<<

. 4
( 5)



>>

remote. What he says in the DC, however, suggests that the remoteness
that puts the celestial world beyond our grasp cannot be reduced to mere
physical distance:
(1) since no circular motion is contrary to circular motion, (2) we have to inquire
why there are several motions; (3) we have to try, though we are far removed,
to make an inquiry; (4) we are far removed not only in place but much
more because of the fact that we have perception of very few of their features (286
a 4“7).

I postpone discussion of the signi¬cance of clause (1) and focus, for the
time being, on the rest of the passage.2 In clause (2) we are told that we
have to explain why celestial motion does not consist in simple circular
motion but is articulated in a plurality of circular motions. But in clause
(3) we are warned that we may encounter some obstacles in our attempt to
provide the relevant explanation for this fact. Once again, perceptual
remoteness is identi¬ed as the main source of the problem. But this time
Aristotle is more informative. He makes it clear that what he has primarily
in mind is not physical distance. Remoteness is mainly due to the fact that
we have only a limited access to the celestial region “ clause (4). In other
words, no matter how extensive and careful our observations may be, they
will provide only a little information about few features of the celestial
bodies.3 However elusive and cryptic this further passage may be, it


2 I shall return to the claim that there is no (circular) motion contrary to circular motion in the
discussion of celestial matter. See below, pp. 106“7.
3 In the passage I have quoted in the epigraph, Aristotle insists that it is worthwhile seeking to
attain more understanding concerning these bodies, though the resources at our disposal are few
and we are at such a great distance from what happens in the heavens (292 a 14“17).
The limits of Aristotle™s science of nature 87
con¬rms that physical distance alone does not supply a full account of the
dif¬culties encountered in the study of the celestial world. The celestial
bodies are at a great distance from earth and too far away to be accessible
to us by perception. But this is not the whole story. If, therefore, we want
to understand why Aristotle is so pessimistic about the extent to which we
can know the celestial world, I suggest that we turn to his conception of
this world. It is in fact the very way in which Aristotle conceives of the
celestial bodies that makes them remote from the natural bodies we
experience on earth, and this quite independently of any considerations
about the physical distance that may exist between them and us. Put
differently, the celestial bodies are conceptually, and not simply geograph-
ically, remote. Moreover, the conceptual remoteness in question ultimately
depends on the fact that the explanatory resources at our disposal,
according to Aristotle, are not adequate to provide a positive characteriza-
tion of certain important features of the celestial world. More directly, the
celestial bodies are conceptually remote because they possess a nature
different from, and not completely reducible to, the natures we ¬nd in
the sublunary world.

discontinuity
In antiquity it was commonly held that the celestial world is a somehow
special region of the sensible world. Stability and incorruptibility were
often offered as the differentiating features of the celestial world. But the
position of Aristotle is more speci¬c and in fact stronger than a generic
commitment to the existence of some difference between the celestial and
the sublunary world. So far I have argued that the natural world is
understood by him as a causal system of a speci¬c type. I now would like
to add that this causal system admits an important discontinuity between
the celestial and the sublunary world, and that as a result of this discon-
tinuity there is unity without uniformity in the natural world. An example
may help to illustrate his position. Aristotle is committed to the view that
the characteristic behavior of the celestial bodies “ their complex but
regular motion around the earth “ requires the existence of a speci¬c type
of material principle. He is persuaded that the distinctive motion of the
celestial bodies can be explained only on the assumption that they are
made of a simple body that naturally performs circular motion. For him,
this celestial simple body is distinct from the ultimate material principles
out of which everything in the sublunary world is constituted. Very few in
antiquity were prepared to postulate the existence of a material principle
Aristotle and the Science of Nature
88
different from, and not reducible to, earth, water, air, and ¬re. This was
ultimately due to the fact that very few in antiquity were prepared to share
with Aristotle the belief in the existence of an important discontinuity
between the celestial and the sublunary world. The belief in the existence
of unity together with an important discontinuity explains not only why
Aristotle takes the view that the celestial bodies are made of a special
simple body; it also explains why he does not admit the existence of a
nature over and above the celestial and the sublunary natures. Aristotle at
times speaks of nature, and says that nature does nothing in vain, or that it
always does the best possible thing.4 Both slogans may be understood as
claims about a cosmic nature, but in general they are better understood, I
think, as claims about a collection of particular natures. It is notoriously
dif¬cult to explain why this interpretation is to be preferred.5 My sugges-
tion is that Aristotle does believe in the existence of celestial as well as
sublunary natures, but does not think that nature is a uniform principle of
motion and rest. On the contrary, he is persuaded that the celestial
natures are in some important respect different from, and not completely
reducible to, the natures we experience in the sublunary world.
The beginning of Lambda is dogmatic but instructive on this point.
Lambda is an investigation into substance on the crucial assumption that
there are different kinds of substances. According to Lambda, there are
sensible and immovable substances. First of all, a relation is established
between the fact that the sensible substances are subject to motion and rest
and the fact that they are sensible. The idea is that these substances need
to be realized in some matter or other in order to be subject to motion
and rest. By being realized in some matter or other, they are sensible. By
contrast, the immovable substances are immaterial and non-sensible, as
they are not subject to motion and rest. Secondly, sensible or material
substances are themselves divided into two parts: eternal sensible sub-
stances (celestial bodies) and perishable sensible substances (animals and
plants). In other words:
(1) there are three kinds of substances: (2) one that is sensible “ which all admit,
and of which one subdivision is perishable, namely plants and animals, and

4 In the physical writings the assumption that “nature does nothing in vain” occurs in DC 271 a 33,
291 b 13“14; DA 432 b 21, 434 a 31; PN 476 a 13; PA 658 a 8, 661 b 24, 691 b 4“5, 694 a 15, 695 b
19“20; IA 704 b 15, 708 a 9, 711 a 19; GA 739 b 19, 741 b 4, 744 a 37“8, whereas the assumption
that “nature always does the best possible thing” is found in Phys. 260 a 22“3; DC 288 a 2“3; GC
336 b 27“8; PN 469 a 27“8; PA 658 a 23, 687 a 16“17; IA 704 b 15, 708 a 9“10.
5 On this point see, for example, Balme (1987: 275“85); Furley (2003: 71“84, in particular 73);
Preus (1975: 221“48).
The limits of Aristotle™s science of nature 89
another eternal “ of this we must grasp the elements, whether one or many;
(3) and another that is immovable (1069 a 30“3).6
This passage con¬rms that Aristotle conceives of the natural world as one
region divided into two parts: animals and plants form one subdivision of
the sensible substances, and they are contrasted with the celestial bodies.
The division into sensible and immovable substances casts some light
upon another fundamental, though often neglected, truth: for Aristotle,
the natural world is only one department of reality, not the totality. Both
in the Metaphysics and in the Physics, Aristotle puts himself in direct
continuity with the activity of the physiologoi. He presents the activity of
his predecessors as a search for the explanatory principles of the natural
world. There is no doubt that Aristotle is in essence right. From the very
beginning, and independently of Aristotle, the investigation of the natural
world consisted in the search for the relevant explanatory principles of a
variety of natural phenomena on the basis of the fundamental assumption
that the natural world is to some extent intelligible to us. There is,
nevertheless, at least one important difference between the activity of
the physiologoi and the investigation of the natural world as it is under-
stood by Aristotle. The pre-Platonic inquiry into nature was rooted in the
conviction that the natural world is the totality of reality. By contrast,
Aristotle is committed to the view that the natural world is not a causally
closed system and can be adequately explained only by an appeal to a
certain number of extra-natural principles. In Book 8 of the Physics,
Aristotle provides an argument for the existence of a type of principle
which is a principle of change but itself stands outside any actual and
possible change. This argument is required for a fully adequate account of
change, and as such it is an essential piece of Aristotle™s science of nature;
however, it also takes Aristotle outside of the natural world.

the boundaries and the scope of aristotle™s study
of the soul
Aristotle™s conviction that there is an important discontinuity within the
natural world between the celestial and the sublunary world leads him to
the further view that the celestial and the sublunary natures cannot be
explained in the very same terms. I would like to provide evidence for this

6 The transmitted text may be corrupted. I follow Michael Frede and transpose he pantes
¯
homologousin before hes he men phtharte and keep the second occurrence of aidios. For a
¯¯ ¯
discussion of the text, see Frede (2000: 78“80).
Aristotle and the Science of Nature
90
further view by looking at the way Aristotle™s account of the soul is
generated in the DA. From the DC we have learned that celestial bodies
are not mere bodies but intelligent living bodies engaged in motion (292 a
18“22). I have argued that Aristotle credits celestial bodies not with mere
life but with intelligent life because they are engaged in a special type of
motion which can be adequately explained only if these bodies are
credited with thought and desire.7 But how much of what Aristotle says
in the DA about the soul is relevant to the study of celestial life?
In the DA Aristotle is concerned with the soul on the crucial assump-
tion that the soul is the provider of life. Though life is a phenomenon that
cannot be reasonably denied, what counts as life is far from being clear.
Part of Aristotle™s enterprise consists in seeking clarity about the soul and,
accordingly, life. One of the most important results delivered by the DA is
the view that living and being alive (zen) is said in many ways (413 a 22).
¯
But this view cannot be presupposed or anticipated at the outset of the
investigation. In the opening lines of the DA we are told that the study of
the soul will result in knowledge of the soul, and that this knowledge is
relevant to all the truth, but in particular to truth about nature, for this
knowledge is relevant to the study of zoia.8 On the interpretation I have
¯
recommended, the most generous reading of zoia is to be preferred. In
¯
this context, zoia means all the living beings that there might be, including
¯
any living beings that there might be superior to human beings.9 The
most general and inclusive reading of zoia does justice to the fact that
¯
the DA is a systematic investigation of life, and as such it is not restricted
to animal life. Aristotle is persuaded, rightly, that any such restricted
investigation would prevent the investigator from arriving at a full

