What surprised me was the realisation that since Russell, nobody in mathematical logic ever claims anything except that his formal logic is logically consistent. That is, they don't claim that their theories are anything beyond consistent theories, which they are. That is, they don't claim that their theories are "correct", i.e. that they would be the correct model of human logic. So, in effect, they're not even wrong and they won't ever be. Pointing out that their theories don't mean anything is just a waste of time.PeteOlcott wrote: ↑Wed May 08, 2019 5:39 pmMathematical logic is currently considered to be the ultimate foundation of the notion of truth. Because of Tarski and Gödel truth conditional semantics is "known" with certainty to be either incomplete or inconsistent thus putting a huge damper on the possibility of any meaningful success of any AI projects. Through the understanding that all analytical truth is simply the tautological connections between concepts that are validated through deduction the broken system of mathematical logic is corrected.Speakpigeon wrote: ↑Wed May 08, 2019 2:09 pm I broadly agree with that but I would need to look at the detail of both Tarski 's and Gödel's proofs. So, pending that, I have to suspend my judgement.
Of course, on the bright side, Tarski 's and Gödel's proofs only apply to standard mathematical logic. So, assuming your system is correct, you would need to identify some conjecture you could prove with your system that standard mathematical logic could not due to Tarski 's and Gödel's proofs.
Yeah, I know, sounds like a lot of work...
The real shame is that none of them could come up with anything like a proper method of logic. They keep inventing ever more idiotic theories that don't represent any real logic. Why not try to model proper logic for once?
To date, the only methods of logic that are correct are Aristotle's syllogistic and the Stoics' logic (or Abelard's). Complete the picture with a nice formalisation of these two methods using the form used in mathematical logic and you have the best which has ever been done in the field. And this after 2,500 years of the history of formal logic! Whoa. Really impressive.
Still, perhaps someone found something and nobody noticed... I wouldn't be surprised.
EB