Consider the following proposition: For each existing religion, one can easily set out evidence of its wrongness that would (1) be very convincing to the majority of people who are not already positively disposed towards that religion and (2) be good reasoning in the abstract; if we combine these, we get a strong argument that no existing religion is close to the truth, but this argument will not convince most people because most people are adherents of some religion or other, and it is extremely difficult for adherents of any religion to appreciate the strength of arguments against that religion.
It seems to me that that proposition may very well be true, and that if it is true then it’s correct (aside from the fact that “proofs” is too ambitious a word) to say both that it’s difficult to convince people that all existing religions are wrong, and that the proofs of that fact are not hard to verify.
For each existing religion, one can easily set out evidence of its wrongness that would (1) be very convincing to the majority of people who are not already positively disposed towards that religion and (2) be good reasoning in the abstract;
The same is true for atheism, and certainly for utilitarianism.
(I think Eugene is downvoting your comments on this thread.)
I’m not sure this is true, for a number of reasons.
By “one can easily set out evidence of its wrongness” I presume that you meant one particular set of evidence. In that case, I am not sure that you can in fact choose one particular set of evidence which would be very convincing to the majority of people. It’s true that if you take all the people who disagree with a religion, then take any individual among them, you can easily find something which is convincing to him. But that may not be the same thing which is convincing to someone else. And since people who are religious typically think there is some evidence for their religion, they will disagree with some arguments against other religions because they will see that those arguments could also refute their own. They will see them as “proving too much.” So it is not clear at all that you can take one particular set of evidence and convince the majority.
On the other hand, if you meant you can find a set of evidence that matches each individual, your argument is in fact too wide, because it would apply to nearly all philosophical views, including contrary ones, as time mentioned in his comment in regard to atheism and utilitarianism.
If you are avoiding that consequence by means of the “good reasoning in the abstract”, I have various issues with that. First, it is possible to argue for something with “good reasoning in the abstract” in a way which would be convincing to many or most people, but which would become unconvincing if they were aware of other evidence and arguments for the opposite position. If good reasoning is supposed to mean that you mention all of the evidence, then your argument is question begging—you are saying nothing more than that you have considered all of the relevant evidence that you can find, and you consider the most reasonable position to be that all religions are false. I agree both with this assessment of the evidence and the conclusion. But I am not going to assert that I am the only one who engages in good reasoning, and I realize that there are people who have considered basically all of the evidence that I have considered, and who are basically reasonable people, and nonetheless think that some religion is true.
Finally, I am not sure that it is true to say “if we combine these, we get a strong argument that no existing religion is close to the truth,” even given the other things, because we can make arguments that are convincing to most people that any particular person is not going to win the lottery, but this is not a strong argument that no one is going to win.
(I think Eugene is downvoting your comments on this thread.)
Eugine is downvoting my comments everywhere. My 30-day karma is currently at −44; if I am interpreting the percentages right then my “non-Eugine” 30-day karma is probably somewhere around +270. (But I suspect that may be inflated a bit, because some people may be applying “corrective” upvotes to some of my comments that have been Eugined.)
I am not sure that you can in fact choose one particular set of evidence which would be very convincing to the majority of people.
That’s a good observation. My feeling is that the “argument from evil” against religions that claim there’s a supremely good supremely powerful being is extremely convincing to almost everyone who doesn’t belong to such a religion—but of course lots of people who don’t belong to one such religion may belong to a different such religion and therefore be unfavourably disposed towards the AfE.
If good reasoning is supposed to mean that you mention all the evidence, then your argument is question begging
It is supposed to mean something along those lines, but if you think I’m begging the question then I think you are probably misunderstanding the nature of my argument. Let me recap, and try to make the connective tissue of the argument more visible; my apologies if this makes it laborious.
DanArmak said (roughly) that atheism is not only correct but in some sense obviously correct; in other words, that the answer to “why are so many people religious?” is less “because anti-religious arguments are subtle and hard to understand” and more “because religious people are unable to accept even easy clear arguments if they lead to irreligious conclusions”.
Various people disagreed.
