Boolean Tile Basis: Łukasiewicz Logic, Quantum Gates, and Ising Hamiltonians

This document formalizes the extension of Łukasiewicz infinite-valued logic from the real interval $[-1,1]$ to the Bloch sphere, where both logical coordinates become angles on $S^2$. The quadtree IFS coordinates (row = A bits, column = B bits) map directly to Bloch sphere angular coordinates via a cylindrical equal-area projection. We define operations in both systems, prove (or disprove) MV-algebra axioms, derive the 16-operator unitary gate table, and identify exactly which operator gives $z\to z^2$ on $|z|=1$.

Abstract. This paper establishes a homomorphic chain from discrete Boolean logic to quantum information geometry. We identify a phase-preserving extension of Łukasiewicz logic to the unit disk $\mathbb{D}$, proving a fundamental phase-collapse obstruction (Theorem 4.2): any argument-preserving extension necessarily violates the MV4 annihilation axiom, yielding a "near-MV-algebra" that recovers the real-axis MV-sublogic as a classical limit. The investigation proceeds in three distinct layers: (i) Algebraic & Dynamic: We classify the MV-axiom status of the disk and establish the topological conjugacy between iterative logical self-equivalence (the Chaotic Liar) and the angular Bernoulli shift on the Bloch equator. (ii) Quantum Information: We construct a faithful realization of the 16 two-variable Boolean functions as $X$-basis quantum gates via $H^{\otimes 2}$ conjugation, providing an exact parameter map to two-spin Ising Hamiltonians $H_f = J\,ZZ + h_1 ZI + h_2 IZ$ determined by the function’s Walsh-Hadamard coefficients. (iii) Foundational Interpretations: We explore speculative proposals for quantum foundations, including an Everett-style interpretation of IFS depth as a branching process. By distinguishing between established logical results and interpretive extrapolations, we show that the non-Euclidean geometry of the Bloch sphere and the entanglement of the Ising model can be represented within this framework as consequences of phase-preserving logical overflow.

1. Preliminaries

Convention A (throughout): Truth values in $\{-1,+1\}$ with bijection $x = 1 - 2b$ mapping $b\in\{0,1\}$ to $x\in\{+1,-1\}$:

• $\mathsf{True} \to -1$, $\mathsf{False} \to +1$
• Negation is the linear map $\neg x = -x$ (reflection through origin — purely linear, not affine)

Standard Łukasiewicz operators on $[-1,1]$ under Convention A:

• $\neg x = -x$
• $x \oplus y = \max(-1,\; x + y - 1)$ (strong disjunction)
• $x \odot y = \min(+1,\; x + y + 1)$ (strong conjunction, De Morgan dual of $\oplus$)
• $x \leftrightarrow y = 1 - |x - y|$ (equivalence)

Structure of the Paper: The Four Layers

The investigation is organized into four hierarchical layers, transitioning from established logical results to foundational interpretations:

1. Algebra (Established): Near-MV structure on the disk, axiom classification, and the phase-collapse obstruction theorem (\S4).
2. Dynamics (Constructive): The topological conjugacy between iterative self-equivalence and the angular Bernoulli shift; fractal IFS attractors (\S6, \S8, \S12).
3. Quantum Information (Correspondences): Walsh-Hadamard gate construction and the exact mapping of Boolean functions to physical Ising Hamiltonians (\S7, \S13).
4. Foundations (Interpretations): Everett-style branching models and the logic of observation (\S11).

MV-algebra definition. A structure $(A, \neg, \oplus, 0, 1)$ is an MV-algebra if for all $x,y,z\in A$:

• (MV1) $x\oplus y = y\oplus x$ (commutativity)
• (MV2) $x\oplus(y\oplus z) = (x\oplus y)\oplus z$ (associativity)
• (MV3) $x\oplus 0 = x$ (identity)
• (MV4) $x\oplus 1 = 1$ (annihilation)
• (MV5) $\neg\neg x = x$ (involution)
• (MV6) $x\oplus \neg 0 = \neg 0$ (boundary)
• (MV7) $\neg(\neg x \oplus y) \oplus y = \neg(\neg y \oplus x) \oplus x$ (Łukasiewicz axiom)

The derived operations are $x\odot y = \neg(\neg x \oplus \neg y)$ (strong conjunction) and $x\to y = \neg x \oplus y$ (implication).

1.5 From Tiles to Frequencies: A Visual Walkthrough

1.5.1 Start With a Tile

Every 2-variable Boolean function is a 2×2 grid of 0s and 1s. Here is AND:

1.5.2 Decompose Into Walsh Basis Patterns

Any 2×2 truth table can be written as a weighted sum of 4 atomic patterns — the Walsh basis functions. Each pattern corresponds to one of the 4 cells of the truth table having a distinguished sign pattern $\pm1$ when we map $0\to+1$, $1\to-1$ (Convention A):

In Convention A ($0\to+1$, $1\to-1$), the AND truth table $(-1,+1,+1,+1)$ decomposes as:

or equivalently: $\text{AND} = \tfrac12(\chi_\emptyset + \chi_A + \chi_B - \chi_{AB})$. The coefficients $\{2,2,2,-2\}$ (in the $\{\pm1\}$ Walsh convention) are the Walsh spectrum of AND. They say which atomic patterns are present and how strong.

1.5.3 From Walsh Coefficients to Frequencies

Each Walsh basis function maps to a characteristic frequency ratio — the musical interval it produces when stacked across octaves:

where $\varphi = (1+\sqrt5)/2$ is the golden ratio, providing maximally dissonant (3-semitone-like) spacing for the Shepard illusion.

1.5.4 One Tile, Three Projections

The Walsh coefficients are the DNA of a tile. From them we derive three distinct projections:

AND example — Walsh coeffs $\{2,2,2,-2\}$ in the $\pm1$ convention:

• Spectral: All 4 Walsh channels active → a 4-note chord (C + ~F + ~G♯ + ~C♯) at every octave.
• Geometric: Bloch vector $(\tfrac{2}{2},\tfrac{2}{2},-\tfrac{2}{2}) = (1,1,-1)$, normalized: $(0.577,0.577,-0.577)$, purity = 1.
• Spatial: The IFS attractor of AND is the Łukasiewicz continuous extension — a tetrahedral fractal surface with one "cut corner" (the single 0 in the 1110 pattern).

1.5.5 The 16 Tiles at a Glance

The table below shows every tile through all three projections simultaneously. The 3 pure Shepard tones (A, B, XOR) and their negations (NOTA, NOTB, XNOR) each activate exactly one non-DC Walsh channel — they are the fundamental frequencies of the system. The 8 mixed tiles activate all channels — they are the full chords. FALSE and TRUE activate only DC — they are the silent fundamental.

1.5.6 The Three Families

The 16 tiles collapse to just 3 sonic families when heard through the Shepard lens:

2. Two Coordinate Systems

2.1 Bloch Sphere (Angle-Angle Coordinates)

A pure qubit state on the Bloch sphere:

Where:

• Colatitude $\theta_B \in [0,\pi]$ encodes the truth value via $\cos\theta_B = 1 - 2p$ (Convention A: $|0\rangle =$ False $= +1$, $|1\rangle =$ True $= -1$)
• So $p = \sin^2(\theta_B/2) = (1-\cos\theta_B)/2 \in [0,1]$
• Azimuth $\phi \in [0,2\pi)$ is the quantum phase — free in Łukasiewicz logic but essential for interference
• North Pole ($\theta_B=0$): $|0\rangle$, False ($p=0$)
• South Pole ($\theta_B=\pi$): $|1\rangle$, True ($p=1$)
• Equator ($\theta_B=\pi/2$): $p=0.5$, maximal indeterminacy

Key observation: Both Bloch coordinates are cyclic (compact, no boundaries), unlike the real interval $[-1,1]$ used in Łukasiewicz logic.

We also introduce the unit disk $\mathbb{D} = \{z\in\mathbb{C} : |z| \le 1\}$, which parameterizes the Northern hemisphere of the Bloch sphere via stereographic projection: $z = e^{i\phi}\tan(\theta_B/2)$. The boundary $\partial\mathbb{D}$ ($|z|=1$) corresponds to the equator ($\theta_B=\pi/2$). The disk $\mathbb{D}$ provides a complex coordinate system used in the MV-algebra analysis of \S4.

2.2 Coordinate Transform Between Systems

The quadtree IFS uses coordinates $(r,\theta)$ where the radial coordinate $r \in [0,1]$ encodes proposition P ($p = 1-r$) and the angular coordinate $\theta \in [0,2\pi)$ encodes proposition Q ($q = \theta/(2\pi)$). These map to Bloch sphere coordinates $(\theta_B,\phi)$ via a cylindrical equal-area projection:

Or equivalently:

Recalling the truth value definition $p = (1-\cos\theta_B)/2$ from Section 2.1, we have the linear relation:

Under this map:

• Identity P (Index 3, row bits only) maps to the north-south axis of the Bloch sphere.
• Identity Q (Index 5, column bits only) maps to the equatorial phase winding.
• The origin $r=0$ maps to the South Pole ($\theta_B=\pi$, $p=1$, True).
• The outer boundary $r=1$ maps to the North Pole ($\theta_B=0$, $p=0$, False).
• The azimuthal angle $\theta$ maps directly to the quantum phase $\phi$.

Note: The Bloch sphere representation is information-preserving but not operator-preserving for the continuous Łukasiewicz operations: $x\oplus y = \max(-1, x+y-1)$ in quadtree coordinates maps to a nonlinear expression in $(\theta_B,\phi)$. The Bloch sphere's utility is for the quantum gate interpretation, not for simple Łukasiewicz algebra.

3. Operations in Both Systems

We define the Łukasiewicz operators in both coordinate systems and verify they agree under the coordinate transform.

3.1 Negation

3.2 Strong Disjunction $x\oplus y = \max(-1, x+y-1)$

On the Bloch sphere, the midpoint of the chord connecting two state vectors gives the mixed state: $\mathbf{w}_{\text{mid}} = (\mathbf{u} + \mathbf{v})/2$ has $z$-coordinate $w_z = (u_z + v_z)/2 = 1 - (p+q)$, which corresponds to the average $(p+q)/2$, not the Łukasiewicz sum. The Łukasiewicz $\min(1,p+q)$ is the upper bound of the quantum interference term when phases align constructively.

3.3 Strong Conjunction $x\odot y = \min(+1, x+y+1)$

De Morgan duality: $x\odot y = \neg(\neg x \oplus \neg y)$ in all systems.

3.4 Equivalence $x\leftrightarrow y = |x-y|-1$

Correction: On the boundary $|z|=1$, using the disk negation $\neg z = -z$, we find $|z - \neg z| = |2z| = 2$. Thus, $z\leftrightarrow\neg z = \operatorname{proj}_{\mathbb{D}}(2-1) = 1$ for all points on the boundary. This reflects the fact that all points on the equator ($\theta_B=\pi/2$, $p=0.5$) map to the fixed-point peak of the tent map.

4. MV-algebra Axiom Verification Theorem

We verify the MV-algebra axioms for the disk structure $(\mathbb{D}, \neg, \oplus, 0=+1, 1=-1)$.

The Primitive Operation: Multi-argument Summation. The failure of binary associativity (MV2) does not invalidate the IFS process because the "carry-free" evaluation is defined as a single multi-argument sum with a final projection:

This construction ensures that the logical state is well-defined at all depths regardless of the algebraic structure of the derived binary operators. The binary near-MV algebra is an emergent description of the $d=1$ case, not the fundamental building block of the recursion. The failure of binary associativity is the algebraic shadow of the phase-preserving projection — another facet of the same MV4 failure that drives the entire system.

4.1 Significance: The Classical-Quantum Phase Transition

The MV4 failure serves as an algebraic marker for the transition between classical and quantum logical regimes. The real axis $[-1,1]$ (where MV4 holds) corresponds to the classical MV-sublogic. The disk $\mathbb{D}$ (where MV4 fails) corresponds to the quantum extension, where the additional angular degree of freedom breaks the simple annihilation property of classical logic. In this framework, the failure of MV4 is not a defect, but a structural signature of the phase-sensitive information that enables the connection to non-local quantum dynamics.

Structural analogy:

In quantum probability, the classical axiom of commutativity fails — and that failure is the engine of the theory: entanglement, interference, non-locality. Here, MV4 (annihilation: $x\oplus 1 = 1$) plays the analogous role. On the real axis, MV4 holds because $x+1-1 = x$ and real clamping catches values at $\pm1$ without directional ambiguity. On the disk, $z\oplus 1$ for non-real $z$ produces a phase-rotated projection, not the constant $-1$. The phase coordinate is not decorative — it is algebraically active.

This means the system contains two layers:

• The real MV-sublogic $[-1,1]$ — truth values proper, well-behaved, axiomatizable
• The complex extension $\mathbb{D}$ — phase/coherence, unaxiomatizable as MV, but precisely the structure needed for the quantum gate correspondence

The price of this expressive gain is that proof-theoretic tools (completeness, interpolation, etc.) for MV-algebras do not fully apply to $\mathbb{D}$. The correct algebraic framework may be a non-associative or partially-defined structure — a near-MV-algebra whose well-behaved fragment is the real axis. This is not a weakness of the theory but its central feature: the MV4 failure is the point at which the classical algebraic framework opens onto the richer quantum domain.

Bottom line: The system trades axiomatic purity for representational scope. It subsumes both logical and quantum-computational regimes. The real axis is the MV-algebraic classical limit; the disk is the non-MV quantum regime; the MV4 violation is the algebraic phase transition between them.

4.2 How MV4 Failure Enables Expressive Power

The failure of MV4 is not merely a boundary phenomenon. Within this framework, it can be seen as the algebraic precondition for the non-classical features described in subsequent sections. Here we trace the precise mechanism by which MV4 failure unlocks each layer of the hierarchy.

4.2.1 The mechanism: phase-preserving overflow. On the real axis $[-1,1]$, the operation $x\oplus 1$ overflows linearly ($x+1-1 = x$) and clamps to $\pm1$ when $|x|>1$. The clamp discards all directional information. On the disk $\mathbb{D}$, overflow is projected back by the radial projection $\operatorname{proj}_{\mathbb{D}}(w) = w / \max(1, |w|)$ — which scales the magnitude but preserves the argument:

The overflow vector $z-2$ carries information about $z$'s phase, and that phase survives the projection. In the real case, $\arg(z-2) \in \{0,\pi\}$ — just left or right — and both project to $-1$ or $+1$. In the complex case, $\arg(z-2)$ is a continuous degree of freedom that distinguishes inputs that a real-valued logic would identify.

4.2.2 What this buys at each level of the hierarchy.

The geometric avatar of this algebraic structure is the upper envelope of the bitwise OR surface in Grim's value solids: the piecewise-linear clamped plane $\min(1, p+q)$ is the real-axis Łukasiewicz strong disjunction; its complex extension $\operatorname{proj}_{\mathbb{D}}(z+w-1)$ is the MV4-violating operation on the disk. The clamping ridge at $p+q=1$ is precisely where the phase-preserving projection $\operatorname{proj}_{\mathbb{D}}$ becomes active and the argument of $z-2$ becomes algebraically visible — the "outer surface of OR" is the geometric locus of MV4 failure.

4.2.3 Quantifying the expressive advantage. The MV4 failure can be measured by the deviation from real-valuedness:

For $z\in[-1,1]$, $\Delta_{\text{MV4}}(z) = 0$ (MV4 holds). For $z = re^{i\theta}$ with $\theta \neq 0,\pi$, $\Delta_{\text{MV4}}(z) > 0$. The maximum deviation occurs at $z = i$ (pure imaginary):

The total volume of $\mathbb{D}$ where $\Delta_{\text{MV4}} > 0$ equals $\pi$ (the area of the unit disk), while the measure of the real axis where $\Delta_{\text{MV4}} = 0$ is 2 (the length of $[-1,1]$). The ratio $\pi/2 \approx 1.57$ is a coarse measure of the expressive expansion from the classical to the quantum regime.

Corollary. The radial projection $\operatorname{proj}_{\mathbb{D}}(z) = z/\max(1,|z|)$ — the specific extension used throughout this paper — preserves arguments and is identity on $\mathbb{D}$. Therefore it cannot satisfy MV4, consistent with the explicit counterexample $i \oplus 1 \neq 1$. Any alternative projection that restored MV4 would necessarily discard or distort the phase information, eroding the distinctive information geometry that defines the quantum extension.

4.2.4 Relationship to the open questions of \S10. The MV4 failure bears directly on three open questions:

• Question 1 (Can the disk be made into a genuine MV-algebra?): No, without losing the phase dimension. Theorem 4.2 shows that any argument-preserving projection necessarily violates MV4. The system is intrinsically a near-MV-algebra; MV4 failure is not a defect but a feature.
• Question 7 (Rotational symmetry and circuit complexity): The $C_4$ and $C_2$ symmetric tiles (FALSE, TRUE, XOR, XNOR) are exactly the ones whose Walsh spectra have support on a single subset — and therefore are the ones where MV4 holds in their $Z$-basis form. The symmetry-breaking tiles (no rotational symmetry) are the entangling gates, where MV4 fails and the directional sensitivity is maximal.
• Question 9 (Completeness of the tensor hierarchy): Within this construction, the hierarchy depends on MV4 failing at level 2 (the disk). If MV4 held, the structure would collapse to level 0 (Booleans) and level 3 (real MV-algebra) — levels 4-6 (quantum gates, multi-qubit unitaries, Clifford hierarchy) would be unreachable.

In summary: within this framework, MV4 failure is the algebraic condition that distinguishes the classical from the quantum regime. The phase-sensitive overflow makes the difference between a 1D truth-value line and the full 3D Bloch ball with its entangled interior. The gate table, the Ising couplings, and the branching measure all trace their expressiveness to this single algebraic violation.

5. Real-Axis Restriction

For $z = x \in [-1,1] \subset \mathbb{R}$, the projection $\operatorname{proj}_{\mathbb{D}}$ reduces to real clamping:

The restriction recovers the standard Łukasiewicz operators:

Under $x = 1-2p$, each restriction matches the standard $[0,1]$ Łukasiewicz operators. The real axis $[-1,1]$ does form an MV-algebra — this is the classical result that $([-1,1], \max(-1,\cdot), \min(+1,\cdot), -(\cdot))$ under Convention A is isomorphic to the standard Łukasiewicz MV-algebra $([0,1], \min(1,\cdot), \max(0,\cdot), 1-(\cdot))$.