7 I have also argued that Aristotle never offers an argument in support of the claim that the celestial
bodies are engaged in intelligent life. Apparently, Aristotle thinks that the explanatory bene¬ts
depending upon the assumption that the celestial bodies are living bodies are also an indirect
argument in support of the assumption itself. In chapter 3 I have made an attempt to provide
(some of ) the reasons that may help us to understand why Aristotle credits celestial bodies with
life.
8 I have reported and discussed this passage in chapter 1, “The unity, structure, and boundaries of
Aristotle™s science of nature.”
9 I have already pointed out that in the Timaeus the stars are zoia, on the assumption that they are
¯
alive. More directly, they are immortal living beings (92 c), and as such they are also divine living
being (40 b). Moreover, the sensible world as a whole is a zoion (29 b; 30 d; 33 b, etc.). The
¯
sensible world contains all the immortal and mortal living beings that there might be (92 c), and
for this reason it is also called the perfect living being (32 d). I have also pointed out that
Xenocrates in his Life of Plato ascribed to Plato the view that there are ¬ve types of zoia, one for
¯
each of the ¬ve elements. See Simpl., In DC 12.22“6 and In Phys. 1165.35“9 (¼ Heinze, fr. 23). In
this case zoia is used in its most general and inclusive sense to refer to all the living beings that
¯
there might be, including demons and gods.
The limits of Aristotle™s science of nature 91
understanding of life. Consider, however, the next occurrence of zoion in
¯
the DA:
For those who now speak and inquire into the soul seem to study the human soul
only. But we have to be careful not to overlook whether the de¬nition of the soul
is one, just as in the case of zoion, or different for each soul, as in the case of
¯
horse, dog, man, god; zoion, the universal, being either nothing or posterior
¯
(similarly with regard to any other common predicate) (402 b 3“9).
At this preliminary stage of the investigation we only know that Aristotle
is about to embark on an investigation that is not restricted to any
particular class of souls. But the unity of the soul is a real problem for
an investigation with the aspiration to be unrestricted. In this passage,
Aristotle recommends considering whether there is one de¬nition for the
soul as there is one de¬nition for zoion because the possibility that horse,
¯
dog, man and god do not form a genus cannot be ruled out. By now the
insertion of god along with horse, dog, and man should be no surprise.
This insertion is not only intended to make the case for equivocity more
vivid; it is also dictated by the logic of the Greek language. Needless to
say, this insertion raises a genuine concern about the scope and the
boundaries of the investigation conducted in the DA.
One thing that this passage makes very clear is that the problem of the
unity of the soul (and, accordingly, life) requires a ¬rm grasp of the scope
and the boundaries of the investigation conducted in the DA. Interest-
ingly enough, in the DA Aristotle does not concern himself with all the
living beings that there might be. By his own admission, he restricts his
investigation to perishable living beings (413 a 31“2; 415 a 8“9). The fact
that the study of the soul is programmatically con¬ned to the soul of
perishable living beings must not be understood as evidence for the view
that life manifests itself only in the form of perishable life. Aristotle™s
natural science is hospitable to both perishable and imperishable life. Like
Plato, Aristotle is prepared to speak of celestial life. Unlike Plato,10
Aristotle never refers to the celestial bodies as zoia. But there is no doubt
¯
that he is prepared to ascribe a certain form of life to the celestial bodies.
This is immediately relevant to the scope and the boundaries of the
investigation conducted in the DA. As a dedicated student of life, Aristotle


10 In the Timaeus, Plato recognizes the celestial bodies as zoia (39 a; 39 b). Elsewhere he is more
¯
tentative. In the Laws, the explanation of celestial motion requires a soul of a certain kind (897
b“c). But this time Plato leaves it open whether this kind of soul is directing the body from
inside, or pushing the body from outside, or conducting the body in some other way (899 a).
Accordingly, he is no longer sure that the celestial bodies are zoia (899 a“b).
¯
Aristotle and the Science of Nature
92
is interested in life in all its manifestations; unlike his predecessors, he
does not arbitrarily restrict his investigation to any class of living beings.
At the same time, however, Aristotle does restrict his investigation to a
study of perishable life, to the exclusion of imperishable life. Why does
Aristotle restrict the scope of his investigation to perishable life? How can
he restrict his investigation to this type of life? Why is this restriction not
an arbitrary one (like, for example, the restriction of the investigation to
the case of animal or human life)?
One way to answer these questions is to re¬‚ect on the analogy between
souls and rectilinear ¬gures which Aristotle offers just before engaging in a
study of the nutritive soul. The centrality of this analogy for the correct
reading of the DA cannot be disputed.11 Among other things, this analogy
reveals that Aristotle is a systematic investigator of the soul in the sense
that he has a plan for the study of the soul and this plan dictates not only
the order but also the boundaries of the investigation. Aristotle argues that
just as the rectilinear ¬gures are ordered in a series beginning with the
triangle, so are the souls beginning with the nutritive soul (414 b 20“1).
The analogy with the rectilinear ¬gures provides Aristotle with a method
of studying the different types of souls (and, accordingly, the different
forms of life). Just as the triangle exists potentially in the rectangle, so the
capacity for nutrition, growth, and decline exists potentially in the cap-
acity for perception. But this crucially depends on the fact that self-
nutrition, growth, and decline are constitutive of perishable life. Note
that in this context Aristotle is not merely speaking of capacities of the
soul; he is speaking of types of souls (414 b 22 and 24“5). The speci¬c
accounts of the nutritive, sensitive and intellective souls are secured on the
crucial assumption that the souls are ordered in series.
I do not deny that what Aristotle says in the DA may be relevant to a
study of celestial life, celestial thought, and celestial desire. But I contend
that the celestial souls go beyond the scope of the investigation offered in
the DA.12 The celestial souls and the celestial bodies are intractable by the

11 Ward (1996: 113“28) rightly insists on the “logical interpretation” of the analogy. Aristotle is not
only denying that there is a generic soul over and above the different types of souls; he is also
denying that the different types of soul constitute a genus and that they can be studied in the way a
genus is studied.
12 The ancient debate on the scope of the De anima is re¬‚ected in the Prologue to the commentary
on the DA which is traditionally attributed to Simplicius. See [Simplicius], In DA 3.21 “ 4.11. In
the Aristotelian tradition, psychology is programmatically restricted to the study of the soul of
perishable living beings. Like Aristotle, Alexander does not deny that there are celestial souls, but
he is persuaded that the celestial and sublunary souls are merely homonymous. See Alexander of
Aphrodisias, DA 28. 25“8. For Alexander, no unitary account of the soul would be possible, if the
The limits of Aristotle™s science of nature 93
conceptual resources developed and re¬ned in the study of the sublunary
world. More directly, Aristotle cannot embark on a study of the celestial
souls because no serial relationship exists between the perishable and the
celestial souls. Moreover, there is no serial relationship because the
celestial creatures are not engaged in any of the activities that are minim-
ally constitutive of sublunary life.13 The celestial creatures do not take in
nourishment, and as a result of this they are not subject to growth and
decline.
By this point I hope to have shown that Aristotle has to restrict his
investigation of life to the case of perishable (¼ sublunary) life. Since
Aristotle is convinced that life as encountered on earth and celestial life
are not continuous, this restriction is not arbitrary. But can the results
achieved in the DA be extended to the celestial world? By looking at the
activities that are constitutive of celestial life, and what is distinctive of
each of them, we may come to appreciate how dif¬cult it might be for
Aristotle to extend the results he has achieved in the DA to the case of the
celestial souls. I shall focus on celestial motion, celestial thinking, and
celestial desire, on the assumption that distinctions between natures
become evident in different kinds of life. More speci¬cally, if the celestial
bodies do not partake in the activities that are constitutive of life as it is
encountered in the sublunary world, or they do partake in some of the
same activities but their activities cannot be reduced to the corresponding
sublunary activities, then this will be suf¬cient proof that the celestial
natures which are responsible for governing these activities, and shaping
them into one and the same behavior, cannot be reduced to the sublunary
natures.
On Aristotle™s account of celestial motion, the celestial bodies are living
bodies which perform motion from one place to another, but which are
not engaged in any of the forms of animal motion we encounter in the
sublunary world. From the DA we learn that the animal motion we are
familiar with in the sublunary world is progressive motion, in Greek
poreutike kinesis (432 b 14) or poreia (432 b 25). This is the capacity of a
¯ ¯
living body to move around by walking, swimming, and the like. More-
over, poreia (or poreutike kinesis) is a case of motion for the sake of a
¯ ¯

investigation of the soul was extended to the celestial soul. Alexander is here speaking in his own
voice. Aristotle never says, in the DA or elsewhere, that the celestial and sublunary souls are
homonymous. But I shall show that Alexander is in essence right: Aristotle does credit celestial
bodies with life, but this life has little in common with the life Aristotle studies in the DA.
13 I say “minimally” because from the DA we learn that perishable life takes different forms and there
are different types and, ultimately, different gradation of perishable life.
Aristotle and the Science of Nature
94
speci¬c goal (432 b 15“16). Finally, it consists in the displacement of a
living body equipped with an appropriate desiderative and cognitive
apparatus. The capacity for this type of displacement involves the posses-
sion of desire and phantasia (432 b 16). Since phantasia is the ability to
form representations on the basis of perception, and is causally dependent
on perception, we can say that perception together with phantasia form
the minimal cognitive equipment required for progressive motion.14 It is
not dif¬cult to see why Aristotle thinks that both perception and phanta-
sia are required for progressive motion. Progressive motion is a case of
navigation from one place to the other; and at times this motion even
requires highly sophisticated navigational abilities: for instance, the ability
to return home from an unfamiliar territory, or the ability to migrate
from one place to the other. While perception provides the animal with
sensitivity to the environment, phantasia presents it with the goal of
motion, which also happens to be the object of desire (433 a 15) “ e.g.
home or food. From the De memoria we learn that phantasia also plays a
crucial role in the formation of memory. And there is no doubt that
perception, together with phantasia and memory, provide all the concep-
tual resources we need to explain even the most sophisticated navigational
achievements.15 In the Timaeus Plato credits the celestial bodies with
poreia (Tim. 39 b 4, and d 8). To my best knowledge, Aristotle never
credits the celestial bodies with progressive motion or poreia. I suspect
that the lack of ¬‚exibility of celestial motion is the ultimate reason for his
silence. As a matter of fact, the celestial bodies do move, but they are
unable to stop moving or to move in any other direction or way than they
actually move. I also suspect that this lack of ¬‚exibility explains why
Aristotle never credits the celestial creatures with perception or phantasia.
Aristotle regards perception and phantasia as the minimal cognitive
equipment required for navigating from one place to another. But the
motion that the celestial bodies perform regularly in the sky does not seem
to require any sensory apparatus for navigating from one place to the
other. Evidently, Aristotle does not regard celestial motion as a case of
navigation from one place to the other. But if celestial motion is not a case
of navigation, the celestial bodies need not be sensitive to the surround-
ing environment. And if they need not be sensitive to the surrounding

