I was first exposed to Aquinas’ 5 proofs of God’s existence as a college freshman — a strongly religious theistic freshman, at that — yet immediately I saw that his proofs were flawed. They didn’t work to prove God at all. My thought at the time was that if you substituted the human mind for God in the proofs, they worked just as well. The general conclusion I came at the time was that the type of God the proofs addressed was wrong: that our concept of God was too tainted with, too similar to, the human mind itself. The solution had to be in finding a better definition of God than the traditional one.
Surprisingly, at the time rejecting God never occurred to me as an option. Instead, I determined that the nature of God had to be quite different than traditionally conceived. God was not a creator-God, not a logos-God, but had to be some other kind of entity. I spent the next couple years trying to figure out what that entity might be.
Eventually I resolved the difficulty: by becoming atheist.
The Cosmological Argument
To give an idea of some of the stumbling blocks I perceive in the idea of God, let me quote Terry Miethe, himself paraphrasing Aquinas’ “Second Way” or second proof of God’s existence. All quotes come from Does God Exist? A Believer and an Atheist Debate by Terry Miethe and Antony Flew, 1991, New York: HarperCollins.
1. There are efficient causes in the world (i.e. producing causes).
2. Nothing can be the efficient cause of itself (for it would have to be prior to itself in order to cause itself.) [Does God Exist?, p. 130]
But already this dooms the concept of God, since it means that God cannot cause Himself, or in other words, one state or moment of God cannot proceed from a previous. To avoid this objection, God must be defined as absolutely unchanging and timeless. But if God is thus constrained, two inevitable conclusions follow: (1) God can’t behave or happen or act — that is, God can’t exist in any normal sense of the meaning of that word when applied to a being, and (2) God can’t be the first cause, since being a causal agent requires action. Something which can’t have changing states, or can’t exist one moment after another (therefore one moment being the cause of the next), can’t cause.
3. There cannot be an infinite regress of (essentially related) efficient causes, for unless there is a first cause of the series there would be no causality in the series. [p. 130]
If applied to God (as it seems it would have to be, since God is going to be placed into the series of “efficient causes” of the world) we are left with the conclusion that God, as well, must have a prior cause. And that prior cause, itself, must also have a prior cause. Ad nauseum.
Another way to restate this premise, simply, is thus: Every series must have a beginning, for without a beginning the series could never get started and therefore never come into existence. Since infinite series don’t have beginnings, they can never get started, and therefore can’t exist. But everything can be seen as a series. Even an infinite series can be seen as a series of finite series. Even God can be seen as a series, or placed into one.
Infinite Time & Infinite Space
If infinite series can’t exist because, as the premise asserts, they lack a beginning, then an infinite series of moments cannot exist. Time then, according to this premise, cannot be infinite, and it must follow then that there cannot be an infinite God, speaking temporally. Of course, an infinite series of space, or place, must also be ruled out, if we are to follow this premise, since otherwise we have no place to begin our series with. So infinite God is also rulled out spacially.
Aquinus, of course, wants to apply this premise to the world, yet exempt God from it. It certainly looks like special pleading. For if we claim the world for logical reasons cannot exist forever (and the premise, despite all its fancy words, is nothing other than that), then the same necessity should mean that God cannot exist forever.
Any claim that an infinite series is impossible is as harmful to God as it is to the world, and makes God as dependent as it thus makes the world.
And for any possibility that God can be uncaused (as we will see is claimed next), it is just as possible that the world itself can be uncaused.
If it is claimed that God is different from the world because God is non-physical, motionless and outside of time, then it becomes impossible for God to be a cause of the world. For without time you cannot change or move, and without the ability to move, you cannot act. Indeed, if you exist outside of time, you cannot be, much less create and cause.
4. Therefore, there must be a first uncaused efficient cause of all efficient causality in the world. [p. 130]
As I see it, this means that the series begins uncaused (uncaused God being the first item in the series), but in that case the series never comes into existence in the first place. It never happens. It never gets its first drop of movement, and time never starts for it.
