This week we discussed Schaffer’s ‘Truthmaker Commitments’, which is a critique of a certain view of ontological commitment associated with Armstrong and with Ross Cameron. It’s worth reading Cameron’s reply to Schaffer as well. No presentation to put here as yet; but the papers themselves are quite clear and concise. Various thoughts follow:
Gonzalo raised an issue about the notion of ‘implication’ being used by the quantifier view. If a theory’s ontological commitments are what it says exists, as Schaffer glosses it, then a theory is committed to certain entailments of the particular sentences or propositions which explicitly make it up. But a theory’s commitment should not include necessary existents, whose existence is entailed by any set of sentences. Perhaps, though, this is a problem more with this gloss on the quantifier view than an objection to Quine’s own view.
Schaffer could have said more in defence of the quantifier view against the ‘linguisticism’ charge. The quantifier view by itself decides no ontological questions, if correct. All it does is to say how the true theory is connected with the true ontology. Given the quantifier view, what our ontology is is determined by which theory is true. The quantifier view is compatible both with theories on which there are only particles arranged tablewise, and with theories on which there are particles arranged tablewise, and there are tables, and the tables are not identical with any particles or arrangments of particles. So the quantifier view does not itself determine what exists.
However, even granting this stronger point Schaffer could have made, I think Cameron can stick with the reply he actually gives. This is that the objectionable linguisticism is in the thought that the syntactical structure of the true theory of the world, as expressed in our language, must mirror the ontological structure of reality. Cameron presumably thinks that the process of translating the true theory of the world from Ontologese to English will involve generating some true English existence-assertions where there were no isomorphic existence-assertions in Ontologese. I think this claim does move the debate forward and I’m tempted to agree.
There is another way of responding to this point on behalf of the quantifier theorist. It is that we should stick with English, but English analysed in whatever way it takes to get the true underlying quantificational structure. That is, we could give an analysis of English according to which it has a quantificational structure exactly isomorphic to the quantificational structure of Ontologese. Perhaps higher-order quantification will make this possible for Schaffer; perhaps, as Gonzalo suggested, we might also have to give up eg the association between classical first-order individuals and objects. It looks like the issue between Schaffer and Cameron here is how far we ought to complicate the semantic and logical structure of English in our analyses, and how much we should ascribe to quantifier variance.
Gonzalo also pointed out that some mereological nihilists might be committed to tables even on the truthmaker view. For if they allow many-one quantification, eg ‘the particles=the table’ then even the truthmaker view is committed to tables. So the friend of the quantifier view has a stronger case here than Schaffer realizes.
We thought Cameron’s reply to Schaffer on the ‘denumerably many electrons’ objection was right.
There was more-or-less a consensus that the grounding view of truthmaking (Tgro) does best of the three accounts Schaffer discusses. He dismisses the use of Tgro by a truthmaker theorist, saying ‘Tgro does not concern what there is, but only concerns what is fundamental.’ It seems to me that this can be easily resisted by a truthmaker theorist, as it is by Cameron.
Schaffer’s dilemma for the truthmaker theorist – if existential quantification is not generally ontologically committal, it is mysterious why the specific existential quantification in TNec is committal – didn’t seem persuasive at all. The truthmaker theorist claims to have presented arguments for his view; if good, those arguments are ipso facto good arguments for a distinction between existential quantification in general and existential quantification over truthmakers. Given this, horn A of the dilemma seems to come down to prejudice that there should be no distinction between types of existential quantification.
Much of the issue between Schaffer and Cameron does seem to come down to what we should mean by ‘ontological commitment’. They each offer us clear enough candidate meanings for the phrase – Schaffer takes it to mean ‘what the theory says exists’ while Cameron proposes ‘the demands that the theory places on ontology’. As Cameron says in his reply, it would be dogmatic for Schaffer to simply refuse to recognise a potential distinction here. And once we do recognise the potential distinction, I think there is a case to be made for both candidate meanings for ‘ontological commitment’.
One moral we might draw from this is that philosophers have sometimes been at cross-purposes over a (comparatively new) technical term. It may have been legitimately used in both senses in the past; if so, it might be better to drop talk of ontological commitment altogether. The idea of commitment is still a useful one, for it makes sense of implied or presupposed existence claims, but perhaps we can replace the troublesome term ‘ontological commitment’ with a pair like ‘derivative commitments’ and ‘underlying commitments’.