Experiment and Theory in the Popular Imagination

A little while back, I posted about the pro-theorist bias in popular physics, and Ashutosh Jogalekar offers a long and detailed response, which of course was posted on a day when I spent six hours driving to Quebec City for a conference. Sigh.

Happily, ZapperZ and Tom at Swans On Tea offer more or less the response I would’ve if I’d had time and Internet connectivity. Tom in particular gives a very thorough exploration of some of the reasons why experiment gets downplayed in popular physics. I particularly liked this bit:

I’m going to put forth a possibility: maybe we have a harder job, in terms of popularizing or telling our story (I’m not claiming the science part is easier). What I mean by this ties back to a story from a few years back, when we were saying goodbye to a colleague who had decided to leave to go to grad school in physics. Someone asked him if he was going to do theory or experiment, and the two physicists at the table pointed out that this is a false division: there are people who do theory, and there are people who do both experiment and theory. There is basically no category of physicist who does only experiment. If I am doing an experiment, I have to be aware of what the theory is, and use it, in order to set the experiment up and to properly analyze the data. While I don’t have to create the theory, I am not insulated from it.

That’s absolutely true, and a point that’s seldom brought out. Experimentalists are, for the most part, also required to be mediocre theorists, because you need at least a rough idea of what’s going on in order to design and carry out an experiment. On the other side, though, theorists don’t really need to know the details of the experiment to quite the same degree.

One of my favorite anecdotes from grad school illustrates this point fairly well: in the mid-to-late 1990’s, the effort to produce Bose-Einstein Condensation at NIST in Gaithersburg involved both a theory group and the experimental group that I was associated with (I wasn’t directly involved in the BEC stuff, but it was the Hot Topic of the day, so I followed what they were doing), and for a long time they had weekly meetings of both groups. This ended, however, after an hour-plus meeting during which the experimentalists debated whether to make the coils for a particular electromagnet out of copper tubing with a round cross-section, or copper tubing with a square cross-section. After that, the theorists insisted on splitting off a second, experimentalist-only meeting for those sorts of discussions.

But the interesting point for the current discussion is this: while there was a separate experimental meeting that none of the theorists went to, all of the experimentalists continued to go to what was nominally the “theory” meeting (in another building, across an open field strewn with goose shit, no less). Both groups needed to know the basics of the theoretical developments, but the experimentalists needed a bunch of extra stuff on top of that.

(Now, you could argue that the theoretical analogue of square wire vs. round wire is not what was being discussed at the theory meetings– that would be stuff like details of numerical integration techniques, or whatever, and whether to do the calculation in C or Fortran. I still think it’s true that the experimentalists needed to know more about the theory than the theorists needed to know about the experiment, though.)

There are also some interesting interactions between this discussion and the history-of-science reading I’ve been doing recently for the book-in-progress. The division between theory and experiment wasn’t always quite as sharp as it seems to be these days. Jogalekar cites Faraday and Rutherford as examples of outstanding experimentalists who weren’t good at theory, and it’s certainly true that Faraday didn’t have the mathematical background he needed to fully complete his discoveries. In another sense, though, this sells both men short. Maxwell eventually put Faraday’s ideas on a sound mathematical footing, but Faraday’s intuitive ideas about fields turned out to be dead on (though they were scoffed at at the time). Einstein famously had pictures of only three scientists in his office: Newton, Maxwell, and Faraday, which gives some idea of how he regarded the quality of Faraday’s ideas. And Rutherford, for all his disdain for professional theorists, was no slouch at manipulating symbols, and was adept enough to realize that the experimental results of Marsden and Geiger implied a “solar system” type atom (and also that such a thing was classically untenable).

It’s also been very striking, as I read about the history of the development of quantum electrodynamics (QED), to note how close the connection was between theory and experiment in the first half of the 20th century. In Schweber’s QED and the Men Who Made It, there’s a strong emphasis in the stories of all three of the eventual Nobelists for QED on “getting numbers out.” Tomonaga, Schwinger, and Feynman all came from a physics culture that placed huge emphasis on concrete calculations, and nothing was regarded as complete or well understood unless it agreed quantitatively with experimental results. Some of this was the fact that the problems were simpler and easier to connect to measurements, and some of this might be traceable to WWII, where even theoretical physicists were pressed into service in vast engineering projects. Schwinger is widely viewed as a theorist’s theorist, but he made very practical contributions to the development of radar technology, and Feynman somehow found time between lock-picking and code-breaking to run an extremely successful numerical calculation program at Los Alamos, providing concrete numbers for use in bomb development. And Willis Lamb, one of the people cited as a relatively unknown experimentalist by Jogalekar, was a theorist by training, but got pulled into experimental work by making magnetrons for the wartime development of radar.

(Of course, there were also people like Pauli, who was so inept in the lab that an experimental failure was once attributed to the fact that Pauli had been changing trains in the same city as the lab when the apparatus broke…)

This is getting kind of long, but I want to mention one other aspect. Jogalekar notes that most popularizations seem to come from the theory side:

The other big reason why for the public seems to downplay the key role of experiments is the bias in physics popularization toward theory. And here at least part of the blame must be laid at the feet of experimentalists themselves. For instance if we ponder over who the leading physics popularizers in the last twenty years are, the names that come to our minds include Brian Greene, Lisa Randall, Leonard Susskind, Brian Cox and Sean Carroll. Almost no experimenter makes the list;

I think Tom has this about right:

Another possible reason is that it may be a little easier to fit popularization into your schedule if you can do some of your work while out popularizing. I can’t do an experiment if I’m on a plane, or at a hotel. Theory is somewhat less constrained to being in one particular place. Perhaps that lends itself, in a small way, to this kind of outreach.

I’ll offer one additional point in the same general direction: if you look at the world of physics blogging, you’ll notice that most of the regular bloggers are theorists, and the handful of experimentalists out there tend to come from particle physics or astronomy. Why these fields, given that other specialties are more common in physics as a whole? I would guess that it’s partly due to the fact that in these fields, experimental work is done at a computer. They’re writing code to process images, or sift through terabytes of collision data for events that fit a particular profile. While that’s going on, there’s some slack time that’s easily used for blogging. The smaller scale experiments done in other subfields involve a little more hands-on, in-the-lab time, which reduces the number of experimental bloggers.

I’ll close with one final observation, about fields of research. Jogalekar writes:

There is no doubt that experimental physics has seen some amazing advances in the last two decades, so there’s certainly no dearth of stories to tell. For instance just last year the Nobel Prize in physics went to Serge Haroche and David Weinland who have achieved amazing feats in trapping ions and atoms and verifying some of the most bizarre predictions of quantum mechanics. Yet where are the books which elaborate on these successes?

Well, if you’re reading this on the ScienceBlogs site, you could look over in the right sidebar… If you’re on an RSS reader, here’s a helpful link. And lest this be just unseemly self-promotion, I’ll also put in a plug for Anton Zeilinger’s book (another experimentalist, it should be noted, one of the very best in the world).

It’s true, though, that there’s less popular material out there on experiments in quantum physics. Even popular books that get a good deal of press, like Cox and Forshaw’s are weirdly scanty about modern experimental physics– their book gives almost no indication that quantum physics is still an active area of research, or that anything interesting has been done in the field since the 1960’s. I’m really not sure why that is, and I’d love to see it change. Which is, after all, why I wrote my book the way I did…

Anyway, that’s a lot more than I thought I was going to write when I sat down to start typing. And now I need to run off to the opening session of the DAMOP meeting, to hear about awesome recent developments in experimental physics…