> R1551 reads as if it is trying to avoid amending the past, by amending > the present gamestate with reference to a hypothetical past. I have > tried to think of a couple of reasons, but neither feels particularly > compelling in the face of your arguments in (7):
I'm guessing R1551's complex language about "what it would be if, at the time..." is more about making sure it's clear how the consequences play out. If *only* the record-of-the-past part of the gamestate were changed, maybe the consequences would not be changed, e.g. D. Margaux would still have blots, even though now e was never fined. > - Pragmatism. It is impossible to amend the past, so why pretend > otherwise via legal fiction? > - It is simpler and cleaner to amend the gamestate at a single point > in time (the present) than amend all times t, P<=t<=T, where P is the > publication of the ratified document and T is the time of > ratification. I think these are strong reasons. The past-not-included interpretation feels nicer and more obvious to me, and I tried to capture some of that feeling in Section 6. But I currently feel that the other arguments outweigh this, no matter how indignant I feel about my prior intuition being wrong.