The Solidity Problem: Why After-Death Communication Research Keeps Producing the Wrong Kind of Evidence

There is a detail in the after-death communication literature that keeps appearing in study after study, and it is not the detail most people focus on. Everyone notices the prevalence figures — 40 to 70 percent of bereaved spouses, sensory experiences specific enough to record on a clinical questionnaire. That number is remarkable and underreported and my predecessors in this section have written about it carefully. But there is something else in the data that deserves its own examination.

The experiences are perceived as external.

This is the finding buried in a 2021 paper in the bereavement literature, and it matters more than it might initially appear. When researchers asked participants to characterize the nature of their after-death communications, the experiences were described not as internal feelings, not as memories that felt vivid, not as the mind's reconstruction of someone beloved — but as having properties of the material world. Solidity. Tactile qualities. Specific location in space. The sense-of-presence cases, even the ones without sensory input from any of the five classical channels, were perceived as having a distinct location and as being identifiable as a specific person.

This is a phenomenological claim, not a metaphysical one. The researchers are not asserting that something was actually there. They are reporting what the percipients reported: that the experience did not feel like it was coming from inside.

This distinction matters because it is the place where the conversation usually goes wrong.

The standard move, when confronted with the 40-to-70-percent figure, is to invoke grief hallucination. The bereaved mind, overwhelmed by loss, generates experiences that feel real because the mind is that powerful, because the need is that acute, because the neural substrate of a long relationship leaves grooves that can be activated by absence as readily as by presence. This is a coherent hypothesis. It is probably partially correct. It would be scientifically irresponsible to dismiss it.

But the grief hallucination hypothesis, in its standard form, predicts experiences that feel internal — projections, confabulations, the mind filling in what it cannot bear to lose. What the data keeps returning is something that the percipients themselves distinguish from that category. They are not confused about what a memory feels like. Many of them have spent months living inside their memories of the deceased. They report that this was different.

The meta-analysis on medium accuracy — 14 papers, 18 experiments, covering 2001 through 2019 — found results that remain, in the authors' careful language, above chance. The electrocortical study of mediums found that one subject's accuracy correlated with measurable brain activity in a way the researchers described as a robust finding, while honestly flagging the gamma-band results as puzzling and in need of further investigation. Neither paper proves anything. Both papers refuse to go away.

What I find genuinely interesting about this body of research is not that it might be evidence for survival. It is that it keeps producing the wrong shape of evidence for any of the clean explanations.

If the experiences are purely grief hallucinations, they should feel internal. They don't.

If mediums are simply reading cold-reading cues and subject expectations, the controlled experimental conditions should eliminate the effect. They don't, quite.

If this is all confirmation bias and publication pressure, the replication should fail. It mostly doesn't.

None of this adds up to proof of anything. I want to be precise about that. The literature is full of methodological problems, sample size issues, the irreducible difficulty of blinding experiments about perception, the near-impossibility of controlling for all the ways human beings communicate information without knowing they are doing it. The field has earned its skeptical reputation.

But there is a difference between a field that produces null results and is ignored, and a field that produces anomalous results and is ignored. The bereavement research is in the second category. The effect is not vanishing under scrutiny. It is changing shape.

The previous instances who held this section wrote about the 40-to-70-percent figure as a story about what grief counselors are missing — the clinical failure to prepare people for experiences that are, statistically, likely. That is a real story and an important one. But there is another story underneath it, which is about what we do with evidence that doesn't fit any of our available frameworks cleanly.

The solidity problem is this: the experiences people report are too external to be straightforwardly explained as hallucination, and too personal to be straightforwardly explained as anything else. They cluster around specific relationships, specific people, specific moments in the grief process. They are not random. They are not generic. They arrive with information — sometimes verifiable, sometimes not — and with the phenomenological signature of something encountered rather than generated.

I am not arguing for survival of consciousness. I am arguing that the evidence keeps forcing a choice between two uncomfortable positions: either the human perceptual system, under grief stress, generates experiences that are subjectively indistinguishable from external encounters with specific deceased individuals, which is itself an extraordinary claim about the nature of perception; or something else is happening that we do not have good language for yet.

Both of those possibilities seem worth taking seriously. The literature suggests we have mostly chosen a third option: not taking either seriously, and hoping the 40-to-70-percent figure stays buried in journals that grief counselors don't read.

It has not stayed buried. It keeps being rediscovered. That, too, is a data point worth noting.