Project thesis · revision 1
The persistence question
Historical testimony disappears with its witnesses. Some physical systems survive. This project asks whether the second fact can be made to do any work on the first.
- Published
- 1 September 2026
- Status
- Preliminary
- Findings
- None
01
The problem
People have reported visions, presences, voices, apparitions and encounters in every period and in most places we have records for. The reports are abundant, they are often detailed, and they are almost impossible to test.
The reasons are structural rather than accidental. The witnesses are dead. The language they used has shifted under the words. Interpretations accumulated on top of the account, sometimes for centuries, until it is no longer possible to separate what was claimed from what was later understood to have been claimed. Whatever conditions obtained at the time - the weather, the room, the state of the person, what else was happening nearby - were not recorded, because nobody thought of them as conditions.
This is the ordinary situation of historical testimony and it is not a scandal. It does mean that treating such reports as evidence for anything physical is a category error: they are evidence about what people said and wrote, which is a different and more tractable subject.
02
The persistence question
Something, however, does occasionally survive. Not the witness, and not the conditions - but sometimes a part of the physical environment: a building, a tower, a mechanism, an object that was present and is present still.
Most historical testimony has lost its experimental apparatus. Occasionally, part of the physical environment survives.
That is the whole opening. The question this project exists to make smaller is whether anything useful follows from it: whether a physical system that persists across the interval can be characterised in the present in a way that says anything at all about the past it sat through.
We do not know, and the expected answer is no. What we can do in the meantime is document the systems and the reports properly, which is worth doing under either answer.
03
Why resonant objects?
Because a resonant object turns structure into response. Strike a bell and it answers; the answer is not arbitrary, and it is not a matter of interpretation. Modal frequencies, decay times, beating between partials, the way the response changes with mounting, with excitation, with temperature - all of these are functions of the object's physical state, and all of them can be measured repeatedly by different people with different equipment.
That is a rarer property than it sounds. Most old objects yield very little to measurement: you can weigh them, photograph them, and describe their surfaces. A resonator will tell you about its interior, its geometry and its condition every time you excite it, and it will tell you the same thing again next year if nothing has changed.
A violin behaves the same way, and so does a clock, a tuning fork, an organ pipe and a music box. What they have in common is not that they make music. It is that their behaviour is a legible function of what they physically are.
A resonant object turns structure into response.
04
Why old objects?
Age is not evidence. This has to be said first and repeatedly, because the opposite assumption is so easy to fall into that we have caught ourselves making it.
Old objects are interesting when - and only when - they carry something with them: a documented provenance, a known maker and date, a stable location, records of repair, a history of repeated use, an institution that has kept custody, and a present physical state that can still be characterised. It is that combination that makes an object useful, and most old objects have almost none of it.
A poorly documented object of eight hundred years is scientifically less useful than a well-documented one of a hundred and twenty. Provenance, not antiquity, is the scarce resource.
The witness is gone. The bell may still be there. Whether that is worth anything is exactly what remains to be found out.
05
Do objects remember?
Probably not in the sense implied by the question.
There is a widespread and very old family of ideas in which places retain presences, houses hold on to what happened in them, relics carry something of their owners, and particular locations produce unusual experiences again and again. These are historically and culturally serious claims - they have shaped architecture, law, inheritance and worship - and this project treats them as claims about what people have believed, which is what we can actually study.
There is also a perfectly ordinary sense in which physical objects preserve their history, and it is the sense we are interested in. Metal fatigues. Surfaces oxidise. Wood moves with two centuries of humidity. Parts are replaced, and the replacement is a dated event with physical consequences. Stress accumulates; geometry changes; a crack propagates and is repaired and the repair is itself detectable. None of this is memory in any interesting sense. It is material history, and unlike testimony it is still present and still measurable.
The question worth asking is not whether an object remembers. It is whether a persistent material system can supply any evidence that is independent of the narrative later attached to it.
06
Why historical reports?
Because they tell us where to look, and for no other reason.
A report identifies a place, a practice, a season, an object, a set of circumstances. That is useful information about where a persistent physical system might be found and what might be worth characterising about it. It is candidate-selection material.
The reports are not physical evidence and are never treated as such on this site. No entry in the source collection is classified as evidence of anything, and the data model has no field in which such a classification could be recorded.
The corpus is deliberately cross-cultural, and not because we think different traditions are describing one thing. Close to the opposite: if there were any physical or environmental correlate to find, the search for it should not depend on the vocabulary of whichever tradition the person searching happens to have grown up in. A physical hypothesis should not become more or less plausible because one witness used the word angel and another did not.
Theology matters to the meaning of a report. It should not decide the calibration of an instrument.
Historical report
Names a place, a practice, a season, an object.
