[TEXT NOT REPRODUCIBLE IN ASCII] and division in Aristotle's Generation of Animals.
Concerning the role and significance of the greatest kinds, Arthur Peck suggests that Aristotle is 'clearly ... not intending to produce any taxonomic scheme.' Rather, he thinks Aristotle 'is taking the obvious large, main groups as a convenient basis for marshalling his material.' (2) Aristotle does make use of the greatest kinds in organizing his discussion of the various features of animals. However, most scholars (perhaps including Peck), take the identification of the greatest kinds to be significant to Aristotle's project in other ways as well. Despite this, there is no agreement concerning the significance or role of the greatest kinds in Aristotle's zoological works. There is even no agreement as to whether the identification of these greatest kinds is supposed to be a starting point or a goal for the Aristotelian biologist.
David Charles, for one, argues that Aristotle has two main aims in his History of Animals: to 'separate distinct genera and (where required) species and to determine which properties belong per se to them.' (3) On this interpretation, the identification of genuine animal kinds, including greatest kinds, is a goal of Aristotle's historia.
Charles reads the list of greatest kinds in History of Animals I 6 as originating in common usage. Nevertheless, he holds that Aristotle does not merely accept this list for his investigation, but that he works to provide 'a justification for treating these [i.e., the greatest] genera as correctly marked out by common usage.' (4) On this view, Aristotle 'reconciles himself to the standard beliefs provided they are correct ... [and] he appears to take as correct only those popular claims which are underwritten (in some way) by his scientific or metaphysical theory.' (5)
According to Charles, '[o]ne should not conclude from the incompleteness of the taxonomy provided in the Historia that Aristotle had no interest in classifying the genera and species to which general features belong.' (6) If Charles is right, the classification of animals is inseparable from the task of explaining the causes of their various features. As he demonstrates this, using the example of fish, we come to know simultaneously that (a) that fish constitutes a genuine kind and (b) the cause of the various features common to fish (e.g., possession of fish parts, the manner in which a fish eats and breathes, and so on) when we identify the essential fish feature (which Charles identifies as the activity of swimming). This is because the identification of the essential fish feature confirms that the animals we call fish constitute a genuine [TEXT NOT REPRODUCIBLE IN ASCII], and because the essential feature unites all the other features common to fish as such. Thus, on Charles' interpretation, one of the Aristotelian scientist's goals is classification. Without this, explanation would not be possible.
In contrast, Allan Gotthelf argues that Aristotle takes the identification of the greatest kinds as starting point--rather than a goal--of the History of Animals. This interpretation, which builds on the work by David Balme, posits that the 'division of animals into large classes here is a working one for convenience of achievement of [Aristotle's] actual goals.' (7) Aristotle's History of Animals records the many similarities and, especially, the many differences among animals and focuses on differences among their parts, habits, modes of living, and actions. (8) According to Gotthelf's interpretation, one main purpose of Aristotle's History of Animals is 'to identify the widest class that possesses the feature under discussion.' (9) Frequently, the widest class possessing a feature corresponds to one of the greatest kinds. A study of feathered wings, for example, is restricted to a discussion about the parts of birds: for all animals with feathered wings are birds (HA I 5, 490a13). Yet, as Gotthelf's study reveals, the widest class possessing a given feature is sometimes wider or sometimes narrower than the greatest kinds identified in History of Animals I 6. The possession of lungs, for instance, is common to all animals that breathe air in and exhale it out (HA II 15, 506a1). This group of animals, which includes man, cetacea, four-footed animals, snakes and birds, is wider than any of the greatest kinds. In contrast, the possession of spurs is restricted to some, but not all, of the birds (HA II 12, 504b7). Gotthelf, following Balme, thus emphasizes the way in which the work in History of Animals serves Aristotle's causal explanations of the various features of animals as these are 'later' provided in Parts of Animals, Progression of Animals, and Generation of Animals. (10) Though Aristotle's discussion in History of Animals frequently proceeds kind-by-kind, according to the greatest kinds, we can make sense of departures from the kind-by-kind survey on Gotthelf's interpretation too. Both Balme and Gotthelf insist that Aristotle is not concerned with producing a classification of animals. (11) In this respect, their view is in opposition to that of Charles.
The first aim of the current work is to clarify the nature and significance of Aristotle's greatest kinds. In Section I, I aim to answer the question, "What makes a kind a greatest kind?" On the assumption that not all kinds are to be identified as greatest kinds, I work to pick out what distinguishes the greatest kinds of animals from other genuine kinds. I consider the three passages in which we find explicit reference to greatest kinds: History of Animals I 6, 790b7, History of Animals II 15, 505b30, and Parts of Animals IV 8, 683b26. Focusing on some of the interpretive challenges raised by these passages, I argue that the greatest kinds within a specified domain are the most general kinds containing multiple forms. Consideration of these same passages leads me to argue, further, that it does not matter whether Aristotle labels any specific kind a 'greatest kind' for the reason that Aristotle does not use 'greatest kind' as a technical term or, if he does, it is a technical term of little significance in his zoological investigations. I maintain that the kinds Aristotle identifies are significant, but that it does not matter whether each of these kinds is counted as [TEXT NOT REPRODUCIBLE IN ASCII].
Next, I consider the fact that Aristotle makes regular use of certain kinds of animals in the organization of the biological treatises. It is now well established that certain kinds, those seven identified as greatest kinds in History of Animals I 6 (12) in addition to humans, four-footed live-bearers, four-footed egg-layers and snakes, feature into the organizational structure of History of Animals and Parts of Animals. (13) In Section II, I comment on two extended discussions in the Generation of Animals, with the goal of making clear the extent to, and the manner in which, these kinds factor into the organization of Aristotle's treatise on animal generation. I demonstrate that, although these kinds do play a conspicuous role in the organization of Aristotle's explanations of the reproductive parts of males, and females, and of the various modes of reproduction, the discussion of each of these phenomena is organized, primarily, according to divisions specifically related to the cause of the features being explained. As a result, we find that Aristotle sometimes groups many kinds into a single class and sometimes divides a single kind, including those he calls greatest kinds, into more than one class.
In Section III, I explain the relationship between the various kinds of animals and the divisions around which Aristotle organizes his explanations. Despite the fact that the divisions of animals in Generation of Animals do not correspond directly to the divisions among the greatest kinds, there is significant overlap. I argue that Aristotle's appeal to commonly recognized kinds is, for one thing, a matter of convenience or efficiency. I then work to explain the overlap between the divisions Aristotle identifies in his account of animal generation and the greatest kinds. Finally, I aim to reconcile, to the extent possible, the competing interpretations set out above. I argue that the classification of animals is a goal for the Aristotelian biologist in so far as Aristotle's classification of animals into the widest classes possessing the various animal features is necessary for meeting the biologist's primary goal of explaining these features.
I [TEXT NOT REPRODUCIBLE IN ASCII]
Of the three passages in the Aristotelian corpus speaking of greatest kinds, the most extensive is found in History of Animals I 6. The passage runs as follows:
The greatest kinds of animals into which other animals are divided are: one of birds, one of fish, another one of cetacea. Indeed, all these are blooded. Another kind is of hard-shells, called shellfish; another [kind is] of soft-shells, unnamed with a single name, such as crayfish, kinds of crabs and lobsters; another [kind is] of softies, such as squid, larger squid and cuttlefish; another is of insects. But all of these are bloodless; as many of these that have feet are many-footed; and some of the insects have wings [or: are able to fly]. Of the remaining animals there are no further great kinds; for in one form (14) there aren't many forms, for a form is simple itself having no differences, such as man; or if it has differences, the forms are unnamed. For, all the four-footed and wingless are blooded, but some of these are live-bearers and others are egg-layers. For, on the one hand, as many as are live-bearers all possess hair, on the other hand, as many as are egg-layers [all possess] horny-scales; the horny-scales being similar in position with [fish] scales. Footless by nature is the blooded, land-dwelling kind of the snakes; this is horny-scaled. But, while other snakes are egg-layers, the viper alone is a live-bearer. For not all live-bearers possess hair; for certain of the fish also bear-live; But then for as many as possess hair, all of them bear-live ... There are many, but unnamed, forms of the kind of the four-footed and live-bearing animals; though each on its own is named, just as man is named, lion, deer, horse, dog, and the others are in turn, and yet there is one certain kind alone called the bushy-tails, such as horse, ass, mule, jennet, and the one from Syria called half-ass. (490b7-1a3) (15)
In this passage, we are told that the greatest kinds into which animals are divided are: birds, fish and cetacea--all of which are blooded--along with hard-shells, soft-shells, softies and insects, which are bloodless. In Parts of Animals I 4, Aristotle observes that birds, fish and other animal kinds (he does not specify 'greatest kinds' though this is sometimes assumed (16)) are commonly recognized by morphological similarities and differences (PA I 4, 644b7). Each of these kinds contains multiple forms and animals of a single kind have parts that differ only according to the more and the less (HA 11, 486a21-2; PA I 4, 644b7-15). For instance, all animals of the bird kind have beaks, wings, two legs that fold inwards, and the other features that distinguish birds from other animals. The various forms of the bird kind differ from each other according to variations within these bird features. Some have shorter wings, while some have longer wings; some have longer beaks, while others have shorter beaks; and so on for all of the other bird features. (17)
Some of the kinds listed as greatest kinds in History of Animals I 6 are not only recognized by popular usage but have a single name. This is true of birds and fish. Aristotle identifies other kinds using a descriptive label while noting that they have no single name. Here, Aristotle must mean that they are not ordinarily, or commonly, called by a single name though we can certainly give them a label in the way Aristotle uses [TEXT NOT REPRODUCIBLE IN ASCII] to speak of the kind that contains crayfish, crabs, and lobsters. Aristotle maintains that each unnamed kind --like the named kinds--contains multiple forms. (18)
Aristotle's remark at 490b15-16, that '[o]f the remaining animals there are no further great kinds ([TEXT NOT REPRODUCIBLE IN ASCII]),' suggests that his list of seven greatest kinds is exhaustive. Problems arise, however, if this list is meant to be exhaustive. I consider two of these problems, starting with the general problem of identifying what Aristotle takes to be the nature of the greatest kinds in such a manner that they can be distinguished from other genuine animal kinds.
We find some help for identifying the defining features of the greatest kinds in the assertion that follows Aristotle's claim that there are no further great kinds: 'for in one form there aren't many forms, for a form is simple itself having no differences, such as man; or if it has differences, the forms are unnamed' (HA I.6, 490b15-16). Thus, this assertion constitutes the explanation for why there are no further great kinds. (19) The explanation is that, for the remaining animals, we do not identify one form containing many forms. Aristotle makes clear two cases in which we do not identify one form containing many forms. The first is when the form in question is simple and has no differences. Presumably, Aristotle means that we do not have a greatest kind when the form in question lacks differences that would support the identification of subordinate forms, i.e., when the form in question is atomic, like 'man'. The second is when there are differences and, thus, identifiable subordinate forms, but these forms are unnamed.
Working backwards, we can identify a condition for the identification of a greatest kind. The condition is that a greatest kind is such that in one form we do identify many, named, forms. In other words, the condition is that the one form in question must also be identified as a kind containing many forms that have common names. (20) This brings us to the task of identifying the 'one form' in question, which is no easy task.
It is tempting to interpret Aristotle to be saying that each of the greatest kinds must contain subordinate forms that are, in themselves, kinds containing subordinate forms. On this interpretation, we would expect to find named forms lying between each greatest kind and the atomic species. 'Soft-shelled animals', for example, would meet this condition just in so far as it contains intermediate kinds, including the kind of the crabs that contain multiple forms, i.e., spider crabs, Heracleotic crabs, and the rest. (21)
This interpretation is tempting at least for the reason that it clarifies the distinction between kinds and greatest kinds. On this interpretation, the distinguishing feature of the greatest kinds is that each contains intermediate kinds that are named. (22) This interpretation must, however, be rejected because it is not consistent with Aristotle's use of the term 'greatest kinds' in Parts of Animals IV 8. There, Aristotle offers the following comment about the soft-shelled animals:
There are four greatest kinds of them [i.e., soft-shelled animals], called crayfish, lobsters, prawns and crabs; and of each of these kinds there are many forms, differing not only in shape but also greatly in size; for some are large while others are altogether tiny. (683b26-30) (23)
In this passage, the four kinds listed (i.e., crayfish, lobsters, prawns, and crabs) are called [TEXT NOT REPRODUCIBLE IN ASCII]. This is initially surprising given that Aristotle lists the soft-shelled animals as one of the greatest kinds in History of Animals I 6, and that the four kinds highlighted in this passage are intermediate kinds--those that are subordinate to an identified wider kind and containing multiple forms. Although each of the four kinds contains multiple forms that are commonly named, there is no indication that the subordinate forms have common names and there is no indication that any of the forms contained in the four kinds listed, themselves, contain forms. In the same chapter, we find, for example, two forms of crab for which names are given: spider-crabs ([TEXT NOT REPRODUCIBLE IN ASCII]) and Heracleotic crabs (684a10). (24) The problem for the interpretation being considered is that each of these subordinate forms is simple in the sense that, like 'man', it does not contain multiple forms.
I propose that we interpret Aristotle to be suggesting in History of Animals I 6 only the condition that each greatest kind must have in it multiple and named forms. 'Bird' meets this condition just in so far as it contains 'curlew', 'cardinal', and the many other forms of birds. 'Fish' meets this condition just in so far as it contains many forms of fish. On this interpretation, it is not necessary that a kind must contain intermediate kinds in order to be counted among the greatest kinds.
One of the strengths of the proposed interpretation is that it allows us to understand Aristotle's use of [TEXT NOT REPRODUCIBLE IN ASCII] in the passages cited from History of Animals and Parts of Animals. Not only 'soft-shelled animals', but also 'crayfish', 'lobsters', 'prawns', and 'crabs' can be called greatest kinds in so far as all of these kinds contain multiple, named, forms.
Yet, on first glance, the interpretation I have proposed provides no grounds for distinguishing the greatest kinds from other genuine kinds. Whatever counts as a kind in the logical sense outlined in Categories and Topics (i.e., a kind in which we find multiple forms) also counts as a greatest kind. For instance, if 'hawk' is a form that is a kind in so far as it contains multiple and distinct forms that are named, then it would also meet the requirement of being a greatest kind. At most, then, the identification of greatest kinds according to this interpretation would seem to allow us only to distinguish indivisible forms (which are often referred to as [TEXT NOT REPRODUCIBLE IN ASCII] (25)) from those that are also kinds in the logical sense.
