A somewhat disorderly nature: unity in Aristotle's Meteorologica I-III.
It is especially worthwhile to consider the Meteorologica in view of the recent and successful reintegration of the biological works into the mainstream of Aristotelian scholarship. The study of Aristotle's methods in the biology has enriched our understanding of classification and scientific organization, of essence and unity, of explanation and cause. I want to do something similar for the Meteorologica by arguing for three related claims:
1. In basic organization the Meteorologica is similar to the biological works, especially in the use of those techniques brought to light by Gotthelf, Lennox and others. (3) This fact is important, because most scholars have supposed that those techniques, most importantly, the 'greatest group' system of multiple differentiae theorized in the first book of the Parts of Animals, are specific to the biology. (4) I shall argue that many of these techniques are present in the Meteorologica, and in fact are more appropriate there.
2. Aristotle adapts the system of multiple differentiae familiar in the biological works to increase the unity of the Meteorologica. Here he treats his differentiae as causal, and though in a sense they are primitive, they are also joined to one another by per se relations. From the interactions among the causal differentiae there arise further techniques of unification--analogies, correspondences and hierarchies--which lend the Meteorologica its remarkable coherence. The general nature and purpose of these interactions have simply not been appreciated.
3. Finally I shall return briefly to the biological works and consider how they differ in their organization and structure from the Meteorologica. I shall argue that the addition there of the final cause leads Aristotle to view the realm of animals as a study in the differences upon a common form, rather than, as with the Meteorologica, a study in unity among diverse phenomena.
The System of the Biology and the Meteorologica
In the first book of the Parts of Animals Aristotle lays out his objections to the system of dichotomous division. (5) According to this system a genus is divided by a pair of contrary differentiae, each of which in turn is divided by contrary differentiae that are per se related to the first differentiae. To adapt a favorite Aristotelian example, footed animals are divided into two-footed and four-footed animals, and four-footed animals are divided in turn into cloven-footed, four-footed animals and non-cloven-footed, four-footed animals. (6) In a perfect dichotomous division, each species will be defined solely by its final differentia, since this differentia implies all the previous differentiae. Aristotle objects that in fact such a division may not be possible, and that even if it is possible, the essence of a thing will have to be summed up in one single differentia, to the great impoverishment of our understanding of the object under consideration (I 2, 642b5-9; I 3, 643b9-18; 644a6-8). (7)
In response, Aristotle proposes the system of multiple differentiae. Within some large group of related objects (e.g., animals) we should accept the sub-groups or greatest genera that everyday language has identified (e.g., birds, fish), and define these by a number of mutually unsubordinated differentiae (e.g., having two legs, having scales; I 3, 643b9-13). Each of these generic differentiae taken singly may be of different extension both from that of the great genus and from each other; but taken together they are coextensive with that genus (I 3, 643b17-26). So, for example, the division between viviparous and oviparous animals results in different groups from the division between two-footed and four-footed animals. But all and only birds are both two-footed and oviparous, and all and only humans are two-footed and viviparous (GA II 1, 732b15-31). Once the generic differentiae have been determined, differences of degree may be used to define the several species within each genus, as for example, some species of birds have long wings, others short wings (PA I 4, 644a19-21). According to Aristotle's programmatic passages, the differences between great genera are not characterized by differences of degree, and such similarities as exist are those of analogy (e.g., the fin is analogous to the wing; HA I 1, 486a14-487a1; PA I 4, 644a21-23; 644b7-15).
Now the Meteorologica considers a wide variety of phenomena, including combustions, like shooting stars, comets and the milky way; the phenomena of condensation, i.e., the rain cycle; emanations from sources (winds and rivers) and their related effects (salinity of the sea and earthquakes); violent ejections, like lightning and hurricanes; reflections, like rainbows and sundogs; and the subterranean deposits, metals and minerals. Aristotle organizes these into eight more or less distinct groups, and the three books of the treatise are divided according to these groups: roughly, the phenomena of combustion and of condensation are taken up in Book I, rivers and winds (and their related effects) together with ejections make up Book II, and reflections are treated in Book III. The seventh and eighth groups, the metals and minerals, are barely discussed, but are important for the overall structure of the treatise. These, I claim, function like the greatest groups of the biological works, and I shall consider this proposition sufficiently proved, if we find that
a) there are several groups each with more than one member;
b) these groups are defined by several mutually unsubordinated generic differentiae that cannot form a dichotomous division;
c) the differences among the members of each group are differences of degree, where the differences between groups are not.
First, it is obvious that Aristotle sorts the phenomena into several groups, only the metals and minerals being so briefly treated as to be barely so recognizable. (8) It is also obvious that there are several members within each group, though some groups are admittedly more clearly demarcated than others. (9) While the upper combustions form a distinct group, it is difficult to know whether to treat the winds and rivers separately or together, and where to place the phenomena of the sea. But for the most part Aristotle clearly signals the introduction of each distinct group.
Secondly, the distinctions between these groups are made on the basis of mutually unsubordinated differentiae. In the biological works these differences are based on the outward appearance of the animals, their obvious parts. (10) The meteorological phenomena, by contrast, are divided by three definitory factors: material and efficient cause, and place. (11) These causes constitute the essential features of each phenomenon. At the beginning of many sections throughout the treatise Aristotle asks 'what is x and what are its causes?' (12) That is, the Meteorologica is a [TEXT NOT REPRODUCIBLE IN ASCII.] inquiry, and the answer to the question is the definition of the phenomenon, which is given explicitly and invariably in causal terms. (13) The causes, then, are the defining factors, and it is among these that we must seek the differentiae. And this very fact, namely, that the divisionary factors are different causes, makes it impossible for them to be mutually subordinated in the manner of dichotomous division: material causes, efficient causes and place cannot serve as differentiae to one another.
Let us take up each of these three causal differentiae in turn to show that they all contribute to the nature of the phenomena, and that none is redundant or reducible to another. The material distinction between dry and wet exhalation is an obvious dichotomous distinction and would seem to be a good candidate for the first differentia in a dichotomous series. But if this were the case, we would expect the treatise to be divided into two halves devoted to each exhalation in turn. Instead we find the exhalations taken up as a pair in each of three spaces: above the earth (combustion/condensation), on the earth's surface (winds/rivers) and under the earth (metals/minerals). (14) Moreover, Aristotle makes very little effort to distinguish among species of dry or wet exhalations, even when it would be useful for him to do so. We could imagine him claiming, for example, that the milky way has a hotter and smokier exhalation than the winds. (15) But instead he distinguishes between them in virtue of their place and efficient cause.
