Cosmological Arguments

The Cosmological Argument is perhaps the classic argument for the existence of a God. Thomas Aquinas included it in his famous Five Ways, although over the years his argument has been constantly refashioned. It lives on in several distinct versions. I bring this up because of a “customer review” I came across on Amazon.com of a book by John Allen Paulos. The book is Irreligion: a mathematician explains why the arguments for God just don’t add up. The review is by M. Stringer.

Full disclosure: I read the review, I have not read the book.

Stringer, as it turns out, is quite critical of Paulos and his work.

As for Paulos’ book I would hesitate to describe it as even schoolboy philosophizing as it fails to reach any level of academic respectability and is, if anything, even worse than the aforementioned efforts from the `New Atheists’.

His first area of attack is the ‘first cause argument’ which Paulos states can be slightly amended to become the ‘cosmological argument’;

1. Everything has a cause, or perhaps many causes.
2. Nothing is its own cause.
3. Causal chains can’t go on forever.
4. So there has to be a first cause.
5. That first cause is God, who therefore exists.

There are however two major problems with Paulos’ version. Firstly no one in Western philosophical/theological history has even advanced the first cause/cosmological argument in this form. Paulos appears to have just made it up for this book. Secondly his version is not logically valid as the conclusion (5) does not follow from the earlier statements (1-4). All that is presented is a series of unconnected assertions unrelated to each other.

Stringer goes on to present what he considers a sound version of the cosmological argument (one popularized in recent years by the philosopher William Lane Craig). His seems shorter than what I recall as Craig’s version, but since brevity is a virtue, let’s take a look.

A good example a modern first cause argument is the Kalam cosmological argument rediscovered and improved in modern thought by William Lane Craig.

1. Everything that begins to exist has a cause.
2. The universe began to exist
3. Therefore the universe has a cause

This argument is logically valid. The conclusion (3) follows deductively from 1 and 2.

Now, I’m not interested in contesting Stringer’s characterization of the book he’s reviewing—I for one am in no position to do so. Instead what I prefer to do is comment on this rather succinct version of the cosmological argument.  I am aware of course that Craig is a better source for the modern cosmological argument than an Amazon reviewer plucked out of the hat, but, here goes….

1. Everything that begins to exist has a cause.
2. The universe began to exist
3. Therefore the universe has a cause

The short problem with this is that it assumes in the 2nd premise what it needs to prove, namely that everything (here referred to as “the universe”) began to exist.

Let’s take a closer look. This is supposed to be an argument for the existence of a Creator—and yet, it never mentions God or Creator. Of course, God is ever-present in the background, lurking, waiting for an opportunity to jump in. Let’s see if an opportunity presents itself.

Under the Microscope

The syllogism begins by asserting that everything that begins to exist has a cause. Why the phrase “begins to exist”?. It’s there so we can exclude God from the requirement to have a cause. Since by definition God is eternal, no beginning no end, premise #1 doesn’t apply to him.

That’s important. For the cosmological argument to work, it has to make the case that (A) “everything has a cause” and (B) “except God.” Obviously, a large part of the debate about whether the argument is successful centers on whether or not the exception made for God is warranted. What is unusual about Stringer’s version is that it doesn’t even mention God. Still, by asserting that physical things like the universe begin to exist and therefore must have a cause, the implication is that their cause must be something that does not begin to exist, i. e. God.

Yet, nothing in the argument requires causes to be non-physical. Nothing seems to prevent an infinite chain of physical causes; nothing, that is, other than the author’s bare assumption that premise #2 is correct. Well, not quite “bare.”  Actually, the idea is that premise #2 has been established by astrophysicists as a fact—after all, aren’t scientists in agreement that our universe began in a big bang which itself exploded from a singularity? Didn’t time itself have it’s beginning with that singular cosmic bang?

A glance at cosmology (the scientific study of the origin of the universe) makes it appear premise #2 is widely accepted as true, since most scientists heartily accept the big bang. And yet, for most cosmologists, I would argue, the term “universe” does not equal “all physical existence”. In fact, most scientists take it for granted that there is some kind of prior physical state which led to the singularity (itself a physical state) which led to the big bang and our current universe. And recently, some cosmologists (e.g. Stephen Hawking) are questioning the singularity anyway. Which means the big bang is not only not the beginning of all physical existence, it may not be the beginning of the universe either.

This is not fatal, of course. There is way too much uncertainty about the science of cosmology to say whether science will or will not end up supporting premise #2. The fact remains that if there is a God who created our physical world, then we ought to find ourselves living inside a world that had a definite origin at some specific point in the past, and prior to that point in the past nothing physical should be detectable. In fact, this fits reasonably well with current science. Sure, scientists talk about strings and multiverses in existence prior to the big bang—but at this point that’s just theorizing without evidence.

The Long and Short of It

So much for the short problem with the Kalem cosmological argument. But there is also a long problem—”long” in the sense that it won’t be as easy to explain, I’m afraid. But I will try.

There is a subtle problem with premise #1, and it involves the meaning of saying something has a cause. If one operates from a worldview based on mind before matter, then this premise is a founding principle. However, if one operates from a natural worldview (which rejects the principle of sufficient reason), then the negative of this premise is your founding principle. From this latter point of view, postulating “causes” is merely a useful way of describing the physical world.

Causes, in short, are a form of mental currency and not something “real” about matter. Technically, you might say, causes are imaginary. This viewpoint follows naturally from neurological constructivism and pragmatic empiricism. These approaches to understanding knowledge and science paint a picture of a relationship between thoughts about physical nature and the actual stuff of physical nature which is loose and indirect. In fact, it is just the sort of insufficient relationship evolutionary scientists should expect from “unguided” biological evolution.

Some of the key elements of this relationship can be summarized as follows. Knowledge is a virtual reality; its relationship to physical reality is like that of a useful map to the terrain the map represents; all of the logical relationships indicated by the map pertain to the map, not to the terrain. That is to say, the map is an analytical construction that has a synthetic relationship to the world it models. The map is only “true” to the extent that we find it a more useful model of the world than any alternative mappings we happen to have thought up. Knowledge, in other words, is something we invent to model the physical world by testing for usefulness. The scientific method codifies this process.

If matter comes first and mind evolves later (the premise of naturalism) then “causes” are just descriptions, and we choose our causal explanations based on their predictive usefulness, nothing else. The same applies for any non-causal explanations we might embrace, as well.

Imagine, now, if we were to restate Stringer’s cosmological argument from this natural perspective. It might look like this:

1. Everything that begins to exist can be usefully described.
2. The universe began to exist
3. Therefore the universe can be usefully described.

So we see that only by embracing a worldview which presumes that causal descriptions identify innate causal truths about the physical universe can the Kalam cosmological argument become an argument for God’s existence. But the notion that there are innate causal truths about or contained within physical existence is a notion that stems from a supernatural worldview (from mind before matter). It is inherently incompatible with a natural worldview, and no one with a natural worldview should accept it. (Some misguided atheists do, of course, but they are . . . well, misguided.)

We will find that if one accepts the premises of the supernatural worldview, it follows that the premises of the Kalam cosmological argument seem obviously true. If instead one hews to the premises of the natural worldview, the Kalam premises seem obviously false. We can be sure that the reverse is the case as well. Premises which seem obvious to advocates of the natural worldview will likely seem far from obvious to supernatural worldview advocates.

Here Comes the Judge

What we need, then, is a way to judge between the two worldviews independent of their inherent premises. I think this can be done. It involves first finding conclusions which differ between the worldviews and then comparing those conclusions to what we pretty much all agree are facts about the world. In short, which worldview best fits the facts, as we know them? This is not a philosophical endeavor so much as an empirical one—there will be no definitive answer that all can agree on. After all, pragmatic empiricism is the only tool we have to arbitrate this debate.

Notice that if I am right about this last point, in itself that supports the natural worldview. For the natural worldview entails that all matters of fact about existence must be brokered through pragmatic empiricism, the scientific method. But the supernatural worldview, it seems to me, entails that a shortcut to direct knowledge is possible, indeed that classical logical arguments can reveal facts about the world. I believe this contention can be shown to be unuseful, and has been shown unuseful again and again, as far as the determination of facts (rather than logical truths) is concerned.

There is another way to say this, which perhaps has more biological clarity. Over the course of the natural history of the earth, the brain has evolved into an organ which creates sensations which we refer to as the mind. This evolution has resulted in a relationship between “minding” and the physical reality that is the subject of that “minding” which is synthetic rather than analytic. Because the relationship is synthetic, pragmatic empiricism has become the best route to factual knowledge. Were the relationship analytic instead, then analytic statements would provide factual content about the world, and thus would have become the best route to factual knowledge. Yet things don’t work that way. That’s not the way the mind evolved.  Instead, only empirical statements provide factual content about the world—and this is just what we would expect if the premises of naturalism are true.

So what then are analytic statements “about”? They are about the organization of the mind itself, or perhaps more accurately, the organization of the brain’s “minding” faculty. In a real sense, of course, the brain’s “minding” faculty is something physical. So logical statements do have factual content in that limited sense. If I make an analytical statement, eg, 2 + 3  = 5 , I am making a factual claim about the organization of the minding faculty in my brain. Fair enough, but the organization of the minding faculty in my brain exists for the purpose of developing useful facts—descriptions, explanations and causes—about the physical world which lies outside my minding faculty. 2 + 3 = 5 tells me nothing factual about the world outside my minding faculty. That is precisely why we call math statements like that analytic rather than synthetic.

But this very state of things, it seems to me, supports the natural worldview and does not support—is not what would be expected in the case of—the supernatural worldview. With the latter, we would expect analytic statements, purely logical arguments, to provide factual knowledge about the world outside the mind. They do not, and that is one reason why I believe the natural worldview is far more useful as a worldview, why it “wins” the debate.

Terminology and Necessity

At this point let me say something about my terminology. Note that “fact” and “factual” in my usage do not equal “true”—when we say something is a fact we mean simply that it’s the most useful knowledge we’ve got (so far) on the matter, utilizing the pragmatic empiricism of the scientific method. Logical/mathematical knowledge can be “true” but it cannot, under this usage, be factual. Empirical knowledge, on the other hand, can be factual but it cannot be “true.” We can only continue to call factual knowledge “true” if first we redefine the term as a comparative meaning “more scientifically useful” than the alternatives it competes against. Again, this is just the method of pragmatic empiricism.

Now let me make a comment or two about another argument mentioned the book review above.

1. Everything has a cause, or perhaps many causes.
2. Nothing is its own cause.
3. Causal chains can’t go on forever.
4. So there has to be a first cause.
5. That first cause is God, who therefore exists.

As the reviewer points out, no one makes the cosmological argument this way because premise #1 forces God to also have a cause, and premise #2 prevents Him from being his own cause, which vitiates the conclusion. Note also that premise #1 and premise #3 are in flat contradiction: if everything has a cause then causal chains must go on forever. #4 follows from #3, but neither can be true if #1 and #2 are true.

So theologians try to make the argument work by asserting that premises #1 & #2 don’t apply to God but do apply to the physical world. But this is simply a case of special pleading based on confusing the physical world with our knowledge of the physical world. (I will explain this presently.)

Specifically, theologians traditionally define God as a “necessary” being and define the physical world as “contingent” instead of “necessary.” As I say, this is mere special pleading. But even if we accept it, the argument fails because if God is not a contingent sort of being then God can’t be a cause for contingent things—causality, in short, is a two-way street. Causes must be the sort of thing that can bring about what they cause. I have written about this in discussions of the cosmological argument elsewhere.

What does it mean to say something is “necessary”? Well, what is intended is that God’s existence be logically required, whereas the existence of physical things be not logically required. But really it is only another way of saying that something does or doesn’t have a cause—and we are back to special pleading. Can the theologian make a factual case for this distinction? Is there some way to show it is not special pleading? I don’t see how. Look at it this way: just because God was never created, why does it follow that God necessarily exists? Isn’t it just as possible that if God was never created God does not exist? Moving God outside the causal chain does not transform God into a necessary being.

