The restricted scope of the Cyrenaic criterion constitutes the target of the Peripatetic philosopher Aristocles of Messene, whose criticism of the Cyrenaic theory is reported by Eusebius.

At the outset, Aristocles sketches out the theory which he intends to attack in the following terms.

Aristocles' objections are not targeted against the affirmation that the pathē are apprehensible, but against the restrictive claim that only they can be apprehended. His goal is to show that the Cyrenaic position is inconsistent by arguing that we know many things in addition to our pathē, or that our awareness of pathē entails (or presupposes) that we know things about external objects.

Having answered the most serious criticisms of his account, Aristotle concludes his arguments that the earth must be a sphere – and so concludes De Caelo II - with the particulars of its size. De Caelo III, 1, opens with a brief summary of what has been discussed thus far concerning the first heaven and its parts, the stars carried in the heaven, and the composition and nature of these things, including, finally, that they are ungenerated and incorruptible (298a24–27). Having dealt with the first (and highest) element, aether, Aristotle turns to the other elements.

These elements – each of which possesses its own specific nature – are “by nature.” Natural things are either substances or operations and affections of substance. “Substances” refers to the elements – aether, air, fire, water, and earth – and things composed of them, such as the heaven, as a whole and its parts, as well as plants and animals and their parts; “operations and affections” include movements of each of these according to their proper power as well as their alterations and transformations into one another. Obviously, Aristotle concludes, the study of nature is for the most part concerned with bodies because all natural substances are either bodies or come to be after bodies, and all involve magnitude. The investigation must also include generation and destruction, as “operations and affections” of the elements and all things composed of them (298b8–11).

“Nature is everywhere a cause of order.” This claim, together with the evidence that Aristotle marshals to support it, forms a consistent theme throughout his entire corpus. There has been and continues to be considerable disagreement about Aristotle's various arguments – their topics, what they say, if they are valid. But there can be neither doubt nor quarrel that within his philosophy as a whole this claim is central: nature is everywhere a cause of order.

As Mansion argues, nature gives the world not only order, but intelligibility. I shall conclude that Aristotle's claim that nature is everywhere a cause of order constitutes a first principle that informs his physics as a science, and consequently, the particular problems and solutions within physics. Beyond his physics, it also appears at work within his metaphysics. Indeed, it constitutes one of his most important philosophic commitments.

As a first principle, this claim is never proven and so is not derived by or within physics – or any other science. Rather, as I shall argue, the topics, proofs, and arguments of Aristotle's physics and metaphysics presuppose and work in terms of the claim that nature is everywhere a cause of order. Hence, we see this claim at work because we see the physics that assumes it as a first principle.

When, at the opening of Physics III, 1, Aristotle lists those things without which motion seems to be impossible, the infinite appears first, followed by place, void, and time. He examines the infinite after completing his account of motion (Physics III, 4–8). I omit examination of it here for two reasons. (1) As noted earlier, the infinite, place, void, and time are subordinated to the definitions of motion and nature, but they are not subordinated to one another. Thus it is not necessary to examine the infinite in order to examine place and void. (2) Furthermore, the infinite plays a largely negative role in the order of nature. There is no such thing as an infinite magnitude, and indeed the being of the infinite is matter and even privation (Physics III, 7, 207a35, 208a). However, when Aristotle identifies the infinite with matter and privation, he anticipates his account of place in an important way.

As matter or privation, the infinite is identified with what is surrounded rather than with what surrounds, which is form (207a35–36; 208a3–4). Place, as we shall see, is not itself form, but strongly resembles form: both are limits (Physics IV, 4, 211b13). Ultimately, Aristotle defines place as the first unmoved limit of that which surrounds (212a20–21). As I shall argue, place, as a first limit, serves as a cause of order: it renders the cosmos determinate in respect to “where things are and are moved.”

With the account of the nature and activities of the elements, the analysis of place and the elements is complete. Taken together “the where,” i.e., place and not void, and inclination, i.e., the active orientation of each element toward its respective proper place, exhibit the order of nature in Aristotle's physics. Nature is always a cause of order, and that order is constituted within the world by two principles. Place is the first limit of the containing body and renders the cosmos orderly in respect to direction; thus the cosmos exhibits “up,” “down,” “left,” “right,” “front,” and “back” immediately and intrinsically in itself. Second, inclination constitutes the very nature of each element as an intrinsic source of being moved toward its proper place, e.g., up for fire and down for earth; consequently, elemental motion is never random or irregular because, in the absence of hindrance, each element cannot fail to be moved toward (and to rest in) its proper place.

These principles solve a number of problems and so express the order of nature in three ways. (1) Place renders the cosmos determinate in respect to “where,” while each element is ordered to its proper place. Place and the inclination of each element work together to produce the order of nature in respect to the intrinsically directional motion of the elements.

