Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Arandur comments on How to Convince Me That 2 + 2 = 3 - Less Wrong

53 Post author: Eliezer_Yudkowsky 27 September 2007 11:00PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (390)

Sort By: Old

You are viewing a single comment's thread. Show more comments above.

Comment author: Arandur 27 July 2011 09:10:31PM 0 points [-]

I'm going to go ahead and offer a differing opinion regarding Christianity. o_o; First of all, much of the "God is vile and insane" protest can be mitigated by remembering that, if the Christian God exists, then this life is a tiny piece of our entire existence, and that death is far from the worst possible outcome. If you try to justify the Christian God's ethics based on conclusions you've reached by disbelieving in Him, your premises are contradictory.

And second, God has indicated many times that He does not change. (Yes, a few of those were not from God directly, but it doesn't matter). So I don't know why a brief stint on Earth would change His mind, particularly about morality, since He's supposed to be infinitely just...

I just... no, I really am curious now. This doctrine seems to be all sorts of self-contradictory. In which church did you grow up?

Comment author: Desrtopa 28 July 2011 04:55:00AM *  8 points [-]

I'm going to go ahead and offer a differing opinion regarding Christianity. o_o; First of all, much of the "God is vile and insane" protest can be mitigated by remembering that, if the Christian God exists, then this life is a tiny piece of our entire existence, and that death is far from the worst possible outcome. If you try to justify the Christian God's ethics based on conclusions you've reached by disbelieving in Him, your premises are contradictory.

If you try to justify the Christian God's ethics based on the bible's own assertions of his justice and infallibility, your argument is circular. Conclusions based on belief or disbelief in God are irrelevant to judging his morality according to the portrait painted in holy texts, we already have our own moral toolkit to work with.

I don't see how it remotely mitigates any charges of vileness or insanity to posit that God hands out rewards of eternal bliss or punishments of eternal suffering for beliefs and/or actions without making his expectations clear. That's negligence beyond the scale of criminality. No matter how sure you are that you understand what God wants, if he had actually made his message clear and well evidenced, your beliefs would not be a minority worldwide. Any intelligent and competent human who wanted to get a coherent message across could do better. Hell, scholars have better agreement on what Nietzsche meant, and he was being deliberately obscure.

Comment author: lessdazed 28 July 2011 11:13:04AM 4 points [-]

Conclusions based on belief or disbelief in God are irrelevant to judging his morality according to the portrait painted in holy texts

As morality is entangled with reality, (incorrect) conclusions about anything are potentially relevant to judging morality.

I don't see how it remotely mitigates any charges of vileness or insanity

I wonder how readily believers would, in a sort of plea bargain, accept and embrace the charge of insanity, with all its connotations, if it were thought the obviously attractive horn of the dilemma, with no other alternatives.

I'm imagining what religious institutions and literature would look like if, say, one out of every ten adjectives referring to a deity in hymns, psalms, etc. were replaced with "insane" or one of its synonyms.

Comment author: Arandur 28 July 2011 06:25:08PM 1 point [-]

snerk I'm sad to say that the answer is "rather readily, for some". I've gotten into heated debates - heated because they ended with me throwing my hands up in the air, appalled at the willing ignorance of my opponent - about whether God follows the same basic rules of logic that we do. It's astonishing. I'm sure if you were to start a Protestant movement saying that "God is insane, but we follow Him anyway", you'd get followers by the truckload.

By Occam's razor, we must conclude that the basic tenants of logic we have in this life will not cease to be true in the next. Then again, Occam's razor depends on Occam's razor for its veracity... :3 If you wanna talk about circular arguments.

And I agree with your statement re: incorrect conclusions; thank you.

EDIT: Man, this 10-minute lag is killing me. O_o;

Comment author: christopherj 18 October 2013 08:16:21PM 0 points [-]

Just out of curiosity, if you for some reason valued above all else, faith (of the sort where the more evidence the less faith), how would you go about maximizing it? If you made the faith too universal, that itself might be evidence of its truth (and thereby reduce the value of the faith). Can you prove that there is an absolute morality, such that valuing faith above rationality is necessarily immoral (as opposed to really inconvenient, or seems immoral to most people)?

PS: Many Christians get annoyed if you suggest that you can just believe in Jesus after you die, like people who lived before Jesus. They'll get even more annoyed if you ask them where in the Bible it says otherwise.

Comment author: RobbBB 18 October 2013 09:07:06PM *  0 points [-]

if you for some reason valued above all else, faith (of the sort where the more evidence the less faith), how would you go about maximizing it?

To optimize that kind of faith, just put every person in a separate universe tailored to contain maximally weak evidence for a proposition P, then make them insane (e.g., with torture, brainwashing, or innate cognitive biases) until they delusively believe in P despite the lack of evidence. Since each person gets their own pocket universe, they needn't be exposed to any other person capable of communicating to them supporting evidence for P.

Can you prove that there is an absolute morality, such that valuing faith above rationality is necessarily immoral

Define "absolute morality" and "prove". If nothing can even be evidenced about morality, then I can safely ignore the question and just ask about what actually benefits people. Believing true things generally helps people have happier, more peaceful lives filled with more love and beauty; is the lunatic in the asylum a paradigm of human flourishing?

If you think that's not universally the case, then OK. But you must think so because you have actual evidence for the value of faith. If I have no evidence that Zeus is Lord but think that's OK because faith is a virtue, I still need to provide some evidence that faith in the proposition 'Zeus is Lord' is a virtue. If I appeal to faith in faith, as opposed to just faith in Zeus, then I've made the content of faith arbitrary, which means that any religious claim can be justified, not just my own religious claim.

