<<

. 3
( 5)



>>

¯ ¯ ¯
de¬ned in this way). Therefore, for the Stoics as well, three-dimensionality by itself can, at most,
characterize a mathematical body, a stereon soma. See Diog. Laert., vii 135 (¼ SVF iii Apollodorus
¯
Seleucensis 6 ¼ LS 45 e).
Aristotle and the Science of Nature
54
the case of the Stoics and the Epicureans this latter de¬nition can be
reformulated to say that body is that which is three-dimensional and resist-
ant to contact.47 We can speculate on the motives that may have induced
¬rst Aristotle and then the Stoics and the Epicureans to develop different
strategies with the common objective of separating the notion of the
natural body from the mere body. In my view the dominant motive
behind these different strategies is the intention of obstructing the path
to the geometric reconstruction of the natural world attempted by Plato.
Without a suitable de¬nition of body that allows a pathway from math-
ematical entities to non-mathematical ones (and vice versa), the recon-
struction proposed by Plato is impossible. It is the ambiguity of the Greek
term soma that renders the entire Platonic operation plausible. Moreover,
¯
this is an ambiguity that Plato exploits knowledgeably, using soma to refer
¯
to the four bodies of the Empedoclean tradition as well as to the four
regular polyhedra with which these simple bodies are associated.

47 I have already emphasized how the Stoics possess another de¬nition of body. This second
de¬nition may be reformulated to say that body is that which is capable of acting or being acted
upon. This de¬nition of body was introduced by the founder of the Stoa, Zeno of Citium, in an
open polemic with the Academic and Peripatetic tradition. From Zeno™s point of view, if x is a
body, then x is capable of acting and of being acted upon; and if x is capable of acting and of
being acted upon, then x is a body. In the Stoic causal chain, there is no room for incorporeal
entities. Moreover, in the Stoic world there are entities capable of acting and of being acted upon,
and entities capable of acting or of being acted upon (where “or” is understood in a rigorously
exclusive sense). Remember that in Stoic philosophy there is room for two principles: an active
principle capable only of acting (god), and a passive principle capable only of being acted upon
(matter). In light of the de¬nition of body introduced by Zeno, both of these principles cannot be
anything other than bodies. Note that the two Stoic de¬nitions of body are not extentionally
equivalent. When the Stoics maintain that the active principle (god) is a body, they certainly do
not mean to say that this body is a three-dimensional entity that offers resistance to contact. The
¬rst manifestation of the active principle, pneuma, is emphatically not a body of this type. It acts
on matter, and by virtue of the second de¬nition of body, it is something corporeal. In light of
this last observation, it should emerge more clearly why the Stoics were not able to renounce the
de¬nition of body introduced by Zeno, and why they could not be content with the de¬nition of
body shared with the Epicureans. The Stoic philosophy is anything but a naive materialism. Only
the de¬nition of body introduced by Zeno allows the Stoics to consider corporeal entities both
matter and the active principle that gives a form and a resolution to it.
chapter 3

Motions




Both Leucippus and Democritus speak of the primary bodies as
always moving in the in¬nite void; they ought to say with what
motion and what is their natural motion (Aristotle, DC 300 b 8“10).


natural and non-natural motions
The student of nature assumes the reality of the natural world and
conceives it as a certain arrangement of natural bodies. Within the broad
compass of natural bodies is found a remarkable array of bodies. They
range from the living celestial bodies performing a circular motion around
the earth, to the living sublunary bodies endowed with the capacity for
poreia and displaying the maximum degree of bodily complexity (perfect
bodies), to the stationary living sublunary bodies (inferior animals and
plants), and ¬nally to the inanimate sublunary bodies. The student of
nature is concerned with all these bodies on the assumption that they
are either simple or composite bodies. Composite natural bodies are them-
selves composed of natural bodies. Earth, water, air, and ¬re are the
sublunary simple bodies. They are the ultimate material principles of all
the bodies that we encounter in the sublunary world, including the
arti¬cial bodies.
All these bodies are liable to undergo motion from one place to
another. Consider the case of a stone: if dropped from a hand, a stone
falls downwards. But why? Aristotle™s view is that a stone is composed of
earth, water, air, and ¬re in a certain ratio, and earth so predominates as to
impart its own natural downward motion to the stone. In other words,
downward motion is the natural motion of the stone because it is the
natural motion of the predominating material principle of the stone,
earth. However, by saying that a stone naturally moves downwards,
Aristotle does not deny that a stone can be moved in a circle by a
stick, or that it can be thrown up by a hand. His view is rather that a
55
Aristotle and the Science of Nature
56
stone non-naturally performs the motions in question, and so does the
earth in the stone. Even these few remarks suf¬ce to illustrate that the
student of nature is not only concerned with natural bodies; he is also
concerned with their motion, or better with the explanation of their
motion. Moreover since the bodies in question are constituted by a
nature, it cannot be a surprise to discover that the explanation of their
motions involves an appeal, direct or indirect, to their nature.
It is signi¬cant, I think, that Aristotle begins his investigation of
the sublunary simple bodies with the claim that the ultimate material
principles of the sublunary world naturally perform a speci¬c motion,
and with a criticism of the cosmological doctrines that overlook this
fundamental truth. What Aristotle has to say against Leucippus and
Democritus is particularly interesting.1 The model of atoms perpetually
colliding in the void fails to give a truly explanatory account of the motion
of atoms. For Aristotle, the motion of an atom, as it is conceived by
Leucippus and Democritus, can only be a case of non-natural motion.
But non-natural motion presupposes, temporally as well as conceptually, a
natural motion of a certain kind. More particularly, if an atom performs a
certain motion as a result of a certain number of collisions, that atom
must have performed an original motion prior to all the collisions
undergone by the atom in its history. The nature of the atom must have
manifested itself in that original motion. In short, that motion was the
natural motion of the atom. But Leucippus and Democritus say nothing
about that motion. They seem to be content to state that the atom has
always been in motion. In this way they do not simply fail to provide an
account for the natural motion of the atom. They thus fail to provide
an adequate account of the motion of the atom.
The Aristotelian doctrine of natural and non-natural motion can be
reconstructed as follows. Let us suppose that the natural body x performs
a certain motion F. This motion is either a case of natural motion, NM,
or a case of non-natural motion, NNM. In other words:
1. If x performs F, then either F is the NM of x or F is the NNM of x.
Since the non-natural motion of x conceptually presupposes the natural
motion of x, one can say that:
2. F is the NNM of x if, and only if, F is not the NM of x.


1 I have quoted the text in the epigraph. Aristotle™s criticism of the atomistic account of motion is
collected both in the DC and in the Metaphysics (Lambda 1071 b 31“4).
Motions 57
But (2) does not adequately grasp the Aristotelian conception of non-
natural motion. If F turns out to be a case of non-natural motion, the
body that performs F also performs some other motion, and this latter
motion is the natural motion of the body. In other words:
3. If F is the NNM of x, there is a motion G 6¼ F, and G is the NM of x.
In the DC Aristotle not only claims that, if a body non-naturally
performs F, it naturally performs another motion. Surprisingly enough,
Aristotle adds that F must be the natural motion of some other body. He
argues that circular motion, since it is non-natural to all the sublunary
bodies, must be natural to some other body (269 a 32 “ b 2). In other
words, Aristotle is committed to the following (stronger) thesis:
4. If F is the NNM of x, then there is a y 6¼ x, and F is the NM of y.
aristotle™s arguments for the existence of a simple
body performing circular motion
In the DC, Aristotle introduces his most controversial reform of the
previous cosmological theories: the thesis of the existence of an ungener-
ated and imperishable celestial simple body that naturally performs circu-
lar motion. His arguments, among other things, shed further light upon
the Aristotelian conception of natural and non-natural motion. In recon-
structing these arguments I shall make use of the following, additional
abbreviations: SM ¼ Simple Motion, SB ¼ Simple Body, CM ¼ Circular
Motion, UpM ¼ Upward Motion, DnM ¼ Downward Motion,
E ¼ Earth, A ¼ Air, W ¼ Water, F ¼ Fire.
The ¬rst argument is presented at 269 a 2“9. Aristotle has already
established that circular motion is a simple motion. He now argues for
the existence of some simple body that performs this kind of motion. The
argument goes like this:
assuming, then, that there is simple motion, and motion in a circle is simple, and
the motion of a simple body is simple and simple motion is <the motion> of a
simple body (for if <simple motion> is <the motion> of a composite body, it
will be in virtue of the prevailing <simple body>), there must necessarily be
some simple body that naturally moves with motion in a circle according to its
own nature. By force it is in fact possible that the motion of a body is also the
motion of another; but this is not possible according to nature, since the motion
according to nature of each simple body is one (DC 269 a 2“9).
The argument of Aristotle can be rephrased as follows:
6. There is SM
Aristotle and the Science of Nature
58
7. CM is an SM
8. Every SM is the motion of an SB
9. There is an SB performing CM.
The crucial premise in the argument is (8). At ¬rst sight, it appears to be
questionable. From the fact that we are observing a simple motion, we
cannot conclude that a simple body is moving. Without additional infor-
mation we can conclude only that a body is moving. In other words, from
the fact that we are seeing a downward motion, we cannot decide what
sort of body is performing it. A downward motion can be performed
either by a composite body “ for example, a stone “ or by a simple body “
the earth that is present in the stone. The words reported in brackets serve
exactly to block this possible objection. The simple motion of a composite
body is ultimately to be explained by recourse to one of the simple bodies.
This is the simple body that predominates so as to impart its own
characteristic motion to the compound.
The argument concludes that there must be a simple body that moves
in a circle. Yet it does not prove very much. It simply proves that there
must be at least one simple body that performs circular motion. People
who held that stars and planets are of a ¬ery nature would have accepted
this conclusion. As it stands, the argument allows them to identify the
simple body performing circular motion with ¬re. But this is exactly what
Aristotle does not want. The argument must therefore be revised in order
to prevent the identi¬cation of the body that performs circular motion
with ¬re. This explains why Aristotle introduces the notion of natural
motion, and argues that any simple body has its own natural motion, and
that there is just one natural motion for each simple body. Two premises
are therefore to be added:
10. Every SM is the NM of an SB
11. There is only one NM of an SB.
Thanks to these two additional premises, Aristotle is now able to conclude
that there must be a ¬fth simple body that is none of the four sublunary
simple bodies, and that circular motion is its natural motion. The revised
argument goes as follows:
6. There is SM
7. CM is an SM
8. Every SM is the motion of an SB
10. Every SM is the NM of an SB
11. There is only one NM of an SB
Motions 59
12. There is an SB 6¼ E, W, A, F performing CM, and CM is its NM.
The crucial premise in the revised argument is (11). It is not dif¬cult to see
why there cannot be more than one natural motion for each simple body.
Simple bodies have such a nature that under the appropriate circum-
stances they always move in the same direction. The nature of earth, for
example, explains why some unimpeded part of earth invariably moves
downwards. Earth can be thrown up, but this upward motion is non-
natural with respect to the nature of earth. Furthermore, since the nature
of a simple body is one, its natural motion too must be one. And if the
downward motion is the natural motion of earth, any other simple
motion cannot be natural with respect to earth. A general principle can
be extrapolated from this example. I shall refer to it as the principle of the
uniqueness of natural motion:
13. If F is the NM of x, then any G 6¼ F cannot be the NM of x.
If the ¬rst argument is meant to prove that there must be a simple body
that naturally performs circular motion, the second argument serves to
block a possible reply by those who hold that celestial bodies are made of a
¬ery stuff. This reply runs as follows. Fire cannot naturally perform
circular motion because each simple body performs only one natural
motion, and ¬re naturally moves upwards. But ¬re can non-naturally
perform circular motion. In the light of this fact, there is no need to
introduce a celestial simple body to account for the motion in a circle: ¬re
can non-naturally move in a circle. Aristotle™s argument consists in a
reductio ad absurdum:
Again, if the motion against nature is contrary to the motion according
to nature, and for one thing there is one contrary, then motion in a circle, being
a simple motion, must necessarily be against nature, if it is not according
to nature, for the moving body. If then ¬re or some other such body is that
which is moving in a circle, its motion according to nature must be contrary to
the motion in a circle. But for one thing there is one contrary, and upward
and downward motion are <already> contrary to one another. If some other
body is moving in a circle against nature, there will be some other motion that
is according to its nature. But this is impossible: if this is upward motion, it
would be ¬re; if this is downward motion, it would be water or earth (DC 269 a
9“18).

