The Fantastic Mathematics of τau

"Why was Six afriad of Seven.
Because Seven mod Nine"

Abstract

We study the sequence $f(n) = \lfloor n/\tau \rfloor \bmod 9$, where $\tau = 2\pi$ and $n$ begins at zero. This sequence decomposes into consecutive blocks of length 6 and 7 — and no other lengths. Within each 710-step cycle, 113 blocks appear (81 of length 6, 32 of length 7), arranged as 15 repetitions of a base phrase $\{7,6,6,7,6,6,6\}$ and 2 interrupting phrases $\{7,6,6,6\}$ in the configuration $8 + \text{break} + 7 + \text{break}$. The spacings between the 32 blocks of length 7 form a Sturmian word on $\{3,4\}$: specifically $(3,4)^8 \cdot 4 \cdot (3,4)^7$.

The complete value-and-length cycle spans 6390 steps (1017 blocks), where a set of column-sum coincidences emerges: the sum of block values, the sum of start-indices modulo 9, and the sum of stop-indices modulo 9 all equal 4068, while the sum of (value $\times$ length) equals 25,560 — and $25{,}560/\tau \approx 4068.0003$. The terminal indices 709 and 6389 are both prime. The block count 113 is prime. At large scales, $\lfloor 710 \times 10^6/\tau \rfloor = 113{,}000{,}009$ is prime. These prime appearances are arithmetically notable, though the paper does not prove a causal prime-generating mechanism.

The paper's central result concerns exact recurrence. At offset $q = 1{,}308{,}519$ within a 6390-point renormalized window, the accumulated phase drift remains below every carry threshold, yielding pointwise identity between the base pattern and its translate: $y_{q}(n) - 113 = y_0(n)$ for all $n \in [0, 6389]$. This exact return persists for 25 multiples. At the 26th multiple, the first failures occur at indices $\{4593, 5303, 6013\}$, spaced by 710. Crucially, these same indices appear independently in the empirically documented drift pattern of the 102,240-cycle — where $f(n) \ne f(n + 102{,}240)$ at exactly $\{2463, 3173, 3883, 4593, 5303, 6013\}$, all spaced by 710. Both phenomena are governed by the same carry-threshold mechanism: the accumulated approximation error exceeding the minimum gap between fractional phases and their nearest integer.

These results are framed within Symbolic Necessity ($\odot$): a proposed formal property of structures whose completeness outruns finite derivation but whose bounded projections nonetheless yield exact discrete law.

---

1. Introduction

1.1 Counting circles

Take any natural number $n$ and form $n/\tau$, where $\tau = 2\pi$. A circle with radius $n/\tau$ has circumference $2\pi \cdot n/\tau = n$. So $n/\tau$ is the radius of a circle with integer circumference $n$. The function $\lfloor n/\tau \rfloor$ counts completed turns; $\{n/\tau\}$ records residual phase.

Starting from $n = 0$ is critical. At $n = 0$: $0/\tau = 0$, $\lfloor 0/\tau \rfloor = 0$, $f(0) = 0$. The sequence begins at the origin — zero radius, zero circumference, zero phase, zero value. Every subsequent structure emerges from this origin by adding one unit of circumference.

1.2 The blocks of six and seven

Define $f(n) = \lfloor n/\tau \rfloor \bmod 9$. Since $\tau \approx 6.2832$, each unit step in $n$ advances $n/\tau$ by $1/\tau \approx 0.1592$. The floor increments approximately every $\tau \approx 6.28$ steps: sometimes after 6 steps, sometimes after 7. The block lengths are $\lfloor \tau \rfloor = 6$ and $\lceil \tau \rceil = 7$, and no other values.

1.3 Epistemic status

This paper presents results at three levels:

• Proved results (Sections 2–6): Block decomposition, Sturmian structure, column sums, exact finite-window return, and the 102,240-cycle unification. All computationally verified and algebraically derived.
• Computationally verified observations (Sections 7–8, Appendices B–C): The meta-return at 73,996,200, primes at scale, the Tau-Wave system, repeating decimal tail ratios, Gelfond's constant identification. Confirmed but not derived from first principles.
• Interpretive framework (Section 9): Symbolic Necessity. A proposed conceptual reading, not a theorem.

1.4 Why $\tau$ rather than $\pi$

Any formal treatment of rotation requires a distinguished full-turn constant. We use $\tau$ because: the circle group $S^1$ has period $\tau$; the kernel of $\theta \mapsto e^{i\theta}$ from $\mathbb{R}$ to $S^1$ is $\tau\mathbb{Z}$; unit-speed parametrisation $\gamma(t) = (r\cos t, r\sin t)$ has period $\tau$; and the identity $e^{i\tau} = 1$ represents one complete rotation directly. None of this changes the mathematics — but it simplifies notation and aligns the constant with its geometric role.

---

2. Block Properties and the 710-Cycle

2.1 Formal definitions

Definition 1 (Block). A block is a maximal run of consecutive $n$ sharing the same value of $f(n) = \lfloor n/\tau \rfloor \bmod 9$. Each block has: Block ID (sequential from 0), Value $v \in \{0,\ldots,8\}$, Length $\in \{6,7\}$, Start $n$, Stop $n$, Block Sum (Value $\times$ Length), Start $n \bmod 9$, Stop $n \bmod 9$.

Definition 2 (Increment indicator). $I(n) = \lfloor(n+1)/\tau\rfloor - \lfloor n/\tau \rfloor \in \{0,1\}$ records whether $\lfloor n/\tau \rfloor$ increments at step $n$. Block lengths are the gaps between successive $n$ where $I(n) = 1$.

2.2 The first seven blocks

ID	Value	Length	Start $n$	Stop $n$	Sum	Start mod 9	Stop mod 9
0	0	7	0	6	0	0	6
1	1	6	7	12	6	7	3
2	2	6	13	18	12	4	0
3	3	7	19	25	21	1	7
4	4	6	26	31	24	8	4
5	5	6	32	37	30	5	1
6	6	6	38	43	36	2	7

These 7 blocks span 44 integers ($n = 0$ to $43$) and form the base phrase: $\{7,6,6,7,6,6,6\}$. Block values progress sequentially: $0, 1, 2, 3, 4, 5, 6$.

2.3 The 710-cycle: phrase-and-break architecture

Proposition 1. In $0 \le n \le 709$: exactly 113 blocks (81 of length 6, 32 of length 7), totalling $81 \times 6 + 32 \times 7 = 710$, arranged as:

Phase	Content	Blocks	Integers	Range
Phase 1	8 base phrases $\{7,6,6,7,6,6,6\}$	56	352	$n = 0$–$351$
Break 1	1 break phrase $\{7,6,6,6\}$	4	25	$n = 352$–$376$
Phase 2	7 base phrases	49	308	$n = 377$–$684$
Break 2	1 break phrase	4	25	$n = 685$–$709$
Total		113	710

Check: $15 \times 7 + 2 \times 4 = 113$. $15 \times 44 + 2 \times 25 = 710$. $\square$

The base phrase contains two 7-blocks at positions 0 and 3 (a pair). The break phrase contains one 7-block (a singleton). Normally, 7-blocks come in pairs; at the two break points, a singleton appears. This accommodates $\tau - 6 \approx 0.2832$.

2.4 Continued fraction context

$\tau = [6; 3, 1, 1, 7, 2, 146, 3, 1, 1, \ldots]$. The convergent $710/113$ truncates after partial quotient $a_5 = 2$ and gives $|710/113 - \tau| \approx 5.34 \times 10^{-7}$. The 8 + break + 7 + break phrase structure is consistent with the continued-fraction organisation around this convergent. A rigorous derivation of the phrase counts from the partial quotients is not given here, but the alignment is exact and computationally verified.

2.5 Value progression

After 113 blocks, the value has advanced by $113 \bmod 9 = 5$. Since $\gcd(5,9)=1$, full return requires $9$ shifts, giving cycle length $9 \times 710 = 6390$.

---

3. Sturmian Structure

3.1 The spacing sequence

The 32 blocks of length 7 in one 710-cycle occupy block indices:

$$\{0, 3, 7, 10, 14, 17, 21, 24, 28, 31, 35, 38, 42, 45, 49, 52, 56, 60, 63, 67, 70, 74, 77, 81, 84, 88, 91, 95, 98, 102, 105, 109\}.$$

The 31 spacings between consecutive 7-blocks (noncyclic — measured within the 710-window, not wrapping) are:

$$\underbrace{3,4,3,4,3,4,3,4,3,4,3,4,3,4,3,4}_{(3,4)^8}, \; \underbrace{4}_{\text{break}}, \; \underbrace{3,4,3,4,3,4,3,4,3,4,3,4,3,4}_{(3,4)^7}.$$

The lone $4$ at position 16 is the "double-4 break" — two consecutive spacings of 4 — marking the Phase 1 → Break 1 transition.

Theorem 1 (Sturmian Spacing). The spacing sequence is a Sturmian word on $\{3,4\}$ with slope $\alpha = 32/113$.

