53

In many of the research groups I’ve worked at or visited, there is a culture that endless hours in the lab equal successful researcher. (I am in a theoretical field, so requirement of long-running experiments are outside of the picture. Let's ignore them in this discussion.) In chemistry, a widely-known example of this culture is the following letter:

               enter image description here

I would like to provide for you in written form what is expected from you as a member of the research group. In addition to the usual work-day schedule, I expect all of the members of the group to work evenings and weekends. You will find that this is the norm here at Caltech. On occasion, I understand that personal matters will make demands on your time which will require you to be away from your responsibilities to the laboratory. However, it is not acceptable to me when it becomes a habit.

I have noticed that you have failed to come in to lab on several weekends, and more recently have failed to show up in the evenings. Moreover, in addition to such time off, you recently requested some vacation. I have no problem with vacation time that is well earned, but I do have a problem with continuous vacation and time off that interferes with the project. I find this very annoying and disruptive to your science.

I expect you to correct your work-ethic immediately.

I receive at least one post-doctoral application each day from the US and around the world. If you are unable to meet the expected work-schedule, I am sure that I can find someone else as an appropriate replacement for this important project.


I have fallen prey to this during my PhD, doing very long hours. Now that I manage a research team, what is good advice to help fight this culture in my team? (Obviously, I don't want students and post-docs to get the message that little work is required either.)

Things I already do to that end:

  • When a new group member comes in and they get the out-of-hours building pass (for Sundays and late nights), mention that they are not expected to use it on a regular basis.
  • Avoid planning meetings at unusual hours (bank holidays, week-ends, etc.)
  • 22
    Is that a real letter? – emory Oct 27 '12 at 13:50
  • 17
    I thought USA overcame slavery? – Hauser Oct 27 '12 at 14:11
  • 14
    @emory Yes, the letter is real. For more background, click the letter itself (it is a link to a more detailed discussion). It is famous at least in the chemistry community. The letter’s addressee actually enjoyed a successful career. Interviewed in 2010 for the Boston Globe, the letter’s author “said that he had been advised by a lawyer not to comment on the validity or the context of the letter” (duh!). – F'x Oct 27 '12 at 14:23
  • 10
    @user2643: Then why didn't the letter say that? The letter specifically stated, "In addition to the usual work-day schedule, I expect all of the members of the group to work evenings and weekends", and complained only that "I have noticed that you have failed to come in to lab on several weekends, and more recently have failed to show up in the evenings." It did not include any complaints about a failure to fulfill "the usual work-day schedule" (aside from an explicit vacation request). – ruakh Oct 27 '12 at 16:59
  • 13
    @ruakh I think user2643 was implying Guido was one of those lazy post-docs trying to get away with working 9:30-4:30 on Saturday, Sunday, Christmas, or the day of birth of one of his children. – emory Oct 27 '12 at 20:50
45

In the different groups I worked in, their respective cultures were always the result of the behaviour of the leader. So lead by example. In particular, some useful hints from my experience would be:

E-Mails: don't send e-mails in non-office hours, especially not requests for help with something. No e-mails on weekends and late night are a signal which everybody "gets" after a short while.

Social activities: support group's social life. A group hike to nearby mountains (in office hours on a working day!), or a joint trip to a museum sends a signal that taking time off is an important part of one's life. We used to do it on occasion of having a guest researcher (for a study stay, or just a seminar talk). It's good for group's cohesion and again tells people that rest time is also very useful.

Personal relationship: care for private lives of your group members. Build relationship with their families. Speak to them about their personal activities, about their vacations, kids, etc. and do not forget to give back and speak about yours. These kinds of discussions send a powerful signal that you perceive time off in a positive way and even encourage it, since you take it as well.

Plan well ahead: clear plans and roadmaps in projects allow people to plan their time off as well. Nothing more annoying than a spurt interfering with one's private life to deliver some report, because the boss didn't care to tell the group ahead.

Time in the office: being in computer science, where I do not need to spend time in a lab and can easily work anywhere, I was always lucky to have bosses who cared for deliverables, rather than for my time in the office. After all, most of my good ideas are born in weird places, such as under shower, while on a bike, walking in woods, and by reading stuff outdoors. Making it clear with the group members that you care first and foremost for deliverables and are flexible regarding the actual time in the office (within reasonable bounds and respecting the local laws and regulations) works well. My experience is also that people tend to deliver better if they are given the power to plan their time and process, rather being forced to mantinels set by their supervisor.

Generally, I found it always very comfortable in groups where having kids was something nice and for what the group leader and members cared. Nothing is worse than a group where work interacts negatively with one's family life and where going home before 6pm to care for kids at home is perceived as something bad.

  • 11
    Should I feel wrong for reading this on a Saturday morning? ;) – gerrit Oct 27 '12 at 10:51
  • Good idea regarding emails; I often use boomerang to send emails at a future date (eg 8:30 am on the next working day) – Abe Nov 2 '12 at 2:16
17

Well, can your postdocs/grad students/whoever compete with their peers if they only spend 40-60 hours working instead of 60-80? If the answer is "no", you're actually doing them a disservice by trying to fix the system on your own. Some fields don't require much thought or brainpower. You can plan a thousand hours of experiments in an afternoon, and then you're left to just do it. In that case, your postdocs/etc. will just be outworked by others, which will negatively impact any future careers they might want in science. It's not very nice, but that's the way things are.

On the other hand, if you're in a field where quality of work and depth of insight are really important, you should not merely stress working less, but stress working intelligently and efficiently. This will help your postdocs/etc. to maintain a less absurd and exploitative lifestyle while still producing results that are as good as or better than those who work under incredible pressure.

  • 17
    Doesn't your first argument assume that success is more important that happiness? – JeffE Oct 27 '12 at 10:14
  • 14
    Regarding success vs. happiness, there's a genuine trade-off here. If a student decides they care more about their personal life than career success, then that's reasonable. If a student hopes for a tenure-track job at a research university and doesn't realize that their advisor isn't pushing them hard enough to achieve that, then that's not a positive outcome, even if the student enjoys grad school more in the meantime. – Anonymous Mathematician Oct 27 '12 at 13:14
  • 7
    @DQdlM - You can change the culture by producing equally good science without the obsessiveness. You cannot change the culture by willfully underperforming on principle; this leaves nothing by which to distinguish you from someone who just isn't very good. (Or you can change the system by reclassifying scientific research as work, and requiring overtime pay and such, but this is a legislative not individual action.) – Rex Kerr Oct 27 '12 at 19:38
  • 4
    I'm not convinced it's a just matter of quality vs. quantity. Instead, you get increasing (rather than diminishing) returns. I believe that if you systematically spend more time thinking and learning, you won't just produce more work at the same or lower quality level. Instead, your insights and perspective will increase like compound interest. In the end, you'll do work that's deeper and qualitatively better than what you could have done with less time. If you can manage to work 60 hours per week efficiently, you'll do better research than someone with equal talent who only works 40. – Anonymous Mathematician Oct 28 '12 at 15:58
  • 2
    @AnonymousMathematician - I don't argue with that; the question I'd ask is whether you're going to spend, say, 75 hours a week at a lab bench, or 40 hours at a lab bench and 10 hours a week learning how to program in Matlab and 5 hours reading popular accounts of science you don't know and so on, then the person can possibly both have a more balanced life and be more productive. If you are actually a mathematician, I'd point out that you are supposed to spend your research time thinking and learning. This is not true for the majority of researchers even in the sciences. – Rex Kerr Oct 28 '12 at 16:16

Not the answer you're looking for? Browse other questions tagged or ask your own question.