Bo Meinertsen writes (via e-mail):
I wrote to you in the summer of 2003 regarding Bradley's regress. At that point I was writing up my Ph.D. on facts at Leeds, which I submitted 30 Sept. (degree awarded 19 Jan.). Anyway, a chapter was on truth-makers, where, among other things, I critisized the 'minimalism' of Julian Dodd, though not at any great length. You said you'd written a paper on this topic. As I'm looking at the matter again, I'd be really grateful if you'd send me a copy of this paper.
You write at an opportune time, Bo. I have just uploaded a draft on Trope Theory and Bradley's Regress which would undoubtedly be improved by your comments. Perhaps you could take a look at it. Imagine that you are a referee for a top analytic journal and then pronounce your opinion on publishability.
As for Julian Dodd's minimalism, I think what I said was that a section of my book, A Paradigm Theory of Existence (Kluwer, 2002), is devoted to this topic. What follows is Chapter VI, Section 5 of PTE. I'd love to hear your comments on it, and I hope it is helpful to you.
We have been arguing that (i) the truth-maker principle is sound; (ii) only facts can be truth-makers; therefore, (iii) facts exist. But this argument can be run in reverse: Since there are no facts, and since only facts can be truth-makers, the truth-maker principle is unsound. This is essentially the line taken by Julian Dodd. Our ‘fact’ is short for ‘concrete fact.’ But Dodd uses ‘fact’ as interchangeable with ‘true proposition.’ To avoid confusion, we will employ ‘state of affairs’ for the duration of this section. A state of affairs is a concrete fact.
As I read him, Dodd’s critique of the very idea of truth-making consists of two main claims. The first is that, even if there are both particulars and universals, and even if particulars instantiate universals, we are not forced to countenance states of affairs. Thus if a exists and F-ness exists, and a instantiates F-ness, it does not follow that there is in addition the state of affairs, a’s being F. A similar sentiment is expressed by David Lewis: “If I were committed to universals myself, I would be an Ostrich Realist: I would think it was just true, without benefit of truth-makers, that a particular instantiates a universal.” Dodd’s second claim is that states of affairs are such seriously problematic entities that no theory that invokes them has any right to our attention.
As for the first claim, Dodd will presumably admit that there is a difference between the sum a + F-ness and a’s instantiating F-ness. This is something to which all must agree. If a exists and F-ness exists, it does not follow that a instantiates F-ness. So there is a difference between a + F-ness and a’s instantiating F-ness. But this is just a brute difference, Dodd seems to be saying, not a difference that needs to be explained by positing a third sort of entity, a state of affairs, in addition to a and F-ness. Even if the truth that a is F ontologically commits us to a and to F-ness, it does not commit us to anything in the world that connects a and F-ness. The ‘is’ in ‘a is F’ has no ontological correlate. Thus we are not committed to an instantiation relation or to a nonrelational tie of instantiation. And not being committed to any such connector, we are not committed to the state of affairs a’s being F assuming that this is the product of instantiation’s connecting of a and F-ness. To this one might respond that it is not instantiation that ties a to F-ness, but the state of affairs itself, and that we can dispense with instantiation (and perhaps must dispense with it in the face of Bradley’s regress). But this will not satisfy Dodd either, since he refuses to admit that we need anything in the world to connect a and F-ness. In the world there is at most a and F-ness, but there is nothing in the world that corresponds to the truth that a is F. Truth does not require an ontological ground. Not only is there nothing corresponding to the copula ‘is,’ there is nothing corresponding to the whole sentence, ‘a is F.’
But if nothing in the world connects a and F-ness when a instantiates F- ness, then what is the difference between a’s instantiating F-ness and a’s not instantiating F-ness? What does the difference consist in? Presumably, Dodd must say that there is a difference, but that it does not consist in anything. This however simply begs the question against the truth-maker principle. For truth- makers are introduced precisely to satisfy the felt need for an explanation of the difference in question. Dodd hasn’t succeeded in refuting the truth-maker principle; all he has done so far is to reject it.
