One of Quine’s aims in “Two Dogmas of Empiricism” (TD) is to argue that no sharp distinction can be drawn between analytic statements and synthetic statements. Analytic statements are ones alleged to be “true by virtue of meaning and independently of fact” (TD 21). Truths grounded in fact are known as synthetic truths; statements of such truths are called synthetic statements.
Quine reminds the reader that meaning “is not to be identified with naming” (TD 21). My height and my stature (under one of its definitions) name the same thing and mean the same thing. Likewise for three and drei. In general, however, concepts with different meanings can name the same thing, as my right hand and my writing hand.
Quine writes of terms, but these are terms working in a certain way, terms employed in statements admitting of truth or falsity. Quine’s “terms in statements” would seem not far from “concepts in propositions” which is the technical vocabulary adopted by Rand.
Turning to general terms like right hand, Quine observes that we must distinguish between the meaning of the term and the extension (the referents) of the term. Think of the essence of right hands. Instead of thinking of the essence in the Aristotelian way—as inhering in those hands (actual and possible)—let it inhere in the term. That thought is the meaning of a general term, and, like Aristotelian essence, it is not one and the same as the thing signified (TD 21–22).
I better hit the gavel for Rand at this point. We speak of the meanings of words, but words are only markers for concepts, “and the meaning of a concept consists of its units.” We define concepts “by specifying their referents.” Concepts and definitions are certain ways of specifying referents (ITOE 44).
I look up the word derelict in my dictionary and find one of its meanings: abandoned property; especially, a ship abandoned at sea. Knowing how to apply the latter term (grammatically, a phrase), I might now use this sense of the word derelict. The word being defined and its definition have the same meaning. They are cognitively synonymous.
Quine thought that the useful conceptions of meanings come down to (i) giving synonyms or (ii) making significant utterances (“On What There Is” 11). Rand held that when we make significant utterances that engage concepts, or general terms, those concepts have definitions specifying their referents. (Prior to being able to state propositions in which concepts figure, a concept like ball is [marked by a word and] nested in image and action schemata [ITOE 13, 20, 43; further UM, G]. Presumably, this rudimentary mentation, alternative to explicit propositions and definitions, informs them.)
Quine notices that there are definitions of a sort that are not simply the giving of synonyms, and he calls these sorts of definitions explications.
In explication the purpose is not merely to paraphrase the definiendum [the term being defined] into an outright synonym, but actually to improve upon the definiendum by refining or supplementing its meaning. But even explication, though not merely reporting a preexisting synonymy between definiendum and definiens [the definition], does rest nevertheless on other preexisting synonymies. . . . Any word worth explicating has some contexts which, as wholes, are clear and precise enough to be useful; and the purpose of explication is to preserve the usage of these favored contexts while sharpening the usage of other contexts. In order that a given definition be suitable for purposes of explication, therefore, what is required is not that the definiendum in its antecedent usage be synonymous with the definiens, but just that each of these favored contexts of the definiendum, taken as a whole in its antecedent usage, be synonymous with the corresponding context of the definiens.
Two alternative definientia may be equally appropriate for the purposes of a given task of explication and yet not be synonymous with each other; for they may serve interchangeably within the favored contexts but diverge elsewhere. By cleaving to one of these definientia rather than the other, a definition of explicative kind generates, by fiat, a relation of synonymy between definiendum and definiens which did not hold before. But such a definition still owes its explicative function, as seen, to preexisting synonymies. (TD 25)
When philosophers lay out theories of good definition, they are theories of an explicative kind of definition (see David Kelley’s Art of Reasoning, chapter 3). Consider Rand’s definition of reason as the faculty that identifies and integrates the evidence of the senses. In my dictionary, I find reason defined as the capacity for rational thought, rational inference, or rational discrimination. The terms rational and thought go to already familiar synonymies with reason. The differentia within the rational, in this dictionary definition, are the discriminatory and the inferential.
Rand’s definition stays close to the common usage reflected by the dictionary, but it replaces discrimination and inference by their kin identification and integration, it eliminates the non-explicative rational, and it adds a base for the activities of reason, specifically, deliverances of the senses. Rand’s definition is explanatory of the common usage found in the dictionary, and it is tailored to tie neatly to a particular wider philosophical view.
Quine could say this is a fine explicative type of definition. Rand has given the term reason a new synonymy. The various contexts in which reason under the dictionary definition is properly used remain contexts in which reason under the new, explicative definition is properly used. The new definition covers the processes of drawing distinctions and making inferences. The new definition also applies to the wider processes of identification and integration of sensory evidence, processes in which the narrower processes are embedded. Quine would stress that, nonetheless, “such a definition still owes its explicative function . . . to preexisting synonymies” (TD 25). Quine is being too short here.
Quine’s argument against the idea that there are clearly statements true purely by virtue of meanings, and true independently of fact, hangs on his conception of meaning. We have joined Quine in saying that meaning is distinct from reference. We have not allowed that meaning can be independent of reference. The meanings of my right hand and my writing hand differ, but both meanings are specifications of a referent. Similarly, the meanings of right hand and writing hand differ, but both are specifications of the extensions (the referents) under those concepts.
Quine’s conception of meaning is shriveled into “synonymy of linguistic forms” (TD 22). He allows that a logical truth such as “Every tall man is a man” has a guarantee of truth that rests on more than one’s experiences of facts about men. He realizes that logical truths are sometimes called analytic, but his target is other statements taken for analytic: statements reducible to logical truths by synonymies, statements reducible to logical truths by meaning.
Quine shares with defenders of analyticities the conviction that logical truths are true, and true regardless of particular facts to which they are applied. If there are statements reducible to logical truths by virtue of meaning and independently of fact, then their truth would be guaranteed by virtue of meaning and independently of fact. A candidate analytic statement from Kant would be “Bodies have location.” (A contrasting synthetic statement would be “Orbiting bodies are weightless.”) One who has the concept body knows that having location is part of the meaning of the concept. Substituting “Things having location (and . . .)” for body yields the logical truth “Things having location (and . . .) have location.”
Now, we know “Necessarily, things having location (and . . .) have location.” Does only that sense of necessity attach when we claim “Necessarily, bodies have location”? Quine disputes the idea that purported analyticity of a statement can be adequately explained by cognitive synonymies and logical truth (TD 29–31). Analyticity cannot be explained by a sensible conception of meaning joined with logical truth. An adequate way of distinguishing analytic from synthetic statements has not been produced.
Quine uncovered a narrow, but serious, problem for the synthetic-analytic distinction. It would seem that there are wider problems for the distinction that he passes over because of his cramped conceptions of meaning, definition, and essential characteristics.
(Page numbers for Quine are from the 1980 edition of his collection From a Logical Point of View.)