Duane, whose (ab)normal interests usually settle around ancient epigraphy responds to the blog storm elicited by John Hobbins series of posts on "Canon": Thinking About Canon (Part One) and First Update with a question: Why The Need for a Canon? It is, from a human perspective, a very good question. Working with John's functional definition of a canon:

A writing is canonical if and only if passages from it can be appealed to for the purpose of establishing a point of doctrine.

Duane asks:

...why would anyone or any group want to do that?

and like all good teachers, he answers his own question

A written authority, often, but not always, of obscure origin replaces a human authority. And it does it precisely in those areas of human thought where no human can be authoritative: religious doctrine.

Sociologically it is a good answer, but I think there is a little more to tease out here. A canon is a closed list of varied works - I realise that a canon need not be varied, though the Christian and Jewish ones John is discussing are, and need not be closed as indeed, at least for many centuries the Jewish canon was not (though I suspect that at any time it "felt" closed). As such a list a canon, as authority, allows an interesting mix of stability and flexibility.

Stability is useful to counteract the natural tendency for humans to "go off the rails" - at least with a canon there is some standard (the "rails") to call the human leader(s), or even whole communities, back to. (I can't think of any easy and simple way to avoid that sentence ending with a preposition ;-) Thus having a written authority over against human authority can be a "good thing". Both the reformers and the anabaptists are examples of such a calling back process that used Scripture as their primary tool.

Yet a canon is also flexible, the variety in the writings allows a greater degree, perhaps even a greater kind, of interpretation and thus allows for significant change. Think of the example of slavery, to some extent the battle for the Western mind over slavery was a battle over the interpretation of canon. A canon (of the sorts that Christians and Jews have) is both a tool for stability and for change.

Bob expresses this sort of idea vividly on a personal scale (I'll quote just a few lines here, but there is so much else in his post related to our question of the psychology of canon):

I must admit I like the canon I think I have. And I am not sure I could define it. I have my favorites - Psalms, Leviticus, bits of Genesis, Exodus and Deuteronomy, Job, Lamentations, Ecclesiastes, the Song, Jonah, large chunks of Isaiah, bits of Jeremiah, and in the NT - Romans, John, Hebrews. I am grateful that the forest is large and for a late starter, too large, but I am also grateful that it has a border. I am grateful that the trees are varied.

In his "Second Update" John also explores some of this, with a particular focus on Christian praxis in its relation with canon and doctrine (personally doctrine is a good word, but I am not so happy with "dogma", John) in this he is responding to a post by Doug (which I have not managed to mention above, but should have).

[12 June: Quote corrected to mirror author's correction]