Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

So what do theorists do?

  1. Feb 14, 2008 #1
    This question has been on my mind for a while now, and I don't want to bug any of my professors in case they give me something to do.

    I know what an experimentalist does. Devise some experiment, build it, take data, analyze data. Either your experiment gives evidence for something or shows that it just doesn't happen (say a new particle being detected under XYZ circumstances). Unless you get an "inconclusive" as a result, but that's irrelevant to my point.

    So it's all pretty straight forward. If people start bugging you about your work, you show them where you are at in the experiment, where you hope to be in the future, etc.

    But I can't think of what a theorist does. I can't imagine it's something like "Oh I'm in the process of solving this equation but it will take me another month." or the like. I can see people finding new ways to solve really hard equations (i.e. using Perturbation Theory or something like that), but partial results probably don't get you anywhere, so you can't just say "I tried it, it failed.", like an experimentalist would. Could you?

    This goes for mathematicians, too. I know math is something different, but similar in the research I suspect. And grad students? How do they help a professor in their research?

    Sorry if this is a stupid question. :(
     
  2. jcsd
  3. Feb 14, 2008 #2

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    In a sense, its no different than any other field. The grad students do the lion's share of the work and put the professor's name on all the papers, sometimes the first name on the paper.

    I just got my name on a paper this way. I had the original idea on how to solve a problem and did a bit of the initial math. Except for this initial math, someone else did all the work. Since it was my idea, my name is on the paper. Since someone else did all the work, my name is not first.

    I also didn't get my name on many papers. I helped an intern through a messy problem last summer (I helped him a lot last summer, probably too much, but it was a fun problem). The problem is messy enough to make it the core of his thesis topic. He spent an additional nine months on the problem. I got an acknowledgement in an upcoming paper, which is fine. He did nine months of work without any consultation whatsoever.
     
  4. Feb 14, 2008 #3

    Danger

    User Avatar
    Gold Member

    Having no experience with formal education, I'll just put in my amateur opinion. A theorist comes up with the stuff that the experimentalist tests for.
     
  5. Feb 14, 2008 #4
    Sometimes it really does boil down to such a situation. In almost all seminars I have attended the research has been done with a set of very complicated equations that has been simplified through scaling or other arguments (i.e. steady-state, spherically symmetric, etc). They continue to solve them with as much rigor as possible and adding more complications as results improve from older methods.

    Disclaimer: I am an undergraduate and am just stating my opinions and observation from seminars and my own research experience.
     
  6. Feb 14, 2008 #5

    Ben Niehoff

    User Avatar
    Science Advisor
    Gold Member

    Then there is also "phenomenology" (not to be confused with the Sartrean philosophy of the same name), which is taking the theories and trying to derive testable results from them; i.e., to see what observable phenomena are implied by the equations. Both theorists and experimentalists participate in this.
     
  7. Feb 14, 2008 #6

    Astronuc

    User Avatar

    Staff: Mentor

    Something like that. Experimentalists develop the experiment and equipment to test a theory or model, and a theorist will develop the mathematical model or framework to explain the observation. It's really cool when a theorist can make a prediction, and then have an experimentalist develop an experiment in which the outcome agrees with the prediction.

    Experimentalists and theorists often collaborate, particularly when it comes to applied science. I do simulations because doing many experiments would be prohibitively expensive. So there are a limited number of experiments performed by different groups, and theorists then develop models based on the results of those experiments, but also with other basic research regarding the behavior and properties of materials. We put the models together and then do predictive analysis, and these days, we do a pretty decent job of predicting the behavior of complex systems and interactions.
     
  8. Feb 14, 2008 #7

    Danger

    User Avatar
    Gold Member

    I see... tit for tat (I wonder if the censor software will let that through). I had totally forgotten that an experimentalist can observe something that then has to be explained by a theorist. That's pretty embarrassing. :redface:
     
  9. Feb 14, 2008 #8
    Thanks for all your help, people.

    But that's not what I meant...

    I am asking what it is you actually did.

    Like my friend is doing research for a professor. Right now this professor is trying to figure out whether or not some experiment is viable with the equipment and what they are using, etc. That much makes sense.

    But when you have people coming up with new models, say for example Einstein and his relativity, or Maxwell, or whoever, how do you come up with something completely different like that? Does it just have to click or what?

