1.There are, among things that can exist, degrees of value (goodness, nobility).
2.The assignment of degrees of value consists in comparing things as more or less similar to a possible thing of maximum value.
3.Therefore, there is a possible being of maximum value. (From 1, 2)
4.A being has maximum value if and only if it has maximum existence. (Convertibility of goodness and being)
5.Therefore, there is a possible being M of maximum existence. (Conclusion of the first part)
6.There is a necessary being (or plurality of necessary beings) N that is, in every possible world, possibly the cause of all existing things in that world. (From the Second and Third Ways)
7.Necessarily, if x is possibly the cause of the existence of y, then x has existence to the same or greater degree than y does. (Aristotle’s Alpha the Lesser principle)
8.Therefore, in every possible world w, N has the greatest degree of existence of anything that exists in w. (From 6, 7)
9.A being of maximum existence exists in some world w. (From 5)
10.Therefore, N has maximum existence in w. (From 8, 9)
11.A necessary being has the same degree of existence in every world.
12.Therefore, N has maximum existence in every world. (From 10, 11)
13.Therefore, N has maximum value in every world. (From 4, 12)
 
There is a potential problem with this reconstruction: step 5 (conclusion of the first half) isn’t strictly necessary. If N didn’t have existence to the maximum possible degree, then there would be a world w containing something with existence to a higher degree. But, then, N could not be the cause of that thing.

Perhaps we could take 1-5 as merely establishing the possible existence of a being of maximum value, which provides some independent support to the conclusion 13. But it would be nice to find some way to make 1-5 directly relevant to 6-13.

The heart of the Fourth Way, as I interpret it, is a simple argument:

1. In every possible world, God is the cause of the actual existence of all other actually existent things. (From the Second and Third Ways).
2. The cause of something's existence must be at least as noble, true, and good as the thing itself. (Aristotle's principle from Metaphysics 2)
3. Therefore, God is a noble, true, and good as anything could possibly be.

 
Why think that the Aristotelian principle is true? And what does it mean for something to be truer, nobler, better, or “more existent” than another? We have already seen an appeal to the proportionality of causes in the First and Second Ways. A cause must “have” what it “gives” in some way, either “formally” (by literally have the same characteristic) or “eminently”. Applying this to existence, we could say that the cause of the existence of x must exist either in the same way as x or in a “higher” way.
 
What do “higher”, “nobler” or “better” mean in this context? They must refer to a real, objective hierarchy, not just to our subjective preferences or interest. They must mean something like: having a wider array of powers and capabilities. As Kenny suggests, this corresponds to something like having a more powerful set of cognitive capacities. Despite what Kenny says, St. Thomas’s argument does not depend on everything’s lying on a strictly linear order. It’s okay if some pairs of things are not comparable in their degree of being, so long as there are some things that are strictly nobler than everything else (a common peak in the “Mountain of Being”).
 
I should say something briefly about Aristotle and Aquinas’s notion that some beings are “truer” than others. As Aquinas explains, “true” can be used in two ways: to refer to a cognition that corresponds to some reality, and to a reality that corresponds to some cognition. My idea of a triangle is true if it accurately represents triangles. A triangle is “true” (as a triangle) if it accurately corresponds to the definition of triangle. What would it be for a thing to be “truer” than another thing (as an existing thing)? It might be that it corresponds more accurately to the full, unqualified concept of Being as such. If so, the more nearly perfect or the nobler a thing is, the truer it is.

Like the Third Way, the Fourth Way consists of two principal parts. The first part establishes that there is something that is maximal in being, goodness, “truth”, and nobility. The second part reaches the conclusion that this maximal being is the cause of the being (and goodness, etc.) of all finite things. As in the case of the other Ways, St. Thomas does not suppose that he has yet proven the unity or uniqueness of God. So, we should really say that the first half involves the claim that there is some thing or things maximal in being, and the second half the claim that this thing or things is or are the cause(s) of all other things. For the sake of grammatical simplicity, I shall mostly ignore this important qualification.
 
The Big Question in interpreting the Fourth Way is this: just how Platonic is the argument? In his dialogue Phaedo (100a1-101a5), Plato has Socrates argue that he has discovered that the true cause of the beauty of things is something called Beauty Itself. Similarly, there is Justice Itself, the cause of the justice of all just things, Goodness Itself, Equality Itself, and so on. These are the so-called “Forms” or “Ideas”.
 
