# Probability of God test (comments)

Comments on the Probability of God test – your results, or what you thought about the test.

1. Quick says:

Question 8: Your answer implies that we would be more likely to notice God if she doesn’t exist, which is surely contradictory!

I don’t see the contradiction here.

Question 8
We cannot see God, and no-one has ever designed a scientific experiment that shows he is there. Some people feel God ought to be obvious if he exists, but others feel science cannot measure the supernatural so we should not expect to “see” God. Do you think God not being physically detectible is …..

much more likely if God exists than if he doesn’t?

It should be expected that we cannot see God as the bible states God is a spiritual being. If God is the creator of the Universe we should not expect such an entity to posses physical traits that we who are constricted to this universe possesses. You also have a spelling error on the word “detectable”.

2. unkleE says:

Hi Quick, thanks for commenting. here’s how I see it.

Bayes Theorem, which I used here, requires us to estimate how likely a fact or event would be (i) if a certain hypothesis is true and (ii) if that hypothesis is false. It then calculates how the probability is changed by that fact or event. In this case the fact is that we cannot see God, and the hypothesis is that God exists and the counter hypothesis is that he doesn’t.

As I see it, if God doesn’t exist, we are only slightly likely to imagine we see God. But if he does exist, we are just as likely to imagine we see God, and also slightly likely to see him in reality – so if he exists it is in total a little more likely that we’d see him than if he doesn’t. Since we don’t generally see him, this fact seems to make it slightly less likely that God exists.

Other questions address issues that seem to make God more likely, but this one seems to me to work the other way.

3. Jamez says:

This test doesn’t really work. Each question falls into the same trap eg.

2) Do you think the universe coming into existence would be more likely if God exists than if he doesn’t?

Answering “more likely” does not infer proof as the story of God is about God creating the universe, which could be complete fiction.

Some Aborigines believe chanting from the future generated an egg at the beginning of time, which hatched the universe.

Do you think the universe coming into existence would be more likely if chanting from the future could generate an universe-hatching egg at the beginning of time, than if chanting could not do that?

I’d say the chanting-induced universe egg makes the universe more likely to occur. Doubt you are going to agree that because we have a universe, this proves the eggs existence!

The test is nonsense, I’m afraid.

Thanks

4. unkleE says:

Hi Jamez, welcome to this blog.

Thanks for letting me know your reaction, and I’m sorry you found the test to be nonsense. It wasn’t easy to construct it to give good answers, but I find a major problem is that people don’t follow the instructions well because (I guess) I haven’t been able to write them clearly enough and people don’t understand Bayes Theorem.

So let me analyse your example of chanting creating a universe-hatching egg. To set this up with Bayes Theorem you firstly have to define your hypothesis – which I presume is not simply (1) that chanting has existed or will exist, but also that (2) chanting in the future produced the universe in the past via an egg.

So is the existence of our universe evidence for that proposition? Are you willing to argue that it is? For I am not. Our universe doesn’t seem to allow things to cause themselves in the past. It is true that some quantum events exhibit some strange features, but (i) they don’t cause things out of nothing but only things out of a quantum field, (ii) they aren’t caused by chanting and (iii) they don’t produce eggs.

So I would say the probability of our universe existing via very special chanting is zero, wouldn’t you agree? And so the universe we live in provides me with no reason to believe in the proposition you advanced.

It is quite possible that my little test is poorly conceived and executed, but it doesn’t seem to me that your example demonstrates that, even if I enter into the spirit of it and analyse it. Are you willing to defend it a little further?

Thanks for your input. I thought it was an interesting question, and gave me both some amusement and some head scratching. Best wishes.

5. Jamez says:

Thank you for your response. You say:

“So is the existence of our universe evidence for that proposition? Are you willing to argue that it is? For I am not. Our universe doesn’t seem to allow things to cause themselves in the past. It is true that some quantum events exhibit some strange features, but (i) they don’t cause things out of nothing but only things out of a quantum field, (ii) they aren’t caused by chanting and (iii) they don’t produce eggs.

So I would say the probability of our universe existing via very special chanting is zero”

Our universe doesn’t seem to allow God existing – a supernatural being who is infinite, immaterial, uncaused, omniscient, omnipotent, omnipresent and capable of creating entire universes of planets, starts and life and performing miracles. So you’ve undone your own argument and demonstrated that your God proposition falls at the first hurdle, just as you say mine does.

Bayes Theorem works because it takes prior probability of a proposition as base data and adjusts according to evidence.

Your God proposition is the same as my egg proposition. The prior probability of either proposition is infinitesimally close to zero, most likely precisely zero, rendering the base data we ascribe (50-50: he/it either exists, or he/it doesn’t) completely false. You use the universe as evidence of God, because in your fictitious story, God creates a universe (handy bit of self-fulfilling prophecy aka a logical fallacy). So I too can use the universe as evidence for my egg, as in my fictitious story, the egg creates a universe – just like God! The same can be made true of any evidence, because we just adjust our story to self-fulfil the evidence we see!

Thanks for your response though, it’s nice to see someone of a religious position willing to debate for a change. For what it is worth, as you have probably guessed, I am an atheist. But I have no gripe with people believing what they want/need to believe. I will however counter when people try and miss-use science to prove God; I’m not arguing with you whether or not he does exist here, remember – just that your test fails to prove it.