14 On phantasia, and the relation between phantasia and perception, I refer the reader to Wedin
(1988: 23“63). Here I am content to recall that for Aristotle phantasia is a change resulting from
the activity of perception (DA 428 b 30 “ 429 a 2).
15 For a discussion of phantasia in connection with progressive motion see Wedin (1988: 39“45). See
`
also Labarriere (1984: 17“49).
The limits of Aristotle™s science of nature 95
environment, they need not be capable of phantasia. In fact, phantasia is
causally dependent on perception. To put it in another way, celestial
motion does not require the existence of celestial perception, or celestial
phantasia.16 Nor does it require a speci¬c locomotory apparatus. The
celestial bodies are composed of a body that naturally performs circular
motion and that crucially contributes to the explanation of the eternal,
blissful life that, according to Aristotle, these celestial creatures enjoy.
If we turn to celestial thinking, another constitutive activity of celestial
life, we ¬nd a similar situation. Any attempt to reduce celestial thinking to
human thinking as it is discussed in the DA runs into severe dif¬culties.
On the one hand, Aristotle is explicitly committed to the existence of
celestial thought; on the other, he does not postulate the existence
of celestial perception or celestial phantasia. The exercise of celestial
thought, celestial thinking, does not depend on the possession of percep-
tion and the exercise of phantasia. On the contrary, from the DA we learn
that human thought crucially presupposes phantasia, perception, and
ultimately, a living body of a certain kind. More directly, thinking as it
is encountered on earth is only human thinking, which is crucially
dependent on a bodily organization of a very speci¬c type, a human
body. In the DA Aristotle insists several times on the necessity of phanta-
sia for the exercise of human thought. Right at the beginning of the DA,
for example, he claims that (human) thinking is not without phantasia
(403 a 8“10). Later on in the DA, he insists that human beings never think
without phantasmata (431 a 16“17, 431 b 2, 432 a 8“10, 432 b 12“14). Of
course, this would not be possible if we were not capable of phantasia. But
phantasia depends on the possession and the actual exercise of perception
(429 a 1“2) and, ultimately, the use of one or more of the sense organs.17
There is also evidence that human thinking might be crucially dependent
on perception in a different, though related, way. In his discussion of
thinking, Aristotle appears to be willing to consider the idea that there are
not two distinct discriminatory capacities “ perception and thought “ but
rather one capacity which can be in two different states (429 b 13 and b
20“1). It is also signi¬cant that at the end of the DA, turning to animal
motion, Aristotle speaks of one, and only one discriminatory capacity

16 Celestial motion does not require the existence of celestial memory either. Quite independently of
any considerations about the nature of celestial motion, there cannot be room for celestial
memory because for Aristotle memory is generated by perception and crucially depends on the
possession of phantasia.
17 On the relation between thinking and phantasia, see Wedin (1988: 100“59). On thinking in
general, see Modrak (1987: 209“36) and Wedin (1992: 243“71).
Aristotle and the Science of Nature
96
(432 a 15“16). At this stage of the discussion, having already gone through
perception and thought, this can hardly be just a loose way of speaking of
two distinct discriminatory capacities. It is not necessary to enter into a
discussion of the sort of relationship between thought and perception that
Aristotle might be trying to establish here. For the present discussion it is
enough to realize that, for Aristotle, it is not the case that human beings
are equipped with two wholly independent and disconnected cognitive
capacities, namely perception and thought. On the contrary, the actual
exercise of thought, and perhaps its existence, rests on the exercise of
perception. Also, in the light of these remarks, it should be clear why the
activity of thinking and the capacity for thought are a genuine problem
for Aristotle.18
We come to a similar conclusion when we ¬nally turn to celestial
desire. I have already said that Aristotle does not only credit celestial
bodies with thought but also with desire. The ultimate reason for this idea
is to be found in the assumption that both cognition and desire are
required for an adequate explanation of celestial motion. More speci¬c-
ally, from the DA we learn that Aristotle is committed to the view that
there are three kinds of desire: appetitive desire, spirited desire, and
rational desire (414 b 1“6).19 Celestial desire is obviously a case of rational
desire: since celestial bodies are not equipped with celestial perception,
they cannot be equipped with non-rational desires (either appetitive or
spirited desires). In the light of what I have argued so far, however, it is
clear that the celestial bodies cannot be related to their appropriate object
of desire, a speci¬c object of thought, by virtue of phantasia. There is in

18 The situation seems to be the following. On the one hand, Aristotle seems to be con¬dent that a
uni¬ed account of thinking, that is an account that includes human, celestial and divine thinking,
is possible. On the other hand, he never engages in an attempt to provide this uni¬ed account. By
“divine thinking” I mean the thinking of divine intellects. This latter is to be distinguished from
the thinking of any of the intelligent living creatures that we encounter in the natural world,
including the celestial region of the natural world. For Aristotle, divine intellects are disembodied
intellects whose life consists in thinking. Simply put, they are not engaged in thinking, but they
are thinking. There is evidence that for Aristotle divine thinking, or the thinking of the
disembodied intellects, is not a form of thinking among others. The word “thought” is used
without quali¬cation only with respect to divine thought. In the DA human nous is often
quali¬ed. At least a couple of times, Aristotle refers to it as “the so-called nous,” ho kaloumenos
nous (429 a 22; 432 b 26). Evidently, nous refers primarily to a divine being and the reference to
human thinking is derived from this primary meaning. Needless to say, this fact is relevant for the
project of ¬nding a unitary account of thinking; that is, an account that uni¬es divine, celestial,
and human thinking. I shall not engage in this further project.
19 This tri-partition of desire is part of a more general thesis. For Aristotle, if something is capable of
desire (minimally appetitive desire), then it is capable of cognition (minimally perception in the
form of sense of touch), and vice versa. In other words, for any x, if x is able to desire, then x is
able to cognize; and if x is able to cognize, then x is able to desire.
The limits of Aristotle™s science of nature 97
fact no phantasia in the celestial world. Once again, this conclusion does
not square with the account of progressive motion offered in DA:
(1) In general, then, as it has been said, it is in so far as the animal has the
capacity for desire that it has the capacity for its own motion; (2) but the capacity
for desire is not without phantasia: (3) all phantasia is either connected with
reason or with perception; (4) also the other animals partake in the latter (433 b
27“30).
Clause (2) commits Aristotle to the claim that desire requires phantasia.
Apparently, Aristotle is committed to the view that animals are presented
with their own object of desire by virtue of phantasia. Clauses (3) and (4)
help us to understand that this is a claim about rational as well as non-
rational animals: phantasia is connected with reason or with perception,
and the latter is shared by all animals. However, clause (1) helps us to
qualify this claim. Apparently, this is to be taken as a claim about desire
and phantasia in the context of progressive motion. More directly, the
animals that have a capacity to engage in progressive motion are normally
presented with their object of desire “ rational or non-rational desires “ by
virtue of phantasia.20

DE CAELO
the limits of the science of nature in the
Both in the DC and in the PA Aristotle acknowledges the existence of
limits affecting the extent to which we can know the celestial world. The
study of the soul offered in the DA con¬rms the existence of such limits. If
I am right, these limits ultimately depend on the fact that the investi-
gation into nature is conducted on the assumption that there is lack of
uniformity in the natural world, that is, some important discontinuity. I
would now like to look at the way in which Aristotle proceeds in cases
where, as he himself admits, such limits are particularly acute. In DC ii 5
Aristotle turns to the daily rotation of the heaven of the ¬xed stars, the
¬rst heaven, and focuses upon its distinctive orientation. Aristotle has
already established that this rotation takes place from the right (285 b
19“20). He now engages himself in an attempt to provide an explanation

20 I add “normally” because from the DA 428 a 8“11 we learn that phantasia belongs to many but not
all non-stationary animals. Apparently, grubs (and perhaps ants and bees) are not capable of
phantasia. According to Aristotle, the non-stationary animals that partake in touch only and do
not have the ability for phantasia move indeterminately “ in Greek aoristos (433 a 4“5). I take it
¯
that this claim is equivalent to saying that indeterminate motion is not a case of navigation. For a
discussion of this passage and the relation between desire, phantasia, and progressive motion I
have pro¬ted greatly from reading Lorenz (2001).
Aristotle and the Science of Nature
98
for this particular orientation. Since nothing that concerns the eternal can
be a matter of chance or spontaneity, and the heaven of the ¬xed stars is
eternal, there must be a reason why this heaven moves in the direction it
does, rather than in the opposite direction (287 b 24“7). He goes on as
follows:

(1) Perhaps, then, the attempt to make some statement on certain things, indeed
on everything, passing nothing by, might well seem to be a mark of great simple-
mindedness or of much zeal. (2) Yet it is by no means right to censure all people
alike, but one ought to consider their reason in speaking, and the sort of
conviction involved in their account. (3) If someone hits upon more exact
necessities, then we should be grateful to the discoverers; (4) but as it is, we must
state what appears to be the case (DC 287 b 28 “ 288 a 2).

