How could mere possibility confer a higher probability? I just don’t understand you here. I’ll illustrate the strength of the claims of the logical vs evidential POE by analogy for you. Let’s use Hitler as an example:
Logical
I’m giving you evidence of his wrongdoing to show that he will go to hell (assuming there was one). You say that since I have no absolute proof that he had a greater good in mind and that he wasn’t commanded by God to do those horrific things so that the greater good could be achieved (it was not achieavable otherwise) then I really couldn’t absolutely say that Hitler would go to hell (assuming there was one). You seem to have a point there, I say.
Evidential
I’m giving you evidence of his wrongdoing to show that it is more likely that Hitler will go to hell than to heaven (same assumption). It is a fact that it is still within the realm of possibility that Hitler had incontrovertible knowledge from God that all those bad things he did was necessary for a greater good. How could that fact confer a higher probability to the statement that Hitler will go to heaven? :smack: You couldn’t even begin to argue that. :rolleyes:
Did you see the post I snuck in above yours? I’m starting to think we’ve been working in different frameworks with the evidential & logical POEs. Neither the logical nor the evidential one would convince a believer, if you buy Leibniz argument (which is what I’ve been arguing), but to somebody unsure as to god’s existence, the evidential POE does provide grounds to reject it one a probabilistic basis (the logical problem, however, still having no impact here).
Half Man Half Wit, I’m not going to do a point by point response this time, as I feel we’ve gone too far down that avenue. In my experience on this board, once posters are giving sentence by sentence responses, no useful information will be exchanged.
So to summarise my position again, I think Leibniz’ argument may work against the idea of omnimaxness and evil being fundamentally incompatible.
But I think it only defeats a “strong” version of POE. A weak version, where we simply state that the large amount of evil in the world suggests strongly that there is not an omnimax god, is unaffected by BPW. In fact, BPW is completely tangential to the “weak” POE.
You seem to dispute that the weak / empiricial / evidential POE works at all, so let’s take it to an extreme. Imagine a world where everything is perfect, and every human feels completely happy and satisfied. Now imagine the exact opposite; a world with the maximum suffering conceivable.
If I were a theist living in one of those worlds, the empirical data of those worlds would absolutely affect how I evaluated the claim that god is omnimax, and the POE.
But if I understand your position correctly, the two scenarios have no significance to the hypothesis whatsoever. If the people in “evil universe” lose their faith because of all that evil, then they’re being irrational, in your view. Right?
Essentially, yes. The believers in both worlds have no logical grounds to conclude their belief false. However, the doubters in those worlds would act differently – more eagerly embracing the existence of an omnimax god in the ‘good’ world, telling somebody approaching the with the notion to fuck off in the ‘bad’ one --, and justifiably so.
The evidential problem, as brought forth to a believer, still suffers the Humean problem: there are no grounds to conclude any given act of evil as unnecessary. Think about the knowledge gap between god and us – how are we supposed to be justified in the induction ‘there does not seem to be a good reason for evil act x, hence, there is probably no good reason for evil act x’ in this case? This induction implies that in general, we should be able to discern the motives behind god letting occur evil act x. However, if an inexperienced amateur watches a chess master play, and doesn’t see a good reason for a move he makes, there’s no justification to conclude that thus, there is none (or think about the scientist saying something that sounds nonsensical to the layman). This doesn’t change when there are a lot of such moves in any game of his. A believer can always point to this, and rationally so.
A doubter, on the other hand – somebody who, in this example, is just as inexperienced an amateur, who however is not convinced that the chess master actually is who he claims to be – might see those apparently bad moves, and reasonably bolster his support for the hypothesis that maybe, the guy’s not actually a good player, but rather an imposter – on pain of possibly being surprised by some brilliant combination securing a win in the end.
I think I briefly mentioned the problem of animals dying horribly as prey, thanks to evolution which does not care about suffering (to personify it a bit.) The argument that natural evil is the result of original sin might have worked 300 years ago, but now we know that the same natural processes causing human suffering today were in operation long before there were any people to make moral choices. If God created them to be ready for us after we sinned, there goes our free will, right? Oddly, the writers of Genesis got this one right, in the context of their myth, since there appears to be no predator-prey relationships or natural evil before the Fall. Good for them for writing a consistent story (in this context,) bad for them for getting all the facts wrong.
