(1) I think the lack of engagement with most religious scholarship on religion reflects the New Atheists' impression that religion is unjustifiedly being treated with kid gloves. They don't think it deserves it, and don't think there's any point bothering with subtle arguments when basic, obvious ones are way more than enough to make the point.

Regarding the lack of engagement with people like Atran and Haidt, well, I fault them for that. (Harris especially, since he often is clearly aware of relevant objections and overrules them by restating simplistic viewpoints with more energy.)

(2) I think you missed the context. I didn't say there weren't feelings of transcendence. I said that the existence of such feelings would not be cause to call the New Atheists wrong, because even if true, "this just begs the question: why have that rotten literal text to begin with?"

What I haven't seen is research that compellingly demonstrates that your religious texts have to badly botch all kinds of factual statements in order for you to derive a transcendent feeling as a follower.

And, meanwhile, while some people are benefiting from the transcendent feelings they do get, others are spreading misery because of an overly literal interpretation of the gobbledygook (the thought would go).

This seems like a pretty clear argument against religions. Just get your transcendence some other way! Go look at the stars on a dark night in a location free from light pollution, take some mild psychoactive drugs, meditate, gather together and sing and drum--whatever, just don't believe primitive nonsense and use it to justify banning girls from schools.

Because you didn't address this at all, but rather gave some defenses of transcendent feelings, I think you missed my point.

Again, the only reason I've seen that the New Atheist approach might be wrong is that it actually doesn't much matter what the content of religions is because of how we make decisions.

If religions say to be cruel to unbelievers, but people form non-religious groups with similar frequency and are cruel to the outgroup in similar ways and at similar frequencies, then the argument that religion is a source of harm is much reduced.

That is, if it is true that people might exploit religion to excuse their actions, but similar numbers of people would come up with other excuses for other similarly bad actions anyway, then religion itself is not fundamentally the problem. The problem is human nature. The counterfactual religious beliefs are then not terribly serious, because in any context where it matters, people can be trained to grab on to the actual situation, and they'll be able to rationalize away why it's not a conflict with their religion.

My current impression is that this is likely to be true enough to cast doubt on the New Atheist perspective, especially since in pratice the New Atheists are strongly tribal. If the real problem is division into outgroups causing stronger ingroup/outgroup boundaries, strident anti-religion language runs the risk of causing exactly the same problem that religions cause.

It's not that religions have equal truth to atheism--factually, they don't. Rather, it's that the pernicious aspects come from division regardless of how it is sewn, so if you are more right but equally divisive, your correctness is largely wasted at least as far as inter-human conflict is concerned. (Also, there are plenty of non-religious ways to be dangerously wrong as the far left and right constantly demonstrate to us.)