November 15, 2005

Stringing along

Posted by Arcane Gazebo at November 15, 2005 1:18 PM

Yesterday's colloquium was entitled "String Theory and Cosmology", usually a sign that I can safely spend that hour in the lab trying to get my qubits to work. If I had known that the speaker would be giving the talk from handwritten transparencies I definitely would have stayed away, figuring that the talk was so overly technical that Powerpoint couldn't handle it, and the speaker would be running through some incomprensible morass of equations and text that had been lifted from the Necronomicon and then translated a couple times by Babelfish.

But fortunately I did go to the colloquium, which turned out to be pretty accessible. The speaker, Shamit Kachru, was very good and able to give sort of a hand-wavy outline of what string theorists are up to. String theory is a very difficult and jargon-heavy subject, and there was no way for him to get very technical without losing 95% of the audience (myself included), so I can't say that I gained much understanding of what string theories are actually about. However, I did at least grasp where the boundaries of knowledge are in this field, which I think can best be classified using the epistemological scheme invented by philosopher/poet Donald Rumsfeld:

  • Known knowns: The Standard Model of particle physics, which describes the behavior of particles in certain regimes (i.e. the experimentally accessible ones) to very high accuracy. And general relativity, which describes gravity in observable regimes.
  • Known unknowns: What gravity does at energies where it's comparable to the other three forces (it's normally much weaker). Also, various mathematical quirks and inconsistencies in the Standard Model.
  • Unknown knowns: Various string theories generate universes that look sort of like this one. But it's unknown whether any of them do describe the actual universe, because they only make interesting predictions at energies much higher than could possibly be tested. (I believe the number cited in the talk was 1017 GeV; the best accelerators run at 103 GeV.) The connection to cosmology in the talk was in trying to explain the origins of the universe using string theory; out of all of the potential verieties of theories, a few do make testable predictions on observable phenomena like the cosmic microwave background due to how they address the Big Bang. So if we're lucky enough to live in one of these universes, we could confirm it with certain astrophysical experiments.
  • Unknown unknowns: And then there's the possibilty that string theory isn't the right answer, but rather something no one's thought of yet. As Douglas Adams noted, "There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable.
    There is another theory which states that this has already happened." (String theory always reminds me of that quote.)

Now, for serious string theory blogging you should be reading Cosmic Variance, since I don't really know much at all about the field beyond what can be communicated in an hour-long colloquium. However, I'm starting to understand why it's interesting. (Also, it turns out that the guys shambling down the halls around here muttering about "braaaaanes" aren't zombies but overworked string theorists. Oops.)

I just discovered that there are videos of the colloquia on the physics department website, here, so you can actually watch this talk if you're interested. (It hasn't been posted yet but probably will be within a week.) Another good one from this semester was "Cycles in Fossil Diversity" by Rich Muller, which was a study of what causes species to thrive or die out at apparently regular intervals in Earth's history.

Tags: Colloquia, Physics, Science


I had the chance to see Brian Green speak about this once. That was entertaining. We also had a lecture two of beyond the Standard Model in our QFT class a few years ago, but it was lightning fast and basically at the colloquium level anyway.

As for the talk's medium: I've seen both good and crappy talks in essentially all media. Some of the pure math ones (including in areas outside of mine) really work best on the blackboard and if I am going to include a calculation that I want people to follow, I'll go ahead and give a blackboard talk (and perhaps reserve the projector for one or two graphical things I want to show). [I can always refer people to the Web for other sorts of graphical things...]

The Rumsfeld link was sweet, by the way!

Posted by: Mason | November 15, 2005 5:09 PM

That's true about different media. I took the transparencies (incorrectly) as a sign of mathematical intensity, but a mathematically intense talk given in Powerpoint is often much worse. (People tend to flip slides too quickly in that case.)

The Rumsfeld link is definitely a classic; I believe it spun off into a book of unintentional Rumsfeld poetry. I stole the idea of ironically using his classification scheme from some other blog post I saw a while ago (by Brad DeLong, I think).

Posted by: Arcane Gazebo | November 15, 2005 8:56 PM

Rich Muller's extinction stuff is regarded with a fairly wary eye by the geology department, FYI. Not that it isn't *good*, it's just...not proven.

Posted by: Wren | November 16, 2005 11:13 AM

I did get the impression from his talk that there's a lot of uncertainty surrounding those results. And he really wants them to support his Nemesis theory, but they don't. Regardless, he was a good speaker.

Posted by: Arcane Gazebo | November 16, 2005 4:51 PM

The other problem with the cycle theory is the way that classification works. Paleontologists don't yet have a great idea how biases in the fossil record affect latter studies. Classification of things is, at best, an inexact science (On the days that I'm most frustrated it's simply handwaving and table pounding with a slight pinch of name-calling to liven the debate). Nobody really knows what effect either of these are having, other than that they really should be having one.

On top of that many people don't want to sit down and do much of the math that would be required to figure out what sort of situation can be caused by something like differential exposures and more well studied groups. The work wouldn't be hard, but there's no money in it and it would be boring as all heck.

Posted by: Curtis | November 17, 2005 8:38 AM
Post a comment