Is it possible to apply for a graduate program in theoretical physics without having a bachelor's degree in physics or mathematics? I'm completely self-taught. I'm half way through Jackson electrodynamics and Peskin QFT. These books are taught to students at a graduate level, let's say that I can pass the GRE exam and graduate level examinations. Is it possible that this would substitute for an undergraduate degree?

  • 2
    Have you asked that question to your prospective graduate program?
    – F'x
    Commented Nov 27, 2012 at 11:07
  • Yes , I have sent an e-mail to ICTP (international centre for theoretical physics) and didn't get a reply yet . Commented Nov 27, 2012 at 11:13
  • 3
    Do you have an undergraduate degree in some other field?
    – Lev Reyzin
    Commented Nov 27, 2012 at 18:46
  • Yes,I'm in my final year of a medical degree. Commented Nov 28, 2012 at 5:07
  • on a different note, should you not be considering a second bachelor's program in Theoretical Physics? That should be a cakewalk for you, except that it takes time. Alternately, you should seriously think about interdisciplinary programs linking your present UG preparation with your graduate aspirations.
    – Kris
    Commented Nov 28, 2012 at 6:11

6 Answers 6


I agree with the other answers — admitting a student without formal academic background is very risky.

As a first step, I suggest contacting graduate programs asking if they allow non-degree students to take their classes. If so, taking classes as a non-degree student will let you build up a formal academic background and strong recommendation letters in your chosen field, making your case for admission much stronger. And if you're admitted to the PhD program at the same university, you may be able to use the classes you've already taken toward your degree.

Taking classes will also give you a sanity check on your self-assessment. From personal experience, it can be really hard to spot gaps in the knowledge you build from self-study, simply because you don't know enough to ask the right questions.

But once you get your foot in the door, do not just take classes; do whatever you can to get involved in research!

One significant downside to non-degree classes is that you'll almost certainly have to pay for them out of your own pocket.

  • I agree with this advice. I was in a similar situation and some formal background is necessary. I am switching from CS to ecology and had to take biology and chemistry classes that I never took during my college career. Commented Nov 28, 2012 at 10:33
  • +1, this was the route I used to go from self-study at the HS level to admission in a bachelor's program. Writing the GRE (or SAT) for your subject in conjunction with this can make a very strong case. Commented Nov 28, 2012 at 14:04

...but I think I'm well aware of my areas of weakness and strength.

I don't dispute that at all. But you'd be in the minority. Most students coming into grad school think that they can handle any material, and once they're in a class, find themselves completely swamped.

The admissions committee's point of view is this:

  • Will this person be at sea in their breadth requirements because we have no idea whether they have the right background ?
  • Will they be a drag on instructors in their classes, and in general bring their cohort down (it can be dispiriting to have many students in a batch struggling to make it through the program: conversely, it's great when a batch has a number of bright spots who can pull the others up)
  • Should this person be chosen as opposed to someone else who has a more well-defined profile ?

If you can address the third point clearly ("why should you pick me") and the first ("I really do know my stuff"), then that will go a long way towards alleviating their concerns. For example, if you can point to independent research projects that would demonstrate your command of the material, that would be even more valuable than course credits.

  • Great answer! I'm going to put some effort on independent research because that's what I ultimately want to do although It would be extremely difficult without the help of a good advisor who can guide me through theoretical physics literature! Commented Nov 28, 2012 at 5:14
  • I've watched some recorded lectures at graduate level online . I think that the pace is very good. The lecturers usually explain things in more a elementary way than the textbooks that are required for the class. Commented Nov 28, 2012 at 5:22

Is it possible? Yes. Is it likely? Maybe not.

A lot will depend on your undergraduate program. If you've done something "related," such as mathematics or physical chemistry or some engineering disciplines, then it will be easier to convince an admissions committee that you have the requisite background. Otherwise, it will be up to you, in your letter statement of purpose, as well as your letter-writers in their letters, to make the case why you should be admitted to a rogram in physics when you don't have a background in the subject (or anything close to it).