This passage is symptomatic of a position in which Aristotle at times ¬nds
himself in the study of the celestial world. This position may be described,
tentatively, as one in which Aristotle makes a judgment as to what is the
case, assumes that there has to be an explanation for what is the case, and
¬nally makes an attempt to provide this explanation, though it is clear
that he is not in a privileged position to provide an explanation. Interest-
ingly enough, Aristotle does not think that we have to give up any attempt
to say something about that which we cannot explain. All attempts to say
something are not to be censured alike. In particular, the reasons motiv-
ating a person to claim what he does, and the epistemic attitude involved
in his claim, are also to be taken into account “ clause (2). These remarks
should be understood, I think, in the light of what immediately follows,
and as a sort of justi¬cation for it. What immediately follows is not the
explanation of the particular orientation exhibited by the heaven of the
¬xed stars in its daily rotation. What follows is, as Aristotle himself puts it,
what appears to be the case, in Greek to phainomenon “ clause (4). Clearly
what follows in the text is not a genuine explanation, though it is the
result of a genuine effort to answer the question “why?” What appears to
be the explanation is not too far, I think, from capturing what Aristotle
has in mind. However dif¬cult it might be to understand what Aristotle
has in mind, it seems to me that two possible interpretations should be
rejected. First of all, what appears to be the explanation is not a provi-
sional account that will be, sooner or later, replaced by a genuine explan-
ation. Given the human limitations on knowledge about the celestial
world, nobody can be in the privileged position to provide a genuine
explanation; though somebody in the same situation as Aristotle™s may
perhaps give a better account. Aristotle is envisaging this possibility in
The limits of Aristotle™s science of nature 99
clause (3). Secondly, and more importantly, what appears to be the case is
not what looks plausible to Aristotle as opposed to what may look
plausible to other people. If that were the case, there would be some stress
on the fact that what appears to be the case is apparent because it may be
controversial. From the mere fact that Aristotle explicitly claims that he is
going to provide what appears to be the case, we are not entitled to come
to the conclusion that he is going to provide what appears to be the case to
him. On the contrary, Aristotle is making a genuine effort to supply an
account which is as objective as possible. In other words, Aristotle is
making an effort to arrive at an account which can satisfy us as intelligent,
rational beings who are approaching the study of the celestial world in the
right way.
What Aristotle says at the very beginning of the DC ii 12 is in line with
the interpretation so far offered.
(1) Since there are two dif¬culties about which one might reasonably be troubled,
we ought to make an attempt to say what appears to be the case, (2) considering
the eagerness to do so a mark of modesty rather than of excessive ambition, if,
out of thirst for philosophy, one is content with small solutions in things in
which we have the greatest dif¬culties (DC 291 b 24“8).

Aristotle provides these remarks as a general introduction to a discussion
of two dif¬culties, the ¬rst of which concerns the number of the motions
of the moon and the sun, and the reason why both celestial objects are
moved by fewer motions than some others above them (291 b 29“31), and
the second the relation between these motions and the daily rotation of
the heaven of the ¬xed stars, and the reason why many stars are carried on
by one single motion (292 a 10“14). Admittedly, to say that where the
dif¬culties are the greatest, eagerness is a mark of modesty rather than of
excessive ambition, provided that one is satis¬ed with small solutions, is
ambiguous “ clause (2). We might be tempted to think that where the
dif¬culties are the greatest, to be satis¬ed with the small dif¬culties one is
able to handle, and correspondingly with the small solutions one is able to
provide, is a mark of modesty rather than of excessive ambition. In the
light of what I have so far argued this is hardly plausible. If I am right, for
Aristotle there is nothing wrong in attempting to say something about any
dif¬culty, provided that one has the right epistemic attitude towards what
one is going to say. I propose, therefore, to read what Aristotle is saying as
follows: where the dif¬culties are the greatest, eagerness is a mark of
modesty rather than excessive ambition, provided that one is satis¬ed with
the solutions one is able to offer, even though one is aware that these
Aristotle and the Science of Nature
100
solutions are only small solutions. Admittedly, to say that something is a
small solution of a great dif¬culty is to say very little about the nature of
the solution and the reason for its being only a small solution. To shed
some light upon what Aristotle has in mind, we have to turn to the very
beginning of the passage. Here Aristotle is not saying that there are two
dif¬culties and we must make an attempt to solve them; rather, he is
saying that there are two dif¬culties and we must make an attempt to say
what appears to be the case, in Greek to phainomenon “ clause (1). What
appears to be the case in this context is what appears to be the solution to
a dif¬culty. Aristotle makes it clear that he is not in a privileged position
to provide the solution to these dif¬culties. There can be only one reason
for this. The dif¬culties in question are to be counted among the greatest
dif¬culties to which we are able to provide only small solutions. In the
light of this, we can appreciate, perhaps, what Aristotle has in mind when
he talks of small solutions. Something is a small solution neither because
it is a solution of a small dif¬culty nor because it is a partial solution of a
great dif¬culty. Rather, something is a small solution because the only
reason for Aristotle to accept it is that this is the best solution available to
him, given that he is persuaded that at times the limits on the extent to
which we can learn about, and understand, something are so acute that
our capacity to provide a solution to a certain dif¬culty concerning this
very thing is seriously affected.
A note of caution: we should not confuse the relevant use of to
phainomenon on which I have focused in these pages with the more
common use of ta phainomena. In the DC ta phainomena are the celestial
phenomena, that which can be observed in the sky from the earth (289 b
5, 297 a 4).21 It is signi¬cant, I think, that the relevant usage of to
phainomenon can be found outside the DC. In PA 1 Aristotle insists on
the structure and unity of the study of the natural world: having already
dealt with the celestial world, saying what appears to be the case to us
human beings “ in Greek to phainomenon hemin “ we have to move to the
¯
study of animal nature, trying as far as possible to omit nothing, however
noble and ignoble it may be (645 a 4“7).22 The use of the expression is to


21 This is in line with the traditional usage of ta phainomena in Greek astronomy. See, for instance,
Geminus, Elementa astronomiae 1. 19“22; Theon, Expositio 177. 9ff; Proclus, Hypotyposis 4. 7;
Simpl., In DC 488. 3“24. The slogans “saving the phenomena” and “saving the appearances” come
from this usage. On these slogans, see Lloyd (1991: 248“77) and Goldstein (1997: 2“12). The
celestial phenomena can, but need not, be observational data. See Owen (1986: 239“51).
22 I have discussed this passage in chapter 1, “The unity, structure, and boundaries of Aristotle™s
science of nature”.
The limits of Aristotle™s science of nature 101
be understood in the light of the informational gap that affects our study
of the celestial world and the dif¬culties of understanding the celestial
region on which Aristotle insists in PA 1. Even here Aristotle ¬nds a way
to remind us that in the study of the celestial world we should neither go
beyond what we can say nor stop making an effort to provide an account,
but state what appears to be the case to us, human beings with a limited
access to the celestial world.23

celestial matter
Aristotle is remarkably selective in his study of the celestial world. He
limits himself to a discussion of some topics, while he consistently evades
others. For example, he takes the view that celestial bodies are endowed
with the capacity for thought and desire, but he never engages, in the DC
or elsewhere, in an attempt to say how celestial souls think and desire.
This silence is open to different interpretations, but it is better under-
stood, I think, in the light of Aristotle™s belief in the existence of an
important discontinuity within the natural world. From the DA we learn
that Aristotle is reluctant to extend the results achieved in the study of
plants and animals to the imperishable creatures populating the celestial
world. Evidently, he is persuaded that the celestial souls work in a way
that is different from, and indeed not reducible to, the one described in
the DA, and consequently that the lack of information at his disposal
cannot be overcome by an appeal to what we know about the perishable
creatures.24 But when Aristotle ¬nally engages in a discussion of particular


23 PA 644 b 23“32, DC 287 b 28 “ 288 a 2, and DC 291 b 24“8 are discussed, together with other
texts, in Lloyd (1998a: 160“83). Lloyd takes these passages as evidence that Aristotle was not a
totally engaged researcher but conceived of himself as an amateur astronomer. I do not want to
deny this, but I think that Aristotle would have made these disclaimers even if he had had a better
command of astronomy. If I am right, these disclaimers depend upon a certain conception of man
and the place that he occupies in the natural world, rather than on the state of art of a certain
discipline or the competence that someone may have in it. I ¬nd it useful to compare the situation
in which Aristotle at times ¬nds himself in the study of the celestial world with the one that Plato
puts into the mouth of Timaeus:
Don™t therefore be surprised, Socrates, if on many matters concerning the gods and the whole
world of change we are unable in every respect and on every occasion to render consistent and
accurate account. You must be satis¬ed if your account is as likely as any, remembering that both
I and you who are sitting in judgment on it are merely human, and should not look for anything
more than a likely story in such matters (Tim. 29 c 4 “ d 3).
24 Since antiquity, commentators have often engaged in the exegetical exercise of ¬lling the gaps left
by Aristotle. I have argued that for Aristotle there is no need to postulate the existence of celestial
perception and celestial phantasia. However, Aristotle™s silence has encouraged a debate about the
Aristotle and the Science of Nature
102
features of the celestial world, he makes use of the explanatory resources
developed and re¬ned in the study of the sublunary world. In those cases,
and only in those cases, he evidently feels that the results achieved in the
study of the sublunary world can be safely extended to the celestial world.
It is signi¬cant, however, that even in those cases he makes a considerable
effort to square the case of the celestial bodies with those explanatory
resources, and this effort is not without consequences for the explanatory
resources themselves. Matter is a particularly interesting case. The notion
of matter is ¬rst introduced in the study of change as it is encountered in
the sublunary world, and then extended to the celestial world to account
for celestial motion. But how successful is this extension? I shall try to
answer this question by looking at the way matter is introduced and
discussed in Lambda 2.25
Though Lambda 1 announces the existence of an important discontinu-
ity within the sensible world,26 Lambda 2 begins by pointing to the
feature that eternal and perishable sensible substances have in common:
these substances, in so far as they are sensible, are all subject to change
(1069 b 3). Aristotle™s intention is transparent and consists in identifying
the principles of the sensible substances with the principles of change. In
this context, the reader is reminded that change takes place between
contraries (1069 b 4“5), and that it involves the existence of something
that admits one of two contraries and, under the appropriate circum-
stances, can become the other (1069 b 6“7). This third principle is
identi¬ed with matter (1069 b 8“9). It is not dif¬cult to see why the
notion of contrariety is placed at the center of Aristotle™s investigation of
change. Change is the emergence of a new state to the exclusion of a
previous one. Moreover, this new state is never extrinsic to the change in
question. Recovering from an illness is a standard example. Relying
on ancient medicine, Aristotle thinks of health as an equilibrium of
certain bodily factors (hot/cold, wet/dry), and illness occurs when this


existence of celestial perception and celestial phantasia. There are traces of this debate in
Philoponus, In DA 595.37 “ 598.24, and [Simplicius], In DA 320.17 “ 321.2. Plutarch of Athens
“ who reestablished the Academy at Athens in the late fourth century ce and was the teacher of
Syrianus, who in turn was the teacher of Proclus “ argued for the existence of some form of
celestial perception and celestial phantasia. It is not clear whether Plutarch was committed to the
existence of non-rational celestial desire too. On the other hand, Alexander of Aphrodisias and his
followers defended the orthodox view that celestial natures do not need the capacity for perception
and the ability to form impressions or phantasia.
25 For a detailed study of Lambda 2, see Charles (2000: 81“110). I have been in¬‚uenced more
strongly than the footnotes can indicate by this excellent study.
26 See my discussion of Lambda 1 in this chapter pp. 88“9.
The limits of Aristotle™s science of nature 103
equilibrium is disrupted.27 Recovering from an illness may or may not be
a natural process, but it always takes place between illness and health. In
particular, someone cannot recover inde¬nitely from a certain illness, but
it is in the very nature of the process to terminate at some point; that is,
when the original equilibrium of the bodily factors is restored. Contrar-
iety provides Aristotle with the conceptual resources to express this
fundamental truth as well as to insist on the fact that health and illness
exclude each other and do so in such a way as to point at one another.28
The use of the language of contrariety was not exclusive to Aristotle.
Quite the contrary. Consider, for example, the ¬rst argument for the
immortality of the soul that is advanced in the Phaedo. This is what
Socrates says to Cebes:
<Socrates> Do not con¬ne yourself to human beings, if you want to understand
this more readily, but take all animals and all plants into account, and, in short,
for all things that come to be, let us see whether they come to be in this way, that
is, from their contraries if they have such, as the beautiful is contrary to the ugly
and the just to the unjust, and a thousand other things of the kind. Let us see
whether those that have a contrary must necessarily come to be from their
contrary and from nowhere else, such as, for example, when something comes to
be larger it must necessarily become larger from having been smaller before.
<Cebes > Yes. <Socrates> Then, if something smaller comes to be, it will come
from something larger before, which became smaller? <Cebes> That is so.
<Socrates> And the weaker comes to be from the stronger, and the swifter from
the slower? <Cebes> Certainly. <Socrates> Further, if something worse comes
to be, does it not come from the better, and the juster from the more unjust?
<Cebes> Of course. <Socrates> So we have suf¬ciently established that all
things come to be in this way, contraries from contraries? <Cebes> Certainly
(Phaedo 70 d 7 “ 71 a 10, translation Grube with alterations).
In this passage Socrates is making a general point about what comes to
be (according to an alternative translation, closer to the text, about what is
becoming). It is signi¬cant that he does not distinguish the case of
something which comes to be something or other “ which acquires a
property “ from the case of something which comes to be tout court “

27 The view that health consists in the right balance of the relevant bodily factors was common
ground in ancient medicine. See, for instance, the author of Ancient Medicine, xiv, 4.16, and the
author of The Nature of Man, iv, 172.15 “ 174.4. Outside of the Hippocratic corpus, see the
imaginative language used by Alcmeon of Croton: health is the isonomia or equilibrium of the dry
and the wet, the hot and the cold, and the like; whereas illness is the monarchia or domination of
one of these bodily factors alone (DK 24 b 4). See also the Timaeus, 82 a“b, for the same idea.
28 For language of contrariety in ancient medicine, see also the author of Breaths and his claim that
“contraries are cures for contraries” (i, 4.10“11). For the importance of the language of contrariety
in the Hippocratic tradition, see Jouanna (1988: 31“4).
Aristotle and the Science of Nature
104
which comes into existence. His point is about coming to be (alterna-
tively, about becoming). It was not unusual in the pre-Platonic investi-
gation of nature to blur this distinction and talk of becoming in such a
way as to make it equivalent to coming into existence.29 Socrates, or
rather Plato, is simply content to register this widespread tradition.
Aristotle acknowledges his dependence on this tradition. At the beginning
of the Physics, Aristotle presents himself as continuing, and indeed com-
pleting, the work of his predecessors. But his doctrine of change is a
substantial revision, rather than a mere re¬nement, of the previous
re¬‚ection on change. First of all, Aristotle makes it clear that, strictly
speaking, there is no such thing as coming to be as such. Coming to be is
said in many ways. More precisely, coming to be something “ coming to
acquire a certain quantity, size or place “ and coming into existence are
distinct processes. Secondly, and more importantly, Aristotle, unlike his
predecessors, does not limit himself to making use of the language of
contrariety. He attempts a comprehensive analysis of this notion and
works out a theory of contrariety.30 To put it in another way, when
Aristotle says that “everything that comes to be comes to be from, and
everything that passes away passes away into, its contraries or something
in between” (Phys. 188 a 22“4), he is making a substantive claim about all
forms of change, and this claim is ultimately supported by a general
theory of contrariety. A recovery of this theory goes beyond the scope
of the present discussion. I am content to say that this theory must supply
Aristotle, minimally, with:
1. a de¬nition of contrariety;31
2. a classi¬cation of the different types of contrariety (e.g. contraries that
admit intermediates and contraries that do not);32
3. a rational way to move from the plurality of and variety of contraries
to a primary contrariety.33

29 On this point see also chapter 1, “The unity, structure, and boundaries of Aristotle™s science of
nature”.
30 For an attempt to recover Aristotle™s theory of contrariety see Anton (1957) and more recently
Bogen (1992: 1“21).
31 For Aristotle, contrariety or enantiosis is “the greatest difference” (Metaph. 1055 a 4“5), and
contraries or enantia are “the things that differ most in the same genus” (Cat. 6 a 17“18, Metaph.
1018 a 27“8 and 1055 a 27“8).
32 Many topoi involving contraries are collected in the Topics. These topoi document the existence of
a classi¬cation of contraries. On contraries that admit contraries and contraries that do not, see in
particular Top. 123 b 1“37.
33 In the Physics, Aristotle insists on the reduction of the contraries to the primary contrariety.
However, he does not provide the details of this reduction. Nor does he provide them elsewhere.
Was this reduction offered in one of his lost books on the contraries? Perhaps so. In the
The limits of Aristotle™s science of nature 105
This theory enables Aristotle to reconsider, critically, the views of his
predecessors and detect their common failure. For example, from the ¬rst
book of the Physics we learn that they all failed to ¬nd a rational way to
reduce “ in Greek anagein (189 b 27) “ the plurality and variety of
contraries to the two primary contraries. This criticism is to be under-
stood in the light of point (3) above. In the natural world we are
confronted with fundamentally different contraries. These contraries are
fundamentally different in the sense that they cannot be explained away
or eliminated, though they can be understood in the light of some
conceptual apparatus whose generality enables us to grasp what they all
have in common.34 Aristotle™s predecessors failed to work out the concep-
tual apparatus that is needed for an adequate analysis of the fundamen-
tally different contraries. More explicitly, they did not possess a theory of
contrariety to deal successfully with the complexity of the natural world.
According to Aristotle, the way they selected the primary contrariety was
random: some of them identi¬ed the primary contraries with hot and
cold or wet and dry, the Pythagoreans with odd and even, and Empedo-
cles with love and strife. Against all of them, Aristotle argues that form
and deprivation constitute the primary contrariety.
Let us return, also in the light of these remarks, to Lambda 2. Here
Aristotle makes use of the language of contrariety in the attempt to
provide a general description of change which applies to all natural
processes. It is easy to see that the language of contrariety is not neutral
with respect to his speci¬c theory of contrariety. Matter is de¬ned as “that
which has the capacity for both <contraries>” (1069 b 14“15). Form and
deprivation are identi¬ed as the contraries in question (1069 b 33“4). But
Aristotle is not content to appeal to the language and theory of contrar-
iety. Interestingly enough, he makes an effort to extend the language, and
indeed the theory, of contrariety to the celestial world. Since the celestial
bodies are moved in a circle, they must possess an appropriate type of
matter: that is, a matter endowed with the capacity for this particular type