I’m not sure I understand the distinction you are making. Now, a trimax god is logically inconsistent because of the limitations omnibenevolence places on omnipotence, but I don’t think that is what you are getting at.
A few posts ago you said that we assume that this world is not optimally good. Actually, we don’t, since we give evidence and plausible scenarios against this supposition. Not proof, of course, but it should be responded to by other evidence. If the response is that this evidence contradicts our world’s position as best, and thus cannot be true, we can see where the true assumption lies.
To put this in the ever-popular mother model, if a mother causes a child to feel pain because of a shot, and someone submits evidence that this means she is not the best possible mother, one can easily provide good reasons why this pain is required and thus does not violate the conditions of optimality. If on the other hand the mother whacks the kid for no good reason, and the response to this showing she is not the best is that it must be justified because she is optimally good, then we can see that the only argument for optimality is the assumption of optimality. If there was only one or a few cases where our world doesn’t seem to be maximally good, we might grant the argument, but there are billions at least.
Or think about it this way: you’re living on the bad world, and suddenly, you’re teleported away, into some land of light in which you feel beyond a doubt the presence of some impossibly powerful, wise and good being. This being then explains to you:“Look mate, really sorry about all the shit’s been happening and all, but I’m actually a cool dude once you get to know me – the reason you’ve all got to go through this is —”, which he then explains to you, and which makes complete sense. In the end, you’re left with justified belief – that also happens to be true – that there is an omniscient, omnipotent, omnibenevolent creator god, and shit just happens because it has to. (This is, of course, just meant to be an example; substitute any other story by which you can come to said justified belief.)
Now, in your belief, no evil you see gives you rational justification to abandon it. However, you telling others probably won’t go over too well, the world being a hellhole and all. But then, some clever chap approaches you, and confronts you with the (logical) problem of evil. Absent any refutation – assuming Leibniz’ argument does not go through–, he proves to you that, indeed, omnimaxness and evil are impossible to coexist; then, the only rational thing to do is to abandon your belief, since it has irrevocably lost its justification. Whatever happened, if anything happened at all, can’t have possibly been what it seemed to be – it’s logically impossible that it was. Nobody bringing forth merely the evidential POE could accomplish the same thing, as you have justified belief that it’s all in the name of some greater good, or whatever.
The Leibnizian argument only gets started if some belief, justified to the believer, has already been established, and it shows that neither logical nor evidential POE should overturn such belief. In the case of prevalent doubt, however, it can’t achieve the same feat – again, if you’re watching somebody play chess who may or may not be any good, you’re perfectly justified in counting his apparently bad moves against his supposed abilities.
This ‘confusion’ is YOURS. Or, more specifically, it’s burned into the soul of the argument you are presenting, which is harping very strongly on the fact that (based on the argument that precedes it) earth being the best possible world remains possible. Possible possible possible. Lovely word, ‘possible’. But what does it MEAN?
From the standpoint of formal logical argument, it means nothing. It means “we have not yet got around to disproving one possibility or the other”. It certainly doesn’t mean “a counterargument to your alternate arguments that it is not possible.”
But that’s exactly how you’re trying to use it.
Your position regarding this argument can be summarized as a group of people discussing the possibility of whether it’s raining.
Person A: Well, we know that it’s either raining, or it’s not raining. So it’s possible that it’s raining outside right now.
Person B: Well, I just looked out the window, and it’s totally dry out there.
Person C: And it’s the middle of summer.
Person D: Also, we happen to be located in the middle of Death Valley.
Person E: And I think I read something in the newspaper about a large dome being erected over this area to prevent rain getting in anyway.
Person A: Look you people, weren’t you listening? I just proved that it’s possible that it’s raining! So obviously anything you say to the contrary must be wrong!
And based on the fact you deliberately chose not respond to it when quoting my post, I take it you agree that the following is an accurate characterization of the argument you are presenting?:
Voyager, I’m a bit unsure as to where exactly in the debate to place you right now. Do you still maintain that the logical POE establishes the impossibility of a tri-omni god, i.e. that Leibniz’ best of all possible worlds argument doesn’t work?
