A History of Western Philosophy

by Bertrand Russell

  The Atomists

  THE founders of atomism were two, Leucippus and Democritus. It is difficult to disentangle them, because they are generally mentioned together, and apparently some of the works of Leucippus were subsequently attributed to Democritus.

  Leucippus, who seems to have flourished about 440 B.C.,* came from Miletus, and carried on the scientific rationalist philosophy associated with that city. He was much influenced by Parmenides and Zeno. So little is known of him that Epicurus (a later follower of Democritus) was thought to have denied his existence altogether, and some moderns have revived this theory. There are, however, a number of allusions to him in Aristotle, and it seems incredible that these (which include textual quotations) would have occurred if he had been merely a myth.

  Democritus is a much more definite figure. He was a native of Abdera in Thrace; as for his date, he stated that he was young when Anaxagoras was old, say about 432 B.C., and he is taken to have flourished about 420 B.C. He travelled widely in southern and eastern lands in search of knowledge; he perhaps spent a considerable time in Egypt, and he certainly visited Persia. He then returned to Abdera, where he remained. Zeller calls him “superior to all earlier and contemporary philosophers in wealth of knowledge, and to most in acuteness and logical correctness of thinking.”

  Democritus was a contemporary of Socrates and the Sophists, and should, on purely chronological grounds, be treated somewhat later in our history. The difficulty is that he is so hard to separate from Leucippus. On this ground, I am considering him before Socrates and the Sophists, although part of his philosophy was intended as an answer to Protagoras, his fellow-townsman and the most eminent of the Sophists. Protagoras, when he visited Athens, was received enthusiastically; Democritus, on the other hand, says: “I went to Athens, and no one knew me.” For a long time, his philosophy was ignored in Athens; “It is not clear,” says Burnet, “that Plato knew anything about Democritus…. Aristotle, on the other hand, knows Democritus well; for he too was an Ionian from the North.”* Plato never mentions him in the Dialogues, but is said by Diogenes Laertius to have disliked him so much that he wished all his books burnt. Heath esteems him highly as a mathematician.†

  The fundamental ideas of the common philosophy of Leucippus and Democritus were due to the former, but as regards the working out it is hardly possible to disentangle them, nor is it, for our purposes, important to make the attempt. Leucippus, if not Democritus, was led to atomism in the attempt to mediate between monism and pluralism, as represented by Parmenides and Empedocles respectively. Their point of view was remarkably like that of modern science, and avoided most of the faults to which Greek speculation was prone. They believed that everything is composed of atoms, which are physically, but not geometrically, indivisible; that between the atoms there is empty space; that atoms are indestructible; that they always have been, and always will be, in motion; that there are an infinite number of atoms, and even of kinds of atoms, the differences being as regards shape and size. Aristotle † asserts that, according to the atomists, atoms also differ as regards heat, the spherical atoms, which compose fire, being the hottest; and as regards weight, he quotes Democritus as saying “The more any indivisible exceeds, the heavier it is.” But the question whether atoms are originally possessed of weight in the theories of the atomists is a controversial one.

  The atoms were always in motion, but there is disagreement among commentators as to the character of the original motion. Some, especially Zeller, hold that the atoms were thought to be always falling, and that the heavier ones fell faster; they thus caught up the lighter ones, there were impacts, and the atoms were deflected like billiard balls. This was certainly the view of Epicurus, who in most respects based his theories on those of Democritus, while trying, rather unintelligently, to take account of Aristotle’s criticisms. But there is considerable reason to think that weight was not an original property of the atoms of Leucippus and Democritus. It seems more probable that, on their view, atoms were originally moving at random, as in the modern kinetic theory of gases. Democritus said there was neither up nor down in the infinite void, and compared the movement of atoms in the soul to that of motes in a sunbeam when there is no wind. This is a much more intelligent view than that of Epicurus, and I think we may assume it to have been that of Leucippus and Democritus.*

  As a result of collisions, collections of atoms came to form vortices. The rest proceeded much as in Anaxagoras, but it was an advance to explain the vortices mechanically rather than as due to the action of mind.

  It was common in antiquity to reproach the atomists with attributing everything to chance. They were, on the contrary, strict determinists, who believed that everything happens in accordance with natural laws. Democritus explicitly denied that anything can happen by chance.† Leucippus, though his existence is questioned, is known to have said one thing: “Naught happens for nothing, but everything from a ground and of necessity.” It is true that he gave no reason why the world should originally have been as it was; this, perhaps, might have been attributed to chance. But when once the world existed, its further development was unalterably fixed by mechanical principles. Aristotle and others reproached him and Democritus for not accounting for the original motion of the atoms, but in this the atomists were more scientific than their critics. Causation must start from something, and wherever it starts no cause can be assigned for the initial datum. The world may be attributed to a Creator, but even then the Creator Himself is unaccounted for. The theory of the atomists, in fact, was more nearly that of modern science than any other theory propounded in antiquity.

  The atomists, unlike Socrates, Plato, and Aristotle, sought to explain the world without introducing the notion of purpose or final cause. The “final cause” of an occurrence is an event in the future for the sake of which the occurrence takes place. In human affairs, this conception is applicable. Why does the baker make bread? Because people will be hungry. Why are railways built? Because people will wish to travel. In such cases, things are explained by the purpose they serve. When we ask “why?” concerning an event, we may mean either of two things. We may mean: “What purpose did this event serve?” or we may mean: “What earlier circumstances caused this event?” The answer to the former question is a teleological explanation, or an explanation by final causes; the answer to the latter question is a mechanistic explanation. I do not see how it could have been known in advance which of these two questions science ought to ask, or whether it ought to ask both. But experience has shown that the mechanistic question leads to scientific knowledge, while the teleological question does not. The atomists asked the mechanistic question, and gave a mechanistic answer. Their successors, until the Renaissance, were more interested in the teleological question, and thus led science up a blind alley.

  In regard to both questions alike, there is a limitation which is often ignored, both in popular thought and in philosophy. Neither question can be asked intelligibly about reality as a whole (including God), but only about parts of it. As regards the teleological explanation, it usually arrives, before long, at a Creator, or at least an Artificer, whose purposes are realized in the course of nature. But if a man is so obstinately teleological as to continue to ask what purpose is served by the Creator, it becomes obvious that his question is impious. It is, moreover, unmeaning, since, to make it significant, we should have to suppose the Creator created by some super-Creator whose purposes He served. The conception of purpose, therefore, is only applicable within reality, not to reality as a whole.

  A not dissimilar argument applies to mechanistic explanations. One event is caused by another, the other by a third, and so on. But if we ask for a cause of the whole, we are driven again to the Creator, who must Himself be uncaused. All causal explanations, therefore, must have an arbitrary beginning. That is why it is no defect in the theory of the atomists to have left the original movements of the atoms unaccounted for.

  It must not be supposed that their reasons
for their theories were wholly empirical. The atomic theory was revived in modern times to explain the facts of chemistry, but these facts were not known to the Greeks. There was no very sharp distinction, in ancient times, between empirical observation and logical argument. Parmenides, it is true, treated observed facts with contempt, but Empedocles and Anaxagoras would combine much of their metaphysics with observations on water-clocks and whirling buckets. Until the Sophists, no philosopher seems to have doubted that a complete metaphysic and cosmology could be established by a combination of much reasoning and some observation. By good luck, the atomists hit on a hypothesis for which, more than two thousand years later, some evidence was found, but their belief, in their day, was none the less destitute of any solid foundation.*

  Like the other philosophers of his time, Leucippus was concerned to find a way of reconciling the arguments of Parmenides with the obvious fact of motion and change. As Aristotle says:†

  Although these opinions [those of Parmenides] appear to follow logically in a dialectical discussion, yet to believe them seems next door to madness when one considers the facts. For indeed no lunatic seems to be so far out of his senses as to suppose that fire and ice are “one”: it is only between what is right and what seems right from habit that some people are mad enough to see no difference.

  Leucippus, however, thought he had a theory which harmonized with sense-perception and would not abolish either coming-to-be and passing-away or motion and the multiplicity of things. He made these concessions to the facts of perception: on the other hand, he conceded to the Monists that there could be no motion without a void. The result is a theory which he states as follows: “The void is a not-being, and no part of what is is a not-being; for what is in the strict sense of the term is an absolute plenum. This plenum, however, is not one; on the contrary, it is a many infinite in number and invisible owing to the minuteness of their bulk. The many move in the void (for there is a void): and by coming together they produce coming-to-be, while by separating they produce passing-away. Moreover, they act and suffer action whenever they chance to be in contact (for there they are not one), and they generate by being put together and becoming intertwined. From the genuinely one, on the other hand, there could never have come to be a multiplicity, nor from the genuinely many a one: that is impossible.”

  It will be seen that there was one point on which everybody so far was agreed, namely that there could be no motion in a plenum. In this, all alike were mistaken. There can be cyclic motion in a plenum, provided it has always existed. The idea was that a thing could only move into an empty place, and that, in a plenum, there are no empty places. It might be contended, perhaps validly, that motion could never begin in a plenum, but it cannot be validly maintained that it could not occur at all. To the Greeks, however, it seemed that one must either acquiesce in the unchanging world of Parmenides, or admit the void.

  Now the arguments of Paimenides against not-being seemed logically irrefutable against the void, and they were reinforced by the discovery that where there seems to be nothing there is air. (This is an example of the confused mixture of logic and observation that was common.) We may put the Parmenidean position in this way: “You say there is the void; therefore the void is not nothing; therefore it is not the void.” It cannot be said that the atomists answered this argument; they merely proclaimed that they proposed to ignore it, on the ground that motion is a fact of experience, and therefore there must be a void, however difficult it may be to conceive.*

  Let us consider the subsequent history of this problem. The first and most obvious way of avoiding the logical difficulty is to distinguish between matter and space. According to this view, space is not nothing, but is of the nature of a receptacle, which may or may not have any given part filled with matter. Aristotle says (Physics, 208 b): “The theory that the void exists involves the existence of place: for one would define void as place bereft of body.” This view is set forth with the utmost explicitness by Newton, who asserts the existence of absolute space, and accordingly distinguishes absolute from relative motion. In the Copernican controversy, both sides (however little they may have realized it) were committed to this view, since they thought there was a difference between saying “the heavens revolve from east to west” and saying “the earth rotates from west to east.” If all motion is relative, these two statements are merely different ways of saying the same thing, like “John is the father of James” and “James is the son of John.” But if all motion is relative, and space is not substantial, we are left with the Parmenidean arguments against the void on our hands.

