[7.2.4.] Quine’s Argument Against the Analytic/Synthetic Distinction.
This argument takes up sections 2 through 4 of “Two Dogmas.” The following summarizes Quine’s reasoning:
1. If there is a real, legitimate distinction between analytic and synthetic statements, then we must be able to explain analyticity in a non-circular fashion without relying on the notion of meaning.
That is, we must be able to give a philosophically satisfying account of what it is for a statement to be analytic, an account which itself does not rely on the notion of analyticity.
2. But analyticity cannot be explained in this way.
a. It cannot be explained in terms of synonymy, since synonymy itself cannot be explained in a non-circular way …
i. Synonymy cannot be explained in terms of definition, e.g., “two terms are synonymous when they have the same definition.” One of the points that Quine makes here is that when a dictionary editor defines a word in a certain way, she is recognizing the synonymy between the definiendum (the word being defined) and the definiens (the definition of that word). So we can’t explain what synonymy is in terms of dictionary definitions of words.
ii. Synonymy cannot be explained in terms of interchangeability without change of truth value (“salva veritate”) in an “extensional language” (a language in which two predicates having the same extension can be interchanged without changing the truth value);
Here Quine is using concepts that we’ve already seen and that should be part of your “Philosopher’s Toolkit” by now…
The Principle of Substitutivity (df.): Co-referring expressions (i.e., expressions that have the same reference) can be substituted for one another without changing the truth-value of the sentence in which the substitution is made.
· Leibniz’s formulation (quoted by Frege on p.13): “Eadem sunt, quae sibi mutuo substituti possunt, salva veritate”: “Those things are the same which may be substituted mutually everywhere without change of truth value.” This formulation makes it sound like it is the substitution of things, rather than linguistic expressions, which is at issue. For this reason, the modern statement of the law given above is preferable.
· Quine defines it in terms of linguistic predicates: “any two predicates which agree extensionally (i.e., are true of the same objects) are interchangeable salva veritate.” (523)
extensional context (df.): a linguistic context, e.g., a sentence, in which you can exchange co-referring expressions and not change the truth value; in other words, it is a context for which the Principle of Substitutivity holds.
E.g., “Superman can fly.” (T)
“Clark Kent can fly.” (T)
intensional context (df.): a linguistic context, e.g., a sentence, in which you cannot always exchange co-referring expressions without change of truth value; i.e., it is a context for which the Principle of Substitutivity does not hold. [Such contexts are also called opaque contexts and oblique contexts.]
E.g., “Lois Lane believes that Superman can fly.” (T)
“Lois Lane believes that Clark Kent can fly.” (F)
[This example illustrates something we saw earlier in the semester, when first studying Frege: reports of propositional attitudes create intensional contexts.]
The proposed explanation of synonymy that Quine is rejecting is this: two expressions are synonymous if, in a completely extensional language (a language with no intensional contexts), they can be interchanged without changing the truth value of the statements in which that change is made.
My example: “More than 65% of the earth’s surface is covered in water.” (T)
“More than 65% of the earth’s surface is covered in H2O.” (T)
We can explain what it is for “water” and “H2O” to be synonymous by saying that the two terms can be exchanged for each other in a wholly extensional language without changing the truth value of the sentences in which the exchange is made.
Quine rejects this explanation because there are counter-examples, statements that meet this criterion but that are not analytic, e.g., “creatures with hearts” and “creatures with kidneys.” These expressions can be exchanged salva veritate in an extensional language, but they are not synonymous:
“All creatures with hearts are creatures with hearts.” (true)
“All creatures with hearts are creatures with kidneys.” (true)
The underlined statements meet the criterion for synonymy under consideration, yet they are not synonymous. So synonymy cannot be exchangeability in an extensional context without change of truth value.
iii. It cannot be explained in terms of interchangeability without change of truth value (“salva veritate”) in an intensional (i.e., non-extensional) language containing the adverb “necessarily.”
