Aristotle the philosopher of nature

Aristotle the philosopher of nature David Furley 1 THE TREATISES ON NATURE The subject-matter of the present chapter is what Aristotle has to say about the natural world—the subject that in classical Greek is most accurately rendered as ta physika. But of course this includes many topics that would not now count as natural science—indeed Aristotle’s own book called Physics contains discussions that according to twentieth-century categories belong rather to philosophy or metaphysics. Book 1 criticizes the views of Aristotle’s predecessors on the first principles of natural objects, and defends his own view that they are three—matter, form, and privation. Book 2 analyses the kind of explanation that is to be expected of the natural philosopher, introducing the doctrine of ‘the four causes’. The third book deals with motion and change, and infinity; the fourth with place, void and time. The second quartet of books seems to form a separate entity —or perhaps two. Books 5, 6 and 8 are sometimes referred to by commentators under a separate title: On Change (kinêsis—the word may denote motion or change in general). Book 5 analyses concepts essential to the study of motion, book 6 deals with continuity, Book 8 argues for the eternity of motion and an eternal mover. Book 7 (part of which has been transmitted in two versions) perhaps contains a preliminary version of Book 8. In the traditional ordering of Aristotle’s works, Physics is followed by three theoretical treatises concerned with different aspects of the cosmos: On the Heavens, On Generation and Corruption, and Meteorologica. After a short essay On the Cosmos, generally and rightly held to be spurious, these are followed by a sequence of works on biology, which constitutes one fourth of the surviving Corpus Aristotelicum. First comes the treatise On the Soul (the principle of life), and a collection of related short essays concerning sensation, memory, sleep, dreams, etc., known as the Parva Naturalia. Then follow the three principal works of zoology: History of Animals (Zoological Researches would be a more appropriate modern title), Parts of Animals, and Generation of Animals. (The traditional Corpus contains also a number of works on the natural world now held to be spurious: On Colours, On Things Heard, Physiognomonics, On Plants, On Marvellous Things Heard, Mechanics, and Problems.) 2 ARISTOTLE’S SCIENTIFIC METHODS IN POSTERIOR ANALYTICS AND ELSEWHERE Before entering upon a discussion of Aristotle’s researches into the natural world, something must be said about the book in which he theorizes about scientific proof—the Posterior Analytics.¹ The book sets out a system of proof by syllogisms. We have scientific understanding of something, says Aristotle, ‘when we believe we know the cause (the aitia)² of the thing’s being the case—know that it is the cause of it—and that it could not be otherwise’ (1.2, 71b10–12). From premisses that are known to be true, the scientific theorist draws a conclusion that is then also known to be true because it follows necessarily from the premisses. If the argument is to qualify as part of a science (epistêmê), its premisses must have certain qualities: they must be ‘true and primitive and immediate and more familiar than and prior to and explanatory of the conclusion’ (1.2, 71b22–24, tr. Barnes). Now when one turns to the treatises in which Aristotle sets out his philosophy of nature (the treatises listed above in section 1), it is at once obvious that they do not even attempt to meet these conditions. They are, in general, inquiries, or the records of inquiries, rather than proofs. They do not confine themselves to necessary truths, which cannot be otherwise. In many cases, particularly in the biological works, they start from propositions based on observation. They do not proceed by syllogistic proofs alone. It is clear that we are dealing with two different phases in the presentation of science, and it is important that this be recognized if the reader is not to be disappointed by the apparent difference between the ideal set out in the Analytics and the more dialectical nature of the other treatises. The Posterior Analytics are generally held to describe the way in which a completed science should ideally be presented; the treatises on the natural world present the inquiries or researches that are preliminary to the finished product. ‘In a perfect Aristotelian world, the material gathered in the Corpus will be systematically presented; and the logical pattern will follow the pattern of the Posterior Analytics’ (Barnes [1.28], p. x). It should be added that the pattern of the Analytics evidently suits the mathematical sciences rather than biology, and Aristotle would be in difficulties if he confined his biology to the knowledge that could satisfy exacting demands for necessary truths and syllogistic proof. In the two treatises (Physics and Generation and Corruption) that deal with the concepts most fundamental to our study of the natural world, Aristotle uses methods that are based neither on the scientific syllogism nor directly on empirical studies of natural phenomena. Most typically, he starts from the views expressed by others—by his philosophical predecessors, or by educated and thoughtful ordinary men in general.³ For example, in book 4 of the Physics he analyses the concept of place. We should assume, he says (4.4, 210a32), whatever is rightly believed to belong to it essentially: i.e. that it is the first thing surrounding that whose place it is, that it is not a part of the thing, that it is neither bigger nor smaller than it; and that it is detachable from its content when the latter changes place. It is only because of locomotion, he adds, that we enquire about place. The object of the enquiry is to determine what place is in such a way that the problems are solved and the beliefs about its properties are shown to be true, and to show the reasons for the difficult problems about it. The first of Aristotle’s statements about place—namely that it ‘surrounds’ (periechein) its contents—turns out to be highly significant. This at once distinguishes ‘place’ from ‘space’; Aristotle’s place is a surface —the inner surface of a container that is in contact with the outer surface of the contents. Thus place is not measured by its volume, as space is, or as space would be measured if Aristotle allowed its existence. In fact, he denies it: it is not necessary, he claims, for the analysis of locomotion, because the concept of place will supply all that is needed (and he finds other problems with the idea of space). It follows, in Aristotle’s view, that there can be no such thing as the void. The void could only be an empty place: but place is a container, and a container is nothing if it contains nothing. When something changes place, its former place is occupied pari passu by something else, or else the former container collapses on to itself as an empty bag does. In this analysis there are no experiments, no measurements, and no observations other than those of ordinary everyday experience. What we have is a study of descriptions of motion, and of the assumptions underlying these descriptions. We have also an exhibition of the problems arising from alternative and incompatible descriptions in terms of space rather than place. There is a somewhat similar but more far-reaching conceptual analysis in book 1 of the Physics. It begins by asking: what are the principles of nature? That is to say, what are the things that are essential to the existence of any natural object? To find the principles, we have to start with what is familiar to us, because the principles themselves are not accessible directly to our minds, nor universally agreed. It is not principles that we are directly acquainted with, but the changing compounds of the natural world. After a criticism of the ideas of earlier philosophers of nature about the principles, Aristotle continues with reflections on our common notions about the essential features of change, since change is a necessary feature of everything in the sublunary natural world. Change takes place between opposites: things are said to change from hot to cold, for example, or from dry to wet, or from unmusical to musical. So opposites must be among the principles. But it is false to say that hot changes to cold: it is not the opposites themselves that change, but something that is characterized first by one opposite, then the other (or if not from one extreme to the other, from one position on the continuum between the two to another position in the direction of the other). What, then, is the ‘something’, the substratum, presupposed by such change? Aristotle’s answer is ‘matter’ (hylê). His concept of matter is one that would be thought of now as belonging to metaphysics rather than to physics. Matter is an abstraction: it is arrived at, in thought only, by stripping away from a physical object all the attributes that belong to its form. It never exists in separation from all attributes. The simplest kind of object with substantial existence in Aristotle’s hierarchy of existent things is a piece of one of the four elements: but any such piece is analysable in theory into matter and certain qualities that give it form. In the sublunary world, as opposed to the heavens, everything that exists is liable to change, from a quality to its opposite, from a given size to a larger or smaller one, or from being what it is to being something else (for example from being a table to being a heap of firewood, from being firewood to being smoke and ash, etc.). What underlies physical change is matter: matter has the potentiality for losing one form and taking on another. A favourite example of physical change in Aristotle’s works is the making of a piece of sculpture. An amount of bronze or stone is the matter: it has the potentiality for becoming an image of a man, and the sculptor gives it that form in actuality. But this is rather too static an analysis: at each stage of the process of making the statue, the material in its penultimate state is matter (potentiality) for the actuality of the next stage. Matter and form, and potentiality and actuality, are pairs of relative terms. The elements themselves, better named ‘the primary bodies’—earth, water, air, and fire—have the potentiality for changing into each other. For example water has the potentiality for vaporizing into ‘air’ or for solidifying into ‘earth’—the names themselves in Aristotelian usage each denote a range of solid, liquid, gaseous, and fiery substances.