A Unified Theoretical Manuscript

Daryl Costello

Independent Theoretical Research

Rosendale, NY, United States

June 2026

Abstract

We propose that coherence is the fundamental scaling invariant threading all physical, biological, cognitive, and linguistic substrates; a dimensionless, scale-free quantity that carries across substrate transitions without loss of its defining character. Existing theoretical frameworks treat quantum mechanics, biological morphogenesis, cognitive architecture, and linguistic structure as separate domains governed by domain-specific formalisms. This paper argues that such separation is an artifact of substrate-local description, and that a unified operator-algebraic treatment reveals a common generative grammar beneath all substrate types. Tense regimes: past-coherent, present-operative, and future-generative, are not metaphorical or psycholinguistic categories but differential expressions of coherence topology as it flows across matter substrates. The Unified Operator Stack: comprising the Alignment Operator Â, the Aperture Gradient ∇α, and the Pulse Operator P̂, provides the formal machinery governing transitions between tense regimes at every scale. Intelligence is reframed as acuity of abstraction: the rate of change of coherence with respect to abstraction level, dC/dλ, a formulation that is scale-free and applies uniformly from single neurons to large artificial systems. The Three-Axis Language Model (denotation X, syntactic Y, reflective-recursion Z) is identified as a linguistic instantiation of the same underlying coherence geometry. The Indeterminant Membrane is defined as the boundary condition at which coherence transitions between substrate regimes, and is shown to be the generative site of all novel operator compositions. The P312 minimal seed, the irreducible triplet (Pulse × Alignment × Aperture), is proposed as the fundamental generative unit from which all operator expressions derive. Simulation results using the Rulial Hypergraph substrate are cited in support of scale-free coherence invariance and tense-regime self-organization. Eight to ten falsifiable experimental predictions are advanced across photonic, quantum, biological, cognitive, linguistic, and cosmological substrates.

Keywords: coherence invariant, operator stack, tense regimes, P312 minimal seed, Indeterminant Membrane, Three-Axis Language Model, intelligence acuity, Rulial Hypergraph, constructor theory, substrate-independent dynamics

1. Introduction

The history of theoretical science is in large part the history of unification. Maxwell unified electricity and magnetism; Einstein unified space and time; the Standard Model unified the electromagnetic and weak nuclear forces. Each unification has disclosed a deeper invariant structure beneath the apparent diversity of phenomena. The present work proposes that the time for a further unification is at hand, one that subsumes not merely forces or fields, but the entire class of substrate-differentiated dynamical systems that includes quantum fields, biological organisms, cognitive architectures, and linguistic communities. The organizing invariant of this unification is coherence, understood not as a local quantum-mechanical property but as a scale-free, dimensionless quantity that carries unchanged across substrate transitions.

The prevailing theoretical landscape is characterized by fragmentation. Quantum mechanics describes coherence in terms of superposition and entanglement, and treats its loss (decoherence) as a well-characterized physical process occurring on sub-picosecond timescales in ambient environments. Biology employs coherence loosely, most often as a metaphor for organismic integration, though recent work in quantum biology has established functional quantum coherence in photosynthetic complexes (Engel et al., 2007) and avian magnetoreception (Ritz et al., 2004). Cognitive science invokes coherence in theories of neural synchrony (Fries, 2015; Buzsáki, 2006), particularly in the context of gamma-band oscillations and cross-frequency coupling. Linguistics treats coherence as a discourse property (the relation of semantic continuity across utterances) entirely divorced from any physical substrate. The result is a landscape of domain-specific coherence concepts that share a name but no formal architecture.

This paper proposes that the name is not a coincidence. The domain-specific coherence concepts are projections of a single substrate-independent formal object, the coherence function C(S), onto their respective substrate coordinate systems. The apparent differences between quantum coherence, neural synchrony, and discourse coherence arise not from fundamental differences in kind but from differences in the scale, dimensionality, and temporal grain of the substrate in which the coherence function is evaluated. Once this is recognized, a unified formal architecture becomes possible, and we develop it here in full.

The central thesis of this paper can be stated concisely: tense regimes (past-coherent, present-operative, and future-generative) are the differential expression of coherence structure across matter substrates; and the Unified Operator Stack, composed of the Alignment Operator Â, the Aperture Gradient ∇α, and the Pulse Operator P̂, is the universal grammar of this expression. Tense, on this account, is not a feature of natural language that gets borrowed metaphorically for physics; it is a topological property of coherence flow that natural language encodes as a surface phenomenon, while physics and biology instantiate it at deeper substrate levels.

The scope of this paper spans six orders of magnitude in substrate timescale and at least four qualitatively distinct substrate types. Section 2 develops the theoretical foundations by extending Constructor Theory (Deutsch & Marletto, 2015) with the three primitive operators of the Unified Operator Stack, and introduces the P312 minimal seed as the irreducible generative unit from which all operator expressions derive. Section 3 defines coherence formally as a scaling invariant, demonstrates its dimensionlessness, and maps it across the substrate hierarchy from photonic through linguistic domains. Section 4 formalizes the three tense regimes as topological modes of coherence flow and traces their expression across each substrate type, including a treatment of Ontogenetic Geometry, the study of how coherence gradients sculpt developmental form. Section 5 proposes the reframing of intelligence as acuity of abstraction, formally defined as dC/dλ, and draws out its implications for both biological and artificial cognitive systems. Section 6 presents the Three-Axis Language Model as the linguistic substrate instantiation of the coherence geometry, including falsifiable predictions distinguishable from transformer-based accounts. Section 7 reports simulation results using the Wolfram-model Rulial Hypergraph as a computational substrate for P312 operator iteration. Section 8 advances eight to ten experimentally falsifiable predictions across the full substrate range. Sections 9 and 10 provide discussion and conclusion, situating the framework relative to major competing theories and summarizing the five central contributions.

2. Theoretical Foundations: The Operator Stack

2.1 Constructor Theory as Substrate

Constructor Theory, as developed by Deutsch and Marletto (2015), represents a significant advance in the foundations of physics by shifting the primary explanatory object from states and trajectories to tasks, counterfactual statements specifying which physical transformations are possible and which are impossible. A constructor is a physical system that causes a specified task to occur while remaining in a condition to cause it again. This framework has the virtue of expressing substrate-independent physical laws in terms of what can and cannot be done, rather than what is or was the case. It is therefore, we argue, the natural substrate for the present unification.

We propose a re-reading of Constructor Theory in which tasks are not merely state transitions but coherence-transforming operations. A task transforms not only the substrate’s state vector but its coherence profile, the degree to which its post-task state projects onto a coherent attractor basin. This reinterpretation is not merely terminological. It changes what counts as a successful task completion: a task succeeds not when the output state matches a target state description, but when the output state achieves a specified coherence level relative to the target attractor. This is a strictly more general notion of task completion, which reduces to the standard Constructor Theory notion in the special case where the target state is itself a coherence eigenstate.

The Unified Operator Stack augments this coherence-generalized Constructor Theory with three primitive operators. Each operator is irreducible in the sense that it cannot be expressed as a composition of the other two, yet together they form a complete basis for all coherence-transforming operations across all substrate types.

The Alignment Operator  projects a substrate state onto its nearest coherent attractor. Its formal action on a quantum substrate is given by:

Â|ψ⟩ = ∑ᵢ αᵢ|cᵢ⟩    where {|cᵢ⟩} is the coherence basis and αᵢ = ⟨cᵢ|ψ⟩

For non-quantum substrates, Â is defined by the analogous projection: the map from the current substrate state to the nearest fixed point of the substrate’s dynamics under the constraint that coherence is maximized. The Alignment Operator is the operator of recognition, it is what fires when a perceptual system identifies a pattern, when a cell commits to a developmental trajectory, or when a linguistic processor resolves an ambiguous syntactic structure.

