"The very idea of God is a product of the human imagination. It is quite understandable why primitive people, who were much so much more exposed to the overpowering forces of nature than we are today, should have personified these forces in fear and trembling. But nowadays, when we understand so many natural processes, we have no need for such solutions. I can't for the life of me see how the postulate of an Almighty God helps us in any way." - Paul Dirac, as related by Werner Heisenberg
"If a belief is just formed in opposition to other beliefs, it can't be fundamental? It can't be that deep." - Unnamed, The Witness
That's the tape recorder with the Hardest Puzzle in a surprisingly good game. No, not The Challenge, though it is not-coincidentally near it. The Witness isn't strictly about any single theme (excluding, perhaps, being pretentious), but struggles with the numinous and the mysterious, along with whether to search for, reject, invent, or depose it, play a big role. And they're far from alone in struggling with the idea of God: dabbling with religious philosophy is a bit of a scientist's curse, along with bizarre linguistic theories.
A lot of the quotes aren't... good, strictly speaking. Einstein and Feynman were giants in their fields and outside of them, but as philosophical arguments go they're very much holding beakers of acid over words and trying to remember the mnemonic. And as unoriginal as Dirac's analysis of religion as the opiate of the masses, or Einstein and Feynman's religion-as-awe might be, at least he wasn't Sagan.
Nor is wrestling with religion limited to the brilliant scientists and religious scholars. Some do it for the puns, some for dramaturgy and poetry, some for hope and despair.
Dirac's quote hits a more interesting sphere as an aside. His point was to say that the purpose of religion was bad. Heisenberg, relating the quote into the post-War period, as he turned increasingly to religion and philosophy, may have had his own reasons to relay his conversations into Physics and Beyond. To get there, however, they both assumed religion had a purpose. That's actually not a given, among skeptics, and even for telos little more complicated than the misfiring of an overoptimized guessing machine.
A rather less-celebrated author considered that particular claim in deeper detail in "Religion's Claim to be Non-Disprovable". It's not clear whether he considered it an original thought, or a summary of others already experts in the field -- he might even have been responding to Heisenberg's Science and Religion, given contemporaneous interest in quantum mechanics. But in it, Yudkowsky describes religion's historical domain circumscribing answers, not merely as fearful response but a reasoned, if wrong, one. Where Dirac considers Zeus a primitive superstition regarding lightning and chance, Non-Disprovable considers the broader realms where religion once held authority:
"Early Egyptologists were genuinely shocked to find no trace whatsoever of Hebrew tribes having ever been in Egypt—they weren’t expecting to find a record of the Ten Plagues, but they expected to find something. As it turned out, they did find something. They found out that, during the supposed time of the Exodus, Egypt ruled much of Canaan. That’s one huge historical error, but if there are no libraries, nobody can call you on it...
Not only did religion used to make claims about factual and scientific matters, religion used to make claims about everything. Religion laid down a code of law—before legislative bodies; religion laid down history—before historians and archaeologists; religion laid down the sexual morals—before Women’s Lib; religion described the forms of government—before constitutions; and religion answered scientific questions from biological taxonomy to the formation of stars. The modern concept of religion as purely ethical derives from every other area’s having been taken over by better institutions. Ethics is what’s left.
Or rather, people think ethics is what’s left. Take a culture dump from 2,500 years ago. Over time, humanity will progress immensely, and pieces of the ancient culture dump will become ever more glaringly obsolete. Ethics has not been immune to human progress—for example, we now frown upon such Bible-approved practices as keeping slaves. Why do people think that ethics is still fair game?"
Note that this isn't just the is-ought problem, or whether morals real. Yudkowsky isn't asking for a mathematical proof or philosophical perfection. Were he responding directly to Dirac, he'd be claiming that early scientific models of lightning, for all their many imperfections, provided predictions at least good enough to displace the model of a bearded man in a toga and a kink for swans. The focus here is instead broader: not merely that the religious models are bad, but that are bad enough that their strict read mixes "thou shalt not kill" and "yay, slaughter children". That they're not merely wrong, but obviously wrong, such that naive readers can and do come away displeased or even horrified.
And while that shorthand is not, bluntly, exactly treating the subject terribly fairly charitably to Christians, it's not as though doing so makes the problem go away. There are deep thinkers among the Christian set who can explain or at least answer the questions arising from death of Ancient Egyptian children or endorsements of slavery, but as a whole Christians internally haven't found it easy to handle paradoxes like "turning the other cheek" and "selling a cloak to buy a sword". And that's not some Christian-specific problem: Judaism's internal struggles with consistency and observed reality are legendary in the literal sense, such that theodicy (and, uh, nitpicking rules to the extreme) is a regular past-time.
((Some other religious practices are different enough in scale and/or goal to be beyond the scope of this argument, or at least my knowledge of it.))
That post itself tends to get overlooked over a bit in the decade since as a bit of enthusiastic triumphalism from the pre-Atheism+ days (though it did survive into the eBook), not helped by Yudkowsky's dive into increasingly complex (and self-referential) theories of meta-ethics or quantum mechanics or the broader collapse of New Atheism with the death or dishonor of atheism's Four Horsemen. There's a remaining thread in some corners descended from it, especially amount the tumblr ratsphere, where Chaitian philosophical and ethical questioning became so common that it became a meme, but it's a narrow thread.
Still, the broader goal of Solving Ethics Slightly Less Awfully did have long-term ramifications in the broader Rationality project. It's not the only reason that so many became cheerleaders for Effective Altruism, or join animal or human rights areas, but it's part of the process that ran that direction. The rationalist sphere tried to take it more seriously, leading people to wade through an increasing sea of bizarre acronyms contrasting minor differences or abstract scenarios. In the broader world, skeptics, universalists, and 'spiritual-but-not-religious' people often, in that time, joked about how fading observance of organized religion could lead to a deeper and more unified understanding of each other, which is perhaps just following the 'radical internet openness will break down exclusion and hate' being the sort of predictions you'd give Croesus.
But Claim to Be Non-Disprovable was written in August of 2007, and it's not exactly a solo act or some bombshell revelation. I might not expect a post-religious equal to Aquinas to come about in a decade, but this isn't looking for the Ultimate Solution To Life, The Universe, and Everything, either; merely a more coherent story than a kit-bashed version that got past King James. Sure, there are modern-day religious morality tales, and no small part of the Philosophy section of your local library or bookstore are devoted to it. Yet even those aiming for a universal message, rather than focusing on a specific narrow problem, seldom bother with much more coherence if they bother at all.
One easy answer, and unfortunately the one with no small amount of supporting evidence, holds that religion's claim to morality isn't being measured by its coherency, not in the sense of non-overlapping magisteria, but in the sense that people don't want or need logically compelling systems of morality. People no more expect today's social policies to be disprovable than they do the Bible's sometime-confusing relationship with denim; trying to measure on this scale is just missing the point. And even some of Yudkowsky's examples of 'obvious' historical error don't seem quite as clear if given a skeptical eye: slaughtering first-born children to free slaves actually has some defenders, given how bad slavery can get!
Yet that seems, at the same time, to be too cynical. Where Yudkowsky claims that "they think they can get away with endorsing Bible ethics; and so it only requires a manageable effort of self-deception for them to overlook the Bible’s moral problems", it turned out that the next ten years had those advocates of Biblical ethics struggling deeply with the conflicts between their beliefs and the experiences of people around them, and often found self-deception hard. Gay marriage (and the decriminalization of same-sex behavior before that) is the best-studied example, where once people found out that quite a lot of their friends and neighbors and sometimes family were different, they changed their minds. There's an argument that at least some people 'really' changed their positions based on social pressure, but in many cases the entire social spheres for these people are similarly Christian. But while that's the fastest social change in recent memory, it's far from the only one: divorce, proper dress, the role of women in business or the family, and many attitudes for childrearing have changed, often dramatically, and sometimes without such clear social pressure.
On the other side of the aisle, groups that attempted to build morality from the ground up haven't had much success producing coherent systems of ethics. Most successful ones simply haven't tried, and the group closest to LessWrong trying to figure it out, Effective Altruism, has melted down into what, at the most charitable, could be described as at least incoherent to outsiders (and some overt fraud). Even many limited-scope ethical codes, such as those for academics or engineers, often do not. Even identifying what counts as genuine fundamental belief or evolved conclusion can rapidly turn into debates over the number of angels dancing on the head of a pin.
There's something that matters, but it's not so simple as uncovering ancient historical artifacts or measuring physical phenomena or showing a logical paradox. The mechanics and philosophy can be interesting, but ultimately the interactions with reality matter far more. Which sounds nice, until you see how many differences in reality people run into.