I've put online a draft of a paper I'm writing for Phil Compass on the grounds of necessity. Comments on anything in it are welcome, but I'm going to post here some of the stuff I say about Kit Fine's reduction of modality to essence, since I'd especially welcome thoughts on that. So here's Fine:
"Necessity has its source in those objects which are the subject of the
underlying essentialist claim. . . We should view metaphysical necessity
as a special case of essence. . . . The metaphysically necessary truths
[are] . . .the propositions which are true in virtue of the nature of all
Here are three thoughts (none intended as anything like insurmountable objections, just things to think about):
1) Prima facie, the view seems to require us to accept the existence of things like properties and relations, and thus appears to be incompatible with nominalism. For what entity can we plausibly say has a nature such as to guarantee the
truth of ‘If there are some things, there is a set of those things’ if not the
relation being a member of? No collection of actual individuals guarantees the
truth of that, because the claim says something about what happens no matter
what individuals are around.
2) If it’s necessary that there couldn’t be certain (kinds of) individuals
(universals, say, or God) then we must admit that some of the things that exist
have natures that exclude the existence of other things. You might find this
harder to accept than the claim that some things have natures that guarantee
the existence of other things. (Cf. the familiar objection to admitting
truthmakers for negative existentials: intuitively, they are true because some
things don’t exist, not because some thing does. Similarly, impossible
existents are impossible, intuitively, because there’s something about them
that’s impossible, not because, e.g., there’s something else whose essence is
such as to make them impossible.)
3) It’s easy to see how the essence of an entity e can account for the necessity of
a conditional the antecedent of which says that e exists. So my essence
grounds the truth of, hence accounts for the necessity of, ‘If Ross exists, he is
a human’. From this, it’s easy to see how unconditional necessities can be
grounded if the thing whose essence accounts for its truth has existence as part
of its essence. So were I an essential existent, my essence would account for
the necessity of the antecedent of the above conditional as well, and hence
account for the necessity of the consequent. But we might want to allow for
cases where an unconditional necessity is ‘multiply realized’ in the following
way. Suppose 2+2=4 is actually true in virtue of the essence of the numbers 2
and 4. So we account for the necessity of ‘If the numbers exist, 2+2=4’. But
it’s not just conditionally necessary that 2+2=4, ‘2+2=4’ is itself necessary.
But on this view, that’s not because the numbers exist necessarily: on this
view, while our actual world is Platonist, and mathematical truths are true
because of the numbers, structuralism is possibly true and ‘2+2=4’ is true in
virtue of the essence of certain structures, and maybe in some worlds there are
brute mathematical laws, and ‘2+2=4’ is true in virtue of these laws. So there
are multiple possible grounds for the arithmetical truth, and the truth is
necessary because it’s necessary that there is some ground or other. But what
actual things have essences such as to ground this last necessary truth? The
worry is that Fine can only account for conditional necessities or
unconditional necessities which are unconditionally necessary because there is
some essential existent that accounts for their truth in any possible circumstance.
Wednesday, September 23, 2009
Subscribe to: Post Comments (Atom)
Re 1: If the relations are internal, then one could just do without them. The natures of the relata are sufficient to fix the relational truths. Relational truths without relations.
That still requires properties. I'm not a nominalist, so I'm not sure if this is what they would want to say, but why not suppose that whatever makes it the case that a is F is of such a nature that it grounds the essential truths? Be sure, of course, to read 'nature' nominalistically. Perhaps there is no such reading, but I don't yet see why. If there is a nominalistically acceptable reading of 'a is F', I'm not sure why there can't be one for 'a has an essence such that...'.
RE 2: If one held a powers view of modality, where what is possible is grounded in what powers there are and what they are powers *for*, then the right thing to say here is not that there is an essence that excludes the existence of that which necessarily fails to exist, but that no power is such that it is a power to bring it into existence.
(I have no idea what sort of crazy person would hold *that* view, but, you know, it's true, so at least it has that going for it.)
This doesn't fit well with your intuition that when something's an impossible existent, what makes it impossible is something about *it*, but that's false anyway. : )
I'm not sure I'm getting where you're coming from on (1). Grant that the 'being a member of' relation is internal, and so no addition of being over the relata. How does that help Fine account for the necessity of 'if there are some things, there's a set of those things'? I would have thought it would have made it worse, since we can't now say it is the membership relation whose essence grounds that necessity, since there is no such relation.
I'm not insisting here that 'a is essentially F' demands the existence of the property F, only that it demands the existence of the entity a. Do you deny that?
On (2) - again, maybe I'm missing something, but it sounds like what you've done is describe *another* view and how *it* would handle necessary non-existents, rather than say how Fine's view can. Am I missing the point?
On 2, you're right. It's a different view. But I think Fine could say something similar. (Whether he would or not, I don't know.) Suppose N is a necessarily non-existent. Fine could say that's because no essence is such as to make possible the existence of N. No weird exclusion there, right? It needs to be necessarily true, so it would have to be true that there are no essences that are such as to make possible the existence of an essence that is such as to make possible the existence of N.
Perhaps the "makes possible" talk is going to saddle Fine with something he shouldn't be saddled with? (I read it as "confers a power to bring about", but that's certainly not in Fine.)
On 1, I was suggesting something like this: Every essence (or collection thereof) is such that it bears an internal relation to its singleton. It's the fact that every essence bears such a relation that makes it true that ACTUALLY(if there are some things, there's a set of those things). But, on the proposal, since internal relations aren't bits in the ontology, it's just the essence and the set that together make it true.
Now we just need to modalize it. I don't know what Fine would say, but here's a suggestion: No essence is such as to make possible an essence that would fail to bear an internal relation to its singleton. (As mentioned above, this might not fit well with Fine's view—I'm not sure.)
