II ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ O NLY to’those unacquainted with Aristotle’s treatise on the development of animals will it seem strange that in a survey of the more important theories concerning development and heredity we hark back to such early beginnings. Aristotle’s biological knowledge, though remarkably extensive in its way, was hopelessly imperfect when it came to the minute study of the early stages of development; he had naturally no glimmering of an idea that the semen, which played so great a part in his theory of development, was composed of minute living organisms, the spermatozoa, nor that the female also produced, not a mere material secretion, but organized living unities, or eggs. His theory would have been transformed had he known these to us elementary and commonplace facts. He knew, and could know, nothing of the cell-theory, nor of segmentation and the formation of germ-layers. Even such incidents of early development as could be observed with the naked eye were not always noted by Aristotle with accuracy — he, like so many of us, was inclined to see what he wanted to see, and to ignore the rest — witness his inaccurate observation that the heart is the first organ to be formed in the embryonic Vertebrate. He was convinced on first principles that this must be so. As a storehouse of observed and established fact his treatise is then practically valueless at the present day.
Yet such was the sheer intellectual power of the man that he was able to work out from inadequate and largely incorrect data a theory of development which must even today be treated with respect. The form which he gave it can hardly be accepted now, but no one after him has put so much fundamental brain-work into the discussion of the central problems, and for this reason it is worth our while to consider Aristotle’s ideas in some detail. We shall take first his famous analysis and refutation of the theory of pangenesis put forward by Hippocrates and Democritus.
12 ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’
Bear in mind that it was the general view of the ancients that the embryo was in some way formed from the semen of the male and a corresponding secretion in the female; they knew nothing of ova and spermatozoa, nor of any cellular or vital continuity between one generation and another (hence, incidentally, the idea of spontaneous generation did not seem in any way strange to them, whereas to us it is almost inconceivable). Now Hippocrates had suggested that the semen came from all parts of the body, much as in Darwin’s theory 2,000 years later — how thought repeats itself! — all the parts produced pangens which were collected and stored up in the generative ceils.
‘The proofs’, writes Aristotle, 1 ‘from which it can be argued that the semen comes from each and every part of the body may be reduced to four. First, the intensity of the pleasure of coition ; for the same state of feeling is more pleasant if multiplied, and that which affects all the parts is multiplied as compared with that which affects only one or a few. Secondly, the alleged fact that mutilations are inherited, for they argue that since the parent is deficient in this part the semen does not come from thence, and the result is that the corresponding part is not formed in the offspring. Thirdly, the resemblance to the parents, for the young are born like them part for part as well as in the whole body; if then the coming of the semen from the whole body is cause of the resemblance of the whole, so the parts would be like because it comes from each of the parts. Fourthly, it would seem to be reasonable to say that as there is some first thing from which the whole arises, so it is also with each of the parts, and therefore if semen or seed is cause of the whole so each of the parts would have a seed peculiar to itself’ (721 b).
Aristotle deals satisfactorily with the first rather quaint argument; with regard to the second he exhibits a quite modern scepticism as to the evidence, and remarks profoundly that ‘if mutilated young are born of mutilated parents, it is for the same reason as that for which they are like them’ (724 a). It is, however, the third and fourth arguments that interest us particularly. We shall follow carefully Aristotle’s reply to these arguments, for he deals here with quite fundamental ideas as to the method of explaining hereditary
1 De Generation Animalium. Translated by Arthur Platt, Oxford, 1910.
ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ 13
resemblance, and the discussion has direct relevance to all later theories of representative particles, and indeed to all theories postulating a particulate germ-plasm.
