Addressing the synonymity raised in Quine’s Two dogmas of empiricism

Philosophy of Language
1

Quine’s objection to the analytic/synthetic distinction seems to be focused on a lack of understanding of how words get their meaning. From Quine’s POV Bachelor (x) and ~Married(x) are synonymous and somehow define each other in a circular way.

When we understand that new words are created to be short-hand for an existing set of ideas then it is easier to see that the previously meaningless Bachelor(x) can be stipulated to mean:

Bachelor(x) := ¬Married(x) ∧ Male(x) ∧ Adult(x) ∧ Human(x)

Then Quine’s Two Dogmas of Empiricism objection to the analytic/synthetic distinction based on synonymity dissolves.

1 Like
2

That is a very concise and apt critique. I’ve always thought that a lot of the shortcomings of Analytic Philosophy could be summed up as a lack of philological acumen—as if words and concepts just descended from Heaven ready-made.

1 Like
3

Well that’s certainly one caricature of the dogma. Did you feel Quine had failed to address it?

Section 2, “Definition”?

4

the explicitly conventional introduction of novel notations for purposes of sheer abbreviation. Here the definiendum becomes synonymous with the definiens simply because it has been created expressly for the purpose of being synonymous with the definiens. Here we have a really transparent case of synonymy created by definition // From Quine’s II. DEFINITION

Quine certainly does not capture my reformation of analytic truth:
“true on the basis of meaning expressed in language”

When we translate all empirical truth into expressions of language that are stipulated to be true then we have merged the analytic and synthetic and have:

(a) A set of “atomic facts” of empirical and other knowledge.
(b) Expressions semantically entailed syntactically from (a) and/or (b).

5

A good topic.

Quine goes a bit further.

Here’s the article: Two Dogmas of Empiricism

What you have pointed to is discussed in the first and second sections. Have a look at the third and fourth sections, about substitution and holism.

6

I will do that. My focus has always been making
“true on the basis of meaning expressed in language”
reliably computable for the entire body of knowledge.

Thus the definition of terms is an exhaustively complete specification in a language such as CycL or Montague Grammar. This is stored in a knowledge ontology like Cyc or in a simple type hierarchy. Because each unique sense meaning is assigned its own GUID synonyms simply have the same GUID.

When we say that:

a statement is analytic when it is true by virtue of
meanings and independently of fact.

Besides being equivocal it may also be somewhat incoherent.
Even the Peano Axioms are themselves a kind of fact.

1 Like
7

The intuition in Two Dogmas is that there is something muddled with “by virtue of the very meanings of their words”. There’s a circularity in that necessity is explained by meaning, but then meaning is being founded in necessity.

Truth is not the sort of thing that can be grounded in meaning. It needs to stetch out past our words. Quine offered holism, Wittgenstein, language games - there are other options. But…

8

And we arrive back at grounding again…

Any reason why we should not accept infinite regress?

9

“true on the basis of meaning expressed in language”
It took me 27 years to come up with those words.

When we start with a comprehensive set of “atomic facts”
of general knowledge then the entire
body of general knowledge expressed in language.
can be derived by semantic entailment specified syntactically
from these atomic facts.

10

:grinning_face_with_smiling_eyes:

Cans of worms.

No, I suppose an at least denumerably infinite regress would technically be valid… no statement in the regression would be dependent on the next, and the truth of the whole be indeterminate.

I can’t quite see what to make of an indenumerably infinite chain of inference… best not to think about it.

11

You know that’s what Russell and the Early Wittgenstein tried?

So back to logical atomism, but with more sophisticated machinery.

I can see why Quine is threatening.

12

“logical atomism” I utterly reject its metaphysics.
I am proposing a 200 petabyte list of things such
as “cats are animals”.

It seems that specific concrete counter-examples are categorically impossible.
see if you can find one.

One key aspect of all of the is to utterly and completely reject “truth conditional semantics” and replace it with “proof theoretic semantics”. PTS sees meaning as terminating inference steps within a well-founded justification tree. Hence my expressions of language defined in terms of other expressions of language.

Then it seems to be more clear that “true on the basis of meaning expressed in language” is in its most basic essence is relations beteen finite strings.

13

Ok. Not seeing the difference.

Oh, I recall our previous chat. Hadn’t made the connection.

