The argument which follows is a recapitulation of a tradition of arguments found in Aristotle’s Physics (VII.1 and book VIII), as well as Aquinas’s “First Way” or proof from motion.
Here is the argument in nuce, in its Aristotelian form.
1. Every mobile in motion is put in motion by another mover.
2. Every mover (moving another and itself in motion) is necessarily put in motion by a first mover.
C: Every mobile in motion is put in motion by a first mover.
In what follows, we will first expound the ancient argument on its own terms. Second, we will transpose any perennial elements into a terminology more familiar to contemporary cosmology. This philosophical key change is enabled by an Aristotelian-Thomistic philosophy of energy in physical systems. In short, the free energy required for any given motion cannot be derived as such from a physical system; the causal dependence required engages the same sort of regress argument proposed by the ancient argument and must be terminated in the same fashion.
(1) Every mobile in motion is put in motion by another mover. — Aristotle or Aquinas argued for this premise in at least three ways. A first approach was an inductive argument. The second proof argues that movers must be in act, while mobiles in motion are as such in potency, and thus cannot be self-movers in the required sense. A third argument is based upon the divisibility of things in motion and is the one proposed by Aristotle in Physics, VII.1 and defended by Aquinas in Summa contra Gentiles, I.13. Sometimes called the “moving part” argument, its central idea is as follows: Because every mobile in motion is divisible, but nothing divisible can move itself as a whole first and foremost, then no mobile in motion can move itself as a whole first and foremost. Thus, every such mobile must be put in motion by another. It is this last strategy which will be updated below.
(2) Every mover (moving another and itself in motion) is necessarily put in motion by a first mover. — The argument defending this premise is rather standard by Aristotelian terms. However, since the third sort of argument for premise (1) is at issue here, the defense of premise (2) as found in Physics, VII.1 is of an odd sort. This argument supposes that the entire universe is, as it were, one single mobile system. This concatenation of mobiles and movers into a single whole, of sorts, is argued to be insufficient to cause its own motion. Thus, some mover causally prior to this whole is required. For if it were itself of the same order as those group of mobiles and movers being put in motion, the regress argument would not be terminated and no causal efficacy would be found.
The conclusion to the argument is that every mobile in motion is put in motion by a first mover. The argument, thus baldly stated, tells us precious little about the characteristics of such a mover. Is it some sort of substance? Is there only one or are there many? What other features does it have? The argument does not say, unless further premises are added. However, our aim is not to explain this ancient argument on its own terms or to defend those terms in a contemporary context. Rather, how can such an ancient terminology be transposed into modern terms as such? Can the argument be restated in philosophically suitable concepts such that physical naturalism can be defeated?
The transposition can be effected because the modality of the ancient argument does not depend for its probative force upon any specific cosmological model, but rather upon principles that any model cosmology must assume, principles explicable in terms of potency, act, subjects of motion, and the agency necessary for their change. The chapters above have explore such a transposition in other areas of cosmology. It remains to complete this task with regard to the physical agency necessary and sufficient for cosmic motion.
Here is the translation key:
a) mobile in motion = a physical system in motion
b) put in motion by another mover (not in motion per se et primo) = a physical system cannot be a closed system in motion (no perpetual motion machines)
c) put in motion by a first mover = physical systems cannot causally nest spatiotemporally in a closed whole
Some initial explanation is required. The first term (a), notes a certain equivalence between the abstract mode under which the ancient argument treats mobiles and the abstract character of a physical system in motion. That is, on the one hand, Aquinas notes that the argument in Physics, VII.1 treats mobile bodies in abstraction from their specific natures, “denuded,” as it were, of their specificities and thought of only as mobile quantity. On the other hand, modern physical systems are frequently characterized only in terms of quantities such as mass or energy, prescinding from their specific characteristics as living or non-living, compound or elemental, etc. Thus, I claim that a mobile in motion is a physical system in motion, for the latter is merely the encoding of act-potency relationships of a mobile in motion through the formalism of mathematical physics.
An objection to this transposition is that, on the Aristotelian-Thomistic approach, mobiles are as such per se unities: substances or wholes. By contrast, physical systems need not be per se unities. For instance, the entire Solar System can be treated as if it were one thing in a calculation. To respond to this, note that, while it is true that a physical system can be defined in any arbitrary manner, this does not guarantee that such a definition is useful or relevant to nature. The mathematical physicist, in holding the mirror up to nature, ought to hew closely to the lines defined in the natures of things. Thus, for example, one could define a “system” whose parts happened to break the conservation of energy. However, if such a “system” is defined in a way closer to the nature of things, no such untoward results arise. That is, physical systems are in their truest form when they pay heed to what are true per se unities or mobiles in the cosmos.
As for (b), its explanation follows closely upon arguments such that the laws of nature encode the cursus naturae and the act-potency relationships in mobiles and movers. A mobile cannot be in motion per se et primo; these conditions deny, first, that the mobile is in motion per accidens and, second, that the mobile is in motion due to its parts. That is, in order to circumvent the motor causality principle, a mobile would have to be truly in motion precisely as a whole and not in virtue of any of its parts. For a physical system, this would mean that neither its parts nor its surroundings are the source of the free energy required for its motion (whether physical, chemical, or biological). Any spontaneous process requires free energy (i.e., Gibbs free energy). Any non-spontaneous process by definition has a source of energy outside the system which produces the change in question. In either case, some source outside the system is required, either to produce the conditions of spontaneity or “work uphill” in the absence of free energy. Another way to make this point is by noting that no perpetual motion machine is possible.
