Five Proofs of the Existence of God
Edward Feser
(Ignatius, 2017)
330 p.
Arguments for the existence of God are of perennial interest, and over the centuries many different lines of reasoning have been proposed. You have your cosmological argument, your ontological argument, your argument from design, your argument from moral objectivity, your argument from Bach – a personal favourite of mine — and many others. As a group, they proceed from a variety of different premises, and, naturally, particular arguments are convincing only to those who accept the relevant premises.
Some of the arguments are probabilistic and lead to the conclusion that God’s existence is likely; arguments proceeding from observations of particular features of the world, such as the argument from design, are of this kind. But there is a class of arguments which are deductive in form and proceed from premises which are difficult to deny; these arguments claim to be demonstrations of God’s existence — “proofs”, to use Feser’s word. In this book, Feser puts forward five arguments of this type and defends them against critique.
*
Before proceeding to outline the arguments in question, it might be worthwhile to clear the ground of a few possible misgivings. There are some who see arguments for God’s existence as quixotic, and this can be for different reasons. If God – an all-powerful, all-good, eternal being who created the universe – exists, wouldn’t it be obvious? Or, from a different point of view, doesn’t belief in God’s existence properly belong to faith, and so isn’t it pointless, or even an impiety, to try to prove it by reasoning? Or, from yet another line of approach, aren’t philosophical arguments like this futile? Haven’t people been arguing these questions for centuries with no clear winner?
To the first point, the Christian tradition denies that God’s existence is obvious, in the sense of self-evident. St Paul does say that His existence can be known from the things that have been made, but doesn’t spell out how, and the Catholic Church actually holds as a dogma that God’s existence can be soundly demonstrated, but doesn’t specify what the sound argument is. In consequence, the Church denies that belief in the existence of God is something necessarily held on faith; rather, she contends that we can know God exists, and that this knowledge is part of the “preamble to faith” (even if, in practice, many people believe in God intuitively or on the authority of the Church). And as to the futility objection – well, the same could be said for many, or most, philosophical arguments. Either you have the taste for this sort of thing or you don’t.
*
The five arguments Feser presents are not original with him, but are drawn from the long tradition of philosophical reflection on this topic. He calls them the Aristotelian proof, the Neo-Platonic proof (from Plotinus), the Augustinian proof, the Thomistic proof, and the Rationalist proof (from Leibniz). It should be obvious, therefore, that these are not just the “five ways” of St Thomas Aquinas.
He devotes a chapter to each argument, and each chapter has a consistent structure. He begins with a discursive argument introducing the main premises, clarifying the relevant concepts, and giving the form of the argument which establishes the existence of some entity or reality X. He then considers what the nature of X is and why it is reasonable to describe this as “God”. In the next stage, he formalizes the argument into a set of specific logical steps (between 27 and 50, depending on the case) linking the premises to the conclusion (“So, God exists”), making the shape of the argument as clear as possible. Finally, he entertains and answers a variety of objections that have been raised against the argument over the centuries.
To attempt to describe the arguments in detail here would reproduce the book, so I’m not going to do that. But I will sketch the basic idea of each argument, and then focus on a few particular points that interested me.
**
The Aristotelian argument begins from the premise that change occurs, and argues that this cannot be intelligible unless there exists, here and now, an unchanging changer (or, in the slightly more technical Aristotelian language, an “unactualized actualizer”) to ground it all. The argument goes like this: a given object (say, my drink) can change in various ways: it can become warmer or cooler, spill on the table, be poured into another cup, be chemically transformed by interacting with other objects, and so on. It could not change in these ways unless the change was caused, and the cause must be something outside it. Now, this thing causing the change must itself be either changeable or unchangeable. If the former, then the argument repeats; if the latter, then we have arrived at some entity which is unchangeable but causes the changes in other things. It is crucial to the argument that this causal series is not a temporal one (not an “efficient cause” in the Aristotelian sense) but what Feser calls a “hierarchical” series, one existing here and now and operative simultaneously. For example, my drink is cooling down, and this is because of heat exchange, here and now, with the surrounding air. This heat exchange is just energy transfer between the molecules of my drink and the molecules of air, and this energy transfer occurs because, here and now, there is an electromagnetic force which exists and acts between the molecules when they become sufficiently close to one another. And this electromagnetic force is, here and now, a result of the electroweak quantum field filling space, and this quantum field fills space, here and now, because… well, we don’t know, but we do know that the quantum field itself is contingent, exists in space and time, which are themselves contingent realities, and that they must be grounded in some deeper reality. Down, down we go, and the argument contends that we cannot go on to infinity, but must ultimately arrive at some level of reality which can “cash out” this hierarchy of simultaneous causes or conditions for the cooling of my drink. This is the unchanging changer (or unactualized actualizer).
Why identify this “thing” (if it is a thing) with God? There follows a set of arguments to unfold the nature of this unactualized actualizer, and these arguments are neat (in the sense of tidy and tight) and also neat (in the sense of interesting). Feser argues that this “thing” must be outside space and time, immaterial, unchangeable, good (in the sense of ontologically good – fully itself, all it can be – rather than morally good), omnipotent (because it has power to cause all changes or actualize all possible realities), intelligent (in the sense of containing in abstract form the patterns of all it actualizes), and one. The validity of these subsidiary arguments are to be evaluated separately from the argument for the mere existence of the unactualized actualizer, but enough has been said to justify the identification of this reality with what we normally call God.
Feser considers numerous objections to the argument, from Hume, Kant, Russell, quantum mechanics, and others, and defends his formulation of the argument against them.
*
The Neo-Platonic argument begins from the premise that things of our experience are composed of parts, and argues to the existence of a reality which is itself non-composite or, to use the philosophical word, simple. Plotinus called it “the One”. In a way similar to the previous argument, the claim is that whenever we have a composite entity we have a hierarchical series of simultaneous causes responsible for combining those parts here and now, and that this hierarchical series cannot go on to infinity but must terminate in something which (by the nature of the argument) is not itself composite.
This is not just an argument about patching together material bits to make a composite object, but is a philosophical argument that applies to metaphysical parts as well, such as (to invoke Aristotelian concepts) form and matter, if they exist.
In classical theism, simplicity is the hallmark divine characteristic. God has no parts because, if He did, the particular way in which they were disposed would require an explanation outside God, and God would depend for His existence on something prior to Himself (in which case that thing would be God). But something which is absolutely simple is one, immutable (because there is nothing about it that can change), eternal (because there is no cause prior to itself that can bring it into or out of existence), uncaused but itself acting causally on all things, and purely actual.
*
The Augustinian proof is quite different. It concerns the nature of abstract entities like numbers, propositions, and universals. From the premise that at least one abstract object exists, the argument proceeds to the conclusion that an intellect exists, and that the features of this intellect justify calling it the divine intellect. Feser begins by laying out a battery of arguments (10 of them) to support the premise that at least some abstract objects exist, and then argues that this implies the existence of a mind in which these objects exist. But this mind must not be contingent on any material reality (since the abstract objects are themselves independent of material reality). Since at least some abstract entities seem to have necessary truth (like mathematical propositions), they must exist in a mind that is not contingent but exists necessarily. And so, step by step, the argument proceeds to the conclusion that the intellect in which these abstract objects exist is omniscient (containing all possible true propositions), eternal, etc., and that therefore it can be reasonably identified with the divine intellect.
*
The Thomistic proof proceeds from the distinction between essence — what a thing is — and existence — that a thing is — and argues to the existence of an entity in which essence and existence are not distinct. The essence of this entity is to exist, or, rather, the essence of this entity is existence itself. As such, this entity exists necessarily and all other things exist by participation in it.
Essence and existence must be distinct in most things because we can know their essences without knowing whether they exist (ie. a unicorn), because they are contingent (which would not be possible if an essence automatically implied existence), and because (which requires separate argumentation) there can be only one thing in which essence and existence are not distinct, and yet we observe many things.