    I have another professor who does biophysics stuff, so I assume a lot of his time is spent modeling behaviors of various cells and whatnot. I understand how you can say "Okay, assume it's spherical, in a vacuum" etc, then one by one add more complexities to the situation and try to account for them.

    But it just baffles me how someone can do theoretical research on some new theory or something completely different. Same with math. How do you do research for math, to prove something or come up with new math? It just seems so crazy.
     
  10. Feb 14, 2008 #9
    They also spend a lot of time scavenging for funding...
     
  11. Feb 15, 2008 #10
    Einstein was a theorist.
     
  12. Feb 15, 2008 #11
    A physicist who is a theorist has 2 main jobs.
    For areas where there is an established model, to compute results for previously untested regions. There may or may not already be experimental evidence for these results, if there is, the results are compared, if there isn't you wait for someone to do an experiment, and then compare the results.
    If there is an established model, but it disagrees with experiment of some sort, then the theory is only approximately correct, or only correct in some region. The theorist then tries to come up with a new theory (or a correction to the old one) which will be valid in a larger region.
    More generally than that, running numerical solutions, computing analytical solutions (or approximations), etc. and so on, and comparing the numerical/analytical solutions/approximations to each other and to experiment.
     
  13. Feb 15, 2008 #12

    Dr Transport

    User Avatar
    Science Advisor
    Gold Member

    There are two kinds of theorists..There are the ones who come up with the brilliant ideas such as Relativity, QM, QED, etc....

    The other type are the ones who do the pick and shovel work applying the theory to other cases and extending the theory.

    Most of the theorists are of the latter not the former.

    Let's give some examples, Einstein, Schrodinger, Dirac and Feynman were of the first type, brilliant leaders.

    Now Bethe was more of the second type, he was brilliant, but his strength was in putting everything together that existed and furthering it.
     
  14. Feb 15, 2008 #13
    I understand what you mean by furthering what we already know. That makes a lot of sense.

    I guess creating new theories and stuff just takes a stroke of genius, then, and not really some logical progression?
     
  15. Feb 16, 2008 #14
    I think you have to have been seriously puzzled about something for some time. Then it has to click!
    There is a very good account of this in the first ten chapters of "Personal Knowledge" by Michael Polanyi (experimentalist cum theorist), (Harper & Row,
    New York, 1964).
     
  16. Feb 16, 2008 #15
    There is an awfull lot to be puzzled about if you just so happen to try and make sense of the last glacial transition. The number of studies in the different areas double about every year and maintaining situation awareness is increasingly hard.

    I think that the key is in understanding why the European, South American and GRIP ice core transition to the Younger Dryas dates to 12,700 years BP and the Cariaco basin - GISP ice core transition dates 12,900 years. I try to write a 6000 word study on it but given the plethora of evidence it gets to at least 9000. And most definitely, something in some suppositions is definitely wrong and that's when it gets very tricky.

    In this case, theorists should be seen as "generalists", who try to maintain overview of the whole picture, not a confined specialism.
     
  17. Feb 16, 2008 #16
    It funny how that works. I have solved some serious problems at work by having things just click -- usually while driving for some reason. I don't know what it is about driving, but I have the best ideas while driving. The best way is on a lonely dark road through the desert. Although I had the solution to one of my most difficult problems come to me while driving through Los Angeles.... Weird.
     
  18. Feb 16, 2008 #17

    Danger

    User Avatar
    Gold Member

    I think that it's just because driving (even in a city) is mostly an autonomous function to a lot of people. Your brain has to remain at a high level of alertness, but the physical actions are pretty much just muscle memory. You don't have to think about what you're doing with your arms and legs. That frees up a lot of the alert brain to think about other things. I'm not sure about that, of course, but it's how it seems to work for me.
     
  19. Feb 16, 2008 #18
    Yes, it works for me too: not driving a car through a desert but rather a bicycle through a forest to and from work. I guess the details are not so important.
     
  20. Feb 16, 2008 #19

    Danger

    User Avatar
    Gold Member

    Same principle, regardless of the vehicle.
     
  21. Feb 16, 2008 #20
    Yeah, I thought it was common knowledge that spinning rubber emits Braintrons, which helps you think.

    Oddly enough, reality TV emits anti-braintrons.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: So what do theorists do?
  1. What to do when sosad? (Replies: 21)

Loading...