Is Aristotle committed, as a theoretical matter, to fixed species in biology? The answer seems to be a resounding no, if we were to infer his theoretical commitments from the actual practice found in his biological works.Footnote ¹ The answer, however, is far from clear, if we turn to the ‘philosophical discussion of biology’Footnote ² found in Book 1 of Parts of Animals. In fact, I shall note that its context suggests that, contrary to some recent interpretations put forward,Footnote ³ the phrase τὰ ἔσχατα εἴδη at 644a23 is best translatedFootnote ⁴ and understood as ‘infima species’, and that such a reading implies that Aristotle favours the methodology that reflects his theoretical commitmentFootnote ⁵ to fixed species (at least in Book 1 of Part. an.).

The crucial passage, containing the disputed phrase τὰ ἔσχατα εἴδη, is the following (644a24–b1; Lennox's translation with slight modifications):

Lennox notes that the opening sentence of this passage is rather difficult and ‘has been variously translated and interpreted’.Footnote ⁶ Traditionally, the phrase τὰ ἔσχατα εἴδη has been translated and/or understood to mean ‘infima species’.Footnote ⁷ But the problem with such a translation is that it is not clear, exactly, to what the word ‘these (ταῦτα)’ at a24 is referring. The natural way to read the Greek is that the reference is to εἴδη, but the implication is that Socrates and Corsicus will then turn out to be ‘infima species’, which is not possible. So ‘these’ must instead refer to Socrates and Coriscus (some suggest a slight emendation to reflect this view).Footnote ⁸ Others argue that, even if ‘these’ refers to εἴδη, Socrates and Coriscus are mentioned so as to exemplify membership in a species (‘mankind’) and are not meant to be identified as such.Footnote ⁹ More recently, however, there has been a shift in strategy. SomeFootnote ¹⁰ now suggest that the term εἴδη should no longer be translated or understood as ‘species’ but instead as ‘form’ in Aristotle's biological works. In this interpretation, the natural reading of the Greek (without identifying Socrates and Coriscus with ‘infima species’) is preserved by understanding the phrase τὰ ἔσχατα εἴδη to refer to individual forms (rather than species) of substances.Footnote ¹¹

The context of the passage (which goes back all the way to the first chapter of the first book of Part. an.), however, suggests otherwise. There Aristotle raises a number of questions concerning the nature of zoological investigation. The first of these questions involves a choice between two methods: either to study the specific attributes (τῶν ἰδίων) of each substance (ἑκάστην οὐσίαν) separately (Method 1) or to study the common (κοινῇ) attributes belonging to many kinds (Method 2) (639a15–b6). It is instructive to list the examples of the objects of Method 1 that Aristotle gives. They are lion (639a18), ox (639a19), horse (639a25) and dog (639a25) from chapter 1; and mankind (644a32), sparrow (644a34) and crane (644a34) from chapter 4. Note that by ‘each substance’ Aristotle does not mean to pick out each individual animal but rather each species. Had he meant each individual animal, he would have indicated it by explicitly using an indefinite adjective τις in describing each animal he lists, as he does elsewhere.Footnote ¹² Method 1, then, is supposed to be studying each species of animals separately.

Aristotle's commitment to fixed species in biology (at least from the methodological standpoint) is revealed by the fact that he considers Method 1 (which studies each infima species) to be the ‘best’ (κράτιστον; 644a30). Some scholars have missed this important point. For example, in an attempt to read Aristotle's ‘deeper motive’Footnote ¹³ surrounding the Part. an. 1.4 passage, Balme suggests that in the background Aristotle has in mind An. post. 2.14, where Aristotle showed that often the fundamental generic attributes are especially important because they reveal or are the causes of the specific attributes. For example, we learn in the An. post. passage that cows do not have upper incisor teeth, but they have a third stomach not qua cows but qua horned animals; it is not any particular attributes that are found exclusively in cows that explain their having an additional stomach and lack of incisors; rather, it is because they are horned animals, for the growth of horns diverts material from the teeth and hence requires an additional stomach for mastication.

