To: Plinio C.
From: Geoffrey Klempner
Subject: Aristotle on the unmoved mover
Date: 14th April 2010 13:12

Dear Plinio,

Thank you for your email of 5 April, with your essay for the University of London BA Greek Philosophy: Aristotle module, in response to the question, 'Discuss Aristotle's proof of an unmoved mover. How does it cause motion?'

This is an impressive piece of work. I have some observations to make, although I would not necessarily regard these as criticisms of what you have actually written.

You conclude that Aristotle's proof 'is based on two dubious premises',

(2) It is impossible for there to be a beginning or end of time

(3) Time is the measure of change (= there cannot be an empty time),

'and an unwarranted inference',

(4) therefore there must be something(s) in eternal motion.

Your argument against (2) -- which is actually the position which Swinburne (whom you cite) argues for in his book 'Space and Time' -- is that there is no contradiction in regarding time as a 'set [i.e. series] of temporal instants' which has a first member. Your argument against (3) is based on a thought experiment by Shoemaker which Swinburne considers, and whose conclusion he accepts.

According to Swinburne there is not the same objection to empty time as one can raise about the hypothesis of an empty space. Positions in time are determined by their relation to prior and posterior events, while positions in space require for their determination actual bodies which occupy space.

You say that the inference is unwarranted because there might just as well be 'an eternal temporal succession of different things that move, this succession being necessarily overlapping such that there cannot be gaps'. However, you are prepared to consider the 'charitable' view that for Aristotle, there would be no adequate reason for this necessity, while mere contingency would not suffice for the argument.

I'll comment on this first. This bears a similarity to Melissus' view that, on the assumption that the universe is temporally infinite, it must be spatially infinite. The argument is that if the universe were spatially finite, then it would be capable of growing or shrinking, and anything that can grow or shrink is capable of shrinking to nothing. So, similarly, if there are different finite things that move whose movements overlap, then in the absence of any adequate reason the permanency of this overlap cannot be relied upon. It cannot be 'necessary'. But in that case, contrary to what you assert, the necessity in the lack of gaps IS more implausible than the necessity that there should be at least one thing in eternal motion.

The problem with identifying (2) and (3) as the questionable premisses in Aristotle's argument (and, by inference, the ONLY questionable premisses) is that you would then be asserting, in effect that (modulo our disagreement over 4) any philosopher who holds (2) and (3) is committed to the existence of an Aristotelian unmoved mover. That is not necessarily a good result.

I am in fact rather uneasy with the idea that you can just postulate time as a finite 'series of temporal instants' and be done with it. Formally, you can define any series you like and call it 'time'. Why should there be only one series of temporal instants, why not two or an infinite number? Why not branching time, circular time etc. etc. All these have been considered in the literature. There is no formal objection. The problem is relating this to what we understand as 'time'. So my view on this would be agnostic. I can sympathize with Swinburne's view. I do agree with him insofar as I don't see why physicists should necessarily have the last word in defining time.

As for (3), it is notable that Shoemaker's thought experiment postulates the very thing that Aristotle is arguing for, laws of nature which we have observed to operate in such a way that we can predict recurring gaps of 'empty time'. Of course, it would be anachronistic to even attempt to map explanation in terms of deduction from laws onto Aristotle's account of what constitutes a physical 'explanation'.

Which brings us to the second part of the question. It is admittedly obscure how the unmoved mover brings about the eternal motion of the celestial bodies. Here, we really do need to make a distinction between Aristotelian physics and modern physics.

Let's stick with the question of efficient causation, because without this the argument equivocates on the notion of a 'mover'. The unmoved mover makes the celestial bodies move. Without this push they would not move. Since Newton, we have become accustomed to thinking of change in motion (acceleration and deceleration, change of direction) as requiring a force which accounts for it, while continuation of motion as such does not need to be 'explained'. In the materialist picture of the universe, God sets everything up, gives it a push, and off it goes.

However, I can well understand that Aristotle would have thought that this is a rather superficial view. (In this connection, one might also consider Descartes, who in many ways reacted against Aristotelianism, holding that the deity not only creates the material universe but continually maintains it in existence through his will.) In addition to any explanation in terms of final causation -- the unmoved mover's aim in maintaining the celestial bodies in perpetual motion -- there must be something that provides an adequate explanation, at every instant of time, of how this aim is realized.

Aristotle is not thinking (as we naturally would) of any 'mechanism' here. Rather, the explanation must appeal to the very nature of the celestial bodies as things which do not themselves contain their own self-sufficient cause of motion (as e.g. men or animals do). I agree with you that the traditional story about their 'love' for the unmoved mover doesn't make a lot of sense. I would see this rather by analogy with the case of intentional action: The celestial bodies are not the unmoved mover. But they are, in effect, the unmoved mover's 'body'.

All the best,

Geoffrey