What are most amazing discoveries in the 20th century?

Does physics research of the mid-late 20th century require significantly more analytical intelligence than physics research of the early 20th century?

• I can't even be sure if many of the "great physicists" of the early 20th century could completely understand quantum field theory in the way that later physicists in the 1950s understand it. Much of the physics of the early 20th century doesn't require too much analytical talent to understand (and to make hypotheses about). Obviously, we shouldn't discount the their amazing achievements, but when we decide to choose what field to go into, it is important for us to get a sense of what's possible for us (and our possible limitations) That being said, the computational techniques of the early 21st century does seem to lower the barrier once again, as you don't need to be a total genius to do computation (and even undergrads can do it, as they often do in astro). Also, computers do make a lot of things easier. Sure, people have much more to learn now, but at least for some computational physics research, they don't have to learn quantum field theory.

• Answer:

  Some of the early twentieth century physicists were alive in the 1950s to make major contributions to quantum field theory, Fermi and Pauli most signifcantly. The younger folks, like Jordan, Heisenberg, Bethe, Landau understood it well, and Bethe was significant in promoting the work of Feynman and later Kenneth Wilson. Einstein didn't study quantum field theory, because he was interested in figuring out the way quantum mechanics could be a statistical description of something else, something we still can't do, and perhaps it's impossible. But Einstein proposed second quantization to Schrodinger in 1924, before modern quantum fields were formulated, and it is clear that both he and Bohr understood field quantization as analogous to the quantization of mechanical systems. Bohr also contributed to quantum field theory, at the foundations, by proposing field quantization in a famous paper with Rosenfeld, when he accepted photons, and later more indirectly, by suggesting to Casimir to calculate his Van-der-Waals force as a vacuum energy change. Bohr was doing nuclear physics in the later years. The reason you think it's easier to do the early 20th century stuff is just because it made it into books, and the methods simplified considerably with the passage of time. Quantum field theory today is about as easy to learn as quantum mechanics was in the 1950s, and string theory today is as difficult as quantum field theory was in the 1950s. As material makes it into books, hard things are removed, and intuitions sacrificed, for the sake of easy presentation. All of these topics are easy in a certain sense, the only sense that matters, because these things are already known. The only really difficult thing is coming up with something totally new. In terms of difficulty of discovery, there is nothing to compare with General Relativity and Quantum Mechanics. To get a sense of how difficult it was to discover these things, consider the experimental situation in 1910, and try just to come up with the Bohr model by yourself, without using anything you already know. Once you try it (you will almost surely fail), you can then read the original paper to see how Bohr did it, and how difficult and non-rigorous (but correct) the reasoning is. Without the Bohr-Sommerfeld quantization, you can't do anything else. Now suppose you know Bohr Sommerfeld quantization. Try to come up with modern quantum mechanics. You will surely fail. If you read the DeBroglie/Einstein/Schrodinger papers and the Heisenberg/Jordan/Born papers, you will see how difficult it really was. In comparison, the quantum field theory work of the 1950s is not so impressive. Although, the path integral is similarly revolutionary, the basic revolutions were already done. The comparable radical shift in physics from this point is S-matrix theory and string theory. The thing that makes quantum field theory difficult to learn is just a political thing: the path-integral was hidden away in the middle decades of the 20th century, because all the founders hated it. Pauli thought it was not interesting compared to Schwinger's stuff, Bohr didn't get it, Dirac thought it was useless because it couldn't be generalized to Hamiltonian with non-quadratic momenta, Heisenberg didn't bother with it, and so on. The only people who really got it were Hans Bethe, and the young people, like Dyson, Schwinger, Kraichnan. It took decades for physicists to get confortable with the method, and it is not easy to make rigorous still today. Without the path integral, you can still do quantum field theory, but it's a pain in the neck.