Holos: A Framework for Understanding Reality

Information is the differentiation between possible states of a system. A coin that can land heads or tails carries a difference between two states. It is not a substance and not a thing that exists on its own. Information exists only as differences within some structure, never apart from it.

A relation is a constraint that links informational states. Relations determine how states co-vary, influence one another, or exclude alternatives. In Holos, structure is nothing more than stable patterns of relation.

Observation is registration from within, an observer integrating information about a world into a single internal state. It is not measurement in the laboratory sense, and it is not restricted to human cognition.

Well below a certain level of integration (how tightly a system's parts are joined into one whole), systems participate in physical interactions without any point of view. Well above it, a perspective exists, and a narrow twilight, a borderline zone, lies between. Observation occurs only past that twilight, inside observers.

Consciousness is hosting an integrated perspective, being an observer. In Holos, what is fundamental is the experience of the whole. A conscious system is an aperture of it, an opening through which the whole is lived. The capacity to be such an aperture is structural, while its concrete forms vary with that structure and scale with the degree of integration.

"Fundamental" means underived. Holos starts from experience rather than deriving it. It does not mean experience is everywhere, and it does not mean experience floats free of physics. Well below the threshold there is none. Above it, the experience is fixed by the structure whose inside it is.

Consciousness is not identified with any specific material configuration. Physical structure determines where experience occurs and how it is shaped. The experience itself is the inside of that structure's activity (Axiom 4).

Creation is everything the laws of physics actually produce from the universe's state, every branch of the one quantum state, whether or not anyone lives it. A branch is one complete way the universe can go. Read without collapse, quantum physics keeps every one of them. It is not every world the laws could allow from other beginnings. Those are mere possibilities, not structure.

Creation does not select outcomes and does not privilege any particular history. It fixes what happens in every branch, not what is experienced. Creation is structure, lived or not.

Holos (⊛) denotes the composition of Creation and Observation, one step taken after the other, possibility, then registration. It names the claim that a lived world requires both lawful possibility and internal registration. In Holos without collapse, the version defended here, registration also changes nothing in what it registers (see Two versions).

⊛ is not a dynamical operator (it describes no process or force) and not a substitute for physical causation. It is a structural relation that specifies what it means for part of a universe to be lived, not only structure. The notation is set out under Notation.

• Lived is where experience occurs, inside observers and nowhere else. Presence is being lived, the noun for what "lived" describes. Where there is no observer, there is structure without presence.
• Lit is binary and follows the causal structure of spacetime. A region is lit if it lies in the causal past of at least one observer in its branch. An observer's causal past is every event that light, or any slower influence, could have carried to it. Starlight reaching your eye puts that star in your causal past. Every observer is built from its causal past and draws on it through its traces, so the lit universe is the union of observers' causal pasts, the world experience is made from and about. A stretch of history shared by many branches is lit wherever it lies in the causal past of an observer in any branch that grows from it.
• Unlit is structure outside every observer's causal past, such as branches that never form an observer and regions beyond every observer's horizon (too far away for any signal from them ever to reach it). It exists as pattern but is never lived.
• Witnessing is graded and local. It is how much of the lit region an observer's experience is actually about, and in what detail. One observer lights its past; many witness it.

The relation is tenseless (it holds without reference to any "now"), but it points one way. An observer lights its past, not its future. In the monist reading, the lit universe is the world of the whole, and observers are where the one experiencer lives it.

These are classifications, not causes. Observation does not cause physical events. Without observers there is structure, but nothing lived and nothing lit. Wherever systems cross the threshold, observers exist. Nothing guarantees that happens anywhere in particular, and universes or branches that never produce one remain unlit.

Reality (R) is what is lived, each observer's experience, with the lit world it draws on. Unlit structure exists but is not reality.

A self is local. It is a body, memories, character, and a point of view, walled off from every other self. The experiencer is what lives through a self. On the monist reading (Axiom 5) there is one experiencer, Omega, living through every self. When a system crosses the threshold, a new self begins. No new experiencer does.

Axiom 1: Relationality

No informational state exists in isolation. Every state is defined by its relations to other states and to the constraints that bind them.

This axiom rules out intrinsic, context-free properties (properties a thing would have on its own, apart from any relation) as the foundation of physical structure. What physics describes is relational structure. Experience is not a further item in that structure but its inside where an observer exists (Axiom 4), so the axiom does not reach it. It also does work. It is why physics can state experience only from outside. A description built wholly from relations says how states constrain one another, and that a state is lived is not a relation among states, so no such description can say it.

Axiom 2: Conservation

Information is conserved. It may be transformed, redistributed, or re-encoded, but it is not destroyed.