Should the resemblance of the whole be explained in terms of the point-to-point resemblance of the parts, or is there some fundamental cause of the total resemblance which explains incidentally the point-to-point correspondence ? This is a fundamental question which is rarely faced nowadays. Yet upon the answer to this question depends the type of theory — integrative or analytic — which we adopt for explaining development and heredity. Aristotle saw this clearly, and dealt with the problem in a way which we must recognize to be masterly, and from which we can learn much. Aristotle had the great advantage over us of not being hindered in his biological thinking by the materialistic conceptions of later centuries, which, we may note incidentally, made havoc of his Physics. His whole philosophy is indeed a biological one. 1
Let us consider how he dealt with the main pangenetic argument. ‘First, then, the resemblance of children to parents is no proof that the semen comes from the whole body, because the resemblance is also found in voice, nails, hair, and way of moving, from which nothing comes. And men generate before they yet have certain characters such as a beard or grey hair’ (722 a). There are two points to be noted specially here: (1) that no representative particle can come from dead products like nails or hair, and (2) that the resemblance lies essentially in functional activity. The latter is the important point. This is shown by a study of the following passage, which I quote here out of its turn in order to get at once into grips with Aristotle’s meaning. ‘For even if it were true’, he writes, ‘that it comes from all the body — as they say — they ought not to claim that it comes from all parts of it, but only from the creative part — from the workman, so to say, not from the material he works in. Instead of that, they talk as if one were to say that the semen comes from the shoes, for, generally speaking, if a son is like his
1 See on this point ‘Aristotle’ by John Burnet, Proc. British Academy , xi, 1924.
14 ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ father, the shoes he wears are like his father’s shoes’ (723 b). It is clear then that resemblance in nails and hair and shoes is a derivative resemblance, depending upon the more fundamental resemblance in manner of total functional activity.
Similarly the point that certain characteristics of maturity can be inherited from a youthful father is inexplicable on the pangenetic hypothesis, but understandable if the hereditary tendencies are inherited as a unitary whole.
Continuing the original argument, Aristotle writes: ‘Further, children are like their more remote ancestors from whom nothing has come, for the resemblances recur at an interval of many generations. . . . The same thing applies also to plants, for it is clear that if this theory were true the seeds would come from all parts of the plants also ; but often a plant does not possess one part, and another part may be removed, and a third grows afterwards’ (722 a). These special difficulties of the theory of pangenesis have in modern times led to its replacement by the germ-plasm theory. It is, however, safe to say that Aristotle would not have accepted the theory of the germ-plasm without transforming it from a morphological into a functional theory. The next point considered by Aristotle is as to whether the semen is derived from the homogeneous parts (tissues) or from the heterogeneous parts (organs). 1 The passage must be quoted in full:
‘We may also ask whether the semen comes from each of the homogeneous parts only, such as flesh and bone and sinew, or also from the heterogeneous such as face and hands. For if (1) from the former only, we object that the resemblance exists rather in the heterogeneous parts, such as face and hands and feet; if then it is not because of the semen coming from all parts that children resemble their parents in these, what is there to stop the homogeneous parts also from being alike for some other reason than this ? If (2) the semen comes from the heterogeneous alone, then it does not come from all parts; but it is more fitting that it should come from the homogeneous parts, for they are prior to the heterogeneous which are composed of
1 The distinction between homogeneous and heterogeneous parts corresponds in the main with our distinction of tissues and organs, but not completely; see my Form and Function , 1916, pp. 12-14.
ARISTOTLE’S ‘DE GENERATIONS ANIMALIUM’ 15
them; and as children are born like their parents in face and hands, so they are, necessarily, in flesh and nails. If (3) the semen comes from both, what would be the manner of generation ? For the heterogeneous parts are composed of the homogeneous, so that to come from the former would be to come from the latter and their composition. 1 To make this clearer by an illustration, take a written name; if anything came from the whole of it, it would be from each of the syllables, and if from these, from the letters and their composition. So that if really flesh and bones are composed of fire and the like elements, the semen would come rather from the elements than anything else, for how can it come from their composition ? Yet without this composition there would be no resemblance. If again something creates this composition later, it would be this that would be the cause of the resemblance, not the coming of the semen from every part of the body’ (722 a— b).
The argument is not too easy to follow, on account of Aristotle’s condensed method of expression and the demands he makes on his reader, 2 so that some exposition is required. As to the first possibility, that the semen comes from the homogeneous parts alone, Aristotle objects that hereditary resemblance is essentially shown in the heterogeneous parts or organs; if then, ex hypothesi, resemblance in these parts can come about without the semen being derived from them, why should not the resemblance in the homogeneous parts also be due to this other cause ? If (2) the semen comes from the heterogeneous parts only, and not from the homogeneous, we are up against difficulties, for we should naturally expect the semen to come from the constituents of the heterogeneous parts, that is, from the homogeneous parts. This difficulty is further discussed under (3), where it is pointed out that to come from the heterogeneous parts cannot only mean to come from the homogeneous parts and their ‘composition’, i.e. their arrangement and fitting together to form the heterogeneous parts. And, speaking in material terms, we cannot conceive of semen coming from this ‘composition’. What is most likely is that the semen
1 i.e. arrangement.