On a closer look, however, the distinction between kinds and greatest kinds is clear. Aristotle's use of the superlative ([TEXT NOT REPRODUCIBLE IN ASCII]) indicates that the kinds identified are the greatest, widest, or most extensive kinds within a certain domain. (26) Thus, the domain becomes a determining factor in which kinds of animals we must identify as the greatest kinds. When considering the division of animals, the greatest kinds in which we find animals divided are those identified in History of Animals I 6, including birds, fish, and the rest. When considering the division of the soft-shelled animals (as opposed to all animals), the greatest kinds into which these animals are divided are those listed in Parts of Animals IV 8, including crayfish, lobsters, prawns, and crabs. In both cases, the kinds identified are the greatest or widest within the domain specified: animals in the first and soft-shelled animals in the second.
Even with an interpretation that allows us to grasp the nature of the greatest kinds and distinguish them from other animal kinds in abstract terms, a second problem persists. It is that Aristotle seems to think of certain kinds of animals, including 'four-footed live-bearers' and 'four-footed egg layers', as greatest kinds despite the fact that History of Animals I 6 seems to rule this out. (27)
In addition to the regular use Aristotle makes of the 'four-footed, live-bearing' and the 'four-footed, egg-laying' kinds in the organization of the zoological works, History of Animals II 15 is thought to be the best evidence that he thinks of both of these as greatest kinds. Against this tradition, it must be realized that the passage is ambiguous (at best). The passage follows Aristotle's claim that he has now described the external parts of the blooded animals and signals that he is moving on to the internal parts. The relevant text runs as follows:
... we must speak of the internal parts possessed in the blooded animals first; for these greatest kinds differ from the other remaining animals, they are blooded and the others are bloodless. These ([TEXT NOT REPRODUCIBLE IN ASCII]) are both the egg-laying and live-bearing of the four-footed animals, and birds and fish and cetacea, and each other one that is not a kind but a simple form for each instance, such as man. (505b25-9) (28)
It seems that this passage can only be taken as definitive evidence that the four-footed live-bearers and the four-footed egg-layers are greatest kinds if we read [TEXT NOT REPRODUCIBLE IN ASCII] at line 28 to refer to the greatest kinds (or even the greatest kinds that are blooded). However, the passage makes better sense if [TEXT NOT REPRODUCIBLE IN ASCII] refers to 'blooded animals' for the reason that the list includes the simple forms, such as man, that are not properly kinds in the logical sense (let alone greatest kinds). (29) Thus, although both the four-footed egg-laying animals and the four-footed live-bearing animals are listed in this passage, they are listed as examples of blooded animals rather than as examples of greatest kinds. Nothing in this passage unambiguously demands that we think of them as among the greatest kinds.
Consideration of the problems related to questions concerning the distinction between the greatest kinds and other animal kinds gives rise to questions related to the significance of the distinction. Given that there is no single, unambiguous passage that identifies four-footed live-bearing or four-footed egg-laying animals as greatest kinds, we might ask, "Does it matter whether Aristotle held that these were greatest kinds?" To this, I think the answer is no. This negative answer does not entail that these kinds are insignificant; only that it is insignificant whether they are identified as greatest kinds. As I demonstrate below, the kind of four-footed live-bearers and the kind of four-footed egg-layers do play a conspicuous role in the organization of explanations in Aristotle's biological treatises. Their role is no less conspicuous than that of the seven greatest kinds explicitly identified in History of Animals I 6. Owing to the truth behind this statement, we might further ask, "Does it matter that any of the kinds Aristotle recognizes are identified as [TEXT NOT REPRODUCIBLE IN ASCII]?" Perhaps, surprisingly, given the focus of this study and the attention that the greatest kinds are given in the literature, my answer to this question is again no.
There are a number of considerations supporting the answer that it does not matter that some kinds are explicitly identified as greatest kinds while others are not. It does not matter for the reason that Aristotle does not consider the term 'greatest kind' a technical term or, if it is a technical term, it is one with very little significance for us as we study Aristotle's zoological investigations. After all, Aristotle does not use the term 'greatest kinds' very often. In fact, precisely three passages can be cited, and these are the ones cited above: HA I6; HA II 15; and PA IV 8.
Another reason to think that 'greatest kind' is not a technical distinction of special importance is that the passage that invites us to think of greatest kinds in a technical sense is followed by the remark that the preceding statements are put forward in outline only, and the promise that the discussion that follows (presumably, in the remainder of the HA) becomes clearer and more accurate (HA I 6, 491a6-10). These comments are intended to clarify that all of the preceding comments, including those of History of Animals I 5-6, are put forward in outline only. Ignoring the chapter breaks, which are not Aristotle's anyway, this puts the comments about greatest kinds on par with a scattered list of observations about the animals to be described in a more systematic way in the remainder of the History of Animals. The remark that these statements are put forward in outline only and Aristotle's emphasis on whether or not each of the kinds has a single and commonly used name suggest that the discussion of greatest kinds in History of Animals I 6 is merely intended to identify those kinds of animals that are commonly recognized. This does not necessarily entail that 'greatest kind' is not used as a technical term. It does, however, suggest that we should expect a clearer and more accurate statement of what it is to be a greatest kind if, indeed, this is significant. No such statement is to be found in the corpus.
The most compelling reason for my claim that it does not matter whether certain kinds are identified as greatest kinds is that nothing hangs on the identification. Especially given that Aristotle does not refer to greatest kinds using this term outside of the passages cited, it is interesting to ask, "Why would Aristotle mention the greatest kinds in a manner that invites us to think they are significantly, even technically, different from other kinds?" To this question, I have no answer. The designation is of no real consequence to the student of Aristotle's biology. Nevertheless, I maintain that it is important for us to recognize the various kinds and their role in the zoological works--it just does not matter whether or not the label 'greatest kind' applies. To clarify the importance of the various kinds, I turn my attention in the next section of this paper to Aristotle's use of kinds and division in the biological treatises.
II Kinds and Division in Generation of Animals
The manner in which Aristotle proceeds in each of the biological treatises is the result of considerable thought. In Parts of Animals I 1, he asks the methodological question of 'whether one should study things in common according to kind first, and then later their distinctive characteristics, or whether one should study them one by one straight away' (639b4-6, trans. Lennox). The question is whether we should explain the various features of animals form-by-form or kind-by-kind. In Parts of Animals I 5, Aristotle provides his answer: 'It is necessary first to divide the attributes associated with each kind that belong in themselves to all the animals, and next to try to divide their causes' (645b1-3, trans. Lennox). One motivation for proceeding kind-by-kind is that it avoids the repetition that would be inevitable if we were to proceed form-by-form (PA I 1, 639a22; 1 5, 645b11-12). As we would expect given these comments, the observations about, and explanations of, the various features of animals presented in History of Animals and Parts of Animals proceed kind-by-kind, for the most part. (30) Commenting on the History of Animals, James Lennox suggests, '[i]nsofar as animal groups are referred to, they serve as a general way of ordering the discussion. Animal groups below the level of Aristotle's "great[est] kinds"... are referred to only to illustrate a certain differentia or correlation between differentiae.' (31)
In contrast, only two parts of the Generation of Animals are organized, in a primary way, around divisions of animals into kinds. By this, I mean that only two parts of this treatise are organized according to specific divisions among, and explicit groups of, animals. The first combines the discussion concerning the presence and nature of the male and the female and the subsequent discussion of the reproductive parts in Book I. (32) The second consists of the discussion concerning the various modes of generation in Books II and III.
