Friday, November 30, 2018

Believing of God that he exists

One formulation of Schellenberg’s argument from hiddenness depends on the premise:

(4) If for any capable finite person S and time t, God is at t open to being in a personal relationship with S at t, then for any capable finite person S and time t, it is not the case that S is at t nonresistantly in a state of nonbelief in relation to the proposition that God exists.

Schellenberg argues that God is always open to personal relationships if he exists, and that there are people nonresistantly in a state of nonbelief to the proposition that God exists, and so God doesn’t exist.

I want to worry about a logical problem behind (4). Schellenberg attempts to derive (4) from a principle he calls Not Open that says, with some important provisos that won’t matter for this post, that “if a person A … is … in a state of nonbelief in relation to the proposition that B exists” but B could have gotten A to believe that B exists, “then it is not the case that B is … open … to having a personal relationship with A”.

It seems that Schellenberg gets (4) by substituting “God” for “B” in Not Open. But “the proposition that B exists” creates a hyperintensional context for “B”, and hence one cannot blithely substitute equals for equals, or even necessarily coextensive expressions, in Not Open.

Compare: If I have a personal relationship with Clark Kent, I then automatically have a personal relationship with Superman, even if I do not believe the proposition that Superman exists, because Superman and Clark Kent are in fact the same person. It is perhaps necessary for a personal relationship with Superman is that I believe of Superman that he exists, but I need not believe it of him under the description “Superman”.

So it seems to me that the only thing Schellenberg can get from Not Open is something like:

(4*) If for any capable finite person S and time t, God is at t open to being in a personal relationship with S at t, then for any capable finite person S and time t, it is not the case that S is at t nonresistantly in a state where he does not believe of God that he (or it) exists.

Now, to believe of x that it exists is to believe, for some y such that in fact y = x, that y exists.

But then all that’s needed to believe of God that he exists is to believe in the existence of something that is in fact coextensive with God. For instance, suppose an atheist believes that her mother is the being that loves her most. Then she presumably believes that the being that loves her most exists. In doing so, she believes of the being that loves her most that it exists. But in fact, assuming theism is true, the being that loves her most is God. So she believes of God that it (or he) exists.

At this point it is really hard to find non-controversial cases of the relevant kind of nonbelief that (4*) expresses. By “non-controversial”, I mean cases that do not presuppose the non-existence of God. For if God does in fact exist, he falls under many descriptions: “The being who loves me most”, “The existent being that Jean Vanier loves the most”, “The most powerful conscious being active on earth”, etc.

It is true that Schellenberg needs only one case. So even if it is true, on the assumption that God exists, that the typical atheist or agnostic believes of God that he exists, perhaps there are some people who don’t. But they will be hard to find—most atheists, I take it, think there is someone who loves them most (or loves them most in some particular respect), etc. I think the most plausible cases of examples are small children and the developmentally challenged. But those aren’t the cases Schellenberg’s argument focuses on, so I assume that’s not the line he would want to push.

The above shows that the doxastic prerequisite for a personal relationship with B is not just believing of B that it exists, since that’s too easy to get. What seems needed (at least if the whole doxastic line is to get off the ground—which I am not confident it does) is to believe of B that it exists and to believe it under a description sufficiently relevant to the relationship. For instance, suppose Alice falsely believes that her brother no longer exists, and suppose that not only does Alice’s brother still exist but he has been working out in secret and is now the fastest man alive. Alice believes that the fastest man alive exists, and mistakenly thinks he is Usain Bolt rather than her brother. So she does count as believing of her brother that he exists, but because she believes this under the description “the fastest man alive”, a description that she wrongly attaches to Bolt, her belief doesn’t help her have a relationship with her brother.

So probably (4*) should be revised to:

(4**) If for any capable finite person S and time t, God is at t open to being in a personal relationship with S at t, then for any capable finite person S and time t, it is not the case that S is at t nonresistantly in a state where he does not believe of God that he (or it) exists, under a description relevant to his personal relationship with God.

This doesn’t destroy the hiddenness argument. But it does make the hiddenness argument harder to defend, for one must find someone who does not believe in anything that would be coextensive with God if God exists under a description that would be relevant to a personal relationship with God. But there are, plausibly, many descriptions of God that would be so relevant.

A different move is to say that there can be descriptions D that in fact are descriptions precisely of x but some cases of believing that D exists are not cases of believing of x that it exists. Again, one will need to introduce some relevance criterion for the descriptions, though.

[Note added later: This was, of course, written before the revelations about Jean Vanier's abusiveness. I would certainly have chosen a different example if I were writing this post now.]

5 comments:

Anonymous said...

I have a question about your statement (just a clarification):

"Now, to believe of x that it exists is to believe, for some y such that in fact y = x, that y exists."

I would think that to believe of x that it exists is to believe that there is an y such that y=x. There might be a problem here regarding the move from a de re state to a de dicto state, but the reason I was wondering about your characterization is that it contains "...that y exists'' in the analysans, and isn't that understood in turn as believing, for some z such that z=y, that z exists, and so on?

steve said...

Suppose, before the fall of the Berlin wall, parents living in E. Berlin want a better life for their newborn child. They give their child up for adoption, by entrusting him to someone who has free passage between E. Berlin and W. Berlin. Perhaps that individual is just a trusted conduit who will convey the newborn to a loving but infertile couple in W. Berlin. The child will never know the identity of its biological parents, yet their action was an expression of sacrificial love.

Heath White said...

Speaking for Schellenberg, I think I would reply this way:

Suppose you are right and lots of people have personal relationships with God albeit not via beliefs that employ the description "God." This fact either is or is not an evil, of the same general kind as hiddenness. The prima facie evil would be that there are people who are non-resistantly open to a personal relationship with God *under that description* and don't have it.

Now, if this is a new form of hiddenness-evil, the argument just starts all over again. But if it is not, we have a powerful reason for doubting that any one religion has *the* story ... after all, by hypothesis there is nothing wrong with having this relationship under a variety of descriptions.

The conclusion is therefore either (a) God does not exist or (b) religious pluralism.

Alexander R Pruss said...

Heath:

It is surely not the case that for every description D that God satisfies, the existence of people non-resistantly open to a personal relationship with God under D but not having it is a prima facie evil.

Moreover, Schellenberg needs more than just "prima facie evil". His argument needs it to be so bad that God couldn't override the reasons from the value of a D-relationship with some others.

It seems to me that there will be three kinds of D:
1. Ones where there is no prima facie evil.
2. Ones where there is a prima facie evil but other considerations can override it.
3. Ones where there is a prima facie evil but very few if any considerations can override it.

I am not sure there are specific descriptions of type D that apply to God. Maybe though there is a family of descriptions such that not having a relationship under some description from that family falls under (3)? That seems somewhat plausible, but it's going to be hard to specify the family.

Alexander R Pruss said...

Fabio:

I don't think so. Suppose that Alice's husband Bob is Russia's best secret agent inside Apple's security division. But Alice doesn't know this, and doesn't even know that Apple's security division contains any Russian secret agents. Then Alice knows of Russia's best secret agent in Apple's security division that he exists. But Alice does not know that there exists a y such that y = Russia's best secret agent in Apple's security division. For if she knew that, she would be able to figure out with a bit of logic that Russia has a secret agent in Apple's security division.