Normally I love The Daily Show. (Especially in HD) But Jon Stewart's interview with Marilynne Robinson last night really fell short of the usual standard. And not just because she wasn't funny. At 2:52 in:

The more you delve into science, the more it appears to rely on faith.

You know, when they start to speak about the universe, they say, well, it's actually, "Most of the universe is antimatter." "Oh really? Where's that?" "Well, you can't see it." "Well, where is it?" "It's there." "Can you measure it?" "We're working on it..."

No. Sorry. That's not science. For one thing, nobody with a clue really thinks most of the universe is antimatter. But the more important point is that observation is the core of science. If you can't see something, can't measure it, can't even produce any observable evidence that it exists, you can't pass it off as scientific fact.

I have to wonder where Jon Stewart is getting this idea that scientists are relying on faith, claiming that things are true without having the evidence to back it up. Maybe he's misinformed. I could sort of understand that; he's not a scientist, and neither are a lot of the people he talks to, so they might not know about our standards for analytic reasoning. Here's how it works: we have theories (or models), which make predictions, which can be compared with observational evidence to support or refute the theory. It's often the case that a theory is easy to explain but the justification isn't (even to other scientists!), because the predictions can be pretty arcane, so we do tend to glaze over the details of the evidence. I can see how it would be easy for a non-technical person who doesn't have background knowledge in the field to hear this as "just take my word for it." But it's important to make the point that with science, the thing people need to take on faith is that the theories actually are being fact-checked by scientists with the appropriate background knowledge.

Unfortunately, that's not always true. The problem is that, especially in high-energy physics and cosmology, the theorists are way ahead of the experimentalists. So we have fields like string theory that, for the time being, are completely beyond the realm of experimental support. That's fine in and of itself, but we have to remember that it's just a theory. It's all too easy to get carried away and start talking about how the universe is made of tiny little strings in hidden dimensions, but let's be honest, we really don't know. This is one area where I feel somewhat let down by my field; when talking to the public, we fail to emphasize that string theory (or whatever) is still in the first stage of the theory → prediction → experiment process, and I think people are so eager to believe in something new and cool that they jump to the conclusion that the stuff is accepted science. I can't blame them. And then of course they start wondering why it's accepted science, and we try to tell them "it's complicated" or "it's going to be proven soon" when we really should just come out with the honest truth: it's not. It's speculation. Guided speculation, yes, and perhaps speculation that could one day be supported by experiments (so it's not useless), but still just speculation.

The point I want to make is that any time we try to pass off an unverified theory as a scientific fact, we're undermining the fundamental principle of science. Experimental verification (plus independent review and some other related stuff), is really what separates science from religion or politics or any of the other "abstract" fields that we like to make fun of. And when people start getting the idea that science is based on faith as much as any of those other fields are, what's the point of having science at all?