‘No truth without truthmakers.’ This is the slogan of the truthmaker movement in metaphysics. The idea seems to be that a truth bearer, e.g. a proposition, could only be true if there were something to make it true. It’s not clear what kind of thing this something should be. Prominent candidates probably include states of affairs or facts, understood in some kind of metaphysically weighty way. There are, no doubt, other candidates as well.
I don’t consider myself to be anything like an expert on truthmaking. I do have some interest in truth, however, so I’m starting to think that I should become better informed about truthmaking since at least some people seem to think that truthmaking is related to truth.
I myself am skeptical on this point. (I am not the first to express skepticism; I believe, for instance, that David Lewis did as well.) Let’s consider a particular proposition:
(a) Some apples are red.
When restricted to (a), the truthmaking doctrine seems to amount to saying that if some apples are indeed red, then this fundamentally consists in there being something, e.g. a particular red-apple-state-of-affairs. It doesn’t appear that we need to bring in truth here. And, this example is, of course, representative. When we look at the situation on a case-by-case basis, truth drops out of the picture. (The point is that we only need the concept of truth as a device of generalization, in the way that Quine taught us.)
We can try to get at this skepticism in another way. The truthmaking doctrine seems to be roughly equivalent to the suggestion that any theoretical commitment is at base an ontological commitment. Again, this does not obviously look like a claim that is distinctively about truth.
I’ll end this post by expressing some concern about truthmaking. Suppose that the rephrase above is adequate so that the truthmaking doctrine is simply the idea that any theoretical commitment is at base an ontological commitment. The rephrase doesn’t strike me as especially compelling. I don’t know why any theoretical commitment should be an ontological commitment. Of course, any theoretical commitment (that isn’t a limiting case, e.g. a tautology) should be a genuine commitment. The one making the commitment should be ‘on the hook’ in some sort of way. But, it’s not clear why commitments need be ontological at base to be genuine.
Moreover, even if all theoretical commitments involve ontological commitments, that is not enough for all theoretical commitments to be ontological commitments at base. For instance, making claims about what is possible is a kind of theoretical commitment. Even if we suppose that such claims involve an ontological commitment to possibilities, it is not at all clear to me that this ontological commitment is more fundamental. It seems plausible enough that the existence of possibilities is fundamentally explained by what is possible, rather than vice-versa.
Hi Ben,
I’m wondering if maybe you’re asking more of the truthmaker principle than its advocates intend for it to supply. It should be uncontroversial that the truthmaker principle (i.e. the principle that x is true iff something exists to make x true) does not yield a substantive definition or theory of truth: the truth predicate appears on both sides of the biconditional, and that smacks of circularity. But it also strikes me as uncontroversial that the principle is nonetheless ‘related to truth.’ You say you are skeptical of this, but your proposed counterexample characterizes the relationship between the truth of ‘Some apples are red’ and a particular red-apple-state-of-affairs in terms of ‘fundamentally consists in’. I don’t think the advocate of the truthmaker principle is committed to such a strong formulation of the relation between a truthbearer and its truthmaker. Again, the principle does not say what it is to be true, it simply says that a truth obtains just in case it has a truthmaker.
So I guess I’m missing how truthmaking is unrelated to truth. You say that ‘truth drops out of the picture,’ but why should that be the case? The proposition expressed by the sentence ‘JD is writing a blog comment at 1100 on 8 Jan 2013’ is either true or false. It is both necessary and sufficient for it’s being true that there is some event (state of affairs, etc) which involves JD’s writing a blog comment at the specified time. This is not exactly earth shattering, but it certainly specifies a relation between truth and the world. Perhaps, however, you’re working with a deflationary conception of truth? I’m not sure what to say about that. I guess it’s not clear to me that a deflationist can even accommodate the truthmaker principle, in which case a rejection of that principle should proceed independently of deflationism. But I don’t pretend to have a firm grip on the deflationary theory…
Regarding your point about explaining possibilities: if you think the existence of possibilities is not fundamental (contra Lewis), then I assume you endorse some version of actualism; but, of course, the actualist does not think that the existence of possibilities is fundamentally explained by what is possible, since she thinks that what is possible is fundamentally explained by what is actual. So it looks like, with respect to this example, you are still committed to an ontological explanation.