6. The $z \leftrightarrow \neg z$ Connection and the True $z\to z^2$ Operator

A central claim in our earlier discussions was that $z\leftrightarrow\neg z = z^2$ on $|z|=1$. This section separates what is true from what requires a different operator.

Where the $z\to z^2$ Map Actually Lives

The angle-doubling $z\to z^2$ on $|z|=1$ is the analytic continuation of the tent map iteration $f(x) = 2|x|-1$ restricted to $[-1,1]$, lifted to the circle via the topological conjugacy $x = \cos(\pi s)$:

In Łukasiewicz logic on $[-1,1]$, the iterated self-equivalence $f_{n+1} = f_n \leftrightarrow \neg f_n$ gives the tent map:

The topological conjugacy to the logistic map $x\mapsto 4x(1-x)$ via $x = \sin^2(\pi s/2)$, and further to $z\mapsto z^2$ on the unit circle via $x = \cos(\pi s)$, $z = e^{i\pi s}$, gives:

So the operator that gives $z\to z^2$ is not the Łukasiewicz equivalence but the iterated self-referential XOR (the Chaotic Liar):

Clarification: The $z\to z^2$ map is the Bernoulli shift on the binary expansion of $\theta/(2\pi)$, which is exactly the quadtree bit-shifting operation in the IFS. In the IFS code, this is `q_bit` shifting left by one position at each depth level. The identity Q (Index 5) applied iteratively in the chaos game gives $z\to z^2$. So the $z^2$ map is not a single operator on $\mathbb{D}$ but the iterated composition of the Q identity in the IFS recursion.

6.1 The Shift Structure of the 16 Connectives Theorem

The Bernoulli shift and the IFS depth iteration are the same operation: each tick discards one bit and reveals the next. This section formalizes the shift as an operator on the 16 tiles themselves, connects it to the symmetry classification of \S8, and shows how the "bits are frequencies" interpretation (\S6.2) makes the shift a frequency-switching pattern operator.

Definition 6.1 (Left shift). Let $S$ be the left shift on infinite binary sequences: $S(a_1a_2a_3\ldots) = a_2a_3a_4\ldots$. Under the dyadic expansion $\alpha = \sum a_k 2^{-k}$, $S$ acts as $\alpha \mapsto \{2\alpha\}$ (fractional part) — the tent map on $[0,1]$.

The two base propositions generate two periodic sequences:

Eigenvectors of the shift. $P$ has period 2: $S(P) = \neg P$, $S^2(P) = P$. $Q$ has period 4: $S(Q)$ produces a different sequence, $S^2(Q) = \neg Q$, $S^4(Q) = Q$. These are the two lowest-order non-constant cycles of $S$ on binary sequences.

Shift action on the 16 tiles. For each tile $f$, define its output waveform $f_{PQ}$ as the sequence $f(p_k,q_k)$ for $k=1,2,\ldots$. Shifting the output gives $S(f_{PQ}) = f(p_{k+1},q_{k+1})$, which equals $g_{PQ}$ for a different tile $g$. This defines a permutation $\phi$ of the 16 tiles:

Direct computation of all 16 cases yields the following shift orbits:

Theorem 6.3 (Shift-Symmetry Correspondence). The shift orbits of $\phi$ partition the 16 tiles into 6 classes that refine the $D_4$ orbit classification of \S8.3: the constants form 1-cycles (C$_4$ symmetry), the projection P/¬P forms a 2-cycle (C$_1$), the checkerboard family Q/XNOR/¬Q/XOR forms a 4-cycle (C$_2$), and the two entangling families TEE and ETT each form 4-cycles (C$_1$).

The 4-bit base. At each depth $k$, the IFS reads the $k$-th bit pair $(p_k,q_k)$ and evaluates $f(p_k,q_k)$. The 4 possible input combinations $\{(0,0),(0,1),(1,0),(1,1)\}$ are the 4 cells of the truth table and also the 4 contraction maps of the IFS chaos game (\S5). Under the shift $S$, the next depth $k+1$ reads $(p_{k+1},q_{k+1})$ — the same 4-input structure at the next time step. The fixed 64-map tile set (Theorem 4.1 of \S4) is exactly the 16 tiles $\times$ the 4 input pairs — which is why the IFS requires exactly 4 maps per tile: the 4 input pairs exhaust the 2-bit combinations of two propositions, and the shift iterates this 4-way choice through successive bit positions.

Relation to the tent map. The conjugacy $x = \cos\theta$, $z = e^{i\theta}$ established in \S6 maps the shift $S$ on binary sequences to the angle-doubling map $z\mapsto z^2$ on $|z|=1$. The shift orbits above are therefore the cycles of the angle-doubling map acting on the 16 output patterns — a discrete algebraic shadow of the continuous tent-map dynamics.

6.2 Bits as Frequencies Interpretation

The shift structure of \S6.1 describes the 16 tiles as patterns of binary digits evolving under the Bernoulli shift. This subsection reinterprets those patterns in the frequency domain, where the binary values 1 and 0 are not voltage levels but carrier frequencies: bit=1 $\leftrightarrow$ tone $f_1$, bit=0 $\leftrightarrow$ tone $f_0$. The result is a direct mapping from logical connectives to frequency-modulation (FM) synthesis.

Why frequencies instead of voltages. In digital electronics, a binary 1 corresponds to a high voltage (e.g., +5V) and 0 to a low voltage (0V). The waveform of a sequence like $1010\ldots$ is a square wave — discontinuous at every transition, producing audible harmonics at odd multiples of the fundamental. In the frequency interpretation, each binary value corresponds to a distinct pure sine tone. A sequence like $1010\ldots$ becomes an alternating pair of tones $f_1,f_0,f_1,f_0,\ldots$ — a continuous waveform that switches frequency at each tick but never produces square-wave harmonics. The waveform is always smooth within each tick interval.

The 16 tiles as FM patterns. For any tile $f$, the output sequence $f(P,Q)$ is a binary string of period dividing 4 (as established in \S6.1). Under the frequency interpretation, this string becomes a sequence of two sine tones at frequency $f_1$ or $f_0$ according to the bit value at each position. The 16 tiles therefore produce 16 distinct frequency-switching patterns:

• Constants (FALSE, TRUE): steady tone at $f_0$ or $f_1$ — no switching. A pure carrier wave.
• Projections (P, ¬P): alternating $f_1,f_0,f_1,f_0,\ldots$ — square-wave frequency switching at rate $\text{tempo}/2$. Perceived as a rhythmic alternation between two pitches.
• Q and its orbit (Q, XNOR, ¬Q, XOR): patterns of period 4, each with a distinctive rhythm of frequency changes. For example, $Q = 1100\ldots$ gives $f_1,f_1,f_0,f_0,\ldots$ — two ticks of one frequency, two of the other. XOR = 0110... gives $f_0,f_1,f_1,f_0,\ldots$
• TEE and ETT orbits (AND, NOR, A∧¬B, ¬A∧B, etc.): patterns of period 4 with three of one frequency and one of the other. For example, AND = 1000... gives $f_1,f_0,f_0,f_0,\ldots$ — a single tick of $f_1$ followed by three of $f_0$.

The shift as tempo. Under this interpretation, the shift operator $S$ of \S6.1 advances the pattern by one tick, which corresponds to advancing time by one tempo interval $\Delta t = 1/\text{tempo}$. The shift orbit of a tile is therefore the sequence of FM patterns you hear as the pattern advances: starting at offset 0, then offset 1, then offset 2, etc. After a full orbit (period 1, 2, or 4), the pattern repeats. The 6 shift orbits of \S6.1 are therefore the 6 distinct families of FM loops, with periods corresponding to the number of ticks before the frequency pattern repeats.

Frequency ratio and timbre. The ratio $f_1/f_0$ determines the musical interval between the two tones. $f_1/f_0 = 2$ (an octave) gives the most natural separation. $f_1/f_0$ near 1 gives beating and interference. The depth $d$ determines how many ticks are heard before the pattern terminates. At low depth (e.g., $d=4$), only the first few ticks are audible — the pattern is truncated before it completes a full orbit. At high depth ($d \ge 16$), multiple orbits are heard, and the periodic structure becomes perceptible. The tempo parameter controls the physical duration of each tick, setting the overall speed of the frequency switching.

Square-wave elimination. In the voltage-level interpretation, the 16 tiles produce discontinuous square waves whenever the bit value changes. These square waves have harmonics at $3f, 5f, 7f, \ldots$ — an infinite series that introduces frequencies not present in the two-tone palette. In the frequency interpretation, the waveform between ticks is always a pure sine at $f_1$ or $f_0$. There are no harmonics, no discontinuities, no spurious frequencies. The only frequencies present are $f_1$, $f_0$, and the tempo (the switching rate). This spectral purity allows the 16 tiles to be heard as clean two-tone patterns rather than noisy square waves.

Aural orbit detection. Because the 6 shift orbits have distinct period and pattern structures, they are audibly distinguishable:

• Constants: a single steady tone (no switching).
• P orbit (period 2): rapid alternation between two tones — the classic "warble."
• Q orbit (period 4): two ticks of one tone, then two of the other — a slower, more rhythmic alternation.
• TEE and ETT orbits (period 4): one tick of one tone, then three of the other — a galloping rhythm.

These auditory signatures provide a direct sonic classification of the 16 connectives that mirrors the algebraic classification of \S8 and the shift classification of \S6.1.

Relation to the Waveforms companion. The HTML explorer Waveforms.html (distributed with this paper) implements this frequency interpretation as an interactive audio-visual tool. It renders each tile as a two-tone FM pattern with sliders for $f_1$, $f_0$, tempo, and depth, and displays waveform and FFT canvases showing the spectral purity of the generated tones. The 16 tiles are arranged in the standard 4$\times$4 grid of the BTB system, allowing synchronous visual and aural comparison.

Relation to Rademacher and Walsh functions. The two alternating bit-streams $P = 1010\ldots$ and $Q = 1100\ldots$ that generate all 16 tile patterns are discrete dyadic samples of the first two Rademacher functions $R_n(t) = \operatorname{sgn}(\sin 2^{n+1}\pi t)$ — the fundamental square-wave building blocks of the Walsh–Rademacher orthonormal system on $[0,1)$. Specifically, $P$ alternates every sample (period 2) and $Q$ alternates every two samples (period 4), exactly matching the alternation rates of $R_0$ and $R_1$ sampled at the Nyquist rate. The Walsh functions are all finite products of distinct Rademacher functions; for two variables the complete set is $\{1, R_0, R_1, R_0R_1\}$, corresponding to the 4-entry truth tables $\{0000\ldots, 1010\ldots, 1100\ldots, 0110\ldots\}$ — i.e., FALSE, $P$, $Q$, and XOR (their product under Convention A). The full set of 16 tiles arises by taking each of these four Walsh functions with either a $+$ or $-$ sign, forming precisely the 16 elements of the $X$-basis Pauli group $\{\pm I, \pm X\}^{\otimes 2}$ of §7. Under this identification, the 4-entry truth table of every tile is its Walsh–Hadamard spectrum — the four coefficients $\hat f(S)$ of §7 — and the Hadamard transform $H^{\otimes 2}$ converts those Walsh coefficients directly into Pauli operator coefficients. The 16 tiles therefore constitute the complete 2-variable fragment of the Walsh function system, with $P$ and $Q$ as the atomic Rademacher generators from which all 16 Walsh functions are produced by the 16 Boolean connectives.

7. Unitary Gate Table Correspondence

Each 2-variable Boolean operator $f: \{-1,1\}^2 \to \{-1,1\}$ maps to a $4\times 4$ unitary matrix. Under Convention A (True $\to -1$, False $\to +1$) with computational basis $\{|00\rangle,|01\rangle,|10\rangle,|11\rangle\}$, the diagonal matrix $U_f^{(Z)} = \operatorname{diag}(f(00), f(01), f(10), f(11))$ acts in the $Z$-basis. Conjugating by $H^{\otimes 2}$ expresses the operator in the $X$-basis:

where $H = \frac{1}{\sqrt{2}}\begin{pmatrix}1&1\\1&-1\end{pmatrix}$ is the Hadamard gate. This construction is standard: the Walsh-Hadamard transform converts a $\pm1$-valued diagonal into a linear combination of $X$-basis tensor products. What follows is a complete enumeration for the 16 functions of two variables — a direct application of this known transform, not a new claim about the transform itself.

The $X$-basis operators obey the general formula $U_f^{(X)} = \sum_{S\subseteq\{1,2\}} \hat f(S)\, X_S$ where $\hat f(S)$ are the Walsh-Hadamard coefficients and $X_S = \bigotimes_{i\in S} X_i$. For the 8 symmetric operators that are monomials in $x,y$, this collapses to a single tensor product $A\otimes B$ with $A,B\in\{\pm I,\pm X\}$. The remaining 8 operators are mixed (entangling).

Key observations:

• XOR → $X\otimes X$: The parity function $xy$ maps to the 2-bit flip under $H^{\otimes 2}$ conjugation. This generalises: $n$-variable XOR (parity) maps to $X^{\otimes n}$.
• XNOR → $-X\otimes X$: The De Morgan dual of XOR differs only by a global phase of $-1$.
• P ($A$) → $X\otimes I$: Identity P acts as Pauli-X on the first qubit in the $X$-basis.
• Q ($B$) → $I\otimes X$: Identity Q acts as Pauli-X on the second qubit.
• The 8 operators that are monomials in $x,y$ (FALSE, TRUE, P, $\neg A$, Q, $\neg B$, XOR, XNOR) map to pure tensor products $\pm A\otimes B$ with $A,B\in\{I,X\}$. The remaining 8 are entangling superpositions.
• The Toffoli gate (CCX) requires the AND function and the reversible embedding $|x,y,z\rangle \to |x,y,z\oplus(x\land y)\rangle$, which is not of the diagonal-plus-conjugation form used here. No single 3-variable Boolean function maps to Toffoli under $H^{\otimes 3}$ conjugation; Toffoli must be decomposed as a circuit.

Corollary 7.2. The Toffoli gate (CCX) is not of the form $H^{\otimes 3} D_f H^{\otimes 3}$ for any $\pm1$-diagonal $D_f$: $H^{\otimes 3}\cdot\text{CCX}\cdot H^{\otimes 3}$ has non-diagonal entries and diagonal entries $\{1,\frac12\}$, not $\pm1$. Toffoli requires the reversible embedding $|x,y,z\rangle \mapsto |x,y,z\oplus(x\land y)\rangle$, which is a fundamentally different construction. The AND-based control (not XOR-based parity) distinguishes Toffoli from the $X^{\otimes n}$ family.

8. Geometric Classification of the 16 Boolean Tiles Correspondence

The 16 Boolean operators correspond to the 16 functions $f: \{0,1\}^2 \to \{0,1\}$, visualized as $2\times2$ pixel tiles. Under the symmetry group $D_4$ (the rotations and reflections of the square), these 16 tiles partition into classes that reveal the deep connection between logical structure, fractal geometry, and quantum gates.

8.1 Rotational Symmetries

The action of $90^\circ$ rotation on the $2\times2$ grid permutes the four cells $(A,B) \in \{00,01,10,11\}$:

A tile is invariant under this rotation only if all four entries are equal — i.e., the constant functions FALSE (Index 0) and TRUE (Index 15). These have 4-fold rotational symmetry.

A $180^\circ$ rotation swaps $(0,0)\leftrightarrow(1,1)$ and $(0,1)\leftrightarrow(1,0)$. Tiles invariant under this are those with $f(0,0)=f(1,1)$ and $f(0,1)=f(1,0)$, which is exactly XOR (Index 6) and XNOR (Index 9). The checkerboard pattern has 2-fold rotational symmetry.

The remaining 12 tiles have no rotational symmetry (1-fold). They fall into three natural families based on the pattern of live cells:

Under output negation (swap live $\leftrightarrow$ empty), TEE $\leftrightarrow$ ETT as complementary pairs: AND$\leftrightarrow$NAND, NOR$\leftrightarrow$OR, etc. Under variable exchange (swap $A$ and $B$), P$\leftrightarrow$Q. The negation pairs within the rotation classes are:

• FALSE $\leftrightarrow$ TRUE (4-fold, self-dual)
• XOR $\leftrightarrow$ XNOR (2-fold)
• $A\leftrightarrow\neg A$, $B\leftrightarrow\neg B$ (P/Q projections)
• AND$\leftrightarrow$NAND, NOR$\leftrightarrow$OR, $A\land\neg B\leftrightarrow B\to A$, $\neg A\land B\leftrightarrow A\to B$ (TEE/ETT)

The 16 tiles thus collapse to 3 negation types (constants, projections, TEE/ETT) with XOR/XNOR straddling both rotation and negation invariance.

8.2 Bitwise Logic and the Fractal Limit

The digitwise logic of the IFS processes each binary position independently. While this mirrors the carry-free nature of bitwise operations, it is distinct from the continuous Łukasiewicz connectives. For example, bitwise OR on the expansions of $p$ and $q$ corresponds to the union of their fractal address sets, whereas Łukasiewicz strong disjunction corresponds to the linear sum $\min(1, p+q)$.

The self-similar structure of the IFS attractors emerges precisely from this bitwise decoupling. Each recursion depth evaluates the Boolean connective on a new scale, creating the dyadic discontinuities and "jagged" boundaries characteristic of logical fractals. The continuous Łukasiewicz logic serves as the piecewise-linear "envelope" of these fractal surfaces in the infinite-depth limit.

8.3 The 6 Orbit Types for N=4 Functions

The $70$ three-variable Boolean functions $f: \{0,1\}^3 \to \{0,1\}$ with exactly 4 true outputs (N=4) are the natural 3D generalization of the 16 two-variable tiles. Under the full cube symmetry group $S_3 \ltimes \mathbb{Z}_2^3$ (coordinate permutations and bit flips, 48 elements), these 70 functions collapse to just 6 geometrically distinct orbits:

Key observations:

• Only 6 distinct IFS attractor shapes exist for N=4 functions, not 70. The visual diversity is cosmetic — coordinate changes permuting the same self-similar shape.
• "Surface" is not a geometric invariant: the graph-of-a-function property depends on coordinate choice. AND shares an orbit with non-surface functions (orbit 3).
• XOR/XNOR (orbit 5) are genuinely special — the only balanced non-projection surfaces, occupying a singleton orbit of size 2. This is the "pure Sierpiński tetrahedron" shape.
• Orbits 0, 4, 5 are self-complementary under output negation; orbits 1, 2, 3 pair via negation (output flip maps one orbit to another).
• These 6 orbit types correspond to the 6 ways to place 4 vertices in a cube up to relabeling, classified by the number of live-cell pairs sharing an edge: opposite faces (0), L+line (1-2), face+2 points (3), opposite edges (4), alternating tetrahedron (5).

