is self-studying advanced physics a silly idea?

redoopi

Hello everybody. I am considering teaching myself physics in spare time, starting from college physics and multivariable calculus, to a level that I can understand most of the research papers on GR, QFT and cosmology.

There are some obstacles that I can foresee:

1. I will certainly get stuck somewhere, especially on the advanced topics. When this happens, I may not find someone to help me out, and have to skip the part that I don't understand. If it happens frequently, I may not be able to proceed.

2. The exercises in advanced-level textbooks usually don't have answers. This means that my understanding will be based solely on the text (and perhaps also by *trying* to do the proof-type problems).

3. The textbooks available on the market may not be advanced/up-to-date enough, so that a self-learner cannot obtain sufficient knowledge to understand research papers by just studying textbooks. (I am not sure about this).

I don't prefer to go back to school due to financial concerns, but I really don't want to self-study for many years, only to find myself stuck in a mud pool and can go nowhere.

Do you think it is feasible to self-study physics to such an advanced level, for a person with average intelligence and can devote 2 hours per day for the next 20 years? or is there really no choice but to study in school? Could you give me some suggestions on how I can overcome the obstacles mentioned above?

Bipolarity

To the OP, welcome to PF! I am curious, what degree do you already have?

I am also curious about the answer to the OP's question since I could be in a similar situation except my undergrad degree would be in math/engineering. Some of my physicist relatives have suggested that advanced physics, like QM, is very important in some fields of EE, particularly optoelectronics.

BiP

ZombieFeynman

I doubt you will find any amount of time you put into learning about nature a waste.

How far can you get on your own? I dont know.

Bipolarity

To the OP:
Since you are very enthusiastic about self-studying, here are some resources you might like:

www.edx.org
www.udacity.com
www.coursera.org

Udacity is especially good for computer science, the others are generic.

BiP

redoopi

Thank you very much BiP. My undergrad degree is computer engineering. I think you are in a much better position than me: Since some of your relatives are physicists, you can call for their help when you have questions on physics.

The web sites you gave me are of introductory level. This is the area that I am less worried about, because there are many textbooks and exercises with solutions at this level, and I can have a higher chance to get help from online forums like this.

redoopi

I wholeheartedly agree with this, ZombieFeynman, but my ultimate goal is to understand how the universe originates and evolves as much as possible. I do look forward to learning about specific phenomena such as the nuclear processes in stars, but to me they are much less important. That's why I am so eager to know if what I am planning to do in the next 20 years is really practical.

A. Bahat

I can't really comment on the feasibility of your plan, but I can address some of your concerns.

Why should you need answers to get something out of the exercises? To be clear, I'm not saying, "Suck it up, answers are for wimps." But the fact that there are no answers doesn't make the exercises worthless. And you can always ask questions here, or elsewhere (the stackexchange websites come to mind) if you are stuck or unsure on some exercise.

Don't worry about how up-to-date textbooks are in physics and math. As a general rule, unless you are basically at a research level, the stuff you will be studying was figured out a long time ago. (On that note, please do yourself a favor and get older/cheaper editions when they are available.) And when/if you do get to a research level, you will be learning from papers, not textbooks.

ZombieFeynman

There are so many layers to the onion and so many pieces to put together, i just think you will get wrapped up in learning about the mundane as well as the cosmic. Then you might discover that what you thought was mundane is teeming with beauty and wonder.

Dont worry about 20 years. Think about the cool stuff you can learn in twenty days.

nabil0

It's certainly possible as I'm learning physics on my own.I've started learning physics two years ago when I had absolutely no physics or math background(except high school math/physics),now I'm working through a textbook on Quantum field theory . Good luck

SolsticeFire

Whoa! That's quite the accomplishment. Congratulations! Which textbook are you using? How did you cover all the mathematical machinery and underlying physics required to tackle QFT in mere two years?!

SolsticeFire

SolsticeFire

To OP:

Although I'm not that good at physics yet(I'm a mathematician!), one thing that helped me tackle advanced mathematics and gauge my skills at particular area of interest (in your case this might be string theory, QFT etc) was to attempt to read articles from arxiv pertaining to my topic of interest and then when I couldn't comprehend things, go back and read/solve problems in that particular area. This might help you. I have gotten stuck on some things for weeks, but I've always managed to dig myself out of the abyss by backtracking and trudging on! :D

Good Luck on your wonderful endeavor!

SolsticeFire

PS: When you get to advanced topics, your book selection skills should be impeccable. Good books on advanced topics are far too few and too far in between. So choose wisely!

ZombieFeynman

If this is true, it is an extraordinary accomplishment.

nabil0

There's not too much physics and math you need for QFT. It's just elementary calculus ,differential equations and linear algebra. That's all the mathematics you need before tackling any physics material . You don't need a whole course on classical mechanics before studying electrodynamics .And if you understand mathematics very well, You can even read Jackson and Griffiths concurrently . I was able to read Weinberg QFT (though not fully understand)in the first year I've have started physics .You need matrix algebra ,tensors and group theory (Which will be easy if you know linear algebra and calculus), special relativity(including relativistic classical fields which I learned from Barut and landau ) and elementary Quantum mechanics (At the level of cohen).I didn't need to understand the whole graduate level stuff in Jackson and goldstein before understanding what's QFT is all about.The point is : you don't need to wait until grad school to learn more advanced areas .The most important skill you need to understand all this is imagination.For example ,With just chain rule and taylor expansions you can solve a lot of problems in advanced physics .

redoopi

Why do you think so? I really can't see the point of doing exercises without being able to verify my answers. But for the proof-type exercises, I can try to solve them and ask for help online as you have suggested. For the advanced topics, I suspect that it is not very likely to get answers within a reasonable timeframe, but I guess this is the best I can get anyway.

redoopi

Thank you very much, SolsticeFire. Have you ever encountered "knowledge gaps" between textbooks and research papers, because the textbooks available are not advanced or up-to-date enough? Is backtracking among research papers via the references a good way to overcome this hurdle (for a self-learner)?

redoopi

You must be very talented to learn so quickly! Do you solve most of the problems in the textbooks, or just a selected few? Have you ever encountered difficulties such as those that I mentioned in the first post? If you have, would you mind sharing how you overcome them?

redoopi

Did you study classical mechanics right after calculus/ODE/LA, without reading a calculus-based general physics textbook like Freedman or Halliday (as most universities require)? Is thermodynamics and statistical mechanics necessary? Do we need to study particle physics before learning QFT? For E&M, is the level of Griffiths enough before learning GR?
(sorry for asking you so many questions and thanks for your generous sharing!)