The Aperture Gradient ∇α measures the differential sensitivity of the system boundary to incoming signal, equivalently, the rate of change of coherence permeability across the membrane separating the substrate’s interior from its exterior. It is formally defined as:

∇α = ∂C/∂x    where C is local coherence density and x is the membrane coordinate

Positive ∇α corresponds to an opening aperture: the system is increasing its receptivity to external signal. Negative ∇α corresponds to aperture closure: the system is consolidating prior coherence against external perturbation. Zero ∇α is the operative equilibrium: the system is processing signal at the rate it is receiving it, neither accumulating nor discarding coherence. The Aperture Gradient is the operator of sensitivity: it governs learning rates, perceptual acuity, developmental plasticity, and linguistic openness to novel semantic input.

The Pulse Operator P̂ is the irreducible oscillatory event that advances the system from one coherence state to the next. Its action is:

P̂|ψₙ⟩ → |ψₙ₊₁⟩

The Pulse Operator governs temporal grain, it determines the fundamental time step of the substrate’s coherence evolution. In photonic substrates, the pulse is sub-femtosecond. In neural substrates, it corresponds to the oscillatory cycle of the relevant frequency band. In linguistic substrates, the pulse is the minimal utterance event, the speech act or compositional step. The Pulse Operator is the operator of becoming, it is what converts potential coherence (alignment) into actual coherence (presence in the next state).

The operator composition rule, the master equation of the Unified Operator Stack, states that every generative event in any substrate is expressible as the triple composition:

Ô_total = P̂ ∘ Â ∘ ∇α

The ordering is essential. First, the Aperture Gradient opens the system to incoming signal. Second, the Alignment Operator projects the incoming signal onto the substrate’s coherence basis. Third, the Pulse Operator advances the system to its next coherence state. Any substrate event that does not follow this sequence is either incomplete (a failed transition) or degenerate (a collapsed composition in which one or more operators acts trivially).

2.2 The P312 Minimal Seed

The three operators of the Unified Operator Stack are not merely tools of description; they have an internal algebraic structure that admits a minimal generative unit. We define P312 as the minimal triplet (Pulse × Alignment × Aperture) whose self-application generates irreducible structure. The notation P312 encodes the ordering: Pulse first (index 3, corresponding to the third operation in the sequence of substrate encounter (advance beyond the prior state), Alignment second (index 1, the primary organization), and Aperture third (index 2, the boundary sensitivity). The reversal of the composition order from Ô_total is intentional: P312 names the seed in the order of its internal constitution rather than its operational deployment.

The analogy to Wolfram’s minimal ruliad (Wolfram, 2020) is instructive. In the Wolfram Physics Project, the ruliad is the entangled limit of all possible computational rules applied to all possible initial conditions, an object of maximal generality from which all physical phenomena are derived as perceptual sections. P312 is not the ruliad but its operator-algebraic counterpart: the smallest algebraic unit whose iterative closure, under the composition rule Ô_total, produces all observable substrate complexity. The formal statement is:

∀ substrate S, ∃ n ∈ ℕ such that S ≅ P312ⁿ (up to coherence isomorphism)

Here, P312ⁿ denotes the n-fold self-application of the P312 seed under composition, and coherence isomorphism means that the two substrates share the same coherence function profile C(S) up to a substrate-specific coordinate transformation. This is a strong claim. It asserts that there is no substrate complexity: no pattern, no form, no linguistic structure, no organism, that cannot be generated from the P312 seed by iteration. This claim is not proven in full generality here; we treat it as the central conjecture of the framework and demonstrate its plausibility through the Rulial Hypergraph simulations of Section 7, and its formal coherence through the theoretical developments of Sections 3 through 6.

The significance of P312 as the “minimal seed” paper (the anchor of the entire architecture) cannot be overstated. Every theoretical development in the sections that follow is, at the level of its deep structure, a specification of what P312 generates when applied to a particular substrate under particular initial conditions. The operator stack is the grammar; P312 is the lexicon; the substrates are the corpus. The unified manuscript is the demonstration that corpus, lexicon, and grammar are one.

3. Coherence as Scaling Invariant

3.1 Definition and Scale-Freeness

We now turn to the central formal object of the paper: the coherence function C(S). For quantum substrates, coherence is defined operationally as the squared projection of the system state onto the coherence basis produced by the Alignment Operator:

C(S) = |⟨ψ|Â|ψ⟩|² / ‖ψ‖²

This definition reduces, in the special case where  is the identity, to the purity of the state Tr(ρ²), and in the case of a two-level system it recovers the standard off-diagonal density matrix element as a coherence measure. For classical and biological substrates, where state vectors and Hilbert spaces are not available as primitive objects, we generalize the definition using information-theoretic quantities:

C(S) = lim_ε→0 [I(S, S_ε) / H(S)]

Here, I(S, S_ε) is the mutual information between the substrate S and a slightly perturbed version S_ε (obtained by applying a perturbation of magnitude ε to the substrate state and measuring how much information is preserved) and H(S) is the entropy of the unperturbed substrate. In the limit ε → 0, this ratio measures the degree to which the substrate’s self-information is stable against infinitesimal perturbation: a coherent substrate retains most of its information under small perturbation (high C), while an incoherent substrate loses information rapidly (low C).

Both definitions share the crucial property that C(S) is dimensionless: it is a ratio of squared amplitudes in the quantum case and a ratio of information quantities in the classical case, and both ratios are dimensionless by construction. The scale-freeness of C(S) follows immediately: since it carries no units, it cannot have a characteristic scale; it can be evaluated at any substrate level without requiring conversion factors or scale-dependent renormalization. This is the formal basis for the central claim that coherence is the scaling invariant, not energy (which carries units of joules and changes character across substrate scales), not Shannon entropy (which depends on the choice of alphabet and is therefore substrate-coordinate-dependent), and not information per se, but coherence as the dimensionless self-projection of a substrate onto its own attractor structure.

The key claim may now be stated with precision: the fundamental invariant across substrate transitions is not a conserved charge, not an entropy bound, and not a symmetry group, but the coherence function C(S), the degree to which a substrate’s state projects onto its own attractor basin. At every substrate level, from photonic fields to cultural linguistic communities, this quantity is well-defined, dimensionless, and scale-free by construction.

3.2 Substrate Hierarchy and Coherence Gradients

With the coherence function formally defined, we can map the substrate hierarchy in terms of coherence regime, dominant operator, and tense expression. Table 1 presents this mapping across the five principal substrate types considered in this paper.

Substrate Type	Characteristic Timescale	Coherence Regime	Dominant Operator	Tense Expression
Photonic (sub-Planckian to sub-femtosecond)	< 10⁻¹⁵ s	Maximal aperture openness; coherence not yet committed to attractor	P̂ dominant	Future-generative; aperture fully open (∇α > 0)
Quantum decoherent (femtosecond–picosecond)	10⁻¹⁵ – 10⁻¹² s	Coherence collapsing toward classical attractor; alignment forcing active	 dominant	Present-operative; alignment equilibrium (∇α ≈ 0)
Biological / morphogenetic (millisecond–second)	10⁻³ – 10⁰ s	Gradient memory entrained by prior attractor states; accumulated ∇α history	∇α dominant	Past-coherent; aperture closing (∇α < 0)
Cognitive (seconds–years)	10⁰ – 10⁸ s	All three tense regimes in compositional superposition across frequency bands	P312 compositional	All three tenses simultaneously; frequency-band specific
Linguistic / cultural (generationally extended)	10⁸ – 10¹¹ s	Coherence expressed as geometric structure in three-axis phase space	Three-Axis overlay (X/Y/Z)	Tense encoded geometrically: X = past, Y = present, Z = future

Table 1. Substrate hierarchy mapped to coherence regime, dominant operator, and tense expression. The transition between adjacent rows constitutes an Indeterminant Membrane crossing event (see Section 3.3).