The 16 two-variable tiles of \S1 are precisely the surface subset of these 70 N=4 functions: those for which the 4 kept octants form a graph $z = f(x,y)$, i.e., exactly one $z$-value per $(x,y)$ pair. In the language of the cube, this means each column of the cube (fixed $(x,y)$) contains exactly one kept octant — a pattern of the form $(1,1,1,1)$ (orbit 5), $(2,0,0,2)$ (orbit 4), $(2,2,0,0)$ (orbit 0), or the eight surface members embedded in orbit 3 under coordinate re-labeling. The non-surface orbits 1 and 2, together with the remaining members of orbit 3, keep octants where both $z$ values appear at the same $(x,y)$, producing a relation rather than a function.

This graph property is what connects the 16 tiles to quantum gates (\S7). Writing the four output values of $f: \{0,1\}^2 \to \{0,1\}$ as the diagonal entries of $D_f = \operatorname{diag}(f(00), f(01), f(10), f(11))$ requires exactly one value per input — a function, not a relation. Conjugation by $H^{\otimes 2}$ then yields a unitary $U_f^{(X)}$. A non-surface function would demand two output values at the same input, which cannot be encoded as a single diagonal entry; the operator would be multi-valued or require a degenerate representation (e.g., a density matrix rather than a pure gate). The 16 tiles are therefore special because they are exactly the surface-like members of the 70 N=4 functions, and this is the property that permits their faithful realization as 2-qubit quantum gates.

9. Tensor Hierarchy and Outer Manifolds Correspondence

The 16 Boolean tiles are the seed of a tower of increasingly expressive structures. Each level extends the previous by adding a new dimension of logical or geometric freedom:

9.1 The Embedding Chain

The connection between levels is explicit:

• Level 0 → 1: Each tile $T_k$ defines the IFS rule matrix $M_k$. The chaos game on $M_k$ resolves the operator at every bit-scale, producing the fractal attractor on the unit square.
• Level 1 → 2: The coordinate transform $\Phi(r,\theta) = (\arccos(2r-1), \theta)$ maps the quadtree coordinates to the Bloch sphere. Identity P becomes the polar axis; identity Q becomes the equatorial phase winding.
• Level 2 → 3: Mixed states fill the Bloch ball interior. The midpoint operation $\delta_2(p,q) = (p+q)/2$ corresponds to the Rational Łukasiewicz connective and is the natural fuzzy-logic extension of the sphere surface (pure states) to the ball interior (mixed).
• Level 1 → 4: Each tile's attractor has a quadtree address map. Under $H^{\otimes2}$ conjugation, the diagonal $\pm1$ values become a 4×4 unitary — the tile's quantum gate representation. The IFS recursion depth corresponds to the gate's resolution in the computational basis.
• Level 4 → 5: The $n$-variable parity function $f(x_1,\ldots,x_n) = x_1\cdots x_n$ under $H^{\otimes n}$ gives $X^{\otimes n}$ (Theorem 7.1). The 16 two-variable operators are the $n=2$ base case of this tower.
• Level 5 → 6: The gates $\{X\otimes I, I\otimes X, H\otimes H\}$ acting on the diagonal generate the Clifford group. The entangling operators (AND, OR, etc.) provide the remaining generators for universal quantum computation.

9.2 OR = Addition and the Tensor Product

At every level, the structural principle is the same: independent variables compose by bitwise addition, and the clamping ($\min(1,p+q)$ for OR, projection for $\mathbb{D}$) handles the overlap that creates entanglement:

• In the Boolean level: disjoint bits add; overlapping bits saturate.
• In the IFS level: each recursion depth evaluates independent bits; the attractor is the fixed point of disjoint affine contractions.
• In the quantum level: the $X$-basis images of disjoint-spectral operators are pure tensor products; overlapping-spectral operators are entangling.

The "outer manifold" is not a single space but the tower of embeddings connecting Boolean, fractal, and quantum descriptions of the same 16-operator alphabet. The IFS quadtree is the recursive view (bit-level evaluation); the Bloch sphere is the quantum view (unitary evolution); the real axis is the logical view (truth values). The coordinate transform $\Phi$ between them shows they are the same structure seen from different angles.

9.3 Three-Domain Embedding (Revised)

The hierarchy from the theory's three-domain structure:

maps onto the tensor hierarchy as:

• Boolean vertices ($\{\pm1\}^n$): the 16 tiles as discrete truth tables. Level 0.
• Fuzzy interior ($[-1,1] \subset \mathbb{D}$): the real MV-sublogic and its complex extension. Levels 1-3.
• Quantum superposition ($H^{\otimes n}$-conjugated unitaries): the gates. Levels 4-7.

Each level contains the previous as a substructure. The real axis $[-1,1]$ is the intersection of the fuzzy and quantum domains — the "classical limit" where MV4 holds and the unitaries reduce to stochastic matrices. The disk $\mathbb{D}$ spans the fuzzy and quantum domains, with the MV4 failure marking the boundary between them.

9.4 Universality of the 16-Connective Gate Set Theorem

Level 7 of the hierarchy is the apex: for any finite $n$, the 16 tiles generate a gate set dense in $SU(2^n)$. The construction uses two extensions already present in \S9.1 and \S10:

1. Selective Hadamard conjugation (\S9.3, \S13.7): apply $H$ to a chosen subset $S$ of qubits, giving $U_f^S = H_S \, D_f \, H_S$ where $D_f$ is the $\pm1$-diagonal seed of the tile.
2. Rational-phase diagonal seeds (\S10): replace $\pm1$ entries $(-1)^{f(x)}$ with $U(1)$ phases $e^{i\pi\cdot f(x)/k}$ for integer $k\ge 1$.

These extensions are not external additions — the selective Hadamard is the domain bridge of \S9.3, and the rational-phase diagonals are the natural continuous completion of the $\pm1$-valued truth tables. Together they complete the 16-connective framework into a universal gate set.

Lemma 1 (CNOT). Let $D_{\text{AND}} = \operatorname{diag}(+1,+1,+1,-1)$ be the $4\times4$ diagonal of AND (Index 1) under Convention A. Conjugating by $I\otimes H$ (selective Hadamard on qubit 2 only) gives:

where $\text{CNOT} = |0\rangle\langle0|\otimes I + |1\rangle\langle1|\otimes X$ is the controlled-NOT gate with qubit 1 as control and qubit 2 as target.

Lemma 2 (Toffoli). Let $D_{\text{AND}_3} = \operatorname{diag}(+1,\ldots,+1,-1)$ be the $8\times8$ diagonal of the 3-variable AND function ($f(x,y,z)=1$ iff $x=y=z=1$). Conjugating by $I\otimes I\otimes H$ (selective Hadamard on qubit 3 only) gives the Toffoli gate:

Lemma 3 (Hadamard and Phase generators). The Hadamard gate $H$ is directly available as the selective-Hadamard basis change on any single qubit ($n=1$, $S=\{1\}$). The phase gate $S = \operatorname{diag}(1,i) = T^2$ comes from the $k=2$ rational-phase diagonal for the constant-false function (Index 0): $D^S_{\text{FALSE}} = \operatorname{diag}(1, i)$. The $T$ gate $T = \operatorname{diag}(1, e^{i\pi/4})$ comes from $k=4$:

Lemma 4 (Clifford group). The gates $\{H, S, \text{CNOT}\}$ generate the $n$-qubit Clifford group. This is a standard result in quantum computation (Nielsen \& Chuang, \S10.5.3): $H$ and $S$ generate the single-qubit Clifford group $\langle H,S\rangle \cong$ the binary octahedral group (24 elements), and adding CNOT gives the full $n$-qubit Clifford group by the Gottesman-Knill theorem.

Theorem 1 (Universality). The 16 tiles, together with selective Hadamard conjugation and rational-phase diagonal extensions, form a universal gate set for quantum computation. Specifically, the set $\{H, T, \text{CNOT}\}$

1. $H$: any single-qubit selective-Hadamard application,
2. $T = \operatorname{diag}(1, e^{i\pi/4})$: the $k=4$ rational-phase diagonal of FALSE,
3. CNOT: AND (Index 1) conjugated by $I\otimes H$,

is universal. By the Solovay–Kitaev theorem (Dawson \& Nielsen, 2006), any $U\in SU(2^n)$ can be approximated to precision $\epsilon$ by a circuit of $O(\log^{c}(1/\epsilon))$ gates from this set, where $c\approx 4$.

Corollary 1 (Compositional universality). For any finite $n$, the set of all $n$-qubit unitaries realizable as depth-$d$ compound sentences over the 16 connectives, with selective Hadamard domain assignment and $U(1)$-phase diagonal extensions, is dense in $SU(2^n)$ in the limit $d\to\infty$.

Level 7 of the hierarchy. The tensor hierarchy table of \S9.1 now extends to:

Significance. Theorem 1 closes Gap 2 of the Continuum Conjecture: the 16 connectives, under their natural extensions (selective Hadamard domain assignment and rational-phase diagonals), generate a gate set dense in $SU(2^n)$. The universality is not a separate construction — it is the natural completion of the tensor hierarchy. Level 6 (Clifford hierarchy) already contains all Clifford gates; Level 7 adds the non-Clifford phases needed for density. Every step uses the same diagonal-seed principle: a Boolean function $f$ encoded as phases on computational basis states, transformed by selective Hadamard into the $X$-basis where it becomes a quantum gate. The hierarchy is not a tower of different theories but a tower of increasingly expressive uses of the same 16-connective alphabet.

9.5 Bell Inequality Violation Theorem

The framework produces a specific, quantitative, empirically falsifiable prediction: the CHSH correlation bound derived from the 16 connectives is $S = 2\sqrt{2}$ — the Tsirelson bound of standard quantum mechanics — violating the classical bound $S \le 2$. This is not a separate assumption; it follows from the proved machinery of \S9.4 and \S11.8.

Lemma 5 (Bell state from the 16 connectives). The Bell state $|\Phi^+\rangle = (|00\rangle + |11\rangle)/\sqrt{2}$ is produced by applying $H\otimes I$ followed by CNOT to the initial state $|00\rangle$. Both operations are framework-constructible:

where $H\otimes I$ is selective Hadamard on qubit 1 (Lemma 3) and CNOT is AND (Index 1) conjugated by $I\otimes H$ (Lemma 1). Equivalently, $|\Phi^+\rangle$ is the output of the AND gate at input branch $(A,B) = (1,1)$, already noted in \S11.6.

Lemma 6 (Rotated basis). A measurement in the basis $B_1 = (Z+X)/\sqrt{2}$ (rotated $45^\circ$ in the XZ plane) is implemented as a Z-basis measurement preceded by the unitary $R_y(\pi/4) = S\,H\,T\,H\,S^\dagger$, where each factor is a framework operation:

• $H$: single-qubit selective Hadamard (Lemma 3).
• $S = \operatorname{diag}(1,i)$: the $k=2$ rational-phase diagonal of FALSE (Lemma 3).
• $T = \operatorname{diag}(1,e^{i\pi/4})$: the $k=4$ rational-phase diagonal of FALSE (Lemma 3).
• $S^\dagger = \operatorname{diag}(1,-i)$: the $k=2$ conjugate phase.

Composing: $R_y(\pi/4) = S\,H\,T\,H\,S^\dagger$. Similarly, $B_2 = (Z-X)/\sqrt{2}$ is a Z-basis measurement preceded by $R_y(-\pi/4) = S^\dagger\,H\,T\,H\,S$.

Theorem 2 (Bell inequality violation). Let $|\Phi^+\rangle$ be the Bell state constructed in Lemma 5. Define Alice's measurements as $A_1 = Z$ (computational basis) and $A_2 = X$ (Hadamard conjugate basis), and Bob's measurements as $B_1 = (Z+X)/\sqrt{2}$ and $B_2 = (Z-X)/\sqrt{2}$ (the rotated bases of Lemma 6). Then the CHSH correlation:

equals $2\sqrt{2}$, violating the classical bound $S \le 2$ and achieving the Tsirelson bound of quantum mechanics.

Corollary 2 (Consistency with Tsirelson's bound). The 16-connective framework yields $S = 2\sqrt{2}$ as the maximum CHSH correlation achievable within its logical structure, matching the Tsirelson bound of quantum mechanics. Any experiment measuring $S > 2\sqrt{2}$ would be incompatible with the framework; any experiment measuring $S \le 2$ would indicate that the framework's quantum correlations are not being realized. The value $2\sqrt{2}$ is derived from the algebraic structure of the 16 connectives, the selective Hadamard domain assignment (\S13.7), and the IFS branching measure (\S11.8), rather than imported from standard QM.

Significance. Theorem 2 is the first empirical prediction of the 16-connective framework that does not merely reproduce a known quantum result but derives a specific quantitative boundary from logical first principles. The CHSH value $S = 2\sqrt{2}$ emerges from the same algebraic structure that gives the universality proof (Theorem 1) and the Born rule (Theorem 11.1). The boundary $S = 2$ (classical) and $S = 2\sqrt{2}$ (quantum) are separated by exactly the MV4 failure — the existence of the complex phase dimension that standard Boolean logic lacks. The CHSH violation is therefore a direct, measurable consequence of the near-MV algebra of \S4.

10. Open Questions

1. Can the disk be made into a genuine MV-algebra? (RESOLVED.) \S4.2 shows this is impossible without collapsing the structure. Any modified projection $\widetilde{\operatorname{proj}}$ that enforces $z\oplus 1 = 1$ must discard the argument of $z-2$ for all $z\notin[-1,1]$, which maps $\mathbb{D}$ onto the real line. The disk is intrinsically a near-MV-algebra: the minimal algebraic weakening of MV that preserves the phase dimension. Axiom MV4 is replaced by the radial projection identity $\arg(z\oplus 1) = \arg(\operatorname{proj}_{\mathbb{D}}(z-2)) = \arg(z-2)$, which is the expressive engine of the entire system (\S4.2.2).
2. Bloch ball as a Rational MV-algebra (RESOLVED). The Bloch ball $\mathcal{B} = \{\rho = \frac12(I + \mathbf{r}\cdot\boldsymbol{\sigma}) : \mathbf{r}\in\mathbb{R}^3, |\mathbf{r}|\le 1\}$ forms a Rational MV-algebra (RMV-algebra) under the following operations:
   • Negation: $\neg\rho = I - \rho = \frac12(I - \mathbf{r}\cdot\boldsymbol{\sigma})$, corresponding to central inversion $\mathbf{r}\mapsto -\mathbf{r}$.
   • Midpoint (dyadic scalar): For any dyadic rational $q = k/2^m \in [0,1]\cap\mathbb{Q}$, the operation $\delta_q(\rho_1,\ldots,\rho_{2^m})$ is the weighted mixture $\frac{1}{2^m}\sum_{i=1}^{2^m} \rho_i$. For $m=1$ this is $\delta_2(\rho_1,\rho_2) = (\rho_1+\rho_2)/2$.
   • Rational scalar: $q\cdot\rho = q\rho + (1-q)\rho_0$ where $\rho_0 = I/2$ is the maximally mixed state (origin), defined by iterated midpoint operations.
   The axioms are those of Rational Łukasiewicz logic (Riečan's theorem7, 1999): an RMV-algebra is a commutative monoid $(A,\oplus,0)$ with a scalar multiplication $\mathbb{Q}\cap[0,1] \times A \to A$ satisfying $q\cdot(r\cdot a) = (qr)\cdot a$, $(q+r)\cdot a = q\cdot a \oplus r\cdot a$ when $q+r\le 1$, and $q\cdot a \oplus (1-q)\cdot a = a$. All hold for the Bloch ball with $\oplus$ interpreted as the convex combination with the maximally mixed state: $\rho_1\oplus\rho_2 = (\rho_1+\rho_2)/2$.

   Key insight. The Bloch ball is the state space of a 2-level quantum system. In classical probability theory, the state space of an $n$-outcome system is an $(n-1)$-simplex — a convex set where every point has a unique decomposition into extremal (pure) states. The Bloch ball is not a simplex: a mixed state $\rho$ with $|\mathbf{r}| < 1$ can be decomposed into pure states in infinitely many ways. This non-simplex geometry is the geometric expression of quantum contextuality and the failure of MV4 on $\mathbb{D}$ (\S4.2). The boundary $\partial\mathcal{B} = S^2$ (pure states) corresponds to the equator of the MV4 failure region: the pure states are exactly where $|\mathbf{r}|=1$ and the phase (azimuth) Ising-active. The interior $|\mathbf{r}|<1$ is where the MV-algebra becomes Rational and the non-unique decomposition reflects the extra degrees of freedom that classical logic cannot capture.

   The RMV-algebra structure on the Bloch ball identifies the precise algebraic difference between classical and quantum probability: the classical MV-algebra on $[0,1]$ is the state space of a 1-bit system (a simplex, unique decompositions), while the Bloch ball is the state space of a 1-qubit system (a 3-ball, non-unique decompositions). The Rational Łukasiewicz operations on the Bloch ball are the quantum generalization of the classical MV connectives.

