But with God, we have to start with the very basics. Does anything really exist at all? Or is the universe an illusion? The latter view is known as solipsism. In this worldview, I, the solipsist, am the only thing that truly exists. Everything is an illusion in my mind. Solipsism is widely held to be irrefutable, but in fact it has one flaw — it doesn’t go far enough. It fails to ask: an illusion in contrast with what? If everything is an illusion, then there is no reality to distinguish it from, and the illusion itself is the realest thing there is. Hence we still have a real universe to deal with. “Ah,” the solipsist might say, “but what about dreams? After all, when we dream, we don’t have any other reality but the dream in front of us.” True enough. We distinguish dreams, illusions, and hallucinations by the fact that they don’t fit in with the rest of our “real” experience.

Which, of course, promptly raises a further question. All of that experience except the present instant lies in the past; and, as Wen the Eternally Surprised wondered,

“I remember yesterday,” said Wen thoughtfully. “But the memory is in my head now. Was yesterday real? Or is it only the memory that is real? Truly, yesterday I was not born.”

— Terry Pratchett, Thief of Time

1. Fire burned the day before yesterday.
2. Fire burned yesterday.
3. Fire burned today.
4. Therefore, fire will burn tomorrow.

We might answer that fire burns by definition. If it doesn’t burn, it isn’t really fire. Does this help? It does not. Now we must ask: are the flickering red tongues in the hearth really fire, or do they just look and sound like fire? Let’s call them phlox, that is, “something that looks and sounds like fire”. We now have the following chain of logic:

1. Phlox burned the day before yesterday, and so proved to be fire.
2. Phlox burned yesterday, and so proved to be fire.
3. Phlox burned today, and so proved to be fire.
4. Therefore, phlox will burn tomorrow, and so prove to be fire.

The philosopher Plato wondered: how do we recognise things? If you look at, say, two different chairs, or the same chair from different angles in different lights, the two will be completely different in shape, colour and size. How, then, do we immediately know they are both chairs? Plato would have it that the chairs participate in what he called the eidos (eidoV) of chairness; a word traditionally translated form, but one that could just as well be rendered essence or nature. Seeing the chair, we recollect the eidos which we already knew, and so recognise it.

Plato argued that the eidē (one eidos, two eidē) were a deeper reality than the objects we actually experience. Never having seen The Matrix, he compared humanity to prisoners in a cave, unable to look out. All we see of the true nature of existence is the bewildering shadows of the eidē, cast on the inner wall of the cave by the unseen blinding light of Absolute Truth, from which we troglodyte seekers of wisdom may or may not deduce some glimmer of veracity.

Can you see the flaw here? Knowledge consists of recollecting our prior knowledge of the eidē. What did that knowledge consist of? Plato maintained that we knew the eidē in a previous existence, but so what? We have merely shifted the question back a step, not answered it. Here begins an infinite regress. Plato would perhaps have replied that, just as all chairs take part in the eidos of chairness, so all eidē in their turn have the eidos of eidosity, the Absolute Truth — the buck stops here. We get into trouble fairly quickly going that way. Not only do we recognise those two chairs as chairs; we also notice all the differences between them, in shape, colour, size, materials, and construction. If all knowledge is from the eidē, then each of those features must have its own eidos. There must be an eidos of blueness and an eidos of redness, an eidos of wood and an eidos of plastic, and so on. And in our hypothetical prior existence, we must have been able to tell the eidē apart somehow. So each eidos must have had its own distinct features, and each distinct feature must have had its own separate meta-eidos so we could tell them apart, and so ad infinitum. They cannot ever be resolved into a single über-eidos of Absolute Truth. We must have had an infinite succession of past existences, which incidentally rather puts the kybosh on the concept of a Creator.

There’s another flaw as well. Suppose one of our recollections was mistaken. Could we ever know it was mistaken? Since our knowledge consists of the recollections themselves, clearly not; we would be stuck in our error forever. We cannot be sure that this hasn’t already happened. Any part of our knowledge might be wrong, and we would be powerless to learn better. We cannot be certain about anything... which takes us right back to solipsism. Plato’s concept is useless to us.

And even if we don’t insist on Plato’s concept, there’s still another problem with Essentialism generally: the distinctions we make between objects are defined by our conceptual capacity, not vice versa. Put it this way: one grain of sand all by itself is not a heap. Two grains right next to each other are not a heap. But three thousand grains all together are a heap. If I drop sand grains, one by one, onto a table, at which precise number do they acquire the Essence of heapness? Clearly, there is no such exact number. I would answer that sand grains may be described as a heap when there are too many all piled together for us to see how many. But in that case, heapness appears earlier for someone with poor eyesight, and later for, say, an autistic savant who can see larger numbers. So if the Essence of a heap exists at all, it is in the mind, not the sand. Similarly, the distinctions between colours differ from one language to another. In English, grass is green and the sky and sea are blue. In Gaelic, grass and the sea are both gorm, but the sky is liath. Philosophers today have relocated Essences to the human consciousness, where they are called qualia. Even there, their existence is debatable: see Daniel Dennett’s What RoboMary Knows and Lovely and Suspect Qualities.

So much for Essentialism. Is there, you may ask, a way around the Problem of Induction? Is there any way to be sure of the laws of nature? Well, as with the horned pigs, absolute certainty eludes us; but again, as with the horned pigs, probability is on our side. Our experience of what fire has done in the past is a very large sample indeed, and so the probability of its doing something different is extremely low.

The laws of nature bring us to another concept: cause and effect. What exactly is the relation between a cause and its effect? As I type, the pressure of my fingers on the keys causes letters to appear on the screen. What does that mean? Well, basically, it means that the letters could not appear if I was not pressing the keys. That’s a law, not exactly of “nature”, but of the system comprising me + the computer. If that rule didn’t apply, if the monitor changed regardless of what I was doing, then we would say that my typing was “having no effect”. Likewise, holding a pencil over the floor and letting go would “cause” it to fall, due to the law of gravity. If we have two events A and B, such that, if event A does not happen, event B cannot happen either, then we say that event A is the cause of event B, and event B is the effect of event A. Cause/effect relations are simply particular instances of the laws of nature. But as we have just seen, such laws are merely statistical. Rather than say that the effect could not happen without the cause, we should say that it has not happened up until now. Once again, though absolute certainty is lacking, probability is still overwhelmingly on our side.

Isn’t there more to Cause-and-Effect than that, though? Couldn’t I say that pressing the keys causes electric circuits to join up, which cause currents in the computer’s interior, and so on? Well, yes, I could; but this changes nothing. Now, instead of one cause and its effect, we have a long chain of causes-and-effects, each, once again, a matter of applied statistical laws of nature. We are dealing with much more universal laws now, such as the laws of electromagnetics, mind you; these rules apply with far higher levels of regularity, so the probability that they will be broken is exceedingly small. But absolute certainty will always be unattainable.

Which leads us to a sobering conclusion: all non-mathematical, real-world knowledge is tentative. Even without the Problem of Induction, in the real world we have to act on our knowledge; we can’t afford to postpone our actions until we have reached absolute certainty. So we have to carry a huge amount of provisional knowledge around in our heads. We have no choice. I don’t think this is any cause for despondency, though. Think of a bat navigating around a dark cave. Bats use a kind of sonar to avoid crashing into the walls, as well as to find food. They emit loud squeaks, inaudibly high-pitched to humans, and then listen. If an echo comes back, there is something near, and the exact pattern of the echoes reveals whether it’s a wall or a moth. If no echo comes after a certain time, the bat concludes that there is nothing there — and keeps squeaking. All it ever has is a provisional picture of its surroundings, so it never stops investigating. Neither should we.

Speaking of cause and effect, let’s begin with a venerable argument for God’s existence: the Argument from First Cause. St Thomas Aquinas’ original formulation of this argument depends on one of Aristotle’s errors in physics, namely that motions require driving forces. Sir Isaac Newton demonstrated that it is only changes in motion that require outside forces. But this is incidental to the argument itself:

1. A contingent being cannot exist without a cause.
2. The universe consists of contingent beings only.
3. An infinite regress of causes cannot accomplish anything.
4. Therefore, the universe was caused by an uncaused, non-contingent being.

