The Fragmentation of Being

some sense one of them is a metaphysical primitive is a different story. Neither
proponent of their relation of ground might take their relation to be metaphysically
primitive, but also neither proponent will take their opponent’s relation as meta-
physically primitive.
At this point, an analogy with quantifier variance (discussed in section 1.4) seems
apt. Recall that quantifier variance was supposedly motivated by the apparent inter-
translatability of ontological systems. The compositional universalist thinks that
whenever there are somexs, there is some object composed of thosexs. The
compositional nihilist thinks that whenever there are somexs, there are only those
xs and nothing further. According to the quantifier variantist, the compositional
universalist can understand the existential quantifier of the compositional nihilist as
being a restriction of the universalist’s quantifier. (In much the same way as the
entity-grounder can understand the relation of fact grounding as a restriction on
entity grounding.) And the compositional nihilist can understand the universalist
quantifier in a variety of ways, such as via an“according to thefiction that compos-
ition occurs”operator. (In much the same way as the fact-grounder has a plurality of
ways of understanding entity grounding in fact-grounding terms.)
Once we see the analogy with quantifier variance, we also see that the potential
problem for the monist is more general. Consider, for example, entity grounding. For
every pair of entities related by grounding, there is a pair of their singletons related
by a different relation, grounding. That grounding could be the content of an
ideologically primitive predicate is far-fetched, but not impossible, and no more
far-fetched than that the property of being grue might be the content of such a
predicate. But regardless of whether such an ideologically primitive predicate is
possible, it is clear that there is something goofy about grounding, which suggests
that grounding is doing better than grounding on some metaphysically important
scale. The question then is whether the grounding monist can account for that scale
wholly in terms of grounding, or whether she instead should embrace another
structuring primitive, such as degrees of being.
One way a foe of quantifier variance can respond to that view is by employing a
notion like naturalness or degrees of being and holding that one of the quantifiers
under discussion is more natural than the other and hence better suited for use in
metaphysical enquiry. How should the proponent of grounding reject quantifier
variantism, if she is so inclined? If she can account for ultimate properties (discussed
in section 8.2), maybe these could be appealed to. Perhapsbeingor its modes of being
are ultimate properties. But if she can’t, she’ll need some other structuring notion
such as naturalness or degrees of being.
Similarly, the proponent of either entity grounding or fact grounding could say
that one of these grounding relations is more natural than the other and hence better
suited for use in metaphysical inquiry. The scale I prefer is the scale of being:
grounding has more being than grounding*, regardless of whether grounding is
perfectly natural. In general, grounding must be more natural than any putatively

 BEING AND GROUND