As for the efficient cause, Aristotle identifies one generic mover, the circular movement of the heavens, which because of its fundamental unity seems unable to serve as a differentia at all. This single generic cause, however, operates in different ways in each kind, and usually in multiple ways even within a single species. It is first and foremost responsible for the production of the exhalations (I 4, 341b6-10). But it operates again to produce further changes. (16) The friction of the movement of the heavens against the upper region, for example, causes the combustion of the dry, smoky exhalation (I 4, 341b18-22); and the withdrawal of the warming power of the sun causes condensation of vapor into water (I 9, 346b20-31). For this reason it might be tempting to view the efficient causes as the first differentiae and the great groups themselves as the species: the phenomena of combustion, the phenomena of condensation, etc. But this cannot be so. First, the flow of the winds and rivers and their associated effects, and likewise the ejections that produce violent phenomena like thunderbolts, can hardly be considered as species of the movement of the heavens. Thunderbolts are moved by compression in a cloud, and not by the heavens (at least not immediately, II 9, 369a12-29). Second, even the specific sources of motion do not completely divide the great groups: condensation, for example, is the common efficient cause of the rain cycle, of the rivers (I 13, 349b27-35) and of metals (III 6, 378a25-b3). Finally, the efficient cause does not adequately capture what the members of each great group share in common. The phenomena of combustion, while they certainly are all heated by friction from the heavens, are still irreducibly dependent on the dry exhalation for their material. Similarly too, the rain cycle requires the wet exhalation, which is an essential part of its nature.
Place, likewise, is an irreducible and essential feature of the meteorological phenomena, for all that its fundamental principles arise out of the material and the efficient causes. Admittedly the distinction among the upper places can be explained by the differences in material. Just under the circular movement of the heavens (I 4, 341b13-14) and extending down to the mountain peaks (I 3, 340b32-6) we find the fiery region, to which the hot, dry exhalation naturally ascends. Below this, extending to the surface of the earth are the less buoyant phenomena common to air and water, the cycle of rain, snow, hail, dew and frost (I 9, 346b16-19). To this extent material explains place. But in the middle zone (though occupying some of the upper space), Aristotle sets another pair of wet and dry exhalations, the rivers and the winds, and their related effects. Again, in a brief coda at the end of III 6 Aristotle introduces the metals and minerals, effects again of the wet and the dry exhalations that remain entirely under the earth. Aristotle makes no effort to account for the difference between the middle and subterranean phenomena on the basis of their material or efficient cause (which they share), and they can be distinguished only in virtue of the peculiarities of their places. Place, then, is an independent defining factor. But, again, neither it nor its differences can serve as the differentiae sufficient to distinguish between all eight great groups, since the violent phenomena and the reflections occur in the same place as the rain cycle.
It is clear then that all three causal factors are necessary, and none is independently sufficient to distinguish the basic groups of phenomena Aristotle considers in the treatise. At the same time none is subordinate to another in the manner of a differentia to a genus. It follows, then, that we have a set of mutually unsubordinated and non-redundant differentiae. It remains now only to show that among the members of each group there are distinctions of degree, of more and less: (17) differences of shape and quantity of material, short and long duration of phenomena, warm and cold, fine and coarse texture. (18) By and large they are not responsible for dividing between the great groups, but as in the biological works, they serve to distinguish the species of phenomena within each great group. So we find both the combustions and the condensations subdivided again, according to quantities of exhalation and longer term (upper) and shorter term (lower) phenomena. (19) Similarly, the difference between thunder and hurricanes (both ejections) is determined by the quantity of material involved (III 1, 370b5-10). The generic differentiae, by contrast, are largely constructed to avoid differences of degree. Within this treatise at least, the distinction between wet and dry exhalation is absolute, and wet exhalation cannot be turned by degrees into dry. Likewise the various manifestations of the efficient cause (combustion, condensation, ejection, etc.) are not related by differences of degree. Only in the case of place might we say that one place is higher than another and even here each place is associated with a peculiar orientation. For the most part, differences of degree give rise to distinctions among the members of each major group of meteorological phenomena and do not make distinctions between the groups.
It is clear then, that the great group method, usually considered specific to the biological works, is a basic organizing principle in the Meteorologica.
Causal Interactions and their Hierarchical Effects
So far, I have outlined the eight great groups into which the phenomena fall, and the generic differentiae (material cause, efficient cause and place), which, only when taken together, are adequate to distinguish them. This fact alone is interesting enough and sufficient to make us rethink the primacy of the biology in the new method outlined in the first book of the Parts of Animals. Now having argued that the causal differentiae are mutually irreducible, I want to show how they are dependent upon one another.
Though in Parts of Animals I Aristotle explicitly introduces the system of multiple differentiae to overcome the weakness of dichotomous division, he does not discuss what is lost on account of his innovation. The great strength of the older, strict form of dichotomous division is that it prevents the kinds of slipshod divisions we find in Plato's Sophist, divisions in which the successive differentiae bear no essential relationship to their predecessors; that is, dichotomous division prevents the definiendum from falling into a heap of unrelated features having no conceptual unity. So, in the Sophist (220b1) swimming animals are divided into winged and water animals, in spite of the fact that swimming bears no essential relationship to having wings. (20) In dichotomous division, however, the logic of the per se relationship solves the problem by making clear why, for example, a cloven-footed animal must have feet.
Now, the new system of multiple differentiae simply reintroduces the problem of slipshod division: there is no reason why a two-footed animal must be oviparous; indeed some are not. (21) There is no direct per se relationship between having so many feet and having some specific mode of reproduction. As a result Aristotle loses the obvious (albeit sterile) unity of the dichotomous system.
In his new system of multiple differentiae, Aristotle maintains the requirement that within each generic characteristic the specific differentiae must be per se related both to the generic characteristic and to each other. So for example, the differentiae, 'nocturnal' and 'diurnal' (HA I 1, 488a25-6) are necessary differentiae of 'times of activity', and there is a necessary relationship between 'at day' and 'at night'. Only rarely does Aristotle deviate from this practice. So, though the differentiae of diet, 'meat-eating', 'grass-eating', 'honey-eating', 'fly-eating' (488a14-16), are necessarily forms of diet, there is no per se relation between them, that is, they are not defined in terms of one another, as day and night are. Aristotle generally avoids this kind of differentiae.
Now the most usual sort of differentiae are not merely per se related--they often form series, in which one differentia is prior to another. Obvious and common cases are those of form and privation: some animals have a dwelling, others do not (488a20-1); other cases involve inclusive series (mute, having voice, having articulate voice, 488a31-2). Aristotle is careful to structure his differentiae in this way in the Meteorologica as well. Each of the definitory factors is divided in such a way that its species form a series (e.g., place is divided into higher, middle and lower) insuring the internal unity of each set of differentiae. So far, though, none of this compensates for the loss of unity produced by the free-floating system of multiple differentiae.
These problems have been frequently noted. What has been universally ignored, however, is the untheorized solution Aristotle uses in the Meteorologica. He shows how the generic differentiae are related to one another causally instead of in virtue of difference. The role of cause in unifying the generic differences marks a clear advance in the multiple differentiae system and provides an appropriately expanded role to demonstration in the unity of the science.