⁴ 3 ARISTOTLE’S WORLD PICTURE We shall begin with an outline of Aristotle’s picture of the natural world as a whole, contrasting it with others of the classical period, and continue with comments on his contribution to each of the major fields, from astronomy to biology. The general character of Aristotle’s interpretation of the natural world is determined primarily by two theses: that the cosmos had no beginning and will have no end in time, and that it is a finite whole that exhausts the contents of the universe. The first main point—that the cosmos is sempiternal—is argued in book 8 of the Physics. The first premiss is that there can be no time without change: change is necessary, if parts of time are to be distinguished from each other. But according to Aristotle’s analysis of change, there can be no first change, and correspondingly no last change. It follows that both change and time are eternal (Physics 8.1). Further argument (in Physics 8. 6) shows that if change is to be eternal, there must be both something eternal that causes change (we shall return to this all-important being in section 7), and something eternal in which this change occurs. This latter being is the ‘first heaven’, the sphere of the fixed stars. Since the rest of the cosmos is determined in its essentials by the motions of the heavens, the whole cosmic order is also eternal. These claims (defended, of course, by arguments to which this bare summary does no justice) distinguish Aristotle from all major philosophers of the classical period, with the possible exception of Heraclitus. Anaxagoras held that the cosmos emerged from a primitive mixture of all its contents; Empedocles that it grows from unity, passes through a period of plurality, and returns to unity, in repeating cycles; the Atomists argued for a plurality of cosmoi, each with a finite lifetime; Plato maintained that the single cosmos is indeed eternal, but he wrote (in the Timaeus) a description of its creation at a particular point in time, which Aristotle at least believed was to be taken literally; the Stoics returned to a cyclic theory. The second of these claims—that the universe is finite—follows from a set of prior assumptions and arguments. In Physics book 4, Aristotle argues that there can be no such thing as a vacuum anywhere in the universe, and hence that there cannot be an infinitely extended vacuum. What people mean when they talk about a vacuum or void, as Leucippus and Democritus did, is an empty place. But Aristotle produced arguments to show that there can be no such thing. The place of a thing is its container, or rather the inner boundaries of its container. According to our experience, when we try to empty a container, either the contents are replaced instantly by something else (usually air), or the container collapses upon itself. In either case we have no empty place. A place is always the place of something or other. It follows from this that there can be no void place within the cosmos, and it follows from Aristotle’s theory of the motions of the elements (which we shall examine shortly) that there can be no place outside the cosmos, since all of the body in the universe is concentrated in the cosmos. In order to show that the universe is finite, then, it remains to show that there cannot be an infinitely extended body or plurality of bodies. This Aristotle aims to do in On the Heavens 1.5–7. He begins with an argument concerned with the ‘first body’—i.e. the body of which the sphere of the fixed stars is composed (for which see section 5). Like most Greeks of the classical period Aristotle believed the earth to be stationary at the centre of the spherical heavens. The fact that it was stationary seemed to be given by experience: once that thesis was accepted, it followed that the heavenly bodies move around the earth. Before Aristotle’s time, it had been established that there was a difference in the motions of the heavenly bodies: the stars appear to move in concert without changing their relative positions, while the sun, moon, and five ‘wanderers’ (planêtai) move around the earth in orbits different from each other and from the ‘fixed’ stars. The appearance of the fixed stars suggests that they are placed on a sphere that rotates as a whole on its axis, with the earth at its centre. We observe that this sphere completes one revolution in a day. If it were infinite in radius, each radius drawn from the centre would sweep an infinitely large distance in every segment traversed. But that is impossible: it is not possible to traverse an infinite distance, since the infinite is ‘that of which there is always more beyond’ (Physics 3.6, 207a1). In dealing with the four sublunary elements—earth, water, air, and fire— Aristotle takes as given his theory of their natural places and natural motions. All earth tends to move towards a single centre, all fire to a single circumference, and the other two to intermediate positions. Consequently there cannot be any portion of the four elements, either simple or in compounds, outside the boundary of the sphere of the stars. But neither can there be any empty place outside this sphere, since, as Aristotle has argued, all place must be the place of something. Hence the universe (not merely the cosmos bounded by the starry sphere) is finite. 4 THE NATURAL MOTIONS OF THE ELEMENTS Aristotle’s theory of the elements is defended in detail in his On the Heavens; books 3 and 4 deal with the four elements that had become traditional since the time of Empedocles—earth, water, air, and fire—while books 1 and 2 introduce what Aristotle calls ‘the first element’ or ‘the first body’ and subsequent writers called ‘aether’, the element of which the heavens are composed. Observation of the natural world suggests a distinction between forced and natural motions: a stone can be thrown upwards, but falls downwards if not prevented; fire and hot vapours rise upwards unless confined by something above them. Aristotle systematizes these simple observations with the help of the geometrical picture of the cosmos described in the last section. ‘Downwards’ is defined as ‘in a straight line towards the centre of the universe’; ‘upwards’ is the contrary direction, away from the centre. These two rectilinear movements are contrasted with motion in a circle around the centre of the universe. The rectilinear motions are natural to the elements contained within the sphere of the heavens—commonly called the ‘sublunary’ elements, since the moon is the innermost of the heavenly bodies. These motions are defined according to the ‘natural place’ of each element. Each element has a natural tendency to seek its natural place, if displaced from it. Earth and water move naturally downwards, towards the centre; fire and air upwards. The tendency to move in these directions is what is meant by ‘weight’ and ‘lightness’ respectively—thus lightness is not a relative property but an absolute one. Earth has more weight than water, and fire has more lightness than air. It is important to note that Aristotle takes the centre, and therefore the elementary motions, to be defined by the spherical shape of the universe as a whole, not by the shape of the cosmos. Later philosophers abandoned Aristotle’s notion that the sphere of the stars has nothing whatever outside it, and posited an infinite volume of empty space around the cosmos. In such a cosmology no centre of the universe as such could be defined, and Aristotle’s theory of natural motion had to be changed. To deal with this problem, the Stoics made the highly significant claim that the body of the cosmos is naturally attracted towards its own centre. This theory of attraction began to make clear what Aristotle never elucidated: what is the cause of the natural motions of the elements? We shall discuss this problem later (section 7). 5 THE STRUCTURE OF THE HEAVENS The natural motions of the four sublunary elements were rectilinear. But the heavenly bodies move in circular orbits, carried around on the surfaces of rotating spheres (we shall describe the arrangement of the spheres in the next sections). But physical spheres must have physical body. So Aristotle is faced with the question: what are the heavenly spheres made of? They can hardly be made of any of the four elements which have rectilinear motions. The motion of the heavens, according to Aristotle’s view in the On the Heavens, requires us to posit a fifth element whose natural motion is not rectilinear but circular. Since he regards it as superior, in more than one sense, to the other four elements, he names it ‘the first body’. But although he made a technical term out of it, the idea of a special element in the heavens was not his alone, and others referred to it with the old word ‘aether’—originally used for the bright sky above the misty air. For convenience I shall adopt this term for Aristotle’s ‘first body’. We can distinguish more than one argument for the existence of aether.