On one line of thought, we do grasp, at least dimly, categorial distinctions between
certain entities. As noted in earlier chapters, some philosophers of the phenomeno-
logical tradition, such as Brentano, Husserl, Meinong, and Heidegger, are sympa-
thetic to this thought, which is why, on their view, phenomenological and ontological
investigations are importantly connected. I’m somewhat sympathetic to this as well,
which leads me to think that, to some extent, the project is also descriptive: we should
account for categorial differences antecedently recognized in our theory of onto-
logical categories. One way in which these pre-theoretically recognized categorial
differences manifest themselves is in our unhappiness with linguistic utterances
called, appropriately, category mistakes. Consider, for example,“the number two is
hungry.”This is a very bad sentence, and its badness does not consist simply in its
metaphysical impossibility. We’ll discuss the nature of category mistakes more fully
in section 4.5.
If we antecedently grasp categorial differences, then there is a question of whether
this inchoate phenomenon of categorial difference can be aptly described. But how
could we grasp categorial differences without having also some sort of understanding
about what some of the categories are? Taking seriously the descriptive aspect of the
meta-ontological project seems to require downplaying neutrality over which list of
ontological categories is correct.
Functionality: To what extent is the project of determining what it is to be an
ontological category an attempt to explicate a theoretical role or an explanatory job
that ontological categories play?As noted earlier, the notion of an ontological category is useful only if theoretic-
ally fruitful. Accordingly, we might proceed not by asking what ontological
categories are but rather what we need them to be. And we assess this by asking
what theoretical role we might want ontological categories to perform. On this way
of thinking, if the notion of an ontological category does no theoretical work—if it
is, so to speak, purely descriptive rather than partly functional—then the meta-
ontological project has no philosophical payoff. Why talk about ontological
categories then? I suspect that this was the perspective of David Lewis who,
despite his status as one of the greatest metaphysicians of all time, seemed to
have so little use for the notion of an ontological category that such a phrase
rarely if ever appeared in his writings. (I think it is appropriate to say that
Lewis (1986, 1991) had, or inclined towards,a two-category ontology consisting
of ur-elements and classes even though he did not put weight on the label
“ontological category.”)
A notion is theoretically fruitful only insofar as it appears in explanatory prin-
ciples. One area in which such principles could appear is in the metaphysics
of modality. Let me illustrate. Some metaphysicians concerned with questions
of metaphysical possibility have embraced a principle of recombination, which
is, roughly speaking, that there are no necessary connections between distinct