2 What we possess of Aristotle are probably merely his lecture notes, not the works he prepared for publication; see Burnet, 1924.
1 6 ARISTOTLE’S T>E GENERATIONE ANIMALIUM’ comes from the primary elements of the homogeneous parts, which Aristotle called fire, water, and air. But hereditary resemblance is all in the mode of composition, in the way in which the elements are combined to form the tissues and in the arrangement of these to form the organs. No derivation of semen from any part can account for the re-constitution of this ‘architectonic’ resemblance. Hence the real cause of resemblance is the same factor that creates this organic architecture. Hereditary resemblance is, as it were, a byproduct of development, and will be explained only when we succeed in explaining development.
The argument is clear and convincing, and is applicable mutatis mutandis to all modern theories that deal with representative particles, genes or the like.
After dealing with some other difficulties of the theory of pangenesis — how, for instance, the representative particles of the semen combine in the proper way to form the germ of a new organism, and how, if they are present in the generative secretions of both parents they combine to form one and not two ‘little animals’ — Aristotle comes back to his argument about the homogeneous and the heterogeneous parts and takes it a stage farther.
‘Some parts’, he writes, ‘are distinguished by possessing a faculty, others by being in certain states or conditions; the heterogeneous, as tongue and hand, by the faculty of doing something, the homogeneous by hardness and softness and the other similar states. Blood, then, will not be blood, nor flesh flesh, in any and every state. It is clear, then, that that which comes from any part, as blood from blood and flesh from flesh, will not be identical with that part. But if it is something different from which the blood of the offspring comes, the coming of the semen from all the parts will not be the cause of the resemblance as is held by the supporters of this theory. For if blood is formed from something which is not blood, it is enough that the semen come from one part only, for why should not all the other parts of the offspring as well as blood be formed from one part of the parent. . . . Why not say that the semen from the very first is of such a kind that blood and flesh can be made out of it, instead of saying that it itself is blood and flesh ?’ (723 a).
Soon after comes the passage, already quoted, about the
ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ 17 semen and the shoes, which gives us the key to Aristotle’s own view of development and heredity and lead us naturally to a discussion of his constructive theory. But before considering his conception of the cause and course of development let us examine the passage we have just quoted. It is notable for the definition of organ as distinguished by ‘the faculty for doing something’. It further points out that the particles of the semen cannot have the same character as the parts from which they come — they must be ‘germs’ of these parts, such that from them the parts can be reconstituted. But it is clear that, if this is the case, the semen theory is quite useless as an explanation of development, for it already assumes development. If the semen-particle re-creates the part it represents, this is just as much a mystery as the development of the embryo as a whole — the problem is not disposed of by being split up in this way. This criticism also is fully applicable to all modern particulate theories of development and heredity.
The actual form of Aristotle’s own theory of development has mainly historical interest at the present day. His knowledge of the essential facts was fragmentary, and he couched his explanation in terms of a philosophy to which we have nowadays some difficulty in attuning our minds. Nevertheless his fundamental idea, that development is the actualization of a functional potentiality, is a profound one and gets down to the root of the matter. We have seen that Aristotle rejected firmly the ‘material’ explanation offered by the theory of pangenesis, after demonstrating clearly its insufficiency. He deduced from his analysis of this theory that development is due to ‘the creative part’, or the creative power, which brings about the ordered ‘composition’ or puttingtogether of the tissues and organs to form the functional whole. This is the real cause both of development and of hereditary resemblance. He gave creative function the priority over structure; he held that the whole was greater than its parts.