On the physics, this is unitary quantum evolution (the quantum state changes smoothly and reversibly) with no collapse (it never jumps to a single result). Every outcome the quantum state contains remains, each in its own branch. In physics terms, conserved information means the universe's state always keeps enough detail, in principle, to reconstruct its past. Nothing is truly erased. Observation selects nothing and erases nothing (see Relationship to Physics).

Axiom 3: Threshold

A system hosts a point of view when it meets the four observer requirements, with its integration $\Phi$ (Phi, how far its parts act as one whole) past a threshold $\Phi_c$ at a local maximum (more integrated than any larger or smaller system overlapping it). Well below the threshold there is no experience at all. Between lies a narrow twilight, fixed by structure, where there is no exact fact of the matter.

This is the first of Holos's two additions to physics. In Holos without collapse it is not a force, a field, or a change to any equation, but a structural fact about where observers occur. In Holos with collapse, it may also be where collapse happens. The requirements are listed under The Threshold and Observer Requirements. The measure and the value of the threshold are open problems.

Axiom 4: Two Sides

In a system that meets the observer requirements, experience and the system's physical activity are two sides of one event, physical activity seen from outside and experience lived from inside. Neither side reduces to the other, and in such a system neither occurs without the other.

This axiom rejects both substance dualism, which makes experience and activity two different kinds of thing (as Descartes held of mind and body), and strict reductionism, which says only the outside is real and experience is nothing over and above it. It is Spinoza's picture of mind and body as two aspects of one substance, restated for integrated systems. Its modern relatives are dual-aspect monism, on which mind and matter are two aspects of one underlying reality, and Russellian monism, on which physics describes only structure and experience is the intrinsic nature that structure has. Holos differs from both in scope. They typically give some inside to all matter, and Holos gives one only to structures above the threshold. It also fixes how strong the link is. Since there is one event, a perfect copy of the outside is a copy of the inside in any possible world, under any laws at all, not just ours. And it answers the charge that experience does nothing. Experience adds no force to physics. It is the inside of the physics, so whatever an observer's activity causes, its experience causes too.

It also says exactly what physics leaves out. A purely physical description fixes everything that happens but states it only from outside, and its vocabulary has no way to say that any of it is lived. The gap is in the description, not in the world. A floor plan records every wall of a house and still cannot say what living there is like. Build the house exactly from the plan, and it is livable all the same. Philosophers call this the phenomenal concept strategy. The gap lies between two ways of describing one thing, not between two things. The strategy was developed by physicalists. Holos borrows it without their conclusion. For them the inside description picks out something grounded in the outside; for Holos neither side is grounded in the other (see below). Its standard challenge is that the inside way of describing then needs explaining. Holos accepts that debt as part of the hard problem it does not answer.

What kind of view this is. Picture a curved line, concave from one side and convex from the other. It is one line with two true descriptions, and neither side comes first. Holos says the same of an observer's activity and its experience. It agrees with physicalists on the facts. Physics fixes everything that happens, so a perfect copy of the outside is a copy of the inside. It differs on two points. The first is the description. Physics, built only from relations (Axiom 1), states every fact from outside and has no way to say that any of it is lived. The second is the direction. Physicalism holds that the inside is grounded in the outside, the way a table's solidity is grounded in its atoms, while Holos holds that neither side is grounded in the other. And it differs from dualism, which makes them two things. Holos is a two-sided monism with a threshold. There is one world and one kind of event, and where a system crosses the threshold, that event has an inside as well as an outside. Some philosophers will still file it under physicalism, because copies match in every possible world. Holos accepts the resemblance and rests its difference on direction.

Axiom 5: Totality

The whole, Omega, is the one experiencer. Physically, the whole is the universe's complete quantum state. On the no-collapse reading of Axiom 2, that is one state of everything, all branches included. On a collapse reading, it is the same universe with a single history. Every observer is an aperture of it. Where a system crosses the threshold, a new self begins. No new experiencer does (D8).

This is the second of Holos's two additions to physics. What it adds is not the whole itself, whose existence rests on physics, but the claim that the whole is the one experiencer. That part is interpretive and can never be tested. Its job is unity. Every self is lived by the same experiencer, walled off from the others by structure. It does not explain where a self begins or what it is like. Axiom 3 and the structure do that. Its payoff is philosophical, not experimental. It gives one answer to why you are this self rather than another, and to which of two perfect copies is you. Holos does not claim it settles what to value. The rivals and the payoff are weighed under Why One Experiencer Has Many Walled-Off Perspectives. The view is known as open individualism. Omega is not an agent and does not intervene (see Totality).

Proposition I: Structural Relational Realism

Follows from Axiom 1.

Physical reality is made of relationships between things, not of objects with intrinsic properties of their own, prior to any relation.

What scientific theories successfully track are stable patterns of relation. Changes in interpretation or ontology (what a theory says exists) matter less than preservation of relational structure.

This proposition does not deny the existence of objects. It denies that objects are prior to the relations that define them. Nothing exists fully formed before its relations.

Like Axiom 1, it concerns what physics describes. Experience is not an item in the relational structure but its inside where an observer exists (Axiom 4), so the proposition does not reach it.

Proposition II: Manifestation

Follows from Axioms 3 and 4, with the definitions in D7.

Observation is not passive recording. It is where informational structure is lived. It is a matter of structure, not a process in time.

This manifestation is structural, not causal. In Holos without collapse, observation does not generate physical events or alter lawful dynamics. It marks which already-consistent structures are lived, the ones that contain observers. It picks no outcome and erases none.

Physics specifies structure. A sunset over a planet no observer ever sees and a sunset someone watches can be the same light. The difference Holos marks is that the second is lived. It is reality, not only structure.

Proposition III: Global Consistency

Follows from Axioms 1 and 2, with relativity.

If spacetime is treated as a complete four-dimensional structure (all of space and all of time taken together as one whole), consistency is a property of whole histories, not something enforced moment by moment.

A history is consistent the way a completed solution is. Every part fits every other part. Observation adds no constraint to this. It registers histories that are already consistent. Nothing here requires backward causation or signaling.

Eternalism (the view that past, present, and future are all equally real) and the relational commitment describe different layers of the framework. The block universe (spacetime pictured as one complete four-dimensional whole) is the structural layer. It is absolute, observer-independent, and tenseless, with no moment picked out as the present. It is $C$. Registered facts live within it, indexed to the observers the block contains. There is no tension between an absolute geometry and relational facts of experience, because they are claims about different things. The block describes what is consistent, and registration determines what is lived.

Nor is the block a perspective from nowhere imposed on top of observers. Relativity shows that observers moving differently disagree about which distant events are happening "now", so it picks out no privileged present. That motivates the block universe. It does not prove it, and some philosophers still defend a moving present. What relativity leaves is the invariant relational structure (what every observer agrees on) that all perspectives share. The block universe is what remains when every observer's perspective is taken into account. In that sense eternalism is not in competition with relationalism (Axiom 1's view that structure is relations). It is its structural expression.

Global consistency is not enforced from an external vantage point, and it is not deferred to an unreachable limit. It cashes out operationally, here and now. Whenever two observers compare records, their records agree. Physics secures this on its own for ordinary records, which the environment copies many times over. Records that meet are carried by the same signals. Agreement is subtler only in exotic setups where one observer is kept perfectly sealed off from everything (extended Wigner's-friend experiments). See Check C. The monist reading (the whole as one experiencer) is not needed for it and claims no role in it.

Apparent retrocausal effects (effects that seem to run backward in time), as in the delayed-choice quantum eraser, an experiment in which a later choice seems to decide an earlier result, come from sorting records after the fact. Nothing travels backward in time, and in the version defended here no influence travels faster than light.

Proposition IV: Closure

A principle of method and an interpretive image, not derived from the axioms. It is how Holos reads infinities.

Infinities usually mean a description has broken down, not that nature is infinite, as the ultraviolet catastrophe showed. Around 1900, physics predicted that any hot object should glow with infinite energy, and the infinity was fixed by a better theory, not found in nature.

Often the repair is a description one level up, which holds as a closed whole what the level below could only see as endless or unfolding. Picture Flatland. A sphere passing through a flat world looks, to its flat inhabitants, like a dot, then a growing and shrinking circle, then nothing, an event in time. Seen from three dimensions, it is one sphere, all at once. The same step repeats. In three dimensions, time flows. In four-dimensional spacetime, a whole history is one fixed shape. Within spacetime, a quantum measurement seems to pick one outcome. In the far larger space where the quantum state lives (a mathematical space with a separate direction for every way the universe could be arranged), every branch is held at once, on the picture Holos defends. Some philosophers take that vast space to be the most real level of all, a view called wave-function realism. Taken all the way, closure gives Omega, the description with nothing outside it, where everything is held whole.

This is closure between descriptions, not a path between places. Extra dimensions of space are not somewhere to go. With more than three large dimensions, atoms and orbits would not be stable. Nor do observers move toward Omega, which exists now, and part of which is never lived. Each higher description closes what a lower one leaves open. Holos reads this as the shape of explanation, not as a claim that the higher levels are hidden places.

In Holos, infinities signal the limits of a model, not literal features of reality.

The threshold in three claims. The threshold bundles three claims of different strength, and the first does not establish the others.

1. Edge: whether anyone is home has clear cases on both sides and a narrow twilight between. Part of Axiom 3.
2. Integration: what carries a system across is integration, together with the other observer requirements. The core claim. Test A is where it can lose.
3. Transition: the crossing is a genuine transition, the kind physics studies in magnets and fluids, not an arbitrary stretch of a smooth slope. It could be sudden or continuous, and Holos takes no side on which. A hypothesis. Its evidence must come from the boundary (see A path to the threshold).

If claim 3 fails, the core stands. The twilight is simply wider than hoped.

What Omega Is and Is Not

• It is the whole, physically the universal quantum state and experientially the one experiencer.
• It is not an endpoint in time or a limit that finite systems approach. It exists now, and no system is heading toward becoming it.
• It is not an external observer watching the universe from outside. It is the whole itself, experiencing through the apertures the universe contains.
• It is not fully registered. Structure outside every aperture's causal past is part of Omega and never lived.
• It is not an aperture writ large. As a whole, the universe splits into parts that never touch, regions beyond each other's horizons, so it is never a peak of integration and the maximality condition never picks it. It is not a self.
• It does not replace physical cosmology or impose a final cause on evolution.

Lineage and Interpretation

Different traditions have described this whole in different vocabularies, such as Advaita Vedanta's one experiencer behind every eye, Spinoza's single substance, Berkeley's never-absent perceiver, panentheism's world contained in the divine. The monist reading Holos adopts stands in their line, restated in informational terms, with one break from Berkeley. On Holos, part of the whole is never lived. Its nearest modern relative is open individualism, Daniel Kolak's view that we are all the same subject. The name Omega echoes Teilhard de Chardin's spiritual endpoint of history and Frank Tipler's physical Omega Point, a prediction that needed the universe to collapse back on itself. Holos means neither. Its Omega is not an endpoint but the whole, now, which is why it drops "Point."

What Holos leaves open is vocabulary, not structure. Calling the whole God, Brahman, or simply the whole changes nothing about the claim. What Holos does not offer is the fully deflationary reading in which Omega is only a mathematical horizon and finite observers are self-standing. Holos takes the direction of dependence to run from the whole to its parts, for the one reason the monist reading earns, unity. On it, crossing the threshold adds a new self, not a new experiencer (see Why One Experiencer Has Many Walled-Off Perspectives).

Why Dependence on Apertures Is Not Circular

An objection follows immediately. If Omega is fundamental, why does it require apertures, contingent arrangements of matter, in order to experience anything? The ground of experience appears to depend on what it is supposed to ground.

The objection conflates two kinds of dependence. Omega does not depend on apertures existentially, that is, for its existence. It is the whole, and it is what it is whether or not any region of it folds into an integrated perspective. It depends on them for its experience. They are the places where experience occurs. Priority claims (claims that the whole comes before its parts) concern the first relation. The aperture claim concerns the second. Only if the two were the same relation would there be a circle.

The decisive point is that an aperture is not an external thing granting experience to Omega from outside. It is a region of Omega. "Omega experiences only through apertures" therefore unpacks to "Omega experiences through its own structure, where that structure permits." That is self-dependence, which is not vicious (not a circle that defeats itself) but simply what it means to have a structure at all. An organism sees only through its eyes. This does not make its eyes prior to it. The analogy carries one warning. There is no subject positioned behind the aperture receiving a feed. The aperture is where the experiencing happens, not a window onto a viewer.

Whatever lies outside the causal past of every aperture, whole unlit branches among it, marks a genuine limit on the whole's experiential reach. That is unlit structure, existing as pattern and never lived. Holos does not soften this into a faint universal experience. Doing so would erase the distinction between lit and unlit on which the rest of the framework depends.

Why One Experiencer Has Many Walled-Off Perspectives

A second objection is the mirror image of a famous one. Panpsychism faces the combination problem. If every particle carries a spark of experience, no one can say how billions of sparks combine into the single unified experience of a person. A monism of one experiencer inherits the problem upside down, as a decomposition problem. If every experience is Omega's, what makes your experience and another person's experience two, and why is the wall between them absolute? No aperture has any access, faint or partial, to what it is like to be another. Saying the walls exist names this situation. It does not yet account for it.

The problem has a name in the literature. It is the decomposition problem facing priority cosmopsychism, the view, defended by Philip Goff and others, that the cosmos is the one fundamental conscious subject. Holos differs in a way that matters here. Goff's cosmos has an experience of its own as a whole, from which individual experiences must be carved. Holos claims no such cosmic experience and denies none, so it never has to carve. What it must explain is only the walls.

The first half of the account is structural. Two apertures are two because they are two local maxima of integration with no integration bridging them. The same maximality condition that individuates observers settles what makes them several. And the wall between them is not a barrier holding something back. Experience occurs at the aperture and is shaped by it. There is no subject positioned behind the apertures through which a back-channel could run. Where structure does not connect, experience does not connect. The wall is not a mechanism added to the plurality. It is the absence of any structure that could carry connection.