If ¬re is non-naturally moving in a circle, then:
14. CM is the NNM of F.
Aristotle and the Science of Nature
60
Since ¬re naturally moves upwards, and upward and downward motion
are contrary to one another, ¬re non-naturally performs downward
motion. In other words:
15. UpM is the NM of F
16. Contr(UpM, DnM)
17. DnM is the NNM of F.
By aggregation of (17) and (14), we obtain that ¬re non-naturally performs
circular and downward motion:
18. CM is the NNM of F and DnM is the NNM of F.
But (18) clashes with the Aristotelian principle that “for one thing there is
one contrary <at most>” “ in Greek hen heni enantion. Consequently,
¬re can neither naturally nor non-naturally perform circular motion.2
The principle that for one thing there is only one contrary at most
turns out to be crucial for the argument. At ¬rst sight this principle is
baf¬‚ing. Aristotle himself appears to provide examples against it.
1. In his ethical theory Aristotle claims that virtue is a mean. But if virtue
is a mean, there are clearly two ways to go wrong: that is, there are two
vices for each virtue, one in the direction of excess and one in the
direction of de¬ciency.
2. Aristotle seems to violate this principle in the Meteorology. On his
account, phenomena such as the shooting stars or the comets take
place in the outer sphere of the sublunary world. This sphere is a
highly in¬‚ammable combination of ¬re and air that under the
appropriate circumstances can be lit by the agency of the immediately
surrounding celestial region. For the present purpose it is not
necessary to enter into the details of Aristotle™s account of these
phenomena. It is enough to recall that the celestial simple body is
naturally moved in a circle, and by so doing it carries around the outer
sphere of the sublunary world. The circular motion of this sphere
appears to be a case of non-natural motion: it is in fact caused by the
agency of an external principle.
3. Even a third argument for the existence of a celestial simple body seems
to be an overt violation of the principle that only one thing is contrary
to another thing. The argument goes like this:


2 I shall return to the language of contrariety in chapter 4. For ancient discussions on this particular
theme in the Aristotelian tradition, I refer the reader to Sharples (1985a: 109“16).
Motions 61
<this is manifest> even on the assumption that every motion is either according
to nature or against nature, and that the motion that is against nature for a body
is according to nature for another, as it happens in the case of upward and
downward motion: for motion against nature for ¬re is according to nature for
earth and vice versa; it is necessary, therefore, that motion in a circle, too, being
against nature for these bodies, is according to nature for some other body (DC
269 a 32 “ b 2).

A possible reconstruction of the argument goes as follows:
19. CM is SM
20. Every SM is the NM of an SB
21. CM is the NNM of E
22. CM is the NNM of W
23. CM is the NNM of A
24. CM is the NNM of F
25. CM is the NM of an SB 6¼ E, W, A, and F.
Aristotle assumes that all sublunary simple bodies can non-naturally
perform motion in a circle. But if they can non-naturally perform circular
motion, the principle that only one thing is contrary to another thing is to
be abandoned. In short, either Aristotle gives up the principle that only
one thing is contrary to another thing, or he holds this principle but gives
up the third argument.
Does the third argument really force Aristotle to abandon the principle
that only one thing is contrary to another thing? More to the point, does
Aristotle really advance two arguments that are mutually inconsistent?
The answer is emphatically no. Aristotle is playing with two distinct
conceptions of non-natural motion. In the third argument circular
motion is non-natural to ¬re because it is not the natural motion of ¬re.
In other words, every simple motion is either natural or non-natural. If
circular motion is not the natural motion of ¬re, then it must be non-
natural to it. Every motion not in accordance with the nature of a body is
therefore non-natural:
26. The NM of x 6¼ the NNM of x.
This conception of non-natural motion does not make any reference to
contrariety “ and a fortiori to the principle that one thing cannot have
two contraries. In other words, any motion that is different from the
natural motion of the body is a case of non-natural motion. By contrast,
the conception of non-natural motion applied in the second argument
crucially depends on contrariety. In this case only one motion of the body
Aristotle and the Science of Nature
62
can be its non-natural motion, and this is the motion that is contrary to its
natural motion. In other words,
27. NNM of x ¼ Contr(NM) of x.
Apparently, Aristotle in the DC makes use of these two conceptions of
non-natural motion without any further comment. He never says that
they are clearly distinct conceptions and that they should not be confused.
But once these two conceptions of non-natural motion are highlighted
and accepted, they can be applied outside the DC. Let us return, brie¬‚y,
to the Meteorology, and to the circular motion that is assigned to the outer
sphere of the sublunary world. Though this motion is a case of non-
natural motion, it does not involve a violation of the principle that for one
thing there is only one contrary at most. As a matter of fact, Aristotle does
not need to make reference to the notion of contrariety to describe this
motion. This motion is non-natural because it is different from the
motion that the air and the ¬re of which the sphere is composed naturally
perform. Put differently, the motion of the outer sphere of the sublunary
world is a case of forced motion: it is the motion of the celestial simple
body immediately surrounding the sphere that forces the air and the ¬re
of which this sphere is composed to move in a circle.3

after aristotle
The thesis of the existence of a celestial simple body was very controversial
in antiquity and did not gain the acceptance one might expect in the light
of the reputation that the same thesis enjoyed in the late Middle Ages and
up until around 1650.4 Proclus, for example, informs us that some
Platonists even reeled back in horror from this thesis because they felt


3 Olympiodorus, Simplicius, and even Philoponus (both in his commentary on the Physics and in
his Contra Proclum), consider this motion a case of motion above nature (or supernatural motion).
Simplicius, for example, compares it to the motion of the planets, which are carried around by the
agency of the sphere of the ¬xed stars:
let us say, even now, that circular motion is not proper to ¬re, since <¬re> is carried around by
the ¬xed sphere, as the motion from the east is not proper to the planets. Nevertheless, it is not
the case that <this motion> is against nature so that it is harmful, but so that it is above nature,
in so far as <¬re> is overcome by something superior and stronger (Simpl., In DC 34. 14“19).
In his commentary to the Meteorology, Philoponus ascribes this solution to the problem to
Damascius (In Meteora 97. 20“2). Wildberg (1988: 129) suggests that Damascius may be the
originator of this solution.
4 For a useful introduction to the fortune of this thesis in antiquity, see Moraux (1964: 1171“263).
For a study of its fortune in the Middle Ages and the Renaissance, see Grant (1994).
Motions 63
there was something barbaric in it (In Tim. ii 42. 9“12). This is clearly an
exaggeration. But, as with all exaggerations, it contains a grain of truth.
The truth is that the overwhelming in¬‚uence of the Timaeus played a
decisive role against the diffusion of this thesis. Very few people in
antiquity were prepared to share with Aristotle the view that celestial
bodies are made of a celestial simple body. Even within the school of
Aristotle, and from the very beginning, the thesis of the existence of a
celestial simple body was resisted. The very unsatisfactory state of the
information at our disposal does not allow us to establish whether
Theophrastus endorsed this thesis.5 But we know that Strato of Lampsa-
cus, the head of the Lyceum after Theophrastus, rejected it and turned to
the Platonic view that celestial bodies are made for the most part of ¬re.6
Xenarchus of Seleucia even wrote a book of objections against the thesis.
Tellingly, this book was entitled Against the Fifth Substance.7 Citations
from this book have come down to us from Simplicius in his commentary
on the DC.8 Simplicius has at least two good reasons for reporting and
debating these objections. First of all, they were a fully developed part of
the exegetic tradition of the DC. Alexander of Aphrodisias had already
reported and debated them in his commentary on the DC. From this
point of view, Simplicius is doing nothing more than following his
reference model, Alexander™s commentary. Moreover, when Simplicius
wrote his commentary, the debate on the Aristotelian doctrine of the ¬fth
substance was anything but closed. This doctrine had recently been
attacked by Philoponus in his Contra Aristotelem. In debating Philoponus™
arguments, Simplicius suggests, venomously, an association between the