The recognition of greatest kinds is not the central organizing feature of either of these parts of the Generation of Animals. Rather, the divisions Aristotle employs in organizing his discussion of the various features of animals are those that best serve his explanatory purposes, that is, his divisions correspond to differences related to the cause of each feature being explained. Considering the reproductive parts of males, Aristotle explains first that some animals, including the live-bearing ones (GA I 4, 717a32-b5), in addition to birds and four-footed egg-layers (GA I 4, 717b5-14), possess testes because it is better for these animals to have testes (GA I 4, 717a15-21). (33) Aristotle then explains that other animals, including fish (GA I 6, 717b34-18a16) and snakes (GA I 7, 718a17), lack testes owing to necessity (GA I 6, 717b33-4).
Aristotle's explanations of the reproductive parts of females are organized around their final cause or that for the sake of which these parts exist: reproduction in its various modes. Differences in the reproductive parts, including uterine shape and, especially, uterine position, correspond to the different modes of reproduction (GA I 3, 716b32-717a10; GA I8-11). In turn, Aristotle's explanations of these modes of reproduction (starting at GA II 1, 732a26) emphasize the causes of their similarities and differences: heat and fluidity.
The order in which Aristotle treats of the modes of reproduction follows the one Aristotle finds in nature: '[w]e should notice how well Nature brings generation about in its several forms: they are arranged in a regular series ...' (733a34-b1, trans. Peck). From the most perfect to the least, the series is as follows: (1) generation of perfect, live offspring resembling the parents; (2) generation of perfect eggs internally and perfect live offspring externally; (3) generation of perfect eggs; (4) generation of imperfect eggs; (34) and (5) generation of larvae. According to Aristotle, those animals that are more perfect (i.e., compared with other animals) partake of a purer principle: all live-bearing animals draw in pneuma (732b30). These animals are, by their nature, hotter and more fluid (732b32-3) and the test for a hot and fluid nature is the presence or absence of a lung that is soft and well supplied by blood. Animals that are deficient in either heat or fluidity generate less than perfect off-spring (e.g., eggs or larvae).
As the result of Aristotle's divisions in Generation of Animals, we frequently find more than one of the commonly recognized kinds in a single class. In considering the reproductive parts of males, for instance, Aristotle groups together the birds and the four-footed egg-layers (717b5-14). He also groups together all of the live-bearing animals (717a32-b5); aside from scattered comments concerning variations in the reproductive parts of certain species, Aristotle treats the live-bearing animals together without distinction. When considering the reproductive parts of females, Aristotle further divides the live-bearing animals to recognize differences between those animals that give birth to live offspring, but lay eggs internally (GA I 10, 718b33-I.11, 719a13) and those that are both internally and externally live-bearing (GA I 11, 13-29). (35) Even there, we find more than one of the commonly recognized kinds in each of the resulting divisions. The former includes some (i.e., the cartilaginous) fish and some snakes (i.e., vipers). We are told the latter includes humans, horses, dogs, and all haired animals along with dolphins, whales and other such cetacea.
Though Aristotle's divisions in Generation of Animals frequently result in classes containing two or more of the commonly recognized kinds, these divisions cut across the recognized kinds as well. In considering both the reproductive parts of females among the blooded animals and the various modes of reproduction, the kind of fish is divided with the cartilaginous fish (that give birth to live offspring) and the scaly fish (that lay eggs) falling into different classes. Similarly, the kind of the snakes is divided with vipers in a class with other live-bearing animals and all other snakes in another class with other egg-laying animals.
Aristotle's comments about the reproductive parts of bloodless animals entail that a complete account of the generation of insects will involve dividing insects into no less than three classes. This is because some insects reproduce in the manner of the blooded animals that produce offspring resembling the parents (e.g., locusts, cicadas, spiders, wasps, and ants); some copulate and produce larvae out of putrefying matter (e.g., fleas, flies, and Spanish flies); and some neither copulate nor generate out of animals (e.g., gnats and mosquitoes) (GA I 16, 721a2-10).
That Aristotle's divisions cut across the commonly recognized kinds (fish, snakes, and insects) demonstrates that Aristotle does not think of these kinds as having inviolable boundaries. Moreover, that two of these (fish and insects) are among the greatest kinds explicitly recognized in History of Animals I 6 indicates that being a greatest kind of animal does not entail any privileged status--even these can fall into multiple divisions. (36) Yet, although they are not given priority in the organization of the discussion in Aristotle's Generation of Animals, the commonly recognized kinds are certainly conspicuous. Thus, we might wonder about their conspicuity. I turn to this in the next section.
III [TEXT NOT REPRODUCIBLE IN ASCII] in Generation of Animals
One obvious reason for Aristotle to mention the greatest kinds in the explanations of reproductive parts and modes of reproduction is efficiency. In noting how each of these kinds fits into the divisions used in each section of the text, Aristotle makes a distinction as to which animals he is describing without having to list the individual forms for each (especially if, as I have suggested above, the kinds Aristotle emphasizes are the popularly recognized ones). For instance, it is more efficient to indicate that birds are egg-layers than to list each and every species of bird one after the other. Given that all birds lay perfect eggs in the same way, a single explanation works for all of them, and Aristotle avoids the repetition that would result from proceeding form-by-form.
Evidence that Aristotle uses the so-called greatest kinds as a convenient basis for marshalling his material, as Peck suggests, is also found in the discussion of the reproductive parts of males and females among the bloodless animals. This discussion begins at Generation of Animals I 14, 720b2 with Aristotle's announcement that there are four kinds of bloodless animals to handle. He lists the soft-shelled animals, the softies, the insects and the hard-shelled animals, and then notes that he is not certain about all of the details concerning copulation and the sexes for these. Aristotle does 'marshal' the material as he briefly outlines facts about the soft-shelled animals, the softies, and the insects. Due to the fact that the hard-shelled animals do not generate by copulating (GA I 1, 715a26), we do not find in them males and females, thus, there is no need for them to be discussed in this section of the text.
Although appealing to the greatest kinds can help in marshalling the material, Aristotle does not always indicate which animals fall into each division by referring to the greatest kinds. Frequently, his practice is to indicate a few examples of the animals in each division. In the discussion of the reproductive parts of males, for instance, he specifies that the class of animals with testes includes 'horses and other such animals' and man (GA I 4, 717a32-b5). Similarly, in the discussion of the various modes of reproduction, Aristotle groups together 'men, horses, cattle, and of marine animals dolphins and the other cetacea' (GA II 1, 732a28-9). In these instances, it would have been more efficient to simply identify which of the commonly recognized greatest kinds fall into each division, yet, for whatever reason, Aristotle does not do this consistently.
Furthermore, in the context of Generation of Animals, marshalling the material according to the commonly recognized kinds often will not do. It will not, for example, serve Aristotle's explanatory purposes to take the fish together and the snakes together when considering the various modes of reproduction. Not only would doing so be inefficient (for it would require stating the same causes as many times as there are kinds with the feature to which the explanation applies (37)), worse, it would obscure the level of generality at which certain causal explanations apply, which is a significant concern for the Aristotelian scientist. (38)
Another reason for the conspicuity of the greatest kinds is that there is a good deal of overlap between these kinds and the divisions according to the cause of the reproductive parts and modes of reproduction. Though I have been emphasizing instances in which Aristotle's divisions group together or cut across the commonly recognized kinds, it is nevertheless true that each of these kinds usually corresponds to one of the divisions used to organize the passages studied. The result is that, though we sometimes find many kinds grouped together (e.g., humans with the four-footed live-bearers, birds with the four-footed egg-layers, and so on), Aristotle often moves from one of the commonly recognized kinds to the next in the discussion. This is seen, for instance, in the explanation of the generation of perfect eggs, where Aristotle addresses the birds before the egg-laying four-footers. Even when a kind is divided, Aristotle can sometimes consider all members of the kind in subsequent sections of the text. This is seen, for instance, where Aristotle considers the means by which all fish reproduce in subsequent chapters (GA III 3-4) despite significant differences in their modes of reproduction.
The order in which Aristotle explains fish reproduction appears, at first, to give priority to the kind of the fish over divisions according to the modes of reproduction being explained. If Aristotle were to have ordered the discussion around the modes of reproduction working exclusively from the most perfect to the least perfect, we would expect him to explain all the animals that give birth to live offspring before moving on to the egg-laying animals. Thus, this would put the explanation of the cartilaginous fish before both the birds and the four-footed egg-laying animals in so far as the cartilaginous fish bear live young and are more perfect than the others which generate perfect eggs. We would further expect Aristotle to explain the other fish after the birds and other egg-layers in so far as the other fish generate only imperfect eggs. However, these expectations are not met. He proceeds by explaining the live-bearing animals then the perfect egg-layers before explaining all of the fish.
Given that this seems to be the only diversion from the order moving from more to less perfect generation, why does Aristotle set out the order of the modes of generation and then proceed in a slightly different order? To this question, I can think of two answers. The first is that Aristotle wants to go through the animals kind-by-kind and avoid dividing a kind by treating the fish together. However tidy this may be, the fish kind is nevertheless still divided by their modes of reproduction.
The second answer is that, by holding the discussion of cartilaginous fish until both the live-bearing animals and the birds are discussed, Aristotle saves himself a good deal of repetition. The cartilaginous fish bear live offspring and generate perfect eggs (albeit internally). By the time Aristotle explains the generation of offspring in the cartilaginous fish, the process involved for both (i.e., perfect egg-laying and live-bearing) have already been described. Since the process of generating perfect eggs has been explained in the section on birds, all that remains is to set out the differences between the birds and cartilaginous fish, and this is exactly what Aristotle does (III 3, 754b20-34). Aside from the differences noted there, Aristotle tells us that the generation of eggs in birds and in cartilaginous fish works in the same way (755a1-2). Thus, by placing the discussion of cartilaginous fish after those of live-bearing animals and egg-laying birds, the rather curious process of generation in cartilaginous fish is easier to grasp. I suspect, then, that Aristotle chose to explain the birds first in so far as they are more familiar to us than the cartilaginous fish or in so far as birds only lay eggs. If this is right, we have as a happy coincidence the benefit of explaining all of the fish in successive chapters.
There are other reasons for the relative overlap between the divisions Aristotle uses to shape each section of the Generation of Animals and the commonly recognized kinds. In the case of the discussion about the reproductive parts of males and females, I think the overlap is easy to explain if we remember that the commonly identified kinds are morphological kinds in so far as each contains multiple forms, with most of their parts in common (PA I 4, 644b7). Due to the fact that each of the kinds is identified according to the similarity of their parts, it should be no surprise to find that these kinds are not cut across when Aristotle considers the reproductive parts. In the same way, we should not be surprised that the bird kind is not divided if we are to consider the parts with which animals eat--to be a bird is, in part, to have a beak where other animals have mouths.
Perhaps less obvious is the reason we find the greatest kinds intact for the most part in the discussion of the various modes of reproduction. Here, too, the explanation goes back to the fact that the kinds Aristotle identifies are those whose subordinate forms have parts that differ only according to the more and less: to understand why demands that we recall Aristotle's insistence that--for the most part--the various parts of animals are for the sake of some function (PA 1 5, 645b16-20). (39) In Generation of Animals, Aristotle is explicit that the reproductive parts are for the sake of generation (see, for examples: GA I 2, 716a23 and I 4, 717a12-15). Thus, given the similarity of parts we find among each kind, it should not be surprising that the functions for the sake of which the animals that have these parts are also similar and vice versa. (40) There is a causal relation between the parts and the modes of reproduction. Indeed, and as I have noted above, this is the fact that informs Aristotle's divisions. This is most obviously seen in Aristotle's explanation of the female reproductive parts, in which differences among the various kinds of animals correspond to the animals' modes of reproduction. In the female, the function of the parts is related to the development of the embryo, egg, or larva. Insofar as the commonly recognized kinds are identified by the similarity of parts, and the parts are similar whenever they are for the sake of the same thing (in this case, the specific mode of reproduction), it should be no surprise that there is little cross-division of these morphological kinds in Aristotle's explanations.
On the face of it, however, not all of the kinds Aristotle identifies are strictly morphological. The four-footed egg-layers and the four-footed live-bearers are somewhat unique. Both of these kinds are identified according to their parts (or in this case, the number of feet), which is a morphological feature, and their mode of reproduction, which is an activity ([TEXT NOT REPRODUCIBLE IN ASCII]) related to their mode of living ([TEXT NOT REPRODUCIBLE IN ASCII]). Moreover, neither of these kinds has a common name in that 'four-footed egg-layer' and 'four-footed live-bearer' are descriptive appellations rather than commonly used names. In contrast, [TEXT NOT REPRODUCIBLE IN ASCII] is commonly used to refer to certain beasts and is, anyway, an obvious differentia. (41) However, it fails to pick out a genuine kind of animals. Among the four-footed animals, many features are analogous rather than similar. For instance, some have hair while others have spiny-scales (490b21). Thus, if there are genuine kinds of four-footed animals, Aristotle is left to identify the differentia/e most useful for explaining the various features of each kind. He does just this when he divides the four-footed creatures according to their relative modes of reproduction; for this feature corresponds to many similarities among those animals including the possession of hair or spiny scales. Thus, the kinds that are differentiated correspond to morphological similarities and differences, even though the differentia and the descriptive appellation are explicitly tied to the different modes of reproduction. This is because among four-footed animals, the live-bearing feature is co-extensive with many other features and the egg-laying feature is co-extensive with still many others. This observation provides us with a way to understand some additional details of the first passage I cited from History of Animals I 6.
Consider again the following extracts from the passage that might first appear to be throw-away comments:
... But all of these are bloodless; as many of these that have feet are many-footed; and some of the insects have wings [or: are able to fly] ... For, all the four-footed and wingless animals are blooded, but some of these are live-bearers and others are egg-layers. For, on the one hand, as many as are live-bearers, all possess hair, on the other hand, as many as are egg-layers, [possess] horny-scales; the horny-scales being similar in position with [fish] scales. Footless by nature is the blooded, land-dwelling kind of the snakes; this is horny-scaled. But, while other snakes are egg-layers, the viper alone is a live-bearer. For not all livebearers possess hair; for certain of the fish also bear-live; But then for as many as possess hair, all of them bear-live ...
These comments invite us to consider the relations among co-extensive features. Do we find a (causal) connection between being many footed and being bloodless? That all bloodless animals that have feet are many-footed suggests this possibility. Is there a (causal) connection between being a four-footed live-bearing animal and the possession of hair? That all four-footed animals that are live-bearing have hair also suggests this possibility. Here, four-footed is too broad in that not all four-footed animals have hair. Moreover, live-bearing is too broad in that not all live-bearers have hair. (42) Is there a (causal) connection between being a four-footed egg-laying animal and the possession of horny-scales? That all four-footed animals that lay eggs have horny scales suggests this possibility.
Since the features being explained do not always map directly onto the commonly recognized kinds of birds, fish and the others, the divisions Aristotle uses to explain each of these features sometimes group together--and sometimes divide--the commonly recognized kinds, including those he identifies as greatest kinds. Thus, it cannot be that the classification of animals according to these kinds is the ultimate goal of the Aristotelian biologist. If it were, we should be struck by Aristotle's failure to present a single genus/species tree. (43)
Nevertheless, Aristotle does regularly divide animals according to the various features being explained. As my study of Generation of Animals indicates, the divisions vary depending on the specific feature being explained, such as male reproductive parts, female reproductive parts, or mode of generation. Thus, as Gotthelf argues to be true of the discussion in History of Animals, one of Aristotle's goals in the Generation of Animals is to identify the widest class of animals having each specific feature being explained. It is not clear, however, whether or not Aristotle's division of animals into the greatest kinds is, as Gotthelf says, 'a working' division for the convenient achievement of Aristotle's actual goals. For the most part, nothing suggests that Aristotle is working from these divisions towards identifying wider or narrower classes. Rather, he starts each section with the division according to the feature being explained in that section. Any mention of the greatest kinds is subsequent to the initial division and indicates into which division each of greatest kinds fall. As I have noted, the only exception lies in his discussion of the male and the female reproductive parts among the bloodless animals in Generation of Animals I.
Do Aristotle's divisions in Generation of Animals reflect classifications of animals? My answer is yes. In so far as the divisions result in multiple and differentiated groups of animals, I think the divisions can be counted as classifications. Does this mean that the classification of animals is a goal of the Aristotelian biologist? Again, the answer is yes. Though it is certainly true that Aristotle nowhere produces, or seems wont to produce, a single taxonomy or systematics as we (after Linnaeus) use these terms, Charles is undoubtedly correct in arguing that Aristotle's use of division is related to his explanations of the various features possessed by animals. Indeed, the divisions Aristotle employs certainly result in the grouping together of animals into classes. Thus, I think Charles is right that the division and classification of animals is a goal of the Aristotelian biologist, so long as we understand that the divisions and resulting classification proceed according to the various features being explained, and that these do not map onto any division according to commonly recognized kinds without some overlap and some cross-division.
On the face of it, my view that division and classification are goals of the Aristotelian biologist is in conflict with that of Balme and Gotthelf who hold that Aristotle is not concerned with classification or taxonomy. (44) On closer review, however, my view can be seen to fit well with theirs. The results from my study of the Generation of Animals supports the Balme-Gotthelf thesis that Aristotle is concerned with identifying the widest class of animals sharing each of the various features Aristotle later explains. We find this thesis defended in Lennox's work too. Building on Balme's work, both Lennox and Gotthelf argue that the goal is not to establish the existence of natural classes, but rather to establish what they call 'widest class generalizations' by identifying the 'co-extensive predications'. (45) In so far as we agree on this, we agree that the division of animals into classes according to their shared features is a goal of the Aristotelian biologist. We can agree, further, that the identification of these widest classes is not Aristotle's final goal in the zoological treatises. The main point of difference is that I take Aristotle's divisions to result in classifications of animals and they do not.
The division of animals into the relevant classes is a goal even if it is to serve Aristotle's loftier explanatory goals. His ultimate, and explicit, goal is to explain the various features of animals. (46) There is general agreement that his explanations of facts about animals rely on division. There is further agreement that his divisions support those of the sort that serve as first principles in epistemic demonstrations. (47) Both Lennox and Greg Bayer have (independently) shown that Aristotle uses division in the process of explaining the various features of animals. (48) Charles' view is similar to the extent that he thinks Aristotle's divisions support the definitions of animals and that these definitions are needed to explain the animals' features. As Charles puts it, 'some kinds possess certain of their features in virtue of their belonging to wider-ranging kinds.' (49) Thus, Aristotle's explanations of the features involve appealing to the wider-ranging kind. Here, Lennox argues that the explanations appeal to features, rather than kinds, that are co-extensive with the features being explained, though the accounts are otherwise similar in this respect. The identification of the widest class with the feature to be explained helps in so far as it helps us identify the appropriate level of generality for understanding the feature to be explained, i.e., the level of generality at which the animals have the feature per se. Knowing that the bluejay on my feeder is a bird, for instance, does some of the work in explaining why it has a beak. (50) The next step is to identify whatever it is about birds that entails the possession of a beak.
Aristotle's aim of identifying the widest class of animals possessing a specific feature for the purpose of explaining that feature is clearly reflected by his effort in Generation of Animals II 1 (starting at 732b15) to identify a division of animals that corresponds to differences in the various modes of reproduction. In this telling passage, Aristotle observes that the various modes of locomotion and reproduction are not co-extensive. From this observation, we conclude that the various modes of locomotion do not provide the divisions needed to explain reproduction; the locomotive parts are not the cause of the differences in modes of reproduction. By implication, if we find a division that corresponds to the various modes of reproduction, we are on the path of identifying the cause or explanation of these various modes. Aristotle goes on to note that it is common to those animals that are internally and externally live-bearing that they breathe air and have moist (or blooded) lungs (GA II 1, 732b30). To know this is to know the level of generality at which we need to work in explaining the feature of bearing live offspring internally and externally. Aristotle makes it clear that this wider class of animals, including man, four-footed live-bearing animals and cetacea, is hot and moist (the possession of a blooded lung is, we learn, the 'test' of natural heat). His claim is that the possession of heat and moisture is the cause of, or explains, bearing live young; we might say that this is the common nature of live-bearing animals. In this case, the widest kind is not one of the greatest kinds, but the wider class of animals possessing lungs.
So what of the greatest kinds? If what I have argued is right, it does not matter that some kinds of animals are referred to as 'greatest kinds'. Moreover, the division of animals into the greatest kinds is not given priority in Aristotle's explanation of features related to reproduction: the reproductive parts of males and females and the various modes of reproduction. For these explanations, Aristotle explicitly divides animals into classes relevant to understanding the causes of these features. Though the commonly recognized kinds are conspicuous throughout the discussion of these features, the significant divisions are those tied to the cause of each feature being explained. That these divisions frequently group together and cut across the greatest kinds suggests that this list of greatest kinds in History of Animals I 6 is of less importance to the student of Aristotle's biology than the divisions he actually uses. Since the goal of the Aristotelian scientist is to explain the various features of animals, our first goal is to divide animals according to these features. (51)
Balme, David. "Aristotle's Use of Division and Differentiae." In Philosophical issues in Aristotle's biology, edited by Allan Gotthelf and James G. Lennox, 69-89. Pittsburgh: Cambridge U.P., 1987.
Balme, David. "Genos and Form in Aristotle's Biology." The Classical Quarterly 12 (1962): 81-98.
Bayer, Greg. "Classification and Explanation in Aristotle's Theory of Definition." Journal of the History of Philosophy 36 (1998): 487-505.
Charles, David. Aristotle on Meaning and Essence. New York: Oxford UP, 2000.
Charles, David. "Aristotle on Meaning, Natural Kinds and Natural History." In Biologie, Logique et Metaphysique chez Aristote, edited by Daniel Devereux and Pierre Pellegrin, 145-67. Paris : Editions du CNRS, 1990.
Gotthelf, Allan. "The Elephant's Nose: Further Reflections on the Axiomatic Structure of Biological Explanation in Aristotle." In Aristotelishce Biologie, edited by Wolfgang Kullman and Sabine Follinger, 85-96. Stutgart: Franz Steiner Verlag, 1997.