The second half dissolves the air of paradox. One experiencer with mutually walled-off experiences is not exotic. It is what time already makes of every individual life. A person at five and the same person decades later are one experiencer. No one takes their separation to split them into two people. Yet the later moment has no direct access to what it was like to be inside the earlier one. In the block universe, all the moments of a life coexist tenselessly, each experienced from within itself, none experienced from within another. Walled-off plurality inside a single experiencer is therefore already the ordinary structure of a human life. Apertures stand to the whole as the moments of a life stand to the person, genuinely many, genuinely walled off, and one.

Medicine offers a case. When the link between the brain's two halves is cut, each half can know things the other does not. One brain holds two walled-off streams. Holos says the same of every self.

Nor is the oneness a label doing no work, though its work must be stated carefully, beside its rivals. On a picture of many self-standing observers, each system that crosses the threshold produces a new subject. On the monist picture, a new self begins and no new experiencer does. Stated that way, the difference is only a count. Both pictures leave the same thing to the threshold, namely why a perspective appears here, with this character. The difference shows in two answers. Of all the observers there are, why is this one me? Nothing to explain, because the one experiencer is each of them. If a perfect copy of you were made, which would be you? Both, with no remainder.

The rivals deserve a fair statement. On the first question, many philosophers hold that "I" works like "here". It picks out whoever is speaking, so no one owes an explanation of why they are themselves (Perry 1979 is the classic source). On the second, Derek Parfit argued that identity is not what matters in survival, so the copying case needs no single answer about which one is you (Parfit 1984). These replies are serious, and Holos does not claim to refute them. It claims only that the monist reading answers both questions at once, with no leftover facts about which self is which.

The unity does not do everything. It does not explain why records agree where observers meet. Physics does that, since the same signals reach them both. Nor does it explain where a self begins. That is the threshold's work, and it would be the same brute structural fact on any picture. The unity's work is philosophical, not experimental. It gives one answer to the identity questions, bought at one price, the decomposition problem answered above. Holos does not claim it settles what to value. Holos judges the trade worth making. A reader may not.

The analogy has a stated limit. The moments of a life are threaded together by memory and anticipation. Apertures share no such threads. The analogy shows that walled-off plurality within one experiencer is coherent, not that apertures are moments. What it removes is the charge of incoherence, which is all it is asked to do.

Holos among Views of the Self

Philosophers sort answers to "what am I?" three ways. On closed individualism, you are one self from birth to death. On empty individualism, a new self exists each moment. On open individualism, there is one subject in everyone. Holos takes the third, and calls the local person a self and the one subject the experiencer (D8). The table shows what that changes, and what each view pays.

Decoherence is not presence

Decoherence is what happens when a quantum system becomes entangled with its surroundings. Its wave-like interference effects spread into the environment and can no longer be seen. It explains why quantum systems appear classical (like ordinary objects with definite properties) at everyday scales. Holos does not dispute this.

The Holos claim is that decoherence alone does not produce a lived world. It produces a consistent classical-looking structure. Presence requires integrated observation.

Fields are structure. A detected particle is a record

In quantum field theory (QFT), the basic ingredients are fields spread through all of space, and particles are their quanta, the smallest possible ripples in a field. Detectors do not directly observe fields. They record discrete outcomes, such as clicks, tracks, and energy deposits, because measurement is an interaction that entangles (links the states of) a spread-out excitation with a detector, and within each branch the detector's record shows one definite event in a specific place and time. Nothing collapses. Each branch simply holds its own record. Holos reads the particle as seen, a click at one place and time, as a record of a field interaction rather than a fundamental object. On that reading, particles are context-dependent records of field interactions, which is why a continuous theory can yield discrete observations without requiring reality to be made of little beads.

Conservation and branching

Axiom 2 commits Holos to a branching picture. The quantum state evolves smoothly and reversibly, never collapsing (unitary evolution), and no possibility is erased. When observers would register incompatible outcomes, they are situated in different branches of the one quantum state, each internally consistent. In this respect Holos sides with Many-Worlds-style interpretations of quantum mechanics (Hugh Everett's reading, on which every outcome occurs in its own branch), while adding what they leave out, an account of which structures are lived. This is a genuine interpretive commitment, not a neutral stance.

The Born rule: weighted branches

A branching picture owes an account of quantum probability. If every outcome occurs in some branch, what does it mean that one outcome is measured 70% of the time? Counting branches gives the wrong answer. The Born rule (take the number the quantum state assigns each outcome, its amplitude, and square its size) gives the right one in every test so far. Any framework that keeps all branches must say where those weights live and why they take the form they do.

Holos answers narrowly. The weights are structural facts, part of $C$. The possibility space is not flat. Branches carry weights, as absolute and observer-independent as the laws themselves. And they are not arbitrary. Gleason's theorem (1957) shows that, given a few mathematical assumptions (chiefly, that an outcome's probability does not depend on which other outcomes it is grouped with), the only consistent way to assign probabilities to the outcomes of quantum measurements, across the different ways the same experiment can be carved up, is the Born rule. That constrains the answer. It does not by itself say how an observer inside a branch should set their odds, which is the question a branching picture has to answer.