Every being that exists either exists by itself, by its own essence or nature, or it does not exist by itself. If it exists by its own essence, then it exists necessarily and eternally, and explains itself. It cannot not exist, as a triangle cannot not have three sides. If, on the other hand, a being exists but not by its own essence, then it needs a cause, a reason outside itself for its existence. Because it does not explain itself, something else must explain it. Beings whose essence does not contain the reason for their existence, beings that need causes, are called contingent, or dependent, beings. A being whose essence is to exist is called a necessary being... God would be the only necessary being — if God existed.

— Peter Kreeft, The First Cause Argument

1. There are no uncaused beings.
2. An infinite regress of causes cannot accomplish anything.
3. Therefore, the universe was caused by an uncaused being.

From the Argument from First Cause, we turn to the Argument from Reason. In Miracles, C. S. Lewis argued that the very existence of reason is proof that there is a reality beyond Nature. For us to have any assurance that the conclusions we reach by thinking are valid, they cannot come from Nature. Atoms don’t think. If we insist that Nature is all there is, we must concede that we don’t have any assurance that our reasoning is valid; and so we must throw all philosophy out the window, including the statement “Nature is all there is”.

That’s intuitively convincing, but intuition cannot always be trusted in abstract matters. If reason is not found in Nature, it must still happen somewhere. It may be a black box to us, but something must be inside; we put premises in and get conclusions out. If we can’t open the box, how can we be sure that what comes out is valid? Mark Rosenfelder rebuts a parallel argument as follows.

Let’s suppose that in the year 2115 neurologists tell us that they’ve figured out how the brain actually understands things. What would that mean? Precisely that they can explain it in terms of components that do not themselves understand.

Perhaps they tell us:

Here’s how the mind understands. The mind is composed of three components, the blistis, the morosum, and the hyborebus. The blistis and the morosum have nothing to do with understanding; the part that understands is the hyborebus.

We don’t have to know what these things are to know that they’ve failed. This cannot be an explanation of understanding, because it simply transfers the problem from the “mind” to the “hyborebus”. It’s like explaining vision by saying that the optic nerve brings the image from the eye to the brain, where it’s projected on a screen that’s watched by a homunculus. How does the homunculus’s vision work?

— Mark Rosenfelder, Understanding the Chinese Room
(emphasis original)

The very concept of a separate supernatural realm is shaky. According to theists, God is supernatural, and transcends Nature. He performs actions in the universe, perceived by us mortals as miracles. So the natural world cannot be entirely cut off from the supernatural; they are connected by God’s actions, at least. That means the natural and supernatural realms are parts of a larger system which encompasses them both. Logically, we should reserve the term universe for that larger system, rather than just one part of it (see below). Hereafter, the distinction between natural and supernatural is simply a matter of labels. In this light, the sophistry of Lewis’s argument becomes clear. Effectively, he defines Nature as “everything but reason”, then holds up reason as proof that there is something beyond Nature.

Fig 1. The transcendental position. God exists outside of the natural universe and produces miraculous effects within it.

Fig 2. Since God produces effects within the natural universe, the two of them can be seen as parts of a larger system.

Fig 3. The label universe can now be assigned to the larger system. The transcendental view is no longer tenable.

To which you could reply that God’s miracles do not follow rules. The interaction between the supernatural and the natural is unsystematic, and cannot constitute a system. But this leaves theism worse than before. If there is no regularity to the miracles, they must occur at random, which implies that God has no purpose in performing them. Indeed it would mean that there is no intention behind them at all, and that we are not dealing with a supernatural persona but with a collection of unconnected events.

Fig 4. A world subject to random miraculous events.

In practice, supernatural means, pretty much, “not available for investigation by nit-picking sceptics”. People who believe in the supernatural usually object to this definition, though, so let me clarify. It’s entirely possible that things exist which we are not used to thinking of as physical. I’m pretty sure human minds exist, for instance. But when somebody says “There must be more to the mind” — or to reality itself, or whatever — “than the boring old scientific facts,” you have to wonder what else they are looking for, and what it would look like if they found it. There is more to the mind than science can currently explain, that’s for sure. Should we conclude that science will never explain it? That there is something spooky about it that doesn’t boil down to mere matter and processes within the brain? I don’t think so. If something is involved that we don’t currently know anything about, some day some bright young researcher will discover it. Whatever it is, it happens in the brain; therefore, once discovered, it will go on the list of “brain matter and processes”. If that sounds wrong, it’s because supernatural does indeed mean simply “not available for investigation”.

One line of persuasion much used by evangelists was formulated in 1660 by the French philosopher Blaise Pascal. Suppose we have no clue whatsoever whether God exists. There is nothing to point us in either direction, so we have an exactly 50% chance of guessing right. If we guess God exists, the possibilities after death are eternal happiness in heaven or nothing; if we guess he doesn’t, the possibilities are eternal misery in hell or nothing. As Pascal himself put it,

Let us weigh up the gain and the loss involved in calling heads that God exists. Let us assess the two cases: if you win, you win everything: if you lose, you lose nothing. Do not hesitate then: wager that he does exist.

— Blaise Pascal, quoted in Kreeft, The Argument from Pascal’s Wager

Nor do matters improve if we assume, as Pascal did, that God has motivations roughly similar to those of a human being. If the traditional concept of hell is correct, then Fred Phelps has the basic facts right: everyone who dies without Christ, including many of our dearest friends and family, are doomed to an eternity of fiery torment. Take two seconds to imagine this happening to someone close to you. Can you honestly say you believe they deserve it? Most Christians I know would distance themselves from such gleeful hate. But if God created hell, then he, being omniscient, cannot distance himself from the reality. What kind of person would not only subject his creatures to the ultimate torture, but, having done so, sit back and watch it forever and ever? What sort of “reward” is that person likely to dish out? And would you want to live forever, knowing that it was because such a sadist had approved of you? Believers may reply that we have God’s promise of heaven just as we have his threat of hell; but this is a God who refuses to provide unambiguous evidence of the dreadful choice before us until it is already too late. Kreeft likens this behaviour to “a lover with a marriage proposal”. Given what’s at stake, an apter comparison would be to a stalker with a death-threat.

So the major metaphysical arguments for God do not stand up under scrutiny. What about arguments against? Is there anything, philosophically speaking, that makes a godless universe a better guess? Well, first of all, there is the scientific rule of thumb known as Occam’s Razor. This is usually phrased as “The simplest explanation is most likely to be true,” but this wording is misleading. There are several things you could mean by simplest, including “the one that takes the least time to explain”; by that definition “God did it” is a pretty simple explanation, but “D-uhh?” is even simpler. What William of Ockham actually said was “Do not multiply entities without necessity.”

This is a matter of probabilities. If some statement is almost certain to be true, the chance that we have guessed correctly whether or not it is true is close to 100%. Conversely, a statement whose truth-value is totally unknown to us is like a coin-toss: true or false, heads or tails? Neither possibility is any likelier than the other, so the probability of each is 50%, one-half. Two unknowns are like two coin-tosses. There are four possibilities: both true; both false; first one true but second false; first false but second true. If the unknowns are truly unknown, all four possibilities have equal probability. Therefore, each one has a probability of 25% — in other words, it is probably not true. Each unknown we introduce into an explanation halves its likelihood. Intuitively this makes no sense at all, but the logic is inescapable. Two “unknown”s make a “probably not”. Hence, in the absence of any evidence one way or another, an account of the universe’s existence which doesn’t include God is better than one which does. Richard Dawkins’ chief argument against God, as expounded in his best-selling book The God Delusion, is grounded in Occam’s Razor: briefly, to design and create the universe, God would have to be at least as complex as the universe itself — supremely complex, and therefore supremely unlikely.

