The first difficulty in reading Aristotle's biological treatises, as often in reading Aristotle, may well be to decide exactly what their purpose is. They are so factual and so comprehensive that it is easy to mistake them for descriptive information, an animal encyclopaedia, as the ancients regarded them. Modern readers have tried to assimilate them to present-day categories, classifying the HA as natural history, the PA as comparative anatomy, the PN as physiology, the GA as embryology. But this assimilation does not fit. One can test it by looking for the facts about any given animal in HA. Without an index (which ancient readers did not have) the facts can only be found by reading through the whole treatise, for they are distributed all over it; and when found they may seem strangely inadequate. A striking example is the blind mole-rat, aspalax, which Aristotle quotes in the De Anima as an interesting case. He twice describes a dissection of its concealed eyes. But the only other fact that he reports is that it is viviparous – not what kind of animal it is, how many legs, what its coat or feet or tail are like, how it lives, nothing. His aim is clearly not to give a natural history of the mole, but to show how it differs from other animals: it is his only case of sightlessness combined with viviparousness.

The range of Aristotle's investigations in zoology is such that a discussion of his use of empirical methods has to be drastically selective. Yet the need to come to some assessment of his performance in this field is all the more pressing in that it has been subject to such divergent judgments. Some of the most extravagant praise, but also some of the most damning criticisms, have been directed at his empirical researches in zoology.

The massive array of information set out in the main zoological treatises can hardly fail to impress at the very least as a formidable piece of organization. But both Aristotle's sources and his principles of selection raise problems. As we have already noted in connection with his use of dissection, it is often impossible to distinguish Aristotle's personal investigations from those of his assistants, although, given the collaborative nature of the work of the Lyceum, that point is not a fundamental one. It is abundantly clear from repeated references in the text that he and his helpers consulted hunters, fishermen, horse-rearers, pig-breeders, bee-keepers, eel-breeders, doctors, veterinary surgeons, midwives and many others with specialized knowledge of animals. But a second major source of information is what he has read, ranging from Homer and other poets, through Ctesias and Herodotus to many of the Hippocratic authors. In general he is cautious in his evaluations of all this secondary evidence.

In the last chapter of her book Physicalism K. V. Wilkes writes:

This view is developed also by D. W. Hamlyn in the introduction to his Aristotle's De Anima (1968): it certainly has the charm of paradox. The common Greek word nous appears to express our concept of a mind. Aristotle speaks of nous as ‘part of the soul’ and says it differs from other parts in three ways: it is not the concern of the natural scientist, it comes in ‘from outside’, and it is ‘separable’ from the body. If that does not give him a mind–body problem, what would? The fact is that while Aristotle's account of psychological concepts generally, his doctrine that soul and body are related as form to matter or actuality to possibility, is attractively undualistic, when he comes to nous dualism seems to seep back. J. L. Ackrill in Aristotle the Philosopher remarks:

Scientific method in Aristotle's biology

The relation between Aristotle's official account in the Posterior Analytics of the nature of scientific knowledge and of the means by which it is reached and his actual practice in arriving at the results presented in his special scientific writings has long been a topic of considerable study. In the recent history of attempts to account for the discrepancies between Aristotle's theory and his practice, or to explain away the apparent discrepancies, the biological works have been assigned a special role. In his famous and still influential treatment of this problem, Jaeger saw in what he took to be the thoroughgoing empiricism of the biological works the final step in Aristotle's emancipation from the Platonic view of scientific knowledge and method found in the Analytics. Students of Aristotle now agree that Jaeger's general account of Aristotle's ‘progress’ away from Platonism is untenable. But it is still widely supposed that Jaeger was at least right that there are empirical elements in the method practiced in the biological writings to which no role is given in the Analytics.

