distinction in modes of being it serves to mark. The terminology continued to be in
currency even in the twentieth century, in which wefind Bosanquet engaged in a
dispute over whetherfinite existents all have the mode of being of“adjectival
entities”;inthisdispute,Bosanquetendorsedanaffirmative answer and argued
that only the absolute is a true substance.^42 Finally, Perszyk (1993: 108–10) suggests
that Meinong had some use for this mode of being as well: some of Meinong’s
incomplete objects enjoy“implexive being,”a mode of being analogous to Aristotle’s
exists-in. (For example, the triangle, an incomplete object, has implexive being in
actual triangles.)
The view also explains why attributes are necessarily dependent on the existence of
substances. The very being of an attribute encodes the information that some
substance exists and exemplifies it: for an attribute to be just is for that attribute to
be exemplified.^43 In the previous section, we looked at a view that grounded the
necessity of a material object’s being spatiotemporally located in the mode of being of
material objects. In this context, it is natural to think of substances as being the
locations of attributes: they are thenexusof inherence.^44 And so by similar reasoning,
the way in which an attribute exists provides the ground for the necessary truth that
modes are always“located”in substances.^45 (This explanation is subject to similar
conditions as the explanation presented in section 2.4.1 of why material things are
necessarily spatiotemporal: it presupposes that entities have their mode of being
essentially.)
Suppose that a necessary condition of being a substance is enjoying an absolute
form of existence. Is it also a necessary condition of being a substance that such
entities donotenjoy a relative form of existence as well? If so, substances cannot
persist viaexisting at times. Given that there are substances, this suggests that time
must be understood either adjectivally or relationally, rather than as consisting
of entities at which objects exist, or the relation between objects and time must not
be an ontological one. However, if substances can enjoy two modes of being,
(^42) Bosanquet et al. (1917–18). See also Mander (2011: 382–91) for further context. Note then that
Bosanquet is not merely a“priority”monist who holds that the One grounds its parts; hence, the notion of
ground does not suffice by itself to fully characterize the dispute between a monist of Bosanquet’s stripe
and his opponents. We’ll have far more to say about ground in chapter 8. On priority monism, see Schaffer
(2007b, 2010). 43
Vallicella (2002: 21) holds that substances and attributes exist in different ways, and the ontological
dependence of attributes on substances is grounded in the mode of being of the attribute. But the key next step
is to provide an explanation of why and how this dependence is grounded in the mode of being of attributes.
The view that attributes have a relative mode of being is the bedrock for the explanation offered here. 44
45 See Hawthorne and Sider (2002) for a full-blooded defense of this analogy.
Knuuttila (2012: 71) notes that, for Scotus, the being of an accident is notbeing in, and this is why
Scotus recognizes the need to postulate a new kind of entity called“inherences”in order to link accidents to
substances. Kok (2014: 528–32) describes Buridan as being in a similar position, and positing a new kind of
entity called“added dispositions”as well. See also Amerini and Galluzzo (2014b: 11). Similarly, Oderberg
(2007: 155–6) holds that accidents merely naturally inhere in substances, but not as a matter of metaphys-
ical necessity, but he is silent on whether this position requires adding to the ontology.