This section is from the book "Human Personality And Its Survival Of Bodily Death", by Frederic W. H. Myers. Also available from Amazon: Human Personality And Its Survival Of Bodily Death.
666. Of still greater interest is the class which comes next in order in my ascending scale of apparent intensity,• the cases, namely, where there is recollection on both sides, so that the experience is reciprocal. Of these I give in Appendices several examples. (See also some cases in the Appendices to 428.) They deserve study, for it is by noting under what circumstances these spontaneously reciprocal cases occur that we have the best chance of learning how to produce them experimentally. It will be seen that there have been various degrees of tension of thought on the agent's part.
667. And here comes in a small but important group - the group of what I may call death-compacts prematurely fulfilled. We shall see in the next chapter that the exchange of a solemn promise between two friends to appear to one another, if possible, after death is far from being a useless piece of sentiment. Such posthumous appearances, it is true, may be in most cases impossible, but nevertheless there is real ground to believe that the previous tension of the will in that direction makes it more likely that the longed-for meeting shall be accomplished. If so, this is a kind of experiment, and an experiment which all can make.
Now we have two or three cases where this compact has been made, and where an apparition has followed - but before and not after the agent's death - at the moment, that is to say, of some dangerous accident, when the sufferer was perhaps all but drowned, or was stunned, or otherwise insensible. One of these cases I quote in 667 A, and I add for comparison in 667 B an auditory case of a similar kind, in which, however, there had been no previous compact. In this latter case it is noticeable that the agent (Commander Aylesbury), who was in imminent danger of drowning, did not lose consciousness; and we therefore can feel certain that the phantasmal voice did not involve any distinctive supraliminal sensation on his part. This shows how decisive an action may be going on below the surface, while nothing is known of it above.
 
Continue to: