Tom -- Welcome!

You raise a valid point, a misunderstanding I should clear up.

Yes, Dawkins here (strangely) grants that "God" could exist, but argues that the theistic God is statisticly unlikely to be that "God" because there are many other possible conceptions of God. He's shifting meaning between two definitions -- the specific theistic God, and anything someone might call "God" -- and I'm simply using "God" in the former sense.

So Dawkins is, in fact, arguing here that the theistic God in specific, being but one conception, is unlikely to be entitled to that monicker* ("God" as highest thing or power) merely because so many other possible conceptions exist.

[* And of course, if the theistic God is not the highest thing, then, by definition, "he" doesn't exist. You'd have something, but not the theistic God. I know you're try to draw a distinction here, but there really isn't one.]

That argument is fallacious, and it's easy to illustrate why: If we doubled the population of the planet, we'd presumably have twice as many possible conceptions of God. (Indeed, this is precisely what he's doing by invoking Alpha Centauri.) If so, would the chances that the theistic God was the "real" God then be cut in half?

By that logic, I could diminish theism's chances further just by writing a computer program to generate new names, and use a dictionary to generate new definitions for "God". By Dawkin's logic, as the possible choices multiply, the chances that the theistic God will "win" the draw -- and thus really be "God" -- will grow smaller and smaller.

Note his exact phrasing: "vanish-ing-ly small" -- transitive, moving, becoming increasingly less probable.

It's the "logic" (or lack of it) underlying his argument that's the problem here, Tom, not his (or my) exact phrasing of the consequence.

A few extra notes, generated while typing the previous response. They get their own comment because I didn't want to muddy the water further...

To be technical (not that anyone has asked) Dawkins is confusing indistinct choices from distinct ones; conflating unconditional choices with conditional ones. If we were blindly picking conceptions of God at random each year, out of a pool of available choices, with each having a uniform weight, then he'd certainly have a valid point.

But that view is wrong for several reasons.

One is that we're not picking at random each time a new possibility is generated. Long ago, when the pool was very small and contained very few entries, one was "tagged", and remains tagged.

Imagine I have three balls: Two are black, and one is red. What are the odds of picking the red ball? 1 in 3. Now let's say I add balls to the urn. (Balls always reside in urns in probability problems, for some reason.) If I add 97 black balls, and pick again, then, of course, my odds are now down to 1 in 100.

But the ball called "theism" was picked a long time ago, when the pool of choices was small, and hasn't essentially changed. We're not re-choosing every year, so it's not like the generation of other apparent possibilities would actually diminishes the odds.

The next problem is that the possiblities aren't uniformly weighted. "Successful" ideas, which work, or at least seem to work for a time, outcompete "unsucessesful" ones, which either hurt those who hold them, or are refuted by more successful ideas.

Theism is certainly a successful idea: most humans alive on earth are theists, in some sense; theistic religions seem to have outcompeted others. That doesn't prove it's right, but it's probably not completely uncorrelated. Theism certainly seems to produce more successful results in adherants (less sickness, greater levels of happiness, better functioning by measurable tests) than some popular alternatives.

[Indeed, if atheism was true, it would be rather unusual, because it would be a case where the wrong model (theism) produced good results, and the right model produced a host of problems. Atheists are, in fact, generally more unhappy, are sick more often, etc. That's not an attack, just a statement of current knowledge.]

We can certainly admit that successful ideas are more likely to be right than easily rejected ones. So it's absurd to treat (for example) Dawkins' flying spaghetti monster, Thor, and YHWH as having equal probability of being correct, since they don't have similar explanatory powers or success rates.

It would be a bit like treating my theory that "objects fall 'down' because dogs exist" as equally likely as Newton's laws of gravity. Yes, they are two different conceptions, but Dawkins would be the first to admit that wide acceptance by many people, if not proof, ought to give us pause, and indicate the two aren't equally likely.

The final problem here, which I mention above, is that the multiplication is merely illusory. Calling the theistic God by two names doesn't make two different gods -- any more than having two words for "planet" means there are now two different sets of planets.

And adding more polytheistic Gods to the pantheon doesn't increase the chance that non-universe-creating polytheistic deities have, among them, the true creator of the universe. (In fact, the odds of that, by definition, will remain zero.)

But Dawkin's core argument clearly implies all of these, and each is clearly fallacious. None of them are very hard to spot -- that Dawkins apparently can't either indicates he's not very bright, or indicates he has no interest in clear reasoning.

There's a fairly well defined set of fallacies. Exactly which one do you think Dawkins has committed?

There's a fairly well defined set of fallacies.

I disagree: there are a fairly well defined set of correct logical operations and transformations. There are an infinite number of wrong things one can do outside that set, and probably an unbounded number of motives for doing so.

There is, however, a fairly well-defined set of common fallacies. There are also various means of classifying similar ones.

But any argument, or form of argument, which can be shown to produce false conclusions from true assumptions can be called fallacious.

