St. Thomas gives no explanation of the “which all call God/a god” premise in his Five Ways, which allows for multiple interpretations. My suspicion is that St. Thomas left the interpretation open since there are multiple true ones. OFloinn states the common interpretation with maximal elegance by saying STA could have just said “This is what all call God: details to follow”. Dovetailing with this, I’ve also argued that the very thing STA is proving, namely deus, gives him a far broader target to hit than a proof for what an English speaker calls God with its noticeable capital “G” and absence of an indefinite article, since one has to prove a complex of properties to show the existence of God which aren’t necessary to show the existence of a god, but one can prove the existence of deus by merely proving the latter.* St. Thomas certainly thinks that the god is proving the existence of is in fact God, but his word deus gave him enough breathing room that he could take his proof as establishing a being that could be shown to be God in the “details to follow”.
So let’s assume this is how we should read the Five Ways: they prove a god or “something divine” exists and then fill out the details showing how it is not just a god but also God. But why does it immediately follow from the proofs that we get to so much as a god? My claim is that all of them conclude to the existence of a sort of cause that no one can reasonably call natural, and I’ll show this for the first three arguments.
The First Way is the most manifest way to conclude to a divinity since, as STA understands things, the mobile is natural and vice versa, and so to conclude to something that acts without being in motion is ipso facto to call it a supernatural cause. Notice that St. Thomas is claiming more than that the first mover in the series is immobile with respect to the peculiar motion it causes, since this is true even of, say, the way a bicycle axle is immobile with respect to the wheel or the way an inertial reference frame is immobile with respect to the things moving in it. He’s saying that the mover conveys an actuality without having to change along with the potency it acts on, so, for example, it is like an axle that could be a cause of the wheel moving without having to be in motion along with the wheel, or an inertial reference frame whose boundaries would not change even relative to the mobile thing in it, or like a magnetic field that wouldn’t change and be in flux along with the ferrous objects it tugged on. As an example of natural activity, this is completely crazy, and imagination fails out in trying to visualize the sort of mover the First Way is arguing for. The motions one sees around himself all presuppose some kind of supernatural activity.
While the First Way arrives at the supernatural by way of negation of natural motion; the Second Way arrives at transcendence by way of agent causality considered formally. Agents can never be the agent cause of what they they have in common with their effects, i.e. a match lighting a candle doesn’t cause fire in the sense of what is common to the match and the candle but only the fire that is peculiar to the candle. Because of this, a first efficient cause is one that has nothing in common with the agents we find around us. If the things around us exert agency by being massive or spatial (like the counterweight on a trebuchet), then the first agent cause cannot have either mass nor spatial extension; if they exert agency by having mass-energy or momentum then the first agent cannot have these things. I insist that this is formally contained in the idea of an agent cause – it is not some additional premise about agency. This is why equivocal causes are described as formally causal whereas univocal ones – which have something in common with their effects – are seen as only causal instrumentally or “materially” (not in the sense of material cause). So here too we arrive at the supernatural, though by a way not tied to the definition of nature but by way of our need to negate whatever is efficiently causal in nature – mass, energy, momentum, etc.
The Third Way concludes to a necessity beyond the necessity that allows for things to exist always in time. It is crucial here to recognize that STA is staring with the generated, that is a thing that did not exist at one time. What this means is that if you took a picture of all the things in the universe at once, then if everything is generated all you would have to do is count back in time to a point where nothing existed. (Please, fertheluva Pete, stop seeing a quantifier shift in this proof. There’s nothing of the kind) All the first stage of the proof concludes to is that there must be something like matter or energy that is not generated, and these things count as “necessary beings” in the context of the proof (please, please, also stop calling the third way the proof from contingency. Matter, the universe, physical law, and logical structures are all non-contingent, but are obviously not God.) What St. Thomas argues is that this sort of necessity is derivative, presumably because one could not establish the necessary existence of mass or energy simply by considering mass or energy as such. St. Thomas calls this the difference between what is “necessary by another” and what is “necessary by itself” – and I take him as understanding the second sort to be “necessary by definition”, i.e. something whose existence you could be certain of just by knowing its definition. But nothing in nature is like this, so you need something outside of it.**
*the Latin of STA’s time did not use capital letters as honorific nor did it need to use indefinite articles.
** Note the fascinating relation between the Third Way and the Ontological argument. The OA says, in effect, we know the definition of God and can conclude from this that he exists. St. Thomas denies that we can know this sort of definition, but he uses this very premise in the Third Way: we know the things of nature, and all the things of nature are such that their definitions do not give us information about whether they exist. Therefore, none of them can be necessary by definition or in themselves, and so natural necessity must be from another.
(Looking at the word count, I notice this post is getting way too long. But interestingly, it is almost exactly as long as the article containing the Five Ways. huh.)