« First They Came for My Chicken Sandwich . . . | Main | Coitus Reservatus and Beyond »

Friday, August 03, 2012


Feed You can follow this conversation by subscribing to the comment feed for this post.


You have introduced the QuineSpeak predicate letter 'D' to translate the English predicate '---depends for its existence on---'. So, if 'a' is the QuineSpeak name for Abraham, say, and 'g' for God, then the QuineSpeak wff 'Dag' translates 'Abraham depends for his existence on God'. The QuineSpeak is no more than a compressed version of the English. Then you go on to say

To translate the target sentences into QuineSpeak one has to treat the presumably sui generis relation of existential dependence of creatures on God as if it were an ordinary external relation. But such ordinary relations presuppose for their obtaining the existence of their relata.
This is where you lose me. Why have we moved from talk of predicates into talk of relations?


I enjoy the blog. First-time commenting. I apologize if the following is asinine, it just occurred to me and it seemed interesting to me. Still not totally sure if it will pan out.

Suppose that van Inwagen concedes that we cannot make sense of classical theism as you have construed it, but he can make sense of a closely related position; classical theism*.

He might insist that 1-t is false, for God bears the dependence relation to himself. He could either (a) say that 1-t is an incorrect translation of 1, on the grounds that the context tells us that what is meant is that God does not depend on anything besides himself for existence. Or he could also say (b) that 1 is false, but what we can affirm is 1*: God does not depend on anything besides himself for his existence.

Either way the translation that he wants to end up with is:

1-t*. (x) [~(x=g) --> ~Dgx]

we could then affirm the following:

3-t (∃x) Dgx

Also, because of the sameness of antecedent we can easily combine 1-t* with 2-t to sum up the classical theists dependency thesis:

4-t (x) [~(x=g) --> ~Dgx • Dxg]

Anyways, because classical theism* affirms 3-t, everything that exists bears the existential dependence relation to something, so we need not posit additional modes of being to makes sense of the view. Further we can still make sense of the derivative and the underivative notion, by saying that something exists derivatively if it bears the dependence relation to something besides itself, and underivately if it bears the relation only to itself. We could also take a page out of mereology and define 'proper dependency' as follows: x properly depends on y if and only if x bears the dependency relation to y and y bears does not bear the dependency relation to x.

I don't think that proper dependency could be used to generate the same kind of problem that you raised in your post, since it is just special type of the general dependency relation, the intuitive force of it requiring modes of being would be lessened. At least, it seems to me.

Perhaps you will insist that this is simply not the view you are interested in entertaining. But then, perhaps van Inwagen would insist that this view is close enough to do the trick, so to speak. The trick being making sense of the intuitions and motivations that generate the classical theists position.


C Gibbs,

Excellent comments! Thanks.

I originally entertained the thought that God is not dependent on anything distinct from himself for his existence, but then I realized that a stronger claim is warranted: God is not dependent on anything (or at least anything concrete) for his existence. (And if God is simple then he cannot be dependent on his attributes, which are poresumably abstract objects.) For we certainly don't want to say that God causes his own existence. To cause his own existence, he would have to exist 'before' (logically speaking) he exists. God cannot be metaphysically prior to himself. If we think of divine creation as a relation, then it is an asymmetrical and irreflexive (not just nonreflexive) relation: if g creates x, then it is not the case that g creates g.

So I would maintain that (1-t) is correct.

More later.

C G,

A question: if God bears the existential dependence relation to himself, then wouldn't Socrates bear it to Socrates, and similarly for every creature? But if it makes a wee bit of sense to say that God bootstraps himself into existence, surely it makes no sense to say the same of Socrates.


Predicates can be 1-place, 2-place, n-place. By deleting the name from 'Socrates is wise' we get the predicate '___ is wise' and by deleting the names (and replacing them with slots for names) from 'Socrates is taller than Crito' we get the 2-place predicate '___ is taller than ---.'

Properties are what is expressed by 1-place predicates and relations are what is expressed by n-place predicates where n > 1.

Thanks, Bill. A couple of further questions, if I may.

1. Don't the QuineSpeak predicate 'D' and the ordinary English predicate '---is dependent for its existence on---' express the same relation? If so, doesn't the 'presupposition' problem from which you say the QuineSpeak suffers also transfer to the English? Or is there a subtle difference between the ways that we are to interpret QuineSpeak and English?

2. I don't understand the presupposition problem. What goes wrong with the following presentation? Initially, just God exists, so the relation expressed by 'D' is empty, ie, there does not exist x, y, such that Dxy. Subsequently, God ('g' in QS) creates and sustains Abraham ('a' in QS). So now Dag obtains and obtains uniquely, ie, Dxy --> x=a & y=g. We have to say that the relation denoted by 'D' has changed, or at least, its extension has changed. It is as if we had two diagrams. In the first a blob labelled 'g' stands majestically alone. In the second a blob labelled 'a' is added together with an arrow from the 'a' blob to the 'g' blob labelled 'D'. I cannot see where the notion of 'presupposition' or 'logical alreadiness' comes in.

The comments to this entry are closed.

My Photo
Blog powered by Typepad
Member since 10/2008



May 2024

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  
Blog powered by Typepad