If you simply disagree, and argue that the series can begin uncaused, then we can as easily begin it with the world as with God. Or we could begin it with the cause of God, rather than with God.
The three premises are simply wrong. The claim that unless there is a first cause of the series there can be no causality in the series, is one without any basis. Why should it be so? And how is the claim that the first cause of the series must itself be uncaused consistent with the claim in premise 2 that nothing can be the efficient cause of itself? For to say that something is uncaused is no different than saying that it causes itself — or else never comes into a state of existence in the first place.
Finally, the claim that there are efficient causes in the world must also be disputed. As a matter of fact, causes (and causality itself) is nothing but a mental phenomena. Causality exists not out in the world, but here internally as part of the currency of thought. What I mean is not that events don’t follow and flow out of other events, but that defining a “series of efficient causes” is nothing but an arbitrary mental game which, once begun, must result in an infinite regress of causes unless we tire of it, arbitrarily deciding to break the rules and declare “God.”
5. Everyone gives to this the name of God. [p. 130]
But it is a God who can’t get started, can’t behave, and therefore in any ordinary sense of the words, can’t create and can’t exist.
My conclusion is that to say something exists is to say that it has a past, or that there is a past which leads to it, as well as a future which proceeds from it. But to say God is uncaused is to say God has no past. I infer that such a God has no future as well.
Existence, in other words, requires that something happen, then something else happen after that. In fact, what I would say about existence is that it essentially equates with movement and change.
To say that something exists is to say that it happens, which is to say that it changes.
If God is to be a cause, God must move. Everything that moves, however, must have first been moved.
So to describe a chain of causality is only to describe the way existence is. To try to break that chain at God is only a hidden way of saying that God is not part of existence, which is an euphemism for saying that there is no God.
How can someone imagine that it is workable for God to be the uncaused cause of everything else?
They imagine it by making a very fundamental mistake, the mistake of thinking that language, a thought, an idea, can create something without itself having to be created. This is the notion behind a “metaphysical” source for physical reality.
This is what Miethe means when he says,
Thomas appeals to a type of causality unknown to Aristotle where existence itself is the effect, a type of causality where the effect is a finite efficient cause. This would be a metaphysical kind of causality rather than a physical one. [p. 131]
Let’s tackle this another way, by bringing in the popular modern term contingency. Miethe says, “Prominent philosophers have reformulated the argument to present what they believe is a true and valid cosmological argument.” [p. 133] Miethe then summarizes Bruce Reichenbach’s version as follows:
(S1) A contingent being exists
a. This contingent being is caused either (1) by itself, or (2) by another.
b. If it were caused by itself, it would have to precede itself in existence, which is impossible. [p. 134]
But I would argue that existing beings almost always precede themselves in existence, since existence must happen over time (involve change). Therefore b is simply erroneous.
(S2) Therefore, this contingent being (2) is caused by another, i.e., depends on something else for its existence.
(S3) That which causes (provides the sufficient reason for) the existence of any contingent being must be either (3) another contingent being, or (4) a noncontingent (necessary) being.
c. If 3, then this contingent cause must itself be caused by another, and so on to infinity.
(S4) Therefore, that which causes (provides the sufficient reason for) the existence of any contingent being must be either (5) an infinite series of contingent beings, or (4) a necessary being
(S5) An infinite series of contingent beings (5) is incapable of yielding a sufficient reason for the existence of any being.
(S6) Therefore, a necessary being (4) exists. [p. 134]
It appears transparent to me is that the necessary “being” or what I would term the “necessity” which the existence of contingent beings compels us to believe in, is what can only be termed the universe or totality of existence, taken as a whole. Everything that exists is contingent on the existence of the totality, both logically and causally; although the totality itself is and must be uncaused, or must be described as self-caused.
Two Pronged Sword
The use of the word being in fact performs a word-trick. It leads us into thinking that at the conclusion we are talking about an agent or person, when in fact all we are really talking about is a “greater existence.” (Alternatively, it is the “infinite series” taken as a totality which fills the role of “necessary” being.)
