But maybe “communal irrational behavior” is the heart of the matter and “a supernatural agent” is just a side show?
His point was that Communism, for example, has many of the traits of a religion, but without a supreme being , and his suggestion was that a non-theistic definition of “religion” which included Communism would be more useful than the theistic one normally used.
Like Razib, but probably more so, I am a “Chamberlain secularist” who does not expect religion ever to disappear. I think that the term
A lot of what is called “communal irrational behavior” I would instead call “social highstakes gambling”. If you cherrypick the disasters (the Branch Davidians, Jonestown, etc.) you have an open-and-shut case against belief. However, if you look at some of the successful social gambles in history (the 1688 Glorious Revolution in England, the adventurism or “total committment rationality” of classical Athens, the Polynesian colonization of the South Pacific, the Church of Jesus Christ of Latter-Day Saints, Christianity) you’ll find that a lot of those people were pretty fucking nuts. The colonization of the South Pacific is my model: fair-sized groups of people gathered all of their belongings and set off on the open ocean toward a destination which they had no reason to believe even existed. Most of them were never heard from again, but the lucky ones colonized Hawaii and New Zealand.
The rationalist assumption is that reality is known, and that progress is grounded on reason. But at any given point, key aspects of reality are unknown, and all enterprises are gambles. The winner of a high-stakes gamble profits enormously, but the losers (the majority, probably the vast majority) are destroyed. Don’t these gambles sound like mutations? Most mutations are harmful, some are neutral, and a very small number are beneficial. I am suggesting a social-history version of Donald Campbell’s “evolutionary epistemology”: blind social variation and selective retention (or, in Gould’s words, proliferation and decimation.)
Social gambling tends to be even crazier than individual gambling, because the followers tend to believe the prophets without really understanding them, so that the leaders’ errors can often be often magnified. But sometimes long-shot gambles work. And (as can be seen with the Mormons and the Hasids, for example) the craziest fanatics are often meticulously rational in significant areas of their behavior, and valuable technical innovations which are part of their grand scheme. The very craziness of a religion increases the selection pressures, thus forcing cult members to improvise survival strategies which prudent moderates would not need. (From this point of view, the down side of religion would just be its cost. There’s no free lunch, but on the net, a successful religion is beneficial).
Religions are a leading component of the cultural part of gene-culture coevolution. Crazy religions are mutants, and most mutants die — often because they kill their followers. But we cannot assume that the religions which have survived were rational at the time of their foundation. New religions never are; at the beginning they are always enormous blind gambles.