Kerim Aydin, Wed, 12 Jun 2013 07:37:53 -0700 :
Some history:
From 2002 (when I started) to 2005 no one thought about paradoxes at all in
this sense. Paradoxical CFJ statements were simply DISMISSED as meaningless.
I think the aforementioned lawyer had a hand in creating this system (before
my time). R2358 didn't exist.
In 2005, we were playing "cards". As a defense card, one card had retroactive
application (it could cancel any recent play). It was used to cancel out the
play that led to it being obtained to play. It was done because it could be,
and hadn't been done, and because the original Nomic rules said that creating
a paradox ended the game. We didn't know what to do: until a proposal papered
over the problem, I (the recordkeepor for cards) tracked to separates states
of the game (where the card had been played and where it hadn't). Someone
suggested we just had to start Agora II.
Players thought it was cool enough (in that particular instance, where it was
a really clear retroactive application) to add it as a win condition. That led
to lots lots more "purely verbal" attempts to win this way (e.g. CFJing on
"this statement is false" etc.) and the rule was constantly tweaked.
[...]
The current case isn't really the same sort of "win attempt", IMO. I think
it's just re-discovering (or re-interpreting according to current Agoran
rules and play) the original legal conundrum that led Suber to invent Nomic.
Probably worth doing that once in an Agoran generation or so!
That's pretty neat to get a historical perspective like that, thanks,
G., for posting this.
> If R2358 were repealed it would probably cut back on such things
> drastically
Well, yeah, based on what you're saying, it seems it took years for a
paradox to come up that actually bothered anyone.
For that matter, is the card paradox still compelling? I had a look at
the current ruleset and I'd guess that nowadays the card paradox would
be resolved by R1030 ("In a conflict between rules...") or R2240 ("In a
conflict between clauses of the same rule...")
> but not eliminate the possibility of logical paradoxes cropping up
> (impossible to do that in a self-referential algorithmic system).
Impossible to do that? I dunno. Of course, the entire ruleset can be
replaced. If no rule is permanent, then you can't eliminate any
possibility. And if any rule did become permanent, then you could say
it's not Nomic anymore. But aside from that, I'd bet that you could
eliminate paradoxes.
-Dan
