I am teaching a course on Modern Physics this semester. When I was a student, this course and Chemistry 101 really bothered me. My perception of science was that it consisted of theories with broad domains of applicability, and there was little or no overlap or conflicts between different theories. Before the Modern Physics course, I had learned of Newton's mechanics, which was about the motion of bodies; I had learned about electromagnetism, which was about a completely different domain of electric and magnetic fields, with some input from Newton's laws to determine the motion of charges; and I learned about thermodynamics, which was all about heating of bodies and such. Each one dealt with separate phenomena that I encountered in the world around me, and I didn't see any conflict among them. Of course, I did touch upon some semi-classical or quantum theory, but it was so badly taught that it didn't really form a strong part of my world view.
Now, in these courses, I encountered a variety of semi-classical theories developed by the Physicists of the late nineteenth and early twentieth century were encountering to explain various phenomena that they encountered as they probed the natural world around them with their rapidly improving experimental technology. These phenomena included the constancy of light, the photoelectric effect, the emission of x-rays by bombarding metals with cathode rays (later found to be electrons), etc. To explain every effect, classical theory was modified in some way to match experimental results. Though, in our retrospective journey through this part of Physics history, the thrust towards the quantumness of various aspects of nature were obvious, I still found it deeply unsettling. The different explanation were often contradictory. For instance, for the photoelectric effect, Einstein postulated that light was not a classical electromagnetic wave, but rather composed of energy bundles, but for the use of x-rays to determine the structure of crystals, we assume that light was indeed a wave.
As a student, this irked me, because it challenged my pristine view of Physics. It wasn't till much later that I really made my peace with this. I had read this piece by Eliezer Yudkowsky in high school, but I didn't really absorb it until I had enough experience with doing Physics. Basically, if you want to create a reference for a region on Earth, you create a map. That map only has the information you want to include. One map can emphasize the road network in a region, while another can emphasize the vegetation. Different maps for different purposes. Moreover, back in the day (or even today), maps are often restricted to a certain region. There were different maps for different regions, and if there was overlap, they could disagree. Not to mention, maps often disagree with what's actually on the ground. The important thing to remember is that the map is not what's on the ground, but only a compact representation, meant to guide humans in their endeavor to understand the world around them.
Things were quite the same in Physics back in the day. Scientists had a wide variety of disparate phenomena to explain. And they just couldn't come up with a single map (or theory) to put them all on. So they created many different maps, each for a small number of phenomena. There was obviously disagreements between maps, but that's not a problem. A map of reality is not reality, merely a way for humans to understand the physical world and predict experimental results. If, we can't come up with a single map for all of reality, so be it. We still expect the reality underneath the maps to be be describable by a single map.
As things turned out, the Physicists of the 1920s did find a map that united a number of these disparate maps into one map, called quantum mechanics. This was hugely successful, till further experiments in the 1950s and later revealed black regions in quantum mechanics map. Most of these regions were filled by quantum field theory. But there are still parts where our map runs askew of reality. The goal is to create a new map where there are no black regions.
The lesson is that never confuse our maps of reality with reality itself. Maps are human aids; often useful, but never completely correct. Reality is nature.