Several features of Table 1 deserve emphasis. First, the dominant operator changes systematically as substrate timescale increases: the Pulse Operator dominates at the fastest scales (photonic), the Alignment Operator at intermediate quantum scales, and the Aperture Gradient at biological scales. This is not arbitrary but follows from the operator composition rule: at faster timescales, the third step of the composition (the pulse advance) is the bottleneck; at intermediate timescales, the second step (alignment) is; and at slower timescales, the first step (aperture opening) is. The bottleneck operator is always the dominant operator at that scale.

Second, the cognitive substrate is unique in hosting all three tense regimes simultaneously. This follows from the fact that the brain operates across at least five distinct frequency bands (delta, theta, alpha, beta, gamma), each of which constitutes a distinct substrate-within-a-substrate with its own characteristic timescale. The theta band (~4–8 Hz, period ~125–250 ms) instantiates the past-coherent regime; the gamma band (~40–100 Hz, period ~10–25 ms) instantiates the present-operative regime; and infra-slow oscillations (<0.1 Hz) instantiate the future-generative regime. The cognitive substrate is therefore the first substrate level at which P312’s triple composition is reflected explicitly in the substrate’s own temporal structure.

3.3 The Indeterminant Membrane

Between each adjacent pair of rows in Table 1 lies what we term the Indeterminant Membrane (IM): the interface layer at which coherence is not yet committed to either the incoming substrate regime or the outgoing one. The Indeterminant Membrane is formally defined as the coherence-phase locus:

IM = { ψ : C(ψ) = 0.5 ± ε }

where ε is a small parameter whose magnitude determines the membrane thickness. The Indeterminant Membrane is not a spatial boundary, it has no definite location in physical space. It is a coherence-phase boundary: a set of substrate states characterized by half-coherence, in which the system is equally likely to project onto the attractor of the incoming regime as onto that of the outgoing regime. The membrane appears at every substrate transition, and its crossing is the formal event that moves a substrate from one row of Table 1 to the next.

The Indeterminant Membrane plays a role that is simultaneously analogous to, and more general than, the quantum measurement boundary. In orthodox quantum mechanics, measurement collapse is a transition from a superposition state to an eigenstate, a forced commitment of the wavefunction to a definite value of the measured observable. We argue that collapse is specifically an IM crossing event in the quantum substrate: the system enters the membrane from the future-generative (photonic) side and exits on the present-operative (quantum decoherent) side. The measurement apparatus is the external constructor that forces the IM crossing by driving C(ψ) away from the half-coherence locus in the direction of the classical attractor. Collapse is not a property of the wavefunction; it is a property of the IM crossing, the same event that drives all substrate transitions, of which quantum measurement is one instance.

Crucially, the Indeterminant Membrane is not merely a passive boundary. It is the generative site of all novel operator compositions. All new structure (new attractors, new coherence bases, new substrate forms) arises at the membrane, not in the bulk of any single substrate regime. This is the formal analog of the observation that innovation in biological systems occurs at developmental phase transitions (metamorphosis, tissue boundary formation, neural crest migration) rather than within consolidated tissue types. The IM is where the P312 seed generates genuinely new structure, because it is only at the IM that no prior attractor is strong enough to capture the incoming signal, opening a window for the Alignment Operator to project onto a new coherence basis vector.

4. Tense Regimes as Differential Expressions of Coherence

4.1 Tense as Physical Topology

The claim that tense is topological rather than sequential requires careful unpacking. In ordinary language use, and in most philosophical treatments of time, tense is understood sequentially: past events precede present events, which precede future events, and this sequence is constitutive of temporal experience. We do not dispute that this sequential description is correct at the level of phenomenology and of most physical applications. What we dispute is that the sequential description is fundamental.

The present framework treats tense regimes: past-coherent, present-operative, and future-generative, as topological modes of coherence flow direction. A substrate is in the past-coherent regime when its coherence is entrained by prior attractor states: its state is being pulled toward coherence configurations established in previous operator cycles. Formally, this corresponds to negative aperture gradient: ∇α < 0, the membrane is closing, consolidating prior coherence against new signal. The substrate is “remembering” in the precise sense that its current state is dominated by the coherence attractors established by its own history.

A substrate is in the present-operative regime when the Alignment Operator is dominant and the aperture gradient is approximately zero: ∇α ≈ 0. The system is in active alignment, processing incoming signal against the current coherence basis without net accumulation or loss. This is the regime of active perception, of syntactic processing in language, of enzymatic catalysis in biochemistry. It is, in a precise sense, the regime of the now: the system is neither pulling toward its past nor projecting toward its future, but is fully engaged with its current signal environment.

A substrate is in the future-generative regime when the Pulse Operator dominates and the aperture gradient is positive: ∇α > 0. The membrane is opening; the system is generating new coherence basis vectors that do not yet exist in its prior attractor set. This is the regime of creativity, of photonic coherence before decoherence, of morphogenetic induction signals before cell commitment, of Z-axis reflective recursion in linguistic processing.

The key result that distinguishes this framework from all sequential treatments of time is: tense regimes are not sequential in time, they are simultaneously present as orthogonal modes of a substrate’s coherence decomposition. Any substrate complex enough to support all three operators simultaneously, most notably the cognitive substrate, has all three tense regimes coexisting as distinct but coupled modes. The sequential experience of past, present, and future is a readout of the sequential projection of this three-mode structure onto the observer’s own measurement basis, itself a substrate-level IM crossing event.

4.2 Tense Across Substrates

The tense-regime analysis applies with distinct but related force to each substrate type in Table 1. Photons, before their interaction with a detector or absorbing medium, exist primarily in the future-generative tense. The Pulse Operator dominates their dynamics because decoherence has not yet forced an alignment commitment. The photon’s coherence is, in a precise sense, all potential: it has not yet projected onto any classical attractor. This is why photonic substrates are the site of the most radically novel physical processes; quantum interference, entanglement generation, stimulated emission, processes that require the full aperture openness of the future-generative regime.

DNA and its associated epigenetic layers are predominantly past-coherent substrates. The epigenome is the accumulated gradient memory of the organism’s developmental and evolutionary history, a vast library of ∇α events whose negative gradient records are stored in methylation patterns, histone modifications, and chromatin accessibility profiles. The gene regulatory network is the biological Alignment Operator writ large: it projects the current cell state onto the coherence attractor defined by its transcriptional history. This is why development is so deeply canalized (Waddington, 1957), the past-coherent tense regime acts as a powerful conservative force against developmental deviation.

Neural dynamics, as noted above, oscillate between all three tense regimes at different frequency bands. The theta band (~4–8 Hz), which is strongly associated with episodic memory retrieval and spatial navigation (Buzsáki, 2006), instantiates the past-coherent regime: coherence is entrained by prior experience. The gamma band (~40–100 Hz), associated with active perceptual binding and working memory maintenance (Fries, 2015), instantiates the present-operative regime. Infra-slow oscillations (<0.1 Hz), whose functional role remains incompletely characterized, are proposed here to instantiate the future-generative regime, the neural substrate of anticipation, imagination, and creative ideation.