⁵ The main argument in Aristotle’s On the Heavens is the argument from motion that we have just described. A second argument is also found there: it may be called the argument from incorruptibility. Earth, water, air, and fire are perishable in that they are all liable to change into each other. But the heavens are eternal: they must therefore be made of a different element. This argument can be found, in rather disguised form, in Aristotle’s On the Heavens 1.3 (there is a very similar statement of it in Meteorologica 1.3). It is disguised in this sense. Aristotle first states the argument for the existence of what he calls ‘the first body’ from the need for a body endowed with natural circular motion. He then deduces that it must be ungenerated, indestructible, and unchangeable. His reasoning is that all generation takes place between opposites, opposites have opposed motions, and there is no opposite to circular motion (it is not clear why he dismisses the notion that clockwise has its opposite in anticlockwise—if we may use such modern terms). Hence, the body that moves in a circle is not liable to generation and destruction. He continues the chapter with some less technical thoughts about this element. These include the idea that ‘according to the records handed down from generation to generation, we find no trace of change either in the whole of the outermost heaven or in any of its proper parts’. Moreover, he says, the name ‘aether’ was given to the first body ‘by the ancients…choosing its title from the fact that it “runs always” (aei thein) and eternally’ (270b13–24). It is not, in other words, circular motion that is the primary characteristic of this element, but eternal motion. These ideas at least produce the materials out of which the incorruptibility argument for the existence of the fifth body can be constructed, and the etymology suggests that in Aristotle’s view this might have been the earliest argument for its existence. There are indications that Aristotle rather tentatively gave a role to aether in the sublunary world as well as in the heavens. Cicero knew something to this effect, from his acquaintance with some of the works of Aristotle that are now lost: He [sc. Aristotle] thinks there is a certain fifth nature, of which mind is made; for thinking, foreseeing, learning, teaching, making a discovery, holding so much in the memory—all these and more, loving, hating, feeling pain and joy—such things as these, he believes, do not belong to any one of the four elements. He introduces a fifth kind, without a name, and thus calls the mind itself ‘endelecheia’, using a new name—as it were, a certain continual, eternal motion. (Cicero Tusculan Disputations 1.10.22) It is hardly likely that Aristotle identified the mind with aether, but it is possible that at some time he wrote of the soul, or some of its faculties, as being based in an element different from the usual four. There is some confirmation of this in his own more cautious words: Now it is true that the power of all kinds of soul seems to have a connexion with a matter different from and more divine than the socalled elements; but as one soul differs from another in honour and dishonour, so also the nature of the corresponding matter differs. All have in their semen that which causes it to be productive; I mean what is called vital heat. This is not fire or any such power, but it is the breath included in the semen and the foam-like, and the natural principle in breath, being analogous to the element of the stars. (Aristotle Generation of Animals 2.3, 736b29–737a1) The evaluative strain in this quotation is significant. The extra element is called ‘divine’ and is associated with the ranking in ‘honour’ of the soul that is based on it—this refers, no doubt, to a scala naturae which puts man, the rational animal, at the top and grades the lower animals according to their faculties.⁶ Aether is not merely the element endowed with the natural faculty of moving in a circle, which is the main emphasis in the On the Heavens. It is also eternal, and therefore divine, and free from the corruption of the earthly elements. Aristotle was committed to a dualism as sharp as Plato’s distinction between the intelligible and unchanging Forms and the perceptible and perishable material world. The heavens are the realm of a matter that moves eternally in circles, is incorruptible, unmixed, divine. With the possible limited exception of the material base of the animal soul, everything in the cosmos inside the sphere of the moon—the sublunary world—is made of different materials, all of them rectilinear and therefore finite in motion, perishable, liable to mixture and interchange among themselves. This was a dualism that lasted, notoriously, until the time of Galileo and Kepler, when the telescope revealed the moon to be not so very different from the earth, and the idea of circular motion at last released its powerful grip on the astronomers’ imagination. 6 THE BORROWED ASTRONOMY Plato (said Sosigenes) set this problem for students of astronomy: ‘By the assumption of what uniform and ordered motions can the phenomena concerning the motions of the planets be saved?’ (Simplicius De caelo 488.21) Aristotle followed Plato in analysing the motions of the heavenly bodies entirely into circles with the earth as centre. The motions of the ‘fixed’ stars, during the time they are visible at night to an observer on the earth, are arcs of circles, and they are assumed to complete their circular paths in the daytime, when they are invisible. But the planetary bodies, including the sun and the moon, appear to ‘wander’ (in Greek, planân) with reference to the fixed stars in the course of a year. In fact, however, they do not wander, Plato had said; Aristotle agreed that their paths could be analysed as being circular, but adopted a much more complex account of the circles than Plato’s. The basis for his account of the heavens was the work of two contemporary astronomers: Eudoxus of Cnidos and Callippus of Cyzicus.⁷ They worked out what was basically a geometrical model of the paths of the heavenly bodies. Aristotle added what he considered to be necessary for a physical model (to be described in the next section). The essence of the geometrical model is as follows. The fixed stars are assumed to be set rigidly in the outermost sphere of the heavens, which turns at a constant speed about its north/south axis once a day. Inside the outermost sphere are seven sets of concentric spheres, one set for each of the five known planets and the sun and the moon. The innermost sphere of each set carries the planetary body on its equator (this applies to the geometrical account: the physical model is still more complex). The outermost sphere of each set moves on the same axis and with the same direction and speed as the sphere of the fixed stars. It carries with it the poles of a second sphere, concentric with the first, rotating about its own, different axis at its own constant speed. The axis of the second sphere is inclined to that of the first so that its equator, as it rotates, passes through the middle of the signs of the zodiac (i.e. along the ecliptic circle). The second sphere of each of the planetary bodies has the same orientation relative to the fixed stars and the same direction of rotation as each other; they differ in the time taken to complete a rotation. But the planetary bodies are observed to deviate from regular motion on the ecliptic circle: they do not keep to the same path. To account for the differences, Eudoxus posited a third and fourth sphere for each planet, nested inside the first two, rotating on different axes and completing their rotation in different times. The planet is assumed to lie on the equator of the fourth, innermost sphere. The third and fourth spheres are so arranged that the planet follows a path (relative to the ecliptic) known as a ‘hippopede’ or ‘horse-fetter’, roughly equivalent to a figure 8.⁸ All that is visible to the observer, of course, is the light of the heavenly bodies: the spheres are invisible. The visible heavenly bodies themselves do not move at all; they are carried around by the motion of the sphere in which they are set. The seven sets of spheres are nested inside each other, in the order Saturn, Jupiter, Mars, Venus, Mercury, sun, moon.⁹ In Eudoxus’ scheme, there are no eccentric spheres and no epicycles, as in later astronomical theories. Consequently it was assumed that all the heavenly bodies remain at a constant distance from the earth: it is a weakness in the system that it has no way of explaining differences in the brightness of the planets at different times. This, then, was the astronomical model taken over by Aristotle. He acknowledges his debt to the mathematicians, but there are numerous obscurities in his account which raise doubts about the depth of his understanding of contemporary astronomy.¹⁰ What is clear is that he constructed a physical description of the heavens, in which the spheres were not geometrical postulates but material bodies, and the most important element in this body of theory is his examination of the causes of the motions of the spheres. 7 FROM ASTRONOMY TO PHYSICS AND THEOLOGY The astronomical model, as we have seen, used the motion of the sphere of the fixed stars as the base on which the other motions were overlaid. For the construction of a physical theory, this created a difficulty concerning the motions of all the planetary bodies except the outermost one, since the sets of planetary spheres are implanted in each other. Jupiter’s set, to take an example, is inside the set of Saturn’s spheres. But in the astronomical model the motion of the innermost of Saturn’s spheres—the sphere that carries Saturn on its equator—is obviously not identical with that of the sphere of fixed stars; its function is precisely to justify Saturn’s deviation from that motion. To preserve the geometrician’s scheme, however, Jupiter’s outermost sphere must move with the motion of the fixed stars. Consequently the physical theory must return to this base, by interpolating a set of spheres whose motions cancel out the special motions of Saturn. Let S1, S2, S3, S4 be the spheres that explain Saturn’s motions; S4 is the one that carries Saturn. Then Aristotle postulates, inside S4, a sphere S−4, which rotates on the same axis and at the same speed as S4, but in the reverse direction. Its motion is thus identical with that of S3. He postulates S−3, and S−2, in similar fashion. Now S−2 has the same motion as S1—i.e. the motion of the fixed stars. The first of Jupiter’s spheres, J1, has its poles fixed inside the sphere S−2. For some reason, a complete set of spheres, starting from the motion of the fixed stars, is postulated for each planetary body. The point is this. The outermost sphere belonging to Jupiter, J1, moves with the motion of the fixed stars. But so does its outer neighbour, S−2. So one of these is redundant. The same applies to all of the inner planetary bodies. It is not clear why Aristotle did not economize in this way. In fact, Aristotle took over Callippus’ modifications of the Eudoxan system, and held to the thesis of a complete and separate set of spheres for each planetary body. They can be listed as follows (positive followed by counteracting spheres): Saturn 4 + 3 Jupiter 4 + 3 Mars 5 + 4 Mercury 5 + 4 Venus 5 + 4 Sun 5 + 4 Moon 5 No counteracting spheres are required for the moon, since there are no heavenly bodies beneath it; so the total is 55. It seems that the outermost sphere of Saturn is identical with the sphere of the fixed stars, which is not counted separately.