I am agreeing (roughly and provisionally) with Dan, and suggesting a way to make his claim more precise: the idea is that (1) there are anti-religious arguments convincing to any given person (you just have to avoid arguments that they can see refuting their own religion) and (2) that isn’t because you can find crappy arguments tailored to any given person’s biases and errors—the arguments in question are actually good ones in some objective sense.
The people disagreeing with Dan mostly agree with him that in fact religions are probably wrong, and I think they agree with him that there are “good in the abstract” arguments leading to that conclusion. The point of their disagreement is that those arguments are unconvincing to many (otherwise) intelligent and reasonable people, and that if you want to claim an argument is easy to verify then the fact that it fails to convince many intelligent and reasonable people is a big obstacle.
The point of my comment, therefore, was to suggest a way to surmount that obstacle, by providing evidence that each person’s unconvincedness is an artefact of their own religious commitments rather than an indication that the arguments are weak or difficult.
And the reason why I stipulated that the arguments in question are “good reasoning in the abstract” is just to exclude the possibility that the individually-tailored arguments are cheaty ones that target individuals’ cognitive weaknesses, rather than anti-cheaty ones that avoid individuals’ cognitive weaknesses.
lottery
That’s a valid point in the abstract, but its validity depends on the actual numbers in question. If you have a million people and a million arguments each showing Pr(a given person wins) ⇐ 10^-6 then indeed that isn’t enough to be any sort of evidence that no one will win; but if you could show that each person’s winning probability was ⇐ 10^-9, it would be. In the case we’re dealing with here, I suggest that we can probably find (let’s say) no more than 100 varieties of theism (I know that some earlier comments said “religion” but I think “theism” is what’s actually under discussion here), and an argument for each showing Pr(that version of theism) ⇐ 10^-4, in which case it follows that Pr(any of them) ⇐ 1%.
I’m still not sure that your idea is true even in when it is limited to “everyone who doesn’t belong to such a religion.” It might be true about LWers but they are not a representative sample. Americans seem to be pretty consistently more likely to identify as agnostic as atheist, for example. Now of course that might be because American doesn’t like atheists and they want to avoid the social consequences. And it might be different (as far as I know) in other countries, since I just checked the statistics for the US according to various polls. But prima facie, it suggests that “most people who don’t believe in any god at all are not extremely convinced by any argument against the existence of a god.” I am not asserting that this is definitely the case, but it is plausible to me, and supported at least by this fact. It’s possible you could establish your claim with better data, but so far I’m not convinced.
There is still the problem that if you limit it to people who “don’t belong to such a religion,” then you appear to be saying “everyone who thinks that God doesn’t exist thinks that there is a convincing argument against the existence of God,” and even if that were one particular argument, it would be similar to saying, “everyone who accepts the error theory of morality finds such and such an argument convincing.” Even if there is one such argument in the case of error theory (which might not actually be the case), that hardly establishes that it is easy to prove that error theory is true, or that it is true at all, for that matter.
I personally think the argument from evil (and various similar arguments) is evidence against the existence of a personal God, but I don’t find it extremely convincing. A large part of the reason for that is when I did believe in Christianity, I had an answer to that argument which I found reasonable, and which still seems reasonable to me, not in the sense of “this is the case”, but in the sense of a reasonable possibility. Now you could say that this is an “artifact of religious commitments” in a historical sense, but I don’t think this is actually the case, given the fact that there are plenty of other issues where I never thought I had a good response in the first place. There is also Scott Alexander’s “Answer to Job.” Now I think that if you consider such responses carefully (my previous personal answer which I haven’t gone into here, or Scott’s answer, or almost any other reasonable response) they actually fit better with an impersonal principle than with a personal God, but they are not inconsistent with a personal one. So that prevents me from considering the argument extremely persuasive.
“Evidence that each person’s unconvincedness is an artefact of their own religious commitments.” You can certainly provide such evidence for many or most people. But theists can also provide evidence that atheistic convictions are often the result of non-evidential commitments as well. And in any case this is not going to apply to every single individual, and especially the case of converts to religion like Leah Libresco or Edward Feser. That doesn’t mean that all sorts of non-rational influences aren’t present in the case of converts. But it isn’t true that e.g. “they aren’t convinced by the argument from evil because they accept religion,” given the fact that the argument from evil did not prevent them from going from not accepting religion to accepting religion in the first place.