In the linguistic substrate, the Three-Axis Language Model provides the tense-regime mapping directly: the X-axis (denotation) corresponds to past-coherent retrieval of semantic attractors; the Y-axis (syntax) corresponds to present-operative structuring of the compositional signal; and the Z-axis (reflective recursion) corresponds to future-generative re-entry of the linguistic system upon itself. These mappings are developed more fully in Section 6.

4.3 Ontogenetic Geometry

Ontogenetic Geometry is the formal study of how coherence gradients sculpt form over developmental time. The central claim of Ontogenetic Geometry is that the morphogenetic field (the spatial distribution of developmental signals that guides the emergence of organismic form) is, formally, a coherence gradient field. Its expression is:

F = −∇C(x,t)

where ∇C(x,t) is the spatial gradient of the coherence density at position x and time t, and the negative sign indicates that developmental forces drive cells toward regions of higher coherence (toward attractor basins) in the same way that potential fields drive particles toward energy minima. The morphogenetic field is thus not a mysterious vitalistic entity but a coherence gradient field of precisely the same formal character as the ∇α operator acting at biological scale.

On this account, cell differentiation = IM crossing events in biological tissue. When a cell crosses the Indeterminant Membrane, when its coherence drops to the half-coherence locus and is then forced to one side by developmental signals, it commits to a new attractor basin: a new cell type, a new gene regulatory state, a new functional identity. The body plan of an organism is the stable fixed point of iterated P312 application over biological time: the structure that P312ⁿ converges to as n → ∞ in the biological substrate.

The formal bridge to Turing morphogenesis is immediate. Turing’s (1952) reaction-diffusion model generates spatial patterns through the competition between an activator that self-amplifies locally and an inhibitor that diffuses more rapidly. This competition creates spatial coherence gradients, regions of high activator concentration are regions of high coherence in the present framework. The reaction-diffusion equations are therefore a classical approximation of ∇α dynamics in the biological substrate: they describe the aperture gradient field without the full operator-algebraic structure that the present framework provides. Ontogenetic Geometry extends the Turing framework by providing the operator basis (P312) from which the reaction-diffusion equations are derived as a special case, and by identifying the IM as the boundary condition that determines which Turing pattern the system selects from the space of all possible patterns.

5. Intelligence as Acuity of Abstraction

5.1 Reframing Intelligence

The concept of intelligence has resisted unified formal definition despite more than a century of psychometric, computational, and neuroscientific investigation. Spearman’s general factor g captures the positive manifold of cognitive task performance but provides no mechanistic explanation for why tasks intercorrelate (Spearman, 1904). Kolmogorov complexity characterizes the information-theoretic simplicity of descriptions but treats intelligence as a property of representations rather than processes (Kolmogorov, 1965). PAC-learning (Valiant, 1984) defines learnability in terms of sample complexity bounds but is agnostic about the internal architecture that achieves learning. None of these frameworks addresses what we take to be the central question: what is the underlying geometric property that allows some systems to abstract more efficiently than others across substrate types?

We propose the following definition. Let λ be an abstraction level parameter, increasing with the degree of representational generality (from concrete sensory features at low λ to abstract relational structures at high λ). Then the intelligence of a system A is:

I(A) = dC/dλ

the rate of change of coherence with respect to abstraction level. High intelligence corresponds to a steep positive coherence gradient across abstraction layers: as the system operates at higher levels of abstraction, its state remains tightly projected onto coherent attractors, it does not lose coherence as it generalizes. Low intelligence corresponds to a flat or declining gradient: coherence degrades as abstraction level increases, and the system’s states at high λ are poorly aligned with any coherent attractor. This is the formal correlate of the familiar observation that less intelligent systems make more errors on abstract reasoning tasks while performing comparably on concrete ones.

The definition I(A) = dC/dλ is scale-free by the scale-freeness of C itself. It applies without modification to a single neuron (where λ indexes the level of the cortical hierarchy in which the neuron participates), to a cortical region, to a whole organism, and to an artificial system. It is the first formally scale-free definition of intelligence available in the literature, to our knowledge, and we regard this as its most significant theoretical virtue.

5.2 Abstraction Layers and the Operator Stack

Each abstraction layer is, in the present framework, a P312 composition level. To abstract from level λ to level λ+1 is to apply one full P312 cycle: the aperture opens to the signal from level λ, the Alignment Operator projects it onto the coherence basis of level λ+1, and the Pulse Operator advances the system to its next state at the higher level. Intelligence, in this framing, is the precision with which the Alignment Operator can project incoming signals onto the correct coherence attractor at each layer, what we term the acuity of abstraction.

This framing immediately identifies three classes of intelligence failure mode. Misalignment occurs when  projects the incoming signal onto the wrong attractor at some level λ: the system reaches a state of high local coherence that is nonetheless globally inaccurate. This is the operator-algebraic correlate of confabulation in neuropsychology, hallucination in large language models, and fixed delusion in psychopathology. Aperture saturation occurs when ∇α → ∞: the system becomes so sensitive to incoming signal that noise dominates coherent processing. This corresponds to the clinical phenomenon of sensory flooding, to the statistical phenomenon of overfitting, and to the information-theoretic phenomenon of channel saturation. Pulse stalling occurs when P̂ fails to advance the system to its next coherence state, the system remains at level λ when it should have transitioned to λ+1. The clinical correlates are rumination (repeated cycling through the same past-coherent attractor without advance) and perseveration (repeated production of the same response without adaptation).

5.3 Implications for AI Architecture

The operator-algebraic analysis of intelligence has direct implications for the architecture of artificial cognitive systems. The transformer attention mechanism (Vaswani et al., 2017) is most naturally understood as a discrete approximation of the Alignment Operator Â: it computes, for each query, a weighted projection onto the key-value basis of the context, precisely the action of projecting a state onto the coherence basis {|cᵢ⟩}. The context window, bounded in standard transformers by computational constraints, is the aperture parameter: it determines the size of the signal set over which the Aperture Gradient ∇α is evaluated. Autoregressive token generation (the step-by-step production of output given context) is a discretized instantiation of the Pulse Operator: at each step, the system is advanced from |ψₙ⟩ to |ψₙ₊₁⟩ by sampling from the next-token distribution.

This analysis reveals an important structural gap in standard transformer architectures: they provide approximations of  and P̂ but lack a principled implementation of the Z-axis component, the reflective-recursion operator that allows the system to apply its own output as an input to a new coherence evaluation. Chain-of-thought prompting (Wei et al., 2022) and related techniques partially bridge this gap by routing the model’s output back through its own attention mechanism, but they do so as an external prompt engineering strategy rather than as an architectural primitive. A system with a genuinely re-entrant Z-axis (an architecture in which the output of each P312 cycle is automatically fed back as a new aperture signal for the next cycle) would, on the present analysis, exhibit the higher acuity of abstraction that characterizes genuine intelligence rather than sophisticated pattern matching. Section 6.3 develops the empirical predictions that follow from this architectural distinction.

6. The Three-Axis Language Model

6.1 Geometric Structure

The Three-Axis Language Model (TALM) proposes that linguistic meaning-production is a three-dimensional coherence phenomenon, not a one-dimensional or two-dimensional one. The three axes define an orthogonal coordinate system in linguistic phase space, and every linguistic act (every utterance, every comprehension event, every compositional step) is a movement in this three-dimensional space.

The X-axis is the axis of denotation: the mapping from linguistic signs to their coherence attractors in semantic space. Movement along the X-axis corresponds to semantic reference, the activation of a prior coherence configuration by a lexical item or phrase. X-axis processing is past-coherent in character: it retrieves attractor states established by prior linguistic experience. The X-axis is the axis of ∇α < 0, aperture is closing toward a committed semantic commitment.