So to understand the nature of some specific attributes, an expositor must correctly identify the fundamental generic attributes that are causes of these specific attributes, for, ‘without the generic attributes, explanation cannot begin’.Footnote ¹⁴ But once the correct generic attributes have been identified, there is no point in repeating the explanation for every instance; and here lies the reason for the absurdity of repetition—the expositor who repeats the explanation in every instance has failed to understand the fundamental character of the cause.

Indeed, the appeal to An. post. 2.14 is a plausible way to understand the choice between the two methodologies, given that both Part. an. 1 and the An. post. passage cover similar topics, such as the method of division, the use of common names (such as birds) and the discussion of analogy.Footnote ¹⁵ In addition, the examples that are illustrated in An. post. 2.14 are all taken exclusively from biology. Lennox, who is sympathetic to Balme, also believes that An. post. 2.14 ‘is in all likelihood in the background’Footnote ¹⁶ of Part. an. 1.4, because the question raised in Part. an. 1.1, 639a16–19 ‘is so reminiscent of An. post. 2.14–18 that it is difficult not to see its methodological suggestions in the background’.Footnote ¹⁷ Furthermore, Lennox declares that the An. post. passage provides us with a ‘powerful reason for seeking to grasp common attributes according to kind, rather than case by case’.Footnote ¹⁸

Despite the plausibility of his suggestion, Balme himself points out its shortcoming: ‘If this is what was in Aristotle's mind it is admittedly odd that he did not say so.’Footnote ¹⁹ Balme then speculates that this apparently trivial reason—the tediousness of repetition—may be ironical, since Aristotle ‘was not ready with evidence’ to support this theory in his zoological enquiry. As mentioned earlier, the fundamental problem with Balme's explanation is that it does not address why Aristotle describes Method 1 as the ‘best’ as if to favour Method 1 over Method 2. On the contrary, Balme exclusively focusses his attention on what he perceives to be the merit of Method 2 when in fact Aristotle is much more concerned with stressing the merit of Method 1; hence, this discrepancy of the emphasis—Balme on Method 2 and Aristotle on Method 1—can easily account for the oddity noticed by Balme. Before we resort to an ironical reading of the passage in question (even if it may simply be an obiter dictum), we should first provide an account that explains why Aristotle thinks Method 1 is the ‘best’.

Lennox attempts to offer an account that acknowledges the merit of Method 1, but he does so by contrasting it with the merit of Method 2. According to him, Aristotle is explicitly framing the tension between Method 1 and Method 2 in terms of the tension that exists between ‘the greater reality of the particular’ and ‘the greater epistemic value of the universal’Footnote ²⁰ found in Metaphysics घ. The nature of this conflict is that the particular (which is more real) cannot be an object of unqualified knowledge, but the universal (which is an object of unqualified knowledge) cannot be a substance. The conflict that is supposedly played out is as follows: metaphysical considerations ‘might incline us to study each indivisible form of a kind … [f]or formally indivisible beings are most real, while what is most universal is furthest from reality’; but ‘from the standpoint of methodological economy, this will be silly and long-winded’.Footnote ²¹

A close inspection of the passage, however, shows that the distinction between the two methodologies is drawn only in part in terms of the particular (644a31) and the universal (644a28), and it is far from clear that this is the heart of the contrast. The main reason for this is that the objects of both Method 1 and Method 2 are universals and not particulars. According to Lennox, the contrast between Method 1 and Method 2 is ‘not between more particular and more general animal kinds (for example, between lion and quadruped) but between individual animal natures studied independently and common attributes established ‘according to something common’ (639a18) and ‘in common according to kind’ (639b5).Footnote ²² But it is difficult to see how studying individual animal natures means studying some particular forms and not something common among individual animals. Surely, when one is studying individual animal natures (such as the natures of lion), even if one is studying them independently from the natures of other quadruped animals, one is still studying common attributes shared by many individual lions. Similarly, when one is studying sparrows or cranes independently from each other, one is still studying common attributes shared by individual cranes or individual sparrows.

