In Two Dogmas of Empiricism, Quine examines various definitions of analyticity. Respectively, he proceeds to reject each for their lack of clarity. In fact, a dominant approach of the essay is characterized by a question: What does “analyticity” (given x definition) mean?
First, the most succinct definition of analyticity: (1) statements whose denials are self-contradictory1. An attack of clarification, analyticity hinges on the notion of self-contradiction, but what is “self-contradiction”? Then, the Kantian definition, (2) an analytic statement as one that attributes to its subject no more than is conceptually contained in it2. The characterization is limited, only dealing with propositions of subject-predicate form. And the typified critique, that “it appeals to a notion of containment which is left at a metaphorical level”3, not clear enough. This is transformed into a stronger definition, one that “contains” it: (3) a statement is analytic when it is true by virtue of meaning and independently of fact4.
Proceeding with (3), Quine elaborates on the two classes of analytic statements: trivial logical analyticity and all else. No unmarried man is married, substituting the variables it reveals a tautological logical truth. Thus, conforming to (1), (3) – thereby (2) as well. The interesting case, the non-trivial, concerns statements as follows: No bachelor is unmarried. Why is the statement analytic? One answers: for bachelor means an unmarried person. Thus, we have the synonymous substitution criteria.
Synonymous substitution criteria: Given proposition p, if through substituting synonyms, it can be transformed into trivial logical analyticity, then proposition p is analytic.
Attack of clarification proceeds. “Analyticity”, given (3), requires synonymy. What is “synonymy”? For the proverbial bachelor, how do we ground the substitution, “how do we find that ‘bachelor’ is defined as ‘unmarried man’”5? Immediately, the prospect an obvious empirical ground is dismissed:
The lexicographer is an empirical scientist, whose business is the recording of antecedent facts; and if he glosses ‘bachelor’ as ‘unmarried man’ it is because of his belief that there is a relation of synonymy between these forms, implicit in general or preferred usage prior to his own work.6
For it still assumes a pre-existing synonym, the question of “synonymy (?)” still left open. What about definitional activity? Carnap’s explication: to not merely paraphrase the definiendum into definiens, but to improve, refine, supplement the meaning. However, this still falls back on pre-existing synonymy. For refinement implies that certain pre-existing synonymy are preserved, while others are modified. A transfer, a preservation of synonymy, from the definiendum to the definiens. All that’s left is the case of an “extreme”.
There does, however, remain still an extreme sort of definition which does not hark back to prior synonymies at all; namely, 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; would that all species of synonymy were as intelligible. For the rest, definition rests on synonymy rather than explaining it.7
Has not Quine solved the problem here? We know that synonyms can be created explicitly, say E-synonymies. The problem facing us is the genesis of pre-existing synonymies. Suppose that these pre-existing synonymies form a recursive chain originating from type E-synonyms. Not to be confused with a comment on the structure of language, rather a possible scenario of construction, term it Language E-synonymy. For an analogy emphasizing the incoherence –
Now if we are to take these words of Quine seriously, then his position as a whole is incoherent.. It is like the position of a man to whom we are trying to explain, say, the idea of one thing fitting into another thing, or two things fitting together, and who says: “I can understand what it means to say that one thing fits into another, or that two things fit together, in the case where one was specially made to fit the other; but I cannot understand what it means to say this in any other case.8
However, it is easy to dismiss Quine’s question by relegating it to incoherence. What’s more interesting is investigating the source of the question: Why does Quine ask what “synonymous” means? Clearly, it must expect an account of the genesis of “synonym” other than the explicit conception. We notice that a property of E-synonymy is contingency. And perhaps the question becomes: What does “synonymous” truly mean?
Standard of Necessity
How do we confirm that the attacks of clarification are aimed at necessary formulations of analyticity? Consider a possible formulation of “synonymy” that Quine offers.
