The Eleatics (the more famous of whom were Parmenides and Zeno of Elea) gave arguments against the possibility of plurality and change that were entirely a priori, that is, based on logical and not empirical considerations. Melissus argued against the existence of motion as follows (Simpl. in Phys. 112, 6 [DK 30 B7]):
1.
There is no void.
2.
If there is motion, there is void.
* * *
3.
Therefore, there is no motion.
The basic consideration from which many Eleatic arguments begin is the apparently truistic “What is, is, and what is not, is not”. Melissus applies this to void. If void is just nothingness, it is “what is not”. But to assert the existence of what is not is contradictory. So void does not exist. Since void is a necessary condition on motion, however, it follows that motion does not exist either.
Leucippus and Democritus both argue that there is nothing incoherent about the notion of void. Void is defined privatively: it is where there is not body. And so, in some sense, void is non-being and nothing – that is, it is not a being, not a thing — but it does not follow that void does not exist at all or lacks all properties. It is simply empty space, and so we can say where it is, and say that as empty space it is yielding, in the sense of allowing bodies to enter into it and giving way with no resistance. Aristotle reports that Leucippus and Democritus were happy to advance the seemingly (but not actually) paradoxical claim that “what is not” exists no less than “what is” (Arist. Metaph. 1.4 985b4 [DK 67 A4]).
The Epicureans would agree with the above, but they take a different tack: if you think you are having trouble conceiving what void could be, all you need to do is think of th
e empty space around you right now, through which you could toss a rock if you wished. That is what void is like. The analogy is imperfect, of course, since, unlike void, the “empty” space in rooms is not absolutely empty. (This can be shown by waving a fan near your face and feeling the breeze against your skin. You feel the “blow” of the air because the space around you is full of corporeal bodies; if it were a perfect vacuum, you would feel nothing at all.) Although this “empty” space is not really void, you normally do not see the air around you, and it provides little resistance to the solid bodies moving through it, so it provides a good analogue to the microscopic stretches of absolute void. This strategy of using things at the macro-level to provide analogies of what occurs at the micro-level is quite common in Epicureanism.
So much for void: on to atoms. Before giving Epicurus’ argument for their existence, let us describe what they are. The Greek word atomos is formed from the root tomos, from cut or split, plus the so-called “alpha privative”, as in words such as “atheist” (one who believes there is no god) or “apathetic” (lacking in feelings). So, if one wanted to translate the Greek word atomoi and not simply transliterate it, “uncuttables” would be a good candidate. Ordinary objects, such as coffee cups, can be broken up into smaller parts. This process of division cannot go on indefinitely, however. Eventually one gets down to the smallest units or building blocks of matter, which cannot be broken down or split up, and out of which all compound bodies are composed. These are atoms.
The Epicureans give at least three arguments for the existence of these indivisible bits of body. The first (Ep. Hdt. 41, DRN I 540–47) is that, if all bodies are liable to be split up, then eventually they would all be reduced to nothingness. And since the universe has existed forever – a point we shall return to later – if this reduction to nothing would eventually happen, it would have happened by now, which is inconsistent with our observations. Unfortunately, it is unclear why an indefinite series of divisions would entirely annihilate things rather than simply produce infinitesimal bits of matter. Later Lucretius makes a slightly different point: if division could continue indefinitely then over time the bits of matter would be worn down to such a extent that, even if they were to exist, they would be unable to produce the complex compound bodies, especially living things, that we observe (DRN I 551–64).
The second argument (DRN I 526–39) depends on the Epicurean theory of how division occurs. Ordinary compound bodies are made up of smaller pieces that are entangled with one another in various ways, but they also contain void spaces. In such cases, a blow from outside can force the pieces apart and spilt the body. Eventually, however, you will come to a piece that is all “solid” space and no void; imagine a perfectly solid, tiny, cubical hunk of matter. In such a case, a blow from another body would make the hunk of matter as a whole bounce away, but without any void spaces to force sub-pieces apart, it would not fissure.
Finally, the existence of an enduring set of atomic constituents with fixed shapes is needed in order to explain the regularities we observe at the macroscopic level (DRN I 584–98). This, in turn, depends on the widely accepted general principle, first explicitly formulated by Parmenides, that nothing comes from nothing. Epicurus says that we must accept that nothing comes to be from what is not, because otherwise everything would come to be from everything (Ep. Hdt. 38). This seems not to follow. Luckily, Lucretius gives a more extended discussion of this principle, in which he tries to give it empirical support through the phenomena of biological generation (DRN I 159–214). We see that things come to be from certain sources (e.g. pears from pear trees), at certain times (e.g. roses in the spring, grapes in the autumn), in certain places (e.g. fish in the water) and in certain manners (e.g. adulthood following adolescence). But if we were to give up the principle that nothing comes to be from nothing, then anything would be able to come to be from anything, in any manner whatsoever. Lucretius lists some of the absurdities we might then encounter, such as human beings springing from the sea, and children too young to talk instantly becoming young adults. So the general principle of “nothing from nothing”, which Parmenides and many others take to be an a priori truth, is given empirical support by the Epicureans.