Gotthelf, Allan. "Historiae I: Plantarum et Animalium." In Theophrastean Studies, edited by William W. Fortenbaugh and Robert W. Sharples, 100-35. New Brunswick: Rutgers University Press, 1988.
Henry, Devin. "Aristotle's Pluralistic Realism." The Monist 94 (2001): 198-222.
Lennox, James G. Aristotle on the Parts of Animals I-IV: Clarentdon Aristotle Series. New York: Oxford U.P., 2001a.
Lennox, James G. "Kinds, Forms of Kinds, and the More and the Less in Aristotle's Biology." In Aristotle"s Philosophy of Biology: Studies in the Origins of Life Science, edited by James G. Lennox, 160-81. Cambridge: Cambridge U.P., 2001b.
Lennox, James G. "Between data and Demonstration: The Analytics and the Historia Animalium." In Science and Philosophy in Classical Greece, edited by Allan Bowen, 261-95. New York: Garland Press, 1991.
Lennox, James G. "Notes on David Charles on HA." In Biologie, Logique et Metaphysique chez Aristote, edited by Daniel Devereux and Pierre Pellegrin, 169-183. Paris: Editions du CNRS, 1990.
Lennox, James G. "Divide and Explain: the Posterior Analytics in Practice." In Philosophical Issues in Aristotle's Biology, edited by Allan Gotthelf and James G. Lennox, 90-119. Cambridge: Cambridge U.P., 1987.
Meyer, Jurgen B. Aristoteles Thierkunde: ein Beitrag zur Geshichte der Zoologies, Physiologie und altem Philosophie. Berlin: Druck und Verlag von Georg Reimer, 1855.
Peck, Arthur L. Aristotle: Historia Animalium. Vol. I: Books I-III. Loeb Classic Library, London: Heinemann, 1965.
Pellegrin, Pierre. "Logical and biological difference: the unity of Aristotle's thought." Translation by A. Preus in Philosophical Issues in Aristotle's Biology, edited by Allan Gotthelf and James G. Lennox, 313-338. Cambridge: Cambridge U.P., 1987.
Pellegrin, Pierre. La Classification Des Animaux Chez Aristote. Paris: Les Belles Lettres, 1982. Translated by Anthony Preuss. Aristotle's Classification of Animals. Berkeley: University of California Press, 1986.
Pellegrin, Pierre. "Aristotle: A Zoology Without Species." Translation by Anthony Preus in Aristotle on Nature and Living Things: Philosophical and Historical Studies presented to David M. Balme on his Seventieth Birthday, edited by Allan Gotthelf, 95-115. Pittsburgh: Matheus Publications, 1985.
Byron J. Stoyles: Department of Philosophy, Lady Eaton College, Trent University, Peterborough, ON K9J 7B8, Canada, email@example.com
(1) Although other translations might be argued for, I use 'greatest' to capture [TEXT NOT REPRODUCIBLE IN ASCII] as the superlative of [TEXT NOT REPRODUCIBLE IN ASCII]. In this context, 'biggest', 'widest', or 'most extensive' kinds would also work.
(2) Peck (1965), lxvii.
(3) Charles (2000), 329. See also Charles (1990)
(4) Charles (2000), 317.
(5) Ibid, 342.
(6) Ibid, 329.
(7) Gotthelf (1988), 108. For Balme's contributions, see especially Balme (1987). See also Lennox (1987). Pierre Pellegrin (1982,1985, and 1987) shows, among other things, that Aristotle's use of [TEXT NOT REPRODUCIBLE IN ASCII], and differentia/e in the biological works is consistent with the logical treatises.
(8) These axes of comparison are introduced by the programmatic remark at HA I 1, 487a10-13: 'Animals differ from one another in their modes of subsistence, in habits, and in their parts. Concerning these differences we shall first speak in broad and general terms, and subsequently we shall treat of the same with close reference to each particular genus.' (trans d'A.W. Thompson)
(9) Gotthelf (1988), 108
(10) See HA I 6, 490b11 for the view that the identification of causes follows the work contained in HA. This is to indicate the order in which Aristotle intends a student of biology to read these texts and nothing about the order in which these texts have actually been written.
(11) Lennox (1987) also accepts Balme's thesis that Aristotle is not concerned with generating a classification of animals. I consider Lennox's work in more detail in the final section of the paper.
(12) Cetacea are listed in HA but not PA.
(13) For work on History of Animals, see Gotthelf (1988 and 1997) and Lennox (1987 and 1991). For work on Parts of Animals, see especially the notes accompanying Lennox's (2001a) translation of this work starting with the note on 642b34-5 found on pp. 156-7. See also Pellegrin (1982).
(14) I use 'form' to translate [TEXT NOT REPRODUCIBLE IN ASCII] throughout this paper despite the fact that 'species' is often used. The use of 'form' avoids confusions that arise if we were to use 'species' in so far as we use 'species' to pick out only the lowest level of taxonomy while, for Aristotle, [TEXT NOT REPRODUCIBLE IN ASCII] is used at the various levels of generality. Following Pellegrin (1985), I think Aristotle's use of [TEXT NOT REPRODUCIBLE IN ASCII] in the biological treatises is consistent with the use of [TEXT NOT REPRODUCIBLE IN ASCII] in the logical works (most notably Categories and Topics), in which each [TEXT NOT REPRODUCIBLE IN ASCII] falls under a kind. As Pellegrin (1982 and 1985) argues, what I am calling the 'logical sense' is consistent with the way these terms are used in Aristotle's divisions (see, e.g., Metaphysics Z 8-12 and Parts of Animals I 2-4).
(15) Translations are mine unless noted. 490b7 [TEXT NOT REPRODUCIBLE IN ASCII], ...
(16) Charles (2000), 342, for instance, thinks that this passage refers to the greatest kinds. I read this passage to include all animal kinds, including the greatest kinds.
(17) For an excellent discussion of these differences, see Lennox (2001b), 160-81.
(18) At PA I 4, 644b1-6, Aristotle claims that some groups are not known by a common name though each, 'like a kind', contains the forms within it. It is not clear whether we are to conclude from this that kinds must have names. More likely, Aristotle means to contrast the kinds that are identified in common parlance and those kinds that are not, but are similar. Since Aristotle regularly calls these 'kinds' (rather than 'like kinds'), and since he does so in the passage identifying the greatest kinds, I refer to both named and unnamed kinds without qualification.
(19) Peck's translation does not reflect Aristotle's use of gar at 490b16. I read this use of gar as an indication that what follows explains why we do not find greatest kinds in the remaining animals.
(20) In response to a similar suggestion from Gotthelf, Pellegrin states '[t]hat seems to me difficult simply because Aristotle would then call a class a form at the same time that he says it is divisible, which he does nowhere else' (1987), 326. However, Pellegrin is wrong unless he means Aristotle to suggest that a form, as such, is not divisible. Aristotle does allow that some, [TEXT NOT REPRODUCIBLE IN ASCII] are also [TEXT NOT REPRODUCIBLE IN ASCII]. See Categories 5, 2b23 and, especially, Physics 227b11.
(21) It is notoriously difficult to identify a list of intermediate kinds that Aristotle uses with any degree of consistency. J.B. Meyer (1855) has done the most notable work in trying to identify genera lying between the greatest kinds and the atomic [TEXT NOT REPRODUCIBLE IN ASCII]. Balme (1987) is critical of any such attempt, including Meyer's. Balme takes this to be evidence that Aristotle is not concerned with taxonomy.