The chief difficulty that final causes present to modern philosophers lies in reconciling them with what Aristotle calls ‘necessity’, that is the automatic interactions of the physical elements. It is difficult to see, first, how laws of nature can be directed towards goals and still remain ‘necessary’; and, secondly, what could be the author and the means of such direction. The modern cybernetic model, and the concept of elaborate genetic coding, have not altered the problem; they have merely shown that some apparently teleological processes may in fact be necessary outcomes. It is arguable, as we shall see, that in the GA Aristotle himself was moving towards such a position. But there is no sign of it in PA, nor is there any sign in his writings generally that the relationship between finality and necessity could be a difficulty of the sort that we feel.

The novelty in Aristotle's theory was his insistence that finality is within nature: it is part of the natural process, not imposed upon it by an independent agent like Plato's world soul or Demiourgos. This is what allows him to claim that none of his predecessors had recognized the final cause with any clarity. Anaxagoras called his primary cosmological cause ‘Mind’, and for this Aristotle likened him to a lone sober man among drunks; Plato offered cosmic teleological causes in the Timaeus, Philebus and Laws; Xenophon argued for the popular belief in providential guidance of natural phenomena.

Aristotle is often characterized, by both philosophers and evolutionary biologists, as the fountainhead of a typological theory of species that is absolutely inconsistent with evolutionary thinking. D'Arcy Thompson, on the other hand, in his remarkable On Growth and Form, claimed that the idea of using quantitative methods to help understand morphological relationships among animals of different species took root in his mind during his work on Aristotle's biology:

A theory that asserts that species of a genus differ only in the relative magnitudes of their structures sounds very different, and might be thought to be incompatible with, a theory that claims that there are complete discontinuities between one eidos and all others. Can Aristotle consistently have held both these views? He can, and he did. To understand how he did so, one must understand the way in which he used the Academic technical notion of ‘the more and the less’ in his biology.

A longstanding problem concerning Aristotle's philosophy of science is the extent to which there is a serious conflict between the account of scientific explanation and investigation in the Posterior Analytics and the explanations and investigations reported in treatises such as the Historia Animalium, Parts of Animals, and Generation of Animals. I shall not here mount a frontal attack on this question, preferring instead a rearguard action. This will consist of (i) articulating a familiar epistemological distinction between unqualified and sophistic or incidental understanding, which plays a fundamental role in Aristotle's philosophy of science; (ii) relating that distinction to the methodological recommendations of Posterior Analytics 11.14; and (iii) presenting three sorts of evidence from the PA and HA indicating that the zoological works owe a great deal both to the above epistemological distinction and to its methodological implications. The evidence for (iii) will consist of the organization of information found in the HA, the relevance of that organization to the explanations of PA, and the theoretical and practical concern with the way in which lack of an appropriate zoological nomenclature can hamper the achievement of understanding. The evidence points to a much more direct relationship between the APo. and the biology than recent commentators have suggested.

Animals, in Aristotle's view, are paradigm instances of substance-being. We may wonder whether Aristotle began with that conviction and shaped his ontology in the light of it, or arrived at it as a result of what his ontology revealed the nature of substance to be, and that question in turn may be related to our views about when and with what attitude he did his biological work. But whatever we decide to say on these questions, our answers will have to take account of his clear conviction that natural living things are ‘above all substances’.

We should therefore expect that Aristotle thought the most important features of substance to be exemplified in the case of living entities. This need not mean that he claimed that only animals are substances. On the whole, Aristotle was less concerned with the correct identification of a class of entities which are substances than with the proper understanding of the principles and modes of being by virtue of which those entities which we commonly understood to be substantial beings are substantial. He was, we might say, less interested in substances than in substance-being, less concerned with the question of what beings are substances than with the question of what it is to be a substance. But among the beings that Aristotle saw as commonly (and therefore in some sense correctly) understood to exemplify substance-being, animals came first, and those features which metaphysics reveals as characteristic of substance-being should therefore be evident in the being of animals.

In his famous exhortation to biological study in the last chapter of Parts of Animals 1, Aristotle offers his students several reasons for pursuing the study of plants and animals.

While the study of the heavens has the superior objects, he begins, the information about living things is ‘better and more plentiful’, and their study thus ‘take[s] the advantage in knowledge (epistēmē)’ (645a1–2).