The conclusion that the elements are generated from one another raises the question of how this generation occurs. This question – indeed, the issue of the generation of the elements generally – presupposes a prior problem: what differentiates the elements? This problem is serious for two reasons. First, whatever differentiates each element must be generated when the element is generated. Second, whatever generates each element renders it unique and so makes the element be what it is according to its definition. Hence an account of what differentiates each element is central to the nexus of topics concerning the generation of the elements, their natures and motions. This “prior problem” is solved in the remainder of De Caelo III and IV, and the final account of the generation of the elements appears in the De Generatione et Corruptione.

Aristotle first criticizes his predecessors (De Caelo III, 7 and 8). The followers of Empedocles and Democritus explain the generation of the elements as an excretion of what is already there; this view reduces generation to an illusion – as if it requires a vessel rather than matter (De Caelo III, 7, 305b–5). And Aristotle quickly shows that it entails that an infinite body is contained in a finite body – which is impossible (305b20–25).

On other accounts, the elements change into one another, by means of shape or by resolution into planes (305b26).

Aristotle's arguments about the void present special interests and problems because of their long history in Aristotle's commentators. A full accounting of the responses to these arguments lies beyond the scope of my analysis, although some special cases will be taken up. One point however has been crucial for interpretations of these arguments: Euclidean geometry requires a three-dimensional infinite space. Euclid flourished (probably) one generation after Aristotle, and his geometry was enormously influential. Because Aristotle defines place as the limit of the first containing body, place was often thought to be both finite and, as we have seen, a two-dimensional surface. Hence on this view, Aristotle's notion of place fails on both counts to meet the requirements of Euclidean geometry. Taken together, the requirements of Euclidean geometry and the apparent failure of Aristotle's account of place to meet them often motivated first a commitment to the void and then criticism of Aristotle's arguments rejecting the void. Therefore, these criticisms derive from the conjunction of Euclidean geometry and a common misunderstanding of Aristotle's account of place. My interest lies in a direct analysis of the arguments concerning the void in Physics IV.

Place has not been established as the exclusive answer to the question “where are things?” because void [κενόν, i.e., “empty”] presents an alternate answer. Consequently, void must be examined, and, Aristotle says, the same questions must be posed for it as for place, namely, “if it is or not, and how it is and what it is.”

Before turning to Aristotle's arguments on place and void, we must consider the structure of Physics IV as a logos and its relation to the other books of the Physics, especially Physics II and III. In Physics II, 1, Aristotle defines nature as a source of motion and rest (192b14); Physics III, 1, opens with the claim that in order to understand nature, we must also understand motion (200b12–14). In effect, Physics III starts from the subject of physics established in Physics II. In short, motion and nature are coextensive, they are found together, and they, and those things required by them, form the primary subject matter of physics as a science. Hence, Aristotle concludes, it is clear that universal and common things must be examined first, namely, motion and those things without which motion seems to be impossible, including the infinite, place, void, time, and the continuous.

The Structure of the Arguments

Aristotle proceeds accordingly. First, “motion” is defined (Physics III, 1-3), and then “the infinite” (Physics III, 3–8). “Place,” “void,” and “time” occupy Physics IV, while “the continuous” along with the related notions “in contact” and “in succession,” “points,” and “lines” occupy Physics V and VI. These terms follow motion and nature not only rhetorically – they are next in the text – but also logically. That is, the examination of them presupposes the definition of motion that he has just established. Hence, place and void are considered, defined, and evaluated in terms of Aristotle's definitions of motion and ultimately nature.

With this account of “the where” – place, not void – I turn to the De Caelo and its topics, particularly the elements. As a topic, the elements present a problem: the elements – themselves unquestionably “things that are by nature” – appear in the Physics only within the examination of other topics, but are never themselves examined as a topic. A direct examination of them appears in the De Caelo. But historically the coherence of the De Caelo as a set of logoi, the definition of its topic (s), and the relation of its arguments to those of the Physics have been problematic. A consideration of the Physics and the De Caelo as topical investigations solves both problems, substantive and historical.

The Topic of the De Caelo

In the Physics, Aristotle investigates strictly defined topics. Physics II, 1, identifies things that are by nature: animals, their parts, plants, and the elements earth, air, fire, and water (192b9–11). Nature is a principle of “being moved and being at rest in that to which it belongs in virtue of itself” (192b21–22); consequently, Aristotle asserts at Physics III, 1, we must know motion if we are to understand nature, and he lists the “common and universal things” without which motion seems to be impossible (200b20–23). The investigation of proper things, he says, will come later because universal things, including the continuous, the infinite, place, void, and time should be investigated first (200b23–25).