A natural suggestion, deserving close examination, is that the synonymy of two linguistic forms consists simply in their interchangeability in all contexts without change of truth value; interchangeability, in Leibniz’s phrase, salva veritate.9
Synonymy is necessary synonymy. Between two linguistic forms synonymy is interchangability without exception, that it must be interchangeable and no other way. This definition is an accepted formulation, whether it is granted true or not, while E-synonymy, the extreme case, is not considered. Therefore, one wonders if the attack is against ‘analytic’, or against another the lack of which would leave ‘analytic’ groundless. As Putnam writes:
But Quine also considers a very different notion: the notion of an analytic truth as one that is confirmed no matter what. I shall contend that this is the traditional notion of apriority, or rather, one of the traditional notions of apriority
…
I think that Quine’s attack on this notion was correct.10
Specifically, the attack is against necessary a priori. That analytic \(\iff\) a priori and necessary. Contingent a priori, the E-synonymy here, is not even considered as ‘analytic’. This sentiment is made clear given the focus on incorporating Fallibilism with historical scientific examples.
Even a statement very close to the periphery can be held true in the face of recalcitrant experience by pleading hallucination or by amending certain statements of the kind called logical laws. Conversely, by the same token, no statement is immune to revision. Revision even of the logical law of the excluded middle has been proposed as a means of simplifying quantum mechanics; and what difference is there in principle between such a shift and the shift whereby Kepler superseded Ptolemy, or Einstein Newton, or Darwin Aristotle?11
Is there a response from Quine, against analytic grounded in contingent a priori? Perhaps not. But another from the same camp, who still persists with the argument, even if it is construed “as an argument to show that words in natural language and scientific language are ambiguous— that “man” is synonymous with “rational animal” in one situation and with “featherless biped” in another”11. The renewed analytic is compared to the Millian take, to reject the philosopher who holds “The one and only true meaning” view of analysis, relegating meaning to the context where the word appears.
The Millian takes as his fundamental metalinguistic statement-form: “X is synonymous with Y in situation S,” whereas his opponent apparently refuses to relativize synonymy. The opponent merely says: “X is synonymous with Y.” What I want to emphasize, however, is that by so relativizing the notion of synonymy he is still far from meeting the difficulty I have raised. For now it may be asked how we establish synonymy even in a given situation.12
White’s concern is answered by E-synonymy. Much like Putnam, certain strains of Quine and White are sympathized with, the critique only holds for analytic as a priori and necessary. And even that, if a priori and necessary statements exist at all. Given that it is contingent, that genes merely is, a capacity exists in language to assert synonymy. Even beyond synonyms, the Capacity of Contingent Assertion. This question warrants a separate investigation.
Language Games: Quine, White, and Others
Language Games, in the Wittgensteinian sense, that is in sense as such. They are contingent a priori, or contextual a priori as Putnam puts it.
Such statements have a sort of ‘apriority’ prior to the invention of the new theory which challenges or replaces them: they are contextually a priori. 13
We are concerned with language games we are born into, much like the context of variants, in scientific theories and logic that is discussed above. Even Quine acknowledges, even utilizes, language games as critiques against analyticity. Thus, confirming our notion that the attack of clarification is against analyticity as a priori and necessary.
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”.14
Similarly, White also admits to the lack of a problem given various formalized languages. “It must be distinguished from the case where we artificially construct a language and propose so-called definitional rules. In this case we are not faced with the same problem.”15. Consider formal language L₁ and L₂. For L₁, rational animal is an E-synonym for man. Featherless biped can never be an E-synonym for man. For L₂, it is vice-versa. Here “it is easy to see that we can construct a language L₂ in which the reverse situation prevails and in which a linguistic shape which was analytic in L₁ becomes synthetic in L₂, etc.16
Given (3) definition of analyticity, if meaning is never necessary a priori – Quine’s (and White’s) critique of analyticity becomes irrelevant. And if they are critiquing necessary a priori, I would assume there are better ways to go about it.
Putnam’s Criterion of Analyticity
In ‘Two Dogmas’ Revisited, Putnam refers to a previous essay where he outlines a theory of analyticity. Analyticity of the second class of analytic statements – All bachelors are unmarried, All vixens are foxes, etc. A statement is analytic given that there is an exceptionless law associated with the noun – ‘vixen’ if and only if it is a female fox and so on. But this exceptionless law has two characteristics.