Furthermore, not only are the differentiae joined together by causal relations; the series among the differentiae also contribute to a number of hierarchies among the great groups. So, as we shall see in greater detail, serial differences in place cause hierarchical differences in the orientation of the several great groups. These hierarchies among the great groups are invariably the result of causes working severally in each group, and never is a hierarchy itself used to explain features of a specific great group.
I want to take these issues up in order, first, showing in more detail the nature of the causal interaction between the differentiae at a generic level, then examining the causal interactions among the species of material, place and efficient cause. I shall then show how these species form series, and argue that they cause hierarchies among the phenomena.
First, one generic characteristic can serve as the partial cause of another generic characteristic. The phenomena of the treatise occur between the source of the efficient cause (the circular movement of the heavens) and the source of the material cause (the earth), which produces the exhalations. The efficient and material principles are, therefore, the causes of the extent of place in the treatise. Place is defined by the combination of the efficient and material causes.
Again, the exhalations, dry and wet, are the material principles, but they are not the first principles of the science. The first and only independent cause is the efficient cause, the movement of the heavens (I 2, 339a23-4). The exhalations are generated by the heating effect of the movement of the heavens upon the earth and upon the moisture in and on the earth (I 4, 341b6-10). In so characterizing the exhalations, Aristotle has made an important choice. He might have chosen as material cause something independent of the efficient cause: earth, water, air and fire are simply preexisting and independent of the movement of the heavens. But instead, he chose a special material, which has to be worked up by a specific warming cause, and which is distinct from the preexisting materials. The effect of this choice is that the exhalations are already special to the science, and their existence is not a conclusion of any of the higher sciences. Obviously, too, the material cause is not causally independent, but is already a caused cause even within its own science. There are, then, not two independent first principles, matter and heavenly mover, but only one, the heavenly mover. In this way, also, the plural material causes are defined in terms of the unitary efficient cause without being species of it. At a general level, then, the efficient cause is the cause of the material cause, and the two together are the causes of place.
Coincidentally, this allows us to understand, if not entirely solve, a long recognized problem in the Meteorologica, namely what is the relationship between the four simple bodies we meet in the Physics, de Caelo and the Generation and Corruption, and these exhalations and the spaces they occupy. (22) Aristotle explicitly states that the general nature of the exhalations arises partly out of the material principles of the higher sciences (I 2). But however much the principles of our treatise may depend upon the higher sciences, the four-element theory is clearly ill at ease among meteorological phenomena. This tension suggested to Strohm that the elements of the higher sciences are forced into new natures in the Meteorologica. (23) But if we keep in mind that Aristotle is deliberately creating his special material cause out of the preexisting elements, and that (in spite of what Aristotle sometimes leads us to think in the opening chapters) (24) the exhalations do not entirely fill the sublunary world, we can see that Aristotle wanted to mark off this material as different from, though causally related to, the four basic pure elements.
Turning from the general causes to the specific, we see again both how the species of each cause are caused by the other causes and how they are serialized by the other causes. Aristotle's two material causes (I 4, 341b6-10), the wet and the dry, are of course, per se related insofar as they are contraries. But by calling on the other causes, Aristotle makes the dry prior to the wet in a variety of ways. First, the dry exhalation is superior, since its natural place is higher and closer to the heavens.
Again, the dry exhalation is superior in virtue of the efficient cause, both because the dry exhalation can carry the wet exhalation upward (I 9, 346b24-31; cf. II 9, 369a13-24), and because the dry exhalation can burn and in turn heat the earth and generate both the dry and the wet exhalations (I 9, 346b23-6). Finally, Aristotle says that the wet exhalation comes from the moisture on and in the earth (I 4, 341b9-10). Why should he mention the earth here and not merely the surface of bodies of water? The earth is essential as a source of wet exhalation, because otherwise it would be impossible to account for the origin of the rivers and the metals in the earth. The effect of this is to join causally the two material causes in their place of origin and to establish the logical priority of the dry over the wet exhalation: the wet exhalation contains in its definition the fundamental source of the dry exhalation, the earth. In these ways the serial order of the material exhalations and of the phenomena that arise from them is caused by efficient and local considerations.
The various species of place likewise form an obvious series from top to bottom beginning in the noblest location next to the heavenly movement, and ending below the earth. In view of the doctrine of the de Caelo and the Generation and Corruption we should expect a simple correspondence of exhalation with its natural place, the hot, dry exhalation occupying the top half of the sublunary cosmos, and the hot, wet exhalation occupying the lower half. As such, its order would be determined merely by the matter. Instead, each of three places is correlated with the ability of the exhalations to attain their natural places: some of the exhalations are forced to remain in the earth becoming minerals and metals, others escape part way becoming winds and rivers, others again fully escape to their natural place to become the combustions and condensations. The fact that each of the three places contains both exhalations shows that differences in place are not determined solely by the material cause.
Moreover, although Aristotle nowhere theorizes the fact, the exhalations clearly exhibit different behavior according to their places, most strikingly in the orientation of the phenomena. These different orientations are caused only partly by the material cause; in part too they are determined by the efficient cause, and in part they are primitive and without prior cause. In the upper place both the materials tend to ascend vertically. The phenomena of combustion and condensation take up their respective higher and lower natural places dictated by their material natures, and are clearly separated from one another at the peaks of the mountains, at which point the rotary motion of the heavens translated to the upper dry exhalation peters out (I 3, 340b32-6). The dry exhalation rises and does not fall again unless forced to. (25) But the wet cools again and descends, forming a cycle similar to the cyclical river of Ocean (I 9, 347a2-8). Aristotle provides a material explanation for this cycle by calling these phenomena the common affections of air and water, indicating that they can only be understood in that combination. Their recurrent movement is also necessitated by their efficient cause, the recurrent movement of the sun in the course of the seasons (I 9, 346b20-3). (26)
As we descend in the series of place, we come next to the second pair, the winds and the rivers, which are horizontally and radially, rather than vertically, oriented. Rather than rising and falling, they seem to flow. Aristotle feels no need to point out that the lateral flow of rivers is caused by their relatively heavy material nature. The lighter, dry exhalation that forms the winds is above the rivers again for obvious material reasons. What is not obvious is why the winds flow in a way similar to rivers. But all the same, Aristotle works very hard to bring out the spatial correspondences between them: (27) both winds and rivers arise from sources (II 4, 360a27-33); there are different winds just as there are different rivers (I 13, 349a20-26); neither is merely matter in motion (I 13, 349a16-20; II 4, 360a29-31); they both gather strength as they go (I 13, 349b30-35; II 4, 361b1-5); when blowing, winds are situated above rivers; but they are opposite in their sources, rivers arising from the high mountains (I 13, 350a2-4), winds from the low-lying areas (I 3, 340b36-341a1). The differentiae of the winds and the rivers are again in both cases horizontal and radial, corresponding to the points of the compass and the solstices. In their vertical places and their differentiae, the rivers and the winds are not as sharply distinguished from one another as the upper pair: properly both winds and rivers belong above the surface of the earth, but they both have a subterranean component as well (I 13, 350b30-351a18; II 8). (28) The orientation of the winds has a very weak aetiology. Aristotle does provide an external efficient cause, though almost as an afterthought: winds move horizontally because of the movement of the heavens (II 4, 361a22-5). But this cause is obviously at odds with the radial directionality of the winds and with the separation of the atmosphere from the kapnosphere (my term for the upper region), as Alexander points out. (29) Aristotle clearly felt that he needed to supply some efficient cause that was specific to the winds and that did not make their flowing dependent on the movement of the rivers.
