Although I am intrigued by Dasgupta’s proposal, Ifind his notion of being apt to
be grounded obscure. Well, most metaphysics is obscure; it is what it is. But still, I’d
like to have a better understanding of what Dasgupta means by“aptitude.”Dasgupta
draws an analogy between causation and grounding to help us grok the idea of
aptitude for grounding. Here’s one example of the kind of analogy he uses: the initial
state of the physical universe is uncaused but apt to be caused, whereas the causal
laws that govern the universe are neither caused nor even apt to be caused. Similarly,
there could be facts that are ungrounded but apt to be grounded—perhaps the
fundamental physical facts are like this—while there are other facts that are
ungrounded and not even apt to be grounded.
I’m a little uncomfortable with taking the notion of aptitude to be grounded as
primitive. So let’s see whether we can give a theory of it. We’ll begin by exploring
further the analogy Dasgupta employs between grounding and causation. Let’s set
aside whether the causal laws that govern the universe are apt to be caused: perhaps
they could, in principle, be caused by a god? Let’s focus on the initial state. Here are
some reasons to think that the initial state of the universe is apt to be caused even if in
fact uncaused. First, the initial state of the universe is one of the relata of the causal
relation, and one might think that if one is apt to cause, then one is apt to be caused.
(In general, however, one can’t always assume that being a relatum in a dyadic
relation makes one apt to be in either slot of a dyadic relation.) Second, there is a
possible world in which either it or a duplicate of the initial state in fact has a cause.
So the initial state of the universe is possibly caused, or something just like it is. Third,
the initial state of the universe is apt to be caused because it is an instance of a kind of
thing that has possible instances that are caused. (What is the relevant kind of thing?
I suppose it is events involving physical properties.)
Let’s see whether we can develop a heuristic for determining when facts are not apt
to be grounding. To that end, let’s consider parallel claims. Thefirst parallel claim is
that any fact that is a relatum in a grounding relation is apt to be grounded. Dasgupta
definitely cannot accept this claim. Facts about essences are relata of the grounding
relation. Suppose it is of the essence ofxthatF. This fact grounds the fact that either
it is of the essence ofxthatForP, for anyP.^42 (Disjunctions, I suppose, are always
grounded in their disjuncts.)
The second parallel claim is that an ungrounded but apt to be grounded fact has a
duplicate that is grounded, or is itself possibly grounded. This is harder to assess.
Perhaps the ungrounded is necessarily ungrounded. (Though, in general, a grounded
fact can have different grounds in different possible worlds. Why then must a grounded
fact be necessarily grounded or an ungrounded fact necessarily ungrounded?) If
something ungrounded is necessarily ungrounded, then the only way in which
ungrounded physical facts can be apt to be grounded (given the second parallel
(^42) Dasgupta (2016) distinguishes between mediated and non-mediated essence claims, and argues that
the former are not autonomous.