His actual theory was that the female provided only the material for development, while the male provided the efficient and formal cause, the principle of putting together
18 ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ or composition — as we should say, the formative impulse. The material possessed, at least potentially, the ‘nutritive soul’, rendering possible vegetative existence and sometimes growth; the male semen carried potentially the ‘sensitive soul’, which is the ground of movement and sense-perception, the distinguishing characters of the animal as compared with the plant. The semen derives its potentiality from the actuality of the male parent. Aristotle’s line of thought in this connexion is rather difficult to follow, and for us it can only have the value of an analogy, derived from the consideration of human activity; it is fairly clearly put in the following passages. After speaking of ‘automatic machines’ in which an effect is produced by the linked action of one part upon another, he goes on:
‘As, then, in these automatic machines the external force moves the parts in a certain sense (not by touching any part at the moment, but by having touched one previously), in like manner also that from which the semen comes, or in other words that which made the semen, sets up the movement in the embryo, and makes the parts of it by having first touched something though not continuing to touch it. In a way it is the innate motion that does this, as the act of building builds the house. Plainly, then, while there is something which makes the parts, this does not exist as a definite object, nor does it exist in the semen at the first as a complete part’ (734 b).
Commenting upon the first part of this passage, Platt gives the following useful note : ‘The male parent makes the semen and somehow imparts to it a potentiality of setting up movements in the embryo; this power given to the semen is like the impulse given to a piece of clock-work by pushing a wheel. Father = watchmaker, first wheel = semen, other wheels moved by the first = the parts developed by the semen.’ He justly adds : ‘We cannot solve the riddle any better at the present day; we can only say that no sooner has the spermatozoon penetrated the ovum than there is set up in the latter a series of movements which differentiate it and develop the parts one after another.’ But let us continue the quotation :
‘How is each part formed ? We must answer this by starting in the first instance from the principle that, in all products of Nature or art.
ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ 19
a thing is made by something actually existing out of that which is potentially such as the finished product. Now the semen is of such a nature, and has in it such a principle of motion, that when the motion is ceasing 1 each of the parts comes into being, and that as a part having life or soul. For there is no such thing as face or flesh without life or soul in it; it is only equivocally that they will be called face or flesh if the life has gone out of them, just as if they had been made of stone or wood. And the homogeneous parts and the organic come into being together. And just as we should not say that an axe or other instrument or organ was made by the fire alone, so neither shall we say that foot or hand were made by heat alone. The same also applies to flesh, for this too has a function. While, then, we may allow that hardness and softness, stickiness and brittleness, and whatever other qualities are found in the parts that have life and soul, may be caused by mere heat and cold, yet, when we come to the principle (Aoyos) in virtue of which flesh is flesh and bone is bone, that is no longer so; what makes them is the movement set up by the male parent, who is in actuality what that out of which the offspring is made is in potentiality. This is what we find in the products of art; heat and cold may make the iron soft and hard, but what makes a sword is the movement of the tools employed, this movement containing the principle of the art. For the art is the starting-point and form of the product; only it exists in something else, whereas the movement of Nature exists in the product itself, issuing from another nature which has the form in actuality’ (734 b-735 a).
The semen is not even a necessary link in the chain, for:
‘The male does not emit semen at all in some animals, and where he does this is no part of the resulting embryo; just so no material part comes from the carpenter to the material, i.e. the wood in which he works, nor does any part of the carpenter’s art exist within what he makes, but the shape and the form are imparted from him to the material by means of the motion he sets up. It is his hands that move his tools, his tools that move the material; it is his knowledge of his art, and his soul, in which is the form, that move his hands or any other part of him with a motion of some definite kind, a motion varying with the varying nature of the object made. In like manner, in the male of those animals which emit semen, Nature uses the semen as a tool and as possessing motion in actuality, just as tools are used in the products of any art, for in them lies in a certain sense the motion of
1 The text is doubtful here.
20 ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’
the art. Such, then, is the way in which these males contribute to generation. But when the male does not emit semen, but the female inserts some part of herself into the male, 1 this is parallel to a case in which a man should carry the material to the workmen. For by reason of weakness in such males Nature is not able to do anything by any secondary means, but the movements imparted to the material are scarcely strong enough when Nature herself watches over them. Thus here she resembles a modeller in clay rather than a carpenter, for she does not touch the work she is forming by means of tools, but, as it were, with her own hands’ (730 b).
