Wonderful program, as usual.
Three points:
1. Conversations about ?emergence? (like any conversation of anything that tries to explain something) assumes that we can converse/communicate about it. Thus, emergence is as constructed as anything else. [Yes, I am a radical constructivist.] The workings of the physical world, of which we are a part, is only present, because we communicate about it. How this communication is constructed/organized become (my) the philosophical problem. Simply: speaking of emergence assumes a position of an observer outside of what is observed. All I can observe, however, is you observing.
2. Niklas Luhmann, especially in his (untranslated) ?Society and its Science? (Die Wissenschaft der Gesellschaft) deals quite comprehensively with this.
3. Stanislaw Lem, in his ?The Invincible? (first published 164, translated into English 1973) deals in a harrowing way with the ?supremacy? of the non-thinking ?whatever-it-is? over the thinking human.

Mike, your question (at 10:46) is a good one. I think you are right up to a point: getting an empirical grip on the impact of an emergent or holistic property on a system's behavior requires that you have a very good understanding of the workings of the system's parts, so that you can determine just what the difference is from what you would have expected in the absence of the property.
That doesn't mean, though, that you have to accept a thorough-going reductionism. It's just that you have to start with the underlying processes. This suggests that scientists should be 'methodological reductionists' -- proceed on the assumption that there is a great deal to be learned about the system in terms of the inner workings of its parts. And that of course is how sciences do in general proceed. But doing so is consistent with the belief that there are, or might be, limits to this approach.

At several points in the discussion this morning, Ken used the word ?mysterious? to convey his discomfort with the notion of emergence. Mystery is revered in theology, but seems to be a term of derision in the scientific sphere. Yet, reductionism has its own share of deep mysteries, and Ken seems quite comfortable with his ?firm faith? that reductionism will ultimately triumph over its many mysteries. Might this be a case of inserting a materialist god into the same gaps where theists have inserted theirs?
One caller introduced the notion of intelligence evolving out of the quantum flux of the big bang and the universe becoming aware of itself. While I appreciate the religious sentiment behind this creation myth, I believe it?s no more or less valid a statement of faith than that of many sacred traditions. Whether we garb our beliefs about the deep mysteries of existence in sacred or secular language, we are nonetheless venturing beyond the domain of certainty and are firmly in the realm of the unknown and possibly unknowable.
Science and religion share common mysteries, and it?s important that both approach the unknown with rigor and humility, not substituting ?firm faith? for theories that can be verified through experiment. This is not to discount faith, but rather to separate hopes, ideals and wishes from what is actually so in the world.

Mike--I don't think you need to accept reductionism to understand emergentism. In fact, I'd consider myself an emergentist--of a sort, although my form of emergentism is probably too mild to satisfy most emergentists.
Here's why, in some sense, there are very clearly facts about the world that are not explicable in purely physical terms:
I think the following sentence is true: "Heat is molecular motion." That sounds pretty reductionist, right? But there's a catch--I don't think it's *analytically* true. "Heat" certainly doesn't *mean* "molecular motion" (despite what the Churchlands may have you believe); it is perfectly possible to understand the concept of heat without understanding the concept of molecular motion, or even molecules--people did so for tens of thousands of years.
So if "Heat is molecular motion" isn't analytic, it must be synthetic--a fact about the actual world, not a mere definitional or logical tautology. Now, is it a fact that can be described in purely physical language?(Er...let's not count "heat" as a physical term here, though it sort of is--I think most reductionists think everything should be reducable to talk of particles and motion, or substitute your favorite modern-physics equivalent.)
No. The closest we can come is "Molecular motion is molecular motion." And *that* sentence clearly *is* analytic--it doesn't express a fact about the world; it's a pure logical tautology. So there are, at least, some truths that cannot be fully reduced to, at least, dynamics. If I were more awake, I'd come up with something similar for physics in general, but you can see how the argument goes, at least.

I've polished up the above argument and posted it here:
http://avromandina.net/avrom/2008/11/the-unreducability-of-reductionism/