Metaphysics, Aristotle mentions a Selection of Contraries (1004 a 2) and a Division of Contraries
(1054 a 30). In his commentary on the Categories, Simplicius several times refers to a book On
the Opposites. Following Rose (1886), Ross collected all the extant testimonies under the title On
the Contraries (Ross, 1955b). He himself decided on this title presumably on the basis of the
catalogue of Aristotle™s books preserved by Diogenes Laertius. This catalogue lists the title On the
Contraries (Diog. Laert., v 21 (30)). But the situation may be decidedly more complicated. On this
point, I refer the reader to Guariglia (1978).
34 In other words, Aristotle™s reduction is not a case of elimination of the complexity of the natural
world but rather an attempt to enrich our understanding of this complexity. For a convenient
introduction to Aristotle™s notion of reduction, see Byrne (1997: 23“5).
Aristotle and the Science of Nature
106
of motion. In Theta Aristotle refers to this matter as “topical matter,” in
Greek topike hyle (1042 b 5“6). But in Lambda 2 he is more speci¬c and
¯¯
refers to this matter as pothen poi matter; that is, “matter for motion from
one place to another” (1069 b 26). By referring to celestial matter as pothen
poi matter, Aristotle is not only claiming that a certain orientation is
intrinsic to celestial motion, and that celestial matter is that which has the
capacity for that particular motion rather than for the motion that takes
place in the other direction; he is also making an effort to square celestial
matter with the language and theory of contrariety. Admittedly, he does
not say that celestial matter is that which has the capacity for both
contraries. Rather, he says that celestial matter is that which has the
capacity to move from one place to the other. But Aristotle cannot go
any further than this. He cannot characterize celestial motion as a motion
between contraries. In the DC, Aristotle provides arguments for the thesis
that there is no motion contrary to circular motion. He does so by
showing that circular motion cannot be integrated into a system of
contrary motions. More directly, contrary motion occurs between con-
trary places. But there is no motion contrary to circular motion. The
language of contrariety simply does not apply to circular motion. This
motion does not take place between contraries but, as Aristotle himself
says, from the same place to the same place:
that motion [¼ circular motion] is from the same to the same, whereas contrary
motion was de¬ned as motion from one contrary to the other contrary (DC 271 a
20“2).35



35 We must not underestimate the importance of this thesis. That there is no motion contrary to
circular motion is crucial for Aristotle™s view that the celestial body which naturally performs
circular motion is not subject to generation, perishing, growth, and alteration. Consider the
following passage from the DC:
Similarly it is also reasonable to assume of it [¼ the celestial body which naturally performs circular
motion] that it is not subject to generation, perishing, growth, and alteration; on the ground that
everything that comes to be comes to be from a contrary and from some substrate, and is destroyed
by a contrary and into a contrary, given that there is some substrate (270 a 12“15).
In the DC, chapter 4, Aristotle offers a battery of arguments for the view that circular motion has
no contrary. If there is a motion contrary to circular motion, this is either a rectilinear or a circular
motion. Let us suppose that a rectilinear motion is contrary to circular motion (CM). Since
rectilinear motion is either upward (UpM) or downward motion (DnM), then either Contr
(UpM, CM) or Contr (DnM, CM). But UpM and DnM are contraries to one another; that is,
Contr (UpM, DnM). Therefore UpM and DnM would be contraries to one another and to CM.
But Aristotle is committed to the principle that for one thing there can be one contrary at most “
in Greek hen heni enantion (for this principle I refer the reader to chapter 3, “Motions,” pp. 60“2).
Moreover, no circular motion can be contrary to circular motion. In the DC, Aristotle argues that
the motions which take place in opposite directions along the same circle are not contraries to one
The limits of Aristotle™s science of nature 107
To put it in another way, Lambda 2 contains an attempt to provide a
uni¬ed account of change that crucially depends on the applicability of the
language and theory of contrariety to the case of celestial motion. But this
language and this theory, originally designed to account for sublunary
change, cannot be easily extended to the case of celestial motion. Doubts
about the applicability of the language and theory of contrariety to
the celestial world were already being raised in antiquity. Alexander of
Aphrodisias was one of the few people who accepted, and indeed defended,
the thesis of the existence of a celestial simple body that naturally performs
circular motion.36 But even Alexander was reluctant to admit the existence
of celestial matter. Here is the evidence in our possession:
1. In his commentary on the Metaphysics Alexander insistently repeats
that the substratum of the heavens is body, not matter (e.g. In Metaph.
22. 2“3; 169. 18“19; 375. 37 “ 376. 1)
2. At times Alexander contrasts the divine body with the material things
of the sublunary world.37 This contrast is to be understood, I think, in
the light of the dif¬culties that arise once the concept of matter is
applied to the celestial world.
3. Quaestio i 10 documents the existence of a debate internal to the school
of Alexander on the applicability of the concept of matter to the
celestial world. It sheds some light on the reasons that might have led
Alexander and his school to contrast the case of the divine body to the
sublunary material things. Some dif¬culties are here connected to the
fact that matter is that which is receptive of both contraries in turn (20.
31“2). This is a reference to Lambda 2. There is also no doubt that the
author of this quaestio is not fully persuaded by Aristotle™s attempt to
transfer the concept of matter as it is developed and re¬ned in the
study of the sublunary change to the case of celestial motion. He even
proposes that this concept of matter be replaced by recourse to the

another. In antiquity Aristotle™s arguments for this view were subjected to close scrutiny and
criticism by Philoponus. See Simplicius, In DC 156.25 “ 201.10 (¼ Philoponus, Contra Aristotelem,
frr. 81“107). A study of this aspect of the ancient debate on the DC is not needed for the present
discussion. Quite independently of the soundness of the arguments that Aristotle offers in support
of the thesis that there is no motion contrary to circular motion, the tensions and problems that
this very thesis causes for the doctrine of celestial matter should have made Aristotle reconsider,
and eventually soften, the original thesis. But there is no evidence that he ever reconsidered, or
even rejected, this thesis.
36 From Simplicius™ commentary on the DC we learn that Alexander, in his (lost) commentary on
the DC, devoted much time and effort to defending this thesis against the objections moved by
Xenarchus.
´ ´
37 See also Bodnar (1997a: 190“205, in particular footnote 3). In this context, Bodnar stresses the
importance of De mixtione 229. 6“9.
Aristotle and the Science of Nature
108
characterization of matter as the ultimate shapeless substrate. This
alternative characterization of matter is not found in the Metaphysics or
elsewhere, but it is a creative interpretation of what Aristotle himself
says in Phys. 192 a 21 and 193 a 11.
A ¬nal clari¬cation is needed. By extending the language and theory of
contrariety to the celestial world Aristotle makes a signi¬cant effort to
provide a uni¬ed account of matter which applies to the variety and
complexity of the natural processes. But this effort does not commit
Aristotle to the existence of one uni¬ed matter which is the substrate of
all the natural processes.38 On the contrary, in Lambda Aristotle explicitly
says that
everything that is subject to change has matter, but <it has> different <matter>:
of the eternal substances those that are not subject to generation but to motion in
place <have> not genetic [geneten] but rather pothen poi <matter> (1069 b 24“6).
¯¯
In this compressed text Aristotle is doing several things at once:
1. He credits the eternal sensible substances, that is to say the celestial
bodies, with pothen poi matter, whereas he denies them a share in
genetic matter, since they are not subject to generation (and perishing).
2. He acknowledges that genetic matter is the primary matter of
perishable sensible substances. In other words, if plants and animals
possess matter for motion in place, they must possess genetic matter
too.
3. He proves that pothen poi matter is a special case of matter and cannot
be reduced to matter for motion in place as it is encountered in the
sublunary world (from (1) and (2)).
One might accept that Aristotle in this passage introduces pothen poi
matter as a distinct case of matter alongside genetic matter. Yet one might
insist that this does not prevent Aristotle from postulating the existence of
some other matter shared by both eternal and perishable substances. This
other matter would be located at a more basic level, and it would be prior
to the distinction between genetic matter (which applies to perishable
sensible substances) and pothen poi matter (which applies to eternal sensible
substances). Richard Sorabji has recently argued that, when Aristotle says
that celestial bodies have different matter, he need not mean that their
primary matter is different, but only that their elemental matter is

38 On this point see Charles (2000: 81“110). As Charles remarks, the demand that there be one
uni¬ed type of thing is not to be confused with the demand that there be one uni¬ed thing (97).
The limits of Aristotle™s science of nature 109
different. The difference would ultimately be due to the fact that celestial
bodies are made of an element which has the capacity for motion in place
only, whereas the other four elements can undergo other kinds of change,
including generation and perishing.39 Admittedly, Aristotle does not
exclude, expressis verbis, the existence of some primary matter shared by
both eternal and perishable substances. But his silence is to be understood
as denying that there is such a further level of analysis. The logic of the
argument offered in Lambda 2 strongly suggests that pothen poi and
genetic matter represent the very last level of analysis, and that there is
no further level of analysis on which the unity of matter is secured. This is
a consequence of the identi¬cation of matter with that which has a
capacity for change. Since the only change that Aristotle admits in the
celestial world is circular motion, it is dif¬cult to see how the analysis of
matter could be carried out in order to reach some more basic matter than
celestial and genetic matter.40

lack of intelligibility to us
By this point I hope to have established that for Aristotle there is a region
in the province of the science of nature where we are confronted with
dif¬culties exceeding our capacity to provide a solution to them. It should
also be clear that the idiosyncratic conception of the natural world held by
Aristotle is the ultimate reason for this lack of intelligibility to us of
important features of the celestial world.41 However, one may still argue
that the dif¬culties in question, given that they affect only a limited region