This is a big risk for a department, and especially if the department you're applying to is small and therefore needs to be more selective in who they admit. Anything you can do to show that you won't be a risk for them will help your application.

  • But I have a background that I have acquired by self-study (although I haven't had any math/physics courses at university).Why does undergraduate credits would be of so much importance for my case if I have more than the minimum required background in physics and math? Commented Nov 27, 2012 at 11:40
  • 1
    @self-learner because credits are an important part of the way your knowledge is assessed in the existing academic framework
    – F'x
    Commented Nov 27, 2012 at 12:11
  • 8
    @F'x: Or more accurately, it's hard to assess achievement based on self-study. I've interviewed a few candidates for open positions in my group, and some of them were almost delusional in over-assessing their qualifications and their suitability for the position in question. You simply can't afford to hire the wrong person, so having the "pedigree" and demonstrable achievement is important.
    – aeismail
    Commented Nov 27, 2012 at 13:39
  • 1
    @self-learner I would rather spend time considering what specific knowledge they would be looking for, and ensuring that you are up to par (for this purpose defined by "could earn an A or B in a final exam given on that topic") in those areas.
    – eykanal
    Commented Nov 27, 2012 at 15:35
  • 5
    The issue is you need some sort of documentary evidence of your knowledge in those fields. Otherwise, anybody could try to make the case "I've studied so and so on my own." I've done self-study courses in high school--but at the end of the day, I had the standardized exams that I could take that demonstrated what I learned. Based on those exams, I was allowed to opt out of the introductory coursework in those fields. Without the exams, my word would not have been accepted. Maybe the schools have a placement exam or something similar you can take. Otherwise, you're probably out of luck.
    – aeismail
    Commented Nov 27, 2012 at 19:20

Is it possible to apply for a graduate program in theoretical physics without having a bachelor's degree in physics or mathematics?

Yes, I believe it is.

let's say that I can pass the GRE exam and graduate level examinations. Is it possible that this would substitute for an undergraduate degree?

First, depends what you mean by "pass" the GRE. I'd hope that you had very high scores (90th+ percentile) on at least the Quant. and Analytical parts of it, but hopefully all three.

Also, I'd edit your question to be clear that you do have an undergraduate degree already; just not one in physics.

My guess would be that if you find the right program they would be willing to really consider your application favorably if you could provide something like the following list of goodies:

  • Very high GRE scores
  • A glowing letter of recommendation from a physicist or some "hard" science prof speaking to your abilities to do heavy duty physics.
  • A pristine statement of purpose that goes into specific details about your research interests within theoretical physics and your program of self-study. Here you'd really want to emphasize your aptness for that particularly graduate program in physics, based on your interests that match well with what (at least one of) their researchers do.
  • Some research experience, if you can get some.
  • A very strong academic transcript with a significant amount of hard science and math courses and mostly As (or the equivalent) in those subjects.

If you can't provide most of a list like this, one might reasonably question why you would want to go on to graduate school in this field.


Depending on the field and department you are thinking of applying to, one piece of evidence that you have independently acquired a strong background in the proposed field of study can be a strong score on the GRE Subject Test in that field.

In computer science, a number of graduate programmes I've looked at recommend this for applicants that don't have a CS degree. For instance, cf. UC San Diego's PhD admissions FAQ: http://www-cse.ucsd.edu/node/195 under "Should I take the Computer Science GRE Subject Exam?"

I do not know for sure to what degree this would apply in Physics.


The answer to your question regarding the GRE is "it will help". You have not specified if it is for a PhD or Master's. However what will count the most is your research experience if you are going for a PhD. If you want to apply to a Master than it will be definitely easier to get accepted with high GRE scores.

Sometimes students with different backgrounds are actually well seen by the department, especially if it is a interdisciplinary one. However you have to substantially prove how you will be able to apply your knowledge into your new field and how this makes you competitive. Students with a different background might provide insights and different perspectives on how to approach research topics.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .