Quine's objections to modal logic 2 - use and mention

Quine once said that modal logic was "conceived in sin". In this video, I explain exactly what that sin was.
Note: this is more a criticism of Lewis's motivation for developing modal logic, than of modal logic in general. You can skip this video if you want to go straight to the really substantial objections. (However, later videos will probably be easier to follow if you've seen this one first.)

[@latinomodesto8813]

we need more videos like this, thanks

[@eyezuel5307]

I hate to leave vapid comments but I love these videos and the dry humor just like Shaun, and the clear way you present logic

[@am101171]

Love your tutorials, hope you keep it up. thanks.

[@mistymouse6840]

That was very interesting, thank you.

[@metanosis]

Excellent!!! Thank you....

[@ffhashimi]

Very important ;
thank you

[@noninvasive_rectal_probe8990]

Wow, that looks exactly like multi-context Lamba calculus with linearity
www.fos.kuis.kyoto-u.ac.jp/~igarashi/papers/pdf/lambdaCB-PPDP06.pdf
And also Edward Zalta's Abstract Object Theory where two kinds of things exist: those which encode and those which examplify

[@christiandarmanin6192]

Well done !

[@svenkalmar]

very cool..thanx

[@newname2600]

One can argue about implication. One cannot argue about the meaning of the conditional. If you think you are arguing about the meaning of the conditional, you aren't: you're arguing about implication.

[@cliffordhodge1449]

Quine's objection seems quite valid, but none of this changes the fact that "If the moon is made of cheese then FZ was a musician," appears to be nothing but a comical non-sequitur. As relates to the seeking of knowledge or understanding, it seems to be of no more help than, "If 7 minus 3, then FZ was a musician," or "If nothingness, then FZ was a musician." In any of these instances, the antecedent and consequent elements could hardly have less to do with one another. In what way would the proposition as a whole be said to constitute a good representation of thought process, an executable mental algorithm? If you take a name (proper name like 'Frank' or a name like 'if the moon is made of cheese') to be simply a pointer to its referent, then it is not clear that the mention/use distinction actually cuts any ice here. In one case you have a single pointer pointing at an object; in the other you have a pointer pointing at another pointer which points at an object. The problem seems to be that you can create a chain of pointers, but if the last pointer does not pick out an object, you have nothing. You may say the object language itself has only syntactic - not semantic - function, but then it seems to be a system purely conventional, wholly artificial, in the sense that it is not meant to represent objects which have logical priority to the system which refers to them.