are innumerably many relations that satisfy the same formal principles as those
grounding is purported to satisfy but are not relations of grounding.^67 Conversely, if
the generic grounding relation is either systematically variably polyadic or system-
atically variably axiomatic (recall section 2.3), then the case for grounding pluralism
(rather than monism) is quite strong.^68 The case for grounding pluralism (rather
than monism) is weaker the more similar the specific relations are to one another.^69
Moreover, it’s not clear to me that the thick pluralist can even say that the various
small-g grounding relations play similar roles, and it is clear that many of the
purportedly small-g grounding relations are formally dissimilar in many respects.
Both the thick and thin pluralists could say that each grounding relation induces a
relation of ontological superiority on its relata, and that this is what accounts for their
unity. I welcome this response—perhaps the grounder has a higher degree of being
than the grounded, or the groundedexists inthe grounder, and so forth. (This is
more or less roughly what I believe.) But since we are exploring considerations
independent of those of the previous sections, set this possibility aside.
Can the grounding pluralist simply take as ideologically primitive the relations
of determinate to determinable, of analogue instance to analogous property, and of
mere disjuncts to mere disjunction? And once they have taken these relations as
ideologically primitive, simply say that it is a brute fact the specific grounding
relation stand in one of thefirst two of these relations (rather than the third) to the
generic grounding relation?^70 It might make sense for a thin pluralist to say that the
specifics are determinates of the generic, whereas the thick pluralist should probably
say that they are analogue instances of the generic.^71
Frankly, I did little better than take these specification relations as ideologically
primitive in chapter 2, but I did do a little better! There, I said that in each case, if
some object or objects instantiate some specification of the generic, then it or they
instantiate the generic as well, and if some object or objects instantiate the generic,
then there is some specification such that this same object or these same objects
instantiate it too. But I also said that what distinguishes these three relations of
specification is that determinates and determinables are always equally natural,
(^67) Wilson (2014: 569–70) notes this, but also worries that there is not much formal unity to the various
small-g relations she discusses. 68
In a similar vein, if some putative grounding relations induce modal ties between their relata, whereas
others do not, the case for pluralism is strengthened. In this context, consider that deRosset (2010), Fine
(2012a), and Rosen (2010) all endorse the claim that ifxgroundsy, then in any world in whichxexists,x
grounds 69 y, while Schaffer (2010b) denies this claim.
Contra Cameron (2014: 52), even if all grounding relations had the same formal features, this would
not seal the deal against grounding pluralism. Fine’s (2012a: 39–40) case for pluralism is based on an
intuition about a single case in which one kind of grounding allegedly is absent, rather than via an appeal to
differences in formal features. I do not know how widely this intuition is shared. 70
Do not assimilate determinables to disjunctions of their determinates, for that is a bad theory of
determinables. See Wilson (2012). 71
Bliss and Trogdon (2014) explore whether grounding is a determinable. Cameron (2014) explores
whether grounding is analogous.