Why (some) physics professors become managers

This post addresses a complaint I often hear and read about: namely that some physics professors, after a certain stage of their careers, no longer do calculations themselves (if they are theorists), or actually spend time in the laboratory tinkering with the apparatus (if they are experimentalists).

Instead of performing any practical work, they become managers of students and postdocs and collaborations, overseeing the work of those who carry out the actual scientific labor. This move is perceived as a copout, signaling the fact that the professor has reneged from doing serious research or real science, in some sense.

Before responding to the above, I would like to clarify that once they occupy a permanent position, there is no formal compulsion on a professor to function in any particular way - I often say there are at least as many ways of doing physics as there are tenured professors. Indeed, I have seen professors choose a variety of operating modalities, including both the ones mentioned above. All categories contain big names, so I do not think a case can be made one way or another. But a comparison of the two styles referred to above might be revealing.

As graduate students and as postdocs, most scientists have to do their own calculations and spend (usually a lot of!) time in the laboratory doing their own experiments. They have the benefit of not having any other duties, largely speaking - e.g. no grant writing, no department administration, no teaching (usually). Typically many people in such positions are young, single, and without children at this point in their lives. Due to all these reasons, they can, as they are expected to, devote large chunks of the work day to (often tedious and) uninterrupted scientific research.

A tenure-track assistant professor who is setting up a theoretical or experimental research program might still have to do substantial calculations (writing papers themselves) or spend time in the lab, building apparatus, while discharging the grantsmanship, teaching, and service duties mentioned above. This is an intense workload and cannot be sustained indefinitely without inviting cardiac arrest. Usually it lasts about five years, until tenure is secured.

What happens next depends on how big a group the professor decides to run. Some professors decide to restrict the size of their group to a couple of graduate students, and maybe one postdoc. This typically allows them to perform as much hands-on research as their other duties permit. Even then, the ensuing research output can be quite substantial. I have seen theorists who operate like this publish single-author papers, and experimentalists basically lead their teams by turning the knobs in their lab.

With few exceptions, this implies that the range of research topics is not very large for theorists (due to finite resources and the need to stick to one line for funding reasons - a common reason for being denied a grant is that the professor has not demonstrated prior expertise in the area, which makes it difficult to switch topics). Likewise for experimentalists there is only one 'machine' (the sum total of equipment aimed at realizing the physics being aimed at) that produces results from their group.

Larger groups, however, require more grants. Since not all grant proposals are awarded (the success rate for applications to the NSF is about 15%), this means a substantial amount of faculty time goes into raising funds. Such groups also typically require the advisor to provide more supervision time - assuming this amount of time is the same per student - since there are more students.

If the students are highly capable, or the postdoc(s) are good leaders who can train the graduate students, the theorist can still find time to do some complete calculations on their own, though this is rare to see. The experimentalist now of necessity has to divide time between multiple - usually complicated - experiments and cannot be present everywhere all at once. Also, typically, in this state of their careers, faculty often have young children. The family also now needs some of the attention that science could secure earlier totally for itself.

The above facts mean that in a large number of cases, theorists may only do exploratory calculations (or even just come up with ideas which others can translate into mathematical models) which are then handed off to graduate students or postdocs for further development. Of course, this development happens under consultation with the professor, with a weekly group meeting, and more frequent encounters/electronic contact closer to the submission date of the article. Likewise for experimentalists, the basic idea of the experiment to be done is laid out by the professor and technical advances - and often reversals -:) - are discussed regularly.

This necessarily means some distance arises between faculty and their research. A theory professor may not get to inspect every line of code written by their graduate student. And I have known graduate students who did not allow their experimental advisor near the (complicated) apparatus once it was working!

Both styles have advantages. Hands-on means more control, the pleasure of working things out for yourself, experiencing personally the delight of discovery, indulging your stylistic choices in writing, etc. Hands-off means you can use other people to learn about the world, explore many topics, think about research in a more global way, ensure the people you mentor turn into independent researchers. etc.

Personally, I find I rarely have the long stretches of time required for detailed scientific work - every 20 minutes I have to run away to put out some new fire: teach, grade, write grants, publish papers, review articles, host visitors, travel to give talks, build new collaborations, file reports, supervise students, organize workshops.... Nonetheless, I feel I can keep a strong connection to science by coming up with new ideas, which source everything else. With experience, wholly formed papers can come out of your brain, and in fact also grant proposal ideas with enough papers to generate multiple theses. Actually - and this might lead some people to ask to check my brain - writing grants is my favorite part of the job. But more about that in another post!