Richard M. Gale, On the Nature and Existence of God (Cambridge UP, 1991), p. 11 :
First, because God is a supernatural being, he seem to defy being
indexically pinned down or baptized. There are no lapels to be
grabbed hold of by a use of 'this.' Some would contend that we can
ostensively pin down the name 'God' by saying 'this' when having or
after just having a mystical or religious experience, in which
'this' denotes the intentional accusative or content of the
experience. This would seem to require that these experiences are
cognitive and that their objective accusative is a common object of
the experiences of different persons as well as of successive
experiences of a single person.
Suppose Abraham or someone has an experience the intentional object of which he dubs 'God.' Suppose the experience is not 'cognitive,' i.e., not veridical: nothing in reality corresponds to the intentional object, the accusative, of the experience. Then there will not have been a successful reference to God. Successful reference is existence-entailing: If I succeed in referring to X, then X exists. Pace Meinong, one cannot refer to what does not exist. Reference is in every case to the existent. It therefore seems that Gale is right when he says that a successful baptizing of God requires the veridicality of mystical experience.
Andrew V. Jeffrey (Faith and Philosophy, January 1996, p. 94) responds to Gale as follows:
. . . the religious language-game could be played as if theistic
experiences were both veridical and cognitive even if they were
not; i.e., people could play the referential game even with a
radically misidentified referent.
It seems to me that this response misses the point. Suppose the referent has been radically misidentified: Abraham dubs his Freudian superego, or an overwhelming sense of anxiety, or what have you, as 'God.' Then no successful reference will have been achieved. Is a long disquisition necessary to explain that God cannot be a feeling of anxiety?
And if you say that all baptisms are successful in that, after all, something gets baptized, then I say that this shows the utter hopelessness of the causal theory of reference. For the question to be answered is this: How in the utterance of a name does the speaker succeed in referring to an object? Under what conditions is successful reference achieved? A theory that implies that one always succeeds, that there are no conditions in which one fails to succeed, is worthless.