If you do, I’m not sure I can say anything I haven’t already said in order to continue the discussion; to me, the logic seems sound. Perhaps you can point to where you disagree:
The lPOE asserts that there would be necessary no evil if an omnimax god exists. To show this false, one has to show that there would to the contrary not necessarily be no evil in this scenario, i.e. that there possibly would be evil. To do this, it suffices to exhibit a possible scenario in which there is an omnimax god and evil exists. Thus, one postulates that this is the best possible world – that hence, even an omnimax god could not create any better one, if he is to obey some basic principles, such as his inability to actualize true contradictions. Especially, this may be the only possible world – it is consistent with the laws of logic for this to be the case, and hence, a valid possibility. True, it is equally consistent with the laws of logic for there to exist infinitely many better worlds than this one, but the possibility that this is not so needs not be actualized for the construction to work – it needs merely be a valid possibility.
I believe you mean the post directed to Mijin? This merely concerned his particular framing of the evidential problem of evil – he explicitly stated that he assumed that the world is not the best possible one.
If you’re debating the evidential problem of evil, and debating with a doubter, then yes, I’ve by now come around to thinking that’s right. But to somebody having a pre-established justified belief in an omnimax god, the burden of proof is squarely on you to show the inconsistency of the proposition that, for instance, there are no other, or no better, possible worlds.
The distinction I make here is essentially the same as between somebody who knows (or more exactly, has a justified belief) he’s watching Anand play, and somebody who’s unsure as to the skill of the player. Apparently bad moves do nothing to convince the former that Anand’s not a good player, as they shouldn’t; however, they reasonably convince the doubter that he’s probably watching somebody not very skilled. The analogy is flawed in that in reality, there’s no ‘perfect’ chess player; and with god and the world, we can’t just up and wait till the game’s over, but I think it gets the gist across.
“Thus, one postulates that this is the best possible world” - this is where you assume your conclusion.
But that’s not your real problem, which is perhaps-innocent equivocation over the word “possible”, as evinced in “To show this false, one has to show that there would to the contrary not necessarily be no evil in this scenario, i.e. that there possibly would be evil. To do this, it suffices to exhibit a possible scenario in which there is an omnimax god and evil exists.” This statement is utterly incorrect from a logical standpoint; “(Omnimax God) implies that (there are NO possible worlds containing evil)” is not disproven by “(Omnimax God) implies that ((there are NO possible worlds containing evil) or (there ARE possible worlds containing evil))”.
Reducing the statements to variables, the statements are (G -> ~E) and (G and (~E or E)). It is of course impossible to logically derive ~(G -> ~E) from (G -> (~E or E)), as is demonstratable by the following truth tables:
G E | (G -> (~E or E))
T T | T
T F | T
F T | T
F T | T
G E | (G -> ~E)
T T | F
T F | T
F T | T
F T | T
As one can plainly see, there are cases where (G -> (~E or E)) is true and (G -> ~E) is not false, so the first being true is not a disproof of the second.
So, why does the argument presented assert something so blatantly untrue? Well, because the apologist that orginally devised the argument was inconvenienced by the way proper logic kept disproving his god, and so was forced to use improper logic to get around that fact. The better question is, why doesn’t this error seem obvious to all at first glance? The answer is, as indicated above, because it leverages the multiple meanings of the word “possible” to give the word seeming importance in situations where it utterly and completely lacks it.
Review the following statement: “It is possible that this world is the best possible world”. Note that the word “possible” appears in two places in this statement - and they do not mean the same thing. The latter is part of the term-of-art “possible world”. This refers to a single element of the set of "possible world"s, which is the complete set of scenarios that are possible under…some specific set of known facts and circumstances. This is a concept from modal logic and does indead have logical meaning.
It’s worth stressing that which worlds are possible depends on which knowns are in our premise set. For example, the moment an omnimax god rolls onto the scene the set shrinks dramatically; all suboptimal worlds become impossible, since an omnimax god cannot allow them to exist. So if one wanted to frame this argument correctly, one should take care to be clear which set of possible worlds is in play at any point in the argument. Framing the argument correctly was not one of the priorities when it was originally created, though; it deliberately avoids refining the set of possible worlds, as doing so casts light on the farcical nature of the argument’s claim that we might be living in the best possible world.