5 A convenient review of the information at our disposal is offered in Sharples (1998: 88“94). See
also Sharples (1985b: 577“93).
6 Stob., Ecl. i 200. 21“2 (¼ Aetius ii 11. 4 ¼ Wehrli, Straton 84), Stob., Ecl. 206. 7“8 (¼ Aetius ii 17.
¨ ¨
2 ¼ Wehrli Straton 85).
7 As for the title of the book, see Simplicius, In DC 13. 22; 20. 12; and 21. 33. Note that Xenarchus
refers to the celestial simple body as the ¬fth substance. This fact suggests that by this time it was
already customary to refer to the ¬rst substance (¬rst element, ¬rst body) as the ¬fth substance
(¬fth element, ¬fth body). Regarding the life of Xenarchus, the information in our possession is
scarce and originates almost entirely from the geographer Strabo. Cf. Strabo, Geo. xiv 5. 4. (670).
Xenarchus was originally from Seleucia, Cilicia, but spent most of his life teaching philosophy,
¬rst in Alexandria, then in Athens, and ¬nally in Rome. Xenarchus reached the zenith of his
career as a philosopher and teacher in Rome, where he was introduced at court and even enjoyed
a friendship with Augustus. Arius of Alexandria (Arius Didymus?) must have had an important
role in Xenarchus™ career. Strabo tells us that Arius and Xenarchus were friends. Presumably,
Arius introduced Xenarchus to Augustus. On the basis of this scant information, it is possible to
date Xenarchus™ activity to the second half of the ¬rst century bce.
8 For a convenient presentation of the work of Xenarchus, see Moraux (1967: 1420“35 and 1984:
197“214). See also Sambursky (1962: 125“32) and more recently Hankinson (2002“3: 19“42).
Aristotle and the Science of Nature
64
work of Philoponus and that of Xenarchus.9 In doing so, Simplicius
emphasizes, polemically, how Philoponus™ arguments are not original,
but rather are the result of a reelaboration, if not an outright plagiarism,
of Xenarchus. Here I am content to stress that Philoponus is only the last
link of a longer and more complicated chain. His dependence upon
Xenarchus documents how persistent and well known the critique of
Xenarchus was throughout antiquity. Apparently, this critique became
an essential source for the debate on the existence of a celestial simple
body. However, there is yet little clear evidence that Xenarchus did not
limit himself to raising objections against the doctrine of the ¬fth sub-
stance. He also provided a positive doctrine of natural motion. I am
persuaded that this doctrine was designed to ¬t the conception of the
sensible world offered in the Timaeus. This doctrine had considerable
in¬‚uence among the Platonists of late antiquity. Both Proclus and
Simplicius credit Plotinus with this doctrine and, in all probability, it
was Plotinus and his decision to make this doctrine an essential part of his
interpretation of the Timaeus that is ultimately responsible for the fortune
of Xenarchus in late antiquity. I shall return to this point in due course.
It is a substantial claim of Aristotle™s that every simple body performs a
simple motion. Alongside the claim that the four simple bodies of the
sublunary world naturally move either upwards or downwards, Aristotle
argues for the existence of a celestial simple body that naturally performs
circular motion. Since upward and downward motion are types of recti-
linear motion, the claim that every simple body naturally moves with a
simple motion can be rephrased as follows:
28. If x is an SB, then x naturally performs either CM or RM.
“Either . . . or” is here to be taken with an exclusive sense in virtue of the
principle of uniqueness of the natural motion. From Simplicius we learn that
Xenarchus attacked (28) by distinguishing the element or the simple body
from something that is becoming the element or the simple body:
Rectilinear motion is not the motion according to nature for anything that is
already one of the four elements, but only for something that is becoming <one
of the four elements>. What is becoming is not without quali¬cation: it is
something between being and not being, like what is moving: for this is between
the place to be occupied and the place previously occupied, and becoming is of
the same genus as motion, being itself some kind of change. For this reason we


9 Simpl., In DC 26. 31“3 (¼ Philop., Contra Aristotelem, fr. 1) and 42. 19“20 (¼ Philop., Contra
Aristotelem, fr. 18).
Motions 65
do not say that the so-called ¬re that is moved upwards is, properly speaking,
¬re, but that it is becoming <¬re>. Once it has reached its own place, and has
risen over all the other bodies, it has become, properly speaking, ¬re: for it has
realized its form, in so far as it is light, and in virtue of that position. And earth
is, properly speaking, earth only when it has settled below all other bodies, and
occupies the middle place. Water and air: air when it has risen over earth and
settled below air, and air when it has risen over water and settled below ¬re. It is
therefore false that the motion according to nature of a simple body is simple, for
it has been shown that motion is not an attribute of something that is, but of
something that is becoming, <one of the four elements>. But if some motion,
and a simple one, is to be assigned to what is already <one of the four
elements>, circular motion is to be assigned, since these motions are only two,
motion in a circle and rectilinear motion, and rectilinear motion is the motion of
something that is becoming, but it is not one of the four elements: it is thus not
absurd to assign circular motion to ¬re and rest to the other three <elements>
(Simpl. In DC 21. 35 “ 22. 17).10


10 A note of caution. Since Simplicius is the only source for the objections leveled by Xenarchus, we
have to accept that we are not able to reconstruct a text that is independent of his testimony. For
the same reason, it is impossible for us to evaluate just how liberal Simplicius is being in his
reporting of these objections. The fact that Simplicius introduces some of these objections with a
phesi “ making reference to direct discourse “ does not prove that we are reading ¬rst-hand
¯
citations, if not actually word for word, from Xenarchus™ book. The casualness with which the
ancients reported the citations of others is a well-known fact. Citations from memory, or even
second-hand citations (or, if one prefers, citations of citations), are the rule in the ancient world.
In particular, one should not forget that many of the citations from Xenarchus (but certainly not
all) have been copied from Alexander™s commentary together with the defense of Aristotle
proposed by the latter. This method of proceeding would have had the advantage of being
extremely practical. Instead of ¬rst copying the objections from Xenarchus™ book, and then
consulting Alexander™s commentary for his defense of Aristotle, Simplicius could have consulted
only Alexander™s commentary for both Xenarchus™ objections and Alexander™s defense. In the
worst-case scenario, we would be dealing with citations largely copied from Alexander. In
the most favorable scenario, we would instead be dealing with ¬rst-hand citations which
nevertheless would not exclude the possibility of some reformulations of the text. Moreover, these
reformulations can take very different forms: from the simple alteration of the order of
appearance of words, to the substitution of several elements with others originally absent in the
text, without obviously excluding the presence of abbreviations or additions. Nevertheless, the
study of cases more renowned than that of Xenarchus suggests a cautious optimism. Simplicius™
citations from the Contra Aristotelem by Philoponus are particularly encouraging. A study
conducted on these citations has revealed Simplicius™ habit of abbreviating and selecting the
material for citation. Simplicius quotes word for word in only about ten cases, and in these
cases the citations are usually introduced by paragraphein or paratithesthai. See Wildberg (1993:
187“98). In the others, the citations from the Contra Aristotelem are in reality paraphrases of the
text. In these cases, we can justi¬ably say that we are dealing with testimonies rather than
fragments. But this does not mean that they are inadequate or unfaithful testimonies. Simplicius
seems to have been an accurate witness and even when he offers a paraphrase, he does it while
trying to leave the spirit of the text unaltered. In the absence of indications to the contrary, there
is no reason to doubt that Simplicius proceeded with the same scrupulousness and the same
liberty in the case of Xenarchus as well. The citations from Xenarchus are probably neither true
and proper fragments nor unfaithful paraphrases. Yet, even in this case, the word “citation” must
be given a signi¬cantly broad sense in order to take into account the possibility of reformulations.
Aristotle and the Science of Nature
66
Apparently, Xenarchus relied on a notion of simple body that can be
characterized as follows: x is a simple body if, and only if, x satis¬es all the
conditions that Aristotle posits for being a simple body, and in addition to
that, x is in its own natural place. In order to understand what Xenarchus
was trying to do, it is important to bear in mind that Aristotle describes
the natural motion of a simple body as a motion towards its actuality
(Phys. 255 a 29“30 and DC 310 a 3), or towards its form (DC 310 a 33“4).
Simply put, the natural motion of a simple body, as it is conceived by
Aristotle, is never an unbounded process. On the contrary,
1. this process always has a starting and an ending point; and
2. the ending point of the process is to be identi¬ed with the culmination
or perfection of the process.
Xenarchus attacked Aristotle where his doctrine of natural motion is
weaker. That earth, water, air, and ¬re come to rest once they have
reached their natural places is a fact that we do not see or experience.
Xenarchus attacked this aspect of the doctrine by exploiting Aristotle™s
idea that the end of a process is also its culmination or perfection. More
directly, by introducing the distinction between a simple body and what is
becoming a simple body, he suggested that the statements about the
nature of a simple body should be made with reference to the simple
body in its natural place. In fact, only in its natural place is the nature of
the simple body fully realized. But once the simple body has reached its
natural place, at least for Xenarchus, this simple body either is at rest or is
moved with circular motion. But since the circular motion in question is
performed by the perfected simple body, this motion is the natural motion
of the simple body. In other words:
29. If x is an SB, then either x is at rest or x naturally performs CM.
However, Xenarchus was not content to state (29). He added that recti-
linear motion is performed by a simple body when this body is away from
its natural place and, properly speaking, is not yet a simple body. Since
rectilinear motion is articulated in upward and downward motion, this
¬nal claim can be rephrased as follows:
30. If x is becoming an SB, then x performs either UpM or DnM.
Both the original doctrine of Aristotle and the revised version of
this doctrine proposed by Xenarchus are supported by ordinary observa-
tions only to some extent. Both of them go well beyond what we see
and experience in the sublunary world. First of all, when Aristotle and
Motions 67
Xenarchus claim that ¬re regularly moves upwards, they do not mean to
say that the ¬‚ame of the candle or the ¬re burning in the ¬replace move,
under the appropriate circumstances, upwards. They mean to say that the
simple body that is liberated from that ¬‚ame or the ¬re burning in the
¬replace does it. But we never see a simple body performing a simple
motion. We always see a certain behavior of a certain body that we
conceptualize as the simple motion of a simple body. Secondly, it is even
more dif¬cult to establish what the simple body does once it has reached
its natural place. For Aristotle, a simple body comes to rest once it has
reached its natural place; for Xenarchus it occupies that place either by
staying at rest or by being moved in a circle. Both claims are equally
dif¬cult to verify. Claim (29), together with (30), enable Xenarchus
1. to dispose of the thesis of the existence of a celestial simple body
distinct from earth, water, air, and ¬re, and
2. to incorporate the concepts of natural place and natural motion into a
Platonic conception of the sensible world.
Since Strato of Lampsacus had abandoned the doctrine of the celestial
simple body, it has been suggested that Xenarchus was under his in¬‚u-
ence, or alternatively that he was in¬‚uenced by the Stoics.11 Though there
may be points of contact between the Hellenistic theories of motion and
the positive doctrine of Xenarchus, this doctrine is not reducible to any
of the previous theories. Claims (29) and (30) represent a creative inter-
pretation of the doctrine of natural motion presented in the DC. It is
signi¬cant, I think, that Simplicius, who is notoriously well documented,
never says that Xenarchus depends for his positive doctrine upon the
exegetical activity of someone else. On the contrary, Simplicius presents
Xenarchus as the originator of a doctrine that had a certain success in
antiquity:
it is to be known that Ptolemy in the book On the Elements and in the Optics, the
great Plotinus, and Xenarchus in the objections Against the Fifth Substance, say
that rectilinear motion is <the motion> of the elements that are still becoming,
that are in a place against nature, that have not yet reached the place according to
nature (Simpl. In DC 20. 10“15).
From this passage we learn that Ptolemy and Plotinus endorsed the claim
that rectilinear motion is performed by simple bodies when they are away
from their own natural place. We also learn that they took the view that