In fact, Lennox's descriptions of the choice between the two methodologies imply that universality is involved in both: ‘whether we may also discuss kinds in common as well, and not just animals of the same form’.Footnote ²³ As soon as animals are described as having ‘the same form’, we leave the realm of particulars and enter into the realm of universals. So these individual animal natures that Lennox speaks of are universals and not really particulars. The two immediate implications of individual animal natures being universals are that: (1) they are also objects of unqualified knowledge; and (2) they are particular only in relation to more universal attributes belonging to many different kinds of animals. Both (1) and (2) do not seem to square with Lennox's interpretation that we are really dealing primarily with the tension between particulars and universals.

The problem with the interpretations of both Balme and Lennox is that their readings revolve around stressing the philosophical import of Method 2 (as opposed to Method 1). The straightforward reading of the text, however, suggests otherwise; that is, Aristotle does not mention any advantage to Method 2 other than the fact that it avoids the tedious repetition of Method 1. But why did Aristotle believe that Method 1 is the ‘best’? Unfortunately, he does not explicitly tell us. I would like to offer, however, an alternative, original and plausible interpretation that is well supported by the passage and by his texts elsewhere: that Aristotle thinks Method 1 is the ‘best’ because it has both metaphysical and epistemological advantages over Method 2.

The crucial detail of the passage worth noting is that, when Aristotle emphasizes the nature of the last forms as substances, he does so in terms of their being undivided (ἄτομον), which is mentioned three times (at 644a30, 644a31 and 644a33), in contrast to their being particular (καθ’ ἔκαστον), which is mentioned only once (644a31). Aristotle's emphasis on indivisible or undifferentiated form, rather than on its particularity, implies that he is emphasizing the unity rather than the being of substance or form. In Metaph. घ.8, for example, Aristotle points out that, although Callias and Socrates are said to be distinct in terms of their matter (i.e. in terms of their flesh and bones, etc.), they are nevertheless the same in terms of their form (ταὐτὸ δὲ εἶδος), because the form is indivisible (ἄτομον γὰρ τὸ εἶδος) (1034a5–8). A similar point is made about the nature of the ultimate indivisible (τὸ ἔσχατον ἄτομον) in Metaph. Ι.9: one man may be distinct from another man in terms of their flesh and bones, but they are not distinct in terms of their form (εἴδει δ’ οὐχ ἕτερον), because there is no contrary in their formula (ὅτι ἐν τῷ λόγῳ οὐκ ἔστιν ἐναντίωσις) (1058b5–10). Since sameness is a kind of unity (Metaph. Δ.9, 1018a7), the contrast is framed in terms of formal unity, that is, one in form (which is indivisible) and one in genus (which is divisible or differentiable in terms of different forms). Furthermore, in contrast to undivided forms, Aristotle indicates the divisibility of genus when he explicitly mentions about ‘bird’ at 644a33 that ἔχει γὰρ εἴδη τὸ γένος. Method 1, then, is best primarily because of the nature of the indivisibility of ‘species’ (in contrast to the divisibility of ‘genus’).

The objects of Method 1 (‘infima species’) do enjoy a metaphysical advantage (although not in the way Lennox explained). Aristotle twice emphasizes the nature of substances at 644a24 and 644a29. Since species are ‘more real’ in the sense that they are closer to individual substances, although both species and genus are universal, the former is more particular in relation to the latter. Explanations based on these forms, then, have an ontological advantage over those at a higher level of generality. That is the reason why Aristotle draws a distinction between particular and universal, but (as we pointed out previously) he does so only to an extent. The metaphysical advantage that is implied by the passage is not the focus of his attention and is only a secondary concern.

