Time, Records, and Dimensional Interfaces

Correction

An earlier version of this essay claimed that time dilation follows from a universal three-dimensional embedding threshold: squeeze an observer's dimensional aperture below three and its clock must slow. That argument was wrong. A strictly monotone-radius spiral can encode cyclic phase and traversal history injectively in two dimensions, so the claimed threshold does not exist.

The universal threshold claim is withdrawn. This essay asks what survives once proper time, readable records, and subjective duration are kept separate.

This remains Part 3 of a series. Part 1 introduces projection as a modelling device. Part 2 discusses its limits in quantum-gravity intuitions.

Dimension glossary

• D: dimension of a model's full state space
• d: spacetime dimension in a physical theory
• k: effective dimension of a specified observer-interface channel

These numbers answer different questions. None can be substituted for another without a model connecting them.

Part I: The idea worth keeping

The useful intuition is about records. A system can revisit a recognizable phase without returning to the same history-enriched state. An interface may preserve the phase while losing which visit it is observing.

Write a recurrent source state as $(\phi,\tau)$:

• $\phi$ answers where in the cycle?
• $\tau$ answers which traversal?

I have called $\tau$ meta-time. The name is suggestive, but it needs a strict limit: $\tau$ is an epoch, history, or traversal coordinate. It is not a second physical clock. It supplies no duration, rate, metric, or arrow of time by itself.

A circular readout,

$F(\phi,\tau)=(a\cos\phi,a\sin\phi),$

keeps phase and discards epoch. Every traversal lands on the same circle.

A planar spiral,

$F(\phi,\tau)=r(\tau)(\cos\phi,\sin\phi),$

is injective whenever $r$ is strictly monotone. Radius stores traversal history. Two dimensions can therefore preserve global identity, even over recurrent motion.

That correction leads to a better question. A bounded spiral can keep each epoch mathematically distinct while making late epochs progressively harder to distinguish through noise. The relevant separation is the radial difference $|r(\tau+\Delta)-r(\tau)|$ compared with the readout uncertainty. For the logistic radius used here, $r'(\tau)\to0$ along the tail. More generally, a bounded monotone differentiable radius cannot maintain one uniform positive lower bound on $|r'(\tau)|$ over an infinite epoch axis. Identity can survive while uniform practical resolution does not.

The simulation compares two deliberately chosen records:

This is the meta-time spiral in its defensible form. A finite interface can preserve the order of visits while spending range and statistical resolution to do it.

Three questions that should not be collapsed

The old essay treated three questions as one:

1. Identity: can two source states map to the same observation?
2. Local separation: how far apart are nearby history records in the chosen readout?
3. Operational resolution: do those distinctions remain usable across a bounded noisy channel?

The planar spiral can preserve identity while a bounded version loses uniform noisy resolution over an infinite horizon. There is no universal three-dimensional floor for recurrence, complexity, memory, or observers.

Three meanings of time

The original slogan also blended three meanings of “time.”

Proper time

Proper time is what an ideal physical clock measures along a worldline. In general relativity it is fixed by the spacetime metric:

$d\tau_{\mathrm{proper}}^2=-\frac{1}{c^2}g_{\mu\nu}dx^\mu dx^\nu.$

For a stationary observer outside a Schwarzschild black hole,

$\frac{d\tau_{\mathrm{proper}}}{dt}=\sqrt{1-\frac{2GM}{rc^2}}.$

This coordinate comparison tends to zero as a stationary worldline approaches the horizon. A freely falling observer still reaches and crosses the horizon in finite proper time. Calling the external view “time freezing” is a shorthand about signals and coordinates, not a literal halt in the infaller's local dynamics.

Special-relativistic time dilation follows from Lorentzian spacetime geometry. Length contraction and time dilation share the same Lorentz transformation, but that does not make either one a loss of effective state-space dimension.

Record time

A physical system can order events by changing a memory state: a counter increments, a chemical mark accumulates, or an environmental degree of freedom retains a trace. Meta-time can label that order in a model.

The label becomes physically meaningful only when a realization carries the record forward. Monotonicity in the model does not establish microscopic irreversibility or thermodynamic entropy production. Those require a concrete mechanism.

Subjective time

Human reports of duration depend on attention, arousal, memory, expectation, and brain state. Neural effective dimension may covary with some of these processes, but a covariance participation ratio is not a clock rate. Sleep, flow, anaesthesia, and psychedelic states should not be placed on a single high-to-low-dimensional axis without data and a task-specific definition of the measured neural state.

