Can theists be rational?

[Ivor the Engineer]

"Relaxed" sounds pretty generous! Pull a number out of thin air (no calculations and don't even say what that number actually means), and then do math with it to "prove" something is "relaxed" to the point of illogic.

I agree. Without defining what God is, any probability being assigned to it existing is nonsense. For example, when the probability of life on alien worlds is calculated, the alien life-forms are assumed to be vaguely similar to what we have experience of life on earth.

Seriously, in this type of "probabilty" what does it mean to say there is a 1 in a million probability for the existence of God? A million whats? Is it sort of like a weatherman's prediction (ignoring any calculation for how he came up with it)? If there's a 50% chance of rain tomorrow, it means that on days when conditions are as they are now, we expect that 50% of them will have rain the following day. So does the God one mean that when we've got universes like this one, we expect that in one of them out of every million of them, God exists?

The probability of a god existing without a definition of what God is is meaningless.

And doesn't that assume the existence of God (in some universe) the same way the weatherman's prediction assumes the existence of rain?

Once a definition has been provided of what God is, one can assign subjective probabilities to express one's degree of belief in its existence.

If you can live without "million" corresponding to anything, and say it's really only talking about God in our universe, you've still got the problem that I mentioned. If God does not exist, the probability God exists is zero. The Forster & Marsten argument denies the possibility of the probability being zero in the premise.

More importantly it doesn't even state what God is. For all we know it could be a super intelligent alien who can support our conciousness after we die outside our brains on some impossible (for us) to imagine technology.

Perhaps the question should be what is the probability an alien life-form exists that we cannot tell the difference from our ideas of what a god would be like?

Even for Bayesians, there surely must be a difference between talking about probabilities of events and probabilities of existence (when you're making an argument on the question of existence).

Bayesians are honest statisticians. They explicitly state that the probability of a hypothesis given a set of data is dependent on the probability of the hypothesis being true in the first place. I.e. Ultimately, all probabilities are subjective.

 

[Kthulhut Fhtagn]

Then you should be reading other posts in this thread.

Again, you'll forgive me if I won't allow the directions that others are heading off away from the OP to dictate the questions that should be asked.

Accordingly, I more exist to pour scorn than argue logic on the subject, and why would I bother when others are so eloquently writing the reasons why it's illogical?

By all means pour scorn, but I won't consider it anything other than simply being irrational.

 

[fls]

@Joe - I agree. I'm trying to find different ways of saying the same things.

The Odds for God, given that life is present equals:

Odds for God x p(life in the presence of God)/p(life in the absence of God)

I think everyone is thinking that life is more likely in the presence of God? So the likelihood ratio (LR) (the ratio of the probabilities), will be greater than 1. How much - 2, 10, 100, 1000, 1,000,000?

You can see that whether or not the LR gets the Odds for God up to even odds depends upon what odds you decide to start with, and what value you give to the LR. And that pretty much anything from one in a million to a million to one is possible within the constraints of our wild-assed guesses. I.E. even if we all agree that the presence of life increases the Odds for God, it still doesn't narrow it down to a range that is even remotely useful.

Linda

 

[Ivor the Engineer]

Linda,

Who came up with the definition of a god in your sig?

 

[Bri]

I agree. Without defining what God is, any probability being assigned to it existing is nonsense. For example, when the probability of life on alien worlds is calculated, the alien life-forms are assumed to be vaguely similar to what we have experience of life on earth.

But isn't that just a rejection of the premise?

If one were to use Drake's equation as an argument for the existence of aliens by assigning values that result in an output showing a probability greater than 50% of the existence of aliens it would be a similar situation wouldn't it? You can reject one or more of the premises (even by claiming that one of the values is pure speculation), but the argument itself isn't incoherent.

Would such an argument qualify as circular reasoning (i.e. would the conclusion be assumed in the premise)?

-Bri

 

[JoeTheJuggler]

On a regular die, tossed fairly, the probaility of any number is approx. 1/6.

Very good. Now on an otherwise regular die with the numbers 1, 2, 4,5,6, and 7, what is the probability of rolling a 3?

You said before that it is not zero. You still buy that?

 

[JoeTheJuggler]

The probability of a god existing without a definition of what God is is meaningless.

Amen to that. I think that was actually my first response to the argument.

I'd still like to know what the "1 in 1 million" means. Assume we're talking about the Catholic God as defined by the Baltimore Catechism (which I quoted above).

If someone said to take as a premise that the probability that that God exists is a 1 in 1 million, what do the numbers mean?

Also, the second premise, that the probability that God created the universe is also 1 in 1 million--surely that begs the question, doesn't it?

At any rate, the argument is still a validating argument in that even if you put in something that was well-defined in place of the word "God", you'd come out with the same conclusion about that thing's existence. You can't really put anything in there where "God" is and not get the near certainty that the thing exists in conclusion.

I know, someone's going to say that you can't put "a square circle" in the place of "God". But that's because the term "square circle" is internally contradictory--not the rest of the argument. It is also trivial to point out that you can't put the term "P and not P" in there in place of God. With self-contradictory things, we know already that the probability of the thing's existence is zero.

For things whose existence is unknown, we don't know what the probability is, and we can't invent one without making assumptions about its existence.

In Linda's million-sided die, you can't say what the probability of getting the "0" is unless you know exactly what it is you're trying to find out, "Does "0" exist?"

 

[Ivor the Engineer]

But isn't that just a rejection of the premise?

It's a rejection of using undefined terms in a premise. Is it your idea of God, mine, Joe's, Linda's, etc. which is being assigned the prior probability of 1 in a million?

If one were to use Drake's equation as an argument for the existence of aliens by assigning values that result in an output showing a probability greater than 50% of the existence of aliens it would be a similar situation wouldn't it?

Yes, but we have more experience with life-forms than we do with gods, so that would tend to make our estimates of the probabilities more accurate for a definition of a life-form we have some familiarity with.

You can reject one or more of the premises (even by claiming that one of the values is pure speculation), but the argument itself isn't incoherent.

The argument is nonsensical because 'God' is not defined. As it is, the term 'God' is just three letters of the alphabet next to each other.

Would such an argument qualify as circular reasoning (i.e. would the conclusion be assumed in the premise)?

-Bri

The premise is not a premise because it does not define what the term 'God' means.

 

[JoeTheJuggler]

Bayesians are honest statisticians. They explicitly state that the probability of a hypothesis given a set of data is dependent on the probability of the hypothesis being true in the first place.

You mean they're honest about their arguments being circular? Because the people defending this argument on this forum have been reluctant to admit that. cj at least admits the argument is meaningless and that it's a GIGO thing. Although he also said it is a rational argument for God (which I take to mean that it's reasonable to believe God exists based on this argument).

I.e. Ultimately, all probabilities are subjective.

All probabilities in Bayesian statistics, right? There's nothing subjective about the 1:6 probability of rolling a 3 on a fair die.

 

[JoeTheJuggler]

The premise is not a premise because it does not define what the term 'God' means.

I agree, but I don't think that's the only problem with the argument.

Consider the one I proposed which removes the question of the undefined term "God".

We've got an otherwise normal (fair) six-sided die, but we don't know what 6 numbers are on the faces. We don't know if there is a 3 on there at all. We have no evidence of the 3 being on there, and no way of testing it to see if a 3 is on there.

Now if a Bayesian wants to make an argument for the existence of the 3 (or even the probability of a 3 being rolled), and he starts his argument by saying there is a 1:6 (or 1:12 or 1:1 million) probability of rolling a 3, what does that figure mean?

Since the case of the non-existence of 3 means the probability of rolling a 3 is 0, taking any non-zero probability of 3 assumes that a three is on the die.

If you then do further arguments and somehow come to the conclusion that rolling a 3 is highly probable (basically concluding that 3 exists), don't you think this is circular reasoning?

ETA: Mine is the over-simplified die analogy. Linda's die with an unknown number of faces also works for this question.

 

[fls]

Linda,

Who came up with the definition of a god in your sig?

A brilliant, yet under-appreciated amateur philosopher who makes a mean beef stew.

Linda

 

[cj.23]

Right, as regards the problem of defining God - I took the argument out of context from it's original place in a (roughly 450 I page i think) book in which the authors very clearly define their notion of what God is. As my sole point was to demonstrate one can make a rational argument -- not one i would choose to employ, but nonetheless highly interesting -- for a God. Note rational is not the same as correct.

Anyway, Having stripped the argument from it's context and not having the book in front of me i think I should acknowledge that any fuzziness is my fault, NOT the authors.

And for anyone still interested in Bayesian Epistemology, and critques thereof, I found this earlier
http://plato.stanford.edu/entries/epistemology-bayesian/

I think it might at least help you see what Malerin and i are saying...

cj x

 

[Ivor the Engineer]

<snip>

For things whose existence is unknown, we don't know what the probability is, and we can't invent one without making assumptions about its existence.

<snip>

I'm of to bed in a minute, but what probabilities do we know with certainty for events in the real world?

