In The Traveler’s Almanac (the first section of which I recently published to this blog), I wanted to emphasize pulp fiction, partially as an update to Alan Moore’s focus on Victoriana in The League of Extraordinary Gentlemen, and partially because there seemed to be something very American about the golden age of pulp magazines. The project quickly swelled to include all sorts of fiction, but I did start off doing lots of research into pulp writers.
One of the authors I discovered in this research was Manly Wade Wellman. The description of his writing intrigued me, and after some searching, I ordered his anthology “Worse Things Waiting” through inter-library loan. It’s a terrific collection for anyone who likes old fashioned ghost stories, and anyone interested in stories set in the Appalachian mountains. By coincidence, the book was also the topic of a recent video collaboration between several book-tubers.
Anyway, the stories aren’t the main thing I want to talk about. I want to talk about the introduction written by Wellman himself, and specifically the following quote”
“Most of us, as I think, have the sixth sense of wonder and want to hear, even to believe, stories of wondrous terror. Montague Summers insists that you must believe in ghosts to tell good stories about them, but M.R. James, who wrote better and more frighteningly than he, says only that he is ready to consider evidence of ghosts. If I may intrude into such splendid company, I’ll confess that I have felt strongly persuaded now and then.”
The quote reminded me of a similar quote from Stephen King. In ‘Salem’s Lot, the character Father Callahan says the following:
“The ordinary fellow isn’t half so leery of the supernatural as the fiction writers like to make out. Most writers who deal in that particular subject, as a matter of fact, are more hardheaded about spirits and demons and boogies than your ordinary man in the street. Lovecraft was an atheist. Edgar Allan Poe was sort of a half-assed transcendentalist. And Hawthorne was only conventionally religious.”
At a glace, there’s a bit of a contrast. Wellman suggests that horror writers bend towards belief in ghosts/the supernatural, while King suggests that horror writers are uncommonly skeptical. But a closer look shows that even among the example each author cites there is considerable variability. In these two quotes, Wellman and King actually recapitulate much of the research on who believes in the paranormal and why.
I first looked into who believes in paranormal stuff (ghosts, ESP, etc.) in undergrad. A class assignment required me to pick two variables and do a mini-literature review on them. I picked religious belief and belief in the paranormal. I found a handful of small studies, with almost an equal number finding a positive correlation between the variables versus finding no correlation. In taking another look at the literature for this post, I found similar variability.
Published in 2003, the results of a nationwide survey in the United States found “the correlations between belief in religious phenomena and other paranormal phenomena are largely insignificant.1” Meanwhile, a nationwide survey of Canadians from around the same time found people with stronger religious beliefs were likely to also have stronger beliefs in the paranormal2. Over at The Skeptical Inquirer3, a researcher failed to replicate earlier findings, also from that periodical, that had initially found a correlation between religiousness and belief in the paranormal4.
Things don’t get any more clear when you compare paranormal belief to other variables. Charlotte Dean et al. published a great review article in 2022 that looked at the relationships between paranormal beliefs and various aspects of cognitive functioning5. They broke cognitive functioning into six categories: perceptual and cognitive biases; reasoning; intelligence, critical thinking, and academic ability; thinking style; executive function and memory; and other cognitive functions. In summarizing each of these six categories, the authors say almost the exact same thing each time: findings were inconsistent.
The authors go on to conclude that roughly 65% of the papers they reviewed showed that paranormal beliefs were associated with some sort of poorer cognition. But I don’t find that number useful. The authors fully admit they simply counted up the papers that showed a “poor cognition” result, but they didn’t calculate a combined effect size. (So it’s possible, for example, that only the small studies that found poor cognition results, and when you combine them with the big studies that didn’t have that finding, the overall effect disappears.) It’s also important to note the authors didn’t do much to account for publication bias. To their credit, they know this is an issue. They cite an interesting study that found psychologists were less likely to complete research that seemed like it would support belief in a paranormal phenomenon (in that case, astrology).
So why is it so hard to get consistent results about who believes in paranormal phenomenon? I think the main reason is that it’s just plain a bad variable.
When researchers do studies on who believes in paranormal stuff, the assumption is that they’re measuring a specific kind of person, just like gun-owners or leukemia patients are specific kinds of people. People in those groups differ from their fellow group members in plenty of ways, but they are assumed to have latent similarities. These are similarities that aren’t immediately obvious (so, not the headline fact about them all owning guns or having leukemia) but can be discovered with research. Apparently, people who believe in paranormal stuff aren’t a specific kind of people. I think they’re more like “people who have scars.”
Imagine all the ways people get scars. Maybe they were born with cleft palate and got it surgically corrected. Maybe they crashed their bike as a kid. Maybe they got into a fight or went to war. Maybe the dark lord tried to cast a killing curse on them when they were an infant. Researchers could occasionally find similarities among people with scars (especially if they, say, recruited their participants from a veterans hospital). But other researchers would try and fail to find similarities, because there’s just so many ways to fall into the “scars” group.
We come by these bad variables in different ways. In all cases, however, we have a reasonable, common sense idea about what things are similar, and that idea turns out to be wrong. Once you know how to look for them, these bad variables are everywhere. In psychology, there are huge, ongoing questions about whether labels like “schizophrenia” really capture a single phenomenon6. Some of the mental activities we call “memory” (for example, learning how to play the piano versus learning somebody’s name) are actually controlled by completely different systems in the brain. And while two members of the same race might have certain similarities in their appearance (e.g. skin color), there’s no genetic reality underlying racial classifications7.
Again, variables aren’t bad just because there are differences between its members. If you’ve ever watched politically liberal people talk about politics, you know they can be a diverse, even fractious bunch. And yet, they have latent similarities. A 2020 study found liberals are more likely to be “night owls,” even after controlling for age, sex, education, income, and aspects of personality8.
Roughly two and a half millennia ago, Plato encouraged people to “carve nature at its joints” when deciding how to classify different species of animals. (The metaphor is one of a butcher, looking for an easy place to cut to divide a carcass into parts.) His advice continues to resonate today.
- https://onlinelibrary.wiley.com/doi/abs/10.1111/1468-5906.00163 ↩︎
- Strangely, the same study found that greater church attendance was significantly, negatively correlated with paranormal belief. https://onlinelibrary.wiley.com/doi/abs/10.1111/1468-5906.00118 ↩︎
- As far as I know, not a peer-reviewed publication, though I think peer review is enormously overrated as an indicator of accuracy. ↩︎
- https://cdn.centerforinquiry.org/wp-content/uploads/sites/29/2001/09/22164801/p50.pdf ↩︎
- This really is a great paper, written an an accessible and engaging way. Even if you’re not used to reading research, you should check it out if you’re interested. https://pmc.ncbi.nlm.nih.gov/articles/PMC9067702/ ↩︎
- https://psychiatryonline.org/doi/full/10.1176/appi.ajp.2014.14040435 ↩︎
- https://en.wikipedia.org/wiki/Race_and_genetics ↩︎
- https://www.journals.uchicago.edu/doi/10.1086/705927 ↩︎
Leave a Reply