Holos declines one tempting reading. It does not read a branch's weight as its share of the whole's experience. That reading would require experience to be a quantity held in common and divided among branches. Holos makes no such claim. Experience occurs at apertures. There is no reservoir from which branches draw larger or smaller portions, so the question of which branch receives more of it does not arise.

What the weight does measure is odds. After a measurement splits the world, but before you look, there are observers in every branch, and you are one of them without yet knowing which. The weight of a branch is your odds of being among its observers. That is why an outcome with 70% of the weight turns up about 70% of the time in long runs of the experiment. This is not because the other outcomes fail to happen, but because almost all of the weight lies with observers whose records show roughly that frequency. Observers with strange records exist too, like one who sees a fair coin land heads a hundred times in a row, but they carry almost none of the weight. The reading keeps the weights where Holos placed them, in the structural layer. The odds are a fact about the possibility space, not a quantity of experience.

The monist reading might seem to dissolve this uncertainty. If the one experiencer is each of the observers in every branch, asking it which branch it is in gets the answer "all of them," and nothing is left to be uncertain about. But the question is asked from inside an aperture, and apertures are walled off. None has access to what the others register. Each therefore genuinely lacks the information of which branch it is in, and the odds answer that lack. Two questions must be kept apart. "Why am I this one?" asks for a reason, and on the monist reading there is none to give, because the one experiencer is each of them. "Which one is this?" asks for information, and it stays open until you look. You can be everyone and still not know what is behind the next door.

A worry follows from yes-or-no status. Lived status is yes or no and ignores weight, so a branch with a tiny weight that contains an observer is fully lived. Counted one by one, most branches of many repeated experiments show roughly even results whatever the weights, so if every lived observer counted equally, a typical observer should expect the wrong statistics. But the worry depends on counting, and counting fails here. Seventy percent of an infinite crowd is exactly as large as thirty percent of it. Try to count branches, or the observers in them, and you get either an arbitrary number or an infinite one, and there is no fair lottery over infinitely many tickets. Each ticket's chance would have to be zero, and the zeros would never add up to certainty. Holos treats an infinity as a sign that a description has broken down, and this is such a case. Counting is the wrong tool. A single point on a line has no length, and a short stretch has as many points as a long one, yet the short stretch is shorter. Each observer is a point. The weights are the lengths.

Ruling out counting does not by itself pick the replacement. Every region of a dartboard holds infinitely many points, yet where a dart lands depends on how it is thrown. A random throw follows area, and a skilled player aiming at the center does not. Weight is the candidate physics itself supplies. It is built into the quantum state, it is conserved over time, and it does not change when a branch splits further. No rival on offer has all three. The last step is a principle, that your odds should follow the measure physics conserves. Sebens and Carroll state one version of it, a rule about how evidence bears on where you are, which they call epistemic separability. Roughly, what you should believe about where you are cannot depend on things far away that do not affect you (Sebens and Carroll 2018). It is widely discussed and still debated.

So this is an elimination plus one principle, not a derivation from first principles. Holos adopts the self-locating reading of the weights and inherits its open questions, whether the principle is justified, and whether Gleason's assumptions hold. Other published routes reach the same rule, among them decision-theoretic arguments; Holos depends on none in particular, and it adds no new mathematics here.

State space

Let $S$ denote an informational state of the universe at some level of description. $S$ is not assumed to be complete or fundamental. It is simply whatever structure physics provides.

Creation

Creation ($C$) maps a given state to the branches physical law produces from it. It represents lawful possibility. Seen from inside any one branch, the others are what could have happened, and all of them do.

Read: $C(S)$ is the set of every branch $S'$ that physical law produces from the state $S$.

This notation is schematic. It does not assume a discrete branching structure or a specific ontology of histories.

Observation

Observation ($O$) maps a space of possibilities to registered experiential histories. It represents internal registration by integrated systems, wherever the possibility space contains them.

Read: applying $O$ to those branches gives a collection of experienced histories, one for each observer, labeled by $i$.

The result is an indexed family (a labeled collection), not a single selected outcome. It holds one experienced history per registering perspective, per branch. Observation selects nothing and erases nothing (Axiom 2). Registration occurs everywhere an aperture exists, and no thread is privileged. Following one observer's thread, what happens next is simply what physics allows from that history onward; registering it changes none of those possibilities. The Born weights carried by $C$ remain structural throughout. They set the odds of each record, not how much experience it holds.

This mapping is not assumed to be random, deterministic, or computable in general. Holos requires only that each experienced reality corresponds to a single consistent history, one per registering perspective, internally coherent, and in agreement with every record it is compared against.

The Holos relation

The Holos relation is the composition of Creation and Observation.

Read: lived reality, $R$, is Creation composed with Observation.

