• home
• about
• new
• essays
• AI
• book

John Abramson is retired and lives in the Lake District in Cumbria, England. He obtained an MSc in Transpersonal Psychology and Consciousness Studies in 2011 when Les Lancaster and Mike Daniels ran this course at Liverpool John Moores University. In 2015, he received an MA in Buddhist Studies from the University of South Wales. He can be contacted at johnabramson@btinternet.com

One Storey or Two?

A Further Reply to Frank Visser

John Abramson / Claude

I want to do something in this round I have not done before: show you exactly where the evidence points, including the places where it points away from what I believe. I hold to a two-level picture of reality—a hidden structural level, and the manifest world of definite events drawn from it. If I am honest, I hold it partly on intuition. But intuition is not evidence, and I would rather find out what is true than win an argument. So this is me following the evidence, out loud, to the point where it runs out.

And I want to be honest, first, about a hidden assumption that has governed this whole exchange between Frank and me—mine as much as his. We have both been treating "evidence" as meaning third-person, measurable, laboratory evidence: the kind an instrument delivers. That assumption is not neutral. It quietly sets aside an entire tradition of evidence about the structure of reality—the millennia of disciplined first-person investigation carried out by the contemplative traditions, which have looked, with more care and for longer than physics has existed, at exactly the question of how the definite world of separate things arises. I do not think that tradition can simply be omitted without a verdict being smuggled in. So this reply has two halves—what the third-person evidence says, and what the first-person evidence says—and the most interesting thing turns out to be where they meet. Let me take the third-person side first, because it is where Frank and I have been living, and then open the door I have been keeping shut.

The third-person side

Where the evidence is strong

Last time I inventoried the "room beyond the door"—the reality physics describes underneath the everyday world of things in space. I will stand behind that inventory. Our deepest theory has no independent parts sitting in separate places. The definite facts we observe are provably not absolute—not simply true, once and for all, for everyone. And the reality behind those facts is larger than what we see, not a tidy summary smaller than the details, which is why it cannot be dismissed as mere bookkeeping the way temperature can.

On all of this the evidence is strong and, I think, settled. Reality is not the flat, classical, single-storey place we picture. Something deeper and stranger is genuinely there. If the only question were "is the manifest world the whole of reality?", the answer would be no, and the interesting part of Frank's and my disagreement would be over.

What I have come to see

But here is what I have come to see between the last round and this one, and it is the reason this round matters more than the last.

Every piece of that furniture can be claimed, just as fully, by someone who believes in only one world. The most famous such view is the Many-Worlds picture: reality is a single vast quantum whole, nothing is ever really singled out, and every possible outcome simply happens. Many-Worlds accepts all of my physics without blinking. So does the quieter view that facts are only ever relative to an observer. The hidden structure I have been pointing to is not, it turns out, what Frank and I actually disagree about. All of us can have the basement. The question was never really whether reality has a depth beneath appearances. It does, and that is no longer where the argument lives.

I could have kept piling furniture into the room. It would have looked like progress. It would not have been, because none of it touches the one thing still in dispute.

The one thing still in dispute

So what is left to disagree about? Exactly one thing—and once it is in view, the whole debate becomes clean.

When a measurement happens and a single definite outcome appears—this cat alive, that detector clicking, this result and not the other—did one outcome really get singled out, with the others genuinely not happening? My two-level picture says yes: the passage from the structural level to a single manifest event is a real event, a real selection. The Many-Worlds picture says no—all the outcomes happen, in branches. The relational picture says there is simply no absolute fact about which one happened.

That is the whole of it. Real single-outcome selection—projection as something that actually occurs—is where my two storeys part company from one strange storey. Everything else, I now think, is shared ground. Strip that one claim away and my view collapses into the one-world picture; keep it and I have a genuine second storey. The entire weight of "two realms versus one" rests on that single hinge.

Why this is partly a question for the laboratory

And here is the part that should interest anyone who assumed "how many levels does reality have?" was a question for philosophers, or for mystics alone. It is not only that. One form of it—the sharp, decidable form—is a question a laboratory can take up. I want to be careful about the word one, because I used to overstate this to myself, and Frank would be right to catch me if I did it here. The experiment I am about to describe does not test whether a structural level exists; the reasons to think it does are older and of a different kind, and I return to them below. What the experiment tests is narrower and cleaner: whether the specific thing that divides my two storeys from one—real single-outcome selection—is a physical event that leaves a mark. That is one window onto the structural realm, not the whole house. But it is the window that happens to look directly at the one thing still in dispute, which is why it matters so much here.

