I understand that there are in languages three ways of expressing
grammatical case: preppositional (like modern English),
postpositional (Japanese and some others) and flexion, in which the
endings of words and sometimes the internal vowels of words (mostly
nouns) are changed to reflect their function in a sentence.

My question is: is postpositional the granddaddy of flexion? Could it
have been that in "proto" languages, the case ending was a discrete
word, while later it blurred into a flexional ending?