Thanks Pierre! Yeah, I did fumble over that a little, and in hindsight I would probably phrase that differently. Anyway, some initial commentary: among the things I like here - apart from Quine's wonderful writing style - is Quine's implicit 'autonomizing' of language, as it were, construing it as a domain unto it's own without needing to 'mirror' the world. To have ontological commitments to some kind of thing or another is just to
say that some kind of thing or another is. Ontology is relativized to 'the semantical plane', which, in turn, becomes decoupled from 'things' in order to operate autonomously. This impulse will also be taken up by Davidson - and, independently, by much of the 'continental tradition' - who will try to do away with 'representational' function of language, treating it on it's own terms.
I do wonder, though, about the 'voluntarist' - or perhaps better, rationalist - terms in which Quine frames this autonomy. In the following passage for example, the common denominator is that our ontological commitments follow as a result of our
saying that such and such 'are': "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 if - and I am inclined to make this move - 'saying' is simply one type of action among others, can we not implicitly commit ourselves to ontologies in our ways of
acting? Can one say, in a political vein, that a city commits itself to an ontology of home owners when it chooses to ignore the building of facilities to accommodate the homeless (or even, implement 'anti-homeless' features, like spikes on flat surfaces, as has been done in certain cities?).
To make a move like this of course is to once again 'substantialize' ontology in a way that Quine would probably find unpalatable, relativizing ontology not to language, but to a broader realm of 'significance' more generally. One consequence of making a move like this would also be to relativize language itself as one type of sense-making apparatus among others (which might include, to continue with our example, the structuring of movement and rest by our architectural and planning decisions, which can in turn structure the intelligibility of the populations who reside in a certain territory). Anyway, the point is: do our ontological commitments need to operate at the level of explicit enunciation (at the level of 'saying' that such and such is), or can they operate also at the level of implicit commitment, at the level of behavior, action, habit, and practice more generally? And if there is indeed a reason to make such a distinction, what in Quine would authorize it?
I also want to say that the above is in some way a response to
@Ciceronianus the White's question about why questions about being can matter so much. If the above is correct, and being cannot be delimited to the field of language alone, it might will be the case that our "ontological commitments" are normative through and through, not at the level of what we say, but at the level of what
do. One rather disastrous effect of 'deflationary ontologies' like Quine's might in fact be to disavow the fact that ontology operates in a manner that goes beyond mere intellectual debate, and flows right into the way in which power is both sustained and exercised across various domains of life, in which what one 'says' is not at all the issue. It's a nice, 'respectable' exercise of course, to confine questions of being to the parlor where we debate about Pegasus and so on, but some debates take place on the streets, conducted in a key other than language - perhaps sometimes violence.