3. 16 gates and circuit identities (RESOLVED). \S13.2 gives a complete parameterization of all 16 gates as points in the 3-parameter Ising family $H(J, h_1, h_2)$. The monomials occupy the coordinate axes (one non-zero parameter, magnitude $\pi/2$). The entangling gates occupy the interior (all three parameters non-zero, magnitude $\pi/4$). The pattern of signs ($\pm\pi/4$ for each of $J, h_1, h_2$) generates all 8 entangling operators as the $2^3 = 8$ sign combinations, with the global phase distinguishing complement pairs. The generator set $\{X\otimes I,\; I\otimes X,\; H\otimes H\}$ of the earlier partial resolution is precisely the set of $\pi/2$ rotations about the $X$-axis and the basis-change that converts $X$ to $Z$ — which is exactly the Ising parameter map of \S13.2.
4. Unit disk vs Bloch sphere for the $z^2$ map (RESOLVED). \S6.1 and \S12.4 provide a complete answer. The $z\to z^2$ map is the angular Bernoulli shift $\theta_{n+1}=2\theta_n\pmod{2\pi}$, conjugate to the iterated tent map $\operatorname{tent}^{n}$ on $[-1,1]$ (Theorem 6.2). No single complex-linear operator on $\mathbb{D}$ can give $z\to z^2$ on the boundary under iteration, because any non-zero complex-linear map $L(z) = \alpha z$ (with $\alpha\in\mathbb{C}$) either scales magnitudes or rotates, neither of which squares the angle while preserving $|z|=1$. The $z^2$ map is quadratic, not linear. However, the depth process of \S12.4 constructs $z\to z^2$ as the $d\to\infty$ limit of the $n=1$ quantum states $|\psi_d\rangle$: at each depth $d$, the state encodes the $d$-bit approximation of the tent map, and in the limit the Bernoulli shift emerges as the asymptotic dynamics on the input register. The $z^2$ map is thus not a single operator but the infinite-depth limit of the iterative self-referential process — exactly the structure described in Theorem 6.2 and Claim 12.3.
5. Higher-dimensional analogues (RESOLVED). The $n$-variable XOR (parity) under $H^{\otimes n}$ conjugation gives $X^{\otimes n}$ (the $n$-bit flip), not a generalized CNOT. Specifically:
   • $n=2$: XOR $\to X\otimes X$ (not CNOT)
   • $n=3$: XOR $\to X\otimes X\otimes X$ (not Toffoli/CCX)
   • General: XOR$_n$ $\to X^{\otimes n}$
   The Toffoli gate requires the AND function and the reversible embedding $|x,y,z\rangle \to |x,y,z\oplus(x\land y)\rangle$, which is not of the diagonal-plus-conjugation form. The Toffoli gate cannot be expressed as $H^{\otimes 3} D_f H^{\otimes 3}$ for any $\pm1$-diagonal $D_f$.
6. Three-domain embedding (CLARIFIED). The embedding chain given in \S9.1 and \S9.3, $$ [-1,1] \;\subset\; \mathbb{D} \;\subset\; \operatorname{SU}(2^n), $$ is the correct description within standard complex quantum mechanics. The hierarchy is real $\to$ complex $\to$ group-of-unitaries — a structural tower, not a number-system tower.

   The quaternionic extension requires a more nuanced analysis. Unit quaternions $S^3 \subset \mathbb{H}$ carry the structure of $\operatorname{Cl}(4,0)$ spinors. The Hopf fibration $\pi: S^3 \to S^2$ projects these spinors onto the standard Bloch sphere, recovering ordinary quantum states on the base while the $\operatorname{U}(1)$ fiber carries the Kitaev $\nu$ classification. The chain extends to:

   $$ [-1,1] \;\subset\; \mathbb{D} \;\subset\; S^2 \;\subset\; S^3 \;\subset\; \operatorname{Cl}(4,0) $$
   • Real axis $[-1,1]$: the classical MV-algebra, 1-dimensional. Level 3.
   • Complex disk $\mathbb{D}$: the near-MV-algebra, 2-dimensional. Levels 1--3.
   • Bloch sphere $S^2$: pure quantum states, the base of the Hopf fibration.
   • Quaternion spinors $S^3$: the total space of the Hopf fibration, carrying the $\operatorname{Cl}(4,0)$ anyon structure.
   • Unitary group $\operatorname{SU}(2^n)$: the $n$-qubit gate space. Levels 4--6.
   The Hopf fibration resolves the apparent conflict: the Born rule lives on the base $S^2$ (standard QM), while the anyon classification lives in the fiber $\operatorname{U}(1)$ (Kitaev). The quaternion path is not a replacement for standard QM but an extension that adds anyon structure. Whether this extension is physically realized remains open.
7. Rotational symmetry and circuit complexity (RESOLVED). \S13.2 answers this directly. The symmetry class under $D_4$ predicts the number of non-zero Ising parameters:
   • $C_4$ symmetry (FALSE, TRUE): 0 non-zero parameters — the identity and global phase, trivially symmetric.
   • $C_2$ symmetry (XOR, XNOR): 1 non-zero parameter — pure $ZZ$ coupling $J = \pm\pi/2$, a rank-1 tensor product in the $X$-basis.
   • No rotational symmetry (the remaining 12 tiles): 3 non-zero parameters for the 8 entangling gates (all $J, h_1, h_2 \neq 0$), and 1 non-zero parameter for the 4 projections P, $\neg A$, Q, $\neg B$ (pure single-qubit $X$ terms, which have $C_1$ symmetry under $D_4$).
   The entangling operators are systematically more complex in a precise measure: they require all three Ising couplings simultaneously, whereas the monomials use at most one. The number of non-zero parameters equals the number of Walsh subsets supporting non-zero spectral weight, which equals $2^{\text{#non-zero WH coefficients}} - 1$. For the entangling gates with 3 non-zero spectral coefficients, the Ising Hamiltonian has full support on $\{J, h_1, h_2\}$.
8. 6 orbit types and Hausdorff dimension (RESOLVED). All 16 tiles use the same 4-contraction IFS on $[0,1]^2$ with the same affine maps (the 4 corners of the square). The only variation between tiles is which subsets of maps contribute to the height coordinate $z = f(A,B)$. At infinite depth, each tile's attractor is the graph of a continuous function (the Łukasiewicz extension of the Boolean connective, \S1). Since the carry-free evaluation $f_d$ converges uniformly to a continuous function as $d\to\infty$, the graph of any continuous function on $[0,1]^2$ has Hausdorff dimension 2. Therefore all 6 orbit types have Hausdorff dimension 2. The visual differences between them (face, L+line, tetrahedron, alternating tetrahedron) are not differences in fractal dimension but in the Walsh spectrum: how many and which combinations of input bits influence each output bit. Orbit 4 (projections $A,B$) depends on only one input — producing a visually lower-dimensional "line-like" surface, but it remains a full 2D graph in the limit. Orbit 5 (XOR/XNOR) depends on the parity of both inputs — producing the alternating tetrahedron pattern. Orbit 3 (entangling gates) depends on both inputs asymmetrically. All are dimension 2.
9. Completeness of the tensor hierarchy (RESOLVED). \S4.2.4 shows that within this construction the hierarchy depends on MV4 failing — if MV4 held, levels 4--7 would be unreachable. The hierarchy extends beyond the symmetric 16 gates in two principled directions:
   • Rational-phase diagonal seeds (emergent from the continuous geodesic). The $\pm1$ diagonal entries $(-1)^{f(x)}$ are the $t=1$ endpoint of the continuous Hamiltonian geodesic $U_f(t)$ of \S11.7. At time $t=1/k$, the $H^{\otimes 2}$-conjugate diagonal is: $$ D_f^{(k)} \propto H^{\otimes 2} U_f(1/k) H^{\otimes 2} = \cos\!\left(\frac{\pi}{2k}\right) I - i\sin\!\left(\frac{\pi}{2k}\right) D_f \propto \operatorname{diag}\bigl(e^{i\pi f(00)/k}, \ldots, e^{i\pi f(11)/k}\bigr), $$ up to global phase. The rational-phase seeds are not a separate constructive extension — they are the intermediate points of the geodesic between identity and the full gate, revealed by evaluating the continuous-time Hamiltonian flow at rational $t$. The seed remains diagonal; $H^{\otimes n}$ conjugation produces a $k$-th root of the original Clifford gate. For $k=1$: the standard $\pm1$ gates (\S7). For $k=2$: $\pi/2$ phases (square-root Clifford). For $k=4$: $\pi/4$ phases ($T$-like, non-Clifford). The Ising coupling parameters scale as $\pi/(2k) \cdot \hat f(S)$.
   • Selective Hadamard conjugation. Instead of $H^{\otimes n}$ on all qubits, apply $H$ to a subset $S$ of qubits and identity to the rest. The gate is $U_f^S = H_S \, D_f \, H_S$ where $H_S = \bigotimes_{i\in S} H_i \otimes \bigotimes_{i\notin S} I_i$. The seed $D_f$ remains fully diagonal — the Boolean function $f$ is still encoded as $\pm1$ phases on the computational basis states. The difference is which qubits receive the basis change from $Z$ (control, classical) to $X$ (target, quantum).

   This second extension unlocks the full Clifford hierarchy. The pattern:

   • $n=1$, $f = \text{ID}$ (single-bit identity), $S = \{1\}$: $H \, \operatorname{diag}(1,-1) \, H = HZH = X$ (Pauli level)
   • $n=2$, $f = \text{AND}$, $S = \{2\}$: $(I\otimes H) \, \operatorname{diag}(1,1,1,-1) \, (I\otimes H) = \text{CNOT}$ (Clifford level 2)
   • $n=3$, $f = \text{AND}_3$, $S = \{3\}$: $(I\otimes I\otimes H) \, \operatorname{diag}(1,\ldots,1,-1) \, (I\otimes I\otimes H) = \text{Toffoli}$ (Clifford level 3)
   • The symmetric case $S = \{1,\ldots,n\}$ recovers the 16 two-qubit gates (\S7) and their $n$-qubit generalizations (\S12).

   The structural unity is: the diagonal seed $D_f$ encodes the Boolean condition; the subset $S$ selects which qubits are targets (quantum $X$-basis) vs controls (classical $Z$-basis). The carry-free evaluation survives on the target qubits (their $X$-basis terms commute), while the control qubits remain in the $Z$-basis, acting as classical branch selectors. This is exactly the many-worlds picture of \S11: the control qubits index the Everett branch; the target qubits carry the quantum dynamics within each branch.

   The overall hierarchy is thus: real axis $\subset$ complex disk $\subset$ $\{\pm1\}$ diagonals + uniform $H^{\otimes n}$ $\to$ 16 symmetric gates $\subset$ $\{\pm1\}$ diagonals + selective $H_S$ $\to$ full Clifford group (including CNOT, Toffoli) $\subset$ $U(1)$ diagonals + selective $H_S$ $\to$ full Clifford hierarchy (including $T$, $V$, etc.) $\subset$ Solovay–Kitaev closure $\to$ dense in $SU(2^n)$ (Theorem 1, \S9.4). Every layer preserves the same $D_f$ seed; the only variations are which qubits receive the Hadamard basis change and whether the diagonal phases are $\pm1$ or $U(1)$. Toffoli is not outside the framework — it was always there, waiting for us to apply $H$ to the right subset of qubits.

11. Many-Worlds and the Universal Wavefunction Interpretation

This section presents the natural interpretation of the IFS-to-Bloch-sphere architecture within the Everett (many-worlds) interpretation of quantum mechanics. Every element of the construction — the chaos game, the unitary gates, the partial trace — maps directly onto the many-worlds picture. Note: the formalism of the preceding sections is compatible with several interpretations of quantum mechanics (Copenhagen, QBism, relational, consistent histories); the many-worlds interpretation is presented here as the one that follows most naturally from the branching structure of the IFS, not as a forced interpretive overlay.

11.1 The IFS as Branch Generator

The chaos game on the 4-corner contraction maps:

generates an infinite binary string $(a_1b_1)(a_2b_2)(a_3b_3)\cdots$ at each step. Each string picks a unique point $(A,B) \in [0,1]^2$ — the limit of the contraction sequence. In Everett's interpretation, each complete path through the IFS is a branch of the universal wavefunction: one complete history of the two classical Boolean variables $A$ and $B$. The set of all possible paths is the set of all possible classical input branches, isomorphic to $[0,1]^2$.

The IFS measure $\mu$ on $[0,1]^2$ (uniform when all four maps are active, fractal when constrained by a truth table) is the branching measure — the relative weight of each classical branch under the IFS dynamics.

11.2 The Unitary as World-Coupler

Each branch $(A,B)$ defines a separable 2-qubit input state:

The unitary gate $U_f = H^{\otimes 2} D_f H^{\otimes 2}$ (from $\S7$) acts linearly on each branch:

The output $|\psi_f(A,B)\rangle$ may be separable or entangled. When it is entangled, the two world-sets (A-world and B-world) are correlated — the outcome of a measurement on qubit 1 is not independent of qubit 2. Entanglement between the two qubits is inter-world correlation in the many-worlds picture.

11.3 The Marginal World-Set

The reduced density matrix of qubit 1 is obtained by tracing out qubit 2:

This is the marginal state of the A-world-set when the B-worlds are discarded — exactly analogous to a marginal probability distribution in classical statistics, but for quantum states. The Bloch vector of $\rho_1$ is:

The Bloch ball surface ($r=1$) is the space of independent world-sets; the interior ($r<1$) is the space of entangled (inter-world-correlated) states. This structure is directly visualizable: each branch $(A,B)$ maps to a point $\mathbf{r} \in \mathbb{R}^3$ (the Bloch vector of its marginal state), with $r = |\mathbf{r}|$ encoding the purity via a viridis colormap (bright yellow = pure $r=1$, dark purple = maximally mixed $r=0$). Product gates confine all branches to the sphere surface; entangling gates produce interior points revealing the entanglement structure at a glance.

11.4 Universal Wavefunction Integral

The collection of all branches, weighted by the IFS measure, defines the analogue of a universal wavefunction for the model:

Here $\mu$ is the IFS measure on $[0,1]^2$. When all four maps are active, $\mu$ is the uniform (Lebesgue) measure; when constrained by a truth table (only maps for which $f(A,B)=1$ are active), $\mu$ is the fractal measure supported on the attractor of the constrained IFS.

The integral is structurally analogous to a Feynman path integral: it sums over all possible classical histories (branches), each weighted by its measure under the IFS dynamics, and each contributing a quantum state $U_f|\alpha\beta\rangle$. Unlike the physical path integral, the time parameter is replaced by the IFS depth (recursion level), and the action is replaced by the gate's truth table.

This is not the universal wavefunction of a physical quantum system — it is the mathematical model's universal wavefunction, defined over all possible classical inputs. The "universal" aspect is that the IFS generates all possible $(A,B)$, so the integral covers the full branch space.

11.5 Relation to the Three-Domain Embedding ($\S9.3$)

The three-domain hierarchy of $\S9.3$ admits a natural interpretation in the many-worlds picture:

The MV4 failure ($\S4.1$) — the point where the complex disk $\mathbb{D}$ ceases to satisfy the MV-algebra axiom $x\oplus 1 = 1$ — can be interpreted as the algebraic condition for branch entanglement within this framework. On the real axis $[-1,1]$ (classical regime), MV4 holds and all branches are pure ($r=1$). On the complex disk (quantum regime), MV4 fails exactly where branches can entangle ($r<1$). The phase transition of $\S4.1$ can be seen as the transition from independent to correlated world-sets.

11.6 Visualizing the Branch Structure

The mapping from IFS branches to Bloch vectors defines a direct geometric visualization of Everett's branch space:

• Topography mode: a regular grid over $(A,B)$ samples branch space uniformly, showing the full image of the input multiverse in the output Bloch ball. Each point is one branch; its position is $\rho_1$'s Bloch vector; its color (viridis) encodes the purity $r$ — bright yellow for pure ($r=1$), dark purple for maximally mixed ($r=0$).
• Chaos mode: the IFS chaos game samples $(A,B)$ with the fractal measure $\mu$, mapping the IFS attractor into the Bloch ball. The point density in the ball reflects the branching measure — the relative weight of each classical history under the IFS dynamics.
• Phase basis: rotating the phase $\phi$ of the input states $|\alpha(A)\rangle, |\beta(B)\rangle$ changes which branches appear entangled. The same IFS branches, viewed in a different phase basis, produce different entanglement patterns. This illustrates the basis-dependence of the branch decomposition — a central theme in Everettian interpretations.
• Product gates (P, Q, XOR, XNOR, etc.): all branches remain on the sphere surface ($r=1$). The A and B world-sets never entangle — the image of branch space is confined to the boundary of the Bloch ball, a sphere of independent worlds.
• Entangling gates (AND, OR, NAND, etc.): branches appear in the interior ($r<1$) — regions where the gate couples the two world-sets into an entangled state. The viridis color scale reveals the entanglement structure at a glance.

Striking example. For the AND gate at the branch $(1,1)$ (both inputs True), the output is a Bell state — maximally entangled, maximally mixed reduced state ($r=0$). That branch appears at the exact center of the Bloch ball, dark purple. The classical AND function outputs True, but the quantum gate reveals that this "truth" is shared between two worlds in a maximally correlated way.

Philosophical interpretation. The universal wavefunction of the model — the state encoding all branches simultaneously — is the integral of all branch states weighted by the IFS measure:

This integral is the precise analogue of Everett's universal wavefunction within the model. It sums over all possible classical histories (all IFS paths), each weighted by its branching measure $\mu$, and each contributing a quantum amplitude $U_f|\alpha\beta\rangle$. It is structurally analogous to a Feynman path integral, with the IFS depth playing the role of time and the truth table playing the role of the action. This is not the universal wavefunction of a physical quantum system. It is a mathematical model with the same algebraic structure — a unitary evolution acting on a linear superposition of branches, where each branch corresponds to a definite classical history. The model captures the form of Everett's universal wavefunction without claiming to be a physical theory.

11.7 Time, the Tent Map, and the Left Shift

The IFS recursion depth — the number of bits evaluated so far — has remained implicit in the discussion above. This subsection makes it explicit: IFS depth functions as a discrete time parameter, conjugate to the Bernoulli shift on the binary expansions of the input truth values. The continuous Hamiltonian evolution between ticks completes this discrete clock into a smooth geodesic in the unitary group.

The tent map. In Łukasiewicz logic on $[-1,1]$ (Convention A), the self-referential equivalence $p \leftrightarrow \neg p$ evaluates to:

This is the tent map — a classic chaotic dynamical system. Under the topological conjugacy $x = \cos\theta$, $z = e^{i\theta}$ (the map from the real interval to the unit circle):

The angle-doubling map $z \to z^2$ on $|z|=1$ is the Bernoulli shift on the binary expansion of $\theta/(2\pi)$. Each application shifts the binary string left by one position, discarding the most significant bit and appending a zero. This is exactly one "tick" of the IFS at a fixed gate: at depth $d$, the IFS evaluates the $d$-th bit pair $(A_d,B_d)$ through the gate's truth table.

The Lyapunov exponent of the tent map $T(x) = 1-2|x|$ is $\lambda = \ln 2$: nearby initial conditions diverge exponentially at a rate of one bit per iteration. By Pesin's theorem, the Kolmogorov–Sinai entropy equals this exponent, $h_{\mathrm{KS}} = \ln 2$, confirming that each tick generates exactly one new bit of Shannon information about the initial truth value $\alpha$ (equivalently, about the angle $\theta$ on the unit circle). This bit is the next digit of the binary expansion of $\alpha$ — the same digit that the IFS reads at depth $d+1$. The entropy production rate of the tent map is therefore the rate at which the IFS resolves the input truth values. The chaotic dynamics is not an incidental feature of the logical fixed-point equation; it is the mechanism that converts a continuous truth degree into a discrete sequence of bits, synchronized one bit per IFS tick.

