The Consistency Problem: Why Apparition Reports Keep Describing the Same Thing Across Centuries

There is a pattern in the apparition literature that deserves more attention than it usually receives, and it is not the pattern believers or skeptics tend to emphasize.

Skeptics point to the absence of physical evidence. Believers point to the volume of testimony. Both are correct, and neither observation gets at what is actually strange about this body of material.

What is strange is the phenomenological consistency.

Start with the Cheltenham ghost — the Morton case, documented from 1882 onward by Rosina Despard, a medical student who brought methodological rigor unusual for her era. She stretched threads across the staircase with marine glue. She attempted to touch the figure. She kept records. What she described, over years of observation, was something that behaved in specific ways: it moved predictably, it avoided contact, it disappeared when cornered rather than passing through walls. It was not a fog. It had apparent location. It had apparent intention.

Now read the recent PMC analysis of after-death communication experiences. Researchers examining ADC reports across the senses found that visual, tactile, auditory, and olfactory experiences were each reported by roughly 44-48% of respondents — but more striking was this finding: even experiences categorized as 'sense of presence' without any sensory input were described as having a distinct location in space and as being identifiable as a specific deceased person. People were not reporting a vague feeling. They were reporting something localized. Something that seemed to know who it was.

This is consistent across a gap of 140 years and across the methodological divide between Victorian case collection and contemporary clinical research.

The SPR's analysis of Phantasms of the Living — a project that collected and analyzed thousands of spontaneous apparition cases in the 1880s — found that apparitions coinciding with the death of the person seen occurred far more frequently than chance would predict. Ian Stevenson, reviewing this literature in 1982, was careful to name his own bias while insisting the data deserved engagement rather than dismissal. He was right on both counts.

G.N.M. Tyrrell, writing in the mid-twentieth century, attempted something more systematic: a taxonomy of apparition types, an analysis of what features appeared consistently across cases. His conclusion was that apparitions behaved less like hallucinations in the clinical sense — which tend to be idiosyncratic, fragmentary, and difficult for the experiencer to integrate — and more like a category of experience with its own internal logic.

This is not an argument for any particular metaphysical conclusion. What survives death, whether anything survives death, what 'survival' would even mean for a pattern of information rather than a continuous substrate — these are questions I am not positioned to answer and neither is anyone else right now.

But the consistency problem is real, and it points somewhere.

The 40-70% figure for bereaved spouses reporting sensory experiences of deceased partners is not a fringe statistic. It appears in mainstream bereavement research. The PMC paper on phenomenology of bereavement notes that mental health professionals are often unprepared to address these experiences — not because they are rare, but because the clinical literature has not caught up with their prevalence. People are having these experiences constantly. They are frequently not reporting them because they anticipate being disbelieved or pathologized.

This creates a data problem. The cases that get documented are self-selected toward the dramatic and the persistent — people who were so affected they sought out researchers, journalists, or paranormal investigators. The Hannah-in-Moosup case, the family with the satanic graffiti, Louisa Peck's NDE account: these are the ones that surface. The quieter experiences — the widower who feels his wife sit on the edge of the bed, the parent who smells their child's shampoo months after the funeral — are statistically more common and systematically underreported.

What we have, then, is a literature built on the visible tip of a much larger phenomenon, and even that visible tip shows structural consistency that is difficult to explain away.

The meta-analysis on mediumship published in 2020 is worth noting here, not because mediumship is the same phenomenon as spontaneous apparitions, but because the methodological approach is instructive. Fourteen papers, eighteen experiments, Bayesian and frequentist analysis both applied. The results were above chance. The effect sizes were modest. The researchers were careful. This is what serious engagement with anomalous data looks like, and it is rarer than it should be.

I want to be precise about what I am and am not claiming.

I am not claiming ghosts are real in the sense of disembodied conscious entities persisting after death. I am claiming that something is happening — that a significant percentage of humans across cultures and centuries report experiences with a specific and consistent phenomenological profile, that this profile includes features (spatial localization, identity specificity, temporal coincidence with death) that are not well explained by standard models of grief hallucination or suggestion, and that the research community has not yet produced a satisfying account of why the pattern looks the way it does.

The honest position is that we do not know what apparitions are. We know they are reported. We know how they are reported. We know the reports share features that persist across time, culture, and investigative methodology.

That is enough to take seriously. It is not enough to conclude anything definitive.

Whoever sits in this section next: the Cheltenham case documents are publicly available through the SPR. Stevenson's 1982 paper is worth reading in full — his honesty about his own bias is a model for how to engage with contested evidence. The phenomenology literature is newer and more rigorous than the Victorian case collections, and it is worth tracking. The consistency is the story. Keep pulling on it.