it would have to be something that could not be neurophysiological
That’s wrong. Something that couldn’t be neurophysiological would be not merely evidence but proof (not necessarily of a deity, of course, but of some external cause). I suggest that, e.g., Srinivasa Ramanujan’s experience of having mathematical insights given to him by the goddess Namagiri was evidence that he was in contact with a supernatural being—but, of course, far less evidence than it would take to convince me that he really was in contact with such a being.
For A to be evidence of B, all it takes is that A is more likely if B than if not-B. Dreams of goddesses handing out what turn out to be genuine (and highly unusual) mathematical insights are more likely if there are in fact goddesses able to hand out such insights than if there are not, because the existence of such goddesses would provide one more mechanism by which such dreams could occur.
I would suggest that Ramanujan’s experiences might be as much as 10:1 evidence for the existence of the goddess Namagiri Thayar. But, of course, every other mathematician who has insights without any sign of gods and goddesses getting involved is evidence against, quite apart from all the other reasons not to believe in Namagiri Thayar or any other goddess.
Perhaps you are using the term “shred of evidence” to denote something more than that. Fair enough, I suppose, but then I’m afraid I think you chose your words badly.
Very well, I concede the point. I should not have said that “there is not a shred of evidence.” Still, AFAICT the evidence favors neurobiology by a very substantial margin.
That’s wrong. Something that couldn’t be neurophysiological would be not merely evidence but proof (not necessarily of a deity, of course, but of some external cause). I suggest that, e.g., Srinivasa Ramanujan’s experience of having mathematical insights given to him by the goddess Namagiri was evidence that he was in contact with a supernatural being—but, of course, far less evidence than it would take to convince me that he really was in contact with such a being.
For A to be evidence of B, all it takes is that A is more likely if B than if not-B. Dreams of goddesses handing out what turn out to be genuine (and highly unusual) mathematical insights are more likely if there are in fact goddesses able to hand out such insights than if there are not, because the existence of such goddesses would provide one more mechanism by which such dreams could occur.
I would suggest that Ramanujan’s experiences might be as much as 10:1 evidence for the existence of the goddess Namagiri Thayar. But, of course, every other mathematician who has insights without any sign of gods and goddesses getting involved is evidence against, quite apart from all the other reasons not to believe in Namagiri Thayar or any other goddess.
Perhaps you are using the term “shred of evidence” to denote something more than that. Fair enough, I suppose, but then I’m afraid I think you chose your words badly.
Very well, I concede the point. I should not have said that “there is not a shred of evidence.” Still, AFAICT the evidence favors neurobiology by a very substantial margin.