Quine's Rejection of the Two Dogmas
Quine’s arguments on analyticity’s circularity and the problem of reductionism in support of his theory on holism, which he believes provides a pragmatic approach that goes beyond the analytic. Submitted for PHIL 2103 on April 14, 2021.
In his famous essay Two Dogmas of Empiricism, W. V. Quine sets out to reject the two dogmas—namely, the analytic-synthetic distinction as well as reductionism—that are fundamental to modern empiricism. In this paper, I will first delineate Quine’s arguments, first on analyticity’s circularity and then on the problem of reductionism. Then I will underline the importance of Quine’s holism for pragmatism, namely, in providing a more thorough pragmatism that goes beyond the analytic. Finally, I will note my view in regards to Quine’s argument.
The purpose of section one through four in Quine’s essay is to argue that the meaning of analyticity has not been adequately established. Quine first defines what is analytic, by what he claims Kant’s intention of the word is, to be “a statement is analytic when it is true by virtue of meanings and independently of fact” (1a). This simply means that, unlike for synthetic statements, the truth of an analytic statement is not contingent upon the truth of other statements that make up the statement, for an analytic statement, which cannot escape the subject-predicate form, does not go beyond what is already contained in the subject and therefore lacks any synthesis at all.
Yet meaning is itself vague and obscure. As Quine notes, we often mistake meaning with naming. Some words, like “Evening Star” and “Morning Star” may have different meanings, but they refer to the same object, others, like the number “9” and “the number of the planets” may have the same meaning, but they refer to conceptually different things. He then purports that we must equally distinguish between the intension and the extension of a general term. That is, because “creatures with a heart” all have kidneys, it extensionally refers to the same entities as “creatures with a kidney,” but the two phrases’ meaning are evidently distinct (2b).
Furthermore, although, Quine concludes, “meaning is what essence becomes when it is divorced from the object of reference”—that is, a correspondence appears present between the “linguistic forms” and Aristotilean “essences”—it does not signify that meanings are objects themselves. This much is clear, Quine contends, once we separate meaning from its object of reference (2b). Hence, Quine proposes that we should instead tackle the problem of meaning from investigating words of linguistic synonymy and statements of analyticity (2b).
An example of a well-known category of analytic statement by “general philosophical acclaim” is “no unmarried man is married.” Supposing a logical inventory of “‘no,’ ‘un-’ ‘if,’ ‘then,’ ‘and,’ etc.,” this category of analytic statements are always true, regardless of how we interpret “married” or “man” (3b). The logical structure itself gives it certainty. Another category of analytic statement gains its truth from substituting synonyms for synonyms, like in the case of “no bachelor is married” (3b). That is, the aforementioned statement is true because we can substitute “bachelor” with its synonym “unmarried man,” and by doing so we have arrived at the first category of analytic statement, which holds analytic truth as far as general philosophy is concerned. But, Quine highlights, the analyticity of this second category depends on the notion of synonymity, which is itself, like the notion of analyticity, lacking a clear characterization (3b). As such, to confront the problem of analyticity, Quine moves to confront the problem of synonymy.
In an attempt to clarify synonymy, some may resort to definitions. If we appeal to a dictionary, we are conceding to the lexicographer’s interpretation of the word as law. Moreover, because the lexicographer is an “empirical scientist” and merely records what is already present in the world, when they record some term as the synonym of another term, they already believe a synonymous relation exists between the two terms. In other words, when we appeal to definitions to explain synonymy, we already presuppose synonymy. The only case in which a definition does not presuppose synonymy is when it involves conventionally created abbreviations, but these abbreviations do not move forward our discussion of analyticity.
Another approach to explain synonymy is to call upon interchangeability. Under this method, two linguistic forms are synonymous if they are interchangeable in “all contexts without change of truth value,” or interchangeable salva veritate. Since this is a universal claim, one counterexample is sufficient to prove it false. One only needs to consider the following phrase “‘bachelor’ has less than ten letters” and substitute ‘bachelor’ for ‘unmarried man’ to see that the two linguistic forms, which appear interchangeable, are not in fact interchangeable salva veritate. Setting aside clever wordplay, Quine emphasizes that what we are after is an account of “cognitive synonymy” by the way of interchangeability that does not suppose analyticity. At first glance, it appears that we may be successful in doing so, with the help of the adverb “necessarily” (7b). Hence, we may say “necessarily all and only bachelors are bachelors” is the same as “necessarily all and only bachelors are unmarried men” (7b). Yet, the lynchpin to this cognitive synonymy is the adverb “necessarily,” and therefore, once again, we are presupposing analyticity in our explanation of synonymy. Irrespective of the angles in our attempts to clarify analyticity by means of synonymy, we are running in circles.
Thus, Quine reckons, we are back at the problem of analyticity without the help of synonymy. This time, Quine presents his argument to dispute the claim that it is the vagueness of our ordinary language that poses a barrier to the distinction between analytic and synthetic statements. Quine contests this claim by appealing to semantic rules of Carnap. Suppose there is an artificial language that tells us exactly what all the analytic statements are. But because this rule contains the word “analytic,” it cannot be used to explain its extension, namely, analytic statements. As such, we still do not understand what analytic means. The notion of analyticity, Quine thus concludes, is also doomed to failure.
Yet, some may still maintain that the verification theory of meaning, or that to determine a statement’s meaning is to confirm or deny it empirically, gives rise to synonymy, and hence “the notion of analyticity is saved after all” (6a). Those who endorse a radical reductionism, the belief that all meaningful statements can be translatable into a statement about immediate experience, must prove that all such meaningful statements are in fact translatable by actually performing the translation, “statement by statement,” into sense-datum language (7a). Although Carnap embarked on this project in his Aufbau, his starting point is not rooted in sense-datum language due to the inclusion of logical notations, and therefore cannot be used to prove reductionism.
Rather than the approach to verify statements individually, like undertaken by Carnap, Quine recommends a form of holism, submitting that we put our statements about the world in the “tribunal of sense experience … as a corporate body” which is the whole of science (8a). To Quine, all beliefs and knowledge, whether it concerns history or logic, are interconnected, as the totality of science is one bound by our experience to the external world. Consequently, revisions to the outer edges of sciences also involve revisions to the inner core beliefs, including those traditionally held to be true in and of themselves in the case of analytic statements, like laws of logic or pure mathematics, as shown in quantum mechanics.
Quine’s holism is undoubtedly momentous in how we understand pragmatism. Having set out to dismiss the distinction between the analytic and the synthetic as well as to repudiate reductionism, instead of stopping at the analytic like some other pragmatic thinkers, Quine makes way for a more thorough pragmatism. Under this framework, by the virtue of our own scientific heritage and of the physical world that continually stimulates our sensations, we are free to amend our beliefs if we find them in need of revision in light of new empirical experiences and scientific evidence. Therefore, so long as it is rational, this broader, more comprehensive approach expands the reach of pragmatism.
I believe Quine’s argument is quite convincing, especially about his meticulous demonstration of analyticity’s circularity. Before reading Quine, I never thought of definition in a circular regard. Although I still do not agree that by appealing to a dictionary we are taking the lexicographer’s belief as law—because I think in order to believe in a conclusive definition, we surely have already exercised our own judgment, however slightly—I have modified my view to be more trusting of my own experience and more open to change. Quine’s essay emerges as a landmark in my philosophical journey, one that I will treasure as I continue forward.