Who came up with the thought that cartridge loading affected tracking? That is rather silly.
A guy named Lenz observed and documented the underlying concept and I'd hesitate to call his work silly. Now extrapolating his work to cartridge loading is interesting. Moncrief showed decidedly different IMD results between the same cartridge lightly and heavily loaded. While I have not been able to replicate his results perfectly, I have seen several areas where loading effects the measured behavior of a MC cartridge.
Stepping back and looking at the big picture One has to ask "What causes IMD sidebands from the record groove needle interface?" The only answer I can come up with is mistracking. A number of tracking ability tests exist. Tests like the Audio Obstacle Course use recored music with increasing levels of one selected instrument and the user can judge where mistracking happens. This is the closest we can come to a real world example but it is important to understand the results are subjective which makes the results unique and hard to relate to others. The other approach as used on the Shure TTR103 and the CBS STR110-112 is more scientific and requires the use of two tones typically 400hz and 4kHz and an intermodulation distortion measurement. Here is the blurb from CBS:
The two things that jump out at me in the above are the use of the phrases "Instantaneous force" and "tracking distortion" both of which seem to be directly related to the topic at hand. The last bit about other system nonlinearities causing IMD is valid but outside the realm of the variable being tested here. (tracking force) If we agree that the load will as ralph states above:
cause the cantilever to be harder to move (stiffer)
It seems fair to follow up with the premise that this can also effect tracking ability. It is important to note that the extreme and clear results given by Moncrief do represent the two extremes where Rcart<<Rload and Rcart>>Rload which just happens to be the exact case of the ideal current mode vs. voltage mode input stage.
dave