Appeal to clarificationlynn_pJanuary 18 2007, 01:18:19 UTC
Perhaps I sould have stated that there exists a breakdown in classical logic. My frame of reference is admittedly bound in another touchy term...reality, and my assertion is that classical logic doesn't always lead to truth (from a Materialst standpoint), that classical logic can be inconsistent with reality. Classical logic is sometimes simply inadequate to solve proofs from premises that are logically rational and consistent yet present anomalies in behavior, especially in the face of quantum logic. In this instance, classical logic doesn't prove to be consistent with the truth of reality; sometimes classical logic can fold in on itself and get caught in a cycle of proving nothing except its own lack of usefulness in a certain situation. What I'm asserting is nothing new. The mathematician John von Neumann disproved the efficacy of distributive law to describe reality, which, from molecular biology to economics, has not a single perfect point of balance, but has several positions of equilibrium. Granted, what truth boils down to is a personal philospohical query, a concept to which no two people might ever agree. The major conclusion (among many) at which I arrived is that classical logic may tell us how we (should) think, but it is woefully inadequate to describe the what and why of all experience. A nonlinear dynamic system requires something more suitable to describe its complexity, and I don't believe that classical logic can cut it.
Re: Appeal to clarificationgalbinus_caeliJanuary 18 2007, 02:21:29 UTC
I can disagree with nothing that you are saying. The problem comes when people try to apply classic logic in situations where it does not apply. The most obvious one is human emotion. Logic cannot reliably predict how people are going to react emotionally to a paticular situation, no matter how carefully the circumstances are controlled. What we call "reality" is another. If we are talking about a finitely bounded subset of reality ("here", "now" and "at human scale") classic logic works perfectly. but as you expand that set to the edges of the universe, or the ends of time, or the infinitly small, all the rules change.
Classical logic is sometimes simply inadequate to solve proofs from premises that are logically rational and consistent yet present anomalies in behavior, especially in the face of quantum logic. In this instance, classical logic doesn't prove to be consistent with the truth of reality; sometimes classical logic can fold in on itself and get caught in a cycle of proving nothing except its own lack of usefulness in a certain situation.
What I'm asserting is nothing new. The mathematician John von Neumann disproved the efficacy of distributive law to describe reality, which, from molecular biology to economics, has not a single perfect point of balance, but has several positions of equilibrium.
Granted, what truth boils down to is a personal philospohical query, a concept to which no two people might ever agree. The major conclusion (among many) at which I arrived is that classical logic may tell us how we (should) think, but it is woefully inadequate to describe the what and why of all experience. A nonlinear dynamic system requires something more suitable to describe its complexity, and I don't believe that classical logic can cut it.
Reply
Reply
Leave a comment