Thanks,

6. unkleE says:

Hi Jamez, thanks for replying. There is actually quite a bit that you say that I agree with, but not all …. 🙂

“Our universe doesn’t seem to allow God existing …. So you’ve undone your own argument and demonstrated that your God proposition falls at the first hurdle, just as you say mine does.”

But here I suggest you have jumped too soon. Your example hypothesises chanting in our universe which produces an egg which is the start of our universe. Everything is within this space-time universe, whose laws don’t seem to allow all that to happen.

But the God hypothesis is quite different It hypothesises that God exists outside our universe so he isn’t bound by the laws that apply within it, and he doesn’t necessarily operate naturally. So there is no reason to suppose that if he exists he couldn’t create.

So there is a major difference between your example and my belief, and thus my belief doesn’t fall at the first hurdle at all. And the example has shown that you can’t, at least not so far, say “The same can be made true of any evidence, because we just adjust our story to self-fulfil the evidence we see!” But it was an interesting try!

But on other matters I agree.If you set the prior probability infinitessimally small, then of course no evidence will be strong enough to make the probability of God more than 50%, though I don’t know why you would do that – on what basis could you justify that? I also agree that my test doesn’t prove God exists – it was never intended to and nor could it ever. It was designed to encourage people to think and clarify their assumptions and discuss, just as we are doing.

As for God being a fictitious story, that is of course just your opinion, not at all a fact as your wording infers, and my opinion differs from yours. I can live with that.

Thanks again for responding, that’s what that test, and this website, is all about.

7. Jamez says:

“my belief doesn’t fall at the first hurdle at all. And the example has shown that you can’t, at least not so far, say “The same can be made true of any evidence, because we just adjust our story to self-fulfil the evidence we see!””

Oh but it does fall at the first hurdle, and I can say that. I just need to amend my hypothesis to self-fulfil and evidence itself in the same way your God hypothesis does.

The cosmic egg exists outside the universe and is infinite and exists in all time at the same time. The chanting in the universe in the present causes the egg to hatch in the past, creating the universe. This is allowed because the egg exists outside of the laws of the universe, just like your hypothesised God! So we run your test again and we find ourselves with a high or at the very minimum a 50% probability that the egg exists! So you see, the test is fallacious, unless of course I’ve now convinced you the egg really does exist! Again, the same can be applied to ANY STORY we like, all we have to do is write the evidence which will self-fulfil the prophesy.

You seem to be confused regarding my second point and prior probability. Let me explain:

Bayes Theorem applies found evidence to base data, the prior probability. Your test frames the base data at 50-50 of God’s existence. You even say that the minimum chance of God existing using your test would at least be 50%! This is because your question is framed thus: The God hypothesis dictates that if God did exist, he would 100% create a universe. This means that the opposing argument, him NOT existing, only has a 50% maximum prior probability. The same can be said of my cosmic egg – as you can see, in my story the egg 100% creates the universe, so if it exists, the universe is definitely created. Thus:

“Do you think the universe coming into existence would be more likely if the cosmic egg existed, than if it didn’t?”

We are using the existence of the universe as evidence to apply to the base data 50-50 prior probability. But as we both really know, the prior probability is NOT 50-50. The base data would be closer to zero for the egg because it is unfathomably unlikely.

Yet the question is framed in such a way that you cannot help but answer ‘more likely’: it would be more likely. At worst, if you want to resist the trap and say “less likely”, lets apply your own rules from your analysis page swapping God for the cosmic egg:

“It is hard to see how the cosmic eggs existence makes the creation of the universe less likely than if it did exist. Perhaps the probability would have to be at least equal?” Equal. 50-50. At least. That’s the prior probability.

As you can see, the test is simply meaningless as it demonstrates a high probability of my egg existing as much as it does God, using completely fallacious base data of 50-50 prior probability. There is not a 50-50 chance that God or the egg exists. It simply doesn’t work and is completely misusing Bayes Theorem.

But like I said, I’m not out to disprove God here. As you say, we hold different opinions for one reason or another. I’d just hate you or anyone to be convinced of his existence using this test as evidence for his probability when it is clearly illogical!

Try it for yourself. Make up a new story which uses an outside-the-universe caveat and dictates that a universe is created. Apply the test. And realise Bayes Theorem cannot be applied in this manner without accurate, rational and logical base data.

8. unkleE says:

Hi Jamez, thanks for continuing your interest in this topic. But I’m afraid you’re going to have to re-think.

Any hypothesis that involves our universe is immediately up against it for two reasons. (1) If it is part of the universe it is difficult to see how it can be a cause of the universe and hence a cause of itself. (2) We know the laws of the universe pretty well, and if it violates them, then the hypothesis can’t really stand.

Your modified hypothesis still involves chanting within the universe causing an egg to hatch in the past, and that is contrary to everything we know about the universe (as well as about causation).

So you’ll have to modify your hypothesis still further, and make everything that is the cause happen outside the space-time universe. So you could say that an egg that existed outside the space-time universe did the chanting and that created the universe. Then you would have finally avoided this problem.

But then of course, you are more or less postulating God but giving him another name! I can accept that. 🙂

Your second point about the 50/50 prior probability is also mistaken – I did not use a 50/50 prior probability, but let you the visitor choose your own prior (in question 1).

So you can see that both of your objections are based on a mistaken understanding. Perhaps the test reveals more about the truth than you think!!!??? 🙂