Commentary Aristotle Physics
Now, the reason that the instrument by which the mover causes motion both is moved and moves is that it partakes of both and exists in a sort of identity with what is moved. This is especially evident in local motion, for it is necessary that, from the first mover to the last thing moved, all must touch one another. Accordingly, it is evident that an intermediate instrument is the same as the mobile through contact and is moved at once with it, insofar as it is in union with it. But it is also in union with the mover, because it is a mover—even though, under its aspect as the instrument by which the mover causes motion, it is not immobile.
Accordingly, therefore, it appears from the premises that the last thing moved is indeed being moved, but it does not have in itself a principle for moving either itself or anything else, and it is moved indeed by something else and not by itself.
Hence, it seems to be reasonable, that is, probable (and, in the present case, we do not care to say that it is necessary), that there be a third thing that causes motion but is immobile.
For it is probable that, if two things are joined per accidens and one is found without the other, then the other might be found without it.
But that it may be found without the other is necessary, because things joined per accidens may happen not to be joined; for example, if white and sweet are joined per accidens in sugar, and if white is found without sweet, as in snow, it is probable that sweet be found in something without white, as in cinnamon. If, therefore, it is per accidens that a mover be moved, and something is found to be moved without moving something else, as happens in the last thing moved, it is probable that one may find moving without being moved, such that there would be a mover that is not moved.
From this, it is evident that this argument does not have force in regard to substance and accident, and in matter and form, and in like things, of which one is found without the other but not vice versa; for accident per se exists in a substance, and to matter it belongs per se to have existence through form.
1045. Third, at “so, too, Anaxagoras” (256b24), he proves the same point on the testimony of Anaxagoras. For since it may be that a mover be found that is not moved, Anaxagoras spoke aright when he said that mind is impassible and unmingled. He said this because he posited mind as the first principle of motion, and the only way it could cause motion and command without itself being moved was that it be unmingled, since what is mingled with something else is in a certain way moved when that something else is moved.
1046. Then, at “we will now take” (256b27), he concentrates on the other part of the division, that whatever is moved is being moved by another that is moved per se and not accidentally.
And he disproves this with two arguments.
The first is this. If it is not according to an accident that a mover be moved, but of necessity, and if it can never cause motion unless it is moved, this must happen in one of two ways:
one is that the mover is moved according to the same species of motion as that which it causes;
the other is that the mover moves according to one species of motion and is moved according to another.
He subsequently explains the first way at “I mean that either that which is” (256b31). We say that a mover is being moved according to the same species of motion if, for example, the thing that causes heating is heated, the healer is healed, or something carrying locally is itself being carried locally.
He explains the second way at “or else what is making healthy is in the process of locomotion, and what is causing locomotion is in the process of increase.” These are examples of moving and being moved according to different species of motion.
Then he shows the impossibility of the first way, at “but it is evident” (256b34). For it is clearly impossible that a mover be moved according to the same species of motion. For it is not sufficient to stop at some subalternate species, but one must divide until he reaches the individual species, that is, the most specific species. For example, if someone is teaching, it is not enough for him simply to be taught at the same time, but he must be teaching and being taught the same; for instance, if he is teaching geometry, he must be at the same time being taught it; or if he is the cause of a local motion called “throwing,” he must himself be moved according to the same motion of throwing. This is clearly false.
Then he dismisses the second way, that the mover not be moved according to the same species of motion but that it move according to one species and be moved according to another, such as if it moves with a local motion and is being moved with respect to growth, or if what causes the growth is being moved by something else according to alteration, or if this mover in turn is being moved with respect to some other motion.
Now, it is clear that motions are not infinite either in genus or species. For it was held in book 5 that motions differ in genus and species according to the differences of the species in which motion occurs. But the genera and species of things are not infinite, as we proved elsewhere; accordingly, neither are the genera and species of motion. If, therefore, a mover is necessarily being moved according to some other genus or species of motion, one will not be able to proceed to infinity and there will be some first immobile mover.
1047. But, because someone could say that, when all the species of motion are exhausted, a return will be made to the first species in such a way that, if the first thing taken as moved was moved locally and we distributed all the genera and species of motion to different movers until these genera and species were exhausted, the remaining mover will then be moved according to local motion: in order to exclude someone saying this, he subsequently says that such a return is tantamount to saying that the cause of alteration is being moved locally (he uses this explanation because above, in his example, he mentioned local motion first and alteration last), the same, I say, as supposing from the very beginning that the mover according to local motion is being moved or that the teacher is being taught not only generically, but specifically.
And that this means nothing more, he proves consequently. For whatever is being moved is moved more by the higher mover than by the lower one, and consequently, much more so by the first mover.
If, therefore, the thing posited as being moved locally is being moved by a neighboring mover that is being increased, and it by a mover that is being altered, and it further by one that is being moved according to place, what is being moved according to place will be more moved by the first one moved according to place than by the second one that is being altered or by the third one that is being increased.
Therefore, it will be true to say that the mover according to place is being moved according to place, and the same for every species of motion. Now, not only is this false, because it is seen to be belied in many cases, but it is also impossible. For it would follow that the teacher is learning while he is teaching—which is impossible. For this involves a contradiction, since it is the property of a teacher that he have science, and of a learner that he not have it. Accordingly, it is clear that it is not necessary for a mover to be moved.
1048. He gives a second argument at “still more unreasonable” (257a14), which does not differ from the preceding one except in that the first leads to certain particular inconsistencies, for example, that a thrower would be thrown or that a teacher would be being taught. But this one leads to inconsistencies in general.
Hence, he says that, although it is inconsistent that a teacher be learning, there is something still more unreasonable, for it turns out that every mover is mobile, if nothing is moved except by what is being moved. For it will thus follow that every mover is mobile, if (for example) one says that whatever has the power to heal, or is actually causing health, is healable, and that whatever has the power to build is buildable—which is more unreasonable than that a teacher be learning, for a teacher could have been learning before, but a builder was never built.
Now, this follows in two ways.
For if it be conceded that every mover is being moved with respect to the same species of motion, it follows that a builder is being built immediately (that is, without intermediary) and that a healer is being healed immediately.
But, if it be conceded that the mover is not being moved according to the same species of motion, it follows that we shall finally come to this after passing through a number of intermediates.
And he explains this. If every mover is being moved by another but not being moved immediately with respect to the same species in which he is causing motion, but according to some other species—for example, if a healer is not at once being healed, but is being moved according to the motion of discipline by learning—yet, since the species of motion are not infinite, by thus ascending from mobile to mover, one will at length reach the same species of motion, as was explained above.
Therefore, of these two, one appears plainly impossible, such as that the builder be immediately being built, while the other is seen as a fantasy, namely, that one come to the same thing through a number of intermediates. For it is unacceptable that what is apt to cause alteration is of necessity apt to be increased in size.
1049. Accordingly, having considered the foregoing arguments (257a25), the first of which concluded that this process—that whatever is being moved is being moved by another—must not go on to infinity, and the second of which concluded that not every mover is being moved, we can conclude from all the foregoing arguments that it is not necessary to infinity that what is being moved be moved by another in such a way that it is always being moved by a mover that is being moved. Therefore, it is necessary to stop at some first. However, this first must either be immobile or be moving itself.
But, if we are considering which is the first cause of motion in the genus of mobiles, whether it is something that moves itself or a mobile that is moved by another, it is held as probable among all that the first mover moves itself. For a per se cause is always prior to what is a cause through another. For this reason, the Platonists held that, prior to things that are moved by another, there is something that moves itself.
And we must therefore consider this thing that moves itself and make of this another beginning of our consideration—that is, we must consider how something moving itself is possible.