Of course, if I disagree with someone’s conclusion, I am almost certainly going to end up explaining how he went wrong. This isn’t Bulverism in a bad sense, given that I don’t assume he is wrong in the first place, but argue for it. And if I am right, it does mean there are better and more reasonable arguments for my position than for their position. But that isn’t the same as calling my conclusion something easily seen to be true.
I also would not say that just because some people disagree who are generally reasonable people, it follows that the matter cannot be an easy one. So for example I would say that young earth creationism is easily seen to be false, even though there are some people who both think it is true, and are generally reasonable people. But note that there are relatively few people like this, compared to the number of generally reasonable people who admit that evolution is true but still think some kind of religion is true.
One difference is this: it does not require anything like philosophical argument to establish that evolution is true. It just requires looking at actually existing plants, animals, and rocks and discussing how they could have got that way. In contrast, philosophy is actually hard for human beings. And I think you cannot prove or disprove something like theism without philosophical argument. This suggests that it is not easy to do so.
It’s true that a lot of people call themselves agnostics, which seems to indicate (1) not being completely convinced by any argument against theism while also (2) not having a commitment to any particular religion. However, I think the great majority of people who call themselves agnostics fall into one of these categories:
People who prefer to avoid too-committal terms like “atheist”, either because there’s a stigma attached to overt atheism where they are or because they think “atheist” implies absolute certainty.
People who haven’t really thought the matter through very much.
People who are agnostic about the existence of some sort of deity but strongly convinced that e.g. there is almost certainly no supremely good and powerful being who takes a personal interest in human affairs.
I would expect people in the first and third groups to share my opinion about arguments from evil, though the third lot would rightly observe that e.g. such arguments tell us nothing about superbeings who just don’t care about our affairs.
People in the second group might well not be very convinced by arguments from evil, but I would expect that if they gave serious consideration to such arguments they would typically see them as very strong.
“Everyone thinks there is a convincing argument”
That’s not quite what I’m saying. I’m saying that there are, in fact, arguments that I would expect to be very convincing if looked at seriously by a sizable majority of people not committed to the religions in question. Of course those who haven’t seriously considered such arguments will not yet be convinced.
Your own epistemic situation
I find it very interesting that you aren’t very convinced by arguments from evil despite having rejected Christianity, but I don’t think there’s anything further I can say without having any idea why it is that you aren’t convinced. You say it’s because there’s a particular answer you find reasonable, but I’ve no idea what that answer is :-).
You do mention Scott’s “Answer to Job”, which is very ingenious. It’s a good answer to arguments-from-evil that end “and therefore it is absolutely impossible that there is a benevolent god”, but I would consider such arguments overreaching even without that particular answer. Is it any good as an answer to “evidential” arguments that take the quantity and distribution of Bad Stuff in our world merely as evidence? Well, I guess that depends on (1) how likely Scott’s scenario is a priori and (2) how credible it is that our world is, so to speak, a random pick from all possible worlds where good outweighs bad, weighted by number of intelligent agents or something of the kind. To #1, I say: not terribly likely, because I am not convinced that good outweighs bad in our world. (Which is not at all the same thing as saying that many people in our world would rather die than live.) To #2, I say: not credible at all; I would expect a random observer from a multiverse containing all possible more-good-than-bad worlds to see something very very very different from what I see. (I suspect the great majority of observers in such worlds are something like Boltzmann brains.)
“Atheistic convictions are often the result of non-evidential commitments”
Oh, absolutely. I wasn’t meaning to make any comment on the relative merits of theists and atheists. I was just addressing an obvious argument against any claim that there’s good evidence against theism, namely that lots of apparently sensible people are theists. (I firmly agree that you likewise can’t reject the claim that there’s good evidence against atheism just because lots of apparently sensible people are atheists.)