Can't there be an infinite number of possible hypotheses to test in any experiment?

If we (well, I, at least) exist as a simulation on an alien's computer, that alien (and any other entities with administrator rights) is God.

 

[Ivor the Engineer]

A brilliant, yet under-appreciated amateur philosopher who makes a mean beef stew.

Linda

Why isn't your housekeeper a member of the forum if she can come up with eloquent ideas such as that one?

 

[Malerin]

Very good. Now on an otherwise regular die with the numbers 1, 2, 4,5,6, and 7, what is the probability of rolling a 3?

You said before that it is not zero. You still buy that?

It's not zero. Only logical contradictions are zero. There's a possibilty the atoms on the die might spontaneously reearange themselves, a latent psi-ability might manifest itself by changing the die numbers, etc.

Only a logical contradiction (or possibly mathemataical one) can be given a zero value.

 

[Ichneumonwasp]

Joe,

A way to forge a meeting of minds might be to examine closely the idea of logical contradiction.

There may be hidden logical contradictions in any magical solution.

For instance, one common example is iron floating on water. Since we can imagine iron floating on water, we cannot call floating iron a logical contradiction (you can't cheat by having the nature of water change or having the nature of iron change or have the iron on a shelf so that it only looks like it is floating). But is that really the case? We can imagine something that looks like a bar of iron in a vague sense floating on water, but since the definition of iron, in some sense, includes its specific gravity and the definition of water, in some sense, includes its specific gravity, is it really even theoretically possible for iron to float on water? Or do we just fudge things in our minds?

Don't know the real answer, but a thorough examination of the details might answer it.

Paul A has a paper about this somewhere using this example (iron floating on water), which is where I got it.

 

[JoeTheJuggler]

I'm of to bed in a minute, but what probabilities do we know with certainty for events in the real world?

If the probability of getting a 3 on a die is 1:6, then the 3 must exist. That's certain.

My point is that events and existence really aren't the same thing. This argument treats them as if they're interchangeable when it's convenient.

 

[JoeTheJuggler]

It's not zero. Only logical contradictions are zero. There's a possibilty the atoms on the die might spontaneously reearange themselves, a latent psi-ability might manifest itself by changing the die numbers, etc.

If that's true, why isn't it true about the probability of getting a 3 on a regular die? You said that it was 1:6. That means you're claiming now that it's impossible for the magic to happen resulting in getting a 7 or a Q or a pigeon.

Why is it possible when the 3 isn't there, but gone when it is?

My point is that you're being contradictory.

You said when it's a regular die (with the 3), the probability of getting a 3 is 1:6. That means there's one possible successful outcome (the 3) and only 6 possible outcomes in total. If the magic is possible, then the probability should be 1: infinity. (1 three: an infinity of possible magically formed non threes.)

ETA: Actually, since the magic makes an infinite number of non-three values possible AND makes the magical creation of additional threes possible, I guess the probability is infinity:infinity which is the same as 1 (I guess). Which means the numbers on the die before you start are completely irrelevant since you'll always get a 3--unless you make the same argument for a different number.

So, is the magic always possible, or only possible when it's convenient for your argument?

 

[articulett]

Here's the problem with the die analogy.

It's a really big die. We don't know how many sides it has, but it's probably at least several thousand. Nobody's seen the whole thing, we just get a quick glimpse of the side that lands 'up' each time it's rolled (and even then, we're not sure that it's the whole side). We've rolled the die billions of times and kept track of the numbers that come up. We haven't seen any new ones for quite some time. Every once in a while someone claims to have seen a new number, but it only seems to happen when nobody else is looking. Occasionally, we get a better look at one of the old numbers and it turns out that it was part of a longer number. We also occasionally discover that what we thought was two separate numbers was two different partial glimpses of one number. A zero has never come up.

What is the probability that you will roll a '0'?

Linda

What's the probability that you'll roll a Q or a frog? What's the probability that you'll roll an invisible entity?

 

[Malerin]

Joe,

On a regular die, tossed fairly, the probaility of any number is approx. 1/6.

1/6 is .16666666666666666666666667
The real odds (when factoring in possibilities like atoms rearranging or God interveneing) might be .16666666666666666666666666
(actually, it's my position that the real odds are ultimately unknowable, but I don't want to derail this into radical skepticism)

So, if I don't use the word "approx" all the time, just assume it's there. A fair coin is not 50/50, no matter how fair it is because there's always the possibiltiy something weird may occur to turn it two-headed or two-tailed. A fair coin is approx. 50/50. All probabilities are approximates.