11 Gottschalk (1981: 1120).
Aristotle and the Science of Nature
68
away from the respective natural places these bodies are not yet, properly
speaking, simple bodies. A few lines below, Simplicius adds that Plotinus
and Ptolemy were also committed to the view that simple bodies, once
they have reached their natural place, either stay at rest or move in a circle,
and in particular that ¬re and thin air occupy their natural place by
moving in a circle (In DC 20. 23“5).
Simplicius is not the only ancient source in our possession to credit
Plotinus with the claim that ¬re is moved in a straight line when it is away
from its natural place, but that it naturally performs circular motion when
it has reached that place. Proclus provides us with the same information in
his commentary on the Timaeus. In defending the Timaeus against
Aristotle and his thesis of the existence of a special simple body distinct
from earth, water, air, and ¬re, Proclus makes an appeal to a doctrine of
natural motion that he explicitly ascribes to Plotinus. This doctrine goes
like this. When a simple body is in its natural place it is either at rest or is
moved in a circle because it is only by being at rest or by being moved in a
circle that this body can occupy its natural place. By contrast, when a
simple body performs rectilinear motion, this body is not yet in that place
or it has just left it (In Tim. ii 11. 24“31). Elsewhere Proclus ascribes the
same doctrine to both Ptolemy and Plotinus (In Tim. iv 113. 30“1). By so
doing Proclus provides further con¬rmation of what we read in the
commentary of Simplicius on the DC. I add only that Simplicius does
not depend on Proclus, and that his testimony is not an abbreviation of
the testimony offered by Proclus in his commentary on the Timaeus.
Unlike Proclus, Simplicius names Xenarchus, and preserves evidence for
the conjecture that Xenarchus was the ultimate source for both Plotinus
and Ptolemy.
There is only one text which Proclus and Simplicius can refer to in
ascribing also to Plotinus the doctrine of motion that ultimately goes back
to Xenarchus. This is the dif¬cult treatise that is transmitted by Porphyry
with the title On Circular Motion [14]. Plotinus is here committed to the
view that celestial bodies are living bodies, and that their motion is to be
explained by recourse to both their body and their soul. By relying on the
Timaeus, Plotinus assumes that the celestial living bodies for the most part
are composed of ¬re. He takes into account, ¬rst of all, the possibility that
¬re naturally performs rectilinear motion. In this case, however, the
circular motion peculiar to the celestial living bodies could be only the
result of the action of a soul which redirects the rectilinear motion of ¬re
and forces this body to move in a circle (ii 2. 1. 14“19). But this is highly
unsatisfactory, especially in light of the fact that the celestial living bodies
Motions 69
are thought to be divine beings enjoying an eternal life appropriate to
their divine status. There is, nevertheless, always the possibility that
celestial ¬re already moves in a circle. The advantage of thinking of
celestial motion as a result of the action of a soul on a body that already
performs circular motion is suggested by Plotinus himself when he points
out that in this way the celestial souls do not get tired of carrying their
bodies around (ii 2. 1. 37“9). But how can ¬re perform circular motion?
Doesn™t every body, including ¬re, move in a straight line? Plotinus
answers these questions by recourse to the doctrine of natural motion
that both Proclus and Simplicius ascribe to him. Fire performs rectilinear
motion until it has come to the place destined to it (ii 2. 1. 19“23). Once
¬re has reached that place, it does not stay at rest but moves in a circle.
Plotinus also provides a reason for this particular behavior: the nature of
¬re is such that ¬re is always in motion (ii 2. 1. 23“4). Plotinus is very
tentative at this point: ¬re can no longer perform rectilinear motion when
it has reached the extremity of the world, either because ¬re would get
dispersed if it always moved in a straight line (ii 2. 1. 24“5), or alternatively
because there is nothing beyond the extremity of the sensible world and
therefore ¬re cannot keep on moving in a straight line (ii 2. 1. 27“9).
There is therefore only one possibility left, namely that ¬re keeps on
moving, but in a circle rather than in a straight line.

looking ahead
Xenarchus was a remarkably independent thinker. His critique of
Aristotle took the form of a point-by-point refutation of the arguments
in favor of the existence of what Xenarchus (in all probability following an
already established tradition)12 calls the ¬fth substance. His knowledge of
Aristotle was vast and solid and was not con¬ned to the DC. He did
philosophy with Aristotle and through a word-by-word exegesis of the
texts of Aristotle. Though this way of doing philosophy may look familiar
to some of us, we should bear in mind that it was relatively new at that
time. Xenarchus, together with Andronicus of Rhodes and Boethus of
Sidon, belonged to the ¬rst generation of interpreters of Aristotle. How-
ever, the case of Xenarchus may be signi¬cantly different from that of
Andronicus and Boethus. They seem to have been two independent and
intelligent men who put in order the work left by Aristotle. Their job
seems to have been that of organizing, clarifying, and defending the

12 I refer the reader to the Epilogue for a discussion of this claim.
Aristotle and the Science of Nature
70
philosophical work of the master. To the best of my knowledge, none of
this is applicable to Xenarchus. It is signi¬cant, I think, that our sources
never make reference to Xenarchus having any direct link with Andronicus
or with Boethus.13 At any rate, the exegesis of Aristotle did not mean
for Xenarchus the cessation of genuine philosophical thought. The au-
thority of Aristotle provided Xenarchus with the starting point for phil-
osophy, not its cessation. He revised Aristotle™s doctrine of motion and
made it acceptable to late antiquity. Apparently, he agreed with Aristotle
that the celestial world is a special, and somehow distinct, region of the
natural world. He shared with Aristotle the view that the celestial bodies
are not subject to generation and perishing. Like Aristotle, he held the
view that these bodies are moved around the earth with an eternal motion.
But he did not see the need to introduce a material principle that is
different from, and not reducible to, earth, water, air, and ¬re to account
for these features of the celestial bodies. He was persuaded that the
motion of the celestial bodies could be explained without recourse to
the postulation of an additional simple body. The study of the reception
of the doctrine of the so-called ¬fth body shows that many people in
antiquity found themselves in the position of Xenarchus. They simply
could not see the need to introduce a special body to account for the
characteristic incorruptibility and stability of the celestial world. They all
thought that pure ¬re, or fully realized ¬re, adequately accounts for these
features of the celestial bodies. Against this background Aristotle emerges
as an extraordinary exception. This explains why the doctrine of the so-
called ¬fth body is recalled over and over again in the doxographical
tradition.14 The truth of the matter is that Aristotle consciously departed

