Quine and Ontological Commitment
Do unicorns exist? I do not think so, but my friend, who believes unicorns exist, claims that my denial is self-contradictory. According to my friend, to say that no unicorns exist is to say that x is a unicorn and x does not exist; I am accused of saying that unicorns both do and do not exist. My friend thinks I must acknowledge the existence of unicorns in order to deny it. Unicorns must exist because the denial of their existence is incoherent. Since we can name unicorns (just by saying "unicorn"), and this name is meaningful (we are talking about white, horse-like animals with a single horn extending from their foreheads, and not some other animal, e.g., tigers), unicorns must exist. Obviously, my friend concludes, unicorns exist or we would not even be having this argument!
I can settle the debate with a little help from Quine and Russell's theory of descriptions. Using Russell's theory of descriptions, I can transform the sentence "The President of the United States is a wimp" into" Something is the President of the United States and is a wimp, and nothing else is the President of the United States and is a wimp." What appeared to be a name in the first sentence (President of the United States) has become a description in the second sentence in the same sense that "wimp" is a description. Note that the meaningfulness of the first sentence is retained in the second sentence; however, the phrase which demands objective reference in the first sentence is "President of the United States," but the word demanding objective reference in the second sentence is "something." "Something" is one example of what is known as a bound variable; "nothing" and "everything" are other examples. Bound variables "do not purport to be names at all; they refer to entities generally, with a kind of studied ambiguity peculiar to themselves." (p. 6; all references herein are to Quine's essay, "On What There Is"). Bound variables are meaningful, but it does not follow that bound variables refer to some existing object.
Thus, armed with Russell's theory of descriptions, I posit the following statement: "Something is a unicorn and is white and is horse-like and has a single horn extending from its forehead." I can then assert that this statement is false without fear of my friend raising the same objection discussed above. My friend has simply confusing naming and meaning: "unicorn" may have meaning, but it is not a name because it has no objective reference.
Now my friend is angry; I have challenged a cherished belief. He still believes unicorns exist, but he cannot make me believe. Things start to get ugly. "Alright, Mr. Smarty Pants," he says. "Does blue exist?" He knows very well that blue is my favorite color. "My eyes are blue, the sky is blue, and your shirt is blue," I say, "but if you are asking me whether blueness exists, then I must say no. Blueness does not exist any more than unicorns exist. Saying that my eyes, the sky, and your shirt all have something in common, blueness, does not commit me to the existence of anything more than my eyes, the sky and your shirt, and does not imply the existence of any entity, even an abstract entity; to claim that these items have something in common is a popular and misleading way of speaking."
My friend remembers our discussion about unicorns, so he knows he cannot claim that I imply the existence of blue simply by using the term. "Do you deny that the word "blue" has any meaning?" he asks with a sneer." Indeed, my good friend Quine and I deny the existence of meanings altogether," I respond. Quine distinguishes between an utterance having meaning and an utterance being meaningful. According to Quine, meaningfulness can be explained behaviorally, as, for example, when my contentious friend and I agree in identifying blue objects; Quine refers to this aspect of meaningfulness as significance. Synonymy, the other aspect of meaningfulness, involves substitution or interchangeability. Quine claims synonymy is is what occurs when we purport to give meaningfulness to a term. As with significance, synonymy is evidenced by behavior, i.e., how terms are actually used. Quine prefers that "we speak directly of utterances as significant or insignificant, and synonymous or heteronymous with another. ... But the explanatory value of special and irreducible intermediary entities called meanings is surely illusory." (p. 12).
My friend is thoroughly exasperated; I have claimed not only that unicorns, but also that universals and meanings do not exist. He pulls some coins out of his pocket. "look here," he says. "Some dimes are brighter than others. Or will you deny this also?" I tell him that I prefer to say "Some of our visual dime-like sensations appear to be more whitish and shiny than other such sensations." Being somewhat of an empiricist, I want to reduce my statements to reports of specific sensory experiences as much as possible; my reformulation is a much more precise statement. My friend's statement commits him to an ontology including dimes and brightness; my statement commits me to an ontology of sensations. My friend's statement implies the existence of the entity "brightness", an entity which I emphatically deny (even though I can speak of it); moreover, the use of a relative term (brighter) requires comparison of two non-existent entities. My friend's statement takes the form of a categorical, and I have not seen enough evidence to support such a strong statement. My version provides more concrete conditions for testing and confirming or disconfirming the statement. My statement involves "the simplest conceptual scheme into which the disordered fragments of raw experience can be fitted and arranged." (p.16).
My friend can bear no more. He stomps off saying, "I'll show you; I'm going to find a unicorn and prove you're full of nonsense." I call after him: "Bring back any universals or meanings you find along the way, too." I am left alone in a desert landscape to ponder the question of ontological commitment. In Quine's view, nothing we say commits us to the assumption of universals or other entities; only the invocation of a bound variable commits us to the existence of an entity. We may say that there is something which my eyes, the sky and my friend's shirt have in common, and thereby posit the existence of an entity. But we may also refuse to make such a statement. Quine states that "To be assumed as an entity is, purely and simply, to be reckoned as the value of a variable." (p. 13). For example, to say that some of my friends are stupid does not commit me to the existence of friendship or stupidity; the statement means only that some things which are my friends are stupid. In order to make the statement true, I only need to produce a stupid friend, who then represents a value for the bound variable.
Producing statements with bound variables does not determine what there is, it only denotes what I am willing to say there is; since ontological debates take place at a linguistic and semantic level, identification of ontological commitments is crucial for understanding the conceptual schemes underlying the debate. Quine points to two useful conceptual schemes in developing an ontology: physicalist and phenomenalist. A physicalist scheme is useful for organizing sensory experiences, and Quine accords epistemological priority to the phenomenalistic scheme. As Quine notes, "the question what ontology actually to adopt still stands open, and the obvious counsel is tolerance and an experimental spirit." (p. 19). Quine does not believe the physicalist and phenomenalistic scheme are incompatible; although each may be suited for particular purposes, the schemes should cross-fertilize each other. The goal is to develop strategies for dealing with the widest possible variety of situations. Quine is, in the final analysis, a pragmatist.
Please use the links below to reach other areas of this site:
Last revised: Apr. 23, 2014.