  Descartes, whose arguments are of just the same sort as those of early Greek philosophers, said that extension is the essence of matter, and therefore there is matter everywhere. For him, extension is an adjective, not a substantive; its substantive is matter, and without its substantive it cannot exist. Empty space» to him, is as absurd as happiness without a sentient being who is happy. Leibniz, on somewhat different grounds, also believed in the plenum, but he maintained that space is merely a system of relations. On this subject there was a famous controversy between him and Newton, the latter represented by Clarke. The controversy remained undecided until the time of Einstein, whose theory conclusively gave the victory to Leibniz.

  The modern physicist, while he still believes that matter is in some sense atomic, does not believe in empty space. Where there is not matter, there is still something, notably light-waves. Matter no longer has the lofty status that it acquired in philosophy through the arguments of Parmenides. It is not unchanging substance, but merely a way of grouping events. Some events belong to groups that can be regarded as material things; others, such as light-waves, do not. It is the events that are the stuff of the world, and each of them is of brief duration. In this respect, modern physics is on the side of Heraclitus as against Parmenides. But it was on the side of Parmenides until Einstein and quantum theory.

  As regards space, the modern view is that it is neither a substance, as Newton maintained, and as Leucippus and Democritus ought to have said, nor an adjective of extended bodies, as Descartes thought, but a system of relations, as Leibniz held. It is not by any means clear whether this view is compatible with the existence of the void. Perhaps, as a matter of abstract logic, it can be reconciled with the void. We might say that, between any two things, there is a certain greater or smaller distance, and that distance’does not imply the existence of intermediate things. Such a point of view, however, would be impossible to utilize in modern physics. Since Einstein, distance is between events, not between things, and involves time as well as space. It is essentially a causal conception, and in modern physics there is no action at a distance. All this, however, is based upon empirical rather than logical grounds. Moreover the modern view cannot be stated except in terms of differential equations, and would therefore be unintelligible to the philosophers of antiquity.

  It would seem, accordingly, that the logical development of the views of the atomists is the Newtonian theory of absolute space, which meets the difficulty of attributing reality to not-being. To this theory there are no logical objections. The chief objection is that absolute space is absolutely unknowable, and cannot therefore be a necessary hypothesis in an empirical science. The more practical objection is that physics can get on without it. But the world of the atomists remains logically possible, and is more akin to the actual world than is the world of any other of the ancient philosophers.

  Democritus worked out his theories in considerable detail, and some of the working out is interesting. Each atom, he said, was impenetrable and indivisible because it contained no void. When you use a knife to cut an apple, the knife has to find empty places where it can penetrate; if the apple contained no void, it would be infinitely hard and therefore physically indivisible. Each atom is internally unchanging, and in fact a Parmenidean One. The only things that atoms do are to move and hit each other, and sometimes to combine when they happen to have shapes that are
capable of interlocking. They are of all sorts of shapes; fire is composed of small spherical atoms, and so is the soul. Atoms, by collision, produce vortices, which generate bodies and ultimately worlds.* There are many worlds, some growing, some decaying; some may have no sun or moon, some several. Every world has a beginning and an end. A world may be destroyed by collision with a larger world. This cosmology may be summarized in Shelley’s words:

  Worlds on worlds are rolling ever

  From creation to decay,

  Like the bubbles on a river

  Sparkling, bursting, borne away.

  Life developed out of the primeval slime. There is some fire everywhere in a living body, but most in the brain or in the breast (On this, authorities differ.) Thought is a kind of motion, and is thus able to cause motion elsewhere. Perception and thought are physical processes. Perception is of two sorts, one of the senses, one of the understanding. Perceptions of the latter sort depend only on the things perceived, while those of the former sort depend also on our senses, and are therefore apt to be deceptive. Like Locke, Democritus held that such qualities as warmth, taste, and colour are not really in the object, but are due to our sense-organs, while such qualities as weight, density, and hardness are really in the object.

  Democritus was a thorough-going materialist; for him, as we have seen, the soul was composed of atoms, and thought was a physical process. There was no purpose in the universe; there were only atoms governed by mechanical laws. He disbelieved in popular religion, and he argued against the nous of Anaxagoras. In ethics he considered cheerfulness the goal of life, and regarded moderation and culture as the best means to it. He disliked everything violent and passionate; he disapproved of sex, because, he said, it involved the overwhelming of consciousness by pleasure. He valued friendship, but thought ill of women, and did not desire children, because their education interferes with philosophy. In all this, he was very like Jeremy Bentham; he was equally so in his love of what the Greeks called democracy.*