The Epicureans also accept the corollary principle that nothing perishes into nothing (Ep. Hdt. 39, DRN I 215–64). If things that were destroyed perished utterly – rather than being resolved into components that could then make new beings – then by now everything would have been annihilated.
We must accept that there is a reason why things occur in the way that they do, and not otherwise. Having a stock of unalterable atomic units of matter allows us to explain the world of orderly change and plurality without violating the Parmenidean sayings. The things we see come into being and pass away. Their ultimate constituents, however, do not come into being, but have instead always existed, and will always exist.
The properties of atoms and void
Only bodies and void exist per se, that is, exist without depending for their existence on something else. A pocket of void space between the earth and the moon, or the cubical atom as it rebounds from a collision, is ontologically basic. All other things that exist are ultimately explicable as attributes of bodies. Motion exists, but does not exist on its own: there must a body that is moving. Likewise, sizes (like two metres long) exist, but are attributes of some body. And time is a measure of motion, an “accident of an accident”. It is a property of motion and other change – there could not be a stretch of time in which absolutely nothing is happening – with motion and other change in turn being attributes of bodies (Ep. Hdt. 68–73, Sext. Emp. Math. X 219–27).
Some of these attributes are permanent, for example the shape of an atom, while others are temporary, for example my present high caffeine level. We shall look at the properties of compound bodies, such as coffee’s being bitter and a stimulant, in Chapter 4. For the moment, let us remain with atoms. Atoms have a very limited stock of properties: size, shape, location, weight and resistance to blows. These properties are simply constitutive of what it is to be a body. Something could not be bodily without being located somewhere and having a shape of some sort, and in order to have a shape it must also be extended. And if a body did not get in the way when another body tried to move into the space it was occupying – if it simply were to give way without resistance – it would not be a corporeal body at all, but simply void, which is incorporeal, that is, non-bodily. In fact, it is this “yielding” that distinguishes void from body, since void space also has size, shape and location.
Epicurus carefully notes that these properties do not exist per se, although they certainly exist. He further claims that they are not “parts” of the atom in the same sense in which the wheels and windows are parts of my car. But the body gets its permanent nature as a body from all of these properties together. So even though atoms do exist per se, it would also be acceptable to think of an atom as being just a complex of size, shape, hardness and so on; that is all there is to being that atom (Sext. Emp. Math. X 257). This allows Epicurus to sidestep a problem that John Locke encounters (although obviously he did not devise his doctrine to avoid Locke’s difficulties). For Locke, physical substance is the underlying substratum that supports bodily qualities such as size and shape. But the substance considered in itself, apart from the qualities it supports, becomes a mysterious “I know not what” (Essay Concerning Human Understanding II 23).
Epicureans assert that atoms do not have properties such as colour and odour (DRN II 730–1022). The main reason for excluding such properties is that atoms are supposed to be the stable building blocks out of which all other things arise, and that do not change in their intrinsic properties at all, but only in their locations and relationships to one another. Other things come to be and perish, while the atoms always are. However, colours, odours, and the like are all variable: the same sea can turn from dark to white when the wind whips it up, even though the atoms that compose it are mostly the same
.
Lucretius adds two points. First, the idea of a colourless body is not incoherent. Just as through touch blind people can form the conception of a body not connected to any colour, so too can we (DRN II 739–48). Lucretius also appeals to the Epicurean theory of how bodies can possess qualities such as odours (which we shall look into later): bodies have sensible qualities because they emit streams of particles that interact with our sense-organs. So a hunk of Limburger cheese smells wonderful because of particles wafting from it that enter our noses. But individual atoms, as indivisible units of matter, cannot emit streams of particles from themselves, and hence cannot themselves have sensible properties (DRN II 842–64).
Minimal parts
Atoms are physically indivisible, for the reasons given above. However, they are theoretically divisible, as they have spatial sub-parts. A cubical atom will have a top and bottom half; in a knobbly atom, each of the knobs would be a distinct (although undetachable) part. The Epicureans, however, think that even this process of theoretical division cannot go on indefinitely. Eventually, we would arrive at absolutely smallest spatial units, or spatial minima. All magnitudes are “composed” of a finite number of these spatial minima.
Why think that space is quantized in this way? The Epicureans give two primary arguments (Ep. Hdt. 57, DRN I 599–634). The first derives from the arguments of Zeno of Elea against motion.1 Before something can move from A to B, it would have to reach the midpoint of A and B. Call this C. But then, in order to reach C, it would have to reach the midpoint of A and C. Call this D. And so on and so forth: since this process of division can continue infinitely, for an object to move anywhere at all, it will have to move across an infinite number of points. And if it is impossible to pass across an infinite number of things in a finite time (as Zeno believes it is; Arist. Ph. 233a21 [DK 29 A25]), then motion is impossible. Epicurus agrees that it is impossible to move across an unlimited number of parts. But instead of accepting the manifestly false conclusion that there is no motion, he simply denies that bodies (or spaces generally) can, even theoretically, be divided without limit. So the path from A to B will contain a finite number of spatial intervals, and Zeno’s conclusion is avoided. The second argument is that any body (or spatial magnitude, more broadly speaking) made up of an infinite number of spatial parts, where those parts are themselves finite in size, would have to be infinite in size. But, obviously, not all bodies are infinite in size.
