Do theists typically believe by faith in God that God exists? Faith is much more than propositional belief. But someone who has faith in a person can believe propositions by faith. What does that mean? I want to start with this necessary condition:
- x believes that p by faith in y only if x believes that p because x takes y to have assured her specifically that p.
But given (1), how could one believe that y exists by faith in y? One would have to believe that y exists because one took y to assure one specifically to that effect. But that would be rationally odd. Granted, I could hear a voice in the dark assuring "I exist", and I could first believe that The Voice is assuring me that it exists, and conclude from this that The Voice exists. But I wouldn't be concluding that The Voice exists because of the specific content of what was voiced, but simply because something was voiced. The connection between the assurance and the belief that The Voice exists is not a connection in the right way for belief by faith.
Granted, it is possible that the voice sounds so trustworthy that I first form the belief that the content of what was said is true, and then because of that I come to believe that The Voice exists. In that case, I would indeed be believing that The Voice exists and doing so by faith. But I would be ignoring an obvious logical inference from one's data and getting the conclusion of that inference by other means. So what we seem to learn from the case of The Voice is this:
- Anybody who believes by faith in y that y exists is in a position to believe by obvious logical inference and not by faith that y exists.
But that doesn't quite answer the question I started with, namely whether theists typically believe by faith that God exists. For it could be that (2) is true, and that most people do in fact go for the obvious logical inference, but the voice of God assuring them of his existence is so powerful that in most cases the belief is overdetermined: they believe both by obvious logical inference from God's assurance and by faith.
There is a second complication. One might have faith in y under one description and believe that y exists under another description. For instance, suppose that The Voice in the dark says: "I am the brother you never thought you had." Then you might believe your brother exists by faith in The Voice. This model works well for Christianity. A Christian might well believe that God exists by faith in Christ, even though Christ is in fact God.
Does the model work outside of Christianity, say in Old Testament times? Well, the Voice/brother case suggests that it might work in cases of religious experience. But it seems implausible that most of the theists in Old Testament times were theists because they had a religious experience whose content included an assurance of God's existence. Maybe, though, one can introduce the notion of indirectly believing by faith, where you indirectly believe something by faith provided that you infer it from something that you (directly) believe by faith. To adapt a Plantinga example, God might give you a religious experience that God forgave you your sins; trust in the "inner voice" (i.e., in God, but you don't know that right away) leads you to believe by faith that God forgave you your sins; and then you conclude by logic that God exists.
I don't have an answer to the question I started with, whether theists typically believe by faith in God that God exists. But I have a story that would have to hold for the answer to be affirmative. Typical theists would either have to be in a rational overdetermination scenario or they would have to be in a position where the difference between two different ways of referring to God can be leveraged to make it rationally possible for them to first believe in an assurer, who happens to be God, and then in God as such.