Do you leave out the possibility that the same reconciliation of the two sets of equations might be achieved by some other, more simple and elegant means without invoking any redundant entities?
If they're necessary to explain the phenomena they're not redundant.
But if you think you can do better, feel free to try.
While you're at it, it's always bothered me that Newtonian mechanics involves all this pesky calculus. Can you come up with some "more simple and elegant" method involving counting on my fingers? Clearly the fact that the math is difficult is a weakness of the theory.