Pointing this out, we of course do not succeed in refuting the Ostrich Realist; we merely highlight the deep conflict of intuitions at the root of the disagreement. The realist about truth cannot shake the sense that truth requires an ontological ground, a sense simply unshared by an Ostrich such as Dodd. With respect to Dodd’s first claim, then, the upshot appears to be a standoff.
We now examine Dodd’s second claim which is essentially that the positing of states of affairs cannot serve as an adequate explanation of how particulars instantiate universals. Whereas Dodd’s first claim is that we have been given no compelling reason to posit states of affairs, his second claim is that nothing is explained even if we do posit them. This is a much more serious objection. If sound, it would appear to refute the truth-maker project.
Given that a instantiates F-ness, Dodd will say that this instantiation is just a brute datum. The truth-maker theorist, however, cannot rest content with this. He feels that there must be something in the world that explains this instantiation of a universal by a particular. So he posits a state of affairs in which a and F-ness are brought together. He posits a state of affairs which just is a’s instantiating of F-ness. For Dodd, however, this is a bogus explanation.
Dodd’s argument is not entirely clear, but it perhaps amounts to something like the following dilemma. Either (L1) states of affairs are composed of constituents that are ontologically more basic than states of affairs, or (L2) states of affairs are ontologically primary, and their constituents are mere abstractions from them. (L1) faces Bradley’s regress and the unity problem, something we will discuss in great detail in the following chapter. Given that a and F-ness are contingently connected, what connects them? The instantiation relation? But how can adding a further constituent establish unity of constituents? Dodd’s point is that if states of affairs face the unity problem – the problem of explaining how they differ from a mere set or sum of constituents – then invoking states of affairs can do nothing to explain how a particular instantiates a universal. It is essentially the same problem all over again. If it is unclear how a particular instantiates a universal, then this cannot be clarified by positing an entity, a state of affairs, concerning the constituents of which it is unclear how they form a unity. In other words, if you say that the difference between a + F- ness and a’s instantiating F-ness is the difference between two items and the same two items connected within a state of affairs, this explanation succeeds only if it is clear how the two items – a and F-ness – are connected within the state of affairs. Since the latter is not clear, to invoke states of affairs to explain propositional truth is to give a bogus explanation.
This throws us onto the other horn of the dilemma, (L2), according to which states of affairs are ontologically basic, and their constituents are mere abstractions from them. This would appear to avoid the unity problem. If a and F-ness are mere abstractions from a primary unity, a’s being F, then there is presumably no problem about what holds them together. But then how could the positing of a state of affairs so conceived explain or ground propositional truth? The truth that a is F is contingent; hence the togetherness of a and F-ness in a’s being F must be contingent. This however leads us straight back to the unity problem which arises because of the contingency of the togetherness of a and F- ness. Since a and F-ness can exist without forming a unity, they cannot be mere abstractions from some ontological primary unity: they are the ontological atoms, the ‘building blocks,’ out of which states of affairs are constructed. States of affairs must therefore be ontologically dependent on the items that contingently form their constituents. It is not states of affairs, but their constituents, that are ontologically basic.
Although Dodd is on to a very serious problem for states of affairs theorists, a problem to be more thoroughly discussed in the next chapter, he has given us no good reason to abandon truth-making and truth-makers. We noted above that his first claim merely begs the question against the truth-maker theorist. Standing pat on our realist intuitions, we are within our epistemic rights in taking the truth-maker principle to show that there must be truth-making states of affairs. And the fact that the ‘compositional’ and ‘noncompositional’ conceptions of states of affairs alluded to in (L1) and (L2) above are faulty does not by a long shot prove that there is no explanatorily adequate conception of states of affairs. For there could be a third conception of states of affairs. Working out this third conception is a task for the following chapter.
Comments
You can follow this conversation by subscribing to the comment feed for this post.