- that no other exceptionless ‘if and only if’ statement is associated with the noun by speakers
- that the exceptionless ‘if and only if’ statement in question is a criterion, “i.e., speakers can and do tell whether or not something is a bachelor by seeing whether or not it is an unmarried man; whether or not something is a vixen by seeing whether or not it is a female fox” 17
Consider the example “Father if and only if he has a child” and “Father if and only if he is an ordained priest outside of x,y,z traditions.” The latter is specified for several protestant denominations that do not call the ordained priest Father. Both are exceptionless definitions. Both can be considered analytic, in their respective senses. If we analyze this supposed counterexample given Putnam’s criterion.
- Both statements aren’t analytic. For “Father” is a polysemy, that there are more than one exceptionless statements.
- The latter definition isn’t exceptionless, for it weeds out exceptions explicitly.
- Note that the former does this implicitly, multiple senses of Father including the Christian, the sexual link, etc., all weeded out.
Therefore, (1). While Putnam’s view is coherent, the explanatory value leaves something to be desired.
I contended that only a few hundred words in a natural language have this ‘one-criterion’ character: most words are either associated with no exceptionless criterion, or with more than one. And I further suggested that all clear cases of analyticity involve these special few hundred words.)
Redefining Analytic Given Language Games
Analytic statements as those that explicate the generative conditions of a language game. Analytic statements can only be construed to be analytic given a language game. Given Quine’s rejection of necessary a priori analyticity specifically, analytic contingent on the language games is not refuted. As Putnam writes –
Can one hold that there are no a priori truths, but there are analytic truths (in the ‘linguistic’ sense)?
The answer is plainly that one can. If one accepts the distributive laws of the standard (Boolean) propositional calculus, for example, then one will accept any statement of the form
\[p(q \lor r) \equiv pq \lor pr\]
as logically true (and a fortiori as analytic). But in quantum logic there are statements of the form \(p(q \lor r) \equiv pq \lor pr\) which are not regarded as logically true (\( p(q \lor r) \equiv pq \lor pr \) is not a ‘tautology’ in quantum logic). So if we change our minds about which logical laws are correct by going over to quantum logic, we shall change our minds about the ‘analyticity’ of these statements.18
Now, an attack of clarification on the two notions: generative conditions and explications. If we use language, there are certain “transcendental” conditions, the quasi-Kantian sense that are unavoidable. The two relevant here:
- There exists language games
- We can discern between language games
Perhaps the unavoidable assumptions and the characteristics of the discern can be discussed further. A fuller treatment of this issue must be deferred.
Generative Conditions (Language Games): Given a language game L, the generative conditions of L include all that which is used to discern L from any other language game, not-L. Not a set of propositional conditions, exhaustible or not, rather an immanent generative set.
Explication: A stating of the generative conditions, the discerning.
To evaluate this given the (3) definitions.
- (1) – For the language game assumes generative conditions, it cannot be denied without self-contradiction
- (3) – Given that meaning is understood through language games, generative conditions do not require fact for its assertion.
- Given that (3), that it is (2).
References
1. Quine, W. V. O. (1951). “Two Dogmas of Empiricism.” The Philosophical Review, 60(1), 20–43. Hereafter referred to as TDE. p. 20.
2. Ibid.
3. TDE, p. 21.
4. Ibid.
5. TDE, p. 24.
6. Ibid.
7. TDE, p. 26.
8. Grice, H. P., & Strawson, P. F. (1956). “In Defense of a Dogma.” The Philosophical Review, 65(2), 141–158. p. 152.
9. TDE, p. 27.
10. Putnam, Hilary. (1983). “‘Two Dogmas’ Revisited.” In Realism and Reason. New York: Cambridge University Press, pp. 87–97. Hereafter referred to as TDR. p. 87
11. White, Morton G. (1950). “The Analytic and the Synthetic: An Untenable Dualism.” In Sidney Hook (Ed.), John Dewey: Philosopher of Science and Freedom. New York: The Dial Press, pp. 316–330. Hereafter referred to as ASUD p. 327.
12. ASUD p. 329.
13. TDR. p. 95
14. TDE. p. 34
15. ASUD p. 321.
16. Ibid.
17. TDR. p. 89
18. TDR. p. 96