The Cults of Physics

There have been some comments recently on two separate, but related phenomena: the worship of genius in physics, and the unspoken hierarchy in physics with theorists at the top of the chain and experimentalists as the supporting casts. It’s certainly a pervasive attitude, and I had trouble trying to reason myself out of it.

A trend I have observed throughout my undergraduate education: The theory-inclined students always top the physics classes that they take. I am told that it is also the case in graduate school that the theorists perform much better than the experimentalists on qualifying exams. The obvious rebuttal to this is that assessment in conventional physics classes is dominated by problem sets and exams. So all that proves is that theorists are better at solving paper problems than are experimentalists. As any experimentalist will tell you, most experimental work isn’t concerned with solving those kinds of problems.

Should we expect, then, that experimentalists perform better at theorists on lab courses? I doubt it, because the problems one has to solve in lab courses are often just not anything like real problems in research. They are usually fixed problems with standard solutions, and most of the difference in credit one gets for lab courses lies in how thoroughly and rigorously analyses the data, and how kosher one’s method of taking data and choice of data was. However, the real ingenuity in experimental physics is finding a new or better way to measure something. This is not an option in lab courses, so they hardly give an indication of one’s potential as an experimental physicist. In any case, at least at the undergraduate level, everyone pretty much scores the same on labs — there are a few people who consistently are willing and able to spend 20 hours a week on their lab report, and these get their 10/10s, but us mortals who can’t bear to spend more than 10 hours fooling around with data they are not particularly interested in get 7/10s.

So the only mark of differential ability between experimentalists and theorists is that the latter tend be better at exams and problem sets. Does this indicate that the latter would be better experimentalists if they turned their minds to it? Only if you believed that being better at exams demonstrates having a higher “general” ability at physics, and that this translates into being better at solving experimental problems. Since I have no empirical evidence to back that last sentence up, all I will say is that it does not seem far-fetched that exam problem-solving does say something about ‘general’ ability (why else do graduate schools look at your grades?), but that the second half of that sentence is rather dubious.

On the flip side, can we say that experimentalists would be worse theorists than theorists if they turned their minds to theory? Obviously, the problems that theorists have to solve are of a rather different nature from exam problems. The latter tend to have fixed, known answers that one can in principle derive from well-established techniques; the former seem to demand extra dollops of “creativity” and physical or mathematical insight. A common assumption, though, is that an ability to solve the former implies an ability to solve the latter — perhaps because an ability to solve problems that have already been answered is a measure of understanding the subject, and one needs as robust an understanding of the subject as possible to solve unsolved problems. Given my lack of understanding of how theorists really work, I will settle for saying that it is plausible that experimentalists would not be as good at theory as theorists are. However, as many have pointed out, so long as experimentalists are good at what they do, it is absurd to say they are inferior to theorists. The ‘ranking’ one assigns to them is largely an artifact of the physics education system.

Another reason for the pervasive cult of theory, I suspect, is that most people who go into physics intially do so because of the beauty of its theories. This would explain why 75% of entering graduate students want to be theorists. Traditional physics education doesn’t help when most of class time is spent on expository lectures on theory. I know of someone here who initially intended to complete a double major in physics and chemistry, but switched later to chemistry because he was an experimentalist and was loathe to fulfill all the class requirements for physics, most of which are lecture-based, when he could get by with lab-only courses in physical chemistry. Graduate schools accept far more students who claim in their personal statements to want to be theorists than they know will be theorists, expecting that as a matter of course most of the wannabe theorists will realise that they are not cut out for it and will switch to experimental physics. Thus a significant proportion of students who end up as experimentalists are wannabe theorists. That is not to say they are failed theorists: they hadn’t even begun, so who knows what they could have been? But certainly the phenomenon of being shunted into experiment against their initial aspirations creates a pervasive atmosphere of being the “lesser” beings.

Another reason why it is tempting to rank experimentalists below theory is that lab courses give the impression that experimental work involves a lot of drudgery, most of which has only an instrumental connection with physics. A great deal of the distaste for experimental work stems from the very justified distaste for lab courses, and negative impressions from short-term research internships where professors generally assign drudge work to interns. Sadly, there isn’t an equivalent of a “theory” course where students get to see the negative aspect of theoretical work. Of course, such a course would not be quite different from real theoretical research, just as lab courses are quite different from real experimental research. Nevertheless, students are tempted, and often do, associate lab courses with experimental research, and doing problem sets with theoretical research. One often encounters difficult problem sets, but one never really encounters difficult lab courses — they may be extraordinarily tedious and demand painstaking data collection and analysis, but they aren’t conceptually difficult the way problem sets can be. And so students grow to think that outstanding theoretical problems must similarly be “more difficult” than outstanding experimental problems, even though the latter demand ingenious solutions that theorists are, in all probability, no better at coming to than experimentalists are.

If anything, writing the above has served to erase just a bit more of the cult of theory from my contaminated mind. As someone who was drawn to physics for philosophical reasons, I was probably more deeply contaminated by the cult than most, but working in real experimental research has helped erase a good deal of its influence. There’s still quite a bit of it left behind, though.

About these ads

One Response to The Cults of Physics

  1. Boaz says:

    Interesting post.
    I’ve also thought about this issue of the cult of theory.
    In some ways theory and experiment are just different types of necessary activity- a separation of duties for a larger project. One could also compare the physics/engineering split, or the theoretical/computational split.
    How could the theory/experiment split be made more healthy?

    I agree that thinking of experimentalists as “lower” creatures is unhealthy, and typically ignorant of the type of work actually going on, as you point out. But I think that theorists also probably think of experimentalists as having their own set of benefits from their work. For example, they may think that experimentalists know how to fix their car, are better at managing money and are in general better able to deal with the real complexity that the world has to offer. The theorist may say: “I’m stuck in this world of pet problems, dependendent on experimentalists to tell me what to work on and what is interesting.” The isolated mind-set of the builder of toy models may need a certain amount of artificial respect in order to maintain it.

    To help rid yourself of the cult of theory mentality, you could try to imagine a cult of experiment situation where the theorists where conceptual models were seen as little more than advanced computer programs. The theorists job was solely to look at the experiments and find attractive ways of presenting the data which the experimentalist would then grab and use to explain his/her work, with a footnote as to its source. I think string theorists may be in danger of doing this. They are applying their skills to RHIC data and will soon do so for LHC data, but they seem to be losing their role as driving the agenda of research.

    This idea of separation of duties has always been uncomfortable to me, but in the end it may be necessary to carry out really large projects. It then falls to other people of a more integrative mind-set to join the groups together and show what sort of creation is occuring overall.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

%d bloggers like this: