Arthur Balfour's Ghosts

Subtitled 'an Edwardian elite and the riddle of the cross-correspondence automatic writings', this is a classic and landmark study of these extensive scripts involving some very distinguished Edwardian characters, including Arthur Balfour, who served as both Prime Minister and Foreign Secretary and was also President of the Society for Psychical Research. For those readers unfamiliar with the cross-correspondences, they constitute over 3,000 scripts from different mediums over a 35-year period, and were the subject of an extensive study by HF Saltmarsh as early as 1938. The cast is very distinguished, both intellectually and culturally, and communicators included SPR pioneers FWH Myers, Edmund Gurney and Henry Sidgwick, all of whom were Fellows of Trinity College, Cambridge. The book provides an accessible treatment of the main themes in the scripts and a closer examination of some examples exhibiting paranormal cognition.

The first part covers the development of the cross-correspondences and describes the major characters. The second asks a number of searching questions: were the scripts unambiguous, consistent and meaningful? Did they exhibit paranormal cognition? Can we be sure that their cryptic nature was not a psychological artefact? Was the communication process interactive and did it signal intention on the part of the communicators? Were normal avenues for acquiring information ruled out? Were the correspondences tweaked by wishful thinking, over subtle interpretation or deliberate selection? And finally were the aims, intentions and long-term predictions of the communicators fulfilled? These are all thoroughly addressed. In the end, it comes down to a choice between the survival hypothesis and some form of super ESP, as with other evidence of this kind. The author explains the ideal and almost unattainable criteria laid down by CD Broad in 1925. He points out that it is in principle almost impossible to distinguish between the living agent psi hypothesis and the survival hypothesis in this or any other case. He argues, correctly in my view, that the best of the scripts stand up well against the Broad criteria. However, they can that best be only highly persuasive, as the reader's judgement has the last word. But readers should take the trouble to read the detail before summarily dismissing the evidence on a priori grounds - the book is a formidable achievement.