This study is also, Aristotle continues, a source of ‘wonder’ and ‘immeasurable pleasures’, since ‘the non-random, the for-something's sake, is present in the works of nature most of all, and the end for which they have been composed or have come to be occupies the place of the beautiful’ (a9, 17, 23–6). The epistēmē that this study of living things offers is an understanding final causes, and the pervasiveness, and perspicuousness, of the final cause among plants and (especially) animals is another source of the biology's value.

In two places in his extant works, in Physics 11.9 and three connected passages of the Parts of Animals 1.1 (639b21–640a10,642a1–13,31–b4) Aristotle introduces and explains his notion of ‘hypothetical necessity’ (anangkē ex hupotheseōs). Judging from this terminology one might think a hypothetical necessity would be anything that is necessary given something else, or something else being assumed to be so (cf. APr. 1.10 30b32–40) – in effect, anything that follows necessarily from something else's being so, but that may not, taken in itself, be necessary at all. But this is not so: the necessity of New York's being north of Princeton given that it is north of New Brunswick and New Brunswick is north of Princeton, is not an example of what Aristotle calls a hypothetical necessity. This is because the hypothesis relatively to which a hypothetical necessity, in Aristotle's usage, is necessary is always a goal posited or set up (hupotethen) as something to be achieved. Hence the necessity in question is always that of a means to some end. A hypothetical necessity, as Aristotle intends the term, is something necessary if some goal is to be attained. Thus, he says, a saw or an axe has to be hard in order to do its job of cutting or splitting, and in order to be hard it has to be made of bronze or iron: hence it is necessary for a thing to be made of bronze or iron if it is to be a saw or an axe, and there must be bronze or iron to hand if one is to carry out the intention to make a saw or an axe.

In this paper I want to explore the application of some of Aristotle's central metaphysical notions to the analysis of living things. Commentators have often remarked that the conceptual pairs of matter / form and potentiality/actuality apply best to the analysis of Aristotle's favorite example, the bronze statue, but fit organic creatures – the true substances – less well. Two distinct problems arise for these more complex cases.

First, it seems impossible to point to something which serves as both matter ‘for’ and matter ‘of’ a living creature, as bronze does for statues. In Physics 1.7 Aristotle tells us that in any change some underlying thing or substratum persists, and it is easy to see that the bronze persists when a statue is molded, the gold continues in the bangle, etc. But what exists before a person, duck, or oak tree comes to be is an embryo, egg, or acorn (cf. Ph. 1.7 190b3–5) – and these surely do not remain in the finished creatures as their matter. This disanalogy is the target of William Charlton's criticism in his commentary on Physics 1 and II:

Introduction

If Aristotle is known for anything it is for his teleology. In the understanding of this issue the biology has proved especially useful, as one might well expect, since the primary subject of teleological explanation for Aristotle is the living thing – its development, structures, and activities.

What, then, is it, according to Aristotle, for a part of a living thing to be for the sake of an end, where that end is explanatory of the part's presence in an organism? And what is it for that part, or that organism, to come to be or to act for the sake of an end, where that end is not consciously aimed at? Earlier interpreters tended to assimilate the general living case to the human case, and they spoke for instance of potential souls and invisible entelechies guiding organic development – of ‘immaterial agencies’ of various sorts. But this, as J. H. Randall, Jr. (1960) pointed out, is to turn the final cause into an efficient cause. More recent interpreters agreed, and went on to speak as if the true causation involved in teleological processes was all at the material level.

Introduction

It is often maintained that Aristotle's practice in such explanatory treatises as the Parts of Animals does not correspond to the theory of science presented in the Posterior Analytics. Two major respects in which this is so, it is held, are (i) the absence in PA of the explicit syllogisms in which, according to APo., the explanations are to be cast, and (ii) the absence in PA of the axiomatic structure of explanation APo. calls for.

