God, Evil, and Best Possible Worlds

[Thomas Metcalf]

Assume that there is a best possible world, or at least, a group of worlds such that it is morally better to actualize any of them than to actualize any of a much larger group. To me, this seems very plausible. Even if there isn't a best possible world, I'd guess there are a set of worlds G and a set of worlds B such that G is a proper subset of all possible worlds and B is a much larger proper subset of all possible worlds, and the unity of G and B is the set of all possible worlds. And it is morally better to actualize any member of G than it is to actualize any member of B. Roughly, only a few of the possible worlds are the sort that a morally perfect being would actualize.

Let us say that G has no more than one one-hundredth the members as B. I think this is a conservative estimate. Given any possible world, it is possible to imagine several much worse worlds. This leads me to believe that there are far more "bad" worlds (or at least "mediocre" worlds) than "good" worlds.

For God to exist, we must be living in a member of G, because only the members of G are the ones that a morally perfect being can actualize. But we must estimate the probability that we are living in a member of G to be at most 0.01. If we had persuasive independent evidence that God existed, this would provide reason to think we are living in a member of G instead of B. The same goes for any direct evidence that we are living in a member of G.

I am aware of no evidence that God exists, or evidence that we are living in one of the worlds that a morally perfect being can actualize. Therefore, I must estimate the probability that God exists to be at most 1%, and the probability that God doesn't exist to be at least 99%. And therefore, positive atheism is justified.

[Satan Oscillate My Metallic Sonatas]

Originally posted by Thomas Metcalf
Assume that there is a best possible world, or at least, a group of worlds such that it is morally better to actualize any of them than to actualize any of a much larger group. To me, this seems very plausible. Even if there isn't a best possible world, I'd guess there are a set of worlds G and a set of worlds B such that G is a proper subset of all possible worlds and B is a much larger proper subset of all possible worlds, and the unity of G and B is the set of all possible worlds. And it is morally better to actualize any member of G than it is to actualize any member of B. Roughly, only a few of the possible worlds are the sort that a morally perfect being would actualize.

Let us say that G has no more than one one-hundredth the members as B. I think this is a conservative estimate. Given any possible world, it is possible to imagine several much worse worlds. This leads me to believe that there are far more "bad" worlds (or at least "mediocre" worlds) than "good" worlds.

For God to exist, we must be living in a member of G, because only the members of G are the ones that a morally perfect being can actualize. But we must estimate the probability that we are living in a member of G to be at most 0.01. If we had persuasive independent evidence that God existed, this would provide reason to think we are living in a member of G instead of B. The same goes for any direct evidence that we are living in a member of G.

I am aware of no evidence that God exists, or evidence that we are living in one of the worlds that a morally perfect being can actualize. Therefore, I must estimate the probability that God exists to be at most 1%, and the probability that God doesn't exist to be at least 99%. And therefore, positive atheism is justified.

...or you could have just saved yourself alot of typing and wrote...
'I assume God doesn't exist' and be done with it.



No need to confuse the issue or make your position sound more reasonable by adding venn diagrams to it.

<shrug>


Satan Oscillate My Metallic Sonatas

[Just_An_Atheist]

"...or you could have just saved yourself alot of typing and wrote...
'I assume God doesn't exist' and be done with it."

Please substantiate your claim that he simply assumes that God doesn't exist, or be quiet.

[Thomas Metcalf]

Originally posted by Satan Oscillate My Metallic Sonatas :

...or you could have just saved yourself alot of typing and wrote...
'I assume God doesn't exist' and be done with it.

"Alot" isn't a word, and the subjunctive you want is "written," not "wrote."

No need to confuse the issue or make your position sound more reasonable by adding venn diagrams to it.

What Venn diagrams?

<shrug>

Yeah, really. I'm not going to respond to you any more unless you can find a tactic other than argument by assertion.

[Jobar]

How many possible worlds are there?

I don't see any possible answers to this question except three. Well, two actually, as one answer is 'indeterminate' and thus not really a valid answer. Or, it could be 'infinite'- and thus there might be infinitely many good worlds, and bad ones. I don't know if we can say that the (infinite) set of bad ones is larger than the (likewise) set of good ones.

Or we can just say 'one' and make things simple, by devaluing all 'possible worlds' scenarios.

I have no idea which of these is true, and I suppose no one does. Thomas, I don't think I've ever seen this argument- is there some alternative I'm missing? (Just an opinion here, but I like the 'many-worlds' interpretation of QM. So my own best guess is that there are infinite (or at least approaching infinite) possible worlds.)

[Thomas Metcalf]

Originally posted by Jobar :

How many possible worlds are there?

A very large number, possibly infinite. But I'm not a metaphysician, and I don't know what the currently accepted answer is.

I don't know if we can say that the (infinite) set of bad ones is larger than the (likewise) set of good ones.

It's not that the set of good ones is smaller than the set of bad ones. It's that the set of "perfect" ones is smaller than the set of "imperfect" ones. I think this is, intuitively, highly plausible. Remember, all mediocre worlds count as imperfect. Do your intuitions agree with mine?

[Jobar]

'Perfect' is one of those absolutes, which as I have said before are intensely slippery concepts. It may be meaningless for us to try to talk about it.

How many things could change about a world, and yet the world still is 'perfect'? Does it matter if a mountain is a few feet taller or shorter, or if an island is a bit bigger or smaller?

Okay, maybe instead of physical perfection, we should focus on the *moral* perfection of the world's inhabitants. Are there multiple moral codes which might be called perfect, or only one?

I'm trying to picture how one might graph the spread of possible worlds from 'most perfect' to 'most imperfect' (evil?). Would it be a normal distribution, with only tiny (relative) numbers of worlds at each extreme, and most falling into the center?

I can't say I'm expecting to get any results out of this line of thought, but it is rather interesting.

[Satan Oscillate My Metallic Sonatas]

Thomas,

Egads...

Originally posted by Thomas Metcalf


Yeah, really. I'm not going to respond to you any more unless you can find a tactic other than argument by assertion.

The irony is devastating.


Your argument is basically...


-Assume A is true.
-Assume B is true.
-Assume C is true.
-IF A and B and C THEN D.

Here...

A = the 'Best Possible World' premise. (ie 'Assume that there is a best possible world...')
B = there are .01 'good' worlds as bad. (ie 'Let us say that G has no more than one one-hundredth the members as B. I think this is a conservative estimate.')
C = evidence for God doesn't exist. (ie 'If we had persuasive independent evidence that God existed...')
D = it is unlikely God exists.

:rolleyes:



This isn't really a sound argument. You simply fabricated some verbage around your assumptions and are passing this off as valid reasoning.



Want proof? What would happen if someone simply didn't agree with premises A, B and C? Would your argument hold any weight?



Didn't think so.



Satan Oscillate My Metallic Sonatas

[bd-from-kg]

Thomas Metcalf

The odd thing about this argument is that it is based on no evidence whatsoever. Is it valid to draw ontological conclusions based on no evidence?

Let me explain what I mean by this. Imagine (per impossible) a being who had no evidence that had any bearing on any ontological question. He could make your argument: he would have no evidence that God exists, and no evidence that he is living in a “G-type” world. If your argument is sound, he could validly conclude that the probability that God does not exist is at least 99%. But this would be an ontological conclusion based entirely on the nonexistence of any evidence one way or the other. Something seems wrong here.

Perhaps you’re just not phrasing the argument properly. Possibly what you mean to say is that given the amount of information we have about the universe, if this were a G-type world we’d expect to see some evidence that it is, and we don’t. And since there’s no direct evidence (or evidence of any kind for that matter) that God exists, we’re entitled to conclude, etc.

Also, as I pointed out on another thread, arguments based on “counting” the number of possible worlds of various types are almost always invalid. For example, the classes of G-type and B-type worlds are (IMO) obviously infinite, and if these classes have a cardinality at all (which is doubtful) they’re almost certainly equal. Which means that statements such as “there are at least 100 times as many B-type as G-type worlds” are meaningless.

[Thomas Metcalf]

Originally posted by Satan Oscillate My Metallic Sonatas :

What would happen if someone simply didn't agree with premises A, B and C? Would your argument hold any weight?

That's a problem with any argument. Do you have any specific reasons to deny any of my premises?

[Satan Oscillate My Metallic Sonatas]

Originally posted by bd-from-kg
Thomas Metcalf

The odd thing about this argument is that it is based on no evidence whatsoever. Is it valid to draw ontological conclusions based on no evidence?


This is (more or less) where I'm coming from. Sure you can believe those things if you wish...but they are basically just assumptions.



Satan Oscillate My Metallic Sonatas

[Satan Oscillate My Metallic Sonatas]

Thomas,
Originally posted by Thomas Metcalf
Originally posted by Satan Oscillate My Metallic Sonatas :

That's a problem with any argument. Do you have any specific reasons to deny any of my premises?

Sure.

A-the 'best worlds' premise is just silly.
B-G=.01B? Uh...<shrug>.
C-I see plenty of evidence for God.



It's perfectly fine for you to not believe in God Thomas, but there is no reason to try to rationalize it AFTER you've made the above assumptions. That's just churn.


Why not just say 'I see no evidence for God' and be done with it?



Satan Oscillate My Metallic Sonatas

[Thomas Metcalf]

Originally posted by bd-from-kg :

The odd thing about this argument is that it is based on no evidence whatsoever. Is it valid to draw ontological conclusions based on no evidence?

Well, almost. I would love to have an a priori argument from evil, but mine does seem to require a modicum of empirical observation: this isn't obviously a best possible world. If it were, we'd rate the chance we're in G to be much higher.

If your argument is sound, he could validly conclude that the probability that God does not exist is at least 99%. But this would be an ontological conclusion based entirely on the nonexistence of any evidence one way or the other. Something seems wrong here.

The more I think about it, the more I wonder whether my argument could actually be presented as a priori. I don't really depend on the observation that this isn't a G-type world. I mean, such an observation would surely undermine my argument, but I think I can leave it out for now.

Perhaps you’re just not phrasing the argument properly. Possibly what you mean to say is that given the amount of information we have about the universe, if this were a G-type world we’d expect to see some evidence that it is, and we don’t. And since there’s no direct evidence (or evidence of any kind for that matter) that God exists, we’re entitled to conclude, etc.

I don't think that really captures the motivation behind my argument. I'm saying it's unlikely that this world is G-type, based on facts about what a G-type world is. But I do think it's true that if this were a G-type world, we'd probably see some evidence of it.

For example, the classes of G-type and B-type worlds are (IMO) obviously infinite, and if these classes have a cardinality at all (which is doubtful) they’re almost certainly equal. Which means that statements such as “there are at least 100 times as many B-type as G-type worlds” are meaningless.

I think this is the only real problem spot for my argument. You will agree, I hope, that if there is only one best possible world (if G is a unit set), the present objection is nugatory.

If G and B are infinite, however, it's harder. But would you agree that if I can show that for any member of G, there is a unique member of B, and an additional unique member of B, then I have proven B to be more pollent than G?

[Jobar]

If God is supposed to exist in *all* possible worlds, then either there should be *no* less-than-perfect worlds, or God is not omnibenevolent. I think that even one bad possible world provides fodder for the classic PoE, and so is all we need to posit anyway. Since we can't number what is unnumbered (and, yes, unevidenced) we can't arrive at any sort of percentage figure.

Thomas, your notion has inspired an idea which is worth its own thread. Look for 'Perfection and the LPOE' please.

[bd-from-kg]

Thomas Metcalf:

I'm saying it's unlikely that this world is G-type, based on facts about what a G-type world is.

Well, on further reflection there seems to be more wrong with this argument than I thought at first. For example, consider this argument, which seems to me to be based on the same kind of idea:

Take any proposed “physical law” – i.e., a proposition that some specific regularity holds absolutely without exception. (We have in mind, of course, the kind of law that can be tested experimentally and which can fail locally in the vicinity of any given point in spacetime, so that the number of ways that it could fail is virtually unlimited.) Call a world “L-type” if this physical law holds, and N-type if it doesn’t. It seems intuitively clear that there are far “more” N-type worlds than L-type ones in some sense.

What’s more, given virtually any amount of evidence, the number of N-type worlds consistent with it is far greater than the number of L-type worlds consistent with it. Even if we had a large number of confirming observations in the near vicinity of every point in spacetime (and no disconfirming ones) there would still be far more N-type than L-type worlds consistent with the observations, because there are far more ways that a regularity can just barely fail to be absolutely exceptionless than there are ways it can be absolutely exceptionless (just as there are far more ways to just barely fail to win a lottery than there are ways to win).

So according to an argument that parallels yours, no matter how much evidence we have that a given physical law holds, based solely on “facts about what an L-type world is” according to your argument we would be justified in concluding that it is far more probable that the physical law in question does not hold.

But would you agree that if I can show that for any member of G, there is a unique member of B, and an additional unique member of B, then I have proven B to be more pollent than G?

No. See Cantor. (Given any such matchup it’s easy to define a different one such that for every member of B there are two [or any other finite number, or even a countable infinity] of “corresponding” members of G.)

[Thomas Metcalf]

Originally posted by bd-from-kg :

So according to an argument that parallels yours, no matter how much evidence we have that a given physical law holds, based solely on “facts about what an L-type world is” according to your argument we would be justified in concluding that it is far more probable that the physical law in question does not hold.

Well, this is a parody, but I'd be interested to see just where you think my argument goes wrong, were you to grant that there are more B-type than G-type worlds.

My argument goes forward a priori, but it is disconfirmable by empirical observations. One might compare it to an ontological argument that proved a perfect being existed, disconfirmed by the imperfect universe (if the universe is imperfect). We can conclude the argument went wrong somewhere because of a disconfirming example. So the (alleged) facts about there being more B-type universes than G-type universes lead us to decide a priori that we are probably not living in a G-type universe, unless we find some reason to think we are. It's simply a probabilistic weight. And I think a large number of observations that the physical law in question seems to hold provides outweighing probability that we are living in an L-type world, chiefly because most observations are not observations of exceptions.

Given any such matchup it’s easy to define a different one such that for every member of B there are two [or any other finite number, or even a countable infinity] of “corresponding” members of G.)

I think my argument, assuming that your above objection fails to militate, will be sound if it is discovered that there is only one world (or some finite number) that a morally perfect being can actualize -- that is, if G is a unit set or finite. But I agree that it's much harder to advance if G is an infinite set.

Let me ask you these. Suppose it were possible to run a computer function to pick some random natural number, but the upper bound was infinite. Do your intuitions say that chances are, the number it picks won't be prime? Is there a principled way to decide?