Anselm offered what seems to be a purely Platonic argument that parallels the Fourth Way in his Monologion. Anselm writes in Chapter 1:
 
“Necessarily, all good things are good through something, and this something is understood to be the same thing in each of the various good things…. And who would doubt that that through which all good things are good is a great good?.... That through which every good thing is good is good through itself…. The one thing that is good through itself is the one thing that is supremely good.”
 
Boethius makes a very similar argument in The Consolation of Philosophy (Book III, Chapter X). Is this exactly the argument Thomas Aquinas is making?
 
I think not, for two reasons. First, the steps of Aquinas’s argument are different. First, he establishes that there is a supreme being, and then he argues that this supreme being is the cause of other beings, which is the opposite of the Plato/Boethius/Anselm order. Second, he explicitly rejects in many places (following Aristotle) the argument from Many to One on which Plato/Boethius/Anselm rely. Human beings, for example, are each made human by his or her own human form, not by a Platonic Idea. It’s true that there is a kind of archetype of humanity (a divine idea) that was involved in our creation, but the divine idea is involved in God’s efficiently causing us to exist, which seems significantly different from the way that good things are supposed to be good “through” the Idea of the Good.
 
Kenny (in chapter 5 of The Five Ways) is pretty good on all this. He brings out the way in which a “common nature” for Aquinas is not some separate thing but rather a way of explaining the commonness of the members of a species in terms of some intimate relationship among their forms—namely, that the forms are not individual in themselves but only through their involvement with prime matter.
 
So, I think it is important to look closely at St. Thomas’s source: the end of chapter 1 of Book 2 (Alpha the Lesser) of the Metaphysics. I am also relying on some interpretive suggestions by Michael Augros in his paper, “Twelve Questions about the Fourth Way.” (The Aquinas Review, volume 12, 2005) In the Metaphysics, Aristotle argues that if something is the cause of all true things, then it must be supremely true. If we translate this into more familiar language about existence and nobility, we could say that if something of the existence of all existing things, then it must have supreme existence (likewise for nobility, goodness, and so on). This fits well with the place of the Fourth Way: we have already established that there is one existing thing that causes the existence of all other existing things (Second Way). Moreover, we know that this thing exists necessarily and causes the existence of everything else in every possible world (Third Way). The Fourth Way adds to this the conclusion that this necessary first cause must have supreme existence (nobility, etc.).
 
If this is right, then the second part of the Fourth Way is really the crucial part. The first part is simply designed to make the ultimate conclusion more plausible, by suggesting that there is something (at least, in the realm of possibility or potentiality) that has maximal existence. The second part assures us that this supreme thing actually exists, and, given the Third Way, actually exists and is supreme in existence in every possible world.
 
Alternatively, as Augros suggests, we could take the first part as simply arguing that there must be something that is, de facto, the highest thing on the scale of existence among the actually existing things. This need not, at this stage, be identified with the greatest possible being. Then the second part establishes that this de facto greatest thing must be the cause of the existence/greatness of all other things in all possible worlds, and so it must be the greatest possible being.
 
If this is right, then the standard translations of the crucial principle of the second part of the Fourth Way are misleading at best. Here’s the Latin original: “Quod autem dicitur maxima in aliquo genere est causa omnium quae sunt illius generis.” Kenny translates it as: "Now whatever is most F is the cause of whatever else is F.” The Leonine translation: “Now the maximum in any genus is the cause of all that is in that genus.” These make the principle a generalization of the form: all things that are maximal in a genus are the cause of the rest of the genus. But the Aristotelian principle in Metaphysics is the converse of this: all things that are the cause of the rest of the genus are maximal in that genus.
 
We should try to fit our translation to this Aristotelian principle, and I think it is possible, since the ‘quod’ is singular rather than general. Aquinas does not write “whatever” (omnia) is most F, as Kenny supposes, but rather “that which” (quod) is most F (or is supposed to be most F). The Leonine translation is better, if we take it to mean: the maximum in a genus and the cause of all that is in the genus (supposing that each of these exist) are one and the same thing. Ways 2 and 3 give us something that is the necessary cause of all other beings, and the first part gives us something that is a supreme in being, and the Aristotelian principle allows us to infer that the necessary cause of being is a supreme being, which is what we want. (Parenthetically, the presence of ‘dicitur’, ‘is said to be’, is some reason to prefer my reading of the first part over Augros’s: the supreme being of the first part is merely a supposed or hypothetical entity.)