On Time

Why time is constituted by observers, not inhabited by them

Time is the most familiar thing in physics and the least understood. We measure it, model it, and build every dynamical law on top of it — but we never explain where it comes from. The framework does not reinterpret time. It derives it. What emerges is not the time of Newton or Einstein, but something more fundamental: a structural ordering created by the observer network itself, with an arrow that is built in from the start, not bolted on after the fact.

The Standard Picture

Newton gave us absolute time: a universal clock ticking in the background, indifferent to what happens in it. Einstein replaced this with relative time — clocks run at different rates depending on motion and gravity — but time remained a geometric ingredient, a dimension of a manifold assumed from the outset. Quantum mechanics uses time as an external parameter, the variable in Schrödinger’s equation that the wavefunction evolves along. In every case, time is presupposed. The equations describe what happens in time, not what time is.

This leaves two deep questions unanswered. First: why does time have a direction? The fundamental equations are time-symmetric — they work equally well forward and backward — yet the world clearly runs one way. Second: why is there a temporal ordering at all? What gives events their “before” and “after”?

The framework answers both questions at once, from the same source.

Time as Phase Ordering

Every observer in the framework is a cyclic structure — a U(1) loop that advances phase with each internal cycle (Axiom 3). When two observers interact via phase transfer, the interaction has an inherent directionality: the positive coherence cost of the transfer prevents backward flow. Each such interaction creates a directed link between two events.

The collection of all directed links across the entire observer network forms a directed acyclic graph — a web of events connected by “this happened before that” relations. The resulting partial order is not a metaphor for time or an approximation of time. It is time — the complete operational content of temporal ordering, derived from the axioms without any temporal parameter being assumed.

This has a striking consequence. A universe with no observers would have no interaction graph and therefore no partial order. Time would not exist — not merely because there would be nobody to notice its absence, but because the ordering structure is constituted by the observer network. Time is constituted by observers, not inhabited by them.

At scales much larger than the Planck length, the partial order approximates a smooth Lorentzian manifold — recovering the spacetime geometry of general relativity as a large-scale emergent description. The framework derives the causal structure that causal set theory takes as primitive.

The Arrow Is Built In

In standard physics, the arrow of time is a puzzle. The fundamental equations are time-symmetric, so the asymmetry must come from somewhere else. The usual answer is statistical: the universe started in an extraordinarily low-entropy state, and entropy has been increasing ever since. But this just pushes the question back — why did it start that way?

The framework dissolves the puzzle by building the arrow into the causal structure itself. Type III interactions generate relational invariants — permanent records of correlations between observers. Once generated, a relational invariant cannot be un-generated. This is not a statistical tendency. It is a structural fact: the invariant is a node in the dependency graph, and nodes do not disappear from graphs.

Along any directed path in the interaction graph, the depth of relational invariants is monotonically non-decreasing. The arrow of time points in the direction of increasing structure — more correlations, more entanglement, more recorded interactions. This direction is irreversible by definition, not by thermodynamic tendency.

The Loschmidt paradox — how can irreversible macroscopic behavior arise from reversible microscopic laws? — dissolves cleanly. Time-reversal symmetry is preserved in the dynamics of individual interactions. The irreversibility lies not in the dynamics but in the accumulation of relational invariants, which is a structural feature of the graph, not a dynamical one.

Entropy Without Statistics

The standard account of entropy is statistical. A system has many microscopic configurations (microstates) compatible with its observable properties (macrostate). Entropy counts these configurations. The second law says the system tends toward macrostates with more microstates, because there are more ways to be there. This is compelling but relies on assumptions — ergodicity, typicality, a well-defined phase space — that are difficult to justify from first principles.

The framework replaces this with a structural account. Entropy for an observer is defined as total coherence minus that observer’s accessible coherence. Coherence is conserved (Axiom 1), but every observer is bounded — it cannot access all the structure in the network. As new relational invariants are continuously generated by Type III interactions, some inevitably fall outside any given observer’s domain. The gap between total and accessible coherence grows.

The second law follows from two facts: coherence is conserved, and every observer is bounded. No statistics, no ergodicity, no phase-space counting. The entropy increase is as inevitable as the fact that a finite window cannot see everything in an expanding landscape.

This has a remarkable corollary: the universe taken as its own observer — the unbounded case — has entropy zero. Total coherence equals accessible coherence when nothing is outside your domain. The second law is a bounded-observer phenomenon, not a cosmic law. Heat death is something that happens to observers, not to the universe.

The Cosmological Arrow

Roger Penrose pointed out that the early universe’s low entropy requires an explanation. The probability of the initial state, measured against the full phase space, is something like one in 10¹⁰¹²⁰ — a number so small it dwarfs any other fine-tuning problem in physics. Why did the universe start in such an astronomically special state?

The framework’s answer: it didn’t. The early state was not improbable. The concept of “improbable relative to the full phase space” is inapplicable because the system was never ergodic in the first place. The early universe occupied the only sector that was accessible — the small set of configurations compatible with the bootstrap hierarchy at that epoch. There was nowhere else for it to be.

As the universe expanded and cooled, new levels of the bootstrap hierarchy became stable — first nuclear bound states, then atoms, then molecules, then macroscopic structures. Each new level opened a vast new region of accessible phase space. The entropy of the universe, measured by any bounded observer, grew monotonically — not because the system was exploring a pre-existing phase space, but because the accessible phase space itself was growing.

The structural arrow (increasing relational invariant depth), the thermodynamic arrow (entropy growth), and the cosmological arrow (expansion) all point in the same direction. This is not coincidence. Expansion enables hierarchy elaboration, which creates the causal structure that defines the temporal order. The three arrows are aspects of a single process.

The Quantum of Time

Action is the coherence cost of a transformation. Every path between two interaction events carries a coherence phase proportional to its action. Physical trajectories are selected by coherence resonance — paths where neighboring paths have nearly equal phase and constructively interfere. This yields the stationary action principle, not as a postulate but as a consequence of coherence accounting.

Planck’s constant is the minimum coherence cost of one observer cycle — the quantum of action, set by the loop geometry of Axiom 3. The Planck-Einstein relation E = ℏω falls directly out: an observer’s energy is its coherence cost per cycle, and its frequency is cycles per unit time. Energy and frequency are two descriptions of the same thing — the cost and the rate of phase advancement.

The uncertainty principle emerges from the same structure. The phase position of an observer’s loop and the cycle count of that loop are conjugate aspects of a single object — Fourier duals of each other. Knowing one precisely makes the other indeterminate. This is not a limitation of measurement technology. It is the structure of the coherence quantum itself — a consequence of the fact that a cyclic process cannot simultaneously have a definite position in its cycle and a definite number of completed cycles.

Not a Background, Not an Illusion

The framework’s treatment of time parallels its treatment of determinism and existence: the answer depends on the level of description, and both levels are real.

At the level of the complete relational invariant network, time is absent. The graph is atemporal — all nodes present, no flow, no before and after. This is the level at which the block-universe intuition is correct.

At the level of any bounded observer, time is constitutive and irreversible. The advancing coherence domain defines what facts exist for that observer. The partial order is genuine. The arrow is structural. The passage of time is the inside view of advancing through a dependency structure — real as experience, and grounded in the physics of relational invariant accumulation.

Time is not the stage on which physics happens. It is not an illusion projected by consciousness. It is the structural ordering that observers create by interacting — as real as anything in the framework, and as observer-dependent as geometry.