¹¹ But before leaving the subject of the heavens, we must raise the question that from some points of view appears to be the most important of all: what is the cause of the motion of the spheres? Since Aristotle concludes that circular motion is natural to the element of which the heavenly spheres are made, it might seem that there is no further cause to be specified: it might be the case that it is just a fact of nature that this element moves in circles, unless something prevents it, and the position of the poles of each sphere and their relation to each other determines what particular circular orbit is traced out by each particular bit of the aetherial element. Since in On the Heavens he attacks Plato's theory that the heavens are moved by their soul, and is silent (in general) about the existence of an external mover, it is tempting to think that in the period when that work was put together Aristotle held a mechanical theory of the motions of the heavens.¹² The whole system of cosmic motions, both in the heavens and in the sublunary world, might then be held to work on the same mechanical principle—the natural self-motion of the five elements. This would fit well enough with one interpretation of Aristotle's well known definition of 'nature', in Physics 2.1, as an internal principle of motion and rest. But it can hardly be so simple. Change in general, including locomotion, is analysed by Aristotle as the actualization of a potency: he insists that there must be some kind of agent that is actual in the required sense, and something that is not yet but can become actual in this sense; and that these two must be distinct. They may be parts or aspects of the same substance, but they must be distinct from each other. The nearest to an example of a self-mover is an animal: what moves it is its soul, what is moved is its body. But he contrasts this example explicitly with the motions of the elements: the elements cannot be self-movers even in this sense, because if they were, they could (like animals) stop themselves as well as put themselves into motion. Aristotle never makes it entirely clear what causes the natural fall of earth or the natural rise of fire; but in the last chapters of Metaphysics 12 (Lambda) he introduces the external mover of the heavenly spheres. God is their mover, himself unmoved whether by himself or any other being. This Unmoved Mover is pure actuality, with no potentiality for internal change. As such, he is the guarantor of the eternity of the motions of the heavens. In the relation between mover and moved, the motion is often brought about in some way that necessitates a motion performed by the mover. for example an artist or craftsman produces something out of the available materials by doing something to them. The prime example of a motion that is not brought about in this way is one that is caused by the thought and desire of the moved object—that is to say, when the moved object conceives of the actuality represented by the mover as good and consequently desirable. This is, remarkably, the model chosen by Aristotle for the motions of the heavens. The model entails a degree of animism in his cosmology: the heavenly spheres, if they are to be capable of thought and desire, must possess souls. Aristotle presents his theology in a notably impressionistic way. It seems (in Metaphysics 12.8) that each of the fifty-five spheres must have its own mover; yet we are not told how such beings can be individuated, and in some of the few paragraphs devoted to this all-important topic it appears that a single unmoved mover is envisaged. At least it is clear that if there is a plurality it is an organized plurality: Aristotle ends the book with a quotation from Homer: ‘The rule of many is not good: let there be one ruler.’¹³ Aristotle’s cosmic deities are remarkably non-providential: their function in his system is to sustain the motions of the cosmos eternally. They have no hand in the creation of the cosmos, since it had no creation but has existed in its present form from all eternity; and they have apparently no thought for the welfare of any particular species or for the whole, except in so far as the eternal survival of the whole system and of all its natural kinds is a matter of concern. In the surviving works of Aristotle there is astonishingly little on this subject, which one might have expected to be crucial. In the theological chapters of Metaphysics 12 (Lambda), he speaks of God in the singular, but introduces plural gods as movers of the spheres without clarifying the change from singular to plural. He describes the activity of the ‘first mover’ in strikingly reverential words: On such a principle, then, depend the heavens and the world of nature. And its life is such as the best which we enjoy, and enjoy but for a short time. For it is ever in this state (which we cannot be) since its actuality is also pleasure. And thought in itself deals with that which is best in itself…. If, then, God is always in that good state in which we sometimes are, this compels our wonder; and if in a better, this compels it yet more. And God is in a better state. And life also belongs to God; for the actuality of thought is life, and God is that actuality. (Aristotle Metaphysics 12.7, 1072b14–29, tr. Ross) But the content of God’s thought is never described, and remains a matter of controversy.¹⁴ 8 MATTER AND ITS QUALITIES IN THE SUBLUNARY WORLD At the end of the fourth century, Democritus put forward the theory of atoms. All of the ‘being’ in the universe, in his view, took the form of unbreakably solid pieces of matter, invisibly small individually but capable of combining temporarily into compounds large enough to be perceived. The only other item in the universe, endowed with a kind of being but sometimes also contrasted with atoms and characterized as ‘not-being’, was void space—itself absolutely without any properties except spatial extension. All the objects in the familiar world perceived by us were composed of atoms with some quantity of void interspersed between them. The perceptible qualities of things were explained as the outcome of the number and shapes of the component atoms, the quantity of void between them, and their motions in the void. Plato, in his cosmological dialogue Timaeus, rejected this simple ‘bottom up’ type of explanation, although he did not entirely abandon the concept of atoms. In his theory, the beings primarily responsible for the characteristics of the physical world are the immaterial Forms, accessible to the mind rather than directly to the senses. Physical objects derive their properties from the Forms that they ‘partake in’ or ‘imitate’. The properties of perceptible bodies are, however, related to the nature of the particles which they contain. Plato describes the mathematical structure of particles of the four traditional elements, earth, water, air, and fire. The quality of heat, for example, is related to the sharply angled pyramidal shape of particles of fire. But Plato’s particle theory is different from Democritus’ atomism in that his particles are not described as having solidity or resistance. They may be regarded as a conceptual analysis of the qualities associated with them, rather than as results of a breakdown of a compound into material components. Aristotle’s theory was in more complete contrast with Democritus than Plato’s, in that he abandoned corpuscles altogether in favour of a continuous theory of matter. He himself analyses the argument which, he says, induced Democritus to introduce ‘indivisible magnitudes’ into his theory. It was a response to the paradoxes of the Eleatic Zeno, and went like this, in brief (De gen. et corr. 1.2, 316a11 ff.). Suppose that there are no indivisible magnitudes: then every magnitude would be divisible ad infinitum. Suppose such a division ad infinitum were completed: then one must be left either (a) with a collection of undivided magnitudes (which contradicts the hypothesis that every magnitude is divisible), or (b) with a collection of parts with no magnitude (which could never be put together to make a magnitude), or (c) with nothing at all. Hence, Democritus concluded, there must be indivisible magnitudes. Aristotle’s response was that every magnitude is indeed divisible every-where, but not everywhere simultaneously. Hence there are no indivisible magnitudes, but in dividing one never arrives at an infinite collection of simultaneous parts. It would be a comparatively easy business to describe his theory if he had made it clear what exactly composes his continuum. Difficulties arise because he fails to make clear whether or not we are to consider the continuum as being composed of ‘prime matter’, without any qualities beyond those of three-dimensional spatial extension and resistance, or as being invariably endowed with further qualities. There is no doubt that he adopted the four elements first clearly identified by Empedocles, and taken over by Plato: earth, water, air, and fire.¹⁵ He rejected Plato’s theory that the four differ from each other because of the mathematical shape of their particles: instead he allocated to each of them (in addition to natural motion, upwards or downwards) a pair of the primary qualities, hot, cold, dry, and wet. Thus earth is cold and dry, water cold and wet, air warm and wet, fire warm and dry. Unlike Empedocles, he held that that the elements change into each other by exchanging qualities. For example, evaporation is analysed as the replacement of water’s coldness by warmth. But water is not simply coldness and wetness: cold and wet are qualities that give form to a substratum: water is something that is cold and wet. The ‘something’ that underlies the qualities is barely described by Aristotle; hence there arises a controversy as to whether or not he had a conception of ‘prime matter’. His theory of elementary change does not require a stage at which there exists prime matter without any qualities: what changes into air, to continue with the example of evaporation, is water, and it changes directly, with no intermediate stage. But each of the four elements has three-dimensional extension and resistance, and these properties remain in place (in some sense, if not exactly) when a given quantity of water changes into air. If that is enough to constitute a theory of prime matter, then it seems undeniable that Aristotle held such a theory. But his account of change requires that there never exists an instance of prime matter without qualities. The four elements are given the familiar names of earth, water, air, and fire, but that is misleadingly simple. The element ‘earth’ gathers in everything that is solid, water everything that is fluid or pliable, air everything that is misty or gaseous. Fire is to some extent sui generis, and does not fit well into this scheme. 