13 In Falcon (2001: 158“74) I argue that the title “Peripatetic philosopher” which the tradition
attributes to Xenarchus is to be taken as an indication of his interest and mastery of the work of
Aristotle. In my opinion, Xenarchus was not a disciple struggling with problems left unresolved by
his master. He was a creative thinker who concerned himself with several of the same themes with
which Aristotle had already concerned himself and who uses the work of Aristotle as a point of
departure for his own philosophy. The truth of the matter is that the return to Aristotle that took
place in the ¬rst century bce did not involve the acceptance of the views stated by Aristotle.
Aristotle was regarded as an authority, not in the sense that he was over and above criticism, but
only in the sense that he deserved to be studied carefully. Xenarchus is too often described as an
“unorthodox peripatetic philosopher.” See, for instance, Hankinson (2002“3: 19“42). I ¬nd this
description misleading: it obscures the fact that there was no orthodoxy in the Aristotelian
tradition at this early stage. The return to Aristotle in the ¬rst century bce took different forms
and involved a variety of distinct positions.
14 Cf. (i) Stob., Ecl. i196. 5“16 (¼ Arius Didymus fr. 9 Dox. gr.); (ii) [Plutarch], Placita 878 b 8“9
¨
and Stob., Ecl. i 128. 4“9 (¼ Aetius i 3. 22); (iii) [Plutarch], Placita 881 e 10 “ f 7 and Stob., Ecl. i 37.
¨tius i 7. 32); (iv) [Plutarch], Placita 887 d 7 “ 11 and Stob., Ecl. i 195. 20 “ 196. 2
16“18 (¼ Ae
(¼ Aetius ii 7. 5); (v) [Plutarch], Placita 888 b 10“11 and Stob., Ecl. i 200. 25 (¼ Aetius ii 11. 3);
¨ ¨
Motions 71
from the tradition of the Timaeus in order to develop an alternative
conception of the celestial world. Evidently, he was persuaded that the
celestial world is not just a special and somehow distinct region of the
natural world. The postulation of the existence of a celestial body that is
distinct from, and not reducible to, earth, water, air, and ¬re, makes sense
only on the assumption that the celestial world is in some important
respect different from, and not completely reducible to, the sublunary
world. The thesis of the existence of a celestial simple body distinct from
earth, water, air and ¬re suggests that Aristotle took the view that there is
an important discontinuity within the natural world. I shall engage in a
study of this discontinuity in chapter 4. For the time being, suf¬ce it to
say that the doctrine of the celestial simple body points to the existence of
an important discontinuity between the celestial and sublunary world, but
it does not provide a reason for the existence of this discontinuity. In
other words, for Aristotle the celestial world is made of a special material
because there is some important discontinuity between the celestial and
the sublunary world, rather than there being some important discontinu-
ity between the celestial and the sublunary world because the celestial
world is made of a special material.

voluntary motion
So far I have argued that Aristotle is committed to the view that motion is
either natural or non-natural. I have also argued that Aristotle admits at
least two different concepts of non-natural motion. He explicitly identi-
¬es the non-natural motion of a body with the motion contrary to the
natural motion of the body:
31. NNM of x ¼ Contr(NM) of x.
But at times Aristotle simply equates non-natural motion with forced
motion. In other words, any motion that a body may perform against its
nature is a case of non-natural motion. This concept of non-natural
motion may be characterized by saying that any motion which is not in


(vi) Athenagoras, Legatio pro christianis 6. 4. 25“30; (vii) [Iustinus], Cohortatio ad graecos 5. 2.
15“20; (viii) Hippolytus, Refutatio omnium haeresium i 20. 4, vii 19. 3“4; (ix) Sextus Emp., PH iii
31 and M ix 316; Diog. Laert., v 32; (xi) [Galen], Historia philosopha 18 (¼ Dox. gr. 610“611); (xii)
Achilles, Isagoge 2. 1 (¼ 30. 25“7 Maas); (xiii) Basil, Hexaemeron i 11 (¼ 18. 17“18 de Mendieta and
Rudberg); (xiv) Ambrogius, Exameron i 6. 23 (¼ 21 d“e Schenkel); (xv) Theodoretus, Graecarum
affectionum curatio iv 12, iv 18, iv 21; (xvi) Nemesius, De natura hominis 5. 165 (52. 18“23 Morani).
Aristotle and the Science of Nature
72
accordance with the nature of the body is a case of non-natural motion. In
other words:
32. NNM of x 6¼ (NM) of x.
I would like to enrich the conceptual apparatus so far developed by taking
into account a testimony preserved by Cicero, whose ultimate source is
presumably Aristotle™s dialogue On Philosophy:
(1) But neither is Aristotle undeserving of praise, in that he thought everything
that is moved is moved either by nature or by force or by will [voluntate]; (2) the
sun, the moon, and all the stars are moved, (3) but the things that are moved by
nature are moved either downwards by being heavy or upwards by being light;
(4) neither of which is proper to heavenly bodies, because their motions are
circular. (5) But neither could it be said that it is by some greater force that the
celestial bodies are moved against nature. (6) For what can be greater? (7) It
remains, then, that the motion of the celestial bodies is voluntary (Cicero, Nat.
deor. ii 44 ¼ On phil. fr. 21b Ross ¼ fr. 836 Gigon).15

The great intrinsic interest of this testimony is the claim that celestial
motion is voluntary “ clause (7). This claim is the conclusion of an
argument whose ¬rst premise is the tri-partition: (i) natural motion, (ii)
forced motion, and (iii) voluntary motion “ clause (1). The desired conclu-
sion is reached by excluding that the characteristic motion of celestial
bodies is either a case of forced or natural motion. Since forced motion is
any motion that is imposed from the outside, it is relatively easy to see
why celestial motion cannot be a case of forced motion “ clause (5). This
would imply the existence of some force greater than the celestial bodies



15 Cicero goes on as follows: “(8) Anyone who sees this truth will show not only ignorance but also
wickedness if he will deny the existence of gods.” Following Rose (1886) and Walzer (1934), Ross
(1955b) prints this clause in 21b. I prefer to follow Gigon (1987), who does not print (8). This
clause does not seem to be part of the argument which ultimately goes back to Aristotle and
describes celestial motion as a case of voluntary motion. The reference to the existence of gods
makes it clear that Cicero is going back to the divinitas of the celestial bodies, which was his
original issue. For a vindication of this interpretation, see Effe (1970: 131). Even if we opt for this
more prudent hypothesis, at least two scenarios are still possible: (1) Cicero read Aristotle™s On
Philosophy and decided to corroborate the Stoic arguments in which he is primarily interested with
the citation from Aristotle (this possibility is advanced in Furley (1989)); (2) Cicero did not read
Aristotle™s On Philosophy but found, in his Stoic source (Poseidonius?), the arguments already
corroborated with Aristotle™s citation. I share the prudence and the skepticism of Moraux (1964:
1222“3), but cf. also Van den Bruwaene (1978: 65n47). With careful examination, the alleged
fragments of an ancient author frequently reveal their nature as testimonies (often only second-
hand). I do not think that the argument reported by Cicero is an exception to this general rule. In
the best case, it is nothing more than a Latin paraphrase of a lost dialogue of Aristotle. In the worst
case, it is a Latin testimony of a Greek testimony of a lost dialogue of Aristotle.
Motions 73
and imposing circular motion upon them “ clause (6).16 It is much more
dif¬cult to see why celestial motion cannot be a case of natural motion.
Yet the testimony is crystal clear on this point: if something is a natural
body, then it naturally performs either upward or downward motion only
“ clause (3). A more careful reading of this clause, however, shows that the
natural motion in question is the motion that a natural body performs in
so far as it is heavy or light. Heavy bodies fall because that motion is in
their own nature; in other words, falling is natural to them. This does not
exclude that the motion a natural body performs in so far as it is a living
body can be a case of natural motion. Suppose that the heavy body in
question is a man. A man is naturally equipped with a locomotory and
sensory apparatus for moving around by walking. For Aristotle, a man
moves around by walking because it is in his own nature to do so; in other
words, walking is natural to him. Of course it is always possible to think
of a situation in which a man is constrained to walk (for example,
somebody threatens him with a gun). Though I think that Aristotle would
count constrained walking as a case of natural motion (because the
ultimate source of the motion is still an internal principle of motion), I
con¬ne myself to the much simpler case of a man who is moving around
by walking and is not constrained to do so. It is hard to imagine that
Aristotle, in his On Philosophy, could deny that this walking is a genuine
case of natural motion. This walking is natural to the man simply because
there is no external force compelling the man to move from one place to
the other. Admittedly, external factors may play a role in the production
of this particular motion, but its basic shape and course are ultimately
regulated by a nature of a speci¬c type: a human soul. More generally, it is
hard for me to imagine that Aristotle could deny that animal motion is a
genuine case of natural motion, if the animal engaged in the motion is not
externally forced or compelled to perform the particular motion it does.
Animal motion is a case of natural motion because it is caused by the
appropriate type of nature, an animal soul. In other words, what is
distinctive of animal motion is that it has a psychological cause.
Let us return, in the light of these remarks, to celestial motion and the
way this motion is presented in the DC. From the DC we learn that the


16 Notice that this force would have to be in¬nite. On Aristotle™s principles, only an in¬nite force
could keep the constraint upon the celestial bodies forever. But there is no such force in the
natural world. Alternatively, we could posit the existence of an in¬nite number of ¬nite forces
successively keeping celestial bodies moving in the relevant way. But this would be regarded, I
think, as a highly questionable assumption.
Aristotle and the Science of Nature
74
celestial bodies are not deprived of life and that their distinctive circular
motion around the earth is the motion of living bodies. More speci¬cally,
we are required to think of celestial bodies as intelligent living bodies
engaged in motion (292 a 18“22).17 Admittedly, Aristotle never offers a
direct argument in support of this claim, in the DC or elsewhere. He
presumably thinks that the explanatory bene¬ts that depend upon this
assumption are also an indirect argument in support of the assumption
itself. Whatever the case may be, it is clear that for Aristotle the study of
the celestial bodies is to be conducted in the same way as the study of all
perishable living bodies is conducted; that is, moving from the activities
that are constitutive of the behavior of these living bodies, and looking for
an internal source governing these activities and shaping them into a
unitary behavior “ the behavior distinctive of the living bodies in ques-
tion. I speak of activities (rather than activity) because from earth the
celestial bodies do not appear to be engaged in simple circular motion,
but they appear to revolve around the earth with a relatively complex
behavior. However, complexity does not involve ¬‚exibility: the celestial
bodies do move around the earth with a relatively complex behavior, but
they are unable to stop moving or to move in any other way than they
actually do. Put differently, lack of ¬‚exibility appears to be a distinctive
feature of the motion of the celestial bodies.18 I postpone discussion of this
lack of ¬‚exibility and its relevance; for the time being, I focus only on the
complexity of celestial motion. Aristotle presumably thinks that the
complex behavior of the celestial bodies can be adequately explained only
by appeal to a psychological principle of unity and intelligibility, a soul of


