Wednesday, April 22, 2009

Explanation and truthmakers

Say that a true proposition is ontologically ungrounded provided that it is not true because of any combination of what things exist, how things that exist exist and what things do not exist. In particular, an ontologically ungrounded proposition lacks a truthmaker.

It would be really nice if one could argue that:

  1. No contingent ontologically ungrounded proposition has an explanation.
For if we could show (1), then we would have a neat argument against those presentist views on which truths about the past are ontologically ungrounded (e.g., Trenton Merricks). For some truths about the past have explanations: that World War II happened is explained by Germany's humiliation after World War I, say. We might also have an argument against the version of Molinism on which conditionals of free will are ontologically ungrounded. For some conditionals can probably be explained in terms of others: That George in circumstances C would greedily accept the bribe explains that he would accept the bribe.

I don't really have a good argument for (1), apart from the fact that it rings true. But, then again, I am no friend of ontologically ungrounded propositions, so my intuitions aren't the best judge here.

However, while (1) is hard to argue for, I think the following is pretty easy to argue for:

  1. No contingent ontologically ungrounded proposition has a causal explanation.
After all, to causally explain a proposition is to give a cause of that in the world which grounds its truth. Causation is a relation between things in the world (possibly including absences, too), while causal explanation is a relation between propositions that depends on causation. I don't know whether (2) is enough to refute Molinism. But it does seem enough to refute the ungrounded past view. That World War II happened is, after all, causally explained by the fact that Germany was humiliated.

Perhaps the presentist will say that World War II was causally explained by the humiliation. But when? As I understand it, when World War II started, the humiliation was already gone. Thus, at no time were World War II and the humiliation both present. Hence, at every time we have a causal explanation relation between two propositions at most one of which reports an ontologically grounded fact. But how can there be a causal explanation relation between two propositions one of which is ontologically ungrounded?

I suppose this is a variation of the old causation objection to presentism.

8 comments:

Maynard said...

Dr. Pruss:

Hello. Thanks for maintaining a fascinating blog. I was wondering if I might temporarily step into it and raise a question. I beg your pardon ahead of time because it's not strictly related to the issue(s) at hand.

My question specifically concerns truthmaker theory on which the grounding objection is putatively based and the alleged relation between truthmakers and truthbearers. Suppose we define truthmaker theory as holding that true propositions (truthbearers) are made true by certain entities (truthmakers). And suppose we identify a truthmaker as that entity in virtue of which a proposition is made true. Then, for any true proposition P, there’s a corresponding entity (or state of affairs) Q which makes P possess the truth-value “true.”

Bearing in mind that many truthmaker theorists deny truthmaker maximalism, along with the supposition that the relation between truthmakers and truthbearers is a causal one, here’s my question: if truthmaker theory has as its content a central proposition, then, according to the theory, does that proposition require a truthmaker? But if this is the case, then truthmaker theory seems to presuppose itself in its very formulation. And how would one avoid engendering an infinite regress of truthmakers? Perhaps a truthmaker theorist will suggest that such a proposition (the central proposition) is among those which don’t require a truthmaker, but I wonder if this reply, while possible, is simply ad hoc and question-begging.

I’m not as acquainted with truthmaker theory as I’d prefer, so it may very well be that my question rests upon a confusion.

Thanks for your time.

Peace,

-- Marc

Alexander R Pruss said...

Thanks for the interesting question.

I don't know that anybody thinks the truthmaker-truthbearer relationship is causal.

I am not sure I see why truthmaker theory's requiring a truthmaker is a circularity. Consider this theory: "All theories are produced by people." This theory, according to itself, requires an author. And that's not a problem.

As for regress, I don't see how one gets off the ground. Let T be the central proposition of truthmaker theory. T has some entity, E, as a truthmaker. But how do I get the second step of the regress? Do I ask what the truthmaker of "E exists" is? That's an easy question to answer. The truthmaker of "E exists" is E. Maybe one should ask: "What is the truthmaker of the proposition that E makes T true?" Maybe E and T taken together?

James said...

Hmmm...How would you define a causal explanation?

Alexander R Pruss said...

It's an explanation of something in terms of the exercise of causation, I guess.

James said...

So, for instance (I choose this example because I'm interested in how your post might or might not relate to Molinism), would you say that Peter's freely doing A is causally explained by the fact that Peter freely chose to do A?

Maynard said...

Dr. Pruss:

Thanks for considering my question.

I realize I wasn't clear about how most philosophers maintain that the relation between a given truthmaker and a truthbearer isn't causal. Robert Adams and William Hasker, however, have offered comments which suggest that they do think the relation is causal, but I can't say whether they still affirm that.

You're right about the circularity problem, and how it's not a problem. The problem with which I'm primarily concerned is the infinite regress issue, which you formulated as I conceive of it. But perhaps mine is simply a misconception.

Following your rendering, let

T = true propositions require truthmakers.

If T is true, T requires a truthmaker.

If we let A be T's truthmaker, then it's either true or false that A is T's truthmaker.

If it's true that A is T's truthmaker, then this propositions itself will require a truthmaker. And so forth.

Perhaps I'm missing something, though.

Peace,

-- Marc

Alexander R Pruss said...

Perhaps A and T are, together, the truthmaker of the proposition that A is T's truthmaker? Yes, there will be a regress, but it may not be any different from the virtuous truth regress (p is true, it is true that p is true, it is true that it is true that p is true, ...).

Maynard said...

Forgive me for continuing to dwell on this . . . and thank you again for engaging my question.

If the relation between the regress of truthbearers and truthmakers is analogous or identical to the relation between the propositions of the truth regress, then the relation involved in truthmaker theory would be, as you suggested, benign, and maybe even virtuous. Once we move beyond a truthbearer and its truthmaker--i.e., once it's established, say, that some truthmaker A makes some truthbearer T true--we seem to embark on such a regress. There appears to be an important sense in which A and T are "self-contained."

Peace,

-- Marc