9 FOUR LEVELS OF MATERIAL BEING 1 The four elements (‘primary bodies’) 2 Homoiomerous bodies 3 Anhomoiomerous parts 4 Organisms The main point of this classification is to distinguish (2) from (3), and the distinction depends on whether the part (meros) has the same name as the whole. If we take a part of a substance such as blood or bone or skin, each of them has the same name as the whole: a bit of bone is bone, and so on. At the next level, the same is not true: a bit of a hand is not a hand (nor ‘hand’), nor a bit of a face a face (or ‘face’). ‘Anhomoiomerous’ means ‘having parts that are dissimilar’. The anhomoiomerous parts are made of the homoiomerous tissues: a hand is made of skin, bone, muscle, etc. This distinction serves only to distinguish level (3) from (4), not (1) from (2). Earth, water, air, and fire are homoiomerous. 10 THE FORMATION OF COMPOUNDS Out of the elements, the tissues: out of these, as matter, the whole of nature’s works. But though they are all out of these said elements as matter, in respect of their real being they are [determined] by their definition. This is always clearer in higher-level things, and in general in things that are for an end, like tools. It is clearer that a corpse is a man in name only; similarly, then, a dead man’s hand, too, is a hand in name only…; such things are less clear in the case of flesh and bone, still less in fire and water, because the final cause is least clear here, where matter predominates. …Such parts, then [sc. the simpler elements of organic compounds], can come-to-be by heat and cold…. But the complex parts composed of these—for example head, hand, foot —no one would believe to be composed in this way. Though cold and heat and motion are causes of bronze and silver’s coming-to-be, they are no longer the causes of a saw or a cup or a box. (Aristotle Meteorologica 4.12, excerpted) Aristotle’s anti-reductionist stance, in strong opposition to Democritus, is clearly announced in this last chapter of book 4 of his Meteorologica (which I take to be a genuine book serving as a bridge between his physical and biological works).¹⁶ He has given an account of the four simple bodies and their motions; he has shown how they combine to make the next layer of his hierarchy of materials—the ‘tissues’, or ‘the homoiomerous bodies’, to use his own technical term. But he wants to make it clear that this ‘bottom-up‘procedure is not the way to analyse the physical world. Material elements are the ingredients, but they do not make the natural compound. Empedocles and Democritus were wrong. Much more important than the material cause is what he designates here as the logos, which I have translated ‘definition’ in the passage above. We shall examine this again in the next section: for the moment, two points must be made. First, Aristotle’s claim is that to know what a thing really is is not just to know what it is made of, by taking it to bits, so to speak, nor just to trace the motions that its ingredients performed in composing it, but rather to know something about it as a present whole. In the case of an artifact, we shall want to know what it is for, the final cause; in the case of a living thing, we shall want to know what it does so as to survive and reproduce. So we know about this object (for example a saw), not when we discover what are the shapes, numbers, and dispositions of its component atoms or other material ingredients (although we shall want to know something about its components), but rather when we see that it is to cut wood and understand how its components enable it to do that. We know about this object (for example a frog) when we see where and how it gets a living and understand how its parts enable it to live and to reproduce. So much is an epistemological point: form, or definition (which puts form into words), takes priority over material ingredients for the purpose of knowledge. But this is true about knowledge just because the same priority operates in reality. Matter-in-motion, by itself, does not make a saw or a cup or a box, still less a head or a hand or a foot. The forms or kinds that exist in nature are the primary data. As causes of the production of individual members of species they take priority over the earth, water, air, and fire that are used in the production. It is form that dominates. How it dominates and operates as a cause is what we must examine. Aristotle’s theory of the roles of matter and form in the processes of nature bears a strong resemblance to Plato’s distinction in the Timaeus between Necessity and Mind. It is true that Plato locates the operation of these two causes in the creation of the physical world by the Craftsman God, whereas Aristotle uses them to explain the continuous cycles of coming-to-be and passing-away. But the function of the two causes is very much the same in both theories. Plato’s Craftsman copies the Forms in a material base; he makes the best possible copies, given the limitations imposed by the Necessity of the materials. In Aristotle’s theory, it is the forms themselves, without the designing mind of a Craftsman God, that shape and guide the potentialities of the four simple bodies and the material compounds formed out of them. For Aristotle, the materials represent Necessity in two guises. Materials with certain definite qualities are necessitated by the nature of the form they are to take on—a saw-blade must necessarily be of metal, not wood, and a bone must be made of something rigid, not liquid. They are also necessitating, in that they necessarily bring with them the whole set of their own properties, whether or not these are all necessitated by the forms. Thus the saw’s metal is necessarily liable to rust as well as being capable of being sharpened, the bone is necessarily fragile as well as rigid, if it is not to be too ponderous.¹⁷ Aristotle’s point against the Atomists is not that simple kinds of matter have no necessitating or causative properties, but that these properties alone cannot bring about the complex forms observed in nature. What they can bring about is described in the fourth book of Meteorologica, where he distinguishes four layers of complexity of natural objects, as we have seen. The point Aristotle makes in the quotation at the beginning of this section is that the necessitating properties of matter become less and less dominant with each step up through the layers. They have the greatest effect in the formation of the homoiomerous tissues from the elements. The active powers of heat and cold in the simple bodies work on the passive qualities of moisture and dryness to produce compounds that differ from each other by being, in different degrees, solidified, meltable, softenable by heat, softenable by water, flexible, squeezable, ductile, malleable, fissile, cuttable, viscous, compressible, combustible, and capable of giving off vapours (this is Aristotle’s list, in Meteorologica 4.8, 385a12–19). The nature of the homoiomerous bodies is determined by these properties, together with the degree of heaviness or lightness imported by the proportions of each of the simple bodies in their composition. Given the heating action of the sun, then, and the seasonal changes in that action brought about by the sun’s motion in the ecliptic circle, we may believe that the continuum of the four simple bodies must be so stirred up into qualitative interaction that many varieties of compound bodies may be formed withot the intervention of other causes. Even at this level, Aristotle’s theory is not reductionist: he did not hold that all these different qualities were ‘nothing but’ different degrees of hot, cold, dry, and wet, nor that the homoiomerous bodies are ‘nothing but’ earth, water, air, and fire in different proportions. They are all to be thought of as real features of the natural world, generated by the interactions of the simple bodies but not reducible to them. But even at this level, the generation of the complex out of the relatively simple is rarely caused solely by matter in motion. Homoiomerous tissues like oakwood, fishskin or cowhide are plainly enough not brought into being by the action of the sun and the natural properties of the four simple bodies, and nothing else. Aristotle says no more than that the causative action of form is less obvious at the lower stages, not that it is entirely absent. 11 THE FOUR CAUSES The four are listed in Physics 2.3: In one way, that out of which a thing comes to be and which persists, is called a cause, for example the bronze of a statue…. In another way, the form or the archetype, i.e. the definition of the essence and its genera, are called causes…. Again, the primary source of the change or rest…. Again, in the sense of end or that for the sake of which a thing is done…. These are traditionally referred to as the material, formal, efficient, and final causes. ‘The causes being four, it is the business of the student of nature to know about all of them, and if he refers his problems back to all of them, he will assign the “why?” in the way proper to his science’ (Physics 2.7, 22–25). But there are reasons for being hesitant about the word ‘cause’ as a translation of Aristotle’s aition or aitia. No single translation is adequate for all contexts—the bronze of which a statue is made, for instance, is not naturally called a ‘cause’ of the statue. The basic idea is to classify those items which are responsible for a thing’s being what it is. Closest to the modern ‘cause’ is the third in Aristotle’s list, the efficient cause—the sculptor, in the case of the statue. But the bronze of which it is made may well be cited as being responsible for some aspects of its nature; so also its form, and the end or purpose for which it was made. In the last chapter of the Meteorologica, quoted at the beginning of the last section, Aristotle insists that it is inadequate to mention material constituents alone as responsible for the nature of the compound: in anything but the simplest objects in the world, form is of much greater importance. But form alone is still insufficient: it is necessary to specify whatever it is that is responsible for giving this form to this matter—the efficient cause. And in many cases, for a full explanation we need to know the goal or end served by the possessor of this form in this matter. This simple schema dominates Aristotle’s studies of the natural world. It guides his inquiries, and gives shape to his presentation of the results. 