Elegant but wrong

Advances in physics in the 1970s and 1980s led to huge improvements in our ability to predict phenomena precisely. Since then, not so much:
"All of the theoretical work that's been done since the 1970s has not produced a single successful prediction," says Neil Turok, director of the Perimeter Institute for Theoretical Physics in Waterloo, Canada. "That's a very shocking state of affairs."
* * *
"I can't believe what this once-venerable profession has become," [writes Sabine Hossenfelder, a physicist at the Frankfurt Institute for Advanced Studies in Germany]. "Theoretical physicists used to explain what was observed. Now they try to explain why they can't explain what was not observed. And they're not even good at that."

4 comments:

E Hines said...

Instant gratification. I want my Maypo.

These guys ignore the clumpiness of human science advances. There were centuries between watching stars and planets move (or not) and figuring out some means of heliocentricity. There were centuries between Greek advances and the Renaissance and centuries more before the fertility of the 20th century. And we've been a couple of decades without major advances? Pf-f-f-t.

It would be cool if we maintained constantly the pace of any of those fertile periods. Had that been the case since we first fell out of the trees and bumped our fannies on the Savannah, we'd be out among the stars today and making our own star/planet systems in convenient locations. But we don't maintain that constant pace.

It would be better if these worthies spent their energies exploring the underlying causes of that clumpiness, whether it's an educational thing, or a cultural thing, or a human nature thing that says I achieved all this stuff and now I'm out of ideas for awhile, or some other thing.

Eric Hines

Grim said...

I think it's charming that they believed that beauty and truth would align. I'd like to believe that, ultimately, they do. But it's not science to say that they do; indeed, science has little to say about beauty, and not as much to say about truth as many believe. Its theories can be falsified but never quite proven; the facts it discovers are still subject to doubts about epistemic access.

So ultimately the question is one of predictability. Can you tell me how this works well enough to get my ship into port? Even the pre-heliocentrics could do that, and originally better than the heliocentrics could, although the heliocentrics could do it more simply. Maybe that's where the love of elegance comes from.

Texan99 said...

Yes--if two theories predict well, and one is more elegant, I prefer the elegant one, too. What's been going on in too many scientific publishing careers in recent decades is a tendency to lose sight of the goal of being able to make predictions that can be tested. It's a way of looking clever on the cheap. We might as well be constructing horoscopes. Bad enough in some debased social sciences, worse in physics.

Mind you, I don't object at all when predictions fail. I object to the fashion of not caring whether they fail or not, or even whether a new theory makes predictions can be tested, before lavishing attention on it because it sounds so dreamy.

james said...

Sometimes a very powerful tool becomes an end in itself. There are very deep connections between some symmetries and physical laws. Studying these was extremely fruitful for many years, and left the tantalizing notion behind that the physics _had_ to be the way it is because of some fundamental symmetries, and the constants like G _have_ to be the values they are. (That's a very attractive notion to some scientists who don't like things being arbitrary.)

String theory has some very attractive ideas behind it, and I follow string theorist Lubos Motl's blog. However, so far string theory predicts things like particle supersymmetry (for which there's no evidence), and I also follow the Not Even Wrong blog of the string theory opponent Woit.

We know the theory we have isn't complete. It is good enough to do some things, and does them very well, but something's missing--most obviously a connection with gravity. Our "canonical theory" isn't one theory but two--quantum mechanics and gravity. They are obviously connected somehow (same universe!), but how? One big problem is that the linkage between the two theories happens at energies or distances that are very hard/impossible to probe, so experiments are right out. We got the "easy" stuff done, and have some nice theories to work with, though.

I suppose string theory is more of a framework than a theory. I can see what makes it attractive--it ties together several things in a natural way and makes some nasty problems like gravitational effects from quantum fluctuations behave nicely. Maybe somebody will find the key that ties the framework down to a solid theory, and then people can go about the business of making predictions and testing them.

String theory predicts that the existing particles (electron, photon, quarks, etc) have "supersymmetric" counterparts. It isn't the only theory that predicts supersymmetry, just the most popular. So far we have found none of these particles, and the limit on the possible mass keeps getting higher and higher. You may draw your own conclusions. I'm willing to be surprised, but I suspect that the way forward is to find a theory that doesn't need the superparticles. Maybe there's such a theory in the string theory framework.

I won't find it--I'm an experimentalist who wound up in IT. ("If once you start down the dark path, forever will it dominate your destiny.")