As we have been discussing ontology as it refers to hidden variables and multiple worlds, I thought there might be some interest in this excerpt from my Hall book, The Roots of Representationism. It focuses on a shift in Quine’s position subsequent to “On What There Is,” but I think it touches on some of the broader questions of ontology and how one ought to investigate it as well.
Postscript on Quine, Alston and Hall on Ontological Commitment
As one might infer from the foregoing, Hall takes the unearthing of the ontological commitments imbedded in philosophical theories to be a tricky, nuanced undertaking. There will generally be found overlaps and lacunae in the evidence (the relevant philosophical statements), and some of this material may require careful interpretation. That position is in stark contrast to Quine’s early pronouncements on this subject, where it seemed one could just look for quantifiers and read off their ranges. Let us therefore pause a moment in our exposition of Hallian categories to explore this difference in approach. In his seminal 1948 paper, “On What There Is,” Quine continued the assault on Meinongian beings that began with Russell’s Theory of Descriptions by arguing that proper names could be converted to definite descriptions and handled in just the way Russell dealt with “the present king of France.” That is, on Quine’s view, finding what seems to be a proper name in some theory doesn’t demonstrate that the theory is committed to entities referred to by that name. Thus far, Hall would agree: the simple search for proper names can only be the beginning of a credible ontological analysis of any philosophical (or scientific or ordinary language) writing. But Quine goes on to claim that if one can “paraphrase away” sentences containing any name “N,” then we need not countenance any such items as might have been thought to be named by “N.” Thus, because one has used sentences containing the word “Pegasus,” it doesn’t follow that one can be accused of (even unwillingly) countenancing flying horses, so long as a good paraphrase is available. In particular, there is no danger to one’s sense of philosophical
parsimony created just by asserting things like “Pegasus doesn’t exist.” Quine (1948 : 8) put it this way:
[T]he singular noun in question can always be expanded into a singular description, trivially or otherwise, and then analyzed out a la Russell. We commit ourselves to an ontology containing numbers when we say there are prime numbers larger than a million; we commit ourselves to an ontology containing centaurs when we say there are centaurs; and we commit ourselves to an ontology containing Pegasus when we say Pegasus is. But we do not commit ourselves to an ontology containing Pegasus or the author of Waverley or [a round square] cupola when we say that Pegasus or the author of Waverley or the cupola in question is not. We can very easily involve ourselves in ontological commitments by saying, for example, that there is something (bound variable) which red houses and sunsets have in common; or that there is something which is a prime number larger than a million. But this is, essentially, the only way we can involve ourselves in ontological commitments: by our use of bound variables.
On this point, Hall demurred; he was not convinced either that the matter was so simple as Quine claimed or that there is anything terribly special about “There are…” or “Some…” statements. He was not alone in taking this tack. For example, William Alston in his little 1957 paper, “Ontological Commitments” pointed out that
[W]hether a man admits (asserts) the existence of possibilities depends on what statement he makes, not on what sentence he uses to make that statement. One admits that possibilities exist whenever he assertorially utters `There is a possibility that James will come’ or any other sentence which means the same (would ordinarily be used to make the same statement.) It is a question of what he says not of how he says it. Hence he cannot repudiate his admission by simply changing his words. A man who was afraid of policemen would be reassured if he were convinced that there are no policemen. But he would not be reassured if he were convinced that one could express all one’s beliefs in a language which took not policemen, but rather policemanship, as values of variables…. (1957: 13)
Hall (1960: 32) attacks from a similar angle. He argues that an absolute idealist such as F.H. Bradley was committed to his Absolute whether or not this commitment could be found either in any quantifier binding a variable or in the range of any acceptable variables. Bradley simply took Reality-as-a-whole to be the implicit subject of every assertion and gave it the proper name “The Absolute.” It seemed to Hall that if that was not a clear case of ontological commitment to some putative entity, nothing ever could be. Thus, any insistence by Quine in 1948 to the effect that “The Absolute” in Appearance and Reality is eliminable has no bearing whatever on an assessment of Bradley’s ontological commitments. Beliefs may not be justly ignored just because they are not put in some particular way. For Hall, all the resources of language must be investigated to determine ontological commitment. In fact, Hall notes that if we could rest with the simple search for quantifiers, we might find that “while Dickens was not committed to the real existence of Pickwick, he was committed to several of Mr. Pickwick’s friends because he quantified over them.” (Hall, 1960: 34)
Furthermore, even the proto-positivist Wittgenstein of the Tractatus held that logical propositions show or present the “scaffolding of the world.” “They presuppose that names have meaning, and that elementary propositions have sense.” (6.124) And such logical features of our language cannot somehow be undone by translation of sentences into a formal mode. We can try, as Carnap (1937: 308) did try, to treat metaphysical assertions as “disguised” syntactical statements, but if “by a transposed mode of speech we mean one in which in order to assert something about an object a, something corresponding is asserted about a [linguistic] object b which stands in a certain relation to the object a” we will be stuck trying to explain the nature of this “certain relation.” As Hall put it, “A correlation is a correlation; every wife has a husband, but to say that she is her husband is indeed beyond the proprieties even of a male-dominated society.” (1959: 56) One running from the law is likely to be no more comforted by “In English, the noun ‘Policeman’ is used for the object over there.” than he would be by “Don’t worry, there are no particular policemen around, only a few exemplifications of policemanship.” For good or ill, ontology does not recapitulate any philology that’s largely a matter of the translation of apparently referring terms into apparently non-referring idioms.
