Properties and a grounding principle

ABSTRACT

In this paper, I advance a lesser known counterfactual principle of grounding in a new kind of way by appealing to properties and the work they do. I then show that this new way of arguing for this principle is superior to another way, describe some of the work this principle can do, defend my use of this principle, and conclude with remarks on why principles like it are needed.

KEYWORDS:

• Facts
• grounding
• properties
• sensitivity

Acknowledgements

For helpful comments, I thank Dan Korman. I also thank my wife, Amy Saenz, for her love and always present support and encouragement in my work.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 For discussion of these principles, see inter alia, Dancy (2004, Ch. 3), Zangwill (2008), Rosen (2010, 118), Chudnoff (2011, 563–7), Fine (2012, 38–40), Trogdon (2013), Leuenberger (2014), Litland (2015), Skiles (2015), and Chilovi (forthcoming).

2 That we should pay attention to what goes on ‘inside’ the facts has been discussed before (Rosen 2010, 119; Audi 2012, 693; Fine 2012, 74–80; Dasgupta 2014). But the focus of such discussions has been on explaining why some facts ground others and not so much on modal principles.

3 In this conditional and others below, ‘x’ is bound by a universal quantifier. Because of this, such conditionals have us quantifying into the operator ‘the fact that’. We could instead state the conditional as follows: x is crimson → (x is red because x is crimson). For those who prefer regimenting talk about grounding in terms of ‘because’, feel free to translate as you see fit. Nothing of substance follows.

4 See Saenz (2018) which involves, among other things, defending it against various criticisms.

5 Cameron (2014, 95) also speaks of the first relatum, as opposed to the existence of the first relatum, giving ground to R's being instantiated. Now this is a different kind of claim and so involves a different kind of relation (again, forget about what relations, if any, we should call ‘superinternal’). It also seems to be a non-starter. Cameron claims that set-membership and composition are such relations. But how can an electron or a plurality of electrons ground that set-membership or composition is instantiated? They do not seem to be the kinds of things that could do this. At any rate, the claim I am targeting here is not that R is instantiated because of the first relatum, but that R is instantiated because existence is.

6 A similar point about composition was made in my (2018). But there, my concern had to do with whether or not there are mereological sums. And I claimed that all summists are committed to thinking that for any sum y, the xs that compose y do so simply in virtue of existing. But issues having to do with sums and summists do not matter here. What does matter is whether or not set-membership and composition are instantiated because existence is.

7 I thank Tom Donaldson for this example.

8 Given how ‘composition’ is standardly defined (the xs compose y ↔ the xs are all parts of y, no two of the xs overlap, and every part of y overlaps at least one of the xs), it follows that everything composes itself. And so for anything, if it is one, it composes itself and so composes something.

9 That this is so is part of the point of Fine’s (1994) influential article on essence. Even if necessarily, the xs form a set or compose a whole, it is not part of their essence to form or compose at all.

10 Finding such features seems to be a harder thing to do for set formation than for composition (thanks to an anonymous referee for pointing this out). What non-essential necessary feature does Socrates have in-virtue-of which he's a member of {Socrates}? Though I reject it, consider this: the xs form y because they are thought of as together. Here, the intellectual activity of collecting or thinking of together is that in virtue of which a plurality forms a set. This is inspired by something Georg Cantor (1932, 282) says

By a “set” we understand any collection M into a whole of definite well-distinguished objects of our intuition or our thought (which will be called the “elements” of M).

An immediate problem with this is the existence of more sets than humans can generate. Some, realizing this, have suggested resorting to God or some God-like agent (Plantinga 2007, Appendix). Here then, a kind of anti-realism about sets provides us with a potential explanation of the instantiation of set formation that appeals to some non-essential, though necessary, feature of pluralities.

11 Here is another worry, raised by an anonymous referee, with Sensitivity (for more worries and some responses to them, see Saenz 2018, 108–11):

At t₁, ship S is composed of the xs. Over time, each x is removed such that at t₂, S is composed of the ys. Finally, at t₃, the xs compose a different ship while the ys, which still compose S, have become rotten. But such a situation is incompatible with Sensitivity since Sensitivity implies that if the xs's (appropriately arranged) being sturdy grounds S's being sturdy, then if it were the case that S is not sturdy, it would be the case that the xs are not sturdy. But this is false since at t₃, S is not sturdy (because the ys, being rotten, are not) and yet the xs are still in good condition.

But Sensitivity does not imply what this worry says it does. Sensitivity only applies to cases where some things are some way F because those very things are some other way G. Here, the things bearing the properties remain but the properties do not (we went from F to G). But in the above worry, we have a case where S is sturdy because the xs are also sturdy. In this case, the things bearing the properties change (we went from S to the xs) but the property does not. Here then, Sensitivity does not apply.

Still, given that the xs compose S, I grant that S is sturdy because the xs (appropriately arranged) are sturdy. I also grant that given this, if it were the case that S is not sturdy, it would be the case that the xs are not sturdy. And this in spite of the fact that at t₃, S is not sturdy (because the ys are not) but the xs are. How can we make sense of this? By denying that the situation at t₃ is the closest situation to the situation where S is not sturdy when S is composed by the xs. Given that S is sturdy because the xs (appropriately arranged) are sturdy, those closest situations where S is not sturdy are ones where S is composed by the xs. They are not ones where S is composed by the ys. And of course, in these closest situations, if S is not sturdy, then neither are the xs.