Is that any more clear?
I'm not sure about the talk of essences making things possible or essences bearing an internal relation to their singletons - this sounds like a pretty unFinean view of essence.
I agree one can appeal to just the essences of the individuals and sets to ground the necessity of 'If the Xs exist, there is a set of the Xs' where the Xs are some collection of *actual* existents. My problem is what actual things have essences that ground the fact that *no matter what* things were to exist, there'd be a set of them, even were there things that don't actually exist. I just don't see how to do that without admitting the membership relation (or by removing the idea of essence as rooted in the idea of being the essence *of some thing*).
RE 1: i presume it is the nature of sets themselves that grouns this
RE 2: Fine is happy with non-existent objects. it is in the nature of God, let's suppose, that she can't exist.
RE 3: similarly, Fine is happy to grant that I am necessarily human even though I only contingently exist.
I am not saying these satify you, but I think this is what Fine would say, more or less.
Lee: on 1. How can that be? I can see how the nature of a set S which has as its members the Xs can ground the necessity of 'If the Xs exist, there is a set of the Xs'. But this much will only get us that for any plurality of actual things, it's necessary that if they exist, there is a set of them. But the necessary truth I'm after is much more general than that.
On 2. I didn't know Fine was happy with non-existent objects which have essences. Fair enough. Seems like a cost to the view to me though.
On 3. I don't see how that addresses the issue. The issue is a structural one about whether it's conceptually possible for there to be multiply realised necessities
In a later paper - necessity and nonexistence - , Fine distinguishes between propositions that are true whatever the case and those that are true regardless of how the world is. maybe the thesis about essence applies only to the former.
I don't have it to hand, but I think he says mathematical truths are in the latter category. This may address your specific example in 3 but not the general idea.
Right, agreed. And it's the general idea I was wondering if he could allow, nothing particularly to do with maths.
I love that necessity and nonexistence paper, but as far as I'm aware Fine never says anywhere how it ties into the essence stuff. (Anyone who can correct me on this, I'd love to know.) But I always supposed that what you suggest is right: that the thesis only applies to those truths true whatever is the case - after all, he says something (if I remember right) along the lines of these being the ones that are really *necessary*, the ones that are true irregardless of what is the case being in some sense amodal.
The thing is, once he makes this move it doesn't seem that we're left with all that many necessary truths (understanding it in the narrow way). So maybe the reduction to essence is fine for this limited class, but it's less ambitious.
It'd be worth thinking about how these different pieces of Fine all go together, but I don't think that's a task for the Compass paper!
This is going to be quick, because I'm in a rush (apologies in advance for any typos and thinkos ;-)):
Somewhat unsurprisingly, I was going to say more or less what Jon said at 3:05 (but he beat me to it)--Fine would be better off adopting a power view of modality.
But, as Ross says, he doesn't. However, there is clearly a link between the two views, which is that both the power view and the essence view are what I like to call hardcore actualist views (i.e. view according to which the truth-makers of modal statements are irreducibly modal features of the actual world not possible worlds) and hardcore actualists have to choose whether necessity or possibility "comes first". The essence view, which Fine seem to favour, assumes that necessity is more fundamental, the power view which Jon and I favour, is for possibility being more fundamental.
This long preamble is just to say that I don't think Fine could say that essences make something possible without abandoning the essence view for some sort of hybrid view.
I think both 2 and 3 can be addressed by considering necessary propositions as essential features of the world itself, though I don't know just how much of a distortion of Fine this is.
It isn't too much of a stretch to take Lewis as agreeing with this, as for him propositions are properties of worlds. It's a small stretch because 'ways a world could be' aren't the same as 'ways THE world could be', so I don't know if Lewis is committed anywhere to all worlds being counterparts of each other. Even so, viewing it this way does help with issues 2 and 3.
For 2, it could be the essence of the world not to have certain kinds of thing as parts, just as it's plausibly essential to a block of cheese not to have any elephants as parts. If there are necessary existents besides the world itself, these are likewise essential parts of the world, just as it is plausibly essential to an elephant to have cells as parts, or to the Solar System to have the Sun as a part. (Of course, the world is an essential part of itelf too.)
For 3 we need to see mathematical truths and the like as general features of the world. It is essential to the world that 2+2=4 is true, though not essential to the world what it is true in virtue of. This would allow mathematical truths to be multiply realisable, but still guaranteed by the world's essence.
Michael: actually, what you say about (2) is similar to the line I take on truthmakers for negative existentials in 'How to be a truthmaker maximalist' - so I'm sympathetic to the general line of thought, although I'm not sure if Fine would be.
On (3). Interesting suggestion! I wonder though: if we're allowed to just say 'it's essential to the world that p', why not just say that about *every* necessary truth p, and let the story end there?
About (3), I suppose the idea is that de dicto necessities are de re necessities to the world. There are ways I had to be, and ways the world had to be. The former are grounded in my essence and the latter in the world's essence. Calling in the world's essence doesn't look like a one-size-fits-all solution now because the de dicto is seen as a special case of the de re.
You can only do what I'm suggesting with de dicto necessities and de re necessities concerning necesary existents. If we try "it's essential to the world that if I exist then I'm human" this creates problems with negative existentials that you don't get with "it's essential to me that I'm human".
It does still seem like cheating to rely on the world's essence, but I think we should expect any account of the grounding of de dicto necessary truths to be a bit degenerate. You can get that if they are meant to have truthmakers too, because they can only degenerately be counterfactually dependent on the existence of their truthmakers because they can't be false.
Perhaps Fine could be brought round by suggesting that everything is essentially part of a/the world, and that there are some ways worlds/the world essentially are/is, so the world's essence is involved in everything else's essence. Perhaps not though.
Post a Comment