Comment author: christopherj 18 October 2013 10:49:34PM 0 points [-]

To optimize that kind of faith, just put every person in a separate universe tailored to contain maximally weak evidence for a proposition P, then make them insane (e.g., with torture, brainwashing, or innate cognitive biases) until they delusively believe in P despite the lack of evidence. Since each person gets their own pocket universe, they needn't be exposed to any other person capable of communicating to them supporting evidence for P.

In other words, little to no practical difference from what we live in. There's widespread belief in God (very weak evidence but excellent privileging of the hypothesis), the threat of future torture (and current social pressures), indoctrination from an early age, and we certainly have plenty of cognitive biases (many of which are geared to preferring things with positive consequences). The main difference with your scenario is the interaction with other people -- but it can be argued that other people are very poor evidence for the existence of God, compared to eg providing a non-theistic explanation to where you came from.

I'm pointing out that P(the world we live in | a God that values faith) is very high, and essentially indistinguishable from P(the world we live in | there is no God). This is no coincidence, as otherwise people who believed in a God that values faith would be constantly surprised by the world we live in. Note that if someone who believes in a God that values faith, claims to have found strong objective evidence for the existence of their God, they're coincidentally also claiming to have found strong evidence that said god doesn't value faith (ie, evidence for some other god).

(I'm not trying to misuse the representativeness heuristic, just pointing out that if our world wasn't representative of some popular god, belief in that god would die out, like with Zeus.)

Define "absolute morality" and "prove". If nothing can even be evidenced about morality, then I can safely ignore the question and just ask about what actually benefits people.

If nothing can be evidenced about morality, than what you just did is called "begging the question".

I wrote my above post because I was intrigued by Desrtopa's claim that it would be immoral and/or insane for God to hand out epic punishments and rewards based on [the person's unevidenced faith] without making it clear to everyone that He wanted [unevidenced faith].

Comment author: RobbBB 19 October 2013 12:33:18AM *  0 points [-]

In other words, little to no practical difference from what we live in.

.. No. If our universe were optimized for maximal faith in Christianity, there would be no Christians other than you, much more conclusive empirical evidence (e.g., impossible-to-fake video footage) showing that Jesus did not rise from the dead, etc., and your brain would just come pre-wired to believe in Jesus with equal and maximal confidence no matter what evidence you encountered. Basically your brain would just be a circuit that chants 'JESUS IS LORD', associated with a strong conviction-feeling, and seeks out and absorbs as much Christianity-refuting evidence as possible, without having any mechanism for modifying the 'JESUS IS LORD' conviction pump in your brain.

If built-in beliefs aren't allowed, then the second best step would be to literally torture you -- like, you've never lived a day of your life without undergoing something roughly as unpleasant as scaphism -- until you're driven literally insane and can do nothing but grind your teeth and gibber endlessly about Jesus being lord. That's much, much closer to what a universe optimized for faith (and faith alone) would look like. We aren't literally in such a universe; any comparison of your own day-to-day existence to actual torture must suggest either a loose grasp on the English language, or a fundamental misunderstanding of just what real torture is like.

Certainly vague, hand-wavey threats are not optimizing for unjustified-belief. Threats can't drive most people past the brink of madness, arguably can't inspire belief at all. Try waving a gun around and threatening to kill someone, or kill eir family, if ey doesn't spontaneously believe that the Moon is made of green cheese. The person might claim to believe the Moon is made of green cheese, but they wouldn't actually believe.

otherwise people who believed in a God that values faith would be constantly surprised by the world we live in

That's not how human belief works. Humans do not automatically notice their confusion in all instances, nor do they automatically form internally consistent belief structures without putting any work into it. 'Person X believes strongly in Y' does not in general suggest 'the evidence available to person X is overwhelmingly predicted by the truth of Y'.

Note that if someone who believes in a God that values faith, claims to have found strong objective evidence for the existence of their God, they're coincidentally also claiming to have found strong evidence that said god doesn't value faith

Doesn't this apply to you? You just claimed to have strong evidence that your God's existence predicts the empirical world approximately as well as your God's nonexistence does. But a God who actually valued faith would make sure that his hypothesis seemed to do a worse job of predicting the world, so that more faith was required to accept him. Every time you refute one of my arguments, your faith weakens.

Doesn't this strike you as a slightly odd obsession for a creator-of-everything? Self-deception and bad arguments, as ends in themselves? It at least strikes me as odd.

Comment author: christopherj 19 October 2013 06:44:19PM 1 point [-]

Doesn't this apply to you? You just claimed to have strong evidence that your God's existence predicts the empirical world approximately as well as your God's nonexistence does.

Yes, but I didn't claim it as evidence of anything other than that such belief isn't particularly maladaptive (consequenceless beliefs won't interfere with your predictive power). If it helps you understand my point, P(the world we live in | invisible unicorns in a far away galaxy fart rainbows) is also very high, which is not at all an argument in favor of invisible unicorns (only that there's no evidence against unicorns nor will believing in unicorns therefore result in poor predictions).

To the extent that belief in a God that values faith is consequenceless, it can't be maladaptive. To the extent that such belief has consequences, it will make predictions about the real world. Making predictions about the real world means they can test the belief, which regardless of the result will be suspect (one way, it is evidence against their belief in God, the other way evidence against their belief that God values faith).

My point is that most Christians don't think through the consequences of claiming that something is evidence for God. Why have faith if they're arguing that their belief is a scientifically verifiable one -- in such a case, faith would be more likely to mislead them than science.

Long story short, if you find yourself arguing with a Christian about whether something is or isn't evidence for God, you could have cut the whole thing short by asking them "Will your suggestion let me test for the existence of God?" Not only is there only one reasonable way they can answer, there's specifically a command against testing God.