Writing of the second point in his little book on Aristotle (1982), Jonathan Barnes explains that

Barnes' formulation makes it clear that he takes the requirement that proper science have an axiomatic structure to mean that it is to be presented in a certain way – with its first principles presented first, and labeled as such, and its deductions following sequentially, ‘in the geometrical manner’. Similarly, the requirement of the full theory of APo. that the deductions be syllogistic in form, Barnes and others take to be a requirement that they be presented in explicit syllogisms. The absence of explicit syllogisms and of an explicit axiomatic structure is the basis of the twofold discrepancy between the APo. theory and the PA practice.

1. In my book, Aristotle's Classification of Animals I tried to show that there is no room at all for any animal taxonomy in the Aristotelian biological project. The various orderings of animals which we find in Aristotle are always relative to the point of view and the immediate objective of the inquiry at hand. Thus Aristotle can at one time order animals according to the growing complexity of their reproductive organs, at another according to the form and disposition of their nutritive organs. Certainly it seems obvious to us that such studies presuppose a distribution of animals into stable and recognized families. That is not how it is for Aristotle, and I tried to present the status, in the biological works, of his various classifications, relative to his different inquiries, none of which is able to claim priority over the others; these are purely empirical procedures, meant to facilitate the work of the biologist, but they remain outside properly epistemic research. I claim that the retrospective projection of the theoretical presuppositions of ‘classical’ natural history onto Aristotelian biology has prevented the best interpreters, and a fortiori the lesser, from grasping the exact functioning of these concepts in that biology. That is true of the notions of genos and eidos and of the relationships which, so to say, follow these two concepts, in particular the relation of analogy.

Two questions

Socrates' conversation with Phaedrus is rich in references to its own setting. As the dialogue opens Phaedrus is about to take the air outside the city walls when he happens upon Socrates, who readily agrees to accompany him in return for an account of his morning's entertainment by the orator Lysias. Their conversation in these opening pages is peripatetic, and much of it is directly concerned with the landscape in which they walk and talk. Where they should sit for Phaedrus to deliver Lysias' speech; what landmarks they pass on the way; and (when they get there) whose shrine they have stumbled upon – such are the questions that exercise them as they stage-manage the speechmaking of the dialogue's first part. Their theatre even has a resident chorus: the ‘chorus (khorōi) of cicadas’ (230c3) whose summery treble Socrates takes note of on arrival. The cicadas' song will be heard to greatest effect later in the dialogue, at the outset of the critique of rhetoric that makes up its second part. Before launching fully into their discussion of rhetoric, Socrates and Phaedrus will break off to consider their physical environment once again, when Socrates – arresting the action, as it were, to let the chorus have its moment – warns Phaedrus against the potentially mesmerising effect of the droning cicadas overhead and tells him a parable in which they are the main characters (see 258e6–259d8).

A change of course

In the pattern of my exegesis so far I have been guided by Plato's sense of drama. He begins both acts of the Phaedrus by fixing the reader's attention on the scenic background, and I have begun my interpretation with an analysis of this topographic ploy. But the time is ripe for emancipation from Plato's script.

I have shown, in a preliminary way, how important to the overall concerns of the dialogue is the effect of sudden downshift from the high rhetorical gear of the speeches on love to the comparatively sedate account of the rhetorical art they exemplify. However, an exegesis of this effect is not obliged to duplicate its jolt on the reader (for all that it cannot wholly escape the expository problems which, I have argued, that jolt brings alive: this we have just seen, in the final section of the previous chapter). Indeed, the order of understanding will be better served if at this point I reverse Plato's order of exposition, and tackle the account of rhetoric in the second part of the dialogue before the love-poetry of the first; for, to anticipate, a grasp of Plato's criticisms of the rhetorical art practised and taught in his society turns out to be an essential step towards our appreciating the following three crucial points about the actual samples of rhetorical art which precede those criticisms. The first concerns the very content of the speeches on love: it will turn out that the character and the message of the fictional speakers of each of those speeches are partly determined by the contrast between the rhetorical and the philosophic approach to the art of speaking.