Ever hear the formal name for the fallacy of random answer generation? I never have. But it's still a fallacy. I flip a coin and choose the answer the coin delivers: Heads means "true", tails means "false". So far, in my preliminary experiments, it seems to give me good answers about half the time -- putting me ahead of professional stock brokers. ;-)

Tom,

I appreciate your forbearance!

You have an attention to detail that I recognise, but you still haven't got the point!

Actually, to me it looks like we're talking at cross-purposes. I'll gladly engage your argument, but it doesn't seem to be a refutation of anything I've said above.

Dawkins says "There could be something incredibly grand and incomprehensible and beyond our present understanding". And some would say the christian god is "incredibly grand and incomprehensible and beyond our present understanding" therefore Dawkins was talking about the christian god! Case proved! No, case not proved.

Um, Tom? Did you miss the part where I said: "My point here isn't whether Dawkins is right or is wrong about God's existence." You seem to want to debate God's existence. That's not at issue here, friend.

But you also can't prove that any other god anyone has ever worshiped or conceived of doesn't exist...

No, you're wrong here. I agree with atheist Jeffrey Jay Lowder that some kinds of god are indeed disprovable.

Ever hear the formal name for the fallacy of random answer generation?

Would that be "Non Sequiter"?

While I don't have time now to read the article, it seems that Dawkins is right that the burden of demonstrating that any divine creature corresponds to the Judeo-Christian God belongs to theists.

What the two of you seem to disagree about is whether that burden has been fulfilled or not.

While I don't have time now to read the article, it seems that Dawkins is right that the burden of demonstrating that any divine creature corresponds to the Judeo-Christian God belongs to theists.

The burden for any statement rests upon the one making it, obviously. If a Christian claims he can prove God exists, then that incurs a burden of proof. But it's also true that if an old-school God-doesn't-exist atheist like Dawkins makes the opposite claim -- God cannot possibly exist, or almost certainly doesn't -- then that also incurs a burden of proof.

I find this simple idea confuses people endlessly. Last time I mentioned it among a group of atheists they were endlessly troubled: "What? Defend my statements and assertions? But I'm not a theist!"

Duuuhhh...

No wonder atheists are often so intellectually flaccid.

What the two of you seem to disagree about is whether that burden has been fulfilled or not.

No, not in the slightest.

Let me quote myself yet again, and state this yet a third time now:

Um, [Ryan]? Did you miss the part where I said: "My point here isn't whether Dawkins is right or is wrong about God's existence." ... That's not at issue here, friend.

This is not a debate on the ultimate question of God's existence or non-existence, something which has been endlessly flogged, and which we won't settle quickly here. And I'm not here to attack Dawkin's personal religious beliefs, or defend someone else's. This is only a debate on the rationality of his argument, highlighted above.

Please, people, let's learn to read a bit first before commenting, eh?

But it's also true that if an old-school God-doesn't-exist atheist like Dawkins makes the opposite claim -- God cannot possibly exist, or almost certainly doesn't -- then that also incurs a burden of proof.

First, in the passage you cited Dawkins is distinctly NOT making the claim that he can prove that God doesn't exist. The phrase 'at least' indicates he's qualifying his statement, saying absent evidence to the contrary his statement is true. And absent any positive evidence, the liklihood of ANYTHING is vanishingly small.

Going off of that point (burden of proof for negative assertions), a person only has to provide evidence for a positive statement. X exists. To make a negative statement (x probably does not exist) lack of positive evidence suffices for as long as no evidence is produced. An atheist (defined here as someone who believes that there is no good reason or evidence to believe in God, and therefore does not) only has to shoot down any positive evidence to 'win' by default. He does not need to prove a negative. Not that Dawkins is even trying to prove a negative here in the first place. Absent evidence in favor of God, Dawkins' statement would be correct by default. Which is why I said "What the two of you seem to disagree about is whether that burden has been fulfilled or not."

This is only a debate on the rationality of his argument, highlighted above.

His argument, as highlighted above, is completely rational. We are reading you, Tim. There didn't seem to be any additional context in the article that you cited beyond what you posted. I assume there's somthing you've read elsewhere about Dawkins that's influencing your statements but which is not posted here.

Um, [Ryan]? Did you miss the part where I said: "My point here isn't whether Dawkins is right or is wrong about God's existence."

I didn't comment on whether Dawkins was right or wrong about God's existance. You said his logic was faulty, and assumed he was writing a probability problem (which may be true, though I think you're overlooking his qualifier.) I said his logic, as far as it went, was sound. Not irrefutable, of course, but sound. It's possible to make a logical argument for somthing, ignore evidence, and be wrong.

Even proof of a supernatural entity would not automatically mean the existance of an entity with the characteristics of the Judeo-Christian God. You'd still have to build that bridge from "existance of the supernatural" to "existance of Judeo-Christian God." Which is what Dawkins said, and which is true.

I'm not disparaging the positive argument that you made. I agree with it. I'm saying that regardless of whether Dawkins was wrong or not or whether he was ignoring evidence or not, he wasn't being illogical.

To clarify, I agree with the majority of the positive argument that you've made on this topic as part of a different post.