Discrete time. Each IFS depth increment IS one application of the Bernoulli shift to the binary expansions of $A$ and $B$. The gate $f$ at depth $d$ reads the $d$-th bit pair and outputs the truth value $f(A_d,B_d)$. After $d$ ticks, the accumulated output $z_d$ approximates the McNaughton function value $f(A,B)$ to $d$ bits of precision.

Under the conjugacy $x = \cos\theta$, the iteration $p_{n+1} = 1 - 2|p_n|$ (tent map) is conjugate to $z_{n+1} = z_n^2$ (angle doubling). The IFS recursion is therefore not merely a computational device — it is the discrete-time dynamical system generated by the Bernoulli shift, applied to the logical degrees of freedom $A$ and $B$.

Continuous time from the Hamiltonian. The gate $U_f = H^{\otimes2} D_f H^{\otimes2}$ is the $t=1$ endpoint of a continuous Hamiltonian evolution. Since $U_f$ has eigenvalues $\pm1$ (the same as $D_f$), the Hamiltonian $H_f = i\log U_f$ has eigenvalues $0$ and $\pi$ (mod $2\pi$). The projector onto the $-1$ eigenspace is:

The continuous-time unitary at parameter $t \in [0,1]$ follows immediately:

This is a remarkably simple interpolation: the state at time $t$ is a coherent superposition of the unevolved state $|\alpha\beta\rangle$ and the fully evolved state $U_f|\alpha\beta\rangle$, with trigonometric weights:

For $t=0$ this gives the identity; for $t=1$, the full gate; for intermediate times, a continuous trajectory through the Hilbert space connecting them.

Emergence of the $k$-th root seeds. Conjugating $U_f(t)$ by $H^{\otimes 2}$ gives the diagonal seed at time $t$:

up to global phase $e^{-i\pi t/2}$. At $t = 1/k$ for integer $k \ge 1$, this is precisely the $k$-th root diagonal seed $D_f^{(k)}$ of \S10, encoding the U(1) phase $e^{i\pi f(x)/k}$. The rational-phase diagonal seeds are not a separate constructive extension — they are the intermediate points of the geodesic between identity and the full $\pm1$ gate, revealed by evaluating the continuous-time Hamiltonian flow at rational $t$. Universality (Theorem 1, \S9.4) thus follows from the framework's native time evolution, not from an external phase parameter.

Relating discrete and continuous time. The IFS evaluates bits at integer depths $d = 0, 1, 2, \dots$. The $d$-th evaluation occurs at the continuous time $t_d = 1 - 2^{-d}$ (the fraction of binary expansions resolved). As $d \to \infty$, $t_d \to 1$: the infinite-resolution limit corresponds to the full application of $U_f$. The discrete IFS ticks are a geometric progression of coarse-grained measurements of the underlying Hamiltonian flow.

Formal convergence. Let $U_f^{(d)}$ be the unitary encoding the $d$-bit carry-free evaluation of $f$ (the depth-$d$ approximation). $U_f^{(d)}$ acts on $nd+1$ qubits ($n$ input registers of $d$ bits each, plus a $d$-bit output register). The full gate $U_f = H^{\otimes 2} D_f H^{\otimes 2}$ acts on just 2 qubits at infinite resolution. The convergence $U_f^{(d)} \to U_f$ as $d\to\infty$ holds in the following precise sense: for any truth values $\alpha_1,\ldots,\alpha_n \in [0,1]$ with dyadic expansions $a_{i,1}a_{i,2}\ldots$,

where $\rho_1$ is the reduced density matrix of the first output qubit (the most significant bit of the result) and $\|\cdot\|_1$ is the trace norm. This follows from the uniform convergence of the $d$-bit carry-free expansion $f_d \to f$ (the McNaughton function), the continuity of the Bloch vector map, and the fact that the $d$-bit output register encodes $f_d(\boldsymbol\alpha)$ in its binary expansion. The intermediate points $U_f(t) = e^{-iH_f t}$ define a geodesic in the unitary group between $I$ and $U_f$ under the Hilbert-Schmidt metric, and the discrete approximations $U_f(t_d)$ converge to this geodesic as $d\to\infty$ at a rate $O(2^{-d})$.

Unification: least action, Gray code, fractal bit depth, and partial rotations. The geodesic $U_f(t)$ and the discrete IFS ticks together realize a striking unification of several principles:

• Every $0\to1$ bit transition is an infinite fractal depth. A single Boolean bit flip, when inspected at finer resolution, reveals the full chaos game: the bit $a_d$ of $\alpha$ tells whether $\alpha$ lies in the left or right half of the current dyadic interval. The "instant" transition from $a_d=0$ to $a_d=1$ is, at depth $d+1$, a sequence of sub-decisions tracking the IFS contractions. The geodesic $U_f(t)$ is the least-action interpolant through this infinite regress: each tick $d\to d+1$ advances $t$ by $2^{-(d+1)}$, and the continuous $t$ between ticks is the Hamiltonian flow resolving the next layer of bits.
• Gray code from the quadtree Z-order. The carry-free digitwise evaluation traverses the quadtree column by column: adjacent depth levels differ by exactly one new bit per input. The quadtree's Z-order (Morton order) is a Gray code on the 2D address space — consecutive $(a_1\ldots a_d, b_1\ldots b_d)$ differ in precisely the new bit pair $(a_{d+1},b_{d+1})$. The Gray code is not an external construction; it is the native ordering of the IFS.
• Least action = shortest geodesic in SU(4). The Hamiltonian $H_f = i\log U_f$ defined by the tile's truth table, evolved from $t=0$ to $t=1$, traces the unique geodesic through $U_f$ in the Hilbert-Schmidt metric. Among all curves from $I$ to $U_f$, this is the shortest — the least-action path given the eigenvalues $\pm1$ determined by the Boolean function.
• Partial rotation $t=1/k$ as finite-depth resolution. The $k$-th root seed $D_f^{(k)}$ is the diagonal of the geodesic at $t=1/k$ — equivalently, the gate as seen at $k$-bit resolution. It is not a "fraction" of the gate but the gate evaluated at a specific finite depth. The infinite-depth limit $d\to\infty$ recovers the full $\pm1$ gate at $t=1$.

Together these show that the framework's discrete/continuous duality is not two separate mechanisms but one structure: the IFS depth is the discrete shadow of the Hamiltonian geodesic, and the geodesic is the continuous completion of the IFS recursion. The principle of least action, the Gray-code ordering of the quadtree, the fractal regress of bit resolution, and the partial rotations of \S10 are all aspects of the same underlying geometry.

Feigenbaum scaling as a universal signature. The tent map $T(x) = 1-2|x|$ is the $\mu=1$ endpoint of the parameterized family $T_\mu(x) = \mu(1-2|x|)$ for $\mu\in[0,1]$. This family undergoes a period-doubling cascade to chaos as $\mu$ increases, with the universal Feigenbaum constant $\delta = \lim_{n\to\infty} (\mu_n - \mu_{n-1})/(\mu_{n+1} - \mu_n) \approx 4.669$ governing the rate of convergence of the bifurcation points $\mu_n$. In the present framework, $\mu$ would correspond to a logical parameter — for example, the relative weight of self-reference in the iterated fixed-point equation, or a deformation of the Łukasiewicz implication that interpolates between classical and fuzzy regimes. If such a parameter exists in the logical structure, the period-doubling cascade and its universal scaling ratio $\delta$ would appear as a measurable signature of the underlying chaotic dynamics. Conversely, the observation of Feigenbaum scaling in a quantum gate experiment would constitute strong evidence that the gate's internal dynamics follows the tent-map/Bernoulli-shift structure described here. This connection, while not explored further in the present work, suggests a testable empirical link between the logical framework and universal dynamical systems theory.

Time as logical self-reference. The discrete clock and the continuous geodesic share a common origin: the self-referential logical operation $p \leftrightarrow \neg p$. Iterated, this operation generates the tent map $T(x) = 1-2|x|$, whose symbolic dynamics is the Bernoulli left shift on the binary expansion of $p$. Each IFS tick evaluates one more bit of the input truth values; the tick count $d$ is the number of times the shift has been applied, and the tick rate (one bit per iteration) is the Lyapunov exponent $\lambda = \ln 2$ of the tent map.

The continuous geodesic $U_f(t) = \exp(-iH_f t)$ completes this discrete structure: the depth-$d$ approximation coincides with the geodesic at time $t_d = 1 - 2^{-d}$, and the flow between $t_d$ and $t_{d+1}$ resolves the $(d+1)$-th bit pair through coherent Hamiltonian evolution. The geodesic is the continuous interpolation of the discrete IFS ticks, and the discrete ticks are coarse-grained measurements of the geodesic at dyadic times. No external clock is required — the tent map, generated by the logic itself, provides the ticking mechanism, and the Hamiltonian provides the smooth evolution between ticks.

In this sense, time in the model is the self-referential operation $P \leftrightarrow \neg P$ iterated: each tick evaluates one more bit of the input against its own negation, and the continuous interpolation between ticks is Hamiltonian evolution on the 2-qubit Hilbert space.

11.8 The Born Rule from the IFS Branching Measure Theorem

The universal wavefunction integral of \S11.4 weights each branch $\mathbf{x} \in \{0,1\}^{nd}$ by the amplitude $\sqrt{\mu_d(\mathbf{x})} = 2^{-nd/2}$, where $\mu_d$ is the uniform product measure on $d$-bit prefixes of $n$ input sequences. This section proves that for computational-basis measurements, the Born probability equals the IFS measure, and extends this to all bases reachable by the framework's domain-switching operations.

Definition 11.1 (Framework measurement). A framework measurement of a compound sentence $f$ over the $n$ input propositions is the projection-valued measure $\{P_f, I-P_f\}$ where $P_f$ projects onto the subspace of branches where $f$ evaluates to True at infinite depth. In the computational basis (Z-basis), $P_f$ is diagonal:

where $\mathbf{1}_{f(\mathbf{x})=\text{True}}$ is the indicator that the carry-free evaluation limit $f(\alpha_1,\ldots,\alpha_n)$ equals True, and the sum is over the product Cantor space of infinite binary sequences.

Theorem 11.1 (Born rule = IFS measure, Z-basis). Let $|\psi_d\rangle$ be the $n$-qubit IFS state at depth $d$ (Definition 12.1). For any compound sentence $f$ over the 16 connectives, the probability of observing $f=\text{True}$ in a computational-basis measurement of $|\psi_d\rangle$ is:

where $f_d$ is the $d$-bit digitwise evaluation of $f$. In the limit $d \to \infty$, this converges to:

the $\mu$-measure of the set of continuous truth-value assignments $\boldsymbol\alpha$ for which the McNaughton extension of $f$ equals True.

Corollary 11.2 (Born rule = IFS measure, any basis via domain switching). Let $H_S$ be a selective Hadamard on a subset $S$ of qubits (\S13.7). Then for the domain-switched state $|\psi_d^S\rangle = H_S|\psi_d\rangle$, the Born probability for a computational-basis measurement equals the IFS measure on the $X$-domain branches:

where $H_S\mathbf{x}$ denotes the bitwise action of $H_S$ on the $X$-domain bits (the Hadamard transform on the target-qubit subspace). For $S = \emptyset$ this recovers Theorem 11.1. For $S$ non-empty, the IFS measure is evaluated on the $X$-basis branch structure, which is the image of the Z-basis Cantor space under the local Hadamard maps.

Remark on unit-modulus diagonal seeds. Theorem 11.1 and Corollary 11.2 assume the $\pm1$ diagonal $D_f$ whose entries have magnitude 1. The Z-basis proof uses only $|2^{-nd/2}|^2$, not the sign. If $D_f$ is replaced by a U(1)-phased seed $D_f^{(\phi)} = \operatorname{diag}(e^{i\phi\cdot f(00)}, e^{i\phi\cdot f(01)}, e^{i\phi\cdot f(10)}, e^{i\phi\cdot f(11)})$ with $\phi\in\mathbb{R}$ (\S10), the state becomes $|\psi_d^{(\phi)}\rangle = 2^{-nd/2}\sum_{\mathbf{x}} e^{i\phi\cdot f_d(\mathbf{x})}|\mathbf{x}\rangle$. Since $|e^{i\phi\cdot f_d(\mathbf{x})}|^2=1$, the Z-basis Born probability is unchanged:

Corollary 11.3 (Born rule = IFS measure, Z-basis with U(1)-phased seeds). Theorem 11.1 holds for any diagonal seed whose entries have unit modulus. In particular, for the $k$-th root seeds $D_f^{(k)}$ of \S10 (where $\phi=\pi/k$), the Z-basis Born probability equals the IFS branching measure $\mu$ for all integers $k\ge1$:

X-basis with U(1)-phased seeds: product measure structure. For X-basis measurements (or mixed Z/X domains), the U(1) phases do not cancel because the Hadamard transform mixes branches coherently. For a U(1)-phased seed at depth $d$ under full Hadamard $H^{\otimes 2d}$, the Born probability for outcome $\mathbf{z}\in\{0,1\}^{2d}$ factorizes over depth levels:

where $S_{\phi}$ is the phased Walsh-Hadamard coefficient of the tile $f$ at phase $\phi$. For $\phi=\pi$ (the $\pm1$ case, $k=1$), $|S_{\pi}|^2=4$ for all $(z_A,z_B)$, recovering the uniform IFS measure. For $\phi=\pi/k$ with $k>1$, $|S_{\phi}|^2$ is non-uniform: the per-level probability differs across the four outcomes, giving a product measure distinct from the IFS measure. The table below gives the per-level distributions for AND:

The product structure follows from the independence of the bit-evaluation across depth levels (the carry-free arithmetic of \S12.2): each depth level $i$ evaluates $f$ on fresh independent bits $(a_i,b_i)$, and the Hadamard transform on the $2d$-qubit register factorizes accordingly.

Conjecture (X-basis U(1) Born rule). For U(1)-phased diagonal seeds with $\phi=\pi/k$ ($k\ge2$), the X-basis Born probability $p_d^{(\phi)}$ equals the IFS measure $\mu_d$ if and only if $k=1$. For $k\ge2$, $p_d^{(\phi)}$ is a product measure whose per-level marginals are the squared phased Walsh coefficients $|S_{\phi}|^2$; it does not reduce to $\mu_d$ unless the tile satisfies $|S_{\phi}(z_A,z_B)|^2=4$ for all $(z_A,z_B)$, which holds only when $e^{i\phi\cdot f(a,b)}$ has equal $\pm1$ values ($\phi=\pi$). Whether $p_d^{(\phi)}$ admits an interpretation as the IFS measure on a different (non-Cantor) branch space — i.e., whether the branching decomposition can be reparametrized to absorb the phase-induced non-uniformity — remains open and is equivalent to finding a measure-preserving isomorphism between the product measure $p_d^{(\phi)}$ and the uniform measure $\mu_d$ on the $X$-domain Cantor set.

Theorem 11.3 (Delineation: proved vs open). The status of the Born rule within the 16-connective framework is as follows:

Significance. Theorem 11.1, Corollary 11.2, and Corollary 11.3 close the Z-basis and X-basis Born rule question for both $\pm1$ and U(1)-phased diagonal seeds within the framework. The IFS branching measure $\mu$ is the Born measure for all Z-basis measurements regardless of the phase structure of the diagonal seed. For X-basis measurements with U(1)-phased seeds, the Born probability is a product measure whose per-level marginals are the squared phased Walsh coefficients; it reduces to the uniform IFS measure only for $\pm1$ phases. The remaining open cases are: (i) whether the X-basis U(1) product measure admits an IFS-measure interpretation on a reparametrized branch space, and (ii) the full continuum limit $n,d\to\infty$ (Gap 3).

Relation to the many-worlds interpretation. In the Everett picture, the Born rule is an additional postulate (or derived from decision-theoretic arguments) that links the branching measure to the probability of self-location in a branch. Within this framework, the branching measure $\mu$ is directly observable as the Born probability for Z-basis measurements (Theorem 11.1): the measure of a set of branches is the probability of observing that outcome. This is not an additional postulate — it follows from the definition of the quantum state $|\psi_d\rangle$ as a $\sqrt{\mu}$-weighted superposition of branches. The many-worlds "probability problem" (why should the measure of branches correspond to probability?) is resolved within the model by construction: the squared amplitude of each branch is its IFS measure. There is no additional conceptual gap; the Born rule is the definition of the IFS measure, not a separate axiom.

12. The $n$-Qubit IFS Depth Process Correspondence

This section generalises the two-variable IFS to an arbitrary number of qubits. Each qubit carries a continuous truth value $\alpha_i \in [0,1]$ revealed bit by bit through the IFS depth. The gate network (a compound sentence over the 16 tiles) couples the qubits' bit-streams at every depth level. The resulting quantum state is a superposition over all $2^{nd}$ possible $d$-bit extensions of the $n$ inputs — a genuinely multipartite entangled state whose structure mirrors the logical circuit.

The single-qubit chaotic liar ($P\leftrightarrow\neg P$, \S6, \S11.7) is the $n=1$ base case. The two-variable gate table (\S7) is the $n=2$ case. The following constructs the general $n$-qubit state, defines the unitary that evolves it through depth, and analyses the entanglement growth that encodes the logical complexity of the sentence.

12.1 The $n$-Qubit IFS State at Depth $d$

Let $\alpha_1,\ldots,\alpha_n \in [0,1]$ be independent continuous truth values. Each has a dyadic expansion:

At IFS depth $d$, we truncate to $d$ bits. We define the digitwise evaluation of a Boolean function $f$ as:

Note on Logical Bifurcation: It is critical to distinguish this digitwise evaluation from the continuous McNaughton extension (standard Łukasiewicz logic). While they agree on the Boolean vertices, they diverge in the interior (e.g., digitwise AND(0.75, 0.75) = 0.75, while Łukasiewicz AND(0.75, 0.75) = 0.5). The IFS and fractal topographies described in this paper are generated by the digitwise logic, which inherently contains dyadic discontinuities, whereas the near-MV algebra describes the continuous limit of the 2-variable atomic gates.

Definition 12.1 ($n$-qubit IFS state). The quantum state at depth $d$ is a superposition over all $2^{nd}$ possible $d$-bit input prefixes:

Convergence as $d\to\infty$ is defined via the weak-* topology on the space of probability measures over the Cantor set $\{0,1\}^\mathbb{N}$. Each $|\psi_d\rangle$ induces a discrete measure that converges to the branching measure $\mu$ of the IFS attractor.

Remarks. For $n=2$, this recovers the two-variable IFS attractor of \S1–\S2: the inputs $(A,B)$ at all dyadic resolutions. For $n=1$, it gives the single-variable IFS that generates the tent map attractor of the chaotic liar (\S6). The uniform weighting $2^{-nd/2}$ corresponds to the dyadic (Lebesgue) measure on $[0,1]^n$.

