Quine's Critique of the Analytic-Synthetic Distinction

Introduction

Welcome to your next lesson. In our previous session, we examined the self-refutation objection to the logical positivists' verification principle. We saw that their core criterion—that a statement is only meaningful if it is either analytic (a definitional truth) or empirically verifiable—could not seem to account for its own meaningfulness. This internal contradiction was a major blow to the positivist project.

Today, we turn to an even more fundamental challenge, launched by the American philosopher W.V.O. Quine in his seminal 1951 paper, "Two Dogmas of Empiricism." Instead of just criticizing the verification principle, Quine dismantled one of its foundational pillars: the very distinction between analytic and synthetic truths. This lesson will focus on the first of Quine's "two dogmas," and by the end, you will be able to reconstruct Quine's critique of the analytic-synthetic distinction.

This attack was devastating because the analytic/synthetic distinction was not just a minor technical point; it was the bedrock of the empiricist program. It allowed them to accept the truths of logic and mathematics as analytic a priori truths—true by definition and independent of experience—without having to appeal to the kind of mysterious rationalist faculties they so despised. If Quine could show that this distinction was untenable, the entire empiricist picture of knowledge would begin to unravel.

1. The Distinction and Its Importance

Before we can understand Quine's critique, we must be clear on what the analytic/synthetic distinction is and why it mattered so much. You'll recall these concepts from our module on Kant, but the logical positivists adapted them for their own purposes.

Let's review the core ideas and their significance for 20th-century empiricism.

As the video explains, this distinction was the empiricists' answer to a critical question: How can we have certain, necessary knowledge (like $2+2=4$) in a world we only know through contingent, uncertain experience? Their answer: such truths are analytic. They aren't deep facts about the world, but are true simply "by virtue of the meanings of their terms."

Quine's central claim is that this seemingly common-sense idea—that truth can be neatly separated into a "linguistic component" and a "factual component"—collapses under scrutiny. He argues that the notion of "truth by virtue of meaning alone" is a metaphor that cannot be cashed out in any clear, non-circular way.

2. Quine's Argument: The Search for a Definition

Quine's method is a process of elimination. He systematically examines every plausible way to define "analyticity" and shows how each attempt either fails or ends up presupposing the very concept it is trying to define. His conclusion is that "analyticity" belongs to a small circle of inter-definable but ultimately unexplained concepts, including synonymy and necessity.

Let's trace the key steps of his argument.

Attempt 1: From Analyticity to Synonymy

Quine first divides analytic statements into two classes:

1. Logical Truths: Statements that are true and remain true under any reinterpretation of their non-logical parts.
   • Example: "No unmarried man is married." This has the logical form No not-X is X.
2. Other Analytic Truths: Statements that are not logical truths on their face but seem to be true by definition.
   • Example: "No bachelor is married."

Quine grants that the first class is unproblematic. His focus is on the second class. The standard way to explain the truth of "No bachelor is married" is to say that it can be turned into a logical truth by substituting the synonym 'unmarried man' for 'bachelor'.

This move seems promising, but it only shifts the burden. We have explained analyticity, but now we need to explain synonymy. What does it mean for two terms to have the same meaning?

Attempt 2: From Synonymy to Definition

A natural next step is to appeal to definitions. Perhaps 'bachelor' and 'unmarried man' are synonymous because one is the definition of the other, as you might find in a dictionary.

Quine argues this puts the cart before the horse. A lexicographer is an empirical scientist who records pre-existing facts about language. When a dictionary states that 'bachelor' means 'unmarried man', it is reporting the fact that these terms are already used synonymously in the language community. The definition presupposes synonymy; it doesn't explain it.

The only exception, Quine notes, is explicit stipulative definition, where we introduce a new shorthand. But the interesting analytic truths of philosophy and mathematics are not mere stipulations.

Attempt 3: From Synonymy to Interchangeability

This brings us to the most technical and crucial part of Quine's argument. Perhaps we can define synonymy without appealing to definitions. A promising criterion, drawing from Leibniz, is interchangeability salva veritate—the idea that two expressions are synonymous if one can be substituted for the other in any sentence without changing that sentence's truth value.