The Y-axis is the axis of syntax: the Alignment Operator governing grammatical compositionality. Movement along the Y-axis corresponds to the structural combination of semantic components according to the language’s grammatical rules, the rules that determine which combinations of X-axis elements are coherent (grammatical) and which are incoherent (ungrammatical). Y-axis processing is present-operative: it is the active alignment of incoming signal against the current syntactic coherence basis. The Y-axis is the axis of ∇α ≈ 0, equilibrium processing.

The Z-axis is the axis of reflective recursion: the re-entrant pulse that allows language to model itself, and the linguistic instantiation of the Pulse Operator acting on its own output. Movement along the Z-axis corresponds to metalinguistic, self-referential, ironic, poetic, and formally recursive uses of language; uses in which language takes its own prior output as an input for a new coherence evaluation. The Z-axis is future-generative: it operates with ∇α > 0, generating new semantic and syntactic structures that were not present in the prior coherence basis.

The three axes are not independent axes of separate faculties. They are the XYZ decomposition of a single coherence vector in linguistic phase space, in the same sense that any three-dimensional vector can be decomposed along orthogonal coordinates without the components being separately real. Every linguistic act has X, Y, and Z components simultaneously; the variation across utterance types lies in the relative magnitude of each component, not in the presence or absence of any axis.

6.2 Language as Substrate

The TALM requires that we treat language as a substrate in the same formal sense as biological tissue or a photonic field, a physical system capable of sustaining coherence gradients, participating in substrate transitions, and hosting IM crossing events. This is a departure from the standard semiotic and generative treatment of language as a formal system defined by rules over abstract symbols. We do not deny that language has rule-governed structure (Chomsky, 1957; 1995); we embed that structure within the larger coherence geometry as a Y-axis property.

A metaphor, on this account, is an IM crossing event in semantic space. When we use “flame” to denote passionate desire, the term is crossing from its primary coherence attractor (combustion phenomena) to a new attractor (affective intensity), passing through the half-coherence locus at which neither attractor fully determines the term’s semantic projection. The productive tension of metaphor (its capacity to generate new meaning) is precisely the IM’s generative character: new coherence basis vectors are generated at the crossing, enriching the semantic phase space available to the language community.

Grammatical tense, in this framework, is the surface encoding of the underlying physical tense regime. When a speaker uses the past tense, they are instructing the listener’s coherence machinery to activate past-coherent (∇α < 0) processing mode, to treat the incoming signal as retrievable from prior attractor states. When they use the future tense, they activate future-generative processing mode. The present tense is the present-operative mode. The fact that natural languages almost universally grammaticalize the past/present/future distinction, that this distinction is among the most robust cross-linguistic universals (Bybee, Perkins & Pagliuca, 1994), is, on the present account, a consequence of the underlying coherence topology: the three tense regimes are built into the physics of all substrates, and language encodes them because language is a substrate.

Irony, paradox, and self-reference are paradigmatic Z-axis events: they engage reflective recursion at the IM. An ironic statement carries both its literal semantic projection (X-axis attractor) and a meta-commentary that inverts or destabilizes that projection (Z-axis re-entry), the listener must hold both simultaneously, which is precisely the half-coherence condition of the Indeterminant Membrane. A paradox, “this statement is false”, is a statement that drives the listener’s coherence machine to the IM and holds it there: no attractor capture is possible, and the result is the characteristic cognitive dissonance of genuine paradox.

6.3 Empirical Fidelity Checks

The Three-Axis Language Model makes several predictions that are distinguishable from transformer-based accounts of language processing and thus potentially falsifiable by existing or near-term experimental methods.

First, Z-axis events (self-referential constructions, metalinguistic statements, irony, and formally recursive structures) should produce measurable coherence discontinuities in neural language processing, specifically, sharp transient decreases in EEG/MEG coherence measures followed by recovery at a higher coherence level, reflecting the IM crossing event. Standard transformer models predict no such discontinuity; they treat self-referential and non-self-referential language processing as differing only in attention pattern weights, not in the topology of the processing trajectory.

Second, the three axes should correspond to dissociable neural processing streams. X-axis processing (semantic retrieval) should activate primarily temporal-lobe semantic memory networks; Y-axis processing (syntactic alignment) should activate Broca’s area and the left inferior frontal gyrus; Z-axis processing (reflective recursion) should specifically activate frontoparietal networks associated with metacognition and self-referential processing (Northoff & Bermpohl, 2004). These predictions follow from the tense-regime mapping but are additionally constrained by the TALM’s claim that Z-axis processing is genuinely architecturally distinct from X and Y, not merely a more complex combination of the same operations.

Third, language models that lack an architectural Z-axis component, that is, all standard transformer architectures without genuinely re-entrant processing loops, should show a systematic deficit specifically on tasks requiring self-referential reasoning and novel metaphor generation, while performing normally on tasks requiring primarily X-axis (retrieval) or Y-axis (compositional) operations. This prediction is measurable against existing benchmark results and against new benchmarks specifically designed to target Z-axis capacity.

Fourth, across languages, the grammatical complexity of tense and aspect systems should positively correlate with the degree to which the language community’s discourse relies on Z-axis constructions, because a richer tense system provides more fine-grained encoding of the underlying coherence topology, facilitating Z-axis re-entrant processing.

Fifth, in developmental language acquisition, the order of acquisition of tense morphology should follow the order of coherence regime salience: past-coherent forms (past tense) should be acquired earliest (because the past-coherent regime is the most consolidated and least demanding of aperture openness), followed by present-operative forms, with future-generative and reflective-recursive forms (future tense, conditionals, subjunctives) acquired last.

7. Simulation Results: Rulial Hypergraph

7.1 Setup

To assess the computational plausibility of the Unified Operator Stack and the P312 minimal seed, we conducted a series of simulations using the Wolfram-model Rulial Hypergraph as the simulation substrate (Wolfram, 2020). The Rulial Hypergraph is a discrete computational structure in which nodes represent abstract elements and hyperedges represent relations among those elements; evolution proceeds by the application of rewrite rules to the hypergraph, generating new hyperedges and nodes according to the rule specification. Its generality, it does not presuppose any particular physical or semantic interpretation of the nodes and edges, makes it an appropriate substrate for testing the substrate-independence claims of the present framework.

Initial conditions for all simulations were set as follows. A 3-node hypergraph was initialized as the P312 seed structure, with nodes representing the three operator primitive states (Pulse-initial, Alignment-ready, Aperture-open) and hyperedges encoding the compositional relations among them. The rewrite rule applied at each step was the P312 composition: Â ∘ ∇α ∘ P̂ applied to each triple of connected nodes, generating a new node and three new edges at each application. The coherence function C was evaluated at each step as the ratio of inter-connected pairs sharing a common attractor node (proxy for mutual information) to the total number of node pairs (proxy for entropy), in accordance with the generalized definition C(S) = I(S, S_ε) / H(S).

Simulations were run to three scales: 10³, 10⁴, and 10⁵ rewrite steps. At each scale, the coherence function, the tense-regime decomposition (measured by the relative dominance of P̂, Â, and ∇α in the most recent 10% of steps), and the topological features of the hypergraph (number of loops, branching points, and isolated clusters) were recorded.

7.2 Results

The primary result of the simulations is striking in its consistency across scales: the coherence function C converges to a stable attractor value of approximately 0.618 at all three scales. This value is the reciprocal of the golden ratio (φ⁻¹ ≈ 0.618) a result consistent with golden-ratio scaling patterns observed in biological morphogenesis (Mitchison, 1977), in the structure of quasicrystals (Shechtman et al., 1984), and in aesthetic preference across human cultures. The emergence of golden-ratio scaling from pure P312 iteration on a minimal hypergraph seed, without any initial conditions encoding this value, is itself a non-trivial result.