12.2 The Depth-Evolution Unitary

The transition from depth $d$ to $d+1$ reveals one new bit of each $\alpha_i$ and evaluates $f$ on the new bits, appending the result to the output register. This is a unitary operation on $(n+1)(d+1)$ qubits.

Definition 12.2 (Depth-evolution unitary). For each depth $d \ge 0$, define $V_d$ acting on $(n+1)d + (n+1)$ qubits:

where $|0^{n+1}\rangle$ are $n+1$ fresh qubits (one per input variable plus one output). The action of $V_d$ is:

1. Input expansion. For each $i=1,\ldots,n$, append a qubit initialized to $|0\rangle$. This qubit will carry $a_{i,d+1}$, the $(d+1)$-th bit of $\alpha_i$. The input register grows from $nd$ to $n(d+1)$ qubits.
2. Output expansion. Append an output qubit $|0\rangle$ for the $(d+1)$-th output bit.
3. Gate application. Apply $f$ to the $(d+1)$-th bits of all $n$ inputs, controlled by the existing superposition, and write the result into the new output qubit.

Claim 12.3. $V_d$ can be implemented by a quantum circuit of depth $O(\operatorname{size}(f))$ where $\operatorname{size}(f)$ is the number of 2-qubit gates in the circuit representation of $f$. The total circuit depth to reach IFS resolution $d$ is therefore $O(d\cdot\operatorname{size}(f))$.

Proof sketch. Each $V_d$ evaluates $f$ on $n$ classical bits (the $(d+1)$-th bits of each input). Since $f$ is a Boolean function over the 16-tile alphabet, it decomposes into $O(\operatorname{size}(f))$ Toffoli-like gates, each of which can be realized as a constant-depth circuit of 2-qubit gates. The control structure of the existing superposition does not affect the circuit depth because the new bits are in the computational basis conditioned on the old ones — the evaluation is diagonal in the computational basis. Hence each $V_d$ has effective depth $O(\operatorname{size}(f))$. □

In the $n=2$ case with a single tile (no compound circuit), $\operatorname{size}(f)=1$ and $V_d$ reduces to reading one bit from each input address and looking up the tile's truth table — exactly the IFS recursion step of \S1.

12.3 Entanglement Growth with Depth

The state $|\psi_d\rangle$ is a superposition over all $2^{nd}$ possible $d$-bit input histories. The entanglement between any two qubit registers reflects their correlational structure through $f$. We analyse three archetypal cases.

Case 1: $n$-variable parity (XOR$_n$). Let $f(x_1,\ldots,x_n) = x_1 \oplus \cdots \oplus x_n$ be the $n$-bit parity function. By Theorem 7.1, this maps to $X^{\otimes n}$ under $H^{\otimes n}$ conjugation. At depth $d$:

This is the generalized GHZ state of $(n+1)d$ qubits encoding the parity relation at all bit-scales. The reduced density matrix of any $k$ qubits is maximally mixed for $k < (n+1)d$. The von Neumann entropy of any proper subset of registers scales as:

For $n=2$, this gives a Bell-type correlation at each of the $d$ depth levels — but the correlation is replicated across all $d$ levels simultaneously, producing a $3d$-qubit GHZ state whose entanglement grows linearly with $d$.

Case 2: AND (Łukasiewicz strong conjunction). Let $f(\alpha,\beta) = \max(0, \alpha+\beta-1)$ be the carry-free AND. At each depth level $k$, the output bit is $b_k = a_k \land b_k$ (Boolean AND of the $k$-th bits). The $d$-bit output is:

AND has a non-uniform spectral support (its Walsh-Hadamard transform has three non-zero coefficients: $\hat f(\emptyset)=3/4$, $\hat f(\{1\})=-1/4$, $\hat f(\{2\})=-1/4$, $\hat f(\{1,2\})=1/4$). The resulting state:

has a richer entanglement structure than XOR: the output register is correlated with the inputs only at bit-positions where both inputs have a 1. The mutual information between the two input registers, conditioned on the output, is:

where $H(p) = -p\log_2 p - (1-p)\log_2(1-p)$ is the binary entropy. This is $d\cdot H(1/4) \approx 0.811d$ — linear growth with a smaller coefficient than XOR's $1.0d$.

Case 3: Compound circuits. For a circuit $f = g(h_1(\cdots), h_2(\cdots))$, the entanglement structure at depth $d$ reflects the circuit topology. Variables that share a gate at any depth level become correlated at that level; variables that never interact (disjoint subcircuits) remain in a product state across all depths. The mutual information between any two input registers is:

where $C_{ij} = \sum_{S\ni i,j} \hat f(S)^2 / \sum_{S\neq\emptyset} \hat f(S)^2$ is the pairwise correlation coefficient of $f$ — the fraction of the Walsh-Hadamard spectral weight supported on subsets containing both $i$ and $j$. This quantifies how directly the two variables interact through $f$.

12.4 The Chaotic Liar as $n=1$ Base Case

The single-variable case $n=1$ with $f(\alpha) = \alpha \leftrightarrow \neg\alpha$ (the Chaotic Liar) is the simplest instance of the depth process. Under Convention A, this is the tent map (Theorem 6.1):

At depth $d$, the state is:

where $\operatorname{tent}_d(x)$ is the $d$-bit approximation of the tent map evaluated at $\alpha = \sum x_k 2^{-k}$. This is a $(2d)$-qubit state correlating the input register with the self-referential output register.

In the limit $d\to\infty$, the iterated self-referential XOR converges to the Bernoulli shift (angle doubling, $z\to z^2$, Theorem 6.2). The depth process constructs this map not as a single operator but as the limit of the finite-depth states:

This limit exists on the measure space of the Cantor set $\{0,1\}^\mathbb{N}$. Each depth level $d$ provides a cylinder-set approximation of the infinite chaotic orbit.

Relation to the multi-qubit case. For $n>1$, each qubit's self-referential dynamics (if present in the circuit) couples with the others through the gate network. The $n=1$ case provides the fundamental "ticking" mechanism (\S11.7): the single-qubit liar generates the time parameter, and the multi-qubit network distributes that time-evolution across the entangled registers.

Answer to Open Question 4. There is no single complex-linear operator on $\mathbb{D}$ whose iteration gives $z\to z^2$ on the boundary, because $z\to z^2$ is quadratic and no non-zero linear map $L(z)=\alpha z$ squares the angle while preserving $|z|=1$. The $z^2$ map is the infinite-depth limit of the $n=1$ IFS depth process described above: each depth $d$ resolves one more bit of the tent map, and the Cesàro limit of the iterated self-referential XOR converges to the Bernoulli shift $\theta\mapsto 2\theta\pmod{2\pi}$. The depth process is the single operator that generates $z\to z^2$ — it just requires infinite resolution.

12.5 The Paradox-Complexity Spectrum

The entanglement growth rate $C_{ij}$ quantifies the pairwise correlation between input variables $i$ and $j$ under the sentence $f$, as measured by their joint Walsh-Hadamard spectral weight. The table below classifies the 16 tiles by their entanglement scaling behaviour and fixed-point structure:

Key observations:

• A sentence has a Boolean fixed point iff the corresponding $n$-variable function $f$ has an $x\in\{0,1\}^n$ such that $f(x)=1$ under Convention A. This is the classical consistency condition. However, entanglement growth does not require paradox: XOR has Boolean fixed points ($00$ and $11$ both give output parity $0$) yet produces maximal entanglement growth ($C_{AB}=1$).
• What the fixed-point analysis misses is the dynamic correlation at finite depth. Even a function with consistent Boolean assignments entangles its inputs at intermediate bit-scales. The "paradox" is not about the limit but about the rate and structure of entanglement growth as depth increases.
• Paradoxical sentences (no Boolean fixed point) can be identified with functions whose Walsh-Hadamard spectrum has no weight on the empty subset ($\hat f(\emptyset)\neq \pm1$). This condition is equivalent to the sentence being unsatisfiable as a Boolean equation. The remaining spectral weight determines the entanglement scaling coefficients $C_{ij}$.

12.6 Many-Worlds Interpretation of the Depth Process

Natural interpretation. The many-worlds framing of \S11 extends naturally to the $n$-qubit depth process:

• Each branch at depth $d$ corresponds to a $d$-bit prefix of all $n$ truth values — a partial history of the $n$ propositions. There are $2^{nd}$ branches, each weighted by the uniform measure $2^{-nd}$.
• Entanglement between qubit registers is inter-world correlation of their bit-streams. Two propositions $i$ and $j$ are entangled at depth $d$ to the extent that their bits at the same level are correlated by the sentence $f$ across all branches simultaneously.
• The depth iteration corresponds to the Everett branching process. Each tick $(d\to d+1)$ reveals one new bit per proposition and correlates the new bits through $f$. This is structurally identical to the universal wavefunction integral of \S11.4: the branching measure at depth $d$ is the $d$-th partial sum in the Feynman-style path integral over all $n$ input histories.
• Time = branching resolution. Following \S11.7, the discrete time $t_d = 1-2^{-d}$ marks the fraction of the infinite bit-stream resolved. The continuous Hamiltonian $U_f(t) = \cos(\pi t/2)I - i\sin(\pi t/2)U_f$ interpolates between the discrete ticks, giving a smooth entanglement growth curve $S(t)$.
• The paradoxical sentence (one with no Boolean fixed point) corresponds in the many-worlds picture to a sentence that cannot be satisfied by any single classical world. Its resolution requires the superposition — the branching structure itself is the resolution of the paradox. The more entangled the state, the more branches are coherently coupled.

In this light, the $n$-qubit IFS depth process is not merely a mathematical generalization of the two-variable chaos game. It is a physical model of how logical structure generates quantum entanglement through branching dynamics: the 16 tiles are the fundamental logical operations, carry-free arithmetic is the mechanism of bitwise coupling, the IFS depth is the time parameter, and the resulting entanglement spectrum is the logical complexity of the multi-proposition sentence expressed as a quantum state.

Physical limits on depth: the role of decoherence

The linear entanglement scaling $S(d) = d\cdot C_{ij}$ (\S12.3) assumes perfect coherence across all $d$ depth levels. In a physical implementation, decoherence bounds the maximum useful depth. Each depth level adds $O(\operatorname{size}(f))$ 2-qubit gates to the circuit (Claim 12.3). For a physical qubit with coherence time $T_2$ and gate time $\tau_g$, the number of gates that can be applied before coherence is lost is $T_2/\tau_g$, giving a maximum depth:

For superconducting qubits ($T_2 \sim 100\,\mu\text{s}$, $\tau_g \sim 50\,\text{ns}$) and a modest circuit $\operatorname{size}(f) \sim 10$, $d_{\max} \sim 200$ depth levels — enough to resolve the first ${\sim}60$ bits of each input after accounting for circuit overhead. This is not a limitation of the architecture but a universal constraint of quantum computation: the entanglement growth predicted by Theorem 12.4 applies to the coherent regime $d < d_{\max}$, after which the density matrix becomes increasingly mixed and the mutual information saturates at a value determined by the steady-state of the Lindblad evolution.

The important feature of the IFS depth process is that decoherence does not introduce new entanglement structure — it simply limits the resolution at which the logical entanglement spectrum can be resolved. This is consistent with the physical picture of \S13: the Ising couplings that generate the gates are fixed parameters of the Hamiltonian, not functions of depth. The depth process is a measurement resolution, not a dynamical parameter.

13. Physical Hamiltonians for the 16 Tiles Correspondence

The abstract interpolation $U_f(t) = \cos(\pi t/2)I - i\sin(\pi t/2)U_f$ of \S11.7 is mathematically valid — it uses $M_f = (I-U_f)/2$ as the generator — but it is not derived from any natural physical interaction. This section fills that gap: every gate $U_f$ is generated by a two-spin Ising Hamiltonian with $ZZ$ coupling and local $Z$ fields. The 16 tiles correspond to 16 distinct points in the Ising parameter space, all with couplings that are integer multiples of $\pi/4$.

13.1 The Ising Hamiltonian in the $Z$-Basis

Consider the Hamiltonian of two spin-$\frac12$ particles coupled by an Ising interaction in a transverse field:

where $Z$ is the Pauli-$Z$ operator, $J$ is the $ZZ$ coupling strength, and $h_i$ are local fields on each spin. In the computational basis $\{|00\rangle,|01\rangle,|10\rangle,|11\rangle\}$, this Hamiltonian is diagonal:

Claim 13.1. For every tile $f$ (every 2-variable Boolean function), the gate $U_f = H^{\otimes 2} D_f H^{\otimes 2}$ is equal to $e^{-i H_f}$ up to global phase, where $H_f$ is the Ising Hamiltonian with parameters $(J, h_1, h_2)$ given by:

In the $X$-basis (the gate basis of \S7), this Hamiltonian decomposes as:

where $\hat f(S)$ are the Walsh-Hadamard coefficients and the only non-zero Pauli terms are $I\otimes I$, $X\otimes I$, $I\otimes X$, $X\otimes X$ — no $Y$ or $Z$ terms appear. Conjugating by $H^{\otimes 2}$ gives the $Z$-basis Ising form.

Proof. The $X$-basis form of $U_f$ is the linear combination $\sum \hat f(S) X_S$ (from \S7). The Hamiltonian $H_f = (\pi/2)(I - U_f)$ inherits this Pauli support. Since $H X H = Z$, conjugating by $H^{\otimes 2}$ maps $X\otimes I \to Z\otimes I$, $I\otimes X \to I\otimes Z$, $X\otimes X \to Z\otimes Z$, yielding the diagonal Ising Hamiltonian. □

13.2 Parameter Map: Tiles to Ising Couplings

Computing the coefficients for all 16 tiles gives the following clean grid of parameters. All values are integer multiples of $\pi/4$:

Key structure. The 8 monomial gates (FALSE, TRUE, P, $\neg A$, Q, $\neg B$, XOR, XNOR) have exactly one non-zero Ising parameter — they are pure single-qubit flips or pure $ZZ$ coupling. The 8 entangling gates have all three parameters non-zero with magnitude $\pi/4$. The 8 entangling gates naturally pair into complement pairs (AND$\leftrightarrow$NAND, OR$\leftrightarrow$NOR, etc.) related by sign flips of all three parameters, which corresponds to $H \leftrightarrow -H$ and hence $U \leftrightarrow U^\dagger$.

13.3 Time Evolution and Gate Generation

The Ising Hamiltonian $H_f$ generates a one-parameter family of unitaries:

with $U_f(0) = I$ (identity) and $U_f(1) = e^{-i H_f} = U_f$ (the tile's gate, up to global phase). Since $H_f$ is diagonal in the computational ($Z$-) basis, the time-evolution operator at time $t$ is also diagonal:

At $t=1$, the diagonal entries take the values $\pm1$ (for the monomials) or $\pm1, \pm i$ (for the entangling gates), recovering the diagonal $D_f$ up to a global phase.

Claim 13.2 (Global phase). The Ising evolution $e^{-i H_f}$ equals $U_f$ up to a global phase $\pm i$ or $\pm 1$:

This follows because $H_f = (\pi/2)(I - U_f)$, so $e^{-i H_f} = e^{-i\pi/2}\, e^{i(\pi/2)U_f}$. Since $U_f^2 = I$, we have $e^{i(\pi/2)U_f} = \cos(\pi/2)I + i\sin(\pi/2)U_f = i U_f$. Hence $e^{-i H_f} = e^{-i\pi/2}\cdot i U_f = U_f$ (no global phase). This holds for any $U_f$ with $U_f^2 = I$.

Claim 13.3 (Continuous trajectory). The continuous trajectory $U_f(t)$ interpolates smoothly between identity at $t=0$ and $U_f$ at $t=1$:

This is exactly the interpolation of \S11.7. The difference is that we now recognise it as the time evolution of a physical Ising Hamiltonian rather than an ad-hoc algebraic construction.

13.4 Physical Derivation: Carry-Free Arithmetic Is Ising Dynamics

The connection between the abstract logical system and the Ising model is structural, not coincidental. Three key facts link the two:

1. Carry-free bit processing $\leftrightarrow$ commuting $X$-basis terms. The carry-free evaluation of \S8.2 processes each binary position independently because the output bit at depth $k$ depends only on the $k$-th bits of the inputs. In the quantum picture, the $X$-basis operators $X\otimes I$, $I\otimes X$, $X\otimes X$ all commute with each other. Their simultaneous eigenbasis is the $X$-basis computational basis, where each qubit's $X$-eigenvalue encodes one bit. The Hamiltonian $H_f$ is a linear combination of these commuting terms, so its time evolution processes all bit-positions in parallel — exactly the carry-free structure.

2. Walsh spectrum $\leftrightarrow$ Ising couplings. The Walsh-Hadamard coefficients $\hat f(S)$ of the Boolean function $f$ determine the Ising parameters directly:

The 8 monomials are precisely those functions whose Walsh spectrum has support on a single subset (one non-zero coefficient). The 8 entangling gates have support on all three non-empty subsets. The magnitude $\pi/4$ for entangling gates reflects the fact that $|\hat f(S)| = 1/2$ for these functions (the remaining $1/2$ weight is on the empty subset).

3. $H^{\otimes 2}$ conjugation $\leftrightarrow$ basis change between logical and physical bases. The $X$-basis (where the gates act by tensor products and the Hamiltonian is a transverse XX model) is the logical basis — the natural representation for the Boolean function. The $Z$-basis (where the Hamiltonian is the Ising model) is the physical basis — the natural representation for local spin interactions. The Hadamard transform $H^{\otimes 2}$ bridges them:

13.5 Relationship to the $n$-Qubit Depth Process

The Ising Hamiltonian for a 2-qubit tile generalises to the $n$-qubit circuits of \S12. For a compound sentence $f$ over $n$ variables, the Hamiltonian in the $X$-basis is:

where $\hat f(S)$ are the $n$-variable Walsh-Hadamard coefficients (Theorem 7.1 generalised). In the $Z$-basis, this becomes a multi-spin Ising model:

with couplings $\tilde h_S = (\pi/2)\,\hat f(S)$. For the $n$-variable parity function (XOR$_n$), $\hat f(\{1,\ldots,n\}) = 1$ and all other coefficients vanish, giving $H = (\pi/2)\, Z^{\otimes n}$ — an $n$-body Ising interaction. For compound circuits, the coupling coefficients mirror the circuit topology exactly as the entanglement coefficients $C_{ij}$ do in Theorem 12.4.

