Learning Physics

Rhett Allain, physics prof, wired writer, and blogger has an article entitled: What do you need to learn upper level physics? YMMV, but I found it spectacularly uninformative, so I decided to see if I could do better.

Firstly, compared to some other academic disciplines and professions, physics is heavy on technique but relatively light on memorizing facts. It also very "vertical" or hierarchical, in that it's built layer after layer one on top of the other, and you often need the lower layers to understand the upper layers. And by "techniques", I mostly mean skill at solving physics problems.

The foundation is mathematics. You can't really begin to study physics at a formal level until you are competent in high school level mathematics: algebra, geometry, trigonometry, and calculus. At that point the student can tackle the basic subjects of physics at an elementary level: mechanics, electromagnetism, thermal physics, and waves. Simultaneously one should be learning some more foundational math: multivariable and vector calculus, differential equations, and linear algebra.

The foundational subjects of classical mechanics, and electromagnetism are usually tackled again in the undergraduate curriculum, at a higher level of sophistication and quantum mechanics added once students know a bit about partial differential equations - or sometimes as an introduction to that subject. Special relativity may be taught at either this level, the previous level, or both.

In many US schools, the finishing layers of these subjects are saved for graduate school, as are such more advanced subjects such as quantum field theory and general relativity, but more ambitious curricula give them to undergrads. Undergrads are also usually introduced to applications such condensed matter and nuclear and particle physics.

Prof Allain is pretty vague on how students go about learning these subjects, alluding to textbooks, lectures, MOOCs and online demos, but he leaves out what is to me the core: working problems. You learn physics by working physics problems. If there is another way, a royal road, I don't know it.