In this post, I will seek to illustrate that the existence of the complex Universe is very powerful evidence against the existence of the Judeo-Christian God. Indeed, the same goes for any other claimed deity who possesses the characteristic of being the Universe's creator.

Assumption: The Universe must have had a cause

In order to consider this issue, we will proceed from the above assumption. That is, we assume that the Universe did not cause itself and did not spontaneously generate with no cause.

Equal or greater complexity of cause

In order for A to cause B (with no other factors at play; no other sources of information), A must be at least as complex as B.

By complexity, I don’t mean structural complexity. Rather, I’m referring to informational complexity.

Imagine an entity P (I will later also refer to it as P1) that causes the Universe. For our purposes, suppose that P is a non-sentient entity; it’s a “thing” that causes the Universe. P causes the Universe by transforming itself into it; in other words it somehow turns into the singularity and then explodes.

We must keep in mind that P is not governed by the physical laws as we know them. That’s because these laws are part of our Universe. P, having existed “prior” to the singularity, is not part of this Universe and therefore there are no grounds to assume that it’s governed by the same physical laws. Note that I don’t anticipate any objections at this point, at least not from theists. Theists posit the existence of a god who is not subject to the physical laws of the Universe. Equally, P is not. In fact, P creates these laws (by generating the Universe).

We don’t know what the structure of P is. But we do know that it’s capable of generating the Universe and that the Universe it will generate will necessarily expand into what we have today, with all its laws and the (very improbable, it is said) set of physical constants.

Hence, whatever the structure of P is (and it might indeed be physically a simple structure, though one might argue that it could not be; but that’s not relevant), P must contain all the information needed to generate something as complex as this Universe. Sure, it doesn’t have to have a map of where Andromeda will be at any specific time, or even that Andromeda will exist. But the precise position of objects in our Universe is not why we call it “complex”. A completely random, chaotic universe can be imagined which nevertheless has lots of objects flying chaotically and randomly with no set laws or constants (eg where each object runs by its own random laws).

And yet, in the sense in which the Universe's complexity (and, ultimately, its unlikelihood) is proposed, such a random chaotic universe would not be considered as complex and improbable as our existing Universe. And if it would, then we have to agree that there's nothing specifically improbable and complex about our Universe in the first place. This would potentially defeat the my argument. But it would also amount to a conclusion that the existence and nature of the Universe (as it is) cannot be used as evidence for the existence of a god; there being nothing particularly unlikely about the Universe's laws and constants.

All that said, it is traditionally claimed that the complexity of the Universe rests in the existence and nature of those laws and those constants. It is in this context and to this extent that P must contain all the complexity of the Universe.

In other words, P must be just as “informationally organised” as the Universe itself.


It is often argued (and theists often rely on this) that the Universe, with its specific set of constants and laws is very improbable indeed. People have proposed numbers to represent this improbability and these numbers are nothing short of mindboggling.

There are many arguments against this proposition. Some claim that these constants are a result of a single property of the Universe (making it much less improbable), others take issue with the significance of the improbability for debates about the existence of a god and so forth. We are not interested in any of that in this discussion. We assume that the Universe, with its set of laws and constants, is indeed very, very improbable.

Now, let’s get back to our hypothetical entity, P. P is the entity that existed (in this scenario) “prior” to the singularity and its only property was that it would necessarily (via laws we don’t know or understand; laws contained in P itself) turn itself into the singularity which then would explode and give rise to the Universe as we know it.

My argument is that P must be at least as improbable as the singularity and as the Universe itself. This should be very simple to grasp and should not meet any reasonable objection. Why? Because for all practical purposes we might as well (we won’t, but I’m trying to illustrate something) say that P is the Universe in its earlier stage. Remember, P does nothing except for turning itself into the Universe as we know it. We could say it’s the Universe in an embryonic stage. P already contains all the data (in some structural form that we don’t and can’t know) for the generation of the Universe.

Since P contains information about every single of the Universal constants and every single of the Universal laws, it is in effect a “packed up” version of the Universe; an embryo. And because it contains al the information that is needed to give rise to this precise Universe (and no other!) it must be equally improbable as the Universe itself. After P transforms into the Universe as we know it, no further information is added. The complexity of information is equal between P and the Universe. No changes in informational complexity. Hence, the improbability must be equal (or even greater for P, but I don’t want to go that far; there’s no need).

To sum up, we have now the following scenario:

1. An entity P exists which then turns itself into the Universe as we know it. It has no capacity to do anything else except to transform into the Universe.

2. P is at least as improbable as the Universe itself.

Let’s now vary P a little.

P2 and U2

Now that we have this construct of P, let’s go a step further. Imagine a different version of P. We’ll call it P2.

P2 is exactly the same as P but it is a little bit more complex. P2 contains the information needed for generation of two alternative universes. One of those is our Universe (call it “U” or “U1”), with all of its complex constants and improbable laws. The other is an alternative universe (call it “U2”), with a different set of constants and different laws. Note that the constants and the laws of U2 are just as improbable (at least approximately so, depending on how many constants there are of course) as those of U. This is not particularly relevant but it is an interesting point to add.

We can play around and assume that U2 is a universe that’s tuned in to only do two things: Big Bangs and Big Crunches. If it gets generated, it will pulsate like that to forward infinity (forever and ever from the point of generation).

It’s not known whether P2 will give rise to U (our universe) or to U2 (the pulsating universe). The decision will be random and the randomiser is based on some property within P2 itself.

P2 is more improbable than P

Having introduced P2, I can now claim that P2 is more improbable than P. That’s because in addition to being able to generate U, it can also (but alternatively) generate U2. On top of this it contains an additional property; the randomiser.

Because P2 contains more information than P1, P2 is more improbable than P1. It’s more complex and more organised. Keep in mind, that it’s the Universe’s complexity and organisation that we claimed makes it so improbable in the first place.

Enter P3 – P2 with a twist

Now, imagine another P. We will call it P3. P3 has the same characteristics as P2 except that it is conscious and intelligent. It can actually think. Its thinking ability is very limited, however. The only thing that P3 can think about is whether to generate U1 or U2. In other words, the decision about which universe to generate is not random; it’s an intelligent decision.

P3 is more improbable than P2

It shouldn’t take more than a primary school diploma to quickly see that P3 must be more improbable than P2. That’s because in addition to the characteristics of P2, it has a third characteristic; intelligence.

A quick side-point about intelligence

We know that intelligence is a very rare thing. We also know how complex it is. There are many claims that the human brain is in fact more complex than the entire rest of the Universe. I don’t know if that refers to structural complexity or processing complexity (the ability to process information in this case) but that’s not important. What is important is that intelligence is an extremely complex and rare thing. If we agree on this, then the difference in improbability between P2 and P3 must be enormous indeed.

But a side-point it was; non-essential to the argument

The above paragraph should very quickly and very intuitively make it clear that P3 must be many times over more improbable than P2. But even without that, the very fact that P3 has this one additional characteristic makes it more improbable than P2.

We can safely conclude that P3 is at least a little bit (though one might say “a whole lot”) more improbable than P2.

Enter P 4

Now let’s imagine an entity P4. This entity is just like P3 except it’s not limited to generating only U1 or U2. P4 can generate any of an infinite (or at least a very very very large number) of potential universes.

Just as P3, P4 can intelligently decide which of these universes to generate. But unlike P3, P4 possesses the information for the generation of a huge multitude of universes, together with their sets of laws and constants. P4 is much more complex than P3.

Since P4 contains much more information than P3, P4 is more improbable than P3. It has at its disposal the complex structures of an almost infinite number of plausible universes (perhaps the number is in fact infinite).

Enter P5

P5 is our final entity. It is just like P4 but with an extra twist. P5 is unlimited in its ability. It can generate an infinite number of things, not just universes. There are no limits for P5. For our purposes, we can say that P5 is omnipotent.

In addition, P5 differs from the other P's in that it doesn't transform itself into the Universe; rather it creates it and then continues with its own, independent existence.

P5 is a great deal more improbable than P4, P3, P2, P1 or U1

Given that P5 has infinitely more characteristics and abilities than each of the other P’s, it is clear that P5 must be more improbable than all the other P’s.

And of course, since P2, P3 and P4 are more improbable than the Universe itself, P5 must be greatly more improbable than U1.

P5 is not God

No, P5 is not God. P5 only has some characteristics of God, at least insofar as the Judeo-Christian variety is concerned. The latter has many more characteristics; capacity to love, ability to make moral judgment etc etc etc.

But for our purposes, P5 has the minimum necessary ingredient of God; creational omnipotence.


I have just demonstrated that an entity with creational omnipotence is greatly (one might say infinitely) more improbable than the Universe itself.

What follows is that, even without the additional traits of love and morality etc, the concept of an intelligent and omnipotent entity that created the Universe (aka God) is a great deal more improbable than the Universe itself.

And it certainly is also a great deal (also possibly infinitely) more improbable than P1.

To sum up:

1. The Universe has a cause
2. That cause could be P1; a non-intelligent entity that just gives rise to the Universe and can’t do anything more
3. That cause could also be an intelligent, omnipotent entity, often referred to as God
4. God is a great deal (maybe even infinitely) more improbable than P1
5. Therefore, it is very, very unlikely that the Universe was created by a God
6. In the Judeo-Christian tradition, God is by definition the creator of the Universe (I will not be supporting this; this is obvious)
7. Therefore, combining 5 and 6, its is very, very unlikely that the Judeo-Christian God, as defined, exists.

Obvious objection followup

“Ah, Allo. But you’ve explained nothing. Just how does this P1 come into existence itself?”

I can just see this coming up in post 2.

The answers is: I don’t know. Just like you don’t know how God comes into existence. Saying “God is beyond time” doesn’t help. P1 is also beyond time as we know it, as is any entity that exists outside the Universe (time is part of the Universe). Saying “God is supernatural” doesn’t help. We might just as well call “P1” supernatural in the very same sense.

God could have spontaneously generated. Or always existed. Or created Himself. You can pick and choose. It makes no difference. Why? Because P1 could also have always existed. Or created itself. Or spontaneously generated.

Whilst this doesn’t answer the ultimate question of “where did the Universe come from”, this is not what this argument has sought to do. Wherever the Universe came from, it is very, very, very unlikely that it was created by an intelligent, omnipotent god. An almost infinitely more likely cause (and we did assume that a cause is needed) is the simple (in comparison, of course) P1.

P1; an entity whose origin we can’t explain any better (or worse) than God’s but which nevertheless is equally capable of generating the Universe and which is enormously (infinitely?) less unlikely than the Judeo-Christian deity.