ฝัง
- เผยแพร่เมื่อ 8 ก.ค. 2024
- For his entire philosophical career, Quine was a confirmed defender of extensionality. In this video, I explain what this means and why it lead to skepticism about modal logic.
At around 4:20, I suggest picking up an intro to phil of language for more info on the extensions of sentences. A good intro is Morris's "An Introduction to the Philosophy of Language"; chapter 2 deals with Frege's account. Morris's book also provides, in chapter 6, an overview of the issues we'll be discussing in this series.
Regarding my comment at 13:18 that violations of extensionality are okay in the case of beliefs: in fact, Quine wasn't happy with these, either. Some of Quine's work on propositional attitudes (propositional attitudes are a kind of mental state, which we express using a verb followed by a "that": "Frank believes that...", "Frank hopes that...", "Frank desires that..." and so on) is very similar to his work on modality. See, for example, his paper "Quantifiers and Propositional Attitudes". An accessible introduction to his views on propositional attitudes can be found in the Morris book, chapter 7.
(More generally, Quine's approach to mental entities is often considered an early precursor to the radical eliminativisms of Dennett, Stich, the Churchlands, et al.)
The point is just that violation of extensionality seems to be more acceptable for beliefs, which we all expect to be messy, than for a completed, naturalistic theory of the world, which we expect to be "pure" and well-defined and so on.
You will be a good Prof. one day.
This is a very nice video series, thank you for doing it. While extensionality is much cleaner, and works perfectly when applying logic to mathematics, I think intensional concepts are necessary if we're going to explore how human beings actually think and reason about the world.
There's no doubt that Quine would've rejected most talk about the ontological status of possible worlds... unfortunately, though, he didn't actually say much about it. He seemed to fall silent on ML in his later life.
For something along the lines of a Quinean critique of various approaches to possible worlds, check out Lycan's article "The Trouble With Possible Worlds" in Loux "The Possible and the Actual".
Thank you, that was very informative and clearly explained. I learned a lot 😃
I always thought that the better naming for this would have been 'extrinsic and intrinsic view'
I never knew these terms. Very interesting to learn them in this context.
It seems to me Kripke's possible world semantics makes Modal Logic extensional. Two modal formulas mean the same thing (are equivalent) if they refer to (or can access) the same set of possible worlds.
Although sympathetic with Quine's doubts about modal logic, absent a thoroughgoing modal realism a la' Lewis, I am not sure where the intensional/extensional line can be drawn. It would seem that linguistic items are poor candidates for designators. Even a name, John Smith, will be ambiguous as to which object it picks out. Even natural kinds would appear to have more than one name. For tigers, you have the everyday English usage and the latin scientific classification nomenclature, at the very least. In the case of John Smith, you might say although he has that name, it does not suffice to differentiate him from others, and in the case of natural kinds, you seem to have what amounts to just another synonymy problem. This problem would not appear in a purely artificial closed system of posited objects and conventional unique names. But if you want to deal in propositions about the world out there, I think there is a problem for this intensional/extensional distinction. Names must apparently be presumed to consist of something that is not linguistic but even more abstract.
Loved the FZ reference :D
Sorry for using this chanel to ask for questions, but i figure that you could help me with my work. Im studying the problem of induction and i was thinking that the same objections made against modal logic could be used to present objections to probability logic. Both use a type of intensional operator, as "it is probable that...". I dont know if i can use this equivalence actually, so thats my first question. Second question is, if im right, so i figure that the problem with probabilistic logic and modal logic is that they fail altogheter in presenting a verofunction interpretation of the concept of implication, then violating the concept of validity. But then i would like to know why D.C. Stove thinks that modal operators, instead of violating the concept of validity, gives a better account of it - but a non-formal account. I dont know, really, how can anyone thinks the concept of validity could benefit from being non-formal. The last question is about the fact that Stove thinks extensionalists are commited to the very implausible thesis that an implication is valid due to a "contingent fact of the world", proved by its instances. Why do i feel its exact the opposite? I.e, that intensionalists diminish the concept of necessity to a contingent condition like "truth in a set of possible words", therefore, they relie in a "metaphisycal" or "rationalistic" fact, something like a intelectual intuition. I would like to know what i am missing, because it seems to me that Quine would also think that intensionalists are the ones relying in contingent presupostions, and not extensionalists. Sorry if i my doubts are not clear enought. Im not a native english speaker also, but i would be glad if you could recomend some literature about this issue.
What does Quine say about the ontological status of possible worlds? Would he have followed Lewis?
of course not
Computer Scientists have no difficulty in defining intension. A computer program describes an intention whereas the input/output relation of that program when it is executed defines its extension. More precisely, the intention of a program is precisely definable as the sequence of states of the machine when the program is run on a given input.
If you accept this definition then it's not intensions that are hard to define, but rather extensions. This happens because we can run the program on the machine and find the state at any point in the program's execution. But we cannot in general know the input/output relation of a program due to the halting problem. (If we could know the input/output relation then we could solve the halting problem. A contradiction).
What is the haulting problem?
@@MathCuriousity The halting problem is basically asking the question: "Given an arbitrary computer program P can you show that P halts on input I" (halt here means that the program stops after a finite amount of time as opposed to running forever).
It's a famous result that this problem is undecidable, which means that there does not exist a general algorithm that can solve the problem. The proof essentially goes like this:
Suppose you had a program that computes the halting problem H. That is, H can determine when an arbitrary program halts or not.
Define a program O (for opposite), that takes basically takes some program P with input I and runs the halting problem over it, so it runs H(P,I). If P halts on input I, then just get O to run forever, and if P does not halt, get O to halt.
Essentially, O does the opposite of whatever H spits out.
The contradiction occurs when you run H and O as input to O. Basically what happens is if the halting problem says O will halt, O runs forever, and when the halting problem says O runs forever, O halts. Effectively doing the opposite of whatever H says. Hence H cannot determine if O halts or not, which contradicts the assumption that H could determine that any program halts. Hence H cannot exist.
Exxxcellent.....
The word s and the word f are synonymes if and only if s and f refer to one thought and only one thought.
Thought is a mental construction,thus it can be false.
And also the mental construction can be something we get from experience or something we get by thinking alone.
And the mental construction can be true.
Reference is a relation relating some thing t to some thing m.
Reference doesn't imply the truth of the thing being refered to.
Kane B 4 president
See Putnam's Theorem, which shows it's possible to construct a situation in which two statements with obviously different meaning can have extensions that are the same in all possible worlds.
Frankly, the while Russell/Ferge/Quine/Carnap/Kripke schtick seems to be barking up the wrong gum tree. As Lawrence Olivier might have said to Dustin Hoffman in an alternate universe, why don't you try idealized cognitive models, dear boy. It's so much easier.