The final pair, the metals and minerals are indiscriminately described as being formed when the exhalations are shut up under the earth (III 6, 378a15-21). Neither is said to be above the other, or in any way related to the other. In these ways, then, and to this extent the distinctive places and orientations of the phenomena are caused by the other causes.
Together the phenomena constitute a series based on a couple of different principles. First, Aristotle is constructing the sublunary world as a cosmos similar to, and prefiguring, his great chain of terrestrial life. Just as we find posture among living things--the highest animals, man, being upright, and lower animals increasingly prostrate, (30) until with the plants, we find organisms that are upside down, their mouths in the ground--so here we find phenomena of different 'postures': the dry combustible exhalation goes straight up and does not come back; the rain cycle explicitly forms a recurrent river; the winds and the rivers are radially oriented on the surface of the earth, and the metals and the minerals move back down into the earth. Aristotle does not explicitly mention this form of unity and hierarchy, but it answers some pressing questions: why does the dry exhalation in the kapnosphere not cycle back down? why does Aristotle work the elaborate analogies between winds and rivers?
A second series is that of the degree of separation between the material causes, where clearer separation is better than promiscuity. The upper dry and wet exhalations have each their own clearly defined areas and their own distinct movements; the winds and rivers are more compressed in space (though still distinct) and are similar in movement; the metals and minerals are indiscriminate in space.
What is remarkable is that these hierarchies are produced (partly) by causes that govern the individual members of the series, but they do not in their turn supply cause, and never does Aristotle account for the nature of a particular phenomenon in virtue of its place in the hierarchy. The case of the wind shows this most clearly. Though it seems obvious to us that Aristotle's winds move horizontally because they have a certain order in his hierarchy, he never adduces this as a cause, citing instead the rotary motion of the heavens. Aristotle wanted the hierarchy to be an effect of causes and not to be a cause itself.
Having now considered the causal interactions associated with matter and place, let us now turn to the primary efficient causes themselves and consider how they form a series, then look at the ways Aristotle uses these to strengthen and reinforce divisions in matter and place, and finally how he uses them to build hierarchies.
Locomotion together with heating and cooling are the fundamental agents of change in the treatise. Through the repeated application and modification of these causes in the various materials and places, the characteristic 'activities' of the eight great groups are explained. Locomotion is the first cause, and is logically prior to the other forms of change. (31) It has its origin in the heavens, and in consequence is more independent than the other causes. Of locomotion there are several kinds arranged in an order of priority that is generated to some extent outside of the causal network of the science. The locomotion of the heavens in general is the first within its kind, though it is a relatively weak cause: it ignites and rotates the kapnosphere (I 4, 341b18-22), prevents clouds from forming above the atmosphere (I 3, 340b32-6) and causes the lateral movement of the winds (II 4, 361a22-5). Next is the locomotion of the sun specifically, the more powerful cause, which generates heat by friction (I 4, 341b6-10). Finally Aristotle adds to the sun's simple diurnal revolution its annual northern advance and southern retreat, which explain seasonal changes and geographical differences (I 9, 346b20-3).
So far, these efficient causes, locomotions all, do not involve any other principles specific to the science. But hereafter efficient causes become increasingly involved with the material cause. The locomotions produce the alterations, which are next in the series and are responsible for every other change, including further locomotions. The friction of the sun warms the earth by some mysterious rays ([TEXT NOT REPRODUCIBLE IN ASCII.]) and produces the exhalations. The heat causes the dry exhalation to rise (I 4, 341b10-12), where it is ignited (i.e., extremely heated); the cooling of the risen wet exhalation causes the condensation and descent of moisture (I 9, 346b31; cooling, therefore, is posterior to warming); wind is produced by simple dry exhalation, moved laterally by the heavens (II 4, 361a22-5); rivers flow as a result of vapor condensing under mountains (I 13, 349b30-5); ejections occur when a cooling exhalation meets a hot one, and moves it violently either downward, or laterally, or in a circle (II 9, 369a12-24); the reflective phenomena remain the odd men out, though even they double back to the sun as that to which the visual ray reflects. These secondary efficient causes are caused by the interaction of the primary locomotions with the other causes.
These causes give rise to hierarchies among the phenomena, the first of which moves from simplicity to complexity. The phenomena of combustion arise from a single efficient cause (the locomotion of the heaven) acting on a single material cause, as is appropriate for the upper phenomena. The phenomena of the rain cycle require two pairs of closely related causes. The double nature of the sun's advance and retreat acts upon the double nature of the material substrate, the common stuff that is air and water (I 9, 347a8-11). Finally, violent ejections arise due to contrary natural motions (II 9, 369a12-29). Since the dry exhalation must rise though the atmosphere to get to the kapnosphere, the atmosphere is the place where necessarily the dry and the wet exhalations coexist. In this case, the contrary motions do not have a single source, as with the sun's double role in the rain cycle; rather the contrary motions result from the nature of the contrary materials. Ejections, formed of complexes of material, have themselves complex efficient causes.
Secondly, Aristotle works the efficient causes into a hierarchy of directness. Since 'anything has more power when it is nearer' (II 4, 361a35-6), we find the agency of the heavens directly operative in the upper places, but only indirectly operative or even inoperative on and below the surface of the earth. It is fitting that the highest and noblest of the phenomena, the combustions, should be caused by locomotion, which is the primary form of change, and the more so the closer the phenomena are to the heavens. Thus not only are the milky way and the comets ignited by the heavenly motion, but they are carried along by that motion as well. Moreover, the combustion of the dry exhalation in general warms the earth in the same way that the sun itself does, producing further exhalations (I 9, 346b25). Thus the highest phenomena of our science are themselves efficient causes in the same way as the external heavenly cause, thus combining effect and imitation in one, and establishing yet again the priority of the dry exhalation over the wet. The phenomena of condensation are never their own efficient cause in this way, but they are their own material cause, since the moisture that returns to the earth becomes in turn the wet exhalation. And though their efficient cause comes from the heavens, it is not the primary locomotion, but by the derivative heat, and it comes in competing contraries rather than as a simple translation.
But below this level direct heavenly agency stops. The winds are simply exhalations moved by indirect horizontal translation. The rivers, earthquakes and salty seas--apart from their material origin--owe even less to the heaven's agency; the metals and minerals owe nothing at all. As with the material causes and places, the series of efficient causes give rise to hierarchies among the phenomena. And like the other hierarchies we have seen, these are generated out of many individual causes, but perform no causal work of their own.
We can see then that the generic differentiae of the meteorological phenomena are dependent upon one another. Both at a generic and a specific level the essential causes of the phenomena interact, that is, they are causes of each other. We see here in practice--though never theorized--Aristotle's solution to the problem of disunity in the system of multiple differentiae, a solution that makes demonstration the central unifying force. But what we have here is more than mere causal interaction. These causal interactions consistently fall into a peculiar pattern in which the specific causal differentiae, themselves in serial order give rise to hierarchical arrangements among the meteorological phenomena.
But before drawing further conclusions from these observations, it is worthwhile to note another remarkable feature of the efficient causes: the way they take up 'cameo roles' in alien matter-places. That is, an efficient cause, that is usually associated with one kind of matter or one kind of place, becomes associated with another. Thus Aristotle describes a kind of shooting star in the lower kapnosphere that is caused, not by the usual combustion, but by ejection just as lightning is (I 4, 341b35-342a13). Again, he provides alternative causes for the non-existence of frost on the mountains, one from within the appropriate causal framework for the rain cycle, namely, that the hot exhalations cannot rise so far without first dumping their moisture (I 10, 347a29-34); and one from outside--the transfer of the heavenly locomotion causes the winds to blow more there, with the result that the frost cannot settle (347a34-35). Indeed, the direct transfer of heaven's locomotion, appropriate as a moving cause within the kapnosphere, is found not only dispelling clouds and frost from mountain peaks, but absurdly as we have already seen, moving winds on their horizontal course across the earth's surface. Again, volcanoes spew fiery ejections from below the earth as if they were casting out lightning bolts (II 7, 367a9-11); and wet exhalations condense within the earth to form metal as if they were rain (III 6, 378a28-31).
These 'cameos' serve to reinforce the rule that the causes must always come from within the genus of the science, and that we may not seek explanations from outside of the general subject matter. But more importantly, they prevent the cosmic whole from breaking down into independent great groups, and give Aristotle reason to claim that meteorology as a whole is a single science and not just the sum of kapnospherics, atmospherics, and so on. As such, 'cameos' tend to the same end as the system of multiple differentiae in general. In dichotomous division the sub-genera become more and more distantly removed from one another with each successive division, and since the sub-differentiae of each sub-genus are unique, the sub-groups have less and less to do with one another. There is little to hold the species together except the abstract genus. By contrast, the system of multiple differentiae permits dualizing between great kinds, whereby major alignments of generic causes are crossed. (32) So just as cetacea share most features with viviparous quadrupeds, but, like fish, live in the water and do not have legs, so the kapnospheric ejected shooting star is similar in place and matter to the upper combustions, but shares its efficient cause (ejection) with the lower thunderbolts. These dualizers contribute to the unity of the science by insuring that there is some overlapping of causes among the great kinds. The result is that, given the group of principles we have, we shall have to deal with all and only the phenomena of the subject matter; and if we try to eliminate even one of the principles, or one of the subdifferentiae, we shall not be able to explain even a whole part or section of the science. So, for example, if we eliminate reflection and ejection from the science, we will not be able to give a full account of kapnospherics, since we need those principles to account for the aurora and ejected shooting-stars. If there were no 'cameos', the science would risk breaking down into its constituent parts and not forming a unified whole. Aristotle can have it both ways: by the limited use of dualizers he can preserve the basic divisions of the great groups, while preventing those great groups from breaking down into semi-independent sciences.
But these 'cameos' have one more function as well. They also allow Aristotle to fill out a matter-place-mover grid to the extent that the phenomena permit. The fifth and sixth great groups, the ejections and the reflections, are both affections of the atmosphere. Accordingly Aristotle sought and found in the kapnosphere corresponding ejections (the lower shooting stars just mentioned) as well as reflections. His discussion of the aurora borealis (I 5) uses the same causal terms of visual rays and reflections (minus the geometry) that we see in his full-scale treatment of rainbows, etc. It is clear, I think, that Aristotle was interested in using efficient causality to preserve the symmetry of his pair of upper places.
The Pursuit of Scientific Unity in the Biological Works
We have seen abundant evidence that the definitory factors of the system of multiple differentiae in the Meteorologica are causal and that they serve as causes for each other both at a generic and specific level. This interdependence among the causes was a stroke of genius on Aristotle's part, since it so elegantly compensates for the weakness introduced when the system of multiple differentiae was brought in to replace the impoverished dichotomous division. It greatly strengthens the less unified classification system and sets aetiology at the core of the scientific enterprise.
But we have also observed that the species of these causes are serially ordered, and that these causal series in turn give rise to a variety of hierarchies among the phenomena, though the hierarchies do not supply cause in their turn. In view of this fact, it is reasonable to surmise that the unification of the system of multiple differentiae was not Aristotle's ultimate goal in introducing this network of causal interaction. Though he certainly accomplished that, he also adapted this network to show how a cosmos, though a little less orderly than the heavens (I 1, 338b20), could arise out of interactions among the multifarious sublunary principles. The significance of this purpose can be appreciated by a brief comparison with his procedure in the biological works.
In both sciences we find the system of great genera and multiple differentiae. In the biology too, just as in the Meteorologica, causes interact with and determine one another. In fact, the material and efficient causes are codetermined in the biology even more intensely than in the Meteorologica, since the primary efficient cause (sperma) is not independent of the material cause (katamenia), as the motions of the heavens are independent of the exhalations. The material is also closely related to the final cause. Aristotle lays out a series of material parts--powers, homoiomerous, anhomoiomerous and gross parts--described both as a series of material inclusion leading to greater complexity (GA I 1, 715a9-11), and as a series involving the final cause--each of the simple parts is for the sake of the more complex (PA II 1, 646b4-10). As such, a part that is itself a material cause for one part can be the final cause of another part. Again, some parts subserve generation (GA I 1, 715a11-12), that is, some material causes have as their final cause an efficient cause. And throughout the Parts of Animals and Generation of Animals, the differences in material causes are made to subserve various final causes.
In a number of ways, however, and for a variety of reasons the interdependence of causes, so evident in the Meteorologica, is obscured in the biological works. First, animals are much more complex than meteorological phenomena. The causes that determine their nature are much more numerous, and consequently a neat set of correspondences such as we find in the Meteorologica is virtually impossible to identify.
Again, Aristotle tends to treat each cause separately. The de Anima treats the form in the most abstract sense, the Parts of Animals treats the final cause in relation to the material cause, and the Generation of Animals treats the efficient cause. These separate treatments make each cause a separate object of study. The exception is the interaction we see in the Parts of Animals between the final and material causes, and this has the effect of conferring on this interaction and the associated issues of teleology and hypothetical necessity a celebrity they do not properly deserve.
Aristotle also distinguishes facts, which he treats in the History of Animals (e.g., that some animal or part has this or that feature), from the causes that explain them. Many of the facts that he explains concern place and orientation (e.g., why ears are on the head, PA II 10, 656b28-9). The result is that place is largely downgraded from an explanatory or definitory feature to a mere fact. Moreover, because Aristotle is explaining why animals have certain features rather than providing comprehensive definitions of them, as he does of the phenomena in the Meteorologica, we do not see how the essence of the animal is conceptually built up out of the interaction and specification of its contributing causes. (33)
Again small differences among animals are often explained by reference to vastly different kinds of causes. So for example the causes for differences in eyes among animals are extremely various: elemental constitution, organic compensation, mobility and environment. (34) For these reasons it is hard to see an orderly pattern of causal explanation.
More importantly, though, Aristotle is interested in explaining the differences among the animals. There is a strong tendency to treat the animals as similar to one another, as sharing a common form even across the boundaries of the great genera. Unity and hierarchy in the Parts of Animals is assumed rather than proved. The meteorological phenomena, by contrast, are explained in such a way that their similarities and order arise out of an obvious diversity.
The different style of unity we find in the Parts of Animals is brought about in the first instance by the addition of the final cause and the peculiar way this cause is deployed. (35) Because Aristotle's tendency is first to insure per se relations among the different species of each cause, the several final causes will most appropriately be unified in virtue of the final cause itself: one final cause will be for the sake of another final cause (just as in the Meteorologica, one efficient cause is the efficient cause of another), and the science will achieve its proper unity when there is one ultimate final cause, to which all the parts and activities make reference. Now, Aristotle could have unified the final cause in ecological terms, where one animal is for the sake of another and ultimately all for the sake of man. Instead he terminated the final cause with each whole adult animal. The various animals and their kinds are related to one another then, not by the final case, but instead by similarity and difference.
Now, the ultimate final causes of animals, that for the sake of which all the parts exist, are the functions of the soul. These functions tend to be common (nutrition, reproduction, sensation, locomotion, etc.), that is, they are widely, though not universally, distributed among animals, and they share common principles. As a result, animals, both in their functions and in the material parts that discharge those functions, display a remarkable similarity, a much greater similarity than we find among the meteorological phenomena. Aristotle's aetiological style in the biology then, is to treat the animals in many major respects as a single kind, and to explain the varieties as deviations from a norm. This norm is not an abstract genus, but the first in a series (man), and this series gives rise to hierarchies, which undercut to a considerable degree the autonomy of each great kind. Two examples will suffice to illustrate this ubiquitous tendency. The lack of the upper eyelid in some birds and in the oviparous quadrupeds is explained by their having hard skin (PA II 13, 657a25-b15). That is, they share with viviparous quadrupeds, which have the upper eyelid, the need to protect their soft eyes, and it would be good for them to have upper eyelids like the viviparous quadrupeds, but their hard skin prevents this. Viviparous quadrupeds, then, are the norm; oviparous quadrupeds are the deviation. More broadly, differences in modes of reproduction (viviparous, oviparous, larviparous) are explained by differences in earthiness, where increase in earthiness accounts for the deviant and inferior forms of reproduction (GA II 1, 732b32-3). In these and many other cases an explanation applying to one kind of animal makes reference to another, better kind. That is, in the biology hierarchies are explanatory and supply cause. This tendency is, of course, not universal. The great kinds have many features that can only be explained by reference to their own peculiar nature. (36) Nevertheless, among animals, where the functions of soul are largely the same, Aristotle has a greater tendency to assume commonality and then explain difference. The Meteorologica by contrast, starts with a bewildering variety of dissimilar phenomena and explains them in such a way that their unity and order become manifest.
But the supervenient order of the meteorological phenomena need not just be explained by the absence of the kind of final cause that soul supplies to animals. It has its own internal cause as well. These phenomena share a kind of Heraclitean, homoeostatic stability: out of the flux of the sublunary world fleeting entities arise, which we call rain, earthquakes, comets, etc. (37) A couple of texts suggest that Aristotle was trying here to give expression to the Heraclitean notion of the hidden harmony, the hidden order that is better than the manifest. In the Politics Aristotle says that harmonia takes the place of soul in inanimate things (I 5, 1254a32-3). For our purposes this is tantamount to saying that harmonia takes the place of the final cause where there is no final cause. Now at de Anima I 4 (408a5-23) Aristotle describes a harmonia as something that is dependent on the composition of the parts of the underlying body. If this sense of harmonia describes the nature of the meteorological phenomena both individually and as a collective cosmos, we can see why the orderly hierarchies in the Meteorologica are made to arise out of the several causes but do not guide this part of the sublunary world toward some good.
Finally, we should note that the system of multiple differentiae and causal interdependence seems much more neatly applied in the Meteorologica than in the biological works. This suggests that, in spite of the biological context of the methodological remarks in Parts of Animals I, the Meteorologica is the more natural application of the method. The fact that Parts of Animals I 2-4 was written for a biological context is no proof that the method was developed first for the biology, and given its more perfect application in the Meteorologica, it may have had its genesis in this prior study of the sublunary world.
Alexander Aphrodisiensis. 1899. In Aristotelis Meteorologicorum Libros Commentaria. Ed. M. Hayduck. Berlin: G. Reimer.
Bonitz, H. 1961. Index Aristotelicus. Berlin: de Gruyter.
Charles, D. 2000. Aristotle on Meaning and Essence. Oxford: Oxford University Press.
Cherniss, H. 1935. Aristotle's Criticism of Presocratic Philosophy. Baltimore: The Johns Hopkins Press.
Coles, A. 1997. 'Animal and Childhood Cognition in Aristotle's Biology and the Scala Naturae'. In W. Kullmann and S. Follinger, eds., Aristotelische Biologie: Intentionen, Methoden, Ergebnisse, 287-323. Stuttgart: Franz Steiner.
Detel, W. 1997. 'Why all animals have a stomach. Demonstration and Axiomatization in Aristotle's Parts of Animals', In Kullmann and S. Follinger, eds., Aristotelische Biologie: Intentionen, Methoden, Ergebnisse, 63-84. Stuttgart: Franz Steiner.
Deslauriers, M. 1991. 'Plato and Aristotle on Division and Definition'. Ancient Philosophy 10: 203-219.
Falcon, Andrea. 1997. 'Aristotle's Theory of Division'. In R.R.K. Sorabji. ed., Aristotle and After. Bulletin of the Institute of Classical Studies. Supplementary Volume 68, 127-46. London.