12 ARISTOTLE’S ZOOLOGICAL WORKS The major works are Parts of Animals (PA), Generation of Animals (GA), and History of Animals. The first of these provides two introductions to zoological studies. PA 1. 5 is a fluently written and rather elementary ‘protreptic’, urging students not to be contemptuous of biology as opposed to ‘higher’ studies such as metaphysics or astronomy, which deal with eternal rather than perishable things. In the realm of biology, we have the advantage of being closer to the subject matter, and are therefore better able to study it. Moreover, the philosophical mind will find great satisfaction in discovering and analysing the causes at work in plants and animals, where Nature offers much that is beautiful to the discriminating eye. PA 1.1 is a discussion of causes, and above all a defence of the view that the final cause is most prominent in the works of nature. Lacking a theory of the evolution of species, Aristotle treats as the starting point for biology the form of the grown specimen—the adult horse or man, the full-grown oak tree. This is in opposition to those who started from the material elements—for example, Democritus. The first step is to understand the mode of life of the animal, and to observe what it needs for survival and for reproduction. These are the two essentials for understanding structure and behaviour. Each animal exists in a particular kind of environment, and the nature of the environment determines what will be good for the animal’s survival and reproductive capacity. The student of nature, therefore, will observe the animal and its parts, and decide first what contribution each part makes to survival and reproductive capacity. This is the ‘cause’ for the sake of which the part exists and has the structure that it is observed to have. The student will understand the nature of the animal when these causes are understood. Aristotle uses the word ‘cause’ (aitia) in its usual Greek sense, as that which is responsible for the phenomenon to be explained. But he does not mean to imply that the parts of animals are caused to grow (in our sense of ‘caused’) by capacities that lie in the future: the hooked beak of the (individual) hawk is not caused by its capacity, when grown, to tear up the flesh of its prey. The key to Aristotle’s teleology, in the biological realm, is the identity of the form of the (male) parent and the offspring.¹⁸ The parental hawk (to continue the example) survived to produce offspring just because its beak was of a kind well adapted to its mode of life; such a beak was an essential attribute of the form of the hawk; and this form is transmitted to the offspring. The final cause—that ‘for the sake of which’ the part of the creature exists—is thus subsumed into the efficient cause. The semen of the parent carries the form of the parent and transmits it, as efficient cause in the process of generation, to the offspring. The mechanism by which this transmission of form is achieved is described in detail in Generation of Animals. Aristotle dismisses the theory that the semen is drawn from all the parts of the parent’s body (pangenesis). That theory, which is set out in the surviving Hippocratic treatise On Seed, and was probably also defended by Democritus, was based on the resemblances of children to their parents. Aristotle argues that this proves too much or too little. Children resemble their parents in characteristics such as their manner of movement which is not determined by physical structure. Moreover children sometimes resemble grandparents or other family members, rather than parents. His own theory depends on his metaphysical distinction between form and matter. The matter of the embryo is provided by the mother, the form by the father. The semen carries in it the ‘movements’ that will cause the parts of the embryo to grow in the proper order and form. These movements are not simply instructions, nor an abstract design or formula: they are derived from the soul of the adult parent, and they are embodied in a material substance carried in the semen, called pneuma. Pneuma, is a concept that plays a large part in Greek physiology, from the earliest times, when it is equated more or less exactly with the breath of life. But Aristotle’s use of the concept is ill defined. He speaks of the ‘connate pneuma’; it is clearly necessary for life, and is especially associated with the faculties of soul such as sensation and movement. It carries also the idea of vital heat. But he does not give it the precise and detailed description that forms an important part of Stoic theory, and he does not explain its relation to the four material elements. There is a single mysterious hint (GA 2.3, 737a1, mentioned above, in section 5) that it is ‘analogous to the element of the stars’.¹⁹ The History of Animals, in ten books, has sometimes been taken to attempt a classification of animals—not by the process of ‘dichotomizing’ (dividing genera progressively by two into narrower classes) practised by Plato but rejected by Aristotle—but by more complicated methods. Recent researches, however, have shown that that the motivation of these treatises is rather to examine the differentiae of animals (for example the shape and size of legs, the apparatus of the senses, the modes of protection) and to relate them to the needs of the animal to get food, to ward off predators, and to bring up the next generation.²⁰ Aristotle uses the terms genos and eidos, which became the standard words for what later biologists denote by ‘genus’ and ‘species’. But it is clear from examination of the texts that genos in Aristotle can denote classes of varying degrees of generality, and eidos is not always subordinate to genos. What Aristotle seeks to do is to identify the kinds of animals there are, as defined by their mode of life in their environment, and to present comparative studies of the structure and organization of their parts as they are adapted to their function. Hence the supreme importance of the final cause. The biologist above all seeks to explain the connection between each of the characteristic actions of each animal kind, and the structure of the parts of the body that enable the animal to perform these actions. 13 PSYCHE Psyche is usually translated by the English word ‘soul’, and it is convenient to use the word in spite of its misleading modern connotations.²¹ Aristotle treats the psyche as the defining principle of life: the four material elements have no psyche, in spite of their natural tendency to seek for their natural place in the cosmos; compounds of the elements have no psyche, unless they possess at least the faculties of nutrition and reproduction. Aristotle constructs a scala naturae in which each higher step of the ladder is distinguished by the addition of further faculties of the soul. Plants and animals have the basic faculties of nutrition and reproduction; in addition to these, animals have sensation, although not all of them have all of the five senses; some animals, but not all of them, have also the capacity to move themselves; man has all of the animal faculties, with the addition of imagination (phantasia) and reason, which are also shared, in some small degree, by the higher animals. There is thus an ascending order of plant and animal species to be found in the world. This is not, however, an order produced by evolutionary processes: on the contrary, all of the species now in existence have always existed, in Aristotle’s view, and will continue to exist. We will discuss the relations between the species briefly in the next section. It is, of course, a crucial ingredient of Aristotle’s theory that the soul is not an entity separate from the body, nor indeed separable in any way except by abstraction in thought (there could be no transmigration of souls in his theory). The soul is ‘the first actuality of a natural body that is possessed of organs’ (On the Soul 2.1, 412b5). If a body is to have soul, it must have the organs that give it the potentiality of carrying out some of the functions of life. The soul is described as the first actuality because it is not necessary for the functions of the living body to be in action to qualify the body as ‘ensouled’. The eyes of a corpse or a statue are not alive, but the eyes of a sleeper are alive although they are not seeing. The soul is a state of readiness, in bodily organs, to perform their function. It can thus be described as a second potentiality, as well as a first actuality. The conception of an ascending order among living species, with the stages defined by the number and complexity of functions capable of being performed by the plant or animal, gives Aristotle the conceptual apparatus for working out a comprehensive classification of species. There is indeed some evidence that such a classification was a goal of his biological work, but it is not achieved in his surviving writings, where he is concerned above all, it appears, with understanding the differences between animals, and especially with putting the differences into relation with the organic parts.²² It has been remarked, too, that many of the ingredients of a theory of evolution of species are foreshadowed in his theory, but he was firmly against such an idea, as we have observed. This is not the place for a lengthy assessment of Aristotle’s achievement in biology, but a few points may be mentioned. He was handicapped by his belief, inherited from some earlier physiologists (against the view of Plato), that the heart, rather than the brain, is the seat of the sensitive soul.