Regarding (c), this transposition defines what it would mean for physical systems to be self-sufficient as cosmological naturalism demands. That is, naturalism requires that physical systems can causally nest spatiotemporally in a closed whole. In other words, the naturalist requires that the phase space of the universe is self-sufficient to interpret the physical causality of the cosmos. If this is false, then this phase space—perhaps in how frame space is defined (Chapter #)—must also reflect the possibility of higher-order causality. That is, if the argument goes through, a corollary to the argument is that the cosmos is not a closed system, but neither is it an open system in a univocal way in which a physical subset of the cosmos can be an open system.
The transposed argument, then, is as follows:
1´. Every S is put in motion by S´.
2´. Every S´ is put in motion by M*.
C´: Every S is put in motion by M*.
In this transposed argument:
S = a physical system in motion (M); conceptually considered as closed, but in fact an open system, because perfectly closed systems as parts of the cosmos are impossible
S´= a physical system (=S) that moves another (= M´) in virtue of itself being in motion (M); thus, S´; it is also open, otherwise it could not put another in motion
M´ = a physical system reponsible for the motion of another system qua open
M* = what is reponsible for the motion of a system but is not itself either S, S´, or M
Based upon the above translation, this transposed argument, as to premise (1´), is subject to elucidation in two ways. The first is a negative proof by contradiction. The second is a positive proof through the requirements of free energy for any system in motion.
First, consider the system S that is M, and let it be a system in motion but by no M*. Thus, it is S* = a physical system in motion that is a closed system. Now, S* can be in motion due to no part of itself, for then that sub-system would be responsible for the motion of S*, meaning that S* is also S, which is contrary to the supposition. However, S* must have parts, for any mobile system is a quantifiable whole, even if it is a natural miminum quantity. Furthermore, S* must have parts, for even its motion has parts due to the various measurable components of physical motion (time, momentum, energy, etc.). Consider any one of those parts, S*-x. It is in principle capable of motion or rest, otherwise S* could do neither. Thus, if it were at rest, then S* would have to be at rest. Thus, S* is not in motion as a closed system, as it is really subject to partitioning into sub-systems S*-x upon which it is dependent. As Aquinas notes, “Thus, just as this conditional is true: if the part does not move, the whole does not move—so also this conditional is true: if the part is not, the whole is not.” Extension is a necessary condition for being mobile, but this means that the mobile whole precisely insofar as it is in motion cannot be the first sufficient ground of its own motion by the very fact of being materially dependent upon its parts. Matter as such is not explanatory of the actual existence of motion. Likewise, being a quantum is a necessary condition for being a physical system in motion, and thus any such physical system is dependent upon its parts and cannot be put in motion of itself (S* is impossible). The possibility of ±ΔE for any part of the system reveals the dependency of the whole upon its parts. (Any truly least quantum part is only definable or able to exist as part of a system of such parts, and so quantum “entanglements” are no obstacle here.)
Second, a proof is possible from the notion of Gibbs free energy. That is, any S or S´ is of the sort that it is in motion (M) and some are systems that move other systems (M´). That is, in another respect, S´ is itself as S in the relevant sense. However, as noted above, any physical system in motion is such either because it is undergoing a spontaneous motion or a non-spontaneous one. Now, the non-spontaneous motion of a system is by definition caused from without. As for a spontaneous motion, this condition must be produced in a system S from without. This is due to the fact that a net gain of free energy available to do work that originates from the system S in question alone would make possible a perpetual motion machine. However, by inductive arguments, these are impossible. That is, any such spontaneous increase would violate the conservation of energy. In other words, the motive source of Gibbs free energy cannot be self-supplied by S but must be from an S´—just as Aquinas stated: “de potentia autem non potest aliquid reduci in actum, nisi per aliquod ens in actu, sicut calidum in actu, ut ignis, facit lignum, quod est calidum in potentia, esse actu calidum, et per hoc movet et alterat ipsum.” For the relevant act corresponds to available free energy, while the relevant potency corresponds to the capacity of the system in need of the very same free energy for motion.
Now, as to premise (2´), we first imitate the ancient argument. Let every S´ and S be composed into a supersystem, C, namely, the entire cosmos. This C is in part composed of systems S´as well as those of S. Some parts of it move other parts; some parts are merely moved, others movers. But, since every S´ is, in another respect, an S, or a physical system in motion (M) conceptually considered as closed but in fact an open system, because perfectly closed systems as parts of the cosmos are impossible, then this means that C, in order to be in motion, must be put in motion by some other S´. However, and first, by hypothesis, there are not other such systems (for C is a cosmos and thus the totality of ordered movers and moved substances); it would result that C is not in motion, but this is empirically false. Second, any such S´ being added would—eventually—make C infinite in extent. But no such system can be in motion.
Why? (1) Because no cosmologist countenances truly infinite spatial extents of subjects with energy or fields of such sort (in a physical sense); (2) because any physical system is subject to entropy, and the degradation of entropy is temporally finite. The composition of parts of the cosmos in space (1) or in time (2) and their active-passive / causal-effectual relationships are subject to the same limits as enunciated in the defense of premise (1´).
Thus, in order for C to be in motion, there must be some source of motion possible, reponsible for the motion of a system such as C, but is not itself either S, S´, or M. Let it by M*. Thus, every S is put in motion by M*. Every mobile in motion in the cosmos is put in motion by a first mover of the cosmos. Now, what are the characteristics of M*? We cannot say immediately. Further information is needed.
John G. Brungardt 