So anyway, back to the point. The second “possible” in the above statement is part of a term of art and thus is meaningful. But what does the first one mean?
The answer to this is, of course, it means nothing. It is an assertion of ignorance; it is a statement of vacuous truth that contains no information because it eliminates no possibilities. Any conclusion that has this sense of the word “possible” in its conclusion is logically equivalent to an empty statement; it is always true and thus cannot disprove any statement that wasn’t already proven to be false in all cases anyway.
So yeah - one possible is meaningful, the second is meaningless. The arguement under discussion hinges on mixing up the two. Were it written in such a way that preserved the specific correct unambiguous meanings of words, such as in symbolic logic…well, the argument can’t be written in such systems, because when the actual meaning of terms is highlighted it becomes impossible to pretend that it reaches a meaningful conclusion.
I appreciate how this analogy supports your point, but let me try to show you it also supports mine.
Say I have a friend who claims to be a chess grandmaster. He claims to have a rating of 2600. I believe him.
So I watch him play an amateur. After an unconventional opening, he loses most of his pieces rapidly, and now his king is in double check with other enemy pieces nearby.
All this has made me lose confidence in my friend’s chess playing abilities. I can’t rule out that he will somehow win the game, and I’m not about to declare my friend a liar. But nonetheless, what I’m seeing carries some weight.
If it’s irrational for me to think this way, then much of inductive reasoning, and therefore much of the way we think, is irrational.
One objection you may have to this reasoning is that if chess playing is our analogy for a universe “setup”, then humans know nothing about chess. We have no basis to judge moves.
However, we know what evilness is, and we can therefore keep score (“material” in the analogy).
Did I? I don’t recall saying that.
Whether our universe is the best possible or not I would see as an empirical fact, I don’t think it should be an a priori assumption.
Belated note: I whiffed my above truth tables; the final line in each should be False and False for the G and E variables, not a repeat of False and True, obviously. The resulting values for the composite statements in question are correct, though, and my main point is unaltered: that the entire argument from the OP is based on a false statement of what one has achieved when they “prove” that (~E or E) is true.
I’m confused whether you are arguing that the POE is invalid because a no-evil situation is impossible, or whether it includes a minimal evil world instead. I think it might be possible to prove that a no evil universe is logically inconsistent, except in the trivial case, but you’d have to define evil and a bunch of other stuff. If you can show a minimally evil world leads to problems, then the no-evil world is just a special case.
You accept the evidentiary argument, so we can drop that.
What you haven’t responded to (I think - I’m pretty busy these days and just grab a moment or two to reply) is the inconsistency of a minimally evil world with a tri-omni god.
First, a minimally evil world, by itself, is logically possible. Take any world, and any characteristic of the world. See if flipping it is logically possible. If not, the world with that condition flipped isn’t logically possible, and the flip can be ignored. If it is possible, see if flipping it results in a more good world. If it does, do the flip. Now, an exhaustive solution to this problem is slightly NP-hard, but since we are talking omniscience, that is okay. So, I agree with you that an omniscient deity can define (and create) an optimal universe - and must do so if he is omnibenevolent.
However, since he is forced to do so, he is not omnipotent. Any action he takes to change this perfect universe causes increased suffering, and thus violates omnibenevolence. Being forced to maintain this perfect universe violates omnipotence. Thus this god is not omnimax - and evidence about the universe has nothing to do with this.
If a person is going to have any fun with the POE, you have to assume that an omnimax god is possible. (Otherwise it’s a very short discussion.) If this requires omnipotence and omnibenevolence to be resolved in the obvious way (“just because god chose not to do something doesn’t mean he lacked the power to do it”), then that’s the price one pays. (The same answer resolves omnipotence and foresight as well.)