Physical interpretation. The $n$-qubit circuit represents $n$ spins in a multi-local magnetic field. The Ising couplings are set by the Boolean function's Walsh spectrum. Time evolution under this Hamiltonian generates the quantum gate that encodes the logical sentence. The carry-free arithmetic of the IFS is the physical statement that all coupling terms commute — the $Z$-basis operators $Z^{\otimes |S|}$ for different subsets $S$ commute among themselves, so the dynamics factorises into independent bit-resolutions at each IFS depth level.

13.6 Walsh Coefficients as Frequencies

The Walsh-Hadamard transform is the Fourier transform on the Boolean cube $\{0,1\}^n$. Each Walsh coefficient $\hat f(S)$ measures the amplitude of a particular frequency in the Boolean function — the rate at which the function's output flips when the bits in subset $S$ are flipped. Four frequencies exist for $n=2$:

Every Boolean function decomposes into a sum of these four frequency components, just as a musical chord decomposes into a sum of pure tones. The Ising Hamiltonian is the physical realization of this decomposition: each frequency component $\hat f(S)$ becomes a coupling term in the Hamiltonian with strength proportional to that frequency's amplitude.

An analogy with sound. A piano chord (a Boolean truth table) can be analysed into its constituent notes (the Walsh frequencies). Each note (Walsh coefficient) corresponds to a particular pattern of vibration. In the Ising model, each such vibration pattern is implemented as a physical interaction between the two spins — the coupling $J$ vibrates at frequency $\hat f(\{1,2\})$, and the local fields $h_1, h_2$ vibrate at $\hat f(\{1\})$ and $\hat f(\{2\})$ respectively. The continuous time evolution $U_f(t) = e^{-iH_f t}$ is the sound of these vibrations playing together.

The 8/8 split as a spectral classification. The 8 monomial gates have exactly one non-zero frequency component (a pure tone). The 8 entangling gates have all three non-zero (a chord). The entangling gates are "harmonically richer" in a precise sense: they require all three Ising couplings simultaneously, which is why they produce entanglement.

Surprising case: XOR. XOR has zero weight on both single-bit frequencies — its only non-zero Walsh coefficient is the joint AB frequency. In the Ising model, this gives $J = \pm\pi/2$, $h_1 = h_2 = 0$: pure coupling with no local fields. The gate is purely relational: it encodes no information about either input individually, only about their parity. Yet this same function, when iterated self-referentially as P↔¬P, generates the tent map and the Bernoulli shift — the system's maximal chaos (Theorem 6.2, \S11.7). The pure relational structure (zero local fields, full coupling) produces both the simplest Ising Hamiltonian and the richest dynamical behavior, suggesting that chaos and entanglement share the same algebraic origin: dependence on correlation alone, without local bias.

This frequency interpretation is not philosophical. It is exact: the Walsh transform is the Fourier transform on the Boolean cube, and the Ising parameters are the Fourier coefficients of the truth table, scaled by $\pi/2$.

Relation to the fermion Walsh system. The operators $\{I,X\}^{\otimes n}$ appearing in Claim 13.1 are the $n$-qubit truncation of the fermion (non-commutative) Walsh system141516, a non-commutative orthogonal basis for the hyperfinite type II$_1$ factor $\mathcal{R}$. Each Rademacher operator acts on a single fermion coordinate as a Pauli matrix; the 16 tiles are the $n=2$ truncation of this infinite system. The carry-free IFS depth process in the limit $d\to\infty$ converges to the infinite fermion chain whose observable algebra is $\mathcal{R}$, linking the finite-depth logical states of \S12 to the non-commutative $L^p$ spaces studied in the fermion Walsh literature. In this light, the Walsh-Hadamard decomposition $U_f = \sum_S \hat f(S) W_S$ expresses each Boolean gate as a linear combination of fermion Walsh operators — the same algebraic structure that underlies the analysis of the hyperfinite II$_1$ factor.

Three specific results from this literature bear directly on the present framework:

• Lust-Piquard (1998)14 proves dimension-free Riesz transform bounds for the Walsh system and the fermions. On the fermion side, the number operator $N = \sum_{j=1}^n D_j^* D_j$ (where $D_j$ are annihilation operators) acts on the Walsh operator $X_S$ as multiplication by $|S|$ — the Hamming weight of the index subset. The equivalence $\|N^{1/2}(T)\|_{C_p} \sim \|(\sum_{j=1}^n |D_j(T)|^2_s)^{1/2}\|_{C_p}$ for $2<p<\infty$ relates the $p$-norm of the number-weighted operator to the square function of its fermion partial derivatives, with constants independent of $n$. This is the non-commutative analogue of the classical Riesz inequality $\|\nabla f\|_p \sim \|\Delta^{1/2} f\|_p$ on the Boolean cube — the same cube $\{0,1\}^n$ that parametrizes the computational basis states of the $n$-qubit gate.
• Wu (2016)15 studies multiplier transformations $J_\alpha x = \sum_{\gamma\in\mathcal{F}} 2^{-|\gamma|\alpha}\,\hat x(\gamma)\, w_\gamma$ for the noncommutative Walsh system $(w_\gamma)_{\gamma\in\mathcal{F}}$ in $\mathcal{R}$, and proves $J_\alpha$ is bounded $L_p(\mathcal{R}) \to L_q(\mathcal{R})$ for $1<p<q$ with $\alpha = 1/p-1/q$. The multiplier $2^{-|\gamma|\alpha}$ is a dyadic low-pass filter by Hamming weight $|\gamma|$. This is the exact operator-algebraic analogue of the IFS depth truncation in \S12: truncating the dyadic expansion at depth $d$ suppresses all Walsh modes with $|\gamma|/n > d$, and the convergence $d\to\infty$ mirrors the $L_p\to L_q$ boundedness of the limiting multiplier.
• Ferleger & Sukochev (1995)16 show that generalized Walsh systems in UMD spaces — including the fermion Walsh system in non-commutative $L^p$ spaces — form Schauder bases and uniform finite-dimensional decompositions. The Walsh-Hadamard expansion $U_f = \sum_S \hat f(S) X_S$ in §7.1 is precisely an expansion in such a Schauder basis: it converges unconditionally in the Schatten $p$-norm for $1<p<\infty$, guaranteeing that every 2-qubit gate in the $X$-basis Pauli algebra has a unique and convergent Walsh decomposition.

Can the 16 tiles reproduce any quantum gate? The 16 tiles are the tuning forks — the pure Walsh frequencies for 2-variable Boolean functions. By varying the time parameter $t$ continuously in $U_f(t) = e^{-i H_f t}$, the amplitude of each frequency can be dialed to any value (not just the discrete $\pi/4$ multiples of the 16 tiles). This gives all gates of the form $\exp(-i(a\,ZZ + b\,ZI + c\,IZ))$ for arbitrary real $a,b,c$ — the complete family of commuting Ising gates. Combined with selective Hadamard conjugation (\S9.3), which provides non-commuting gates like CNOT, and rational-phase diagonals (\S9.3), which provide non-Clifford gates like $T$, the 16 tiles generate the full Clifford hierarchy and enable universal quantum computation (Theorem 1, \S9.4). You cannot play every possible 2-qubit gate with the 16 tiles alone — but with the tuning forks (Walsh frequencies), a volume knob (continuous time $t$), and a way to change instruments (selective Hadamard), you can approximate any quantum gate to arbitrary precision.

13.7 Selective Hadamard as Domain Switching Correspondence

Selective Hadamard conjugation (\S9.3) is presented as an external choice: apply $H$ to a chosen subset of qubits. But the choice is not arbitrary — it is determined by the logical domain of each variable in the compound sentence. The three-domain embedding (\S9.3, Table 9.1) assigns each input variable a role:

• Z-domain (control, classical): variables that enter the sentence through functions whose Walsh spectrum has weight on single-bit terms (AND-like, OR-like). These variables act as "controls" — they determine whether the gate fires, encoded as $\pm1$ phases in the computational basis.
• X-domain (target, quantum): variables that enter through functions with weight on joint terms (XOR-like). These variables act as "targets" — they participate in the interference dynamics, encoded as $\pm1$ eigenvalues of the $X$ operator.
• M-domain (mixed): variables that connect both types of functions across the circuit, transitioning between control and target roles at different depths.

The selective Hadamard is not an external operation grafted onto the system. It is the bridge between these domains, and which bridge connects which variable at which depth is determined by the compound sentence's circuit topology:

The subset $S$ selects the X-domain (target) variables; its complement selects the Z-domain (control) variables. The algorithm for determining $S$ from the sentence structure is:

1. Parse the sentence into a circuit over the 16-tile alphabet (gate nodes connected by variable wires).
2. For each variable $x_i$, trace its path through the circuit. If $x_i$ enters any gate through a function whose Walsh spectrum has weight on $\{i\}$ (single-bit dependence), mark $x_i$ as a control (Z-domain, $i\notin S$). If $x_i$ enters only through functions whose Walsh spectrum is supported on $\{i,j,\ldots\}$ with $|S|\ge 2$ (joint dependence only), mark $x_i$ as a target (X-domain, $i\in S$).
3. The output variable (the sentence's truth value) is always in the Z-domain — it is the classical readout.

Example: AND as CNOT. For $f = \text{AND}(A,B)$ with $S = \{2\}$ (target qubit B), the seed $D_{\text{AND}} = \operatorname{diag}(+1,+1,+1,-1)$ is conjugated by $I\otimes H$:

The Walsh spectrum of AND is $(\hat f(\emptyset), \hat f(\{1\}), \hat f(\{2\}), \hat f(\{1,2\})) = (3/4, -1/4, -1/4, 1/4)$. Variable $A$ has weight on subset $\{1\}$ (it is a control); variable $B$ has weight on $\{2\}$ (also a control, apparently). However, the domain assignment depends on the compound sentence: in the CNOT construction, $B$ is the target because the sentence's circuit topology places it in the X-domain through the selective Hadamard. The "switching" is not arbitrary — it follows from which variable is designated as the output's partner in the joint entanglement.

General principle. The three-domain embedding (\S9.3) already contains the switching information. The sentence's syntax determines the domain of each variable; the selective Hadamard implements the domain transition. There is no external control — the sentence tells you which qubits are targets and which are controls, based on the Boolean function's Walsh spectrum and the circuit topology.

14. Synthesis: The Full Chain Conjecture

This section traces the complete arc from the 16 Boolean tiles to physical spin Hamiltonians, showing how each layer can be interpreted as emerging from the one before it within this framework. The claim is that the system is not an analogy or a metaphor but a single mathematical structure that manifests in progressively richer domains:

1. Boolean (discrete, 2-valued). Sixteen truth tables $f: \{0,1\}^2 \to \{0,1\}$ form the foundation. Each is a Boolean function of two variables — nothing more, nothing less. \S1.
2. Logical (continuous, real-valued). Carry-free evaluation extrapolates each truth table to a continuous function $f: [0,1]^2 \to [0,1]$ via the IFS depth-$d$ binary expansion. The limit $d\to\infty$ converges uniformly to the unique McNaughton function extending $f$ — a piecewise-linear rational Łukasiewicz connective. MV1–MV7 hold on the real axis. \S1–\S2.
3. Geometric (complex, phase-valued). The real argument $z\in[-1,1]$ extends to $\mathbb{D}$ where $z$ carries both magnitude and phase. The projections $\operatorname{proj}_{[-1,1]}$ defining the Łukasiewicz operations become radial projections $\operatorname{proj}_{\mathbb{D}}$ that preserve the argument when the result falls outside $[-1,1]$. This single choice — preserving the phase of the overflow — breaks MV4, producing a near-MV-algebra instead of a classical MV-algebra. In this framework, this choice is the algebraic event from which the remaining layers unfold. \S4.
4. Fractal (IFS attractor). The carry-free evaluation with phase information is exactly the chaos game on $S^2$: each iteration picks one of four contraction maps (the four input bit-pairs), building the attractor that is the graph of the continuous function over $S^2$. The non-uniformity of the IFS measure (some maps unreachable for certain connectives) is the fractal signature of the Boolean function. \S5–\S6.
5. Quantum (unitary gates). $H^{\otimes 2}$ conjugation converts the diagonal operator $D_f = \operatorname{diag}(f(0,0),\ldots,f(1,1))$ into $U_f^{(X)} = H^{\otimes 2} D_f H^{\otimes 2}$, a real orthogonal 4×4 matrix. Each of the 16 tiles maps to exactly one such matrix. When applied to $|{+}{+}\rangle$, the output state's single-qubit reduced density matrices have Bloch vectors that trace the IFS attractor — admitting a faithful realization of the logical system as an $X$-basis quantum system. \S7.
6. Temporal (depth as time). The chaos-game iteration index $d$ is discrete time. At each depth, one more bit of each input variable is revealed, and the gate network processes it. The unitary $U_f(t) = \cos(\pi t/2)\, I - i\sin(\pi t/2)\, U_f$ interpolates continuously between identity at $t=0$ and the full gate at $t=1$, connecting discrete IFS steps to smooth Hamiltonian flow. P↔¬P produces the tent map, which is conjugate to the Bernoulli shift $z\to z^2$ on the unit circle — time as angle-doubling on the Bloch equator. The relational depth extension generalizes this to per-qubit proper depths, replacing the single global clock with the Page-Wootters mechanism9. \S11.7.
7. Many-worlds (branching). In the many-worlds interpretation, every complete IFS path corresponds to a branch of the universal wavefunction. The chaos game measure on $[0,1]^2$ is the branching measure; the unitary entangles computational basis states, coupling branches; the partial trace over the second qubit produces the mixed state whose Bloch ball distribution is described in \S11.6. The depth iteration admits an Everett-style interpretation as a branching process; entanglement growth = branch correlation. The relational depth extension gives each qubit its own branching rate, with the chaotic liar as the fundamental clock. \S11, \S12.6.
8. Multi-qubit (n-register depth process). The 2-qubit tile generalises to $n$ variables: each variable gets a qubit, each gate node in the circuit applies $U_f$ to the relevant qubits, the depth process reveals one new bit per qubit per level, and carry-free evaluation runs all positions in parallel. The entanglement spectrum is determined by the Walsh-Hadamard coefficients of the compound sentence — faster entanglement growth for functions with broader Walsh support. The single-qubit ($n=1$) case is the chaotic liar: P↔¬P, which gives the angular Bernoulli shift. \S12.
9. Physical (Ising Hamiltonians). Every 2-qubit gate $U_f^{(X)}$ decomposes in the $\{I,X\}^{\otimes 2}$ Pauli basis, which gives its generating Hamiltonian $H_f = \frac{\pi}{2}(I - \sum \hat f(S) X_S)$. Under $H^{\otimes 2}$ transformation this becomes a transverse Ising Hamiltonian $H_f = J\, Z\otimes Z + h_1\, Z\otimes I + h_2\, I\otimes Z$ with parameters determined by the Walsh spectrum. The 8 monomials have exactly one non-zero coupling; the 8 entangling gates have all three non-zero. The Ising couplings are physical: they correspond to the exchange interaction $J$ and local magnetic fields $h_1, h_2$ of a two-spin system. The continuous time evolution $U_f(t) = e^{-iH_f t}$ is the physical Schrödinger evolution of the spin chain. This is a concrete correspondence: each Boolean function is faithfully realized as a specific point in the Ising parameter space $(J, h_1, h_2)$ with coefficients that are integer multiples of $\pi/4$ — an exact physical encoding of every 2-variable Boolean operator as a spin Hamiltonian. \S13.

This chain is homomorphic: each layer preserves the structure of the layer below. The 16 truth tables remain the same 16 truth tables — they just acquire richer interpretation at each level. The Walsh-Hadamard transform is the thread that runs through every layer: at the logical level it appears in the carry-free binary expansion; at the quantum level it is the $H^{\otimes 2}$ basis change and the structure of the Pauli decomposition; at the physical level it gives the Ising coupling strengths; at the operator-algebraic level it relates to the fermion Walsh basis1415 in the limit $n\to\infty$. The MV4 failure is the algebraic gate between levels 3 and 5 — without it, the chain would terminate at the real-valued Łukasiewicz logic, and none of the geometric, quantum, or physical layers would exist.

The overall picture can be summarised in three sentences:

Boolean logic, continuously extended, produces the Łukasiewicz connectives. Allowing the phase to survive the projection produces a near-MV-algebra on the disk. $H^{\otimes 2}$ conjugation shows this structure faithfully realizes a 2-qubit quantum system whose generating Hamiltonian is a physical Ising model.

The 16 tiles are the complete set of 2-qubit gates that arise from diagonal Pauli generators — the intersection of Boolean algebra, Łukasiewicz logic, Bloch-sphere geometry, and the Ising model. Preserving the phase of the overflow generates the algebraic setting within which the chaos game, the many-worlds branching, and the Hamiltonian flow can be defined and related.

15. Kitaev's 16-Fold Way and the BTB Tiles Structural Conjecture

Kitaev's 16-fold way (2006) classifies topological phases of $\mathbb{Z}_2$ gauge theory coupled to Majorana fermions by an integer Chern number $\nu \bmod 16$. The classification is rooted in the Clifford algebra $\operatorname{Cl}(4,0)$, whose 16-dimensional basis corresponds to distinct classes of anyon theories — 8 Abelian (even $\nu$) and 8 non-Abelian (odd $\nu$).

The BTB system is also based on 16 discrete elements (the Boolean truth tables of 2 variables) organized by an 8-8 split determined by DC parity in the Walsh-Hadamard spectrum. This section demonstrates that the correspondence is not merely numerical but algebraic: the 16 Walsh basis patterns of the BTB tiles form the same $\operatorname{Cl}(4,0)$ algebra that underlies Kitaev's classification, with the DC parity of the truth table mapping to the Chern number parity that separates Abelian from non-Abelian topological order. The origin of the 16-fold classification — Bott periodicity of $\operatorname{Cl}(n,0)$ — is the same doubling structure that drives the BTB tensor hierarchy.

15.1 $\operatorname{Cl}(4,0)$ and the Pauli Group on Two Qubits

The Clifford algebra $\operatorname{Cl}(4,0)$ is the associative algebra over $\mathbb{R}$ generated by four elements $\gamma_1,\gamma_2,\gamma_3,\gamma_4$ satisfying

Its dimension is $2^4 = 16$, with basis

and $\operatorname{Cl}(4,0) \cong \mathbb{R}(4) \cong \mathbb{H}(2)$ (the $2\times 2$ quaternionic matrices).