The passage in which Aristotle points out that ‘there is no such thing as face or flesh without life or soul in it’ is cardinal to an understanding of his view. It is important to realize that this view was not ‘vitalistic’ in the modern sense of implying a dualism of matter and ‘entelechy’ ; for Aristotle ‘soul’ in this connexion was an expression for the total functional activity of the organic unit or part considered — its activity as a whole. This is certainly true of the ‘nutritive’ and the ‘sensitive’ soul, but must be qualified for the ‘rational’ soul, to which is attributed a certain independence of the body, ‘for no bodily activity has any connexion with the activity of reason’ (736 b). The other two kinds are dependent on the body for their manifestation, for ‘those principles whose activity is bodily cannot exist without a body, e.g. walking cannot exist without feet’ (736 b). Both the semen and the unfertilized egg have the nutritive soul potentially but not actually; the semen has in addition the sensitive soul potentially, but this soul cannot become actual or active until the embryo develops, and the organs of perception and movement are formed.
Abstracting from the particular form of Aristotle’s theory and stating only the essential core, we may say that he regarded what we should now call the fertilized ovum as possessing potentially the functional capabilities of the parents, ana development as being the actualization of these potentialities. The major part of them come from the male parent, for the female contributes only material, containing
1 Ai Aristotle thought was the case in certain insects. Cf. 721 a.
ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ 21
potentially the nutritive soul. Soul = functional activity, and exists in the fertilized ovum in a potential state. Resemblance between parents and offspring is a consequence of their starting out with the same developmental poten- , tialities; it cannot be explained as a point-to-point or particu- , late inheritance due to the transference of material particles. , The physical properties of the homogeneous parts, and of > their constituent elements, come in only as conditioning development, not as causing it. We shall see later, in Chapter X, that this conception is still of fundamental importance. Material properties, such as heat and cold, could not in Aristotle’s opinion account for development. Growth was due to the power of the nutritive soul, ‘using heat and cold as its tools in accordance with a principle’ (740 b).
Some properties of the tissues of the developing organism are due to ‘necessity’ — as we should say, ‘material causes’ — but above all this there is the final and efficient cause in which the real explanation lies (743).
I said above that Aristotle would not have accepted any germ-plasm theory in its modern structural form. But there is a certain analogy between his explanation of hereditary resemblance as due to identity of developmental potentialities, and the modern explanation in terms of identity of germinal constitution — the one theory deals in functions actual and potential, the other in structures; and the structural germinal complexities which modern theory finds it necessary to postulate are merely the translation into material terms of Aristotle’s functional potentialities. In a sense one explanation is as good as the other, but Aristotle’s has the merit of being simpler and not requiring any elaborate hypothetical machinery.
Aristotle was, of course, an out-and-out believer in epigenesis. He based his opinion partly on observation and partly on reasoning. His theory made it unnecessary to assume that any part existed ready-made either in the semen or in the material supplied by the female. He puts the alternatives of preformation or epigenesis as follows:
‘Either all the parts, as heart, lung, liver, eye, and all the rest, come
22 ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’
into being together or in succession, as is said in the verse ascribed to Orpheus, for there he says that an animal comes into being in the same way as the knitting of a net. That the former is not the fact is plain even to the senses, for some of the parts are clearly visible as already existing in the embryo while others are not; that it is not because of their being too small that they are not visible is clear, for the lung is of greater size than the heart, and yet appears later than the heart in the original development’ (734 a).
He raises too the question as to whether the succession of parts is a causal or a temporal succession and concludes that it is merely temporal. His reason is that ‘in all the productions of Nature or of art, what already exists potentially is brought into being only by what exists actually; therefore if one organ formed another the form and the character of the later organ would have to exist in the earlier, e.g. the form of the liver in the heart’ — which is absurd (734 a).
The point is further discussed in a later chapter. After comparing the embryo with the seeds of plants he goes on: ‘So also in the embryo all the parts exist potentially in a way at the same time, but the first principle is furthest on the road to realization. Therefore the heart is first differentiated in actuality. This is clear not only to the senses (for it is so) but also on theoretical grounds’ (740 a).
