The Analytic/Synthetic Distinction: Quine's Critique

This video outlines Quine's arguments against the analytic/synthetic distinction.

0:00 - The analytic/synthetic distinction
8:18 - Quine
9:30 - Containment
11:05 - Contradiction
14:58 - Synonymy
24:44 - Interchangeability
35:45 - Semantic rules
44:35 Against sentence meaning

[GottfriedLeibnizYT]

Wow! I've been waiting so long for this!
Please do Quine's "Epistemology Naturalized" also if you feel like it.

[HerrEinzige]

It is analytically and synthetically true that I have liked and commented on this video.

[EdgarQer]

How strange it is that today I decided to read "Two Dogmas of Empiricism" and understand Quine's criticism, and then this video comes out

[captainstrangiato]

Just got to Kant in my Modern Phil class. Great timing!

[jmike]

Yesss been waiting for this for some time. Thanks Kane you rock 🔴

[martinbennett]

Your reference to Andrew Wiles and Fermat's last theorem interested me because I have used it myself in my own teaching. I was expecting you to follow up on this.

To me the point is that there there is nothing intrinsic to the idea that Aˣ = Bˣ + Cˣ cannot be satisfied where x > 2. However for a long time this was accepted as empirically true. Then (after a false start) Andrew Wiles makes the claim that he has a mathematical/ logical proof, subsequently his proof is accepted by other leading mathematicians. At this point something has changed: although I (along with most others) cannot understand the proof, I can understand that at for those, at the very least, who do understand the proof any further empirical attempts to investigate possible values for x are utterly futile in a way that was not so, before Wiles' work.

Whatever Quine has to say about distinctions between analytic and synthetic, we need a concept to distinguish between the two states of understanding. Perhaps we could contend that propositions are either synthetic a priori or synthetic a posteriori and that analytic is only applicable within strictly defined local situations. Even there I see a problem because I would want to distinguish between statements that triangles have three straight sides (usually accepted as analytic a priori) and that the interior angles of a triangle add up to a straight line (arguably synthetic a priori).

[rv]

I actually read Quine's "Two dogmas" paper. I found it utterly unconvincing.

[JohhnyBoyNu]

This is so awesome lol i literally have an exam on this on thursday

[StatelessLiberty]

There’s an interesting attack on the analytic synthetic distinction in Wittgenstein’s On Certainty, and I find this attack more convincing than Quine’s. Basically Wittgenstein’s point is that Moore’s statement “I have two hands” has the form of an empirical statement and isn’t tautological yet is at the same time empirically unverifiable (because outside of special circumstances it’s inconceivable what it would mean to test it).

[drewhallett]

I remember finding the exact spot in Word and Object where he makes the argument, and I was like "wait, that's it?"

[mobili]

Bro uploaded this when I've almost finished SEP entry on the subject matter 😭

[InventiveHarvest]

Circular arguments are good as long as the circle is progressing. The concept of definitions has been progressing to new knowledge for thousands of years. I think Quine gets confused because he thinks apriori statements must be necessarily true or false, but we can always find counterexamples to conjectures, even the proven conjectures.

[MoleyMoleo]

Hold up, at around 44:34, you note that Quine complains that there is this circle of terms that refer to each other, and hence, through their circularity, we cannot confidently approach knowing if they are real concepts.
Maybe I'm misreading, but that very finding sounds like an analytic argument to me!
I'm genuinely unsure is a point for or against Quine's argument though: maybe the fact that he used analytic reasoning implies it is a possible form of reasining. But maybe the fact that by attempting analyticity he reaches a potential contradiction is an indictment on analyticity. I'm not sure which way this should cut.

[soyoltoi]

Is it really the case that Fermat's last theorem is true over "all worlds" when it's not even clear if it's provable in Peano arithmetic?

[yoramgt]

It seems to me that the argument is that synonymy is a synthetic or empirical property of natural language. It is inferred rather than have it officially defined.

There was an example given of a hypothetical situation in which it was discovered that foxes are asexual and vixen are a certain variety of fox. Such discovery would be impossible if vixen was in fact defined as a female fox.

[hodgeyhodge]

By analogy with the two classes of contradiction on the slide at 12:00 do you also distinguish two separate classes of tautology, one of the form "X is a bachelor and it is the case that X is a bachelor" and a broader one of the form "X is an unmarried bachelor"?

[BumbleTheBard]

To some extent, it may be that Quine and Carnap are talking past each other. Quine's criticism is an attack on *a priori* knowledge, based on his holistic approach to epistemology. Analyticity gets caught up in the criticism because if there are *a priori* propositions then analytic ones are, and the logical positivists denied the existence of the synthetic *a priori*. Defenders of analyticity focus on what makes a proposition true; Quine is concerned with whether a proposition is exposed to revision for empirical reasons. For Quine, analytic sentences, and even definitions, do not have a privileged epistemological status that makes them immune to revision. So there is no dichotomy among propositions, just a spectrum ranging from more or less exposed to revision. For myself, I think that is correct. There are ways to define 'analytic' but it has no explanatory value in philosophy.

[tunahankaratay]

I personally hate arguments that try to exploit the ambiguity of the natural language so much that I lose interest in reading the rest of the text. Congratulations, you are the smartest kid in the class! Want a smiley sticker?

[jamesoneill]

Can you elaborate on what poetic quality is?

[danwylie-sears]

Of course you can make an error in dealing with something that's stipulated. If I'm typing with my left hand out of alignment, and some of my instances of "ff" appear as "gg", that's an error. If I actually state a stipulation that "ff" is to be an abbreviation for "female fox", and you hear me say "ephemeral fox" instead, that's an error. If I make a bunch of abbreviations, and you have to think a bit when you encounter one of them, it's still a stipulated definition. Why should that be any different when the stipulation was done collectively over decades, instead of individually within one lecture?

--

Sydney can't believe that there is a vixen in his garden without believing that there is a female fox in his garden. Believing that there is a vixen in his garden isn't the same as believing that there is something called "a vixen" in his garden. One statement describes only his belief about what's in his garden, and the other describes his belief about the language.