I’m going to come back to this point in a minute, but now let’s consider the contingency side of the matter.

Contingency and Knowledge

The idea behind contingency is that if something has a cause or causes, then had those causes not occurred the something would never have come to exist. While this may seem to be true for individual things in the physical universe, importantly it is not true for the collection of all physical things. The existence of the collection of all physical things is logically necessary—therefore shouldn’t the entire collection (the physical universe in toto) fall into the same category of being necessary rather than contingent—and therefore like God, shouldn’t it be exempt from premises #1 & #2? The special pleading which supposedly exempts God must also exempt the universe taken in its entirety. (Note that the collection necessarily exists even if it’s an empty set.)

I think if we analyze this carefully we see that factual (synthetic) knowledge is “contingent” and analytic knowledge is “necessary”. The distinction is really not about the things known but about the manner in which we know them. Contingent things must be known empirically. Necessary things must be known logically.

There is a problem in this for the theist. It effectively denies that God’s existence is a factual matter and makes it a logical matter instead. That at once puts God into a category that prevents him from interacting as cause with the physical world (the “lack of contingency” problem). 2 + 3 = 5 is necessarily true, but that is because like all analytical knowledge it is not a reference to the world outside our “minding”. It is not a reference to anything factual. So the problem with the subtle cosmological argument is that its premises amount to simply asserting that the central claim of supernaturalism—that mind precedes matter—is true. This assumes what is to be proven, the fallacy of begging the question.

And anyway, it is not at all clear to me why individual physical beings which actually exist aren’t therefore “necessary” beings. True, our knowledge of them is synthetic, therefore merely factual, therefore uncertain to some extent. But it is a fallacy to assume that what it true for knowledge is equally true for the physical subject of that knowledge. We may always know through a glass darkly, but that is because knowing is a synthetic process based on pragmatic empiricism. Regardless of the uncertainty of what we know about a physical being, if it exists then it exists, it necessarily exists.

Whatever “contingent” steps led to your coming into existence, if you exist then you absolutely exist—you necessarily exist. What is, is. Things that exist exist regardless of logical argument or anyone’s factual knowledge of the matter. They exist regardless of what we know about them or how they came into existence.

A Different Necessity

But perhaps theists will reply that this is not what is meant by the term “necessary being”. What is meant is “a being who does not have to have a cause” a being who, if it exists, necessarily exists causeless. To this the special pleading objection obviously applies. For as I pointed out previously, advocates of the natural worldview maintain, as a necessary consequence of that worldview, that “causes” are simply knowledge-descriptions created by our brain’s ”mindings”—that it is a mistake to think that “causes” are true things, or that real physical things have innate causes. They only have the causes our minds find it useful to assign to them—causality literally exists in our minds and not outside our minds. Again, it is the mistake of confusing physical things with our mindings about them.

Thus to say something is contingent is simply to say that we can create knowledge about it through our minding process of pragmatic empiricism. That is, it is something that can be factually addressed. That’s all contingency really boils down to: if something is empirically knowable, subject to synthetic statements, it is contingent. If it is not empirically knowable then it is not contingent. Now we see the problem with defining God as non-contingent. It does serve to effectively distinguish God from the physical world, but at the cost of no longer being able to claim that God factually exists. God only theoretically exists, and the logical arguments which are supposed to “prove” that existence can only do so if we start them with premises which make God necessary rather than premises which do not. They amount to saying, “If things are such that God’s existence is entailed, then it follows that God’s existence is entailed.” True enough. But if things are such that God’s existence is not entailed, then God’s existence is not entailed.

Analytical arguments can’t settle factual questions. And ultimately, God’s existence is a factual question. Pragmatic empiricism, scientific method, is the only way to approach it. But any answer obtained this way will lack the certainty of truth. At best it will only be a fact, and therefore not a final answer.