I’m not actually sure I see this as a fraudulent redefinition of omnipotence, by the way - omnipotence is about capability, not the actual course of action chosen. This can be trivially shown by the fact that otherwise omnipotence would require God to do everything, all at once, including all cases of A and ~A simultaneously. I think we can agree that this is not what we normally think of when we think “omnipotent” - so a god that only does what he decides to do could still in theory be omnipotent, so long as he had the cabability of doing anything else that isn’t inherely self-contradictory.
So yeah - I would say that an omnimax god could maintain omnimaxness so long as he ensured that suffering was minimized at all times. (Note: all times. So if the world now is better than it was at any point in the past, then no omnimax god could have allowed it to be in that prior state, proving that no such god exists. Anybody want to say that the total level of suffering on earth has never changed?)
Well, you’re the one messing up basic logical terms and definitions and making bizarre claims about what does and doesn’t matter in logic…
It’s an elementary unary operator in modal logic which, from the post of yours with the truth tables, you don’t seem to be familiar with at all.
So this means you accept every point I raised that you didn’t respond to?
I spoke somewhat loosely; I meant ‘one proposes the best of all possible worlds scenario’.
No. It’s not. It’s the basic relation between the operators for possibility and necessity, as shown for example in the first formula here.
This is simply parsed wrong. (Omnimax god) implies that (there necessarily is no evil) is proven wrong by demonstrating the truth of (omnimax god) and (there is not necessarily no evil), which is equivalent to (omnimax god) and (there is possibly evil). Again, you simply disregard the modalities, which as I pointed out to you before is where your error lies.
If you want to see the correct symbology, with denoting necessity and <> denoting possibility, G -> ~E, the negation of which is ~(G -> ~E) = G & ~~E, which, by the above-referenced formula, is equivalent to G & <>E.
Well, so you’ve at least heard of it.
If you’re using material as an analogy for evil, then also take into account that we can’t actually see all of it. And as I’ve said, there’s no perfect chess player, so there’s always some grounds for doubt – it’s rather impossible to believe in the perfect abilities of some chess player when such a thing is known to not exist. But, consider another version of the analogy, where you have justified belief that one player is indeed perfect – say, a computer that has ‘solved’ chess, i.e. from every situation can predict the winner using the most logical combination of moves, and hence, in every situation can play optimally. Then, I’d suggest you’re far less likely to experience ‘waning faith’ even if he seems to play badly.
You seemed to me to do, in post #135:
I’m sorry if I misread you, or misunderstood the context.
I argue that the POE is invalid because there is no a priori reason for us to conclude that, given the limitation of logical consistency, this is not the best world it was possible for god to create. Only if it could be established that there is, in fact, a better world, and that this world is logically possible, is there any bite to the POE.
Okay, I can square a circle any time I want to, but I choose not to. How does that work for you?
What we do here is to assume that a god is omnimax, and then find an inherent contradiction based on the assumption. In some cases I accept your point about choice - God can choose for us all to do good, but doesn’t because he prefers a universe of free will. There is no limitation on omnipotence there. But my example is different. He is logically capable of interfering in a perfect universe. But, if he does, he as a consequence of his interference makes it not perfect. So he doesn’t actually have freedom of action. Maybe it is more like Russell’s Barber paradox. The Barber can shave himself, but then he violates the rule. He can choose to not shave himself, but then he also violates the rule. Since God cannot choose to change the world without violating the rules, his choice is so limited that calling him omnipotent is absurd.
I thought that was the whole point. the POE is not a problem of all if you allow God to be evil. The Greeks never had this problem. If you reject omnibenevolence as a requirement for an omnimax god, then the problem goes away. In reading the Bible I read as a kid I never saw anything saying god was omnibenevolent - that seems to be a much later creation, but Christians I expect.
Well, if god instantiates this best possible world, then he is omnibenevolent; if he chooses to instantiate any other world, he isn’t.
By analogy, my having two arms is a property similar to god’s omnibenevolence (think two arms to hug you with! :p), and my capacity to chop of one arm is similar to god’s omnipotence: my having two arms does not restrict my capacity to lop one of them of – I could, if I chose to; similarly, my capacity to chop of one arm does not entail that I must be only one-armed. No?
<Wipes tear from eye> Oh, wow, that’s good. I’m impressed. You have some serious self-confidence in the face of overwhelming evidence of facts to the contrary, don’t you?