14

If you simply hypothesize an exhaustively complete finite list of every “atomic fact” of general knowledge formalized in a language such as CycL and stored in a knowledge ontology such as Cyc, then this by itself seems to mostly prove my point. If you cannot explain how it does not then it would seem that you do not understand.

15

Obvious point is: what counts as an atomic fact? I already pointed you towards Philosophical investigations in this regard.

Second point, this looks to be pretty much logical atomism with a different formalisation

Third point: my lack of understanding might be a result of incoherence in your explanation. More, when someone says “if you can’t see why I’m right, you don’t understand,” they’ve abandoned philosophical dialogue for something else. And this is why I walked away from our last chat.

I thought this was a thread about Quine. Seems it isn’t.

16

I still have to carefully study those two sections of Quine.

An “atomic fact” of general knowledge that cannot possibly be derived from other facts and is not about any specific situation. The only way that we know “cats” [are] “animals” is that it is stipulated to be true. Stipulated relations between finite strings is the foundation of their meanings.

It is probably the case that I did not explain things well enough that is why I need things that are unclear pointed out.

17

A curious fault in that pdf is omission of this revision towards the end of section 1, from

Otherwise there would be a state-description which assigned truth to ‘John is a bachelor’ and falsity to ‘John is married’, and consequently ‘All bachelors are married’ would turn out synthetic rather than analytic under the proposed criterion.

to

Otherwise there would be a state-description which assigned truth to ‘John is a bachelor’ and to ‘John is married’, and consequently ‘No bachelors are married’ would turn out synthetic rather than analytic under the proposed criterion.

The first one caused me some worry, a while back, so I was gratified to find that Quine had later on thought it improvable. Not that they aren’t equivalent, of course. But surprising that the comparison pdf doesn’t include the change.

I say “of course” but I admit that I was confused, about either version. Googling, the only direct discussions I could find were talking about the first version. I claim that I get it, now, haha. This interesting article helped btw: Elementary Propositions and Independence, Bell and Demopoulos 1996

18

IV. SEMANTICAL RULES
Analyticity at first seemed most naturally definable by appeal to a realm of meanings.

The notion of analyticity about which we are worrying is a purported relation between statements and languages:

to begin with, an artificial language L0 whose semantical rules have the form
explicitly of a specification, by recursion or otherwise,

tell us that such and such statements, and only those, are the analytic statements of L0. Now here

the difficulty is simply that the rules contain the word ‘analytic,’ which we do not understand!

analytic statements are supposed to be true.

semantical rule … stipulates, recursively or otherwise, a
… statements which … are to count as true.

a statement is analytic if it is (not merely
true but) true according to the semantical rule.

we are now appealing to an unexplained phrase ‘semantical rule.’

It may be instructive to compare the notion
of semantical rule with that of postulate. Relative
to the given set of postulates, it is easy to say that
what a postulate is: it is a member of the set.
Relative to a given set of semantical rules, it is
equally easy to say what a semantical rule is.

It might conceivably be protested that an artificial language L (unlike a natural one) is a language in the ordinary sense plus a set of explicit semantical rules

Semantical rules determining the analytic statements of
an artificial language are of interest only in so far as we already understand the notion of analyticity;

It is obvious that truth in general depends on both language and extra-linguistic fact.

for all its a priori reasonableness, a boundary between analytic and synthetic statement simply has not been drawn.

First we transform the synthetic side of the analytic/synthetic distinction into a finite list of “atomic facts” (of general knowledge) that form the basis of a verbal model of the conventional world. This utterly ignores any and all metaphysical speculation. We are simply stipulating a set of axioms.

“Cats” [are a type of] “Animal”
“Animals” [are a type of] “Living Thing”
“Banks” [are a type of] “Financial Institution”

These stipulations are the ultimate foundation
of all knowledge expressed in language. Knowedge
expressed in language cannot have any other basis.

From these axioms of natural language we allow the single inference step of semantic entailment using semantic rules (specified syntactically) that operate on finite strings.

The CycL language already has a very rich set of semantic rules that are specified syntactically that operate on GUID place holders for the unique sense meanings of words.

These are arranged in a knowledge ontology / simple type hierarchy. Thus synonyms simply have the same GUID.

19

That is why I defined this explicit scope:
“true on the basis of meaning expressed in language”
Stipulates the scope of all truth that can be expressed as relations between finite strings.

20

What does this say?

How does it pay out?