Start with the class Mammalia. Pick two orders within that class, say rodents and primates. Do all the families within one order (e.g. the mouse family and the mole rat family in the rodent order) show the same degree of difference as all the families within the other order (e.g. the great ape family and and the lesser ape, i.e. gibbon family in the primate order)? Would that imply an absolute classification system?
Sorry for my slow reply - been very busy in work. Others have mostly answered this, but a quick couple of thoughts.
If you look at the ranks assigned to different groups, it's clear that they're not consistent. There are amphibian genera that have been around since the Palaeocene, about 60 million years ago, while many mammal genera are less than 10 million years old -
Gorilla,
Pongo and
Homo probably all diverged from each other within the last ten million years.
Now, it's possible of course that this is because these groups haven't undergone evolutionary change at the same rates. Maybe the great apes have had a burst of sudden diversity, while mudpuppies haven't changed much for the last 60 million years. If this were true, though, it creates a clear problem for the idea of objective ranks, though, if we're trying to fit it into a framework that shows relationships.
Imagine we have one genus, with four species: A, B, C and D. A and B are sister species; C is the sister to the clade A+B, and D is sister to the claide C+(A+B). Over a few million years, B experiences intensive selection pressure and undergoes rapid morphological change, while the remaining three species are relatively static.
At the end of this process, B is now sufficiently divergent from A to justify its own genus, in order to be consistent with how we classify genus. But, if classifications are supposed to represent relationships, C and D must also be given different genera to A, otherwise the genus becomes paraphyletic with respect to B. So, we wind up with four monspecific genera, three of which (A, C and D) are all more similar to each other than the objective standard of difference we've agreed upon to define the idea of 'genus'.