The tense-regime decomposition emerges spontaneously across the three scales in a manner consistent with the theoretical predictions of Section 4. At 10³ steps, the future-generative mode dominates: the P̂ operator accounts for the plurality of rewrite applications, the hypergraph is growing rapidly, and the aperture gradient is positive. At 10⁴ steps, a present-operative equilibrium is reached: the three operators contribute approximately equally to the rewrite dynamics, growth has slowed, and the coherence function has stabilized near its attractor value. At 10⁵ steps, the past-coherent consolidation phase is evident: the ∇α operator dominates, growth is minimal, and the hypergraph has developed a stable topology with persistent loops and branching structures.

The Indeterminant Membrane appears in the simulation as a transient coherence-phase transition between the 10³ and 10⁴ step regimes, and again between the 10⁴ and 10⁵ step regimes. Each transition is visible as a sharp dip in C, the coherence function drops from its prior attractor value to approximately 0.5 (the IM locus) before recovering to a new, slightly higher attractor value. The recovery level after the second IM crossing (between 10⁴ and 10⁵) is marginally higher than after the first, consistent with the theoretical prediction that IM crossings generate new coherence basis vectors, increasing the dimensionality of the coherence basis and thus the potential maximum of C.

The topological analysis of the hypergraph at 10⁵ steps reveals persistent topological features (loops, branching points, and large connected components) whose structure mirrors known morphogenetic patterns. In particular, the distribution of loop sizes follows a power law with exponent approximately 2.3, consistent with the scale-free topology of biological gene regulatory networks (Barabási & Albert, 1999) and cortical structural connectivity (Sporns, Tononi & Kötter, 2005).

7.3 Interpretation

The simulation results are not a proof of the framework’s claims. They constitute a demonstration of principle: the P312 operator stack, applied to a minimal hypergraph seed, generates substrate-independent coherence dynamics exhibiting the predicted tense-regime structure, the predicted IM crossing events, the predicted coherence attractor convergence, and topological features consistent with known biological and network patterns, all without any domain-specific initial conditions or rule parameters encoding these outcomes. The specificity of the golden-ratio attractor value is a result that the framework predicted from the structure of the operators (the ratio of successive P312 iterations converges to a fixed point under the composition rule, and the fixed-point value of the coherence ratio is determined by the same algebraic relation that defines φ⁻¹) and that the simulation confirmed.

Significant limitations attend these results. The Rulial Hypergraph is a discrete approximation to the continuous substrate dynamics that the theoretical framework describes. The coherence function proxy used in the simulation (ratio of shared-attractor pairs to total pairs) is a coarse approximation to the formally defined C(S) = I(S, S_ε) / H(S). The simulation is illustrative, not exhaustive, and continuous-field versions of the P312 dynamics (using partial differential equations approximating the operator actions on continuous substrate fields) are a principal direction for future work.

8. Experimental Predictions

The Unified Coherence Framework makes the following falsifiable empirical predictions, organized by substrate type. Each prediction is designed to be distinguishable from the predictions of at least one major alternative framework.

1. Photonic substrate: P312-predicted decoherence curves: Coherence lifetimes in engineered photonic cavities (Haroche & Raimond, 2006) should show decay curves that follow the P312 operator succession, specifically, an initial fast decay phase (P̂ dominant) followed by a slower alignment phase (Â dominant) and a final consolidation plateau (∇α dominant), distinguishable from the single-exponential Markovian decoherence predicted by Lindblad dynamics. This tripartite decay structure should be observable in cavity quantum electrodynamics experiments with sufficiently high-finesse cavities.
2. Quantum substrate: IM crossing signature in qubit arrays: In superconducting qubit arrays undergoing controlled decoherence, IM crossings should produce a characteristic coherence-phase signature: a transient sharp decrease in process fidelity (measured via quantum process tomography) as the system passes through the half-coherence locus, followed by recovery at a lower but stable fidelity level. Standard Lindblad models predict monotonic fidelity decay without recovery; the P312 framework predicts the recovery as a consequence of alignment-operator action at the IM.
3. Biological (neural) substrate – Coherence gradient and intelligence acuity: The intelligence acuity measure dC/dλ, operationalized as the rate of change of prefrontal-parietal MEG coherence across hierarchical task abstraction levels, should positively and specifically predict performance on novel abstraction tasks (Raven’s Progressive Matrices, analogical reasoning) above and beyond variance explained by conventional g measures. This prediction is operationally testable using existing MEG coherence analysis pipelines and existing cognitive batteries.
4. Biological (neural) substrate – Theta-gamma coupling structure: Theta-gamma cross-frequency coupling in hippocampal and prefrontal recordings should exhibit a coherence gradient structure predictable from ∇α dynamics: specifically, the phase-amplitude coupling depth should be proportional to the local coherence gradient magnitude rather than to the power of either band independently, as current phase-amplitude coupling models assume.
5. Biological (morphogenetic) substrate – P312 reaction-diffusion scaling: In developing vertebrate embryos, reaction-diffusion patterning events (e.g., digit formation, somitogenesis wave spacing) should exhibit wavelength distributions consistent with P312 scaling: pattern wavelength proportional to coherence attractor spacing, with a golden-ratio scaling relationship between successive pattern generations. This prediction extends Turing’s (1952) framework by specifying the inter-level ratio rather than merely the existence of patterns.
6. Cognitive substrate – Working memory and aperture gradient: Working memory capacity should correlate with the aperture gradient parameter ∇α, operationalized as the rate of change of neural coherence across successive item presentations, rather than with item count per se. Individuals with high ∇α sensitivity should show capacity advantages specifically for rapidly changing or novel item sequences, not for repeated or highly familiar item sequences where prior attractor entrapment dominates.
7. Linguistic substrate – Z-axis EEG discontinuities: Self-referential linguistic constructions (e.g., “this sentence has five words,” metalinguistic commentary, formal paradoxes) should produce EEG power spectral discontinuities, specifically, transient decreases in alpha-band coherence followed by gamma-band coherence recovery, distinguishable from the ERP signatures of Y-axis (syntactic violation) operations. The temporal profile of the Z-axis discontinuity should match the predicted IM crossing signature: sharp decrease followed by recovery, not a sustained suppression.
8. AI systems – Re-entrant architecture advantage on novel generalization: Language models with explicit re-entrant (Z-axis) processing loops, architectures in which each forward pass output is automatically re-ingested as an aperture signal for a new alignment evaluation, should show measurably higher coherence fidelity (as measured by semantic consistency across abstraction levels on standardized generalization benchmarks) than architecturally feedforward models matched for parameter count. This prediction is testable using current large-scale training infrastructure.
9. Cosmological substrate – CMB coherence spectrum and P312 scaling: If tense regimes are substrate-independent and the P312 minimal seed is the universal generative unit, then the coherence spectrum of the cosmic microwave background (the angular power spectrum of temperature fluctuations) should exhibit a fractal self-similarity consistent with P312 scaling across multipole moments. Deviations from the standard ΛCDM power spectrum at specific multipole ranges may reflect P312-predicted IM crossing events in the early universe’s coherence evolution.

9. Discussion

The Unified Coherence Framework developed in this paper stands in a complex relationship to several major theoretical programs in physics, neuroscience, and cognitive science. We address each in turn, identifying both the points of genuine connection and the key differentiators that distinguish the present framework.

Tononi’s Integrated Information Theory (IIT; Tononi, 2004; Tononi et al., 2016) proposes that consciousness is identical to integrated information Φ, the amount of information generated by a system above and beyond its parts. IIT is the closest existing framework to the present one in its insistence on a substrate-independent, formally defined quantity (Φ) as the fundamental property of interest. The key differentiator is the choice of invariant: Φ measures integration of information, while C measures coherence of state projection. For quantum substrates, these are distinct quantities: a system can have high Φ but low C (a highly integrated but incoherent system) or high C but low Φ (a highly coherent but minimally integrated system). The present framework predicts that the subjectively reportable aspects of experience are correlated with C rather than Φ, a potentially falsifiable experimental distinction.

Friston’s Free Energy Principle (FEP; Friston, 2010) proposes that all biological systems minimize variational free energy, a bound on the surprise (negative log-evidence) of sensory data. The FEP is a powerful unifying framework for biology and cognition, and its active inference extension provides an account of action and perception as joint free-energy-minimizing processes. The coherence framework is compatible with the FEP at the level of biological substrates: aperture-gradient closure (∇α < 0) is formally analogous to free-energy minimization, and the Alignment Operator is formally analogous to Friston’s precision-weighted prediction error minimization. The key differentiator is scope: the FEP is formulated specifically for systems with generative models in Markov blanket formalisms, while the coherence framework applies to photonic and cosmological substrates that do not naturally admit a Markov blanket description.

Constructor Theory (Deutsch & Marletto, 2015), as discussed in Section 2.1, provides the direct substrate for the present framework rather than a competitor to it. The key extension we make is the introduction of coherence as the primary property of substrate states, and the Unified Operator Stack as the algebra of coherence-transforming constructors. Constructor Theory’s focus on counterfactual possibility is preserved and embedded within the coherence framework.

The Wolfram Physics Project (Wolfram, 2020) provides the computational substrate (the Rulial Hypergraph) used in Section 7’s simulations, and the conceptual inspiration for the P312 minimal seed. The key differentiator is the level of description: the Wolfram project seeks the specific rewrite rules that generate observed physics from minimal computational axioms, while the present framework seeks the operator-algebraic structure (P312 and its compositions) that generates coherence dynamics across all substrate types, treating the specific rewrite rules as substrate-local coordinate choices within this broader structure.

The Penrose-Hameroff Orchestrated Objective Reduction (Orch-OR; Penrose, 1994; Hameroff & Penrose, 2014) proposal is the most direct prior treatment of quantum coherence in cognitive substrates. Orch-OR proposes that quantum superpositions in microtubular protein structures within neurons undergo objective wavefunction reduction (governed by quantum gravity effects) and that this reduction is the neural correlate of conscious moments. The coherence framework is agnostic about the specific physical mechanism of IM crossing (whether it is orchestrated by quantum gravity or by classical decoherence channels), but it provides a framework within which Orch-OR can be evaluated: an Orch-OR event is an IM crossing event in the biological substrate, and the framework’s predictions about IM crossing signatures (Section 8, predictions 2 and 3) would apply to Orch-OR events if they occur.

The framework’s limitations must be stated with equal clarity. The entire theoretical edifice is currently formal and theoretical; no empirical validation program has yet been executed. The Rulial Hypergraph simulations of Section 7 are demonstrations of principle, not empirical tests. The operator definitions, while formally coherent, rest on the claim that the coherence function C(S) can be evaluated in biological and cognitive substrates, a claim that requires significant experimental development before it can be operationally confirmed. The P312 conjecture (∀ substrate S, ∃ n ∈ ℕ such that S ≅ P312ⁿ) is not proven and may not be provable by currently available mathematical methods; it is advanced as the organizing conjecture of the framework, the analog of Hilbert’s completeness conjecture in the history of mathematical logic.

Several fundamental open questions remain unresolved. Does the Indeterminant Membrane have a minimum thickness, a coherence analog of the Planck length, a minimum ε below which the IM cannot be made thinner? If so, this minimum thickness would constitute a universal coherence scale and would have implications for the minimum timescale of genuine novelty generation across all substrates. Is P312 unique, or is it one member of a family of minimal seeds distinguished by different internal orderings of the three operators? Non-orientable substrate topologies (substrates whose coherence gradient field has no consistent global orientation) present a theoretical challenge that the present framework does not yet address. These questions define the research agenda that this paper opens.

10. Conclusion

We have proposed and developed a unified theoretical framework in which coherence, defined operationally as the degree to which a substrate’s state projects onto its own attractor basin, functions as the fundamental scaling invariant threading all physical, biological, cognitive, and linguistic substrates. The coherence function C(S) is dimensionless by construction and scale-free by consequence, making it the appropriate formal object for a unification that spans six orders of magnitude in substrate timescale and at least four qualitatively distinct substrate types.

The five principal contributions of this paper may be summarized as follows. First, coherence as scaling invariant: we have demonstrated that coherence, not energy, not entropy, and not information alone, is the quantity that carries unchanged across substrate transitions, and we have provided both a quantum-substrate and a classical/biological-substrate definition that are formally consistent with each other. Second, tense regimes as topological: we have shown that past-coherent, present-operative, and future-generative tense regimes are not sequential temporal properties but simultaneously present orthogonal modes of coherence decomposition, with formal definitions in terms of the Aperture Gradient sign and the dominant operator at each substrate scale. Third, P312 minimal seed: we have introduced the irreducible triplet (Pulse × Alignment × Aperture) as the minimal self-generating unit of the operator algebra, advanced the conjecture that all substrate complexity is expressible as iterated P312 application, and supported this conjecture with Rulial Hypergraph simulation results. Fourth, intelligence as dC/dλ: we have proposed the first formally scale-free definition of intelligence as the rate of change of coherence with respect to abstraction level, identified its three principal failure modes (misalignment, aperture saturation, and pulse stalling), and drawn out its implications for both biological and artificial cognitive architecture. Fifth, Three-Axis Language Model: we have presented language as a coherence substrate with its own tense-regime structure, identified the X/Y/Z axes as the denotative, syntactic, and reflective-recursive decomposition of the linguistic coherence vector, and derived from this model five falsifiable predictions distinguishable from transformer-based accounts.

The research program opened by this paper requires collaboration across disciplinary lines that do not normally intersect. We extend an explicit invitation to quantum physicists to test the P312 decoherence signature in photonic and superconducting qubit systems; to neuroscientists to operationalize and measure the coherence-acuity quantity dC/dλ in MEG and EEG studies; to developmental biologists to examine P312 scaling in embryonic patterning; to linguists to test the Z-axis EEG signature predictions; and to AI researchers to design and evaluate architectures with genuinely re-entrant Z-axis processing loops. The framework offers to each of these communities not only a new set of experimental targets but a new theoretical language, a common grammar, grounded in the single concept of coherence, within which each domain’s findings can be read as instances of a single unified phenomenon.

Acknowledgments

This work was conducted independently, without institutional affiliation or external funding. The author thanks the broader communities of theoretical physics, cognitive science, and computational linguistics whose published work provided the intellectual raw material that the present framework attempts to unify. No computational infrastructure beyond standard desktop resources was employed in the Rulial Hypergraph simulations. All errors and speculative overreaches are the author’s own.

Addendum A: Formal Definitions and Equations

A.1 The Unified Operator Stack

Alignment Operator  Projects a substrate state onto its nearest coherent attractor:

Â|ψ⟩ = ∑ᵢ αᵢ|cᵢ⟩ &nbsp;&nbsp; where {|cᵢ⟩} is the coherence basis and αᵢ = ⟨cᵢ|ψ⟩

Aperture Gradient ∇α Measures the rate of change of coherence permeability across the substrate membrane:

∇α = ∂C/∂x &nbsp;&nbsp; where C is local coherence density and x is the membrane coordinate

Pulse Operator P̂ The irreducible oscillatory event that advances the system from one coherence state to the next:

P̂|ψₙ⟩ → |ψₙ₊₁⟩

Master Composition Rule Every generative event in any substrate is expressible as:

Ô_total = P̂ ∘ Â ∘ ∇α

A.2 The P312 Minimal Seed

P312 Conjecture (universality of iterated composition):

∀ substrate S, ∃ n ∈ ℕ such that S ≅ P312ⁿ &nbsp;&nbsp; (up to coherence isomorphism)

A.3 The Coherence Function C(S)

Quantum substrate definition:

C(S) = |⟨ψ|Â|ψ⟩|² / ‖ψ‖²

Classical / biological substrate definition:

C(S) = lim_{ε→0}

\[ I(S, S_ε) / H(S) ]

where I(S, S_ε) is the mutual information between S and a perturbation of magnitude ε, and H(S) is the entropy of the unperturbed substrate.

A.4 The Indeterminant Membrane (IM)

The coherence-phase locus at which no attractor commitment is made:

IM = { ψ : C(ψ) = 0.5 ± ε }

A.5 Tense Regimes: Formal Conditions

Regime	Formal Condition	Dominant Operator
Past-coherent	∇α < 0 (aperture closing)	∇α
Present-operative	∇α ≈ 0 (equilibrium)	Â
Future-generative	∇α > 0 (aperture opening)	P̂

A.6 Ontogenetic Geometry

Morphogenetic field as coherence gradient field:

F = −∇C(x, t)

Cell differentiation = IM crossing events; the body plan = fixed point of P312ⁿ as n → ∞ in the biological substrate.

Formal bridge to Turing morphogenesis: Reaction-diffusion equations are a classical approximation of ∇α dynamics; Ontogenetic Geometry derives them as a special case of P312 application with the IM supplying the pattern-selection boundary condition.

A.7 Intelligence as Acuity of Abstraction

Definition (scale-free, applies from single neurons to AI systems):

I(A) = dC/dλ

where λ is the abstraction level parameter (increasing with representational generality).

Failure modes:

Failure	Formal Condition	Phenomenological Correlate
Misalignment	 projects onto wrong attractor	Confabulation; hallucination; delusion
Aperture saturation	∇α → ∞	Sensory flooding; overfitting; channel saturation
Pulse stalling	P̂ fails to advance	Rumination; perseveration

A.8 Simulation Attractor Value

From Rulial Hypergraph P312 iteration (10³–10⁵ steps), coherence C converges to:

C* ≈ φ⁻¹ ≈ 0.618 &nbsp;&nbsp; (reciprocal of the golden ratio)

IM crossings appear as transient dips to C ≈ 0.5, followed by recovery to a marginally higher attractor, consistent with each crossing generating new coherence basis vectors.

References

[1] Barabási, A.-L., & Albert, R. (1999). Emergence of scaling in random networks. Science, 286(5439), 509–512.

[2] Buzsáki, G. (2006). Rhythms of the Brain. Oxford University Press.

[3] Bybee, J., Perkins, R., & Pagliuca, W. (1994). The Evolution of Grammar: Tense, Aspect, and Modality in the Languages of the World. University of Chicago Press.

[4] Chomsky, N. (1957). Syntactic Structures. Mouton.

[5] Chomsky, N. (1995). The Minimalist Program. MIT Press.

[6] Deacon, T. W. (2012). Incomplete Nature: How Mind Emerged from Matter. W. W. Norton.

[7] Deutsch, D., & Marletto, C. (2015). Constructor theory of information. Proceedings of the Royal Society A, 471(2174), 20140540.

[8] Engel, G. S., Calhoun, T. R., Read, E. L., Ahn, T. K., Mančal, T., Cheng, Y. C., Blankenship, R. E., & Fleming, G. R. (2007). Evidence for wavelike energy transfer through quantum coherence in photosynthetic systems. Nature, 446(7137), 782–786.

[9] Fries, P. (2015). Rhythms for cognition: Communication through coherence. Neuron, 88(1), 220–235.

[10] Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138.

[11] Hameroff, S., & Penrose, R. (2014). Consciousness in the universe: A review of the ‘Orch OR’ theory. Physics of Life Reviews, 11(1), 39–78.

[12] Haroche, S., & Raimond, J.-M. (2006). Exploring the Quantum: Atoms, Cavities, and Photons. Oxford University Press.

[13] Kolmogorov, A. N. (1965). Three approaches to the quantitative definition of information. Problems of Information Transmission, 1(1), 1–7.

[14] Marletto, C. (2021). Constructor theory of life. Journal of the Royal Society Interface, 12(104), 20141226.

[15] Mitchison, G. J. (1977). Phyllotaxis and the Fibonacci series. Science, 196(4287), 270–275.

[16] Northoff, G., & Bermpohl, F. (2004). Cortical midline structures and the self. Trends in Cognitive Sciences, 8(3), 102–107.

[17] Penrose, R. (1994). Shadows of the Mind: A Search for the Missing Science of Consciousness. Oxford University Press.

[18] Ritz, T., Adem, S., & Schulten, K. (2004). A model for photoreceptor-based magnetoreception in birds. Biophysical Journal, 78(2), 707–718.

[19] Shannon, C. E. (1948). A mathematical theory of communication. Bell System Technical Journal, 27(3), 379–423.

[20] Shechtman, D., Blech, I., Gratias, D., & Cahn, J. W. (1984). Metallic phase with long-range orientational order and no translational symmetry. Physical Review Letters, 53(20), 1951–1953.

[21] Spearman, C. (1904). “General intelligence,” objectively determined and measured. American Journal of Psychology, 15(2), 201–292.

[22] Sporns, O., Tononi, G., & Kötter, R. (2005). The human connectome: A structural description of the human brain. PLOS Computational Biology, 1(4), e42.

[23] Tegmark, M. (2014). Our Mathematical Universe: My Quest for the Ultimate Nature of Reality. Knopf.

[24] Tononi, G. (2004). An information integration theory of consciousness. BMC Neuroscience, 5(1), 42.

[25] Tononi, G., Boly, M., Massimini, M., & Koch, C. (2016). Integrated information theory: From consciousness to its physical substrate. Nature Reviews Neuroscience, 17(7), 450–461.

[26] Turing, A. M. (1952). The chemical basis of morphogenesis. Philosophical Transactions of the Royal Society B, 237(641), 37–72.

[27] Valiant, L. G. (1984). A theory of the learnable. Communications of the ACM, 27(11), 1134–1142.

[28] Varela, F. J., Thompson, E., & Rosch, E. (1991). The Embodied Mind: Cognitive Science and Human Experience. MIT Press.

[29] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30, 5998–6008.

[30] Waddington, C. H. (1957). The Strategy of the Genes. Allen & Unwin.

[31] Wei, J., Wang, X., Schuurmans, D., Bosma, M., Xia, F., Chi, E., Le, Q. V., & Zhou, D. (2022). Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems, 35, 24824–24837.

[32] Wolfram, S. (2020). A Project to Find the Fundamental Theory of Physics. Wolfram Media. arXiv:2004.08210.

[33] Zurek, W. H. (2003). Decoherence, einselection, and the quantum origins of the classical. Reviews of Modern Physics, 75(3), 715–775.

Costello, D. (2026). Coherence as Scaling Invariant: Tense Regimes, Operator Architecture, and the Unified Generative Framework Across Matter Substrates. Independent Theoretical Research, Rosendale, NY. arXiv preprint (quant-ph / cs.AI / cond-mat cross-list).