I found this site through the posts on decoherence and many-worlds; I haven't yet read them all, and look forward to doing so. Also enjoyed the posts on Bayesian rationality.

But I was disappointed by this one. The main reason is that it implicitly reduces all religious phenomena to matters of belief, which I think is a mistake.

To be clear about where I'm coming from: I don't hold any religious beliefs. Nevertheless, I think that much of what goes on in religion is psychologically or sociologically beneficial. And I think that religious language is often misconstrued (by religious and nonreligious people alike) as expressing beliefs, when it actually (or also) functions in other ways. (It expresses certain kinds of attitudes and perspectives.)

Eliezer's main point is to deny that religion can't be disproven. In order to do this, he paints a picture of religion as essentially a set of beliefs. Addressing people like me who want to save some non-epistemic subset of religion, he says "The orthogonality of religion and factual questions is a recent and strictly Western concept." I want to make two points about this.

The first is that even if it's true, it says nothing about the value of modern people pursuing such non-fact-based activities. Explorations of attitudes and global perspectives can be pursued via religious language in much the same way as it is pursued in non-religious art, literature, poetry, etc. Eliezer takes "ethics" as the core of the non-fact-based questions. His argument against religious ethics is that the Bible contains elements that conflict with contemporary ethics, which has "progressed" since the Bible was written. The argument simply ignores the fact that religious ethics also progresses. In other words, Eliezer implicitly focuses on fundamentalist religion; but many modern religious people explicitly treat the Bible as a literary background to rational reflections taking contemporary attitudes and insights into account. Eliezer seems strangely unaware that many modern religious people have fought against slavery, for women's rights, gay rights, etc.

In fact, he seems unaware that such rational revisions of traditional attitudes have been going on for thousands of years--and this leads into my second point: the separation between myth and morality is not something new. The prophets Hosea and Amos explicitly reject mythology when it overshadows morality; they make fun of people who think that animal sacrifices can atone for bad deeds, or that religion essentially depends on anything beyond morality. The book of Deuteronomy contains many revisions of earlier material in Exodus, turning laws from a mythic to an ethical rationale. And the Talmud contains countless examples wherein Biblical morality is reversed, explicitly or implicitly.

So I think Eliezer is doubly wrong about the orthogonality of religion and factual questions.

I clicked on a link in this post to "believing in". I expected to find an acknowledgment of the purely non-epistemic sense that this phrase often carries ("I believe in the right to organize","I believe in myself", "I believe in America"). Instead I found arguments against people who hold factual beliefs without or despite evidence. But I would hypothesize that many people who affirm beliefs without evidence are actually just affirming an attitude which they are used to expressing in the misleading belief-language. For many, "belief in God" expresses solidarity with a particular community and a set of attitudes toward the world and other people.

The main fault I find with Eliezer's analysis is that it appears blind to the literary character of many Biblical texts. He says "The vast majority of religions in human history…tell stories of events that would constitute completely unmistakable evidence if they'd actually happened." But this is also true of literature in general. It in no way implies that authors or readers of the stories take them literally. Anybody who thinks that Ezekiel literally expected bones to rise from graves, or that the author and audience of the story of Balaam's ass took it differently from, say, Aesop's stories, is operating with a deficient view of how stories work.

Final point: Eliezer sets himself against those who posit "wonder" as a basis and/or effect of religious language. He finds no (or very little) wonder in ancient texts. But from a philosophical point of view, I would nominate a different emotion as the essential religious category: gratitude. The most positive attitude toward the world or one's life must contain gratitude, even when what one is grateful for is something as vague as life itself, and even if one posits no metaphysical entity toward which one is grateful. And ancient religion certainly expresses such global gratitude.