What cannot fill the role is God. As remarked earlier, God is defined so that God has no way of behaving, and therefore cannot be the source of contingency.
Contingency, in fact, is a two-pronged sword. Just as something has to come from something previous to it, the something which was previous to it has to have the sort of nature which can hand contingency, so to speak, to that which it causes.
To be a causer in the world, or even to be the first causer of the world, God has to be made of the kind of stuff that can cause contingent, physical things to happen. A disembodied idea, or even a metaphysical ideal, can’t do it, for it can’t interact with the world. God must be a contingent sort of thing to be a causal agent for contingent beings.
I will comment also on a particularly bizarre claim in the argument presented above.
(S5) An infinite series of contingent beings (5) is incapable of yielding a sufficient reason for the existence of any being. [p. 134]
In the premise immediately prior to this one, the phrase “provides the sufficient reason for” was defined as meaning “causes”. So the premise is really claiming that an infinite series is incapable of causing the existence of any being — or in other words, if the chain is infinite, no causality occurs.
Of course, the real reason for the appearance of the phrase “sufficient reason” is not logical, but emotional. It is designed to make us feel that only God can sufficiently explain why something really exists. But why something exists is different from whether it exists — and it is whether things can exist without God, rather than why they exist, which is the logical subject of this series of premises.
Miethe says that Reichenbach’s argument above is “based on the relationship of causality and sufficient reason.” [p. 134] Next he presents another version which he says is “not based on the principle of sufficient reason but on the principle of existential causality.” [p. 134] Here it is:
1. Some limited, changing being(s) exist.
2. The present existence of every limited, changing being is caused by another.
3. There cannot be an infinite regress of causes of being.
4. Therefore, there is a first Cause of the present existence of these beings.
5. This first Cause must be infinite, necessary, eternal, and one.
6. This first uncaused Cause is identical with the God of the Judeo-Christian tradition. [p. 134]
My first impression is that steps 3, 5, and 6 are flawed. But first let’s let Miethe explain his steps.
1. Some limited, changing being(s) exist. [p. 135]
For example, us. There is certainly nothing to disagree with here.
2. The present existence of every limited, changing being is caused by another.
3. There cannot be an infinite regress of causes of being. [p. 135]
Of course, it seems to me that there can ONLY be an infinite regress of causes of being. But Miethe maintains this argument is valid because “an infinite regress of causes” does not refer to a “linear series of historical causes of becoming, but a vertical series of causes of being, for existence itself.” [p. 135] He explains “existential causality refers to the cause of the being of entities and not the cause of their becoming,” and then clarifies (or rather unclarifies) this by saying that he is . . .
talking about a cause for the very being of a thing, not its coming into existence (its becoming) or the changes it may undergo. It is impossible to have an infinite regress (go backward to infinity, has no first or beginning cause) of existent-dependent causes. [p. 130]
I don’t understand the distinction Miethe is attempting to make here. Perhaps because I cannot envision how “the very being of a thing” and its “coming into existence or the changes it may undergo” can be anything but references to the same thing. Is Miethe trying to say something like: I concede that Jennifer’s “coming into existence” was caused by her parents, those, limited, changing beings called Mary and John, but the source of “the very being” of Jennifer–the “existential cause” of her being, as opposed to the historical cause–has to be an uncaused First Cause? Body by Mary and John; but soul by God?
If this is what Miethe means (and I am not sure it is) then it boils down to the unwarranted assertion that only God can be the cause of a thing’s soul or essence. In which case his argument is really as follows:
(1) At least some things (humans) have a soul or essense.
(2) Only God can be the cause of a thing’s soul or essense.
(3) Therefore God must exist.
The problem here is that (2) is just a bald assertion, and if we define soul as “a god-like aspect”, then (1) becomes the bald assertion instead. (And of course it begs the very question at hand by presuming that God exists in its premise.)