17 Aristotle makes this claim in discussing the following two dif¬culties: why are the sun and the
moon moved by fewer motions than some of the other planets? (291 b 29“31); and (ii) why are so
many stars carried by one single motion “ the motion of the heaven of the ¬xed stars “ whereas
many motions are needed to carry one single planet? (292 a 10“14). We are used to thinking of
celestial bodies as mere bodies and units which do have order but do not have a soul (292 a 18“20).
On the contrary, we should conceive of them as partaking of life and action: in this way what
occurs will not seem to be anything contrary to reason (292 a 20“2). Thinking of something as a
unit means making abstraction of some of its natural properties and conceiving of it as a point. In
other words, the dif¬culties stated above can be successfully treated only if we abandon the
geometrical models offered by Eudoxus of Cnidus and his followers and conceive of celestial
bodies as living beings. Notice that celestial motion is here presented as a case of action (292 a
20“2). From Aristotle™s discussion of the two dif¬culties, it is clear that the relevant notion of
action “ the notion of action on which Aristotle is here relying “ makes a crucial reference to the
good. Apparently, to be alive is to be sensitive to the good, and action is the way in which a living
being is sensitive to the good. Action so understood is attributed to plants, animals, human
beings, and celestial bodies. For this particular notion of action, see also DA 415 b 1“3.
18 We usually focus on the regularity of celestial motion rather than on its lack of ¬‚exibility. But it is
easy to see that regularity involves lack of ¬‚exibility (and vice versa).
Motions 75
a certain type.19 In his view, celestial motion is for the celestial bodies as
natural as walking is for men. In the latter case the ultimate source of
walking is a psychological principle of a speci¬c type: a human soul.
Accordingly, the ultimate source of celestial motion is a psychological
principle of a speci¬c type: a celestial soul. The similarities existing
between the two cases should not obscure the fact that Aristotle admits
important differences too. First of all, walking is a case of progressive
motion or poreia and implies, minimally, the actual exercise of perception
and phantasia. By contrast, celestial motion is not a case of navigation
from one place to the other, and it does not require the actual exercise of
perception or phantasia. I shall return to this doctrine in chapter 4.
Secondly, it is in the nature of man to walk (in the sense, for example,
that it is up to him to cover some distance by walking). But though
walking is natural to him, it is not natural to his bodily parts. To begin
with, we get tired of carrying our body around and need to stop walking
and rest (and eventually eat and drink). Moreover, we can damage our
joints, and ¬nally get injured, by covering excessive distances without
appropriate rest. By contrast, the celestial natures do not make any effort
to move in a circle. Nor do they get tired of carrying around their body.
They are composed of a simple body that already performs circular
motion. It is not dif¬cult to ¬nd a reason for a doctrine that at ¬rst
appears to be bizarre, if not redundant. From the DC we learn that any
account of the celestial bodies should accommodate the belief that the
celestial bodies are divine beings engaged in an eternal blissful life. By
being realized in a body that is naturally moved in a circle, they can enjoy
the eternal and blissful life that is appropriate to their divine status.
By simply accepting that Aristotle conceives of celestial bodies as
intelligent living beings engaged in action, and thinking of this action

19 In connection with this claim I ¬nd it useful to think of the way in which the apparently erratic
motion of the planets was routinely explained by the help of geometrical models, which all went
back, in one way or the other, to the one developed by the great mathematician Eudoxus of
Cnidus. According to the tradition, Eudoxus produced his geometrical models in response to
Plato™s challenge to “save the phenomena.” Roughly, by combining a certain number of
homocentric (or concentric) spheres into a single system, and by giving each sphere a speci¬c
rotation and angle of inclination, Eudoxus and his followers were able to approximate the motion
described by a celestial body in the heavens. Though Aristotle thinks that these geometrical
models can by no means provide an adequate explanation of celestial motion, he is ready to accept
the idea that the complex behavior of a planet can be analyzed into a certain number of simple
circular motions. But if this is the case, an appropriate nature as principle of unity (and
intelligibility) is required to transform these simple activities into a single, complex behavior; that
is, the behavior of the planet. For Aristotle, this nature can be only a speci¬c type of soul, a
celestial soul. On the soul as principle of unity (and intelligibility), I refer the reader to chapter 2,
“Bodies.”
Aristotle and the Science of Nature
76
as a rational activity involving the exercise of the capacity for thought and
desire, we have explained how it is possible for Aristotle to think of
celestial motion as a psychological activity; however, we have not yet
explained how it is possible for him to speak of this activity as a case of
voluntary motion, let alone as a case of voluntary action. From our point
of view, something performing circular motion forever, without being
able to stop moving, or to perform a different type of motion, or to
perform the same motion but in a different direction, is by no means
engaged in a voluntary motion. Our dif¬culty ultimately depends, I
think, on a certain conception of the voluntary. We are inclined to make
the voluntary conceptually dependent on that which can be chosen. For
the sake of illustration, let us return to animal motion as experienced on
earth. We may think that walking is voluntary because it is performed as
the consequence of a choice and conforms to that choice. Though a
walking man can continue to walk, it is in his power to stop walking. It
is in his power to continue to walk because the other option is available
to him. In other words, he can decide to act otherwise. Clearly, this is
not the conception of the voluntary on which Aristotle could rely in
his On Philosophy to claim that celestial motion is voluntary. Though it is
not possible to say exactly to which conception of the voluntary Aristotle
may have had recourse in this lost dialogue, we can always turn to his
ethical works for enlightenment on this matter. There Aristotle is not only
content to make use of the conception of the voluntary; in addition, he
spells out theoretically what the voluntary is. For him, the voluntary is
simply that which is not forced or compelled by the outside, but takes
place in accordance with an internal principle of motion. This is particu-
larly clear from the way the non-voluntary is introduced in the NE: an
action is non-voluntary if it takes place by force or through ignorance
(1109 b 35 “ 1110 a 1). Moreover, an action takes place by force as it comes
from a source external to the agent and nothing is contributed by the
agent itself (1110 a 1“4). Apparently, Aristotle recognizes the voluntary as a
particular case of the natural (and, accordingly, the non-voluntary as a
particular case of the non-natural). This is not the place to enter into a
discussion of the eventual merits and limits of this treatment of the
voluntary.20 What matters is that this particular approach provides us
with the conceptual resources to make sense of the testimony preserved by
Cicero. From Cicero we know that this activity cannot be imposed upon

20 For a convenient discussion of Aristotle™s reduction of the voluntary to the natural, see Broadie
(1991: 132“42).
Motions 77
the celestial bodies from outside. There is nothing in the natural world
that can force the celestial agents to act in the way they do “ clauses (5)
and (6). Celestial motion is therefore to be explained exclusively with
reference to an internal source of motion. But since this activity is
conceived as a case of rational activity, involving the exercise of the
capacity for thought and rational desire, it can be nothing but a case of
voluntary action.21

EPINOMIS
the ¬fth body in the
Up until now, I have spoken of the simple celestial body as if it were a
unique creation of Aristotle. And yet the claim of the existence of a ¬fth
body along with earth, water, air, and ¬re is present from the very
beginning in the Platonic tradition.22 From this point of view, the Epino-
mis is a model document. The notion that this dialogue is not Plato™s but
the work of one of his immediate disciples (Philip of Opus?) dates back to
the sources of Diogenes Laertius.23 Interest in this dialogue resides primar-
ily in the fact that it documents, in an extraordinarily ef¬cacious way, how,
from a certain point on, Plato™s thought, especially as it is presented in the
Timaeus and the Laws, became Platonic doctrine. Part of this doctrine,
though not necessarily part of Plato™s thought, is the thesis of the existence
of a ¬fth body. What is most important for the present discussion is to
ascertain whether there is some relationship between the ¬fth body of the
Epinomis and the simple celestial body of Aristotle. Once light has been
shed on the fact that the author of the Epinomis makes use of a different
conceptual apparatus to introduce his own ¬fth body, it will be easier to
understand why, since antiquity, Aristotle has been presented as the

21 In passing, I point out that Alexander of Aphrodisias recognizes celestial motion as a case of free
motion. See Alexander, De fato 181. 16“20. A ¬nal note on Aristotle™s On Philosophy is needed.
From the scanty information concerning its content, we cannot be certain that in this dialogue
Aristotle was committed to the existence of a celestial simple body that is different from, and not
reducible to, earth, water, air, and ¬re (pace Jaeger, 1948). Certainty in this matter remains beyond
our reach. However, I am inclined to think that Aristotle always thought that the celestial bodies
are made of a special body, unique to them. In other words, this view is not speci¬c to the DC,
and from the little we know about the content of On Philosophy we should not conclude that in
this dialogue Aristotle offers an explanation of celestial motion that is at variance with that of the
DC (pace Guthrie, 19866). On this point see also Moraux (1964: 1210“13).
22 For a survey of the testimonies in our possession, see Moraux (1964: 1184“6).
23 Diog. Laert., iii 37. The ancient debate on the authorship of the Epinomis is partially preserved in
the Anonymous Prolegomena, x 25. 1“10. From this Neoplatonic introduction to the philosophy of
Plato we learn that Proclus was skeptical about the authenticity of the dialogue. For a convenient
´
introduction to the Epinomis, and the ancient and modern debate over its authorship, see Taran
(1975: 3“47). For a recent introduction to the Epinomis, see Dillon (2003: 178“97).
Aristotle and the Science of Nature
78
discoverer of the simple celestial body. This will also explain how, in the
Platonic tradition, a doctrine that contemplates the existence of ¬ve bodies
can coexist with a criticism of the Aristotelian view that the celestial bodies
are made of a celestial simple body which is different from, and not
reducible to, earth, water, air, and ¬re.
The way in which the ¬fth body is introduced and justi¬ed in the
Epinomis has nothing to do with the way in which the simple celestial
body is proved in the DC. While Aristotle establishes a correlation
between bodies and motions, the author of the Epinomis links bodies to
regular polyhedra. More speci¬cally, the speculations of the author of the
Epinomis on the nature and number of bodies are the result of a creative
interpretation, if not a deliberate misunderstanding, of the Timaeus. In
the Timaeus, earth, water, air, and ¬re are associated with four regular
polyhedra.24 These four solids are constructed from two elementary
triangles: the scalene rectangular triangle and the isosceles rectangular
triangle. The icosahedron, the octahedron, and the pyramid are con-
structed from the scalene rectangular triangle, the cube from the isosceles
rectangular triangle. However, geometry at the time of Plato recognized
also a ¬fth regular polyhedron that consists of twelve pentagonal faces and
that can be constructed from either of the elementary triangles, the
dodecahedron. Plato assigns to the dodecahedron a rather mysterious
task: god availed himself of this ¬fth ¬gure in order to decorate the
universe (Tim. 55 c 4“6). Even these few mentions are suf¬cient to
appreciate the importance of the Timaeus as a historical document on
the state of geometry during Plato™s time. From independent sources we
know that several of the results achieved in the ¬eld of the geometry of
solids are attributable to Theaetetus, the gifted mathematician to whom
Plato dedicated one of his most important dialogues.25 Among the reasons
that motivated Plato to make extensive use of Theaetetus™ discoveries, and
of those of his contemporaries and predecessors, was surely the conviction
that geometry could offer a method of studying and analyzing the natural
world. The fate reserved for the dodecahedron is nevertheless instructive.
It helps us to understand how, at least for Plato, geometry could offer a
method for analyzing nature but it could not provide a criterion for
establishing what exists in nature. In particular, the fact that there are