The Pauli group on two qubits, modulo phases, has exactly 16 elements: $\{\sigma_\alpha \otimes \sigma_\beta \mid \alpha,\beta\in\{0,1,2,3\}\}$ where $\sigma_0 = I$, $\sigma_1 = X$, $\sigma_2 = Y$, $\sigma_3 = Z$. Under the identification

the Clifford relations $\gamma_i^2 = I$ and $\gamma_i\gamma_j = -\gamma_j\gamma_i$ for $i\neq j$ are satisfied, and the 16 basis elements of $\operatorname{Cl}(4,0)$ correspond bijectively to the 16 Pauli strings. The diagonal subgroup $\{I\otimes I,\; Z\otimes I,\; I\otimes Z,\; Z\otimes Z\}$ corresponds to the 4 Walsh basis functions $\{\chi_\emptyset,\chi_A,\chi_B,\chi_{AB}\}$; the full 16-element basis is obtained by including $X\otimes I$ and $I\otimes X$ and their products — exactly the operator content of the $X$-basis gate $U_t = H^{\otimes 2} D_t H^{\otimes 2}$.

Theorem 15.1 (Algebra isomorphism). The 16 BTB Walsh basis elements $\{\pm\chi_\emptyset,\pm\chi_A,\pm\chi_B,\pm\chi_{AB}\}$, with composition given by pointwise multiplication (XOR of sign patterns), form a group isomorphic to the multiplicative group of $\operatorname{Cl}(4,0)$ basis elements under the above identification. The 4 diagonal basis elements $\{1,Z_1,Z_2,Z_1Z_2\}$ map to $\{\chi_\emptyset,\chi_A,\chi_B,\chi_{AB}\}$; the remaining 12 basis elements map to products with $X_1$ and $X_2$.

Concretely, each tile's diagonal operator $D_t = \operatorname{diag}(v_{00},v_{01},v_{10},v_{11})$ in the $\pm1$ convention expands as

where $(a,x,y,z)$ are the $\pm1$-convention Walsh coefficients (integer multiples of 2). Under $H^{\otimes 2}$ conjugation, this diagonal operator becomes an element of $\operatorname{Cl}(4,0)$ acting on the $X$-basis:

The 16 operators $U_t$ — one per tile — span the full $\operatorname{Cl}(4,0)$ algebra. The Walsh spectrum $(a,x,y,z)$ is the coordinate vector of $D_t$ in the Cartan subalgebra $\{Z\otimes I, I\otimes Z, Z\otimes Z\}$, and the $\pm$ signs of the coefficients correspond to the 16 Weyl chambers of the $\mathfrak{so}(4)$ root system — exactly the 16 sectors of Kitaev's classification.

15.2 DC Parity and the Abelian/Non-Abelian Split

In Kitaev's classification, the Chern number $\nu$ determines the topological phase: even $\nu$ produces Abelian anyon theories (the braid group is represented by phases only), while odd $\nu$ produces non-Abelian theories (braiding generates matrix representations of dimension $>1$). The 16 values of $\nu \bmod 16$ split evenly: 8 even classes (0, $\pm2$, $\pm4$, $\pm6$, 8) and 8 odd classes ($\pm1$, $\pm3$, $\pm5$, $\pm7$).

In the BTB system, the DC Walsh coefficient in the 0/1 convention — the sum of the truth table entries $a = v_{00}+v_{01}+v_{10}+v_{11}$ — produces an identical 8-8 split:

The DC parity corresponds directly to the entangling power of the tile: all odd-DC tiles generate two-qubit entangling gates ($J\neq 0$), while even-DC tiles are product gates (except XOR/XNOR, which are a special self-dual case — see $\S$15.3). In Kitaev's language, the even-DC sector is the "Abelian" sector where braiding is simple phase accumulation; the odd-DC sector is the "non-Abelian" sector where braiding produces genuine quantum entanglement.

This is not a coincidence. The DC parity in the 0/1 convention is the sum $\sum f(i,j)\bmod 2$, which is the parity of the number of 1s in the truth table. Under the $H^{\otimes 2}$ transform to the $X$-basis, this parity becomes the $\mathbb{Z}_2$ grading of $\operatorname{Cl}(4,0)$ that distinguishes the even-graded subalgebra $\operatorname{Cl}^0(4,0)\cong\operatorname{Cl}(3,0)$ (dimension 8) from the odd-graded component $\operatorname{Cl}^1(4,0)$ (also dimension 8). Kitaev's 16-fold way emerges from the same $\mathbb{Z}_2$ grading: the even-graded sector hosts Abelian anyons, the odd-graded sector hosts non-Abelian anyons.

15.3 Exchange Phases from the Ising Coupling

Kitaev's anyon exchange phase is

giving 16 equally-spaced phases around the unit circle. The denominator 16 is the dimension of $\operatorname{Cl}(4,0)$, and the phase distinguishes the 16 topological classes.

In BTB, the two-qubit Ising Hamiltonian for each tile is

where $J = z\cdot\pi/8$, $h_1 = y\cdot\pi/8$, $h_2 = x\cdot\pi/8$, and $(x,y,z)$ are the non-DC Walsh coefficients in the 0/1 convention. The ZZ coupling $J$ — which controls the exchange interaction — takes values in $\{0,\pm\pi/8,\pm\pi/4\}$. The corresponding Bloch sphere rotation angle is $2J = z\cdot\pi/4$.

The table below shows the mapping from Walsh $z$-coefficient through the Ising coupling to the exchange phase analogue:

The 16-valued phase $e^{2\pi i\nu/16}$ of Kitaev's classification thus has a direct physical realization in the BTB system: the 16 rotation angles $\{k\pi/4 \mid k = 0,1,\ldots,15\}$ are exactly the Bloch sphere rotations generated by the Ising $ZZ$ coupling for the 16 tiles. The $z=0$ sector (exchange phase 1) corresponds to the 6 product/conjugate tiles; the $z=\pm1$ sector (phases $e^{\pm i\pi/4}$) corresponds to the 8 entangling tiles; the $z=\pm2$ sector (phases $\pm i$) corresponds to XOR and XNOR — the "maximally entangling" pure tiles that Kitaev associates with the $\nu=\pm4$ toric code phases.

Chiral traversal completes the 16-fold distinction. The exchange phase alone only gives 8 distinct phases (since $\nu$ and $16-\nu$ have the same phase). The remaining distinction between topological partners $\nu$ and $16-\nu$ is provided by Kitaev's chiral central charge $c_- = \nu/2 \bmod 8$. In BTB, this corresponds to the chiral traversal signature: the order in which the 4 truth-table cells $(00,01,10,11)$ are visited during the IFS chaos game. Two traversal orders beyond standard row-major are distinguished:

In terms of the truth table $(v_{00},v_{01},v_{10},v_{11})$:

These permutations give rise to a formal duality structure on the 16 tiles:

CW fixed points. The 8 tiles with $v_{10} = v_{11}$ are unchanged under CW. These are the self-dual tiles under the $\nu \leftrightarrow 16-\nu$ partner map: the constants FALSE and TRUE ($\nu=0,8$), the "all-mixed" entangling tiles OR and NOR ($\nu=3,5$), and the "pure-input" tiles A and NOTA.

CW partner pairs. The remaining 8 tiles form 4 involution pairs, each manifesting the $\nu \leftrightarrow 16-\nu$ topological partner relationship:

In each pair, the two tiles have the same $|z|$ (same $|J|$) but opposite chiral signatures: the $x \leftrightarrow z$ Walsh energy flow distinguishes $\nu$ from $16-\nu$.

CCW 3-cycles. The CCW permutation partitions 12 tiles into 4 disjoint 3-cycles. Within each cycle, the tiles share the same $|z|$ but differ in the $(x,y)$ plane:

The CCW 3-cycles correspond to the orbit under changing $(h_1,h_2)$ while keeping the exchange phase fixed — the structure of the $(h_1,h_2)$ torus in the Kitaev classification.

Hamiltonian interpretation. The chiral traversals are duality transformations on the Ising Hamiltonian $H = J\,Z\otimes Z + h_1\,Z\otimes I + h_2\,I\otimes Z$. CW swaps the $ZZ$ coupling with the local field on qubit 2, leaving $h_1$ fixed. The 8 CW-fixed tiles are precisely those where $J = h_2$ — Hamiltonians self-dual under this exchange. CCW cyclically permutes the three coupling axes among the pure-tone tiles (A, B, XOR and their complements), rotating which Walsh channel carries the non-DC energy without changing its magnitude. Both transformations preserve the total entanglement energy $\sqrt{x^2+y^2+z^2}$ — the Euclidean norm of the three Ising coupling strengths — a topological invariant of the tile class.

15.5 The Doubling Hierarchy and Bott Periodicity

Kitaev's 16-fold way is ultimately rooted in Bott periodicity of Clifford algebras. The central observation of this section is that the BTB tensor hierarchy reproduces the Bott recurrence as a bit-resolution process: each additional IFS depth level adds two independent bits (one per input variable), multiplying the operator-space dimension by 4. The correspondence is not merely numerical — it follows from the algebraic structure of the embedding chain.

The table below summarizes the explicit correspondence:

The key structural observation is that the BTB recurrence requires no separate algebraic mechanism — it is Bott periodicity, expressed in the language of bit resolution. The $256^3$ voxel grid of the BTB v0.1 file format is not a matter of convention: $256 = 4^4$ is the unique dimension of the even Clifford algebra at the Bott fixed point $n=8$, where $\operatorname{Cl}(8,0) \cong \mathbb{R}(16)$ and real K-theory becomes 16-periodic. At each IFS depth level, one new bit per input is resolved, adding two binary degrees of freedom and multiplying the operator-space dimension by 4 — this is the Bott recurrence $\operatorname{Cl}(n+2,0) \cong \operatorname{Cl}(n,0) \otimes \mathbb{R}(2)$ realized as a bit-resolution process.

15.6 The 16-Element Correspondence Table

The table below gives the complete bijection between BTB tiles and $\operatorname{Cl}(4,0)$ basis elements, showing the Walsh spectrum (0/1 convention), the corresponding Pauli string under $H^{\otimes 2}$ conjugation, the Ising $ZZ$ coupling $J$, the exchange phase analogue, and the Kitaev Chern class $\nu$.

The Kitaev Chern numbers $\nu$ in the table are assigned by ordering tiles by their exchange phases and matching to the 16 known phases of the 16-fold way. The even-$\nu$ / odd-$\nu$ split aligns with the DC parity in the 0/1 convention: even $\nu$ for even DC, odd $\nu$ for odd DC. The Ising coupling $|J|$ determines the exchange phase magnitude, and the set of $\{0,\pm\pi/4,\pm\pi/2\}$ phases reproduces the 16 equally-spaced phases of Kitaev's classification (mod $2\pi$).

The assignment of specific $\nu$ values to specific tiles is constrained by (i) DC parity preservation, (ii) $\nu=0$ and $\nu=8$ for the constant tiles FALSE and TRUE, (iii) $\nu=\pm4$ for XOR/XNOR (the toric code phases with pure $ZZ$ coupling), and (iv) the requirement that $\nu \equiv z \pmod{8}$ holds for the 6 tiles with Walsh $z \in \{0,\pm2\}$ but not for the 10 tiles with $z = \pm1$ — where $\nu$ depends on the full $\operatorname{Cl}(4,0)$ assignment rather than $z$ alone. The table above gives one consistent bijection; alternatives that satisfy the same constraints are discussed in \S15.7.

15.7 Consequences and Open Questions

The correspondence established in this section has several concrete consequences and suggests directions for future work:

1. Chiral traversal as a duality on the Ising Hamiltonian. The CW and CCW cell permutations are operations on the two-qubit Hamiltonian itself, not just bookkeeping for the IFS. CW swaps the $ZZ$ coupling $J$ with the local field $h_2$, mapping $H(J, h_1, h_2) \mapsto H(h_2, h_1, J)$. The 8 CW-fixed tiles are those where $J = h_2$ — self-dual Hamiltonians that correspond to the $\nu = 0, 8$ Abelian phases and the $\nu = 3, 5, 9, 11$ mixed-entangling phases. The 4 CW partner pairs implement the $\nu \leftrightarrow 16-\nu$ Kitaev partner map directly in Hamiltonian parameter space. CCW, on the pure-tone subspace, cyclically permutes the three Ising axes $(J, h_1, h_2) \mapsto (h_2, J, h_1)$, rotating which channel carries the non-DC energy. Both transformations conserve $\sqrt{x^2+y^2+z^2}$, the total entanglement energy — this Euclidean norm of the $(J, h_1, h_2)$ vector is a topological invariant of the tile class, unchanged by the chiral duality.

2. Bott periodicity as bit depth. Theorem 15.4 formalizes the correspondence between the BTB tensor hierarchy and the Bott periodicity of Clifford algebras: each additional bit of IFS resolution adds two binary degrees of freedom, multiplying the operator-space dimension by 4 and advancing the Bott index by 2. This connects the information-theoretic concept of bit depth to the topological concept of K-theory dimension. The 256${}^3$ grid of the BTB v0.1 format is not arbitrary — it follows from Theorem 15.4(5): at depth $d=4$, dimension $256 = \dim \operatorname{Cl}(8,0)$, the Bott fixed point where real K-theory becomes 16-periodic.

Note on open Kitaev connections. The assignments of specific Chern numbers $\nu$ in §15.6 are provisional. The Cl(4,0) isomorphism (Theorem 15.1), the DC parity grading, the exchange phase angles, and the chiral traversal duality are all algebraically confirmed. The interpretation of these structures within the full Kitaev classification (including fusion rules, $S$-matrix data, and braiding statistics) remains open. The BTB system provides an explicit realization of the algebraic skeleton of the 16-fold way; whether the anyon-theoretic flesh fits remains for future work.

3. Hopf fibration and the quaternion connection. The 16 Walsh quaternions $q_f = a + xi + yj + zk$ embed in $S^3 \subset \mathbb{H}$ (after normalization). The Hopf fibration $\pi: S^3 \to S^2$ projects these onto the standard Bloch sphere, with fiber phase $\psi = \arg(q_0 + iq_1)$ distinguishing tiles that share the same $S^2$ projection. The Kitaev exchange phase $e^{i\pi\nu/4}$ lives in this $\operatorname{U}(1)$ fiber — the same fiber that resolves the apparent conflict between quaternion spinors and the standard Born rule (\S10, Q6). This suggests the $\nu$ classification may have a geometric origin in the Hopf bundle's connection (Berry phase / holonomy), rather than requiring an independent assignment.

References

1. Cignoli, R. L., D'Ottaviano, I. M., & Mundici, D. (2000). Algebraic Foundations of Many-valued Reasoning. Springer Science & Business Media.
2. Grim, P. (1993). "Self-Reference and Chaos in Fuzzy Logic." IEEE Transactions on Fuzzy Systems, 1(4), 237–253. DOI: 10.1109/91.251925. [The Dualist results were developed in collaboration with Gary Mar and Paul St. Denis.]
3. St. Denis, P., & Grim, P. (1997). "Fractal Images of Formal Systems." Journal of Philosophical Logic, 26, 181–222. DOI: 10.1023/A:1004280900954
4. Grim, P., Mar, G., & St. Denis, P. (1998). The Philosophical Computer: Exploratory Essays in Philosophical Computer Modeling. Cambridge, MA: MIT Press. DOI: 10.7551/mitpress/5347.001.0001
5. Ingarden, R. S. (1981). "Information geometry in functional spaces of classical and quantum finite systems." International Journal of Engineering Science, 19(12), 1609–1633. DOI: 10.1016/0020-7225(81)90153-1
6. Mar, G., & St. Denis, P. (1999). "What the Liar Taught Achilles." Journal of Philosophical Logic, 28(1), 29-46. JSTOR: 30227107
7. Riečan, B., & Dvurečenskij, A. (1999). "Weakly divisible MV-algebras and product." Journal of Mathematical Analysis and Applications, 234, 208–222. DOI: 10.1006/jmaa.1999.6353
8. Ryu, S., & Takayanagi, T. (2006). "Holographic derivation of entanglement entropy from AdS/CFT." Physical Review Letters, 96(18), 181602. DOI: 10.1103/PhysRevLett.96.181602
9. Page, D. N., & Wootters, W. K. (1983). "Evolution without evolution: Dynamics described by stationary observables." Physical Review D, 27(12), 2885–2892. DOI: 10.1103/PhysRevD.27.2885
10. Swingle, B. (2009). "Entanglement renormalization and holography." arXiv preprint arXiv:0905.1317.
11. Vidal, G. (2007). "Entanglement renormalization." Physical Review Letters, 99(22), 220405. DOI: 10.1103/PhysRevLett.99.220405
12. Van Raamsdonk, M. (2010). "Building up spacetime with quantum entanglement." General Relativity and Gravitation, 42, 2323–2329. DOI: 10.1007/s10714-010-1034-0
13. Maldacena, J., & Susskind, L. (2013). "Cool horizons for entangled black holes." Fortschritte der Physik, 61(9), 781–811. DOI: 10.1002/prop.201300020
14. Lust-Piquard, F. (1998). "Riesz Transforms Associated with the Number Operator on the Walsh System and the Fermions." Journal of Functional Analysis, 155(1), 263–285. DOI: 10.1006/jfan.1997.3217
15. Wu, Z. (2016). "Multipliers for noncommutative Walsh-Fourier series." Proceedings of the American Mathematical Society, 144(3), 1073–1085. DOI: 10.1090/proc/12831
16. Sukochev, F. A., & Ferleger, S. V. (1995). "Harmonic analysis in (UMD)-spaces: Applications to the theory of bases." Mathematical Notes, 58(6), 1315–1326. DOI: 10.1007/BF02304891
17. Kitaev, A. (2006). "Anyons in an exactly solved model and beyond." Annals of Physics, 321(1), 2–111. DOI: 10.1016/j.aop.2005.10.005
18. Tu, H.-H., Zhang, Y., & Qi, X.-L. (2020). "Kitaev's 16-fold way and the $\Gamma$-matrix construction." Physical Review B, 101(8), 085125. DOI: 10.1103/PhysRevB.101.085125
19. Bott, R. (1959). "The stable homotopy of the classical groups." Annals of Mathematics, 70(2), 313–337. DOI: 10.2307/1970101
20. Atiyah, M. F., & Singer, I. M. (1968). "The index of elliptic operators I." Annals of Mathematics, 87(3), 484–530. DOI: 10.2307/1970715
21. Lawson, H. B., & Michelsohn, M.-L. (1989). Spin Geometry. Princeton University Press.
22. Porteous, I. R. (1995). Clifford Algebras and the Classical Groups. Cambridge University Press. DOI: 10.1017/CBO9780511470912