So yesterday I linked to a couple of anti-relativity sites, and tearing down their arguments might be fun, as gg suggests in the comments. But that’s just the tip of the iceberg of people who have a beef with some part of physics. And the folks who think physics is in need of an overhaul aren’t limited to their own little websites, or posting to science forums. They write books, too. There’s a discussion about one such author over at Pharyngula (and the same topic popped up today on SFN) but it’s been dealt with pretty thoroughly over at the JREF forum. I haven’t read the book, so I’m not going to raise specific objections here.

But there’s an attitude that is presented, and echoed elsewhere, that the physics is wrong because it doesn’t explain things. And these people are hell-bent on explaining things. But they go about it in the wrong way — they seem to be mortally afraid of making a specific prediction. Of using math in a useful way.

A common refrain goes along the lines of X is not a mathematical model, and it does not, nor does it need to, reduce mathematically to the physics I’m attempting to supplant

And that’s the problem. It’s great that you can “explain” phenomenon A. But can you also “explain” phenomenon B, that doesn’t actually happen? If you can explain anything, then it’s not very useful. A useful explanation can’t be some vague handwave. The strength of the mathematical model is it allows one to actually calculate things. It’s not good enough to explain the bump on my graph. I want to know if the bump should be at 10 eV, because if your explanation puts it at 15 eV, then you’re wrong and need to go back and change something. Your work has to be able to be tested for being wrong in a verifiable way