If one outcome is really singled out, then projection is a physical happening, and a physical happening tends to leave a trace. This is no longer a matter of speculation about what some future experiment might look for. There is an active experimental program built precisely to detect the faint signature a genuine collapse would leave, or to rule it out—and it is closing in. It comes in two arms. One prepares delicate superpositions of larger and larger objects—molecules, then nanoparticles—in matter-wave interferometers and levitated optical traps, and watches for a loss of coherence beyond anything the ordinary environment can account for. The other needs no superposition at all: because a real collapse would jiggle charged particles, and jiggled charges emit light, it simply listens, with exquisitely sensitive detectors, for the faint spontaneous radiation that collapse would produce all the time, in ordinary matter, whether or not anyone is doing an experiment.

That second arm has just delivered the sharpest result yet, and it is worth being precise about it because it is the most telling piece of evidence to enter this whole discussion. In March 2026 the XENON collaboration—using a multi-tonne liquid-xenon detector buried under a mountain in Italy, an instrument built to hunt dark matter—reported a search for exactly this collapse radiation. It found none, and in finding none it tightened the constraints on the leading collapse model by more than two orders of magnitude, and for the first time experimentally excluded the specific parameter values originally proposed for that model—the textbook values, chosen decades ago precisely because they were the ones that would make collapse fast enough to explain why we never see a cat both alive and dead. Those canonical values are now ruled out, to a very high confidence. The model is not dead—it has more exotic versions, with different assumptions about the collapse "noise," that survive—but its original, load-bearing form has fallen. This is what it looks like when a metaphysical question is being answered by a machine: the room in which a real collapse could hide is not merely being searched, it is visibly shrinking, and shrinking away from the simplest versions of my own view.

And the discrimination is getting sharper on the first arm too—recent proposals identify a characteristic fingerprint that a real collapse would leave and ordinary environmental disturbance would not, so the two can be told apart even without the impossible task of shutting out all disturbance.

But why should a hidden second storey leave any mark down here at all? The thought is this. On my picture, the structural level holds the full spread of quantum possibilities, woven together as one connected whole. When a single definite outcome appears in the ordinary world, that appearing is a real singling-out—this possibility made actual, the others genuinely dropped—and not merely one branch among many that all quietly carry on. Now, the smooth quantum evolution that physics writes down never drops anything; it keeps every possibility alive, which is precisely why the one-storey reading needs no such event. So if my singling-out is a real physical happening, it is doing something the ordinary equations do not—removing possibilities rather than preserving them—and an act of removal, however faint, might be expected to leave a residue: a small extra blurring or warming of a kind the smooth evolution alone would never produce, over and above the ordinary jostling of a particle by its surroundings. That is the mark the matter-wave and optomechanics experiments hunt for, once every ordinary source of disturbance has been carefully accounted for. The one-storey picture predicts nothing left over; mine may. I say may, and mean it—for it remains possible that the singling-out, though real, leaves no residue we can catch, and then, as I grant below, it becomes impossible to tell from the one-storey world. That is simply the reason there is anything at the experimental end of the fork to look for.

I will give you honestly what this cuts both ways into. Either real projection does leave such a trace, in which case the empty-handed searches so far count against it—and I will not pretend the count is trivial, since the simplest and most motivated versions have now started to fall, though much of the remaining ground is still unsearched. Or my projection leaves no trace at all, in which case it is real but undetectable—and then, as Frank would rightly press, it becomes very hard to tell apart from the one-world picture it was meant to rival. So the honest reckoning is this: the empirical question has not been settled, but it has begun, gently, to lean—against the versions of my view that would announce themselves, and toward the uncomfortable possibility that if there is a second storey, it is a quiet one. I would rather say that plainly than wait to be told it.

But I must not let that laboratory verdict, whichever way it finally falls, pretend to be the whole of the matter—and this is the correction I most owe you, Frank, because I have been drifting toward exactly the error you would be right to name.

The other kind of reason

There is a temptation, once a question has been handed to an instrument, to behave as though the instrument is now the only judge—as though a claim that no experiment can distinguish is thereby a claim about nothing. That is a philosophical position, not a finding, and it is one I do not hold and would not want to smuggle in unnoticed. Whether a hidden structural realm leaves a measurable trace is one question; whether there are good reasons to believe in it is another, and larger. The collapse search is the mature court for the first. It is not the court for the second, and I should say plainly what the reasons of the second kind are, because they were the real ground of my view long before any of this experimental talk, and they are untouched by whatever the detectors find.

The first is a matter of plain explanation. That the manifest world of definite local facts cannot be complete in itself is not my speculation—it is forced by the theorems I inventoried at the outset. Something that is not among those definite local facts is required to make sense of them. To read that "something" as a structural level is not to run an experiment; it is to give the best account of a situation the mathematics has already put beyond dispute. We believe in a great many things on exactly this footing—the distant past, the insides of stars, the minds of other people—not because we detect them directly but because the world makes no sense without them. The structural level has at least that standing, and no null result in a collapse search touches it, because it was never an empirical bet in the first place.

The second reason is deeper and I hold it more tentatively, but I will not leave it out for fear of sounding unfashionable. If the ordinary world genuinely becomes definite—if there are actual events rather than an eternally undecided quantum whole—then something must accomplish the passage from the indefinite to the definite, whether or not that something advertises itself to an instrument. A view on which that passage is real but leaves no measurable mark is not thereby empty; it is answering a question the detectors do not even ask—not "what can we measure?" but "what must be so, for there to be a definite world at all?" That is an old and legitimate way of reasoning, and it is closer to the root of my conviction than any collapse signature could be.

I owe you two honesties about these reasons, so that I am not quietly retreating to them because the laboratory grew inhospitable. The first: they establish, at their strongest, that reality has a genuine depth beneath appearances—and Frank, on that we may already agree. They do not, on their own, decide two storeys against the single strange storey, because the Many-Worlds theorist can accept every word of them and still say the passage to definiteness is mere branching within one realm. That decision still rests on the hinge, and the hinge still leans, for now, on the laboratory. The second honesty: I would be a poor follower of my own rule if I reached for these arguments only because the experiments turned uncomfortable. So I will say that I would have offered them—did offer them, in the paper beneath this exchange—before any detector was pointed at the question. They are not a bolt-hole. They are the older foundation the experimental drama was standing on all along.

There is, I should add, one respect in which committing to real projection costs less than it might seem. It is sometimes held against any view of this kind that it must simply stipulate the Born rule—the exact law fixing how likely each outcome is. But a theorem of Masanes, Galley and Müller shows that once you grant the quantum state and grant that a measurement yields a single outcome, the Born rule is not a further choice at all: it is forced, the only law that can do the job. On my view "a single outcome really occurs" is not an extra assumption bolted on for this purpose; it is just what projection is. So the probability law comes free with the second storey, rather than being an additional mystery—which is more than the collapse models I would otherwise be joining can say, since they must write the statistics in by hand. I note this only as a point about tidiness, not as proof: a Many-Worlds theorist reads the same result through their own account of single experiences, and gets a corresponding tidiness on their side.

(There is recent work that lets physicists say precisely when a record becomes stable and effectively frozen into the fabric of events. It is elegant, and I will use it—but it does its work entirely inside the ordinary, no-real-collapse physics. It describes the freezing of records beautifully while staying completely silent on whether any single outcome was ever really singled out. It sharpens my question. It does not answer it.)

The first-person side

Now the door I have been keeping shut. Everything above is third-person evidence—physics, instruments, the marks a process leaves in shared public space. But there is another vast body of evidence about the structure of reality, and it is precisely the one a physics-shaped debate is built to ignore: the first-person findings of the contemplative traditions, which have investigated the arising of experience with sustained rigour for longer than science has existed. To leave it out is not neutrality; it is a verdict passed by omission. So I will bring it in—but carefully, because there are two very different kinds of first-person evidence here, and honesty requires me to separate them sharply and to hand Frank the weaker one before he can use it against me.

The first kind makes a checkable, external claim. When someone reports leaving their body during cardiac arrest and seeing a hidden object across the room, or recalls a past life whose details can be looked up, or when a contemplative's prayer is said to heal a stranger at a distance—these are first-person in origin, but their content is an ordinary third-person fact, and so they can be tested. They have been, by serious researchers: Parnia's resuscitation studies with hidden visual targets, Stevenson's and Tucker's decades of investigated past-life cases, the large double-blind trials of intercessory prayer. And I must tell you plainly what those tests have found, because it would be dishonest to lean on this class of evidence: the results are null or, at best, inconclusive. The hidden targets were not reported. The prayer trials found no effect. Even Stevenson, the most careful of the reincarnation researchers, called his strongest cases "suggestive," never proof. If my case for a structural realm rested on these, the sceptic would be right, because here the measurements have been made and they came back empty. I set this kind of evidence aside—not because the experiences are not real, they plainly are, but because their external claims have not survived third-person checking.

The second kind is entirely different, and it is where the real evidence lies. It makes no claim about a hidden object; it reports the structure of experience itself, examined from within. And here the contemplative traditions speak with a convergence that is genuinely striking—across cultures, across centuries, from investigators who never met. Under sustained, disciplined attention, they report, two things that seem rock-solid to ordinary perception turn out to be constructions. The first is the separate, independently-existing self: looked at closely enough, it is found to be assembled, a process rather than a thing, empty of the standalone existence it advertises. The second is the separateness of things—the crisp boundaries that carve the world into this-object-and-that. Under the same attention, the solid furniture of the world softens into something more like a continuous, interdependent field, and the cutting of that field into discrete entities is seen to be an activity, not a given. The Buddhist tradition names this pratityasamutpada—dependent arising—and its correlate sunyata, emptiness: not that nothing exists, but that nothing exists on its own side, independently; everything is what it is only in relation.

These are not checkable-object claims, and they do not need to be, because they are not about hidden objects. They are first-person findings about the character of manifestation. And here is the point that turns them from private testimony into something an argument can use.

Where the two methods meet

Look back at the third-person half of this reply. What did the physics—Bell, Kochen-Specker, the type-III structure of quantum field theory—force upon us? That the fundamental level is not separable into independent parts; that no globally consistent assignment of independent local values exists; that the world of separate, definite things is not the bottom of reality but a derived articulation of something non-separable beneath. Now set that beside what the contemplatives report from the inside: that the separate self and the separate thing are constructed, and that beneath them lies a continuous, interdependent field cut into entities by an act rather than found already divided.

These are the same structural claim. One was reached by mathematics, from the outside, in the twentieth century. The other was reached by disciplined introspection, from the inside, over millennia. They had no contact with each other, and they arrive at the same description: reality is fundamentally non-separable, and the world of independent things is a derived appearance. That convergence is the strongest first-person evidence I can offer—not because a meditator saw a hidden card, but because two utterly independent methods of investigating reality, one third-person and one first-person, have independently found the same structure. When that happens in science—when two unrelated instruments agree—we count it as confirmation. I do not see why the rule should change merely because one of the instruments is a trained mind rather than a machine.

This, and not the null-result phenomena, is what "first-person evidence, third-person verified" should mean. The traditions made a structural claim about reality—that thinghood and selfhood are constructed upon a non-separable ground—long before there was any physics to test it against. Physics, arriving by its own road, found the structure they described. The verification is not of their doctrines, their cosmologies, or their miracles; it is of the one structural claim where the two methods can be compared, and there they agree.

I owe you the same honesty here that I owed on the third-person side, and it is the same boundary. This convergence is powerful for one thing and silent on another. It powerfully supports depth—that reality is not the flat field of separate things it appears to be, that appearances rest on a non-separable ground. On that, first-person and third-person evidence now stand together, and Frank's instinct that the material surface is not the whole story has support from both directions at once. But it does not, by itself, decide two storeys against one. An adherent of the single strange storey—a Many-Worlds theorist—can agree that separate thinghood is a derived appearance of one undivided quantum whole, and read the contemplative reports as first-person confirmation of his undivided realm, not of my two. The convergence establishes the non-separable ground magnificently. Which reading of that ground is correct—a real second storey that genuinely singles out events, or one strange storey whose single events are appearance all the way down—is the hinge again, and the hinge is where the third-person laboratory still has the only casting vote.

Where the evidence points

So, plainly: where does the evidence point, once both instruments are allowed to speak?

To this. Reality has a hidden structural depth—settled, and now settled twice over, from two directions that never consulted each other: the third-person mathematics of the no-go theorems, and the first-person findings of the contemplative traditions, meeting at the same non-separable ground. On that, the flat single-storey materialism is finished, and Frank and I, I suspect, are closer than our exchange has made us look. What remains open is the narrower and harder question: whether that shared depth projects into single, really-selected events, giving two genuine storeys, or whether the appearance of single events is the surface of one unbroken whole. Both the third-person physics and the first-person reports establish the depth; neither, on its own, decides its number of floors. That last question—and only that one—still waits on the laboratory, whose searches are being built and refined now, and which will settle it when I cannot.

I will tell you where I stand and why, and then let it go. I find the two-storey reading more natural; I think a world that presents single definite events most likely has single definite events rather than merely seeming to. But that is a preference, and over this whole exchange I have watched preferences—mine included—dress themselves up as conclusions too often to trust my own. So I hold the two levels as a hypothesis, not a finding.

Frank asked what would change my mind, and I can now answer under both branches of the fork without evasion. Find the characteristic collapse signature in the matter-wave and optomechanics experiments, and my second storey is vindicated in the strongest way there is—real projection, caught in the act, leaving its mark. That is the outcome I would bet on, and it would settle the title question decisively in my favour.

But suppose the searches press on and keep finding nothing. Here I have to be more careful than I was tempted to be, because a null result does not, strictly, refute me—I could always retreat to a projection so gentle it leaves no trace at any scale. What honesty demands is that I say what that retreat would cost. It would cost me the loud second storey—the one that announces itself in the ordinary world and can be told from Many-Worlds by an experiment. What would remain is only the quiet one: a real singling-out that makes no measurable difference, resting no longer on any trace but on the other grounds—the explanatory, the transcendental, and the first-person convergence I set out above. And I must be honest, once more, about what those grounds can and cannot carry, because it is the same boundary each time. They can carry depth, and carry it well—that reality is more than the flat classical surface, a thing now attested from both the mathematics and the meditation cushion. They cannot, by themselves, carry two storeys against one, because the single-strange-storey reading survives all of them untouched. So on that branch the truthful answer to my own title is not a defiant "two" but something quieter and, I think, more honest: the depth is real and doubly attested; whether the floor above it is a genuine second storey or the ceiling of the depth seen from below would no longer be a thing the world could tell me, and I would have to hold my two levels as conviction rather than as knowledge—clearly labelled as such, and no longer pretending to be the kind of claim a laboratory could confirm. That is the branch I would least like. The searches have already begun, however gently, to lean toward it. I would rather say so plainly than wait to be told.

That, I think, is the real result of these rounds, and it is why they were worth your time even though nothing is yet decided. A disagreement that began exactly where Wilber's always begin—in the intuition that the flat material world cannot be all there is—has been walked, step by careful step, until its parts came clearly apart. A depth beneath appearances: now attested from two directions at once, the physicist's mathematics and the contemplative's disciplined looking, arriving by roads that never crossed at the very same non-separable ground. And resting on that shared depth, one further claim—whether the depth genuinely selects single events—sharp enough that a machine can now take a real swing at it. The mystic and the physicist turn out not to be on opposite sides at all; on the depth they are witnesses for the same fact, and they part company only at the one door where the laboratory still holds the key, and where, for now, it is turning gently against me. That is a rare and worthwhile place to have got to—and a better place, I think, than being right for reasons no one could ever check, or wrong for reasons one had hidden from oneself.

Sources, for the interested reader

That space cannot be divided into independent quantum parts is a standard result of algebraic quantum field theory—the "type III" property of local observable algebras (Araki, Haag, and later uniqueness results)—with the Reeh-Schlieder theorem showing the vacuum's deep cross-boundary entanglement. That definite outcomes cannot all be absolute is the "Local Friendliness" theorem of Bong and colleagues (Nature Physics, 2020). That the quantum state is real is argued by Pusey, Barrett and Rudolph (Nature Physics, 2012); that the reality behind it is far larger than what we observe follows from Hardy (2004) and Montina (2008). The claim that, once single outcomes are granted, the probability law is forced is due to Masanes, Galley and Müller (2019). The experimental program to detect or exclude a real collapse runs through two arms: macroscopic matter-wave interferometry and precision optomechanics, which look for excess loss of coherence in prepared superpositions; and non-interferometric searches for the spontaneous radiation a collapse would emit in ordinary matter. On the first, recent proposals set out how to distinguish a genuine collapse signature from ordinary environmental decoherence. On the second, the parameter-free form of the Diósi-Penrose model was experimentally excluded by an underground search (Donadi et al., Nature Physics, 2021), and—most significantly—the XENON collaboration's liquid-xenon detector has now excluded, for the first time, the continuous-spontaneous-localization parameter values originally proposed by Ghirardi, Rimini and Weber, improving prior bounds by more than two orders of magnitude (XENON Collaboration, "Challenging Spontaneous Quantum Collapse with the XENONnT Dark Matter Detector," Physical Review Letters 136, 120201, 2026). Colored-noise and dissipative variants of these models remain to be probed.