(22) Pellegrin (1982) assumes something like the first interpretation I have put forward. He believes that the greatest kinds are those sufficiently large to be divided several times. In light of the passage in PA IV 8, Pellegrin has since retreated from that interpretation in later work (see, e.g., Pellegrin, 1987).
(23) Translation following Lennox (2001a). I have used 'greatest' rather than 'extensive' to translate [TEXT NOT REPRODUCIBLE IN ASCII]. The bracket is inserted. 683b.26 ... [TEXT NOT REPRODUCIBLE IN ASCII].
(24) cf. HA IV 2, 525a30 where Aristotle treats of the soft-shelled animals. There, he notes that there are many kinds of crabs. He lists the 'granny' crab, the common crab, the Heracleotic crab, the fresh-water crab, and that found on the Phoenecian beaches, while noting that the others (presumably, all other crabs) are smaller and lack specific names.
(25) Balme (1962) notes that [TEXT NOT REPRODUCIBLE IN ASCII] are often used interchangeably and that Aristotle often refers to the lowest species as [TEXT NOT REPRODUCIBLE IN ASCII]. Examples include the several kinds of cicada that are referred to as [TEXT NOT REPRODUCIBLE IN ASCII] at HA IV 7, 532b15 and [TEXT NOT REPRODUCIBLE IN ASCII] at HA V 30, 556a14 and other kinds that are atomic forms or indivisible species.
(26) This point is obscured by the frequent use of 'extensive' or 'great' to translate [TEXT NOT REPRODUCIBLE IN ASCII].
(27) Scholars sometimes interpret Aristotle to say that these are greatest kinds. For examples, see Peck (1965), note a on pp. 32-33 and Lennox (2001a), 157. Charles (2000), 318 thinks these are not to be counted among the greatest kinds. Charles typically notes only the seven kinds explicit in the list found in HA I 6. He works to show why other kinds, such as the kind of snakes, are not greatest kinds. However, Charles does not seem to think that talk of greatest kinds is restricted to the passages in which this label is used explicitly. For instance, he writes of 'the greatest genera' referred to in PA I 4 (p. 342) despite the fact Aristotle does not use that label there.
(28) 505b25 ... [TEXT NOT REPRODUCIBLE IN ASCII]. My interpretation of this passage is, indeed, better supported with the manuscript if unchanged in so far as 505b28 would read: 'These are man, and ...' The inclusion of man at the start of the list would surely mean that [TEXT NOT REPRODUCIBLE IN ASCII] refers to blooded animals rather than greatest kinds. Curiously, Pellegrin (1987) goes with the version in which man is the first of the list of 'these', yet Pellegrin maintains that the [TEXT NOT REPRODUCIBLE IN ASCII] refers to the greatest kinds. Since man is explicitly a simple form, it does not serve as a kind in the typological sense Pellegrin emphasizes (never mind a greatest kind).
(29) Reading the passage as listing the greatest kinds would seem to 'deny the four Bloodless gene mentioned [in HA I 6] the title of megista gene' as Peck (1965), lxvi observes.
(30) See note 13.
(31) Lennox (1999), 19.
(32) Specifically, the discussion with which I am concerned is that concerning the parts instrumental to reproduction. [TEXT NOT REPRODUCIBLE IN ASCII] are treated in a manner that applies to all animals in so far as they possess blood or that which is analogous to blood. Thus, no divisions are required.
(33) The complete explanation is teleological and can be compared to Physics II 8.
(34) The perfection of an egg has to do with whether or not it grows outside of the female--perfect eggs do not whereas imperfect eggs do.
(35) The live-bearing animals and those that are internally egg-laying and externally live-bearing are first introduced in the opposite order (I 9, 718b29-32). They are, however, explained in the order indicated here.
(36) There arises the question of whether these divisions violate Aristotle's claim that natural kinds should not be divided at PA I 4, 642b10. Devin Henry (2011) argues that these divisions do not split natural kinds for the reason that Aristotle is recognizing different, but still natural, kinds (i.e., the live-bearing kind and the egg-laying kind) in Generation of Animals and takes these divisions to reflect that Aristotle is a pluralist concerning natural kinds.
(37) For this concern, see PA I 1, 644a24-7 and PA I 4, 644a34.
(38) See Gotthelf (1988), Lennox (1987), and Charles (2000). I return to this detail below.
(39) The opening lines of GA I 1 indicate that Aristotle would have us read Generation of Animals after we have read Parts of Animals. The exceptions are considered in Book V of Generation of Animals, where Aristotle addresses differences that are not determined teleologically.
(40) It is plausible Aristotle assumes that organs sharing a function have the same shape or form unless there is some reason for the difference. Such differences exist, for example, whenever we observe that different animals have parts that serve the same function but differ in form. These are cases in which the parts in question are merely analogous (as scales are to feathers) rather than one in form. Cf. Charles (2000), 313.
(41) The dichotomists Aristotle criticizes in PA I 2-4, for instance, seem to have divided by the number of feet possessed by the various animals.
(42) At History of Animals II 1, 498b16 Aristotle records similarity and differences with regards to hair: all four-footed, live-bearing animals have hair; likewise, man has hair, but in patterns differing from the animals that are hair-coated. In addition, at HA III10, 517b4, Aristotle mentions that all live-bearing animals with feet also have hair.
(43) Charles frequently suggests that Aristotle would have us develop a descending tree of kinds and subkinds in what he calls genus/species or genus/differentiae trees. See for examples, pp. 238, 241, 243, and 269. Charles suggests that '[k]inds, sub-kinds, and differentiae are all selected (in part) on the basis of their role in explanation', 242.
(44) Pellegrin (1980 and 1987) argues for the same view.
(45) Lennox (1990), 170 and (1987).
(46) Though the History of Animals does not include explanations, it is to serve as an inquiry that aids in the explanation of differences among animals (HA I 6, 491a10). Parts of Animals (II 5, 646b2-3; II 1, 646a8), Progression of Animals (IA 1, 704a5-b10), and Generation of Animals (GA I 1, 715a1-16) are intended to explain the parts, movement and generation of animals, respectively.
(47) Balme (1987) came to this view. For Charles and Lennox, especially, this point is what we would expect given Aristotle's general explanation of [TEXT NOT REPRODUCIBLE IN ASCII] in Posterior Analytics II. The differences in their interpretations of the biological treatises are closely related to the differences in their interpretations of the Analytics.
(48) Lennox (1987) and Bayer (1998). Gotthelf (1988), 105 cites Lennox with approval.
(49) Charles (2000), 219
(50) This is an example of what Lennox (1987) calls an A-type explanation. As Lennox notes, this kind of explanation is not ultimately what Aristotle is after, although it can be a useful intermediate stage in our understanding.
(51) I wish to thank to the participants of the workshop hosted by Devin Henry at the University of Western Ontario who commented on an early draft of this paper, especially Jessica Gelber, Devin Henry, Mariska Leunissen, John Thorp, and Charlotte Witt.
|Printer friendly Cite/link Email Feedback|
|Author:||Stoyles, Byron J.|
|Publication:||APEIRON: A Journal for Ancient Philosophy and Science|
|Article Type:||Critical essay|
|Date:||Jan 1, 2013|
|Previous Article:||The psychological background of the first education in Plato's Laws.|
|Next Article:||Aristotle's critique of Platonist mathematical objects: two test cases from Metaphysics M 2.|