39 Sorabji (1988: 15).
40 Sorabji is led to this conclusion on the basis of a certain reading of Zeta 3. Though it is not my
intention to engage in a discussion of this notoriously dif¬cult chapter, I would like to point out
that no link is here established between matter and change. Matter is presented as that which
remains when all the other things are stripped away. Apparently, the stripping procedure takes
place in two stages. Stage 1: affections and doings and capacities are removed (1029 a 12“16). Stage
2: length, breadth, and depth are subtracted (1029 a 16“19). Aristotle concludes by saying that
matter alone necessarily seems to be substance “to those considering <the issue> in this way”
(1029 a 18“19). These words might be taken as an indication that Aristotle does not ¬nd this
reductio argument particularly attractive. It has also been suggested that this argument is based on
assumptions that Aristotle does not hold. On this point see Charlton (1970: 136“8); Scho¬eld
(1972: 97“101); Frede and Patzig (1988, vol. ii: 44“5); Gill (1989: 26“31). But even if Aristotle
endorses this argument, it is far from clear that the matter left over once the stripping procedure
has taken place provides us with a solution to the problem of the unity of matter. The objects of
the stripping procedure are presumably ordinary objects such as a statue. What is left over once
all the other things have been subtracted is not some primary matter but rather the matter of the
statue. Primary matter is simply not at issue in Zeta 3.
41 I add “to us” because I want to provide Aristotle with the resources to claim that the natural,
though it is not completely intelligible to us, is intrinsically intelligible.
Aristotle and the Science of Nature
110
in the province of natural science, should not force us to modify the
of¬cial image of Aristotle™s science of nature as it emerges from other
passages. There is more than a grain of truth in this argument. Yet one has
to realize the importance of the study of the celestial bodies to Aristotle.
In the exhortation concluding PA 1, Aristotle claims that the celestial
bodies are the most honorable and divine among the substances we
experience in the natural world. This statement is not to be underesti-
mated. Aristotle thinks of the study of the celestial world as the culmin-
ation of natural investigation. This is also the reason why at times he
offers this very study as an illustration of what a theoretical life, dedicated
to theoretical activities, and pursued for its own sake, should be like. One
should also bear in mind that Aristotle admits causal, and therefore
explanatory, unity in the natural world. Aristotle does not think of the
natural world as a mere collection of separate, or only loosely connected,
parts, but as a uni¬ed whole articulated into causally interrelated parts.42
But if Aristotle takes the view that the different parts of the natural world
are causally related to one another, how is it possible for him to con¬ne
the lack of intelligibility to the celestial world? This lack of intelligibility is
transmitted to the sublunary world by virtue of the fact that certain
features of the celestial world play a role in the explanation of features
of the sublunary world. Some distinction needs to be introduced here. If
my reconstruction of Aristotle™s conception of the natural world is correct,
there is no doubt that explanatory factors of important features of the
sublunary world are not completely accessible to us, and consequently an
explanation of these features is, strictly speaking, not available to us.
Nevertheless, the temptation to con¬‚ate the lack of intelligibility experi-
enced in the study of the celestial world and the consequences that derive
from this lack of intelligibility in the study of sublunary world into one
and the same case should be resisted. Let us return to the slogan that it
takes a man to generate a man.43 On the one hand, Aristotle is able to
provide the explanation for the generation of a particular man. He is able
to point out the factors involved in the explanation of that generation: the
father, the sperma, the katamenia, the goal of that particular generation “
that is, a particular form of organization realized in a body of a certain
type. On the other hand, Aristotle is not able to explain why the gener-
ation of that particular man is part of a continuous process of coming into
existence and going out of existence. In order to be able to provide an

42 See chapter 1, “The unity, structure, and boundaries of Aristotle™s science of nature.”
43 See ibid., pp. 9“13.
The limits of Aristotle™s science of nature 111
explanation for the continuity of this process Aristotle would have to be in
a position to say why the celestial revolutions take place in the precise way
they do. But because of the very way in which Aristotle conceives of
celestial motion he is not in possession of the relevant information to
explain why celestial revolutions take place in the precise way they do. I
am ¬nally in a position to specify the claim that some grasp of the celestial
world is not only necessary but also preliminary to the attainment of an
understanding of important features of the sublunary world. I insist on
the quali¬cation “some grasp” because:
1. only some features of the celestial world are relevant for the study of
the sublunary world; and
2. access to these very features is limited to us by the physical distance
which separates us from the celestial world together with the fact that
these bodies cannot be adequately explained on the basis of
the conceptual resources developed and re¬ned in the study of the
sublunary world.
Aristotle™s view that there is a lack of intelligibility to us of important
features of the celestial world is to be understood and appreciated in its
historical context by relation to his predecessors. Concern, if not fascin-
ation, with ta meteora was a conspicuous feature of Anaxagoras™
¯
44
thought. In antiquity, Anaxagoras was often remembered for his auda-
cious views in this ¬eld. For him, the sun and the rest of the celestial
bodies were stony and heavy bodies ignited by the resistance and tearing
of the surrounding aither.45 Though Anaxagoras could not be entirely
¯
responsible for the widespread hostility to meteorological speculations in
the ¬fth and fourth century bce, the audacity of his views surely contrib-
uted to a debate on the extent of what could be investigated and what
could be known about ta meteora. Consider, for example, the way
¯
Xenophon™s Socrates reacts to Anaxagoras™ view that the sun is a ¬ery
stone (Mem. iv 7. 6“7). The problem with Anaxagoras is not simply that
he fails to recognize that there are human limitations to the extent of what
can be known about ta meteora. These limitations are set by god (or gods),
¯
and Anaxagoras™ attempt to go beyond human limitations results in
madness.46 In other words, Anaxagoras is not only seriously mistaken;

44 For the use of the phrase ta meteora to refer to the study of the heavens see ibid., footnote 5.
¯
45 Plutarch, Lysander 12 (¼ DK 59 a 12). This testimony is con¬rmed by Diog. Laert., ii 12 (¼ DK 59
a 1.12). Cf. Stob., Ecl. i 23. 10“15 (¼ Aetius ii 13.3 ¼ DK 59 a 71).
¨
46 For the language of madness in connection with natural philosophy, see Xenophon, Mem. i 1.1
13“14.
Aristotle and the Science of Nature
112
he has gone mad. His madness is the divine punishment for his failure to
recognize and accept the limitations set by god (or gods) to our capacity
to gain knowledge of the world. This is not the place to discuss how
seriously to take this testimony as evidence about Socrates™ attitude to the
study of nature. Much of this may well be idiosyncratic to Xenophon™s
Socrates. But it is revealing of a typical skepticism about the extent of
what we can know of the celestial phenomena, which is well documented
in the ¬fth and fourth century bce. This skepticism seems to be rooted in
the conviction that there are limits to our ability to acquire knowledge of
the world, and the study of ta meteora often results in an attempt to go
¯
beyond these limits.
Both Aristotle and Anaxagoras are engaged in vast and ambitious
projects of investigation of the natural world, on the crucial assumption
that this world is intelligible. They share the same desire to understand
every single aspect of the natural world. They also share the same concern,
if not fascination, with the celestial world. Unlike Anaxagoras, however,
Aristotle recognizes the existence of a gap between what is intrinsically
intelligible and what can be known by us. To the best of my knowledge,
there is no evidence that Anaxagoras is willing to accept this distinction.
Note also that Anaxagoras™ meteorological speculations were often equated
with atheism in the ¬fth and fourth century bce. His study of the celestial
phenomena resulted in a rejection of the common supposition that the
celestial bodies are alive and enjoy the blissful life of gods. By contending
that the sun and the rest of the celestial bodies are heavy and stony objects
Anaxagoras denied not only life but also divinity to the celestial bodies.47
By contrast, Aristotle can accommodate the popular supposition that the
celestial bodies are alive. His considered views about the celestial world do
result in an extensive revision of popular piety, but they do not result in a
rejection of the pre-philosophical (or pre-theoretical) intuitions about the
gods. More speci¬cally, Aristotle™s investigation of the celestial world ends
up in a radical revision of the received views about the heavens; but this
revision also ¬ts certain pre-theoretical conceptions of the heavens as the
seat of the gods.48


47 Anaxagoras was put on trial on a charge of impiety. See Diog. Laert., ii. 12 (¼ DK 59 a 1.12).
48 The rationalization of the religious tradition and its consequent appropriation is not a distinctive
feature of Aristotle™s philosophy. The Stoics shared Aristotle™s view that the best considered view
of the philosophers should meet certain pre-theoretical intuitions about the gods. The Stoic
rationalization of Greek and Roman religious tradition is to be understood in the light of this
idea.
Epilogue




There would be no problems for Aristotle,
if one accepted his assumptions regarding the ¬fth body (Plotinus,
ii 1. 2. 12“13).

aristotle™s language
In antiquity it was common to refer to the celestial simple body as the
¬fth body,1 the ¬fth substance,2 the ¬fth element,3 the ¬fth nature,4 or
even the ¬fth genus.5 This language strongly suggests that there was a

1 Together with the passage in the epigraph, see Stob., Ecl. i 37. 16“19, and [Plutarch], Placita 881 e
¨
10 - f 7 (¼ Aetius i 7. 32): “an aetherial body called by Aristotle ¬fth body.” Stob., Ecl. i 128. 4“9,
¨
and [Plutarch], Placita 878 b 8“9 (¼ Aetius i 3. 22): “some ¬fth aetherial body, not liable to
change”; [Galen], Hist. philos. 54 (¼ Dox. gr. 623. 21“2). Sextus Emp., M X 316: “along with the
four elements they [¼ Aristotle and Ocellus Lucanus] adopted the ¬fth body which revolves in a
circle.” The presence of Ocellus along with Aristotle calls for an explanation. In the Pythagorean
tradition it was customary to assign the Platonic speculations about the ¬fth regular polyhedra to
Pythagoras (and the Pythagoreans, including Philolaus). In all probability, Sextus (or his source)
con¬‚ated the Platonic/Pythagorean and the Aristotelian ¬fth body into one and the same thing. I
have argued that the Platonic/Pythagorean ¬fth body is the result of an attempt to give a place to
the ¬fth regular polyhedron, the dodecahedron, and as such it has nothing to do with the never
stationary celestial simple body which is forever revolving in a circle. See chapter 3, “Motions,” in
particular pp. 77“83. The expression “¬fth body” is frequently used by the commentators. See, for
example, Olympiodorus, In Meteora 2. 19; Philoponus, In Meteora 3. 37; 14. 32, 37.
2 From Simplicius we learn that Xenarchus™ book was entitled Against the Fifth Substance. Cf.
Simpl., In DC 13.22; 20.12; 21.33. Stob. Ecl. i 212. 25“7 (¼ Arius Didymus fr. 10 partim ¼ Dox. gr.
451. 7“8): “the heaven and the things that populate it are constituted of a ¬fth substance, different
from ¬re.” See also Hippolytus, Refutatio omnium haeresium vii 19. 3“4 : “there is a certain ¬fth
substance, according to Aristotle, as it were a certain super-mundane substance. And this has
become a necessity in his system, in order to account for the division of the world.”
3 Diog. Laert., v 32: “<According to Aristotle,> besides the four elements, there is also a ¬fth
<element>, of which the aetherial bodies are constituted.” See also Arnobius, Adversus nationes ii
9; and [Clement] Recognitiones viii 15. Like “¬fth body,” the expression “¬fth element” is
routinely used by the commentators. See, for example, Olympiodorus, In Meteora 2. 14“16; 2. 29“
30; Philoponus, In Meteora 5. 9“10; 5. 13; 9. 31; 9. 33; 31. 29, 58. 8“9.
4 Cicero often refers to the celestial simple body as quinta natura. See Tusc. i 10, 22; Tusc. i 17;
Tusc. i 41; De ¬n. iv 5, 12.
5 Cicero, Acad. i 7, 26.

