- There are several debates in metaphysics that centre on issues of arbitrary boundaries. Consider the question of composition: when do several things compose another thing? We want to draw a line between those sets of things that compose another and those sets of things that don't compose another. But where to draw the line, without being arbitrary? We want to rule in the set that contains (only) the jacket and the trousers of my suit, but rule out the set that contains (only) the trousers of your suit and the trousers of my suit. But this seems, in some inchoate way, metaphysically arbitrary: there seems to be no real, objective difference between the two sets. (The issue isn’t quite the same as the problem of vagueness, since even a vague boundary would seem arbitrary.)
- Three possible views, then.
- Mereological nihilism draws the boundary way to one side: it says that no set of putative parts is a set whose parts compose something else.
- And mereological universalism draws the boundary way to the other side: it says that every set of putative parts is a set whose parts compose something else.
- Finally, restricted composition says that there is a real, objective boundary between those sets whose members compose something else and those sets whose members don't compose something else, and that that boundary lies somewhere between the two extremes.
- There are two ways of developing the thesis of restricted composition.
- The first is to propose some unifying principle (or principles) of composition: putative parts compose only when they are touching each other, or stuck together somehow, or 'caught up in a life' (cf. van Inwagen 1991).
- The second is to reject any such unifying explanation of composition, and say that the many specific facts of composition are brute (Markosian 1998).
- A structurally similar issue arises when we consider the individuation of individual things in modal space. Given a plurality of possible worlds, we can ask after the modal boundaries of any individual thing: what merely possible things count as 'the same thing’ as this particular individual? To stay neutral on issues of trans-world identity (vs. counterpart theory), we can ask which things are versions of a given individual. To ask whether a possible thing y is a version of some actual individual x is just to ask whether x could have had the properties of y.
- The problem of arbitrariness arises here as well. We want to draw my modal boundaries so that I, with a haircut, lie inside the boundary, but so that the atoms that presently compose me, scattered across the universe a few hundred years from now, do not. But this seems metaphysically arbitrary. Three possibilities, then:
- to draw the boundaries of individuals in the extremely minimal way (so that no merely possible things count as versions of me);
- to draw the boundaries of individuals in the extremely maximal way (so that everything counts as a version of everything else); or, finally,
- to say that there are real, objective boundaries which fix the modal boundaries of individuals, and that these boundaries lie somewhere between the two extremes.
- One way of developing this thought is by talking about essential properties. We find this idea in Aristotle, where he writes that ‘the essence of each thing is what it is said to be in virtue of itself,' and that '[what... you are by your very nature is your essence' (Metaphysics Z, 1029b14). The temporal and modal boundaries of a thing extend just as far as those things that share that thing's essence, and it is the possession of this essence that explains why a thing's boundaries lie where they do. (More on this idea later in this chapter.)
- The aim of this essay is to motivate the rejection of individual essences, in their traditional form. My argument against them has two steps.
I then consider alternative solutions to the problem of temporal and modal boundaries (Section 3). The view that I favour is brutal individuation, which is the temporal and moral analogue of Markosian's brutal composition: individuals have objective modal boundaries, but these facts don't admit of any further explanation.
- First, I argue that we have prima facie reason to reject individual essences of two types: sortal properties and historical properties (Section 1).
- Then I criticize three arguments that suggest we need to posit essential properties, in spite of this (Section 2).
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2020
- Mauve: Text by correspondent(s) or other author(s); © the author(s)