Is string theory science? Math? Religion? What is it?

There's lots of buzz about string theory recently. Two books (Not Even Wrong, by Peter Woit, and The Trouble with Physics, by Lee Smolin, an ex-string theorist), blog posts (Slashdot onceand again, Cosmic Variance, Uncertain Principles), and magazine articles (the New Yorker, Scientific American among others), and so on.
Like many, I feel capable of commenting, without the benefit of reading the books, much less understanding the basic prerequisites of QFT or the extremely hairy mathematics of string theory itself. But I do it, not to make a comment about science, but about religion.

I like Cosmic Variance's defense of string theory as an exercise. You can put the case for string theory quite simply: physics is incomplete. There are simple, physically well-posed questions that accepted theories of physics cannot answer: what is the gravitation field of an electron? What happens when you have strong gravity and high energy at the same time, as for example early in the life of the Universe? General relativity works well, the standard model of particles works well, but they are not compatible. Conventional science would do experiments, and be led by data before theory showed up to explain the data, but no data is available, or likely ever will be, for these kinds of questions. So the problem is simply to find a coherent, internally consistent point of view, which is compatible with both the standard model and general relativity. It's been The Problem in fundamental physics for nearly a hundred years now, and it has sucked up some of the best minds of several generations, without a lot to show for it. But if you care about Fundamental Physics, it's the only problem which exists. String Theory is, by collective consensus of the best minds of our age, the only likely candidate for a solution. Certainly experiment can't help, it's a completely theoretical exercise.

Sheldon Glashow's crack is that contemporary physics, in the guise of string theory, is essentially medieval theology. This is also correct: much medieval theology took, as postulates, the fundamental truth of a certain set of written documents, and then worked over the consequences of holding all those valid simultaneously. Very difficult stuff, really, especially considering the documents, and it occupied the best minds of our civilization for a thousand years. String theory is a mental exercise of precisely the same form. I guess, for me, this is a validation, not an attack. Yes, it isn't science, but it is theology, and very respectable theology too. Much better to work through GR + SM than some dusty texts selected by a fourth century committee.

Recently I watched the physics building being knocked down at the university where I work. Not a surprise, it has been outdated for decades. What seems apt, though, is that it will not be replaced. Or rather, it will not be replaced with a physics building. The physics department will be housed in some kind of interdisciplinary science center. Later, I presume, they will share offices and a secretary with the department of philosophy and philology of middle european languages. And they will find a happy home.

Coevolution of dogs and people

We've had Sheba for about a month now. She's getting used to us, and we her. I think the most interesting aspect is how natural, in some sense, it all feels. Yes, she snitches food, chews up valuable objects, and has angled her way into jumping on our couches and sleeping in our bed. But it feels like a normal experience, of primates and canines living together, setting up shared housekeeping with a companion animal. Not normal in the sense of us being used to it, quite the contrary, but normal in the sense of everyone does it, there are lots of dogs, lots of dog owners, it's a primary experience. Normative, perhaps. The state of our household, before dog, could have been called "dog deprived", as a moving Slate piece describes the family in an adoption.

(I learned from another Jon Katz Slate piece that the phrase, "companion animal", is thought by some to be a kind of newspeak for pet. It never struck me that way. I learned it from a vet, for whom the domestic animals his profession cares for fall into two classes: companion animals, and food animals. What we take from each is different, body vs. soul, but we take regardless.)

One of the dog books we got has an extended section on the evolution of domestic dogs. Dogs and people have been hanging out together for longer than was previously though, about 100,000 years or so. (I should really look this up, and get the story from the primary literature. Till then, you get it third hand, hopefully not too distorted.) 100k years is a long time, extending back to before we became quite what we are, in a biological sense. And some of those changes -- loss of excellent sense of smell, etc. -- are exactly the kind of thing which is well balanced by working canines. So it's not exactly the previous story, where dogs showed up, people domesticated them and used them. Rather, dogs and people got together earlier, before we were quite human yet. And sharing life with dogs is part of how we became what we are. So it should feel natural, being with dogs.

It's a comforting thought, particularly when she does some doggy thing, and we react in a natural manner, in contradiction to the carefully thought through guidelines of the training books. No, getting mad may not be optimal. But it probably does work okay, if somehow humans and dogs have been getting along fine, for thousands of generations before the training books were written. Unlike say computers, where one's natural tendencies (fist through screen, eg) may align particularly poorly with problem solving.