Re: DIS: Re: BUS: A little briefer than hoped
Kerim Aydin <kerim <at> u.washington.edu>
2014-10-23 21:15:13 GMT
On Thu, 23 Oct 2014, Alex Smith wrote:
> On Thu, 2014-10-23 at 09:46 -0700, Kerim Aydin wrote:
> > I think about judgements like House Rules.
> > For any boardgame, when the Rules are unclear, you have to figure out how
> > to play. So everyone argues, and either you come to some consensus, or some
> > way of deciding ("the owner of the game decides" or whatever). Nomic just
> > codifies that part (and in doing so, makes it self-referential, but that's
> > not an issue here).
> > So the point of a House Rule is (1) it lets you get on with playing, and
> > (2) when the same thing comes up again, the House Rule doesn't change
> > (wouldn't be fair to stop one person from doing something, then allow the
> > next person). That's the role of precedent. Although precedent always
> > loses to rules... if a player finds an obscure clause that everyone missed
> > that answers the question, bye-bye house rule. (Or you may say "let's
> > finish this game the wrong way because that's how we've been playing, and
> > play the right way next time". Nothing wrong with that - that's more or
> > less what ratification does!)
> > But for precedent to work, you can't just say "this move doesn't work
> > because I own the game and I say so". You need to supply enough reasoning
> > so that players will accept that it makes sense and move on. Otherwise,
> > you don't know how the House Rule applies to other situations, and also
> > it makes people grumpy.
> Well, a CFJ is fundamentally different from a rule. Rules (and
> proposals) don't have any effect before they're enacted; they can change
> the gamestate to simulate the effect of a change having been made
> retroactively (e.g. when we ratify something), but they don't actually
> change what was going on in the past.
> A CFJ, meanwhile, doesn't change what was going on in the past, but it
> does clarify. The aim of the CFJ process is to determine /what actually
> happened/. Anything that politicises CFJs simply makes them less good at
> determining what happened, but it doesn't actually change the past.
> I mean, if it did change the past, it'd be pretty trivial to get a
> paradox out of it.
> Using the mechanism in Rule 2437, I cause rule 2437 to set the
> recommended judgement of every Brief in the current Moot to TRUE. (Such
> Moots are, unless you're using Eritivus' interpretation, entities
> pertaining to the PoA; and the recommended judgement is a property of
> There are a few possible interpretations of what that action just did.
> With my view on matters, what it does is to cause the CFJ to have to be
> decided TRUE, while its statement is actually FALSE; in other words, it
> simply causes us to have a gamestate that doesn't correspond to the
> judgement. And there's nothing wrong with that! Our current CFJ process
> is pretty bad at finding appropriate judgements in case of appeal.
> If we treat this with the "house rules" intepretation, though, what I've
> just done is (when the CFJ is resolved) to cause a situation in which I
> retroactively caused omd's dictatorship to work, which will
> retroactively cause my action to fail, which will retroactively cause
> omd's dictatorship to fail, etc. The resulting paradox means that
> resolving the Moot is illegal (because it causes the result of the Moot
> to become indeterminate). Sadly, I no longer get a win, because we
> repealed wins for paradoxes. (Unless people would want to give me a win
> by proposal?)
> I think I know which interpretation is more sensible.
No, the "House Rules" would not let us move on. Another CFJ would be
called. That happens when the result of a particular CFJ is corrupted.
If what you did is broken, it's because you've effectively pulled a
Lindrum: since everything "pertains" to that CFJ, them whomever
controls that rule controls every interpretation, including any
disputes over that interpretation. That's the problem with construing
"pertains" so broadly - I think I know for sure now that a narrow
interpretation is more sensible.