Atheism and the existence of peanut butter

[polymath257]

God is generally presented as a first cause. If the theist manages to establish the existence of God as a first cause, it's moronic to ask "who created God?" after that. There is no special pleading in this case.

If the theist shows there is only *one* uncaused cause, that the uncaused cause has a personality (and so *can* design), and, of course, that there is such an uncaused cause. ALL of those are in contention.

OK, how do you support that P(T|T&C)>P(T|U)? What is your event space? What measure do you use on it?

And, more specifically, what are the probabilities of P(T|U&(not C))? How about P(U&(not C))? Without those, it is impossible to compare those relative probabilities.

But you are going further and claiming that P(T|U&C)>> P(T|U). How, precisely, do you justify that claim?

First of all, P(non C) is exactly zero. Because the fact that the universe obeys life-permitting conditions is indisputable, and is generally conceded by both the theist and the atheist, therefore P( C )=1, from which we get P(non C)=0. Since the probability of non C is zero, the probability P(U&(not C)) is zero, and the probability P(T|U&(not C)) is undefined. It's impossible to condition on an event of probability zero.

https://en.wikipedia.org/wiki/Conditiona...ility_zero

Sorry, my bad. I was looking at C=Creator, not what you had assigned.

Now for the justification that P(T|U&C)>P(T|U), that's because the existence of a finely tuned universe makes it more probable that an intelligent designer exists as opposed to a "less tuned" universe or some random vacuum. If you have in front of you a work of art next to a random paint splatter, it's clear that the probability of an artist existing is vastly superior for the work of art, given the skillfulness displayed in the first object.

That deduction only holds if the laws of nature are such that it is the case. For example, with gravity, a 'splattering' of mass is far *less* likely than a spherical shape once the mass is large enough. So the conclusion in that case would be invalid.

Also, for example, evolution selects for adaptive structures, so the probability P((not T)& U & C) might be a LOT larger than you give credit for.

And that is part of the point: that you need to know P( (not T)&C&U) and P( (not T) & U) and the relative sizes of those compared to P( T&C&U) and P(T&U) to say which conditional probability is larger. Do you have any way to estimate the sizes of those?

More specifically, you seem to think that the laws of nature cannot themselves push towards complexity and structure even though we know many examples of exactly that.

I disagree. I don't think there *can be* a coherent 'explanation' for something like that. It is like asking for the cause of causality. The question itself makes no sense.

Why ? That's not the same thing. If a very unlikely scenario occurs, like shooting thousands of bullets at point-blank range and still missing, it demands an explanation. How is this request incoherent?

It depends on the nature of the bullets, doesn't it? And the natural laws that they obey. If, for example, the bullets are much smaller than the spaces between the components of the material they are show at, the likelihood of a miss becomes much larger.

In the specific case I am talking about, asking about the cause of causality is clearly a problem. But, if causality only exists within the universe (which is what ALL the evidence points to), then asking for the cause of the universe is problematic as well.

And to expect an answer to that question means that you already have some regularities. And those regularities *are* examples of natural laws.

You simply cannot have any actual explanations without natural laws of some sort. So to e ven ask for the cause of natural laws is a category error.

It's true that we can't have a physical or natural explanation without natural laws, but this is not the issue. We're not looking for some scientific theory of the universe's existence, nobody can hope to apply the scientific method beyond the observable universe, however we still have the ability to think logically about what caused the universe with all its natural laws, given that the principle of causality is universally valid.

Well, clearly to test our ideas observationally requires that we do observation.

But we have found through long experience that logic alone is a very poor judge of what is actually possible or even what actually occurs.

For example, Aristotle thought it logically necessary that heavy things fall faster than light things. But, the simple fact is that they do not. And the reason is that Aristotles axioms, which he thought were logically necessary, are in fact wrong.

I can give other examples. Kant thought that geometry was a type of synthetic a priori. He thought that because he couldn't image anything other than Euclidean geometry. of course, it was later discovered that there are MANY types of non-Euclidean geometries and that what Kant *thought* was necessary was actually not so.

In particular, there is very good reason to believe that the notion of causality simply is not applicable 'outside the universe' or even to the universe as a whole (all of spacetime). The reason is that causality is always *within* the universe, so the idea of a cause for the universe is against all observation and logic.

Again, isn't it problematic for you to allow for violating causality? 

Not at all. Causality is, first of all, a very vague principle. To become more specific about it naturally leads to certain types of physical laws. And those laws need to be *tested* against observation. it may well be that causality is NOT universal or that it simply doesn't act in the way everyone *thinks* is necessary.

More specifically, quantum mechanics is a non-causal physical theory that works incredibly well. because of that, I am more than willing to think that non-causality happens with some frequency.

What makes you think that 'prior in a causal sense' that isn't 'prior in a temporal sense' is even coherent?

It's perfectly coherent. Take the model of simultaneous causation, let's say A causes B, and B causes C, and both these operations happen simultneously. In this case A is prior to B/B prior to C in a causal sense, but no time has elapsed in this model.

You will have to show an example of such. In this case, I would simply say that A occurred before B, which occurred before C. And that *defines* time in this case.

Why would you even think it is *possible* for spacetime to have a cause?.

Because it's irrational to suspend the principle of causality for the spacetime itself.

I disagree. ALL causality we have ever witnessed in within spacetime. That makes it highly unlikely that it is coherent to talk about the 'cause of spacetime'.

This is not 'suspending the principle' any more than claiming there is an uncaused cause 'suspends' that principle.

More to the point, the question arises whether spacetime 'began to exist'. In particular, was there an event in which spacetime did NOT exist? if not, then the same argument for KCA applies to the universe (which we know exists) and not to God (which we do not).

And, in your scenario, the 'rank' is simply a notion of time. Nothing else. Time is the rank function (to the existent a rank function can even be defined).

Not necessarily, time is a rank function, but not any rank function should be time. As above, you can have a chain of causes and effects arising simultaneously, in which it's meaningless to speak about time, but you can still assign ranks to every element of this chain. There is nothing incoherent about doing it.

I disagree. A rank function of the type you claim would be a *definition* of time.

And, again, time only makes sense within spacetime, so that rank function is *part of the geometry of the universe*. because of that *there is no rank outside of spacetime*.

As above, a rank function doesn't necessarily have to be time.

So now you want to have two rank functions? We certainly know time is one such. Why the extra assumption of another? Doesn't the new rank function simply give an alternative definition of time?[/quote]