^{page 244 note 1} I have been helped greatly here by conversations with Donald F. Henze, although I am sure that he would not agree with all that I say.

^{page 246 note 1} The problem of God predication at this point looms large in the minds of many, no doubt. For, they would want to point out, this analysis allows that ‘loves’ is predicated of God in the same sense it is predicated of men. To be brief, so it does, but then so it is. If it were not, then truly we would be lost in our understanding of what we are being told when we are told that God is concerned for us and loves us as a Father.

^{page 246 note 2} In what follows I do not reply to McPherson's point that I do not distinguish between saying that p and ˜˜p are equivalent in meaning and saying that ˜˜p gives the meaning of p. The distinction McPherson is pointing up is valid: often something equivalent in meaning does not give the meaning of what one says to someone else. Thus while ‘Au secours!’ is equivalent in meaning to ‘Help’! it would not give the meaning of ‘Help!’ to one who speaks only German. However I do not think that any confusion regarding this distinction undercuts the argument I propounded to support the idea that assertions must have an empirical denial. If it does, so much the worse for the argument, which of course was not one I advocated as sound, but one I examined as a possible argument a proponent of Flew's views might advance to support the falsification challenge.

^{page 247 note 1} He says that ‘the view that reasons for religious belief are internal to the belief has been persuasively presented by such writers as Professor Peter Winch and Mr D. Z. Phillips. From such a view it could be said that there is, in a sense, evidence for religious belief…’