On Dec 29, 2013, at 3:31 PM, Kerim Aydin wrote: > On Sun, 29 Dec 2013, Tanner Swett wrote: >> I might say something like this: "The behavior of the game is specified by >> the >> rules themselves, not by the logical consequences of the rules. In >> particular, >> if the rules logically entail that an event does or does not take place, >> this >> does not necessarily cause the event to take place or to fail to take place." > > Hmm, in your first proposed sentence, the issue with that wording > is that all of the game is the result of the (common-sense) logical > consequences of the rules, not the rules themselves. The rules don't > specify that "G. is a player". They specify "If someone is a person, > and has registered, then [the logical consequence is that] they are a > player" and we track the instances of that line of reasoning. At least > that's my reading.
Yeah, I guess that "specified by the rules themselves" doesn't really mean what I want it to mean. But I don't think we want the game to be governed by logical consequences, either. Something *like* logical consequences, yes, but something else. I think most of the statements in the rules (other than definitions) can be interpreted as statements of the form "whenever X happens/is true, Y happens". The sentence "The Promotor CAN distribute a proposal which is in the Proposal Pool at any time" means "whenever a proposal is in the Pool and the Promotor announces that e distributes it, e distributes it". The (hypothetical and admittedly contrived) sentence "If a person CAN assume Promotor, then e CANNOT assume Assessor" means "whenever a person CAN assume Promotor, every instruction that e assume Assessor is void", which is clearly different from the other way around. So I might say that the game's behavior, instead of being determined by some unspecified non-classical logic, is determined by some unspecified operational semantics. —Aru