(Highly intelligent converts like Leah Libresco and Ed Feser would be evidence against any claim that the only reason for not being convinced by atheistic arguments is prior religious commitment. But I have not been making that claim, at least not deliberately.)
Consider the following proposition: For each existing religion, one can easily set out evidence of its wrongness that would (1) be very convincing to the majority of people who are not already positively disposed towards that religion and (2) be good reasoning in the abstract; if we combine these, we get a strong argument that no existing religion is close to the truth, but this argument will not convince most people because most people are adherents of some religion or other, and it is extremely difficult for adherents of any religion to appreciate the strength of arguments against that religion.
It seems to me that that proposition may very well be true, and that if it is true then it’s correct (aside from the fact that “proofs” is too ambitious a word) to say both that it’s difficult to convince people that all existing religions are wrong, and that the proofs of that fact are not hard to verify.
The same is true for atheism, and certainly for utilitarianism.
Original thread here.
(I think Eugene is downvoting your comments on this thread.)
I’m not sure this is true, for a number of reasons.
By “one can easily set out evidence of its wrongness” I presume that you meant one particular set of evidence. In that case, I am not sure that you can in fact choose one particular set of evidence which would be very convincing to the majority of people. It’s true that if you take all the people who disagree with a religion, then take any individual among them, you can easily find something which is convincing to him. But that may not be the same thing which is convincing to someone else. And since people who are religious typically think there is some evidence for their religion, they will disagree with some arguments against other religions because they will see that those arguments could also refute their own. They will see them as “proving too much.” So it is not clear at all that you can take one particular set of evidence and convince the majority.
On the other hand, if you meant you can find a set of evidence that matches each individual, your argument is in fact too wide, because it would apply to nearly all philosophical views, including contrary ones, as time mentioned in his comment in regard to atheism and utilitarianism.
If you are avoiding that consequence by means of the “good reasoning in the abstract”, I have various issues with that. First, it is possible to argue for something with “good reasoning in the abstract” in a way which would be convincing to many or most people, but which would become unconvincing if they were aware of other evidence and arguments for the opposite position. If good reasoning is supposed to mean that you mention all of the evidence, then your argument is question begging—you are saying nothing more than that you have considered all of the relevant evidence that you can find, and you consider the most reasonable position to be that all religions are false. I agree both with this assessment of the evidence and the conclusion. But I am not going to assert that I am the only one who engages in good reasoning, and I realize that there are people who have considered basically all of the evidence that I have considered, and who are basically reasonable people, and nonetheless think that some religion is true.
Finally, I am not sure that it is true to say “if we combine these, we get a strong argument that no existing religion is close to the truth,” even given the other things, because we can make arguments that are convincing to most people that any particular person is not going to win the lottery, but this is not a strong argument that no one is going to win.
Eugine is downvoting my comments everywhere. My 30-day karma is currently at −44; if I am interpreting the percentages right then my “non-Eugine” 30-day karma is probably somewhere around +270. (But I suspect that may be inflated a bit, because some people may be applying “corrective” upvotes to some of my comments that have been Eugined.)
That’s a good observation. My feeling is that the “argument from evil” against religions that claim there’s a supremely good supremely powerful being is extremely convincing to almost everyone who doesn’t belong to such a religion—but of course lots of people who don’t belong to one such religion may belong to a different such religion and therefore be unfavourably disposed towards the AfE.
It is supposed to mean something along those lines, but if you think I’m begging the question then I think you are probably misunderstanding the nature of my argument. Let me recap, and try to make the connective tissue of the argument more visible; my apologies if this makes it laborious.
DanArmak said (roughly) that atheism is not only correct but in some sense obviously correct; in other words, that the answer to “why are so many people religious?” is less “because anti-religious arguments are subtle and hard to understand” and more “because religious people are unable to accept even easy clear arguments if they lead to irreligious conclusions”.
Various people disagreed.
I am agreeing (roughly and provisionally) with Dan, and suggesting a way to make his claim more precise: the idea is that (1) there are anti-religious arguments convincing to any given person (you just have to avoid arguments that they can see refuting their own religion) and (2) that isn’t because you can find crappy arguments tailored to any given person’s biases and errors—the arguments in question are actually good ones in some objective sense.