Hi Joseph,
Thanks for this (and Happy New Year!).
A few things:
(A) The ‘fundamentally consists’ is only supposed to signal an asymmetric explanatory relationship. (You seem to acknowledge such a relationship when you speak of “ontological explanation” rather than “ontological consequence.) Without the asymmetric explanatory relationship, truth-making seems to lose its bite. Consider a kind of clever phenomenalist who thinks that there are external objects but also thinks that what it is for there to be external objects is for one’s perceptual or experience to be a certain way were certain conditions to obtain. Or consider a neo-Fregean who thinks that there are numbers but thinks that there being numbers is nothing above and beyond there being facts about equinumerosity between sets. These kinds of positions have a deflationary or anti-realist flavor, but they are strictly compatible with a view that says only that theoretical commitments must have ontological consequences. You need explanatory asymmetry (i.e. ontological explanation rather than merely ontological consequences) if you want to criticize these views on truth-making grounds. Indeed, I think you need asymmetry for truth-making to be interesting. (Of course, my point at the end is just that asymmetry is not something that opponents should readily concede.)
(B) Putting the issue from (A) aside, my example of truth-making is this:
Some apple’s being red requires there being some thing that suffices for some apple’s being red (e.g. there being a particular red apple state of affairs).
There is no mention of truth in this example. That’s the point. Truth is either redundant or drops out completely. If we want to generalize the example to a principle, then we need some way of stating an arbitrary fact (to put in the slot currently being occupied by “some apple’s being red”). The concept of truth gives us a way of doing that, but that makes it seem like the truth-maker principle isn’t really about truth. The concept of truth is simply giving us a way to express something that we could have written schematically without truth as:
(*) p requires there being some thing (e.g. object, state of affairs, etc.) sufficient for p.
where for ‘p’, we insert a nominalized sentence. The schema doesn’t say anything about truth either. Insofar as (*) captures the commitment to truth-making, truth-making isn’t about truth.
(C) It is pretty clear that the deflationists can accommodate truth-making if all that’s required is adopting (*). Deflationism has variations, but roughly it is the view that there isn’t much more to the nature of truth than is captured by the T-Schema, which is a kind of ‘conceptual schema’ (i.e. the analogue of a ‘conceptual truth’):
(T-Schema) [p] is true iff p (where one substitutes a use of the proposition [p] for ‘p’)
Inflationism requires saying more about the nature of truth than the T-Schema. But, (*) doesn’t say anything about the nature of truth. So, deflationists can accept (*).
(D )It’s worth pointing out that deflationists wouldn’t say that the truth of a truth-bearer doesn’t depend on the world. For instance, it follows from the T-Schema that the truth of depends on whether it is raining, which presumably, is a feature of the world. And, it is open to the deflationist to think that whether it is raining or not is either settled by or at least has consequences for what there is. In other words, the world and even ontological dependence of truth is not sufficient to refute deflationism. And, insofar as deflationism is a competitor to correspondence theories of truth, this shows that world/ontological dependence is not sufficient to establish a correspondence theory of truth.
Hi Ben, Happy New Year to you too! Here are some additional thoughts (based on your headings):
(A) Yes, of course the truthmaker principle expresses an asymmetric relation, my point was simply that ‘fundamentally consists in’ is a stronger asymmetric relation than the principle seems to warrant.
(B) The absence of the truth predicate in your example is no consequence of the truthmaker principle, it is a consequence of your omitting truth bearers from the example. ‘Some apple’s being red’ is not a truth bearer, it’s a fact or a state of affairs or an event (depending on your ontology). One would only think that the specific example, and the generalized (*), capture the commitment to truthmaking if one thought that p’s being true was defined in terms of p’s obtaining. This is why it struck me that you must be assuming deflationism. But obviously a truthmaker theorist who was working with a more robust conception of truth would reject the adequacy of (*) as a truthmaker principle.