A thing can begin to exist only by the action of another, for a thing which does not yet exist cannot act on itself or anything else. But whatever causes a thing to exist must either have its existence from itself or from another. All contingent things have their existence from another, and those others must in turn derive their existence from others, and so on. This is another “hierarchical causal series”, happening here and now, not spread out temporally. But then nothing can exist unless this series of causes terminates in something which has its existence from itself, something that exists necessarily, the nature of which is simply to be. But this is a thing in which essence and existence are identified. Therefore the existence of any individual contingent being requires the existence of something that exists necessarily.
Like the other arguments in this book, this is a fully metaphysical argument. That essence and existence are one in God is the classic Christian position, derived from Scripture (“I AM WHO AM”) and fully flowering in the thought of Aquinas. He proceeded to argue that God, so conceived, must be one, the cause of the existence of all things at each moment, itself uncaused, and purely actual or fully realized.
*
The Rationalist proof, which is the one of most recent origin, is based on the principle of sufficient reason. It begins from the premise that everything has an explanation and argues to the existence of a necessary being. This is an interesting one because it requires fewer metaphysical commitments than some of the other arguments, and relies on an uncontroversial principle that underlies all rational inquiry whatever. (Of course, the principle becomes controversial if it seems to imply God’s existence!)
The form of this argument will be familiar to any parent whose children are fond of asking “Why?” to whatever reasons are given them. The PSR (as it is called for short) states that “there is an explanation for the existence of anything that does exist and for its having the attributes it has”. It does not require that we know the explanation, or that the explanation be a deterministic one (ie. quantum mechanical causes do not undermine it). Feser does a good job explaining the incoherent mess one gets into if the truth of the PSR is denied.
The existence of contingent things must have an explanation, given PSR. If this explanation is in terms of another contingently existing thing, then it too must have an explanation, and so on. But this series of contingent things itself requires an explanation, and this passing of the explanatory buck can only terminate in a non-contingent thing – that is, a necessary thing, the existence of which, by its nature, requires no explanation. Absent this necessary being, we can never provide a complete explanation for anything.
The argument then proceeds along lines growing increasingly familiar to conclude that this necessary being must be one, the cause of all things, purely actual, immaterial, etc., and so reasonably identified with God.
**
Again, I have done no more than sketch the arguments here. Feser unfolds them in considerable detail, and devotes many pages to answering objections. I hope that I have summarized them with adequate faithfulness, although it has now been several months since I finished the book.
I find these arguments very interesting. It is a well-known phenomenon that physicists make poor metaphysicists, and I fear I am no exception to that pattern. I feel shaky when handling philosophical concepts like possibility, necessity, potentiality, causation, and so forth. Are these concepts reliable? Do they refer to real things? Do I understand these things clearly enough to reason with them? Personally I don’t feel particularly confident when on this ground. I am wary of any argument which seems to require that I phrase things in a particular way, or use particular concepts rather than others, for fear that the argument is only working by a linguistic formula. But I recognize this wariness as potentially unjust, for does the framing of arguments in particular ways reflect anything other than its making use of the appropriate terms?
*
The worry that some of the metaphysical concepts used in the arguments might be merely conceptual, not corresponding to anything real, is, in its worst and most destructive form, the worm of nominalism eating the metaphysical apple. Feser recognizes that the nominalist habits common to much modern thought act as an acid on the structure of these (and other) philosophical arguments, so he devotes several pages to counterarguments against nominalism, and these arguments I found quite interesting.
One counterargument is that a nominalist cannot state his position without having recourse to universals. A nominalist may say that things do not share a common nature (because he believes there are no such things as natures) but are grouped together for convenience because they resemble one another in some way. But, as Bertrand Russell pointed out, this “resemblance” is itself a universal. And if the claim is that this resemblance, too, is merely a convention or name, then the question arises why particular “resemblances” are grouped together conceptually, since this can only be on the basis of some higher resemblance, and on and on. The nominalist is stuck in a vicious regress.
Another argument is that nominalism’s attempt to evade essences or natures in favour of mere names or words is a non-starter, since words are themselves universals. A nominalist says there is no such thing as blueness, but only things we call “blue” because they have a colour in common. But you say “blue” and I say “blue”; are we saying the same word or not? If we are, then we’re using a universal, and the nominalist has failed. As Feser sums up, “it is notoriously very difficult to defend nominalism in a way that doesn’t surreptitiously bring in through the back door a commitment to universals or other abstract objects, in which case the view is self-undermining.”