24 I have brie¬‚y presented and discussed this doctrine in chapter 2: “Bodies.”
25 Suda Lexicon, s.v. Theaetetus: “Theaetetus of Athens, astronomer, philosopher, disciple of
Socrates, taught at Heraclea. He was the ¬rst to construct the so-called ¬ve solids. He lived after
the Peloponnesian war.” Apparently, Theaetetus was the ¬rst to study the octahedron and the
icosahedron, and it is believed that Book xiii of Euclid™s Elements is based on his work.
Motions 79
¬ve regular polyhedra is not, at least for Plato, a suf¬cient reason for
introducing a ¬fth body alongside earth, water, air, and ¬re. In all
probability, the author of the Epinomis was also convinced that geometry
could not offer the ultimate criterion for deciding what there is in nature.
And yet, using the Timaeus and the results reached by Theaetetus as a
point of departure, he concluded that if the regular polyhedra are ¬ve in
number (981 b 3“4), then the bodies must also be ¬ve in number: ¬re,
aither, air, water, and ¬nally earth (981 c 5“8). Why?26
¯
The operation attempted by the author of the Epinomis will become
clearer if we concentrate on the order in which the bodies in question are
offered. Fire and earth are the two outermost bodies. They are present in
every composite body. But they cannot be mixed without the help of
aither, water, and air. The job of these three intermediate bodies is that of
¯
glue or cement: they serve to hold earth and ¬re together. These ¬ve
bodies are present, in different proportions, in every composite body.
Moreover, a predominant element is easily detectable in each composite
body. In our body, for example, earth predominates. Yet, besides the
earth, a certain amount of water, air, aither, and ¬nally ¬re is also present
¯
(981 c 8 “ d 5). The celestial bodies are no exception to the rule. The sun,
the moon, and the remaining planets are composite bodies. More specif-
ically, these bodies are composed, for the most part, of ¬re.27 But next to
¬re it is possible to detect not only the presence of earth and air, but also
traces of aither and water. Even from these few remarks, it is evident that
¯
the author of the Epinomis accepts some version of the principle that can
be extracted from Tim. 31 b “ 32 c:
P: if x is a body, then x is composed of E, W, A, and F.28

26 Note that the author of the Epinomis calls his ¬fth body aither. Aristotle refrains from using this
¯
name. I refer the reader to the Epilogue for a discussion of Aristotle™s silence. For the time being, I
only note that the name aither is used in the Timaeus to refer to a speci¬c kind of air, not ¬re. In
¯
all probability, Anaxagoras was the ¬rst to use the term aither to refer to ¬re. The (ab)use of the
¯
name aither to refer to different things did not prevent the author of the Epinomis from using it.
¯
27 Here the author of the Epinomis follows the Timaeus. This text was usually read in the light of
Tim. 58 c 5“7, where Plato claims that there are different forms of ¬re.
28 More speci¬cally, the author of the Epinomis accepts the following version of this principle:
P*: if x is a body, then x is composed of E, W, A, Aither, and F.
¯
Is Plato really advancing principle P in Tim. 31 b “ 32 c? At least at ¬rst sight, one is tempted to
answer no. In this passage, Plato is not offering an account of the composition of every single
body. He is rather offering an account of the body of the universe as a composition of earth,
water, air, and ¬re. However, one has also to acknowledge that this particular body is not merely a
body among others but the body par excellence. And this may explain why the whole passage
could easily be understood as offering an account of the composition of every single body.
Moreover, a comparison of the ¬rst two lines of our passage with Tim. 28 b 7 “ c 2 suggests that
Aristotle and the Science of Nature
80
It is less easy to understand up to what point he is aware of distancing
himself from the doctrine reported in Tim. 31 b “ 32 c. The doctrine of
the Epinomis is in fact incompatible with the mathematical speculations
contained in Tim. 31 b “ 32 c on the number of intermediate elements
needed to mix with the outermost ones. The introduction of a ¬fth body
allows one to leap over the mathematical conjectures in support of the
existence of two intermediate bodies (water and air) and two outermost
bodies (¬re and earth). The intermediate bodies that carry out the
function of glue are in fact three: besides water and air he also considers
aither. But is it possible to preserve the doctrine of bodies offered in the
¯
Timaeus by renouncing the mathematical speculations that sustain it? For
the author of the Epinomis evidently yes.
I do not think it is a mistake to present the Epinomis as the attempt to
construct a doctrine of bodies that also reserves a space for that ¬fth
regular polyhedron: the dodecahedron. Nevertheless the author of the
Epinomis does not limit himself to introducing a ¬fth body. He uses this
body to construct a demonology that develops, in a systematic way, the
frequent references in the Platonic dialogues to the existence of intermedi-
ate entities whose primary task is that of functioning as mediator between
humanity and the divine. Once again, the way in which the author of the
Epinomis proves the existence of demons is the result of a creative

Plato is ready to extend his speculations about the composition of the universe to every single
body. I would like to focus on a consequence descending from this reading of Tim. 31 b “ 32 c.
What we usually call earth, water, air and ¬re are not the same as the elements which are usually
referred to with the name “earth,” “water,” “air,” and “¬re.” If principle P holds, any quantity
whatsoever of the body we usually call earth is a composition of earth, water, air, and ¬re, and
therefore the name “earth” is merely an indication of the dominant element, earth. In fact, from
1. P: if x is a body, then x is composed of E, W, A, and F.
By replacing x with E, W, A, and F, one obtains:
if E is a body, then E is composed of E, W, A, and F
2.
if W is a body, then W is composed of E, W, A, and F
3.
if A is a body, then A is composed of E, W, A, and F
4.
if F is a body, then F is composed of E, W, A, and F.
5.
This interpretation of Tim. 31 b “ 32 c is discussed “ rejecting P “ by Alexander of Aphrodisias in
Mantissa 123.4 “ 126.23. I owe this point to Bob Sharples. In all probability, Alexander™s polemical
target is a certain interpretation of Tim. 31 b “ 32 c which was endorsed, among the others, by
Numenius. Cf. Proclus, In Tim. iii 9. 4“5 (¼ Numenius fr. 51 Des Places). For the reception of P
in the Platonic tradition, see Falcon (2001: 123“44). I argue that Tim. 31 b “ 32 c played an
important role in the critique of Aristotle™s simple celestial body. Plotinus and Proclus provide us
with good examples of the way P could be used against Aristotle (Plotinus, ii 1.6.1“21, and Proclus
In Tim. ii 43. 20 “ 44. 18). From GC 334 b 30“1, we learn that Aristotle was ready to accept that a
version of P holds for the sublunary world. In other words,
6. P**: if x is a sublunary body, then x is composed of E, W, A, F.
Motions 81
interpretation, if not a deliberate misunderstanding, of the Timaeus. More
speci¬cally, he associates each element with a particular type of creature.
Earth is associated with a rather broad genus that comprises all terrestrial
creatures (981 c 8 “ d 5). Fire is instead associated with the divine genus of
celestial creatures “ the celestial bodies. Between the terrestrial and
celestial genus, there must exist creatures of an intermediate nature
associated with the intermediate bodies. Aither is the origin of a ¬rst type
¯
of beings, which are speci¬cally called demons (984 e 1). Apparently,
demons are invisible to our eyes because the predominant element in their
body is transparent (984 b 6 “ c 2; 984 d 8 “ e 5). The intermediate nature
of aither places these beings between the earth and the sky, thus allowing
¯
them to carry out that mediating function between gods and humans that
is characteristic of them (984 b 4 “ e 5). After the demons, the author of
the Epinomis considers two other intermediate types of creatures: the
aerial and aqueous creatures, in which air and water respectively carry
out the role of the dominant element. Thanks to the intermediate nature
of their dominant element, even the aerial creatures (like the demons)
seem to share a function that is peculiar to the demons: they can in
fact pass with ease from earth to the celestial region and vice versa (985 b
1“4). Although information on the aqueous creatures is scarce, we under-
stand that their intermediate nature makes them semi-divine creatures
(985 b 4“c 1). However, the author of the Epinomis does not seem to grant
to these creatures any intermediary function between the heavens and
earth.
Even this brief presentation of the demonology of the Epinomis should
suf¬ce to demonstrate how the doctrine of the celestial simple body
defended by Aristotle has nothing to do with the doctrine of the ¬fth
body advanced in the Epinomis. To begin with, the use of the doctrine
introduced in the Epinomis is signi¬cantly different. For the author of the
Epinomis the introduction of the ¬fth body is not a consequence of
convictions about the nature of the celestial bodies. In the vein of the
Platonic tradition of the Timaeus, he is committed to the material unity of
the natural world. The introduction of the ¬fth body is functional to a
task foreign to Aristotle™s concerns, but perfectly comprehensible within
the Academic tradition. Though few, the testimonies in our possession are
suf¬cient for maintaining that demonology was a serious business within
Plato™s Academy.29 With the introduction of the ¬fth body, the author of