So…are you assuming deflationism about truth? I’m certainly not, and so if you are, I think we may be talking past one another with respect to truthmaking.
Hi Joseph,
Thanks for pressing this further.
(A) OK. (By ‘fundamentally’, I meant ‘more fundamentally’ not ‘not most fundamentally,’ so that’s probably just miscommunication on my part.)
(B) I’m neither assuming deflationism nor assuming inflationism. The idea is that truth-making doesn’t bear on this debate (or any other debate about the nature of truth) and vice-versa. Both deflationism and inflationism are consistent with truth-making because truth-making doesn’t involve taking on any controversial commitments about the nature of truth.
I’m a little unclear in this dialogue what the truth-maker principle is supposed to be. You seem to think that it must essentially involve truth bearers, which I wasn’t assuming. But let’s go with you since I don’t really know the literature as well as you. Suppose the truth-maker principle is TM:
TM: If [p] is true, it is because there is a truthmaker that suffices for it.
Note that TM can be effectively broken down into TS and Ontology-First:
TS: Something explains (constitutes? makes-for?) the truth of [p] only to the extent that it explains (constitutes? makes-for?) p*.
(Example: Something explains the truth of [Some apples are red] only to the extent that it explains some apples’ being red.)
and
Ontology-First: If p, p* is because there is a truthmaker that suffices for it.
(Example: If some apples are red, some apples’ being red is because there is a truthmaker that suffices for it.)
(Note on terminology: ‘[p]‘ stands for a truth bearer. You are to substitute the nominalization of a use of [p] for ‘p*’ and a use of [p] for ‘p’.)
Nearly everybody (deflationists and inflationists) should accept TS (and I certainly do). That’s pretty much because everybody should accept the T-Schema, and nobody should think that the left-hand side of the T-Schema (the ascription of truth to [p]) should explain the right-hand side of the T-Schema (the use of [p]). For instance, it just isn’t very natural to think that snow is white because of some (abstract?) proposition or sentence-type has some property or even because some written or spoken token sentence does.
If TS is independently acceptable and TM breaks down into TS and Ontology-First, then the real meat of TM must come by way of Ontology-First. And, indeed, Ontology-First is the controversial claim. It should be unacceptable to people who accept ontological consequences without ontological explanations. For instance, I don’t want to deny that there are (obtaining) states of affairs. But, I don’t think that some apples are red because there is a red apple state of affairs. Indeed, I might even think that there is a(n) (obtaining) red apple state of affairs because some apples are red (i.e. the obtaining of the state of affairs is a pleonastic add-on to the simpler fact). So, if truthmakers are (obtaining) states of affairs, then I’m not really on board. And, while I think that unicorn-like creatures are possible and even that there are possibilities in which there are unicorn-like creatures, I’m not tempted to think that unicorn-like creatures are possible because there are these possibilities or (even worse) because there are possible unicorns. So, it is clearly a kind of ontology-first thinking that I’m disturbed by.
Now notice Ontology-First–the controversial claim–doesn’t really concern truth or truth bearers. So, the real meat of the truth-maker principle doesn’t seem to have anything to do with truth or truth bearers. That’s the point that I’ve been trying to make (while indicating that I am not sympathetic to Ontology-First). A deflationist has no more reason to accept or reject TM than an inflationist (or any other reasonable theorist about truth). And my skepticism about the truth-making principle is not driven by deflationism (which incidentally, I reject).
The bottom line: Advocates of TM might be presupposing the T-Schema, but nobody thinks that the T-Schema is at all in question (modulo issues about the semantic paradoxes). So, that presupposition is not enlightening. And, TM sheds no further light on the nature of truth or truth bearers.