I have not read very deeply about mereology, but I feel that I should have a view about such things.
What follows is more a statement of position than an argument.
Some of my requirements include:
(1) Realism about ordinary objects. This does not mean accepting that ordinary objects survive even microphysical changes, i.e. accommodating a distinction between essential and non-essential change. It just means that the medium-sized dry goods that we treat as wholes really are wholes.
(2) Respecting de re modality – which I think requires transworld identity.
(3) Respecting our best physics… mostly.
It seems to me that the best way to guarantee (1) is to allow universalism about fusions. That is, any two things a and b compose a third thing c, regardless of whether or not a and b are immediately, spatially contiguous. We can try primitive composition, but that just seems too ad hoc.
So, you do get the result that my nose and the Eiffel Tower compose a third object. However, I would not say that mereology is as ontologically innocent as David Lewis did. That is, c does not stand in a relation of plural identity with a and b. This is because it is conceivable that a change in the position of a and b alone may result in the destruction of c and the creation of a new whole d. For example, it seems logically possible to cut a man in half, destroying him and yet not destroying any of the sub-atomic particles that composed him. As the distinction between essential and non-essential changes is being ignored (or rather, all changes are taken to be essential), any change of position of parts will destroy the whole they compose on this view.
Given that a change of position destroys a whole, it is suggested that wholes are not only dependent upon their parts, but also on the world that contains them. Thus, a and b are not enough to compose c, c also requires w – the world containing a and b.
This should not be confused with the dependence monism of Jonathan Schaffer, who claims that the only fundamental entity is the whole world. On Schaffer’s view, c is dependent on a token world w as he rejects transworld identity and genuine de re modality (as I interpret him). On my view, any number of possible worlds contain c by virtue of the infinite possible recombinations of regions of the world beyond the immediate spatial region of c. So, c is dependent on a type of world and not a token world.
A further consequence of my view (the view I am settling for and probably not originating) is that only wholes composed of non-contiguous or non-overlapping parts are dependent on the whole world. Wholes composed solely of contiguous or overlapping parts can constitute a world of themselves and for this reason they are not dependent on any type of world. This is important because my initial view was that wholes are dependent on both simples and the world. However, this committed me to the denial of gunk and I don’t want to be so committed. Instead, I can say that wholes are dependent upon their parts generally, but wholes with spatially diffuse parts are also dependent upon the type of world they inhabit.
Lastly, the world itself is dependent upon its parts generally, but not the parts that are dependent upon it. Thus, the world is only dependent upon its simple parts and its parts that are composed of contiguous or overlapping parts.
I am open to suggestions for revision.
0 Responses to “Current Thoughts on Parts and Wholes”