The people disagreeing with Dan mostly agree with him that in fact religions are probably wrong, and I think they agree with him that there are “good in the abstract” arguments leading to that conclusion. The point of their disagreement is that those arguments are unconvincing to many (otherwise) intelligent and reasonable people, and that if you want to claim an argument is easy to verify then the fact that it fails to convince many intelligent and reasonable people is a big obstacle.
The point of my comment, therefore, was to suggest a way to surmount that obstacle, by providing evidence that each person’s unconvincedness is an artefact of their own religious commitments rather than an indication that the arguments are weak or difficult.
And the reason why I stipulated that the arguments in question are “good reasoning in the abstract” is just to exclude the possibility that the individually-tailored arguments are cheaty ones that target individuals’ cognitive weaknesses, rather than anti-cheaty ones that avoid individuals’ cognitive weaknesses.
That’s a valid point in the abstract, but its validity depends on the actual numbers in question. If you have a million people and a million arguments each showing Pr(a given person wins) ⇐ 10^-6 then indeed that isn’t enough to be any sort of evidence that no one will win; but if you could show that each person’s winning probability was ⇐ 10^-9, it would be. In the case we’re dealing with here, I suggest that we can probably find (let’s say) no more than 100 varieties of theism (I know that some earlier comments said “religion” but I think “theism” is what’s actually under discussion here), and an argument for each showing Pr(that version of theism) ⇐ 10^-4, in which case it follows that Pr(any of them) ⇐ 1%.
Regarding the argument from evil:
I’m still not sure that your idea is true even in when it is limited to “everyone who doesn’t belong to such a religion.” It might be true about LWers but they are not a representative sample. Americans seem to be pretty consistently more likely to identify as agnostic as atheist, for example. Now of course that might be because American doesn’t like atheists and they want to avoid the social consequences. And it might be different (as far as I know) in other countries, since I just checked the statistics for the US according to various polls. But prima facie, it suggests that “most people who don’t believe in any god at all are not extremely convinced by any argument against the existence of a god.” I am not asserting that this is definitely the case, but it is plausible to me, and supported at least by this fact. It’s possible you could establish your claim with better data, but so far I’m not convinced.
There is still the problem that if you limit it to people who “don’t belong to such a religion,” then you appear to be saying “everyone who thinks that God doesn’t exist thinks that there is a convincing argument against the existence of God,” and even if that were one particular argument, it would be similar to saying, “everyone who accepts the error theory of morality finds such and such an argument convincing.” Even if there is one such argument in the case of error theory (which might not actually be the case), that hardly establishes that it is easy to prove that error theory is true, or that it is true at all, for that matter.
I personally think the argument from evil (and various similar arguments) is evidence against the existence of a personal God, but I don’t find it extremely convincing. A large part of the reason for that is when I did believe in Christianity, I had an answer to that argument which I found reasonable, and which still seems reasonable to me, not in the sense of “this is the case”, but in the sense of a reasonable possibility. Now you could say that this is an “artifact of religious commitments” in a historical sense, but I don’t think this is actually the case, given the fact that there are plenty of other issues where I never thought I had a good response in the first place. There is also Scott Alexander’s “Answer to Job.” Now I think that if you consider such responses carefully (my previous personal answer which I haven’t gone into here, or Scott’s answer, or almost any other reasonable response) they actually fit better with an impersonal principle than with a personal God, but they are not inconsistent with a personal one. So that prevents me from considering the argument extremely persuasive.
“Evidence that each person’s unconvincedness is an artefact of their own religious commitments.” You can certainly provide such evidence for many or most people. But theists can also provide evidence that atheistic convictions are often the result of non-evidential commitments as well. And in any case this is not going to apply to every single individual, and especially the case of converts to religion like Leah Libresco or Edward Feser. That doesn’t mean that all sorts of non-rational influences aren’t present in the case of converts. But it isn’t true that e.g. “they aren’t convinced by the argument from evil because they accept religion,” given the fact that the argument from evil did not prevent them from going from not accepting religion to accepting religion in the first place.
