The body count argument is an attempt to address a couple of points—that religion is necessary to make people moral and that religion, or alternatively lack of religion, makes people better than they would be otherwise.
I’d say the evidence is pretty strong that religion is not necessary to make people moral, though the argument that the good residual effects of religion hang on for a couple of generations but fade is harder to disprove. On the other hand, I think it’s close enough to non-falsifiable that I don’t worry about it.
I’ve never seen it argued that religion increases the moral range among people, leading to both more generous behavior (some of it useful) and more destructive behavior, but I don’t find this implausible. I can’t say whether increasing the moral range can be said a priori to be a good or bad thing. Like nationalism, religion is a force concentrator.
On the other hand, I think it’s close enough to non-falsifiable that I don’t worry about it.
A hypothesis which is non-falsifiable has an alternative which is also non-falsifiable. You don’t get to “not worry about it” unless you can show that the prior is low, which is a separate matter from any difficulties of setting up an experiment that everyone will agree is definitive.
The prior has several parts. Would the null hypothesis be that people are moral because there are practical and emotional advantages? That the good effects of religion don’t fade?
And just to go sideways a little, should “religion” be viewed as a single thing for this discussion, or do some religions or some approaches to religion have different moral effects?
Agreed, falsifiability isn’t the issue. But there is a problem with the question being ill-defined. Questions such as “Does atheism make society less moral?” have to be unpacked using counterfactuals, which in turn can be unpacked only if you have some precise conception of the alternative scenarios within some robust theory. These are missing in all discussions of these questions that I’ve seen.
The body count argument is an attempt to address a couple of points—that religion is necessary to make people moral and that religion, or alternatively lack of religion, makes people better than they would be otherwise.
I’d say the evidence is pretty strong that religion is not necessary to make people moral, though the argument that the good residual effects of religion hang on for a couple of generations but fade is harder to disprove. On the other hand, I think it’s close enough to non-falsifiable that I don’t worry about it.
I’ve never seen it argued that religion increases the moral range among people, leading to both more generous behavior (some of it useful) and more destructive behavior, but I don’t find this implausible. I can’t say whether increasing the moral range can be said a priori to be a good or bad thing. Like nationalism, religion is a force concentrator.
A hypothesis which is non-falsifiable has an alternative which is also non-falsifiable. You don’t get to “not worry about it” unless you can show that the prior is low, which is a separate matter from any difficulties of setting up an experiment that everyone will agree is definitive.
The prior has several parts. Would the null hypothesis be that people are moral because there are practical and emotional advantages? That the good effects of religion don’t fade?
And just to go sideways a little, should “religion” be viewed as a single thing for this discussion, or do some religions or some approaches to religion have different moral effects?
Agreed, falsifiability isn’t the issue. But there is a problem with the question being ill-defined. Questions such as “Does atheism make society less moral?” have to be unpacked using counterfactuals, which in turn can be unpacked only if you have some precise conception of the alternative scenarios within some robust theory. These are missing in all discussions of these questions that I’ve seen.