Yeah, I know, there’s a possiblity you might be able to convince somebody else who doesn’t know me that I don’t know what I’m talking about. Maybe. But honestly I have too high an opinion of the general readership here to seriously worry about even that.
Where the point is not answered by the answer to a prior point in the same or recent post? Yes, that’s actually a fair assumption, if not an entirely safe one. For example, were I to deliberately skip over the main part of a post and respond to a small part at the end (proving that I had seen the post and assessed which parts of it required response), then one could with relative safety assume that I had no problems with it.
But if there is anything I have not responded to that you have uncertatinty about my acceptance thereof, feel free to quote it back as I did. You even have my permission to state your clarification request as an assumption that it had been accepted, as I did.
The entire argument is founded and wholly dependent on speaking loosely. Writing it formally would undoubtedly destroy it.
Allow me to ask, since you profess an awareness of formal logic that transcends that demonstrated in the slopshod rhetorical argument given in the OP, what precise logical function does assuming your conclusion have in your argument. The closest example I can think of in actual formal logic is when you propose the negation of your conclusion as an assumption and then shoot for a proof by contradiction. But since the intent probably isn’t to disprove that there is evil in the best possible world, I have to ask, what do you think the goal of introducing the conclusion as a premise, from the standpoint of formal logic? I’m presuming that you don’t share my opinion that ithe purpose is to make a fallacious rhetorical argument, to be able the untrue in a convincing-sounding manner.
Your symbology is incorrect - though I am partly at fault. I should have blasted down the equivocation about the term “possible world” long ago - I seriously didn’t recognize that you intended to pretend this argument was real modal logic, as opposed to the obvious sham of it that it is.
Remember what I said about the set of possible worlds being dependent on what premises are in effect at the time? Yeah, that’s your error here. What set of possible set of worlds do you suppose is being referred to here by and <>? What frame?
I’ll help you out - the frame in the argument shifts, rather painfully, between “possible without the assumption that there’s an omnimax god” and “possible including assumption that there’s an omnimax god”. It should be trivially obvious that the frame assuming that an omnimax god is present includes only a subset of possible worlds in the frame that doesn’t, since omnimax God can’t allow the worlds with unnecessary evil to exist; the frames certainly aren’t interchangeable.
This cavalier shifting and intermixing of differing frames simply doesn’t happen in proper modal logic - at least, not any variant of it that I’ve ever heard of. In real modal logic, the existence of the omnimax god would be a variable within the possible worlds; not something external to it. Gods don’t create ‘logical’ possible worlds. They simply don’t. All possible worlds exist ex nihilo and outside of reality, suspended in the same ephemeral immutable land that all the rest of abstract logic is.
So. If wanted to write these things in real modal logic, you’d pick a frame and stick to it. And that frame would necessarily be one that included possible worlds with and without gods in them; otherwise you couldn’t introduce the existence of our universe into the argument at all, because you have no evidence that our universe has a god in it, and thus our universe may not be included in the set of possible worlds defined as legal in the argument.
Of course, the argument rather explicitly does not use the term “possible world” correctly (a point I should have made in my prior posts, admittedly). It’s brazenly conflating the logical worlds with physically created universes. And then based on that conflation tries to shoehorn our universe in as one that is possible with an omnimax god (an act which itself assumes the conclusion, depending on which frame you’re in).
I invite you to try to write the argument from the OP in symbolic modal logic - presuming you did it correctly, it should school you pretty hard as to the rampant errors in the argument. Most notably, you would rapidly discover that you can’t write statement 8 without first assuming it as a premise. And even assuming it, I think you’d find there’s no way to write line 13 as well.
(In case you want to try, I’ll start you: the correct symbology is actually (G -> ~E), which is logically equivalent to (G -> ~E), which is what I said it was. Its negation is <>(G & E), which you assume in 8.)
I kick ass at it - enough to recognize that it is never used correctly in any proof for God. This argument makes a particular hash of it, enough so to make it literally unrecognizeable, but even the more famous and at-least-sound Modal Ontological argument is build upon equivocation over the meaning of “necessary” and “possible”, in the premises.