But in fact there is very strong evidence against a perfect, all-powerful God: the existence of evil. Either God wants evil to exist, in which case he is not perfect, or he cannot prevent it, in which case he is not all-powerful. Dawkins dismisses the Problem of Evil as a “childishly easy” objection to answer, compared to his own argument from complexity. It can be overcome, he notes, by postulating an evil God, or a God who isn’t concerned about human distress. I disagree. Nobody who believes in God believes that he is evil, at least not according to their own lights; some people’s ideas of God look pretty evil to me (Fred Phelps’, for instance), but not to them. If, on the other hand, there is an evil God as well as the good God, then either God has an equal (Ahriman) and is not really God, or else God created a being who rebelled (Satan) and the Problem of Evil stands. As for the unconcerned God, a believer would argue — at least, I used to — that an infinite, all-knowing God would presumably have an infinite attention span, which closes that particular escape route. Christian theologians have proposed three broad solutions. First, there is the Bible’s answer:

Therefore hath he mercy on whom he will have mercy, and whom he will he hardeneth. Thou wilt say then unto me, Why doth he yet find fault? For who hath resisted his will? Nay but, O man, who art thou that repliest against God? Shall the thing formed say to him that formed it, Why hast thou made me thus? Hath not the potter power over the clay, of the same lump to make one vessel unto honour, and another unto dishonour?

— Romans 9:18–21

The second solution is that God, in Christ, suffers just as we do. Whatever sorrows we have, he shares. The depth of our suffering merely highlights the depth of his compassion. We have no more cause to complain than he has, and he took our griefs on himself willingly.

Are we broken? He is broken with us. Are we rejected? Do people despise us not for our evil but for our good, or attempted good? He was “despised and rejected of men.” Do we weep? Is grief our familiar spirit, our horrifyingly familiar ghost? Do we ever say, “Oh, no, not again! I can’t take any more!”? He was “a man of sorrows and acquainted with grief.” Do people misunderstand us, turn away from us? They hid their faces from him as from an outcast, a leper. Is our love betrayed? Are our tenderest relationships broken? He too loved and was betrayed by the ones he loved.

— Peter Kreeft, What is God’s Answer to Human Suffering?

Finally, there’s the Doctrine of Free Will. God has given humans the ability to choose our own path. That way, our love for him is a gift, not a tax. If we go the wrong way, God is not to be blamed. But then how do we explain evil falling on the innocent? Wouldn’t a just God at least ensure that the consequences of one person’s evil actions fall on them and on them alone? Well, no-one is truly innocent, because we are all born with sin inside us. And there is the fundamental problem with the doctrine of Free Will. We did not choose to be born with sin inside us. Either God made us evil, or we became evil through the actions of others — in the traditional concept, our ancestors — which God could have prevented. Either way, our evil is still at least partly God’s fault.

We may also note another difficulty in passing. Free Will, according to the doctrine, means any choice, including sin, is possible for us. (A problematic concept, but we’ll let that go for now.) Remember, the laws of nature are purely statistical observations. Therefore, if something never ever happens, there is, by definition, a physical law forbidding it. Hence, if something is physically possible, that means that, given infinite scope, it is inevitable. And according to Christianity, we will be given infinite scope, in the form of Eternal Life. So sin is inevitable. After God has cleaned up this world and banished the devil and all his works, it is only a matter of time before someone else starts the whole sorry mess all over again; hardly the work of a wise creator. If sin is somehow to become impossible in the life to come, why didn’t God make us that way to begin with?

The Problem of Evil, then, remains unanswerable evidence that God does not exist. And this, accordingly, must be our conclusion. The answer to the question posed in the title is “No.” Like the bat, however, we cannot stop there. There are still many objections to answer. We must explain, for instance, where the complexity of the living universe came from, and how it was that we came to have the capacity to argue about God in the first place. But now we have a standard to measure the counter-arguments against; if they are to convince us, they must be more compelling than the Problem of Evil. Onward, then, to the Argument from Design.

Last updated: 21 March 2007