And summing up, he writes : ‘Now the parts of the embryo already exist potentially in the material, and so when once the principle of movement has been imparted to them they develop in a chain one after another, as the wheels are moved one by another in the automatic machines’ (741 b). This is not of course to be understood in the modern sense that one stage of ontogeny is the cause of the next, but rather that the ‘movement’ imparted by the semen is the primary and underlying cause of the whole process.
Development, Aristotle noticed, was from the general to the special. The unfertilized embryo, before receiving the activating impulse from the semen, is possessed of the nutritive soul and lives the life of a plant. As it develops after ‘fertilization’ by the semen, the embryo acquires the sensitive soul in virtue of which an animal is an animal. Furthermore,
ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’ 23
‘an animal does not become at the same time an animal and a man or a horse or an y other particular animal. For the end is developed last, and the peculiar character of the species is the end of the generation in each individual’ (736 b). We have here a clear foreshadowing of K. E. von Baer’s law of development.
There is an indication, too, of the truth, emphasized many centuries later by W. Roux, that many organs are formed in advance of functioning. Thus, ‘those which breathe and whose parts are differentiated within the mother’s uterus yet do not breathe until the lung is perfected, and the lung and the preceding parts are differentiated before they breathe. Moreover, all polydactylous quadrupeds, as dog, lion, wolf, fox, jackal, produce their young blind, and the eyelids do not separate till after birth’ (742 a).
There are many other points of interest in Aristotle’s treatise on which much time might be spent. We shall mention only one or two. His ideas on sex were necessarily crude, but he did point out that separate sexes were characteristic of animals that move, whereas sedentary animals, like plants, had no sex (715) or as he explains later (731 a) were male and female combined. ‘In all animals which can move about, the sexes are separated, one individual being male and one female. . . . But in plants these powers are mingled, female not being separated from male. Wherefore they generate out of themselves, and do not emit semen but produce an embryo, what is called the seed’ (731 a). There follows the quaint suggestion that animals ‘seem literally to be like divided plants, as though one should separate and divide them, when they bear seed, into the male and female existing in them’. The rationale of the separation of the sexes in motile animals is to enable them to develop the faculties of the sensitive soul. ‘For to the essence of plants belongs no other function or business than the production of seed; since, then, this is brought about by the union of male and female, Nature has mixed these and set them together in plants, so that the sexes are not divided in them. . . . But the function of the animal is not only to generate (which
24 ARISTOTLE’S ‘DE GENERATIONE ANIMALIUM’
is common to all living things), but they all of them participate also in a kind of knowledge, some more and some less, and some very little indeed. For they have sense-perception, and this is a kind of knowledge’ (731 a). Such knowledge may be very inferior as compared with intellect, but it is ‘most excellent’ as compared with the absolute insensibility of the inorganicworld. For Aristotle there were of course four grades of existence — matter, vegetative existence, thelife of sense perception, and the life of reason. In man all are combined in a hierarchy; in animals the first three, and in plants two only.
The problem of sex-determination had great interest for Aristotle; he discussed all the current views and worked out a new solution of his own, which we need not however consider here.
We have seen that Aristotle realized the important point that the explanation of hereditary resemblance is dependent upon the explanation of development, that this resemblance is a feature of development rather than a separate problem. It is of interest to reproduce here his summary of the facts of heredity, which shows that he had gone very carefully into the matter.
‘(1) Some children resemble their parents, while others do not, some being like the father and others like the mother, both in the body as a whole and in each part, male and female offspring resembling father and mother respectively rather than the other way about. [(2) They resemble their parents more than remoter ancestors, and resemble those ancestors more thananychanceindividual. (3) Some,thoughresembling none of their relations, yet do at any rate resemble a human being, but others are not even like a human being but a monstrosity’ (767 b).
The second generalization has quite a biometric turn, and only wants precise formulation to give us Galton’s law.
In concluding this slight sketch of Aristotle’s views on heredity and development I should like to emphasize the fact that the De Generatione Animalium is a veritable mine of philosophic thought and well worth a much more thorough investigation than has been attempted here. I have merely detached the main ideas which seem to have permanent significance for any and every theory of development and heredity.