Of course, if I disagree with someone’s conclusion, I am almost certainly going to end up explaining how he went wrong. This isn’t Bulverism in a bad sense, given that I don’t assume he is wrong in the first place, but argue for it. And if I am right, it does mean there are better and more reasonable arguments for my position than for their position. But that isn’t the same as calling my conclusion something easily seen to be true.
I also would not say that just because some people disagree who are generally reasonable people, it follows that the matter cannot be an easy one. So for example I would say that young earth creationism is easily seen to be false, even though there are some people who both think it is true, and are generally reasonable people. But note that there are relatively few people like this, compared to the number of generally reasonable people who admit that evolution is true but still think some kind of religion is true.
One difference is this: it does not require anything like philosophical argument to establish that evolution is true. It just requires looking at actually existing plants, animals, and rocks and discussing how they could have got that way. In contrast, philosophy is actually hard for human beings. And I think you cannot prove or disprove something like theism without philosophical argument. This suggests that it is not easy to do so.
Agnostics
It’s true that a lot of people call themselves agnostics, which seems to indicate (1) not being completely convinced by any argument against theism while also (2) not having a commitment to any particular religion. However, I think the great majority of people who call themselves agnostics fall into one of these categories:
People who prefer to avoid too-committal terms like “atheist”, either because there’s a stigma attached to overt atheism where they are or because they think “atheist” implies absolute certainty.
People who haven’t really thought the matter through very much.
People who are agnostic about the existence of some sort of deity but strongly convinced that e.g. there is almost certainly no supremely good and powerful being who takes a personal interest in human affairs.
I would expect people in the first and third groups to share my opinion about arguments from evil, though the third lot would rightly observe that e.g. such arguments tell us nothing about superbeings who just don’t care about our affairs.
People in the second group might well not be very convinced by arguments from evil, but I would expect that if they gave serious consideration to such arguments they would typically see them as very strong.
“Everyone thinks there is a convincing argument”
That’s not quite what I’m saying. I’m saying that there are, in fact, arguments that I would expect to be very convincing if looked at seriously by a sizable majority of people not committed to the religions in question. Of course those who haven’t seriously considered such arguments will not yet be convinced.
Your own epistemic situation
I find it very interesting that you aren’t very convinced by arguments from evil despite having rejected Christianity, but I don’t think there’s anything further I can say without having any idea why it is that you aren’t convinced. You say it’s because there’s a particular answer you find reasonable, but I’ve no idea what that answer is :-).
You do mention Scott’s “Answer to Job”, which is very ingenious. It’s a good answer to arguments-from-evil that end “and therefore it is absolutely impossible that there is a benevolent god”, but I would consider such arguments overreaching even without that particular answer. Is it any good as an answer to “evidential” arguments that take the quantity and distribution of Bad Stuff in our world merely as evidence? Well, I guess that depends on (1) how likely Scott’s scenario is a priori and (2) how credible it is that our world is, so to speak, a random pick from all possible worlds where good outweighs bad, weighted by number of intelligent agents or something of the kind. To #1, I say: not terribly likely, because I am not convinced that good outweighs bad in our world. (Which is not at all the same thing as saying that many people in our world would rather die than live.) To #2, I say: not credible at all; I would expect a random observer from a multiverse containing all possible more-good-than-bad worlds to see something very very very different from what I see. (I suspect the great majority of observers in such worlds are something like Boltzmann brains.)
“Atheistic convictions are often the result of non-evidential commitments”
Oh, absolutely. I wasn’t meaning to make any comment on the relative merits of theists and atheists. I was just addressing an obvious argument against any claim that there’s good evidence against theism, namely that lots of apparently sensible people are theists. (I firmly agree that you likewise can’t reject the claim that there’s good evidence against atheism just because lots of apparently sensible people are atheists.)
(Highly intelligent converts like Leah Libresco and Ed Feser would be evidence against any claim that the only reason for not being convinced by atheistic arguments is prior religious commitment. But I have not been making that claim, at least not deliberately.)