**
Having presented his five arguments and defended each of them, Feser still has several meaty chapters in reserve. He has already treated not only the existence but also the nature of the entity or reality of the thing each argument arrived at, and we have seen that these things shared certain properties in common. In this later chapter Feser draws these lines of argument together to argue that these five arguments do not exist in isolation from one another, but are drawn together by many connections and all converge on a single reality: God. He then systematically treats each of the principal divine attributes — unity, simplicity, immutability, immateriality, eternity, necessity, omnipotence, omniscience, perfect goodness, will, love, and incomprehensibility – and examines how each is supported by the arguments he has presented.
The argumentation in this section is sometimes quite subtle. Feser follows Aquinas in thinking that talk about God must for the most part proceed analogically – that is, words applied to God cannot be understood in exactly the same sense in which we apply them to contingent realities, nor are the meanings unrelated in the two cases, but there is an analogy between them. Feser devotes quite a lot of effort to carefully articulating what this means, distinguishing types of analogy, etc. As a result of reading this section I came tantalizingly close to finally achieving my long-time ambition to understand the “analogy of being”, but I ultimately failed.
The relationship between God and the world is the next major topic of discussion, and here Feser provides helpful overviews of divine conservation (that is, that God conserves the world in being here and now) and the relationship between miracles and laws of nature.
In the book’s final chapter he steps back to consider a huge variety of objections that are commonly raised against the kinds of arguments the book defends. These range from the manifestly incompetent (“If everything has a cause, what caused God?”) to the thoughtful (“Don’t the arguments based on the impossibility of infinite regress merely establish a necessary condition for the existence of a contingent world rather than establishing the existence of a necessary being?”). There is a valuable section critiquing scientism (the doctrine that science alone can provide genuine knowledge) and a good discussion of the problem of divine hiddenness. Indeed, this chapter, which accounts for about 1/5 of the length of the book, is a treasure trove of arguments and counterarguments relevant to natural theology.
*
At the end of this long overview, it is worth summing up some of the general features of the arguments presented. They are all metaphysical arguments. Nothing has been said here about science, or about any particular finding of the sciences. Nothing in these “first cause” arguments says anything about whether the universe is small or large, finite or eternal. It is hard to see how any future scientific discovery could have any bearing on them. Nothing depends on probability, an expanding universe, or the complexity of biological structures. Each argument starts from a single thing that exists here and now: I outlined the Aristotelian argument using my drink as an example, and that same drink could have served equally well for the others (except perhaps the Augustinian, though two drinks would have served there). The arguments are based on very basic ideas: change occurs, things are composed of parts, abstract objects are real, things exist contingently, and there are explanations for things. If the arguments are sound, they establish the existence of God with certainty. They are quite delightful and intriguing arguments.
*
I have been looking for years for a book that would clearly and carefully state classic arguments for theism, and especially for a book that would treat the divine attributes philosophically. I’ve read a few, but this has been the best. It would make a fine textbook for use in seminaries or university courses on philosophical theology. The attention Feser devotes to rebuttals of common misunderstandings and objections is a valuable service (though I can imagine a typical reader will have to hunt patiently for the answer to his objection amid the wealth of material). The writing is clear and concise, and I can see myself returning readily to this volume when I want to revisit the arguments or clarify my thoughts. There is much more here than I was able to absorb on one reading.
The existence of God is more than a matter of merely academic interest. I am myself a Catholic, so I affirm the existence of God. I pray, try to act in ways consonant with my dignity as a child of God, and hope to one day see God face to face. I do not spend a great deal of time, as part of my daily routine, thinking about absolute simplicity, hierarchies of causes, perfect ontological goodness, and so forth. Yet it would be false to conclude that these philosophical arguments are somehow irrelevant to my religious life and devotion. The God whom they reveal to the mind is, to say the least, mysterious, behind and before, replete and bountiful, overflowing with power, closer to me than I am to myself, hidden but present within every aspect of my experience, the great fountainhead of being, in whom I live and move along with everything else. There is substance here for fruitful prayer and reflection.