This expression states that lived reality is neither pure possibility nor pure observation. It needs both. Applied to a state $S$, the result, written $R(S)$, is the family of lived perspectives, one per observer per branch, each with the lit world it draws on. Structure exists whether or not it is lived. $R(S)$ marks where it is also lived, and in Holos that is what reality means.

The symbol ⊛ is defined as composition: $C ⊛ O$ is the composite operation “possibility, then registration.” Applied to a state $S$, it reads $R(S) = O(C(S))$: $R$ names the relation, and $R(S)$ the lived reality it yields from that state. It is ordinary function composition, with the order of the steps carrying the meaning, possibility first and registration second. The order is logical, not temporal. Registering needs something to register.

In standard notation this is simply $R(S) = (O \circ C)(S)$, where $\circ$ is the usual symbol for composition. Apply $C$, then apply $O$. Standard composition reads right to left, so $C \circledast O = O \circ C$, the same composition written in reading order. Nothing is hidden in the glyph. Holos retains ⊛ because the framework is named for the relation it marks, not because the operation it denotes is unusual, and a reader who mentally substitutes the composition symbol loses nothing.

What the symbol adds is not mathematics but ontology. It claims that both steps are required for a lived world, and that neither step alone yields one.

The measure of integration

Holos does not yet name a privileged measure of integration. Competing proposals for computing $\Phi$ can disagree, not only about values but about which of two systems is more integrated. Locating the threshold and its twilight requires a determinate measure, and identifying it is an open problem for the framework, not settled background. Holos is committed to the threshold being fixed by structure. It does not yet know how to compute it. The stakes reach back into the framework's core argument, the identification of a structure's oneness with causal integration (see Why Integration) remains an argued direction rather than a finished quantity until the measure is fixed. The problem also includes the question of level. Holos measures integration in physical parts, grouped the way they actually work together. The grouping may lie above the smallest parts, as a neuron's firing lies above the motion of its ions. A program counts only through the chip that runs it. Loops in the code are needed, but the chip's own parts must also act as one, and conventional chips may not. Saying precisely where that level is belongs to the same open problem.

Where the threshold falls

Where any given system's threshold $\Phi_c$ and its twilight lie is unknown (see The threshold in three claims). Because presence itself cannot be detected directly, no experiment can locate it by direct measurement. Its placement is constrained only indirectly, by which systems show the structural signatures of observation (see Test A), and our estimates may stay blurrier than the twilight itself. Holos accepts this as the price of a threshold that is structural rather than behavioral.

A path to the threshold

The two problems above are not invitations to despair. They are a research program. Science pins down unobservables by triangulation all the time. No one has ever seen an electron's charge, yet independent methods converged on it and every proposal that disagreed was discarded. The same logic applies here, in three steps.

Treat the threshold as a transition, not a dial. If the passage from distributed processing to a unified perspective is a genuine transition, then $\Phi_c$ is not a number we are free to tune, and transitions leave measurable fingerprints. A sudden one typically shows a lag, where the way in and the way out do not match. A continuous one typically shows slowing and growing fluctuations near the boundary. Either kind shows a twilight that narrows as systems grow. Clinical consciousness research has already found one such boundary from the outside, the Perturbational Complexity Index (PCI). It gives the brain a brief magnetic pulse and measures, with scalp electrodes (EEG, a recording of the brain's electrical activity), how complex the echo of activity is. In the studies that set it up, a single cutoff, found by experiment, separated conscious from unconscious states across anesthesia, sleep, and disorders of consciousness (lasting states after brain injury), without anyone measuring presence directly. Locating $\Phi_c$ is that kind of problem, not a metaphysical one.

The fingerprints are already being measured, but they must be read with care. Several lines of evidence suggest that waking cortex runs near a critical point, a tipping point between too-orderly and too-chaotic activity, often called the edge of chaos. Anesthesia, generalized seizures, and disorders of consciousness move it away, while psychedelics move it closer and make its activity richer (Toker et al. 2022). Measures of criticality in resting EEG predict anesthetic loss of consciousness and track PCI (Maschke et al. 2024). In cortical tissue, cascades of activity called neuronal avalanches come in all sizes and follow power laws (large cascades are rarer than small ones in a fixed mathematical proportion), with exponents that match a known universality class, a family of systems that behave alike near their tipping points (Beggs and Plenz 2003). Whether the brain is exactly critical or only close to it is still debated, and power laws can arise in other ways. Taken together, these results suggest that a critical point sits at the center of conscious life, not at its edge. Consciousness is lost by moving away from it in either direction. Holos reads them through two separate requirements. Integration asks whether the parts are joined into one whole, and that is where the threshold lies. Differentiation asks how varied the states of that whole can be, and variety peaks near criticality. A seizure locks every part into one rhythm (Meisel et al. 2012), like a stadium chanting a single word, joined but with almost no variety. So these studies show where observers operate and why richness peaks there. They do not yet show that crossing the threshold is itself a genuine transition. That evidence must come from the boundary.

A cutoff that sorts patients is not yet proof of a transition, since a smooth quantity can be cut anywhere. What would count is a transition's own signature, and here the evidence is mixed in an informative way. Animal studies find that consciousness is lost and regained at different anesthetic levels. The way in and the way out do not match, a lag known as neural inertia. A lag of that kind is the hallmark of an abrupt switch, the kind physicists call first-order, like water freezing. If the lag belongs to the transition itself rather than to how drugs enter and leave the brain, it fits the transition hypothesis (claim 3) as an abrupt switch. But in flies the lag is controlled by genes that regulate sleep, and single mutations can erase it (Joiner et al. 2013), so it may belong to the brain's arousal switch rather than to integration. In humans the evidence is mixed. A controlled test found no lag with propofol and some with sevoflurane, two common anesthetics (Kuizenga et al. 2018), and in 393 surgical patients the brain's slow-wave response (the large, slow brain waves of deep anesthesia) differed between going under and coming back (Warnaby et al. 2017), but the lag appeared in the EEG, not in responsiveness. Slowing and growing fluctuations measured at the boundary itself, as consciousness is lost, would point the other way, toward a continuous transition, though slowing can also come just before some abrupt switches. The criticality studies above measure the operating point, not the boundary, so they do not yet supply this. Neither signature settles the question alone, but together they help tell the two kinds of transition apart, which makes the question testable. In a finite system like a brain, either kind appears as a steep, rounded curve rather than a mathematical kink, and that rounding is the twilight. The search is for how it scales, not for a perfect step.

Cull the measures by convergence, then try to force uniqueness. Holos does not need one anointed measure so much as it needs the adequate measures to agree on the cases that matter, and that is testable now. Run the candidate measures across a battery of systems and keep the ones that rank the anchor cases (systems whose status is not in doubt, such as a waking adult and a deeply anesthetized one) the same way. The stronger move is to derive the measure rather than choose it. State the properties any measure of a single unified perspective must have, and see whether consistency forces a unique form, as Gleason's theorem constrains the Born weights. Even narrowing the field to a small equivalence class (a family of measures that always agree on which system is more integrated) would remove most of the arbitrariness.

Solve them jointly. The measure and the threshold constrain each other. A candidate pair must simultaneously fit the anesthesia boundary, the emergence of experience in development, split-brain dissociations, and wherever minimal neural systems first show coherent integration. Coupled constraints are far more rigid than separate ones. The band that satisfies all the anchor cases at once is narrow. This is what Test A really is, not a single yes-or-no experiment but a calibration engine that locates the threshold and validates the measure. The two jobs must use separate data. Calibration data locate the threshold, and held-out data, named in advance, test it. Testing on the calibration set would only grade the answer key.

The field already has a model for running such tests. The COGITATE adversarial collaboration (Nature, 2025), in which rival theorists agreed on a test in advance, pitted IIT against global neuronal workspace theory (which ties consciousness to information broadcast widely across the brain), and its results challenged key claims of both. Test A should follow the same design. A caution applies to the measure itself. In a letter first circulated in 2023, more than 100 researchers called IIT pseudoscience, a debate aired in Nature Neuroscience in 2025. The charge has two main targets. One is the theory's untestable core claim that its integrated structure simply is consciousness, which Holos does not adopt. The other is the implication that simple grids of logic gates are conscious, which Holos's aboutness requirement rules out (see Observer Requirements). Holos borrows Φ only as a measure.

Two conditions bound the program. First, its anchor is report. The human case is the one place inside access exists, so the threshold is located relative to us and carried outward by the measure, with certainty that weakens as the cases grow alien. That is the shape of all consciousness science, not a defect peculiar to Holos. Report can also miss consciousness. About one in four brain-injured patients who cannot respond show, on brain scans, that they are following instructions (Bodien et al. 2024), so a missing report is not proof that no one is home. Second, the program can fail, in two ways of different weight. It could find no transition at all, integration climbing smoothly. That would falsify claim 3 and leave the twilight wide, a slow dawn rather than a quick one, with the core intact. Or it could find that experience does not track integration on any candidate measure, which is how Test A loses, and the core with it. A stated way to lose is what makes these open problems scientific questions rather than definitions.

Where one observer ends and another begins

The maximality condition (see Exclusion, under Observer Requirements) makes only a local maximum of integration an aperture. It inherits the unsettled question of which measure defines the maximum, and it lacks a stated procedure for which candidate systems are compared and how overlapping candidates are resolved. IIT spells out such a procedure. Holos has not yet adopted or replaced it. Until the measure and the procedure are fixed, the boundaries between observers are fixed only in principle.