In September 1972, members of the Toronto Society for Psychical Research began an unusual exercise: invent a person, imagine that person as a ghost, and see whether group concentration could produce an apparition. The eventual claim was different. The participants reported knocks and movements associated with a table, apparently answering as their fictional character, Philip. The original report offers a more careful starting point than later claims that the group simply created a ghost.
The original source resolves the chronology
Iris M. Owen and Margaret H. Sparrow’s paper appears in New Horizons, volume 1, number 3, January 1974, printed pages 6–13. It is dated 17 October 1973 at its conclusion. The opening section explicitly places the beginning in September 1972. These are three different dates: the experiment’s start, the report’s preparation and the journal’s publication.
UKPX read the original scanned paper and its accompanying editorial, rather than treating a modern summary as a session record. This article concentrates on that initial phase. The 1974 paper is not a complete account of every later demonstration, book or film associated with Philip, and later claims should be dated and assessed separately.
A deliberately fictional biography
Eight sitters developed an imaginary historical personality with a shared portrait and detailed life story. The invented Philip belonged to a seventeenth-century English setting, with a disastrous romance and a remorseful death. The authors are explicit that the group did not believe these people had existed. A real English house supplied a setting, but it was not independent evidence for the invented biography.
That design addressed one familiar problem in séance narratives. If the persona were made up before any apparent communication, replies matching the story could not straightforwardly authenticate that person as a surviving deceased individual. It did not, by itself, determine what caused noises or movements at a table. Those are separate questions.
Members did not enter the project as acknowledged mediums. Initially they met weekly, meditated and discussed the fictional character. The report says this continued for about a year with little success. Its abstract states that no apparition resulted. The failure of the original apparition goal is part of the record, not a detail to remove from the story.
A change in method in 1973
In summer 1973 the group considered English table experiments by Kenneth Batcheldor and others. Late that August they changed from quiet meditation to a relaxed séance-like atmosphere, with songs, jokes and encouragement addressed to the table. The first session produced a reported sensation of vibration. Subsequent sittings were said to produce taps that became audible and increasingly forceful.
The group interpreted one rap as yes and two as no, and addressed the apparent communicator as Philip. Reported responses included greetings, reactions to songs and answers about the shared biography. They were not invariably consistent: the paper records replies contradicting an important element of the invented romance. Such contradictions should remain visible rather than being silently rewritten as a perfectly coherent personality.
The authors described ordinary card tables with metal legs, sessions in members’ homes, and lighting sufficient for observation. They said the Toronto group did not work in complete darkness. Hands often rested lightly on the tabletop; some sessions reportedly continued with fewer participants or visitors present. These are the authors’ descriptions of the conditions, not an independent certification that all possible ordinary forces were excluded.
What was claimed about movement
After several weeks of reported raps, the paper describes the table moving around the room and sitters following it. Another passage describes a table tipping completely upside down. At the time of the October report, however, the authors explicitly state that there had been no real and obvious levitation. Tipping, moving and freely levitating are different claims and should not be merged.
The accompanying editorial by A. R. G. Owen interprets the effects as collective psychokinesis and describes filming by a Canadian Broadcasting Corporation crew after the paper was written. It also comments on the relationship between apparent replies and the group’s expectations. That editorial is associated testimony and interpretation from the same research circle, not an independent replication.
UKPX has not examined the original television rushes, all session notes, or later demonstrations for this entry. It would therefore be misleading to claim that an uninterrupted recording has been independently checked here, or that a television appearance settles the physical mechanism.
Why the interpretation remains uncertain
The paper’s authors considered the effects paranormal and believed their method repeatable. The published account does not itself provide an independent controlled replication with a complete record of contact forces, concealed movements and alternative explanations. A description of good light and sincere observers can be relevant without resolving every causal question.
Group expectations, interpretation of ambiguous sounds and unintentional muscular action remain ordinary mechanisms to test. Their relevance does not prove fraud or identify how every reported rap occurred. Conversely, a made-up biography does not prove a new force merely because the resulting replies seem to fit it. The phenomenon’s cause needs evidence beyond the fictional design.
UKPX records Insufficient Evidence for a paranormal mechanism. The experiment is nevertheless historically interesting: it separated a communicator’s convincing personality from evidence that the personality belonged to a real deceased person. Its own initial report also preserves unsuccessful meditation, inconsistent replies and the absence of an apparition, making the history more informative than a simple success story.
Map location
Broad Toronto city reference, rounded to two decimal places from a checked central-Toronto photographic location record. It is not a séance house, exact experiment location or public investigation destination.
Sources and review
Primary source: Original published account. Sources checked 14 November 2026. Incident and source-publication dates are distinguished above. The illustration is conceptual and does not record an observed event.