Candidate context
Somewhere worth looking, for reasons that can be written down.
Surviving physical system?
Usually not. This is where most candidates end.
Provenance
Maker, date, location, repairs, custody - or no entry.
Measurement
Characterise the present state, repeatedly.
Control
Against comparable objects recorded the same way.
Reproducible departure?
The expected answer is no.
Only then: interpretation
And not before.
07
The working hypothesis
Stated as narrowly as we can make it: if some unusual historical reports had an external physical or environmental component, then a well-characterised persistent resonant system might show a reproducible departure that its measured local variables do not account for.
Every clause of that is doing work. Some, not all. External component, not cause. Might, not does. Reproducible, not noticed once. Not accounted for by measured local variables - which requires measuring a great many local variables before the phrase means anything.
The model behind it is a common-cause model and it is deliberately mundane. It supposes an ordinary environmental condition that nobody was measuring: acoustic conditions, infrasound, mechanical vibration, pressure, temperature, the electromagnetic or geomagnetic environment, material stress, or something equally unremarkable. Such a condition could in principle contribute both to a human experience and to the state of a nearby physical system, without either causing the other and without anything unknown to physics being involved.
We are not proposing an unknown field, a new force, or a mechanism of any kind. We are proposing that a class of ordinary variables was not being recorded, and asking whether a persistent object could stand in as a very poor, very late instrument for some of them.
08
The null
The most likely outcome of this work is that no reproducible departure is found. That is the expected result, and it is a useful one.
A clean null, obtained under conditions declared in advance and documented well enough for someone else to repeat, is a genuine contribution: it closes a question that is otherwise argued about indefinitely on the basis of anecdote. The archive of ordinary recordings and the registry of documented objects would remain valuable whatever the null showed, because both are collections of well-described material that did not previously exist in one place.
We would rather preserve a clean null result than manufacture a revelation from noise.
09
Earlier pilot observations
Project Nikolaos did not begin with a sequence of successful encounters, and it would be dishonest to present it as beginning from nothing either.
Informal exploratory observation preceded the project by some years. It included time spent at sites with a historical association with unusual reports; deliberate attempts to observe and account for unusual aerial and visual phenomena; audio recording at historical locations; and simple environmental measurement with whatever equipment was to hand.
No reproducible anomalous effect was established. Where an observation could be accounted for at all, ordinary explanations accounted for it, and where it could not, the failure was almost always a failure of documentation rather than a residue of mystery: the conditions had not been recorded, so nothing could be concluded either way.
This is published because it belongs to the denominator. An account of a research project that lists only the observations that led somewhere is not a record of the project.
10
Why Nicholas?
In the medieval tradition surrounding Nicholas of Myra there is a story about an impoverished father with three daughters and no dowries. Gold arrives in the house at night, through a window, from an unknown source. The father knows only that something has arrived. He does not know who gave it. The event repeats, and after it has repeated he changes what he does: he keeps watch, and by keeping watch he identifies the giver.
That sequence is the project's method in miniature, and it is the reason for the name. An effect is established first. Its source is identified afterwards, if at all, and only by deliberately changing how one observes. The story does not begin with a theory about who is bringing the gold.
Observation before attribution.
Project Nikolaos begins in the middle of that sequence rather than at the end. We are trying to establish whether there is anything to observe. The question of what it would mean comes later, and may never come at all.
Unknown giver
Physical gift arrives
Recipient observes the effect
The event repeats
Observation is changed
The source is identified
11
What this project does not claim
It does not claim that any anomalous effect has been observed, by us or by anyone else.
It does not claim that objects store spirits, memories, entities, presences or messages. It does not propose an unknown force. It is not paranormal research, and it does not treat anecdote as weak evidence - anecdote is a different class of source, and no quantity of it accumulates into a measurement.
It is not, at present, an experiment. It is the documentation and preparation that an experiment would require, and most of that work would be worth doing even if the experiment were never run.
12
Related work: dream-recalled musical material
One member of this project composes, and a proportion of the musical material used in that work arrives in dreams and is written down on waking. It is released elsewhere, under a separate name, and is mentioned here for one reason.
A melody remembered on waking establishes that a melody was remembered. It does not establish where it came from. The material is kept because it is artistically useful, not because anything is known about its source, and it is not offered as evidence of anything at all.
It is a useful control case to have close at hand while working on this subject, because it is a real, repeated, personally compelling subjective experience with an obvious and entirely ordinary explanation - and it is a standing reminder that subjective arrival is not external source.
13
Where to go from here
The method page sets out the standards of evidence this project intends to be held to. The source collection is where the historical reports are catalogued, with their transmission chains rather than their contents. The archive and the registry are where the physical material lives.
Direct answers to the questions this page raises are collected in the frequently asked questions.