Fobes, F.H. 1919. Aristotelis Meteorologicorum Libri Quattuor. Reprinted Hildesheim: Georg Olms 1967.
Gilbert, O. 1907. Die meteorologischen Theorien des griechischen Altertums. Leipzig: Teubner. Reprinted Hildesheim: Georg Olms 1967.
Gotthelf, A. 1997. 'The Elephant's Nose: Further reflections on the axiomatic structure of biological explanation in Aristotle'. In W. Kullmann and S. Follinger. eds., Aristotelische Biologie: Intentionen, Methoden, Ergebnisse, 85-95. Stuttgart: Franz Steiner.
--. 1987. 'First Principles in Aristotle's Parts of Animals'. In A. Gotthelf and J. Lennox, eds., Philosophical Issues in Aristotle's Biology, 167-98. Cambridge: Cambridge University Press.
Ideler, J.L. 1834-6. Aristotelis Meteorologicorum Libri Quattuor. Leipzig: Vogelius.
Kullmann, W. 1997. 'Die Voraussetzungen fur das Studium der Biologie nach Aristoteles'. In W. Kullmann and S. Follinger, eds., Aristotelische Biologie: Intentionen, Methoden, Ergebnisse, 43-62. Stuttgart: Franz Steiner.
--. 1974. Wissenschaft und Methode: Interpretationen zur Aristotelischen Theorie der Naturwissenschaft. Berlin: Walter de Gruyter.
Lang, H.S. 1998. The Order of Nature in Aristotle's Physics: Place and the Elements. Cambridge: Cambridge University Press.
Lee, H.D.P. 1952. Aristotle: Meteorologica. Cambridge, MA: Harvard.
Lennox, J. 2005. 'The Place of Zoology in Aristotle's Natural Philosophy'. In R.W. Sharples ed., Philosophy and the Sciences in Antiquity, 55-71. Burlington, VT: Ashgate.
--. 1987b. 'Kinds, Forms of Kinds, and the More and the Less in Aristotle's Biology'. In A. Gotthelf and J. Lennox, eds., Philosophical Issues in Aristotle's Biology, 339-59. Cambridge: Cambridge University Press.
--. 1987a. 'Divide and explain: the Posterior Analytics in practice'. In A. Gotthelf and J. Lennox, eds., Philosophical Issues in Aristotle's Biology, 90-119. Cambridge: Cambridge University Press.
Lettinck, P. 1999. Aristotle's Meteorology and its Reception in the Arab World. Leiden: Brill.
Lloyd, G.E.R. 1961. 'The Development of Aristotle's Theory of the Classification of Animals'. Phronesis 6: 59-81.
Peck, A.L. 1961. Aristotle: Parts of Animals. Cambridge, MA: Heinemann.
Saint-Hilaire, J. Barthelemy. 1863. Meteorologie d'Aristote. Paris: A. Durand.
Schoonheim, P.L. 2000. Aristotle's Meteorology in the Arabico-Latin Tradition: A Critical Edition of the Texts. Leiden: Brill.
Solmsen, F. 1960. Aristotle's System of the Physical World. Ithaca, NY: Cornell University Press.
Strohm, Hans. 1970. Aristotelis Meteorologie, Uber die Welt. Aritotelis Werke in Deutscher Ubersetzung, 12. Berlin: Akademie Verlag.
--. 1936. 'Untersuchungen zur Entwicklungsgeschichte der Aristotelischen Meteorologie'. Philologus Supplementband 28: 1-85.
Taub, L. 2003. Ancient Meteorology. London: Routledge.
Vicomercatus, F. 1565. In quatuor libros Aristotelis Meteorologicorum Commentarii. Venice: Guerreus.
(1) I should like to acknowledge the helpful comments of anonymous reviewers and Sylvia Berryman, Victor Caston, Henry Mendell, Stephen Menn, Liba Taub and especially Alan Code (my respondent at the San Francisco APA 2007).
(2) Gilbert, Taub, Lettinck, Schoonheim, all worthwhile in their own way. There is no English commentary or monograph devoted to the work. Barthelemy St. Hilaire's commentary is mediocre at best. Ideler's, now 170 years old, is still useful on some points. Vicomercatus is not widely available or accessible. The best philosophical treatments of the Meteorologica remain Strohm's 1936; 1970 and Solmsen's 1960. That Meteorologica IV is not part of the same scientific subject matter as I-III was recognized from antiquity (Alexander 179 3-11).
(3) Gotthelf (1987), (1997); Lennox (1987a), (1987b); Detel
(4) Kullmann (1997) according to whom these chapters are a specification of the rules we find in the Posterior Analytics: 'Doch scheint mir die Funktion von PA I, in Erganzung der Methodologie von APo. eine Propadeutik fur die Biologie zu geben, um den wissenschaftlichen Charakter dieses Gebiets zu sichern, deutlich zu sein' (51; cf. 58); cf. Charles, 312: 'It is no exaggeration to say that the study of biological kinds precipitated a crisis in Aristotle's thinking about definition.' Lennox (2005) provides a variant on Kullmann's position, namely that Parts of Animals I claims to be providing methodological prescriptions for biological science 'because the standards that apply to animals will include standards that would be missed completely if one were to focus solely on principles that apply to natural change in general' (62).
(5) PA I 2-4. For discussion see Kullmann (1974), 53-79; Lloyd.
(6) Cf. PA I 3, 644a4-6; I 2, 642b5-9.
(7) For recent discussions see Falcon, DesLaurier.
(8) I 4, 341b1-5 announces the subject of flames, shooting stars, torches and goats. And though the aurora (I 5), comets (I 6) and the milky way (I 8) are introduced individually, all the combustions are grouped together in the concluding formula at I 8, 346b10-15, which ends by commenting on their commonality of place ([TEXT NOT REPRODUCIBLE IN ASCII.]). Of these sub-groups, only the milky way is a unique individual. The phenomena of condensation are introduced at I 9, 346b16-20. They share a place and a set of causes and principles. The discussion is concluded with an ending formula (I (12), 349a9-11) that enumerates the five members. Winds, rivers and the sea are introduced together (I (13), 349a12-14). There follows a lengthy discussion of the rivers and the sea, which is officially concluded at II 3, 359b22-6. The winds are reintroduced at II 4, 359b27-8, and wrapped up at II 6, 365a10-13, before earthquakes are introduced as a separate but related topic (II 7, 365a14-15). II 9 introduces lightning and thunder, hurricanes and firewinds (369a10-12). The discussion officially concludes at III 1, 371b14-17. Finally reflections are introduced III 2, 371b18-21. The entire discussion in the treatise so far (phenomena above the earth) is concluded at III 6, 378a12-14. The subterranean phenomena are introduced at III 6, 378a15-16 and further discussion is promised on specifics (378b5-6), but is not delivered.
(9) Combustions: burning flames, torches, goats, shooting-stars, ejected shooting stars, aurora borealis (blood-red colors, chasms), comets (tracking and trailing), milky way; condensations: upper (rain, snow, hail), lower (dew, frost); emanations: rivers (from the various mountains), interchange of sea and land, sea (source and salinity), winds (from the various points of the compass), earthquakes, tidal waves; violent atmospheric phenomena: thunder, lightening, hurricanes, whirlwinds, firewinds and thunderbolts; reflections: haloes, rainbows, sun-dogs, mock-suns and rods; minerals: realgar, ochre, ruddle, sulphur, cinnabar; metals: iron, gold, copper.
(10) PA I 4, 644b7-11
(11) Place is obviously not among the canonical causes (Ph IV 1, 209a18-22; but IV 5, 213a1-4; Cael IV 3, 310a31-b19; see Lang 72-4; 278-9), but it is recognized in the biological works among the other orders of difference, specific and generic. 'These parts respectively are either identical, or diverse, in the ways already described: viz. specifically, or in respect of excess, or by analogy or differing in position' (Peck's trans. HA I 2, 488b30-2; cf. HA I 2, 486b22-487a1).
(12) I 4, 341b1-3 (flames); I 5, 342a36-b1 (aurora); I 8, 345a11-12 (milky way); II 1, 353a32-3 (sea); II 5, 363a18-20 (winds); II 8, 369a7-9 (earthquakes); III 2, 371b18-21 (haloes and rainbows).
(13) Mete I 2 discusses the material and the efficient causes explicitly in those terms.
(14) Strohm 1, identifies three zones as well, those mentioned in I 1: the area neighboring the motion of the heavens, the area common to water and air, and the area of the earth. This division is appropriate as a first enumeration of the commonly considered kinds, but ultimately does not stand up to the critical treatment in the bulk of the treatise.
(15) At II 4, 360a10-20 Aristotle says that the exhalations can be mixed as they rise from the earth, and that mixtures with a preponderance of vapor are the origin of rain, while those that are mainly dry form the winds. But he does not say that there is a difference between the kind of dry exhalation that produces combustions and those that produce winds. Again, at I 4, 341b15-18 he calls the dry exhalation 'fire,' and says 'for we have no common name to cover all the smoky exhalation ([TEXT NOT REPRODUCIBLE IN ASCII.]), but because it is the most inflammable of all substances we must adopt this nomenclature' (modified Lee trans.). However, Aristotle seems here to have in mind the distinction between fire and smoke, not that between different kinds of smoky exhalations.
(16) The upward movement and the condensation of the air as it contracts (to produce ejections) are explicitly called moving causes (I 4, 342a27-30).
(17) I 4, 341b5: [TEXT NOT REPRODUCIBLE IN ASCII.]
(18) For fine texture: III 1, 371a15-16.
(19) Combustions differing by quantity (I 4, 341b24-5); differing by altitude (I 7, 344a34-5); implicitly by duration (shooting stars, comets, milky way I 8, 346a19-23); condensations differing by quantity (I 9, 347a11-12); by time, altitude and cold/hot (I 10, 347a13-28).
(20) So as Aristotle complains (PA I 3, 643b19-23): 'Suppose they make the division into 'wingless' and 'winged,' and then divide 'winged' into 'tame' and 'wild' or into 'pale' and 'dark': neither 'tame' nor 'pale' is a differentiation of 'winged,' but the beginning of another line of differentiation, and can come in here only by accident' (Peck).
(21) GA II 1, 732b15-30; so Charles, 334, on mutual irreducibility of the generic differentiae.
(22) See Strohm 2-12, Solmsen 397-9 for a discussion of these issues.
(23) Strohm 10-11
(24) These passages (cited by Solmsen 397) expressly relate the exhalations to the principles from higher sciences: I 3, 340b23-9; 4, 341b6-21 (where Aristotle lays out the principles of the exhalation at the beginning of the chapter); 7, 344a8-13 (an explicit restatement of the principles at the beginning of a section).
(25) Aristotle does not comment on this remarkable fact, though in Sens 5 he recognizes that [TEXT NOT REPRODUCIBLE IN ASCII.] arises from the smoky exhalation (443a29-30).
(26) And if there were any doubts about Aristotle's intention in characterizing this upper pair as vertically oriented, they are dispelled when we see that the sub-differentiae of the places are themselves vertically oriented. Aristotle divides the kapnosphere (as I call the area of the dry, smoky exhalation) into upper, middle and lower bands. The upper band, which most of all is dragged by the sphere of heaven, and which therefore tracks most closely the motion of the stars, and partakes most of all in the heavenly nature, is devoted to the upper comets and the milky way (I 7, 344a33-b1; I 8, 345b35-346a6). The middle band, which is discussed first, is populated by the ordinary torches, goats, and shooting stars (I 4, 341b35-6). Finally there is the lower band, which because it is partly mixed with the cooler wet air, may give rise to ejected shooting stars (I 4, 341b36-342a3). Similarly too, the atmosphere is divided into an upper region, the place of rain, snow and hail, and the lower region, the place of dew and frost (I 10, 347a13-16).
(27) As Cherniss 128-30 has noted.
(28) Note that, like the combustions, the dry winds do not cycle back.
(29) In Mete 93 32-3
(30) The locus classicus is PA IV 10, 686a24-687a2. For a discussion of this and other scalae in the biology, see Coles.
(31) Ph VIII 7
(32) For dualizing, [TEXT NOT REPRODUCIBLE IN ASCII.], see PA IV 13, 697b1 (and other passages cited in Bonitz 265a8ff).
(33) It has often been noted that Aristotle provides only partial definitions in the biological works, e.g., Gotthelf (1987) 191-2; Kullmann (1997) 58.
(34) PA II 13: Man and viviparous quadrupeds have soft eyes (for sharp vision), which need protection. Thus they have two eyelids. Most birds have two eyelids for the same reason. But in large birds, the earthy material, which would have gone into their wings, is redirected to their skin, which in consequence is hard. This makes it impossible for big birds to have two eyelids. They have a nictitating eyelid in compensation. All oviparous quadrupeds have hard skin, not because their earthy material is redirected, but because of their hard scales. And they do not need any compensating nictitating eyelid, since their sight is duller. Insects and crustacea have hard eyes, which they can further protect by moving. Fish have soft eyes, but do not need protection of eyelids because they do not strike against objects in the water.
(35) I owe this insight to Alan Code.
(36) See Gotthelf (1987).
(37) See Solmsen 408-10 on Heraclitean influences.
Department of Classics
University of Oregon
Eugene, OR, 97403
|Printer friendly Cite/link Email Feedback|
|Date:||Mar 1, 2009|
|Previous Article:||Aristotle on the order and direction of time.|
|Next Article:||Prediction, precision, and practical experience: the Hippocratics on techne.|