Furthermore, Miethe’s argument that premise 3 refers not to a “linear series of historical causes of becoming, but a vertical series of causes of being, for existence itself” in support of the claim that there can not be “an infinite regress of causes of being” is itself inconsistent with the line of argument in premise 2, where it appears clear that a “linear series of historical causes” is meant. Premise 2 states, “The present existence of every limited, changing being is caused by another.” If, instead of referring to historical causation, this is about a “vertical series of causes of being” as it must be if premise 3 is to logically lead us to statement 4, then we are left with no idea in the world what premise 2 means.
Logical vs Historical Causation
Unless, that is, the distinction Miethe is attempting to make between “causes of being” and causes of becoming” is a distinction between logical and historical causation. Miethe writes
When you simply add another dependent being to a chain of such beings, it does not ground the existence of the chain. To say that it does is like saying one could get an avocado by adding an infinite number of pineapples to a basket of pineapples. Adding pineapples to pineapples does not yield an avocado; adding dependent beings to other such beings does not yield a cause or ground for their dependent existence. One contingent being cannot ground another such being. No caused being can be an intermediary in a chain of existential causality. Thus it follows that the very first cause of a caused being must be an uncaused Being.” [135-136]
I can make sense of this only if I assume that in Miethe’s canon “first cause” means “logical cause” while “dependent cause” means “historical cause”. Yet if this is what Miethe is referring to then he has fallen into an obvious fallacy.
Perhaps it can be clarified this way. An oak begins as an acorn, sprouts into a sappling, eventually becoming a fully-grown oak tree dropping acorns of its own which themselves sprout into other oaks. This is what Miethe means by dependent causation. Fine and good. If I interpret Miethe correctly here, he would readily assert that one oak tree causes the next in a series of dependent causation. But what, he seems to be asking, causes the ground of being of oak trees themselves? No matter how many oak trees we add to the basket of dependent causation to explain each individual oak tree’s becoming, the cause of oak trees as a class remains unexplained.
Now surely Miethe is aware of evolutionary theory and how, according to scientists, individual variability and natural selection lead to speciation. Even if one rejects evolution as historical fact, one must admit the theoretical possibility of speciation by evolutionary processes. We must assume therefore that it is not the origin of species — of oaks in this example — that Meithe is denying on logical grounds to dependent causes.
Essence Before Existence
Instead, Miethe must be denying that dependent causes — individual oak trees — can cause the logical class of oak trees to come into existence. Miethe declares, in other words, that evolution can account only for the becoming of oak trees, and not for the ground of their being — and to make any sense of such a distinction we are forced to interpret “ground” as meaning “logical ground”, to wit the initiation of the contingent oak tree, sapling or acorn into the logical class of oaks.
This initiation is a purely mental process, and the relationship of any particular dependent existence or “contingency” to its logical class is a purely mental relationship.
Miethe’s assumption, I take it, is that these logical classes are universal, that they transcend history, and therefore they cannot have a contingent cause. I can find no other way to make sense of his language.
To put this into Sartrian terms, Miethe asserts as a raw premise essence before existence. But that is of course precisely what atheists reject when they assert atheism: to declare that there is no God is to declare that there is no essence before existence, no mind before matter, no logical framework to existence which must serve as a blueprint or logical ground for the physical world. In assuming essence before existence Miethe is simply declaring up front the God he is trying to prove.*
Why then, theists might ask, do we see logical relationships everywhere we look? And why do they appear universal and transcendental? And the atheist answer is simply that logical relationships are the currency of thought itself: the very process of thinking requries and creates them. They come from us, out of our very human brains. Why do causal relationships seem universal and transcendental? Because that’s what makes thought useful. In actual fact, however, every aspect of human thought is species-specific to humans.
The essential problem of the cosmological argument (whether or not modernized) is that one way or another the existence of God is presupposed in the premises, but this is disguised by the use of ambiguous and confusing phrases.
* Of course theists can make the opposite accusation: that in denying essence before existence, atheists are beginning with a premise that excludes God’s existence. But this only focuses us on the real point of contention between theists and atheists: the relationship of thought to the world. It is an even more fundamental divide than the God question.