²³ The nerves had not yet been identified as such, and the blood was taken to be the vehicle for the transmission of messages from sense organs to the centre and vice versa. Blood was thought to be, or to contain, food for the tissues of the body—the circulation of the blood was not, of course, discovered for many centuries after Aristotle. He took respiration to be a way of moderating the natural heat of the body of animals with blood in their system, although he had a use for the concept of pneuma. or breath, which has been mentioned in section 5. The Generation of Animals contains a detailed study of the reproduction of many species. Aristotle did not understand the contribution of the female of the species to the reproductive process: in his theory semen is the vehicle that conveys the formal structure of parent to offspring, while the female contributes only the material constituents of the embryo, and (in some species) a protective site for its development. But there are remarkable insights in his analysis of the function and structure of semen. It contains both the formal and the efficient cause of the offspring: it contains in potentiality the specific form and some, at least, of the individual characteristics that will be passed on from the parent, and it contains also the ‘instructions’ for the motions needed to embody these in the embryo. The transmission takes place not by some crude exchange of materials, but in the form of ‘encoded’ messages. An odd feature of his ‘embryology’ is his continuing belief in the spontaneous generation of members of some species. Some creatures (testacea) originate from sea water; some plants (for example mistletoe) and animals (grubs) from putrefying matter. What is supplied from sources other than parents in these cases is pneuma, which is the material vehicle of life, and warmth. So much is perhaps not hard to understand: there is more difficulty in understanding how matter and warmth alone can supply the form, which in the case of sexual generation requires the subtle and complex contributions of the semen.²⁴ 14 THE UNITY OF THE COSMOS Plato’s cosmos, as described in the Timaeus, was itself an organism—a zôon or animal; Aristotle never talks of the whole cosmos in such terms. He does, however, in various ways and from time to time indicate clearly enough that he regards the cosmos as being appropriately named: the word carries with it the idea of good order. We must consider also in which of two ways the nature of the whole contains the good or the highest good, whether as something separate and by itself, or as the order (taxis) of the parts. Probably in both ways, as an army does. For the good is found both in the order and in the leader, and more in the latter; for he does not depend on the order but it depends on him. And all things are ordered together somehow, but not all alike—fishes and fowls and plants—and they are not so disposed that nothing has to do with another, but they are connected. For all are organised together with regard to a single thing. (Aristotle Metaphysics 12.10, 1075a11–19, tr. Ross, slightly adapted) In the context it would seem that Aristotle draws an analogy between the commander of an army and the supreme deity in command of the cosmos— perhaps the mover of the sphere of the fixed stars, or perhaps ‘the divine’ in a collective sense, meaning all of the movers of the spheres. The good that they achieve is the eternity of the cosmic order. That is to say, they ensure directly the eternal continuity of the motions of all the heavenly spheres, and hence the eternal interchange between contraries in the sublunary world, and the eternal continuance of all living species. Aristotle repeats one brief sentence many times, in various contexts: ‘nature does nothing without purpose’ (matên, sometimes translated ‘randomly’ or ‘in vain’). This is a notoriously puzzling claim: there seems to be no room in Aristotle’s theory for a single personified ‘Nature’, acting purposively like a rational being. Each natural thing has its own nature, and some of the effects of the nature of a thing are purposive only in a very loose sense, if at all. The sentence seems to be a summing up of the manner of biological processes; it does not carry us far towards an understanding of the order of the cosmos as a whole. There is a striking statement in the Politics: The viviparous species have sustenance for their offspring inside themselves for a certain period, the substance called milk. So that clearly we must suppose that nature also provides for them in a similar way when grown up, and that plants exist for the sake of animals and the other animals for the good of man, the domestic species both for his service and for his food, and if not all at all events most of the wild ones for the sake of his food and of his supplies of other kinds, in order that they may furnish him both with clothing and with other appliances. If therefore nature makes nothing without purpose or in vain, it follows that nature has made all the animals for the sake of men. (Aristotle Politics 1.3, 1256b13 ff.) This is a claim that sounds more like Stoicism than Aristotelianism. In the zoological treatises, animals are described in a more autonomous fashion; it is not asserted that the function of any characteristic of oxen, for instance, is to supply beef or leather. Man is indeed the ‘highest’ of the animals, because man shares the divine capability of reason. But the function of the parts of animals is not, apparently, to provide for man, but to provide for the continued life of their own species. What Aristotle has in mind is that we can observe a ‘rightness’ in the constituents of the cosmos and their modes of behaviour. It manifests itself in different ways. In the case of the elements, it consists in their natural motions—towards, away from, or around the centre. In the heavenly spheres, it consists in their positions and in the regularity of their motions. In the case of the sun, it shows itself in the daily and annual cycles of light and darkness, summer and winter, which have their effects on the mode of life and generation of biological species. The complement of these features of the cosmic spheres is that each species has its ‘niche’ in the world. The species did not in any sense find their niche, or grow to fill a previous vacancy: it just is (Aristotle thought) an observable fact that the physical cosmos provides variously characterized environments, and the living species have just those features that enable them to take advantage of them. Aristotle thus differs both from Democritus and from Plato. He differs from Plato’s Timaeus, as we have observed, in denying that the cosmos is the work of a purposive Creator. But he differs even more from Democritus, in denying that the world comes about through accident or material ‘necessity’. The cosmos just is as it is. It is like a well disciplined army, commanded by a good and effective General who keeps his troops up to the mark in performing their various traditional tasks. NOTES 1 See Barnes [1.28]; for more detailed discussion of Posterior Analytics, see chapter 2, below. 2 See section 12 for discussion of the translation of aitia as ‘cause’. 3 On this subject, see especially Owen [1.72, §8]. 4 This subject is continued in section 8. 5 One argument is derived from Plato’s Timaeus, although it was in fact used neither by Plato nor by Aristotle, and those who use the argument do not, it seems, think of aether as the element of the heavens. The argument is that there are five regular solids, and so there should be a fifth element corresponding to the dodecahedron, which was assigned by Plato to the shape of ‘the whole’. This argument is found in the pseudo-Platonic Epinomis (981b ff.) and apparently in Plato’s sucessor Xenocrates (fr. 53 Heinze, from Simplicius). 6 See section 13. 7 Aristotle acknowledges his debt to these two in Metaphysics 12.8. 8 The third and fourth spheres enable the model to accommodate the retrogradation of planets. But Aristotle is quite vague about the details of these motions, being content, apparently, to leave them to the mathematicians. 9 Later Greek astronomers put Venus and Mercury between the moon and the sun. 10 For comparison of Aristotle’s description with astronomical theory, see [1. 63] Neugebauer, History Part II, pp. 675–89, with diagrams in Part III, pp. 1357–8. 11 At 1074a12–14, Aristotle says that if the extra spheres added by Callippus to the sun and the moon are removed, the total should be 47. But something has gone wrong with the text or the calculation. If Aristotle states the condition correctly, the number should be 49. Another interesting puzzle about the numbers may be mentioned at this stage; it was first raised, so far as I know, by Norwood Russell Hanson [1. 87]. It turns on the question whether the axis on which each sphere turns should be regarded as an axle, with a certain thickness in diameter, or as a geometrical line. If it is an axle, and is fixed at its ends in the surface of its outer neighbour, then when its poles coincide with those of the outer neighbour it should rotate along with that neighbour. Thus we have a problem at the junction between two planetary sets. To take the example used above, S−2, which has the rotation of the fixed stars, must impart its own rotation to the axle of J1, and since J1 rotates about its own axle with the motion of the fixed stars, the sphere J1 will be rotating with twice that rotation, i.e. in 12 hours. The first sphere of Mars will rotate in 6 hours, the first of Venus in 3 hours, and so on. The solution to this is simply to treat the axis of each sphere as a geometric construction and its poles as dimensionless points. This is consistent with the physical nature of the spheres themselves, and abolishes the consequence of a double rotation. The points of contact do not rotate, although of course they are carried around with the surface in which they are located whenever they do not coincide with the poles of the superior sphere. 12 For a clear discussion of the problems about motion in On the Heavens, see the introduction to [1.14] Guthrie. 13 Iliad 2.204; Metaphysics 12.10, 1076a5. The problems connected with the unmoved mover or movers of the spheres has of course been very much discussed. Some notable examples: [1.8] Ross, pp. 94–102; [1.14] Guthrie, introduction; [1.85] Merlan. 14 See recent discussions in [1.86] Norman, [1.83] DeFilippo, and [1.74] Waterlow. 15 See section 4, above. 16 See the discussion of this book in [1.62] Furley, chapter 12. Its authenticity has been, and still is, doubted by some scholars. 17 See [1.80] Sorabji, and [1.77] Cooper. 