29 Xenocrates was particularly active in this ¬eld. The best introduction to Xenocrates and his
demonological theory is still Heinze (1892: 78“125). On Xenocrates™ demonology see Plutarch, De
Aristotle and the Science of Nature
82
the Epinomis is able to elaborate a cosmology that gives space to the
demonic beings of intermediate nature between gods and humans which
are frequently referred to in the Platonic dialogues. Secondly, and more
importantly, the theory that underpins the introduction of the ¬fth body
in the Epinomis is incompatible with the conceptual apparatus to which
Aristotle makes reference in arguing for the existence of a celestial simple
body. The author of the Epinomis endorses the theory advanced in the
Timaeus, and in particular the idea that a geometric structure is to be
assigned to each body. But the geometrical account offered in the Timaeus
clashes with the principle of the ever-divisibility of body. Moreover, from
the DC we learn that for Aristotle ever-divisibility (along with three-
dimensionality) is a distinctive feature of a body. Lack of clarity on this
point, admonishes Aristotle, jeopardizes the ¬nal result of our investi-
gation.30 In the DC Aristotle also proves that the idea of assigning a
geometric ¬gure to each body functions only on the condition that one
assumes that the bodies in question are not divisible into ever-divisible
parts (306 a 30 “ b 7).
I would like to end this brief discussion of the doctrine of the ¬fth body
in the Epinomis with the map of ancient dogmatic philosophy that Sextus
draws with respect to the attitude it shows towards the much debated
question of the divisibility of body:
(1) There is an undecidable dispute amongst all the philosophers: some of them
say that body is indivisible, others that it is divisible; (2) and of those who say
that body is divisible some claim that body is in¬nitely divisible, others that the
division stops at what is minimal and atomic (M I 27).
According to Sextus, there are philosophers who believe in the divisi-
bility of body and philosophers who do not “ clause (1). Moreover, those
who claim that body is divisible can be further divided. Whereas some
philosophers believe in the in¬nite divisibility of body, others think that
there are items which cannot be further divided and that the division of


defectu oraculorum 12. 416 b“d (¼ Heinze fr. 23); Proclus, In Remp. ii 48. 4“27 (¼ Heinze fr. 23);
Plutarch, De Iside et Osiride 25. 360 d“f (¼ Heinze fr. 24). Interestingly enough, from Simplicius
we learn that Xenocrates in his Life of Plato ascribed to Plato the view that the zoia can be divided
¯
to arrive at ¬ve stoicheia, which are called schemata and somata: aither, ¬re, water, earth, and air.
¯ ¯ ¯
Cf. Simpl., In DC 12. 22“6 and In Phys. 1165. 35“9 (¼ Heinze fr. 53). The Greek zoia is ambiguous
¯
in various ways. Cf. chapter 1, “The unity, structure, and boundaries of Aristotle™s science of
nature.” In this case it refers to all animals there are, including any living beings which there might
be superior to men. If this is the case, this testimony is part of an attempt to develop a
demonology from the doctrine of the Timaeus.
30 I have discussed the importance of this principle in chapter 2, “Bodies.”
Motions 83
body stops when these items are reached “ clause (2).31 Sextus™ aim is
clear: he wants to provide two claims and two counterclaims on the issue
of the divisibility of body “ body is divisible/body is not divisible; body is
ever-divisible/body is not ever-divisible “ in order to end up in a suspen-
sion of judgment because it is not possible to come down on one side or
the other in the dispute. Of course we do not have to buy into Sextus™
conclusion “ the suspension of judgment. We have only to realize that the
DC and the Timaeus (and therefore the Epinomis) are on different sides in
the dispute over the divisibility of body.

taking stock
Before proceeding we would do well to pause and take stock of the results
achieved so far. The student of nature is concerned not only with natural
bodies but also with the explanation of their motions. For Aristotle, the
celestial bodies are intelligent, living bodies that perform a regular but
complex motion around the earth. He is persuaded that the explanation
of the behavior of the celestial bodies requires an appeal to a psychological
cause, a soul equipped with the capacity for thought and desire. In the
following chapter I shall return to celestial thought and celestial desire.
For the time being, I am content to insist on the following crucial, though
too often neglected, truth: the celestial motion that is naturally performed
by the celestial bodies is not the circular motion that is naturally per-
formed by the celestial simple body. Celestial motion is the motion of a
living body engaged in a speci¬c animal motion, and as such it involves
the reference to a psychological cause: a soul of a certain type. This
motion is complex and involves the exercise of celestial cognition and
celestial desire. By contrast, the circular motion that is naturally per-
formed by the celestial simple body is a simple motion and does not
necessarily involve the reference to a psychological cause. Interestingly
enough, this motion is never described, in the DC or elsewhere, as the
motion of a living being. A gap seems to exist between the circular motion
of the celestial simple body and the celestial motion performed by the
celestial bodies. In order to bridge this gap, we can opt for one of the
following two solutions.



31 I have argued that ancient atomism is a constellation of positions, and that these atomic items may
be conceived in a number of ways, in chapter 2: “Bodies.”
Aristotle and the Science of Nature
84
1. We may insist that the circular motion of the simple body is not only a
necessary but also a suf¬cient condition for the explanation of celestial
motion. We may argue that the arguments offered at the beginning of
the DC do not merely prove that there is a simple body that naturally
performs circular motion but provide also an adequate account of
celestial motion. In this case, we have to identify the nature of the
simple body with a soul of a speci¬c type. Apparently, Alexander of
Aphrodisias took this view. He argued that celestial motion is the
motion that the celestial simple body performs in accordance with its
own nature. He identi¬ed this nature with a soul of a certain type, a
celestial soul.32
2. We may contend that these arguments do not provide an account of
celestial motion but supply only the material condition for celestial
motion; that is, a simple body that is naturally moved in a circle. In
this case we have to specify the contribution of the soul to the
explanation of the distinctive motion of the celestial bodies. In
antiquity, much time and effort was devoted to detecting this possible
contribution. The ancient interpreters of the DC concentrated on the
case of the heaven of the ¬xed stars. Julianus of Tralles argued that the
soul of the ¬rst heaven is not responsible for the production of circular
motion, but only for its being oriented in a certain direction.33
Herminus agreed that the soul of the heaven of the ¬xed stars is not
responsible for the production of circular motion. But he argued that
this soul causes the circular motion of the celestial simple body to be
continuous and everlasting.34
It is signi¬cant, I think, that the ancient interpreters of the DC who
engaged in this exegetical exercise did not have doubts about the involve-
ment of the soul in the explanation of celestial motion. They all assumed
that celestial motion involves a reference to a psychological cause of a
certain type. Disagreement among them was con¬ned to the precise
nature of the involvement.35

32 Simpl., In DC 380. 29“381. 2. See also Simpl., In Phys. 1219. 3“7. For a presentation of the position
´
of Alexander, see Sharples (1983: 62“6) and Bodnar (1997a: 190“205).
33 Simpl., In DC 380. 1“3. We know virtually nothing about Julianus. For a modern vindication of
the position of Julianus, see Judson (1994: 155“71).
34 Simpl., In DC 380. 3“5. For a presentation of the life and work of Herminus, see Moraux (1984:
361“99).
35 I devoted an entire chapter to the discussion of this exegetical problem in Falcon (2001: 187“241).
chapter 4

The limits of Aristotle™s science of nature




It is worthwhile seeking to attain more understanding regarding
these things, though the resources at our disposal are few and we are
at such a great distance from what happens in the heavens (Aristotle,
DC 292 a 14“17).


remoteness
From the opening lines of the Meteorology the science of nature emerges
as a systematic investigation of the natural world. This investigation is
systematic in the sense that it consists of an inquiry into the different parts
of the natural world in the attempt to discover the explanatory connec-
tions existing between its parts. If this investigation is successful, it does
not provide mere knowledge of the natural world; it provides understand-
ing of it. But this investigation is systematic also in the sense that it
consists in a study of the natural world in its entirety. While Aristotle does
not insist on this point in the opening lines of the Meteorology, he is more
explicit towards the end of PA 1. This logos ends with an exhortation to the
study of the entire natural world: the celestial together with the sublunary
world, and this latter in all its parts, plants and animals included (645 a 4“
7).1 Aristotle takes it for granted that the natural world is constituted by a
celestial and a sublunary part, and argues that the study of each of these
two parts has its own appeal. In this logos, however, the emphasis is on the
study of plants and animals. This gives us opportunities for knowledge
that are not available to us in the study of the celestial world:
(1) Among the substances constituted by a nature, some neither come into being
nor perish for all time, and others share in coming into being and perishing.
(2) It has turned out that we have fewer ways of studying the ¬rst type of


1 I have discussed this passage in chapter 1, “The unity, structure, and boundaries of Aristotle™s
science of nature.”

85
Aristotle and the Science of Nature
86
substances, honorable and divine though they are: (3) for both the starting points
of the inquiry and the things we would like to know about present very few
things to perception. (4) We are better supplied with opportunities for
knowledge about perishable plants and animals because we live among them: (5)
for much can be learned about each kind if one is willing to undertake the
appropriate labor (PA 644 b 23“32).
In this passage Aristotle is not content to say that the study of the celestial
world is more dif¬cult. He also shows a remarkable amount of pessimism
regarding the possibility of knowledge about the heavens. He acknow-
ledges the existence of an informational gap affecting the study of the
celestial world “ clause (2) “ and provides remoteness as the reason for this
gap “ clause (3). Admittedly, Aristotle says very little about remoteness in
clause (3). He is content to claim that the celestial bodies are perceptually

<<

. 3
( 5)



>>