Quine shows that this criterion is fatally ambiguous. Its effectiveness depends entirely on the type of language we assume. Your background in mathematics should make the following distinction particularly clear.

• In an "Extensional" Language: An extensional language is one that only allows you to talk about objects and their properties/relations (their extensions). In such a language, interchangeability is too weak a criterion for synonymy. Quine uses the famous example of 'creature with a heart' and 'creature with a kidney'. As a matter of contingent biological fact, every creature with a heart also has kidneys, and vice-versa. These two phrases have the same extension. Therefore, in any simple, non-modal statement, you can swap one for the other without changing the truth value. Yet, they clearly do not have the same meaning.

• In an "Intensional" Language: To distinguish 'creature with a heart' from 'creature with a kidney', we need a richer language that includes modal adverbs like 'necessarily'. We can then form sentences like:

  • Necessarily, all and only bachelors are unmarried men. (True)
  • Necessarily, all and only creatures with a heart are creatures with a kidney. (False)

  This seems to solve the problem! We can now define cognitive synonymy: two terms are synonymous if they are interchangeable salva veritate even within the context of "Necessarily...".

But this leads directly to the circularity Quine wants to expose. For an empiricist, what does an adverb like "necessarily" even mean? They would typically explain it by saying it applies to all and only analytic statements. So, to use 'necessarily' to provide a criterion for synonymy (which we need to explain analyticity), we must already have a grasp on what 'analytic' means. We are back where we started.

The following resources will guide you through this complex chain of reasoning. The video lays out the argument step-by-step, while the reading from Prof. Speaks' notes provides a concise, formal summary that reinforces the logical structure of Quine's critique.

3. The Last Resort: Semantic Rules

Quine considers one final attempt, associated with his mentor Rudolf Carnap. Perhaps the problem is the messiness of natural language. In a precisely constructed artificial language, couldn't we simply lay down "semantic rules" that stipulate which sentences are to count as analytic?

Quine argues that this maneuver also fails to escape the circle.

• If a rule simply says, "The following sentences are in the class 'Analytic'," this gives us a label but no understanding of what the property of analyticity is. We could have called the class 'K' or 'Smargon' for all the explanation it provides.
• A more subtle approach is to have semantic rules that designate certain statements as true by fiat. We could then define the 'analytic' statements as those that are true according to these rules. But then a new problem arises: what distinguishes a "semantic rule" from any other belief we hold to be true? For example, we hold "There are eight planets" to be true. Why can't we just call that a semantic rule?

Quine concludes that the notion of a 'semantic rule' can only be distinguished from other strongly held beliefs by appealing to a prior, intuitive notion of analyticity—the very thing it was supposed to explain.

For a summary of this final step and Quine's overall conclusion, this brief article is very helpful.

Conclusion

Quine's methodical deconstruction of the analytic/synthetic distinction left the empiricist project in a precarious position. He argued that despite its "a priori reasonableness," a clear boundary between truths of language and truths of fact "simply has not been drawn." For Quine, this distinction was not a discovery but an "unempirical dogma of empiricists, a metaphysical article of faith."

Key Takeaways:

• The analytic/synthetic distinction was crucial for empiricism, as it provided a way to account for the necessary truths of logic and mathematics without recourse to non-empirical modes of knowing.
• Quine's central argument is that all attempts to give a clear definition of 'analyticity' are circular.
• Attempts to define analyticity via synonymy only shift the problem.
• Defining synonymy via definition is backwards, as lexicographers report on pre-existing synonymy.
• Defining synonymy via interchangeability salva veritate either is too weak (in extensional languages) or presupposes an understanding of modality ('necessarily'), which for an empiricist presupposes analyticity.
• Defining analyticity via semantic rules in artificial languages also fails, as there is no non-circular way to distinguish a 'semantic rule' from any other firmly held belief.

Preview of the next lesson:

Quine did not just tear down this dogma; he proposed a radical new picture to replace it. Having argued that statements cannot be individually tested against experience as either 'analytic' or 'synthetic', he introduced his own holistic model. In our next lesson, we will explore Quine's second dogma—reductionism—and his famous metaphor of the "web of belief," where knowledge as a whole, including logic and mathematics, confronts experience only as a "corporate body."