18 This is well explained in [1.81] Woodfield. 19 The fullest account of pneuma is found in De motu animalium. See [1.9] Nussbaum, especially pp. 143–64. 20 See especially [1.18] Balme, Introduction to books 7–10, p.17; [1.90] Pellegrin, and [1.64] Lloyd ch.1 and ch.12. 21 See also chapter 3, below. 22 See [1.64] Lloyd, ch.1 and ch.12. 23 See below, chapter 10, for Galen’s refutation of Aristotle’s view. 24 There is a considerable recent literature on Aristotle’s theory of spontaneous generation. See, for example, [1.89] Balme, and [1.90] Lennox. BIBLIOGRAPHY ITEMS RELEVANT TO CHAPTERS 1–4 1.1 The edition of the Greek text of Aristotle, to which reference is standardly made by page, column and line numbers: Aristotelis Opera, ed. I.Bekker, 5 vols (Berlin, 1831–70). 1.2 Index Aristotelicus [Greek word index], ed. H.Bonitz (Berlin, 1870). Complete English translation 1.3 Aristotle: The Revised Oxford Translation, ed. Jonathan Barnes (Bollingen series), (Princeton, NJ, Princeton University Press, 1984). Greek texts with English commentary—some notable editions 1.4 Prior and Posterior Analytics, ed. W.D.Ross (Oxford, Clarendon Press, 1949). 1.5 Physics, ed. W.D.Ross (Oxford, Clarendon Press, 1936). 1.6 Aristotle on Coming-to-be and Passing-away (De generatione et corruptione), ed. H.Joachim (Oxford, Clarendon Press, 1922). 1.7 De anima, ed. R.D.Hicks (Cambridge, Cambridge University Press, 1907). 1.8 Metaphysics, ed. W.D.Ross (Oxford, Clarendon Press, 1924). 1.9 De motu animalium, ed. M.Nussbaum (Princeton, NJ, Princeton University Press, 1978). Greek texts with English translation (Leob Classical Library, Harvard University Press) 1.10 The Categories, On Interpretation ed. H.P.Cooke; Prior Analytics, ed. H.Tredennick (1938). 1.11 Posterior Analytics, ed. H.Tredennick; Topics, ed. E.S.Forster (1938). 1.12 On Sophistical Refutations, On Coming-to-be and Passing-away, ed. E.S. Forster; On the Cosmos, ed. D.J.Furley (1955). 1.13 Physics, ed. P.H.Wickstead and F.M.Cornford (1929–34). 1.14 On the Heavens, ed. W.K.C.Guthrie (1953). 1.15 Meteorologica, ed. H.D.P.Lee (1952). 1.16 On the Soul and Parva naturalia, ed. W.S.Hett (1936). 1.17 Generation of Animals, ed. A.L.Peck (1943). 1.18 Historia animalium books 1–4 and 5–8, ed. A.L.Peck (1965 and 1970); books 7–10, ed. D.M.Balme (1991). 1.19 Parts of Animals, ed. A.L.Peck, with Movement of Animals, ed. E.S.Forster (1937). 1.20 Minor Works, ed. W.H.Hett (1936). 1.21 Problems, ed. W.H.Hett (1926). 1.22 Metaphysics, ed. H.Tredennick, with Oeconomica and Magna Moralia, ed. G.C.Armstrong (1993). 1.23 Nicomachean Ethics, ed. H.Rackham (1926). 1.24 Athenian Constitution, Eudemian Ethics, and Virtues and Vices, ed. H. Rackham (1935). 1.25 Politics, ed. H.Rackham (1932). 1.26 Rhetoric, ed. J.H.Freese (1926). English translations of separate works, with commentary, in the Clarendon Aristotle series (Oxford University Press) 1.27 Categories and De interpretatione, by J.L.Ackrill (1963). 1.28 Posterior Analytics, by Jonathan Barnes (1975). 1.29 Topics, books 1 and 8, by R.Smith (1994). 1.30 Physics, books I and II, by W.Charlton (1970); books III and IV, by Edward Hussey (1983). 1.31 De generatione et corruptione, by C.J.F.Williams (1982). 1.32 De anima, books II and III, by D.W.Hamlyn (1968). 1.33 De partibus animalium I and De generatione animalium I, by D.M.Balme (1972). 1.34 Metaphysics, books Gamma, Delta, and Epsilon, by C.Kirwan (1971); books Zeta and Eta, by D.Bostock; books M and N, by J.Annas (1976). 1.35 Eudemian Ethics, books 1, 2, and 8, by M.Woods. 1.36 Politics 1 and 2, by T.J.Saunders (1995); 3 and 4, by R.Robinson (1962); 5 and 6, by D.Keyt (1999); 7 and 8, by R.Kraut (1997). Bibliographies Recent bibliographies in [1.39] The Cambridge Companion to Aristotle (1995); also Barnes, Schofield and Sorabji [1.53]. General Introductions to Aristotle 1.37 Ackrill, J.L., Aristotle the Philosopher (Oxford, Oxford University Press, 1981). 1.38 Barnes, J., Aristotle (Oxford, Oxford University Press, 1982). 1.39 Barnes, J., ed., The Cambridge Companion to Aristotle (Cambridge, Cambridge University Press, 1995) (contains an extensive bibliography). 1.40 W.K.C.Guthrie, A History of Greek Philosophy, vol. VI: Aristotle: An Encounter (Cambridge, Cambridge University Press, 1981). 1.41 W.D.Ross, Aristotle (London, Methuen, 1923). Proceedings of the Symposium Aristotelicum 1.41a Aristotle and Plato in the Mid-Fourth Century, ed. I.During and G.E.L. Owen, Göteborg, 1960). 1.42 Aristote et les problèmes de méthode, ed. S.Mansion (Louvain, Publications Universitaires, 1961). 1.43 Aristotle on Dialectic: The Topics, ed. G.E.L.Owen (Oxford, Oxford University Press, 1968). 1.44 Naturphilosophie bei Aristoteles und Theophrast, ed. I.Düring (Heidelberg, Lothar Stiehm, 1969). 1.45 Untersuchungen zur Eudemischen Ethik, ed. P.Moraux and D.Harlfinger (Berlin, De Gruyter, 1971). 1.46 Aristotle on the Mind and the Senses, ed. G.E.R.Lloyd and G.E.L.Owen (Cambridge, Cambridge University Press, 1978). 1.47 Etudes sur la, Métaphysique d’Aristote, ed. P.Aubenque (Paris, 1979). 1.48 Aristotle on Science: the ‘Posterior Analytics’, ed. E.Berti (Padua, Antenore, 1981). 1.49 Zweifelhaftes im Corpus Aristotelicum: Studien in einigen Dubia, ed. Paul Moraux and Jurgen Wiesner (Berlin and New York: De Gruyter, 1983). 1.50 Mathematics and Metaphysics in Aristotle, ed. Andreas Graeser (Bern/ Stuttgart, Paul Haupt, 1987). 1.51 Aristoteles’ ‘Politik’, ed. Günther Patzig (Göttingen: Vandenhoecht and Ruprecht, 1990). 1.52 Aristotle’s Rhetoric, ed. D.J.Furley and A.Nehamas (Princeton, NJ, Princeton University Press, 1994). Other collections of essays by various authors 1.53 Barnes, Jonathan, Schofield, Malcolm, and Sorabji, Richard (eds), Articles on Aristotle: vol. 1 Science; vol. 2 Ethics and Politics; vol. 3 Metaphysics; vol. 4 Psychology and Aesthetics (London, Duckworth, 1975). 1.54 Seeck, Gustav Adolf (ed.), Die Naturphilosophie des Aristoteles (Darmstadt, Wissenschaftliche Buchgesellschaft, 1975). 1.55 Gotthelf, Allan (ed.), Aristotle on Nature and Living Things: Philosophical and Historical Studies (Bristol Classical Press, and Mathesis Publications, Pittsburgh, 1985. 1.56 Gotthelf, Allan, and Lennox, James G. (eds), Philosophical Issues in Aristotle’s Biology (Cambridge, Cambridge University Press, 1987). 1.57 Matthen, Mohan (ed.), Aristotle Today: Essays on Aristotle’s Ideal of Science (Edmonton, Alberta, Academic Printing and Publishing, 1987). 1.58 Devereux, Daniel, and Pellegrin, Pierre (eds), Biologie, Logique et Métaphysique chez Aristote (Paris, CNRS, 1990). 1.59 Judson, Lindsay (ed.), Aristotle’s Physics: A Collection of Essays (Oxford, Clarendon Press, 1991). BIBLIOGRAPHY FOR CHAPTER 1 General works on Greek science and philosophy of nature 1.60 Sambursky, S., The Physical World of the Greeks (London, Routledge and Kegan Paul, 1956). 1.61 Dicks, D.R., Early Greek Astronomy to Aristotle (London, Thames and Hudson, 1970). 1.62 Furley, David, Cosmic Problems (Cambridge, Cambridge University Press, 1989). 1.63 Neugebauer, O., A History of Ancient Mathematical Astronomy (Berlin/ Heidelberg/NY, Springer Verlag 1975). 1.64 Lloyd, G.E.R., Methods and Problems in Greek Science: Selected Papers (Cambridge University Press, 1991). 1.65 Sorabji, Richard, Time, Creation, and the Continuum: Theories in Antiquity and the Early Middle Ages (London, Duckworth, 1983). 1.66 ——Matter, Space, and Motion: Theories in Antiquity and their Sequel (London, Duckworth, 1988). General studies on Aristotle’s philosophy of nature 1.67 Aristotle Today: Essays on Aristotle’s ideal of science, ed. Mohan Matthen (Edmonton, Academic Printing and Publishing, 1989). 1.68 Barnes, Schofield and Sorabji [1.53], vol 1, Science. 1.69 Barnes [1.39] The Cambridge Companion to Aristotle. [With good bibliography.] 1.70 Graham, Daniel W., Aristotle’s Two Systems (Oxford, Oxford University Press, 1987). 1.71 Mansion, Augustin, Introduction à la physique aristotélicienne, 2nd edn (Louvain, 1946). 1.72 Owen, G.E.L., Collected Papers in Greek Philosophy, ed. Martha Nussbaum (Ithaca, NY, Cornell University Press, 1986). Especially no. 8 ‘Aristotle: method, physics and cosmology’ and no. 10 ‘Tithenai ta phainomena’. 1.73 Solmsen, Friedrich, Aristotle’s System of the Physical World (Ithaca, NY, Cornell University Press, 1960). 1.74 Waterlow, S., Nature, Change, and Agency in Aristotle’s Physics (Oxford, Clarendon Press, 1982). Causation 1.75 Annas, J., ‘Aristotle on inefficient causes’, Philosophical Quarterly 32 (1982), 319. 1.76 Balme, D.M., ‘Teleology and necessity’, in Gotthelf and Lennox, [1.56], 275– 86. 1.77 Cooper, John M., ‘Hypothetical necessity and natural teleology’, in Gotthelf and Lennox [1.56], 243–74. 1.78 Gotthelf, Allan, ‘Aristotle’s conception of final causality’, in Gotthelf and Lennox [1.56], 204–42. 1.79 Lennox, James, ‘Teleology, chance, and Aristotle’s theory of spontaneous generation’, Journal of the History of Philosophy, 20 (1982), 219–38. 1.80 Sorabji, Richard, Necessity, Cause and Blame: Perspectives on Aristotle’s Theory (London, Duckworth; Ithaca, NY, Cornell University Press, 1980). 1.81 Woodfield, Andrew, Teleology (Cambridge, Cambridge University Press, 1976). Motion and theology 1.82 Bogen, James, and McGuire, J.E., ‘Aristotle’s great clock: necessity, possibility and the motion of the cosmos in De caelo 1.12’, Philosophy Research Archives 12 (1986–87), 387–448. 1.83 DeFilippo, Joseph, ‘Aristotle’s identification of the prime mover as god’, Classical Quarterly 44 (1994), 393–409. 1.84 Kahn, C., ‘The place of the prime mover in Aristotle’s teleology’, in Gotthelf [1.55]. 1.85 Merlan, P., ‘Aristotle’s unmoved movers’, Traditio 4 (1946), 1–30. 1.86 Norman, Richard, ‘Aristotle’s Philosopher-God’, Phronesis 14 (1969), 63. Repr. in Barnes, Schofield and Sorabji [1.53], vol. 4, 183–205. 1.87 Hanson, N.R., ‘On counting Aristotle’s spheres’, Scientia 98 (1963), 223–32. Matter and elements 1.88 Moraux, Paul, ‘Quinta essentia’, in Pauly-Wissowa, Realencyclopädie 24, 1171–226. Biology 1.89 Balme, D.M., ‘Development of biology in Aristotle and Theophrastus: theory of spontaneous generation’, Phronesis 7 (1962), 91–104. 1.90 Lennox, James ‘Teleology, chance, and Aristotle’s theory of spontaneous generation’, Journal of Hellenistic Studies 20 (1982), 219–38. 1.91 Pellegrin, Pierre, La Classification des animaux chez Aristote: Statut de la biologie et unite de l’Aristotélisme (Paris, Les Belles Lettres, 1982).