to be perfectly natural. But I have two concerns about this. Thisfirst might be
somewhat question-begging in this context, but it sounds odd that an ontological
category cuts nature less at the joints than other kinds. Second, and also troubling, we
have no guarantee that there will be properties that uniquely balance generality and
naturalness. So, on a view like this, there simply might not be a fact of the matter
about which kinds are the ontological categories.
The cut-off problem is ameta-ontologicalproblem, and therefore should not be
conflated with thefirst-order problem of determining when some predicate corres-
ponds to an ontological category. It might be hard to determine when some category
is an ontological category. (For example, ismaterial objectanontological categoryor
just a very general natural kind?) By defining“ontological category”as“highly
general natural kind,”oneensuresthat the notion of an ontological category is
vague, just as“wealthy person”is vague.
This consequence of understanding ontological categories as highly general kinds
of things was clearly seen by Bolzano (2014a: 401–2), who explicitly notes the cut-off
problem, draws the conclusion that the concept of a category is therefore vague, and
explicitly states that there can therefore be no determinate number of categories.
I’ll admit tofinding this disquieting, but if the project of determining what it is to
be an ontological category is purely descriptive, one might be willing to set this
disquiet aside. But if the concept of an ontological category is vague, it might not be
suited for realworkin fundamental metaphysics. How could, for example, a vague
notion of an ontological category be of use in formulating and justifying a restricted
principle of recombination? A principle formulated with its help would be vague as
well. Is it vague which states of affairs are metaphysically possible? If we care about
functionality, the cut-off problem should concern us.
One might hope to avoid the cut-off problem by adding additional conditions
beyond generality. For example, perhaps an ontological category is a highly general
natural kind K such that, necessarily, anyxthat falls under K essentially falls under K.
However, there are potential problems with this maneuver. First, not every plausible
ontological system implies that objects belong to their ontological categories essen-
tially. For example, Bricker (2001, 2006) defends the view thatthe actualandthe
merely possibleare two distinct ontological categories. Many (perhaps all) things that
are actual could have been merely possible, and everything that is merely possible
could have been actual. This definition of“ontological category”rules out plausible
ontologies such as Bricker’sbyfiat. Maybe that’sOK—it depends on whether
neutrality is a constraint on our investigations here. (And we’ll have a lot more to
say about whether things belong to their ontological categories essentially in
chapter 9.) Second, it might be the case that electrons are essentially electrons.^27
Being an electron is a highly natural property. Butelectronstill doesn’t seem to
(^27) Compare with Westerhoff (2005: 64).