As far as I know, Quine didn’t specifically respond to either Hall or Alston on this matter (although he does mention the Alston paper a couple of times in Word and Object). But by the publication of his 1966 paper “Existence and Quantification” he had altered his thesis significantly. In this later work he writes, “To show that a theory assumes a given object, or objects of a given class, we have to show that the theory would be false if that object did not exist, or if that class were empty; hence that the theory requires that object, or members of that class in order to be true.” (Quine, 1966 : 93) And, while he still gives significant weight to particular locutions, he now acknowledges that such statements as “There are unicorns” may be misleading:
Thus far I have been playing down the difference between commonsense existence statements, as of rabbits and unicorns and philosophical existence statements, as of numbers and attributes. But there is also a curious difference between commonsense existence statements and philosophical ones that needs to be played up, and it is one that can be appreciated already right in among the rabbits. For let us reflect that a theory might accommodate all rabbit data and yet admit as values of its variables no rabbits or other bodies but only qualities, times, and places. The adherents of that theory…would have a sentence which, as a whole, had the same stimulus meaning as our sentence “There is a rabbit in the yard”; yet in the quantificational sense of the words they would have to deny that there is a rabbit in the yard or anywhere else.…When we…tell him there is a rabbit in the yard, he will know better than to demur on account of his theory; he will acquiesce on account of a known holophrastic relation of stimulus synonymy between our sentence and some sentence geared to his different universe. In practice he will even stoop to our idiom himself, both to facilitate communication and because of speech habits lingering from his own benighted youth. This he will do when the theoretical question is not at issue, just as we speak of the sun as rising….I grant that there are for him two senses of existence; but there is no confusion and the theoretical use is rather to be respected as literal and basic than deplored as a philosophical disorder. Similar remarks apply to our nominalist. He will agree that there are primes between 10 and 20, when we are talking arithmetic and not philosophy. When we turn to philosophy he will condone that usage as a mere manner of speaking, and offer the paraphrase. Similar remarks apply to us; many of our casual remarks in the “there are” form would want dusting up when our thoughts turn seriously ontological. Each time, if a point is made of it, the burden is of course on us to paraphrase or retract. (1966 : 98-99)
I believe this to be a particularly clear statement by Quine of a Hallian distinction between empirical and categorial assertions. It shows why, (to use Hall’s example) the fact that a Hegelian’s references to “The Absolute” can be paraphrased into language that does not seem on its face to require any such thing(s) does not entail that Hegelians would or should change their beliefs regarding what there is—even if they concur with the grammaticality of the restatement. Similarly, if a Quinian materialist allows proper names for various bodies or theoretical entities in his theory that an immaterialist finds she can paraphrase entirely into property talk, such paraphrasis need not move this materialist to endorse properties and dispense with muons. In fact, the Quinian’s reasons for resisting such a conversion would likely have nothing to do with linguistic flexibility at all, but would focus on different matters entirely—parsimony, explanatory value, and criteria of identity to name three. Thus, by 1966, Quine seems to have been in basic agreement
with Hall that to determine somebody’s ontological commitments, the safest course is simply to ask her—whether she is a Platonist, a Bradleyan or a materialist. And, naturally, if one does not have the particular philosopher around to answer this question, one may be required to do considerably more than simply provide the range of the quantifiers in some comprehensive restatement of the theory. An accurate assessment of commitments requires a determination of just which ‘entities’ are such that their non-existence would make the theory false. To return to Dickens, Quine would now be able to say (as he couldn’t in 1948), that no legitimately philosophical bound variable was intended by Dickens when he wrote things like “Some of Pickwick’s friends returned to London.” The upshot is that there’s nothing dispositive about “Some…” or “There are…” statements: philosophical analysis requires one to dig deeper, to investigate the total resources of the language/theory in question.