The primary evidence put forward in support of paranormal reality are accounts of witnesses. Those who believe in the reality of UFOs, Bigfoot, or ghosts are heavily influenced by seemingly legitimate and truthful tellings of strange events people say happened to them. They also contend that there are so many accounts that “there must be something to it” and “all these witnesses aren’t lying”. Investigators collect these reports and then derive ideas, theories, and conclusions from them. Astounding accounts show up in the media, sometimes repeatedly, and those who hear paranormal-themed stories from TV and popular written accounts tend to accept that they are accurate. This is a deeply flawed assumption to make. I recently came across two sources that exemplify why we need to qualify eyewitness accounts as data. Here is the first.
Masefield’s Mystery
Many popular ghost and “true mystery” books relate content from sources that are inaccurate and do not check the primary source for accuracy. In addition, they certainly do not check if the primary account was accurate in the first place.
Sir Peter Masefield was an aviation journalist, writer, pilot and important figure in British civil aviation during WW2, according to Robert Charman writing in the Journal of the Society for Psychical Research [1]. In 1972, Masefield related a personal paranormal experience from 1963 in Flight International. Masefield claimed to have taken a recently-met passenger on a flight where he experienced seeing the replaying in the air of a fatal plane crash he knew had happened. His passenger then disappeared. He later concluded that the passenger was the ghost of Lt. Desmond Arthur who perished in such a crash over the same location, on the same day, in 1913. Charman’s assessment of the case led him to conclude that not only is the primary account by Masefield questionable, but the subsequent retellings included embellishments or added details, even false details, that changed the story. The story is famous as part of the lore of the “Montrose ghost”.
Charman suspects, but can not know for sure, that Masefeld did not have the experience exactly as he related in his original account – he may have been imagining the event and later came to believe it really happened, or he may have fictionalized the account because of the circumstances relating to the recent remembrance of Lt. Arthur. Those are only two options that do not require us to accept a paranormal cause.
Later popular accounts of the tale in books about aviation ghosts (such as Caiden, referenced in Arthur’s Wikipedia article) not only repeated the story but added embellishments that were not in Masefield’s original accounts to add a bit more dramatic flair to the story. These retold accounts get retold again in media products where the authors use only these secondary and tertiary sources, thus perpetuating the fictionalized story as “true”. There have been countless cases where a good paranormal story was squashed by that investigator who dared to find the original newspaper account from long ago.
- Charman, R. (2018). “Did Sir Peter Masefield Fly with the ‘Montrose Ghost’ as Passenger? A Re-assessment”. JSPR. 82(2): 103-115.
For another example, part two of this series, go here.
Just to correct one point. The reference to Sir Peter serving with the USAAF is incorrect. In fact according to his Obit Sir Peter served with the Royal Air Force.
https://www.telegraph.co.uk/news/obituaries/1510672/Sir-Peter-Masefield.html
I thought that was weird! Corrected in the piece.
Cult leaders expound upon pseudoscientific theories about paranormal events but for most believers and indeed the general public the persuasive information is the number of sightings or encounters and the smoke/fire trope.
As this way of thinking implicitly concedes that eyewitness testimony is unreliable, rationalists should be able to come up with a more persuasive response than “eyewitness testimony is notoriously unreliable.”
Some sort of easily understood counter example would be preferred. For example, a corrupt lottery might have millions of adherents, many of whom supposedly came within one number of winning, even if there were no prize Other examples could be imagined, perhaps borrowing from Bayesian probability.