So I'll say first that I've been meaning to write a lengthy response to Kai's earlier comment, but have lacked the steam. Instead I'll sum up what have been my idle thoughts on the matter, and I'll do it briefly. Literature suffers from a problem of a peculiar relationship with agency and representation. An individual can make liberated choices (ideally) that still coincide with stereotypes and have but a marginal effect on the relationships of others to stereotypes. (It is when lots and lots of people make these same decisions that they become normative.) In literature this marginal influence on normativity is magnified by the singularity of the protagonist (particularly in classically heterosexual romances such as His Dark Materials). The problem is then not in accurately presenting or participating in free choice but in depicting a healthy protagonist, who, like it or not, becomes something of a model, which is in turn a norm. Now I think Pullman does as good a job of this as can be expected; my criticisms we nits, not full grown lice. (More honestly, I now think whatever criticism can be leveled would be against nits, but since I haven't re-read my earlier post, I'm not sure if that was the case then or not.) (Thanks, Kai, btw, for chatting about this with me on the phone; it was, as always, very pleasant.)
The more pressing issue on my mind is a wonderful link I found in a Reddit comments thread. I've only had time yet to listen through session 3 (it is a very handing addition to crocheting), which concludes with Carolyn Porco's talk on Saturn. (This includes a picture and description of a shot of Saturn eclipsing the sun--taken from the far side of Saturn--which alone is worth the time.) The conference deals with the question of religion and science, and the majority of the speakers thus far suggest that the former interferes with the latter, and that we'd all be happier (and more alive) in the long run if we could just do away with the former.
Two points are underdeveloped so far, though perhaps the second half will redress this: 1) An understanding of the disciplinary element of religion, and 2) The possibility of a fully non-religious morality.
Now I have a hard time keeping track of what thoughts I've taken the time to elaborate on here, so forgive me once again if I repeat.
Point one is particularly conspicuous in Porco's talk, where she partitions the concern of the conference into God on the one hand and religion on the other; but it is introduced much earlier and involves a common counter-atheist argument. Religion offers community, and it is a sign of my privilege--economic, social, historical, educational, sexual, geographical, political--that I can live happily without a religious community. But in many places, religion binds a community together, giving its constituents communality and place. In short, religion provides social comfort.
Of course, the response to this is clear to anyone with a passing knowledge of the history of the Jewish diaspora, though this is not by a long shot the only handy example. The fact is that while religious communities are capable of tremendous good--the example from the conference is the donations to Katrina victims--they operate on what is sometimes called a politics of the same. This is the politics that enables the Salvation Army to threaten to close its soup kitchens if New York doesn't ban gay marriage, for example. It is a politics of defining community by who is excluded and why (think of the inanity of the Bible study groups, where a sub-group--married women, married men, teens--of the church examine famous passages with the intent of reinforcing their received wisdom).
This is not to say that religion is intrinsically or necessarily disciplinary (in the sense that these groups discipline their members to conform, under the threat of exclusion). Nor is it to say that it is exclusively religious organizations that employ this sort of discipline. The question is whether or not this mode of discipline is in fact fundamental to faith (for example, to be christian you must have a common belief in Christ, but how far does that commonality go?), or, perhaps more broadly, whether a method of disciplinary exclusion can be employed without a tangible connection to ethics (that is, can a church exclude homosexuals without denouncing them, for example). (A lot of work, incidentally, has been done in this vein, but it does not often fall under the banner of science.)
The second point came up most clearly in an exchange between Richard Dawkins, Sam Harris, and Joan Roughgarden (the lone open theist thus far). Roughgarden suggests that a viable source of morality is unlikely, and at the least has yet to be demonstrated, without religion--and if I recall correctly, she means specifically a founding religious text.
I was sure that Harris, who apparently has a degree in philosophy, would come back with the rich philosophical tradition of non-theism, starting with the Categorical Imperative, and bringing it up to the recent (comparatively) formulation of groundless solidarity and the politics of difference (though these last two stray a ways form philosophy proper). Because after all one of the primary preoccupations with intellectuals on the left since WWII has been how to talk about ethics when everything--God, religion, progress, humanism--has died or failed.
And this leads me to conclude with a broader concern regarding these talks. It is clear that they don't all mean the same thing when they say science. At one point Lois Althusser, a French Marxist philosopher from the middle of the 20th century, characterizes the history of Marxism as the struggle of science to clear itself of ideology. By this he means that unbiased, open-ended inquiry is science--and can be philosophy as well as it can be genetics--while inquiries with an interest in what can or can't be the answer are ideologies. From this perspective, the contrast between religion and science is apt, and can even be pushed to the point of equating religion with ideology (Harris talks instead of dogma, but means much the same thing).
Adorno, though, and Derrida subsequently (really this could be a very long list), shows us something a little different. It is unrealistic for any of us to ever presume a clear picture; we all always have biases, and these biases will always form "epistemological barriers," as Althusser says. It is naïve to suppose we will ever perform without them, and dangerous and condescending to claim we do. This is not to say that science is a religion, but that all thought, no matter how radical or scrupulous, is partly habitual--which is almost to say partly linguistic.