The crucial distinction that Aristotle draws in terms of indivisibility (of species) and divisibility (of genus), in fact, implies that the objects of Method 1 (‘infima species’) turn out to enjoy an epistemological advantage (contrary to the interpretation of Lennox). First, Aristotle closely relates indivisibility with knowledge. For example, he identifies indivisibility as a prerequisite for knowledge, for without the indivisibles (τὸ ἴδιον) it is not possible to know (εἰδέναι).Footnote ²⁴ Furthermore, Aristotle explicates what is indivisible (ἀδιαίρετος) in form (εἴδει) in terms of indivisibility in knowledge (τὸ τῷ γνωστῷ καὶ τῇ ἐπιστήμῃ).Footnote ²⁵ Second, when there is a contrast between two accounts, one based on a distinctive characteristic (ἴδιον) (of indivisible species) and the other based on what is common (κοινόν), Aristotle prefers the former on an epistemological basis. This can be seen not only in the Part. an. 1.4 passage but also in a De Anima passage, where Aristotle states that it is ridiculous (γελοῖον) to seek a common account (τὸν κοινὸν λόγον) whether one is studying figures or souls, if we fail to seek a distinctive account (ἴδιος λόγος) and proper properties belonging to ‘indivisible form’ (ἄτομον εἶδος).Footnote ²⁶ Note the contrast Aristotle here is drawing between ἴδιος λόγος and κοινὸς λόγος. The same contrast (which is hinted in the Part. an. 1.4 passage, when he describes the objects of Method 2 as ‘common’ twice at 644a27 and 644a35) is also made explicitly in Part. an. 1.1: ‘… whether one should study things in common (κοινῇ) according to kind first, and then later their distinctive characteristics (τῶν ἰδίων), or whether one should study them one by one straight away’ (639b4–5; Lennox's translation with my emphasis). The import of this contrast is found in Cat. 5, 2b9–14, where Aristotle declares that species is more knowable (γνωρισμώτερον) than genus, because species identifies what is more distinctive (ἴδιον μᾶλλον), while genus only tells us what is more general (κοινότερον), about a given substance.Footnote ²⁷

So, the upshot is that Aristotle thinks that Method 1 is the ‘best’ because it has both metaphysical and epistemological advantages. But in biology he endorses Method 2 for the most obvious reason that he explicitly tells us: to avoid tedious repetition, because the diversity of living species shares the same affections (πάθη) and dispositions (διαθέσεις), such as sleep, respiration, growth, deterioration, death and other attributes (συμβεβηκότα).Footnote ²⁸ An implication of my interpretation is that, although Aristotle was theoretically committed to fixed species in biology (and preferred the methodology that reflects this commitment), for practical considerations he selected the alternative method that deviated from it.

But what are the consequences of accepting my interpretation? First, because Aristotle employed Method 2 for the most part in his biological works, it would not be surprising to discover (as correctly observed by many contemporary scholars who examined them) that εἶδος as well as γένοςFootnote ²⁹ are relative terms applied at the different levels of generality, and that Aristotle uses them as if there are no fixed or privileged levels. We could only expect such a fixed use of the term (at least in the case of) εἶδος to refer to fixed species had Aristotle employed Method 1 exclusively in his biological works. Second, it is also not surprising that any kind of essentialism that is based on fixed species is nowhere to be found in his biological worksFootnote ³⁰ (again for the same reason that for the most part he employed Method 2). Third, the question of whether or not Aristotle was committed to classifying animals (whether the classification was fixed or otherwise)Footnote ³¹ may not be settled, unless we can reconstruct Aristotle's biological works by employing only Method 1 (or at least until we attempt to imagine what such works will look like). Suppose Aristotle decided to use Method 1 exclusively. He painstakingly and tediously examined each species separately and explained the causes of each of the attributes (such as sleep, respiration, growth, etc.). Assuming that he (or for that matter any human being) can live long enough to be able to accomplish such a mammoth task, the result, of course, would be a compendium of enormous absurd repetition of similar explanation ad nauseam. Would such a work result in any kind of classification of animals at all? Or perhaps the question of classification only arises because Aristotle employed Method 2 at various levels of genera (again fixed or otherwise) such that it appears as though he was interested in classification of animals. It is obviously beyond the scope of this article to adequately deal with resolving this issue here.

For now, the main points I want to drive home are: first, it is premature to give up our traditional understanding of Aristotle's theoretical commitment to fixed species (and what appears to be the deep metaphysical and epistemological commitment he invested in their existence) in biology found in Book 1 of Part. an. (as indicated by his apparent preference for Method 1—the ‘best’ method); and second, his actual practice found in his other biological writings (as displayed by his predominant employment of Method 2) may not be a reliable source in determining whether or not he held the view of fixed species as a part of his theoretical commitments.