113
Aristotle and the Science of Nature
114
tendency to think of the celestial simple body as an additional body. In all
probability, from very early on, the view that earth, water, air, and ¬re
were the simple bodies out of which the other bodies are made was largely
accepted and relatively uncontroversial. Disagreement was con¬ned to the
existence of a celestial simple body. This body was perceived as an
innovation whose need was not transparent to everyone and called for
an explanation. In other words, from very early on, the scope of the
debate was narrowed down to, and focused on, the need for another body
alongside earth, water, air, and ¬re. At the beginning of the DC, however,
it is an entirely open question how many simple bodies or elements there
are. It is only in the course of the argument that Aristotle comes to the
conclusion that there is a celestial simple body which is naturally moving
in a circle along with four sublunary simple bodies which naturally
perform rectilinear motion. In the DC, Aristotle ¬rst argues for the
existence of a celestial simple body that naturally performs circular
motion, and then establishes the existence of four sublunary simple bodies
or elements. But the existence of four sublunary simple bodies or elements
is not taken for granted. Aristotle™s arguments for the existence of four
sublunary simple bodies are only postponed until the existence of a
celestial simple body is secured and the discussion of the celestial world
is concluded. Obviously it is not the existence of earth, water, air, and ¬re
that requires an argument, but rather the existence of earth, water, air,
and ¬re as elements (or simple bodies): that is to say, their existence as
the ultimate principles of the natural bodies that we encounter in the
sublunary world.
Though the expressions “¬fth element,” “¬fth body,” “¬fth substance,”
and “¬fth nature” are abundantly attested in the ancient testimonies
concerning the celestial simple body,6 there is no evidence that Aristotle
coined or used any of them. To begin with, Aristotle never uses these
expressions in the extant works. To claim that he did use these expres-
sions, but only in his lost works, is to make a bold assumption for which
there is no good evidence.7 Secondly, and more importantly, Aristotle

6 Alexander of Aphrodisias is the exception to the rule. Alexander tends to avoid these expressions.
Most of the time, Alexander refers to the celestial simple body as “the divine body” (DA 43.9 and
45.3; Quaestiones i 1, 4.2; i 10, 21.7; ii 3, 47.30; iii 12, 106.271) or “the body which moves in a
circle” (Quaestiones i 1, 3.9, 3.14“15, 4.1; i 10, 20.10; i 25, 40.10).
7 Typically, this assumption is itself built of other assumptions: for example, that until Andronicus
published his “edition” the lost works of Aristotle remained the chief source of information about
Aristotle™s thought. For a discussion of Aristotle™s language (as opposed to the language we ¬nd in
the doxography), see Moraux, (1964: 1171“3 and 1209“26); Easterling (1964: 73“85); and, more
recently, Hahm (1982: 60“74).
Epilogue 115
developed his own language to refer to the celestial simple body. In the
works that have come down to us, Aristotle consistently refers to this body
as “the ¬rst element,” to proton stoicheion (or “the ¬rst of the elements,” to
¯
8
proton ton stoicheion), “the ¬rst body,” to proton soma (or “the ¬rst of the
¯ ¯ ¯ ¯ ¯
9
bodies,” to proton ton somaton), or ¬nally “the ¬rst substance,” he prote
¯ ¯¯ ¯ ¯¯
10
ousia. It is routinely suggested that this language implies a value judg-
ment about the celestial simple body. I do not deny that Aristotle
conceives of the simple celestial body as a noble and divine body, and
as such prior to any of the sublunary bodies.11 But I do not think that this
is the primary and most important message that these expressions are
intended to convey. The study of Aristotle™s conception of the natural
world that I have conducted in chapter 1 suggests that there is more to this
language than a value judgment about the superiority of the celestial
simple body over the four simple bodies of the sublunary world. In all
probability, the celestial simple body is the ¬rst element (or the ¬rst body,
or the ¬rst substance) because it comes ¬rst in the order of explanation.
Moreover, this element (this body, this substance) comes ¬rst in the order
of explanation because Aristotle thinks of the natural world as a very
special causal system in which the direction of the explanation is from the
celestial to the sublunary world only.12
In antiquity, it was also common to refer to the celestial simple body as
aither or as an aethereal body.13 Interestingly enough, Aristotle systematic-
¯
ally refrains from using the term aither. In the extant works, Aristotle does
¯


8 DC 298 b 6; Meteor. 338 b 2, 339 b 16“17, 340 b 11.
9 DC 270 b 3, 22; and DC 291 b 32.
10 DC 270 b 11. In Metaph. Zeta, we are told that substance, ousia, is thought to belong most
obviously to bodies (1028 b 8“9). The celestial bodies are substances, ousiai, and are made of a
body which is itself a substance, an ousia. The case of the celestial simple body is not different
from that of the four sublunary simple bodies. They too are called substances, ousiai. See chapter
2, “Bodies.”
11 DC 269 b 13“16.
12 For a discussion of this claim, I refer the reader to chapter 1, “The unity, structure, and boundaries
of Aristotle™s science of nature.”
13 On the celestial simple body as aither, see Stob., Ecl. i 196. 11“12 (¼ Arius Didymus fr. 9 ¼ Dox.
¯
gr. 450. 13). Aither is often identi¬ed with the ¬fth body. Achilles, Isagoge: “a ¬fth body which is
¯
imperishable and pure, and which people call aither.” On the ¬fth body as an aethereal body, see
¯
¨
also Stob., Ecl. i 37. 16“18, and [Plutarch], Placita 881 e 10 “ f 7 (¼ Aetius i 7. 32). Stob., Ecl. i 128.
4, and [Plutarch], Placita 878 b 8“9 (¼ Aetius i 3. 22). These testimonies are reported in footnote
¨
1. Perhaps the most interesting piece of doxography comes from Stob., Ecl. i 195. 20 “ 196. 2, and
¨
[Plutarch], Placita 887 d 7“11 (¼ Aetius ii 7. 5): “Aristotle: ¬rst unaffectable aither, some ¬fth
¯
body; next to it affectable ¬re, air, water, and lastly earth.” At De mixtione 223. 10“11, Alexander of
Aphrodisias refers to the celestial simple body as “the divine, aethereal body which moves in a
circle.”
Aristotle and the Science of Nature
116
not show the least inclination to call his celestial simple body aither. In the
¯
DC Aristotle refers to aither, but only as the traditional name for the
¯
upper part of the world (270 b 20“1). In other words, there is no
appropriation of the name aither, in the DC or elsewhere. Why? In all
¯
probability, Aristotle felt that the use of this name would have been
misleading at best. The early history of aither has been recently told by
¯
Peter Kingsley, to whom I refer for a very helpful discussion.14 Here
suf¬ce it to say that in the earliest Greek poetic tradition the word aither ¯
was used to refer to the clear, bright, upper air, and as such aither was
¯
15
contrasted with aer, the often misty air near the ground. But “as the
¯
scope of the term aer increased,” Kingsley re¬‚ects, “the scope of the
¯
word aither decreased; by the early fourth century bce it was only used to
¯
refer to the highest and most exalted region of air, up in the heavens.”16 In
the Timaeus Plato registers the semantic modi¬cations undergone by the
term aither, which he uses to refer to a special type of air (Tim. 58 d 1“2).
¯
By this time, “instead of aer being a particular example of aither, aither
¯ ¯ ¯
17
has become a particular example of aer.” The successive history of the
¯
name aither is more complicated and has yet to be written. However, a
¯
crucial development in the career of this name was the identi¬cation of
aither with ¬re. It is very hard to establish who was responsible for this
¯
important development. Three times Aristotle refers to Anaxagoras for the
equation of aither and ¬re,18 and Theophrastus credits him with the
¯
identi¬cation of aither with ¬re.19 According to Simplicius, Anaxagoras
¯
misidenti¬ed aither with ¬re because he was relying on the wrong ety-
¯
mology (aither < aithein, that is to say kaiein, burning).20 We do not have
¯
to agree with Simplicius that the etymology of aither from aithein is
¯
wrong (it is the one that Liddell, Scott and Jones (1940) adopt). But it
seems to me possible that the identi¬cation of aither with ¬re might
¯

14 Kingsley (1995a), in particular the chapters on aither and aer.
¯ ¯
15 Iliad xiv 288: “the ¬r-tree reached through the aer to the aither.” For a convenient discussion of
¯ ¯
this passage and the conception of the world that it presupposes, see Kirk, Raven and Scho¬eld
(1983: 9“10). For a convenient discussion of the Homeric use of the words aither, aer, and ouranos,
¯¯
see Kahn (1960: 133“54). For a study of the use of aither in the Greek tragedians, I refer the reader
¯
to West (1982: 1“13). See also Egli (2003), who offers a good discussion of the way Euripides
manipulates intellectual themes borrowed from Anaxagoras, Diogenes of Apollonia, and the
“philosophers” of the time. For her discussion of aither, see in particular 79“120.

<<

. 4
( 5)



>>