Introducing “necessarily” changes a language from extensional to intensional, e.g.:
“Necessarily, nine is nine.” (true)
“Necessarily, nine is the number of planets.” (false)
“Nine” and “the number of planets” refer to the same abstract object, namely, the number nine. Yet they cannot be exchanged in the above sentences without change of truth value. So the word “necessarily” creates an intensional context.
Now, adding “necessarily” to our language does get around the hearts-and-kidneys problem…
“Necessarily, all creatures with hearts are creatures with hearts.” (true)
“Necessarily, all creatures with hearts are creatures with kidneys.” (false)
So the two expressions “creatures with hearts” and “creatures with kidneys” cannot be exchanged salva veritate once we introduce “Necessarily.” This gets around the counter-example that derailed this potential explanation of synonymy.
However, we cannot make sense of “Necessarily” without relying on the notion of analyticity. This strategy smuggles analyticity back in, and it’s the very notion that we are trying to explain! After all, what is being said by “Necessarily, all creatures with hearts are creatures with hearts” other than it’s a logical truth (i.e., an analytic truth) that all creatures with hearts are creatures with hearts?
So the attempt to define analyticity in terms of necessity is objectionable circular.
b. It cannot be explained in terms of semantical rules of an artificial language. [Quine’s arguments in support of this claim are quite complicated and abstract -- we will not consider them.]
3. So, there is no real, legitimate distinction between analytic statements (those that are true in virtue of the meanings of their words) and synthetic statements (those that are true in virtue of BOTH the meanings of their words AND the facts).
[7.2.5.] The Source of the Dogma.
At one point in his arguments, Quine asks: is the statement
“Everything green is extended.” (524)
analytic or synthetic? He concludes that he cannot answer this question. The problem is not that there is something wrong with the word “green” or the word “extended.” The problem is with the distinction between analytic statements and synthetic statements. The fact that there is no answer to the question about the above sentence is a symptom of the illegitimacy of the analytic/synthetic distinction itself.
So what is the source of this dogma? Quine has a suggestion:
It is obvious that truth in general depends on both language and extra-linguistic fact. The statement 'Brutus killed Caesar' would be false if the world had been different in certain ways, but it would also be false if the word 'killed' happened rather to have the sense of 'begat.' Hence the temptation to suppose in general that the truth of a statement is somehow analyzable into a linguistic component and a factual component. Given this supposition, it next seems reasonable that in some statements the factual component should be null; and these are the analytic statements. But, for all its a priori reasonableness, a boundary between analytic and synthetic statement simply has not been drawn. That there is such a distinction to be drawn at all is an unempirical dogma of empiricists, a metaphysical article of faith. (525)
Whether a given statement is true depends on both language (what its words mean) and fact (i.e, extra-linguistic fact, facts other than facts about language)
This suggests (incorrectly) “that the truth of a statement can be analyzed into a linguistic component and a factual component.”
This in turn suggests that there can be statements that don’t have a factual component, i.e. true statements the truth of which depends only on language, rather than on facts and language.
But this is false. Whether a statement is true depends on both language and extra-linguistic fact; however, it does not follow from this that there are true statements that are true only because of language.
Stopping point for Thursday November 7. For next time, no new reading (unless you haven’t finished reading “Two Dogmas,” in which case you should do that!). We will conclude our discussion of “Two Dogmas” next time. Sometime soon I will adjust the online reading schedule accordingly.
 Sometimes known as “the Principle of Substitutivity Salva Veritate” or “the Principle of Substitution.” This is a different principle from both Leibniz’s Law (a.k.a. the Indiscernibility of Identicals), according to which if a = b, then a and b have all their properties in common, and the Identity of Indiscernibles, according to which if a and b have all their properties in common, then a = b. Gallois describes the Principle of Substitutivity as “the linguistic counterpart” of Leibniz’s Law; see Andre Gallois, “Identity Over Time”, The Stanford Encyclopedia of Philosophy (Spring 2005 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/spr2005/entries/identity-time/>. That version of the article has been superseded by this more recent one: <http://plato.stanford.edu/archives/sum2012/entries/identity-time/>.
This page last updated 11/8/2012.
Copyright © 2012 Robert Lane. All rights reserved.