The distinctions matter because a change in records can alter retrospective duration without changing proper time, and relativistic proper-time differences do not require a theory of consciousness.

Part II: Where the mathematics stops

Relational clocks

Canonical approaches to quantum gravity often motivate a relational question: if a global equation has no external time parameter, can one subsystem serve as a clock for another? The Wheeler–DeWitt form is commonly written

$\hat H|\Psi\rangle=0.$

That equation does not prove that time is a correlation rate, and it does not select a dimensional-aperture law. Relational-clock constructions require choices about subsystem factorization, clock states, conditioning, and approximation. A restricted observation channel may degrade a chosen clock's readability. It does not follow that fewer accessible dimensions slow physical evolution.

Effective dimension

For a positive spectrum $\{\lambda_i\}$, a participation ratio

$k_{\mathrm{eff}}=\frac{(\sum_i\lambda_i)^2}{\sum_i\lambda_i^2}$

summarizes how evenly variation is distributed across modes. It can be useful for neural covariance, Fisher information, or another explicitly defined operator.

It is representation- and scale-dependent. Applied casually to coordinate components of a spacetime metric, it need not be coordinate invariant and cannot explain relativistic time dilation. At most, it describes anisotropy in the chosen representation. Any claimed link between $k_{\mathrm{eff}}$ and a clock needs an independent physical law and an invariant measurement procedure.

Fisher information

The Fisher metric

$ds^2=I_{ij}(\theta)d\theta^i d\theta^j$

measures local statistical distinguishability within a model. The Cramér–Rao inequality bounds estimator variance. Neither statement supplies an equation of motion.

A natural-gradient model can stipulate dynamics involving the Fisher metric. If its metric changes, its update trajectory changes because the model says so. That can be a useful engineered clock or controller. It is not evidence that a contracting Fisher volume slows proper time, nor does poor estimation precision prevent an underlying physical state from evolving quickly.

The Meta-Time Spiral uses a simpler noisy-record model. It derives no law of physical or subjective time.

Thermodynamics

Ignoring, marginalizing, or tracing out a degree of freedom is not automatically a physical erasure operation. Landauer's bound applies when a logically irreversible operation, such as resetting a memory, is implemented physically. The heat cost depends on the implementation and thermodynamic conditions.

A record-forming device may dissipate energy. A bounded memory may eventually overwrite old distinctions. Those are promising places to connect meta-time to thermodynamics, but the connection has to be built from an explicit memory cycle. “Lower dimension” alone supplies neither erasure nor a heat bill.

Physical realization and the algorithm claim

There is a broader reason to care about the distinction between an abstract trajectory and its realization.

An algorithm describes an equivalence class at a chosen causal grain. A simulated flag and a fabric flag may share a useful state-transition description while differing in material stresses, fluid coupling, thermal fluctuations, memory, and the interventions they support. If changing phase, noise, contact, or environmental coupling changes the system's responses under interventions, those features are part of the operative mechanism at that grain.

This is a description–realization gap. Noise can even contribute to signal detection through stochastic resonance. That makes noise functionally constitutive in some physical operations rather than mere contamination.

None of this provides a consciousness test. Noise, synchrony, irreversibility, effective dimension, and environmental coupling are not individually or jointly sufficient evidence of consciousness. The narrower claim is methodological: an abstraction is incomplete when it identifies systems that respond differently under the interventions relevant to the phenomenon being explained.

Meta-time fits here as a record coordinate. Whether a physical system genuinely carries that coordinate depends on how its present state constrains future interventions, not on whether we can append an epoch variable to its mathematical description.

What remains of the aperture proposal

A dimensional interface can shape which distinctions an observer can recover, how accurately they can recover them, and how long a bounded noisy channel can retain them. Those effects can alter records and reported duration. They do not derive general- or special-relativistic time dilation.

The next useful tests are narrower:

• compare circle and bounded-spiral records under controlled noise;
• measure collisions, record separation, and epoch error separately;
• model an explicit physical memory before invoking irreversibility or Landauer cost;
• test neural effective-dimension measures against prospective timing and retrospective duration as separate outcomes;
• keep relativistic proper time as a boundary condition, not a result to be reverse-engineered from an aperture proxy.

That program can fail cleanly. It also leaves the worthwhile part of the original intuition intact: recurrence, history, and readable records are different properties, and a finite physical interface makes different compromises among them.