Proof. The irrational rotation on $[0,1)$ by $\theta = 1/\tau \approx 0.15915$ produces an orbit $\{n\theta \bmod 1\}_{n=0}^{112}$ that partitions $[0,1)$ into 113 intervals of exactly two distinct lengths (by the three-distance theorem, applied to the convergent $710/113 \approx \tau$ with error $\sim 10^{-7}$). Of 113 intervals, 81 have the shorter length (6-blocks) and 32 the longer (7-blocks). The distribution of 7-blocks among 6-blocks follows a Sturmian word with slope $32/113$, taking values $\lfloor 113/32 \rfloor = 3$ and $\lceil 113/32 \rceil = 4$. The initial-phase-dependent form $(3,4)^8 \cdot 4 \cdot (3,4)^7$ is verified computationally. $\square$

Structural correspondence. A $(3,4)$ pair corresponds to a base phrase $\{7,6,6,7,6,6,6\}$: two 7-blocks at distance 3, gap of 4 to the next pair. The lone 4 corresponds to a break phrase $\{7,6,6,6\}$: a singleton 7-block with gap 4 on both sides.

---

4. The Hierarchical Tower

4.1 The 6390-cycle

Proposition 2. The range $0 \le n \le 6389$ contains 1017 blocks (729 of length 6, 288 of length 7), comprising 9 complete 710-cycles. Each value $0$–$8$ appears as a block value 113 times and occupies exactly 710 positions. The terminal index $6389$ is prime.

4.2 Column sum coincidences

Theorem 2. Over one 6390-cycle (1017 blocks):

Column	Sum	Relation
Block Value	4068	$= \lfloor 25{,}560/\tau \rfloor$
Block Length	6390	exact
Block Sum (Value $\times$ Length)	25,560	$= 4 \times 6390$
Start $n \bmod 9$	4068	$=$ Block Value sum
Stop $n \bmod 9$	4068	$=$ Block Value sum

Moreover, $25{,}560/\tau \approx 4068.000345$, connecting the block-value sum and the block-product sum through $\tau$.

These coincidences do not hold at the 710 level (where the sums are $\{442, 710, 2777, 459, 444\}$). The 6390-cycle is the minimal level where they emerge.

4.3 Block-sum distribution

The 1017 block sums take 16 distinct values. The frequencies split cleanly: sums achievable only by length-6 blocks appear 81 times each; sums achievable only by length-7 blocks appear 32 times each; and the two degenerate sums (0 from value 0, and 42 from $6 \times 7 = 7 \times 6$) each appear 113 times.

4.4 Prime factorizations

Cycle lengths	Block counts
$710 = 2 \times 5 \times 71$	$113$ (prime)
$6{,}390 = 2 \times 3^2 \times 5 \times 71$	$1{,}017 = 3^2 \times 113$
$25{,}560 = 2^3 \times 3^2 \times 5 \times 71$	$4{,}068 = 2^2 \times 3^2 \times 113$
$102{,}240 = 2^5 \times 3^2 \times 5 \times 71$	$16{,}272 = 2^4 \times 3^2 \times 113$

The irreducible core is $\{2, 5, 71\}$ for cycle lengths and $\{113\}$ for block counts, with powers of 2 and 3 accumulating at each level.

4.5 Preserved approximation error

$$\frac{710}{\tau} \approx 113 + 9.595 \times 10^{-6}, \quad \frac{6{,}390}{\tau} \approx 1017 + 8.636 \times 10^{-5}, \quad \frac{25{,}560}{\tau} \approx 4068 + 3.454 \times 10^{-4}.$$

Error ratios: $9.000$ and $4.000$ to ten figures — the same factors as the cycle-length ratios. The per-unit error is identical across all levels:

$$\frac{710/\tau - 113}{710} = \frac{6{,}390/\tau - 1017}{6{,}390} = \frac{25{,}560/\tau - 4068}{25{,}560} \approx 1.3514 \times 10^{-8}.$$

At the rational level these are scaled copies of the same reduced fraction $710/113$. The nontrivial claim is that the same scaling family reappears in the projected block structure, column sums, and exact finite-window returns.

4.6 Primes at cycle boundaries

The terminal index $709$ is prime. The terminal index $6389$ is prime. The block count $113$ is prime, and $71$ (a factor of $710$) is prime. By contrast, $25{,}559 = 61 \times 419$ and $102{,}239$ are composite. The primality does not persist at all levels. These appearances are arithmetically notable within the hierarchy, but the paper does not prove a causal relationship.

4.7 Block-sum chain

The sum of all block sums in one 6390-cycle ($25{,}560$) equals $4 \times 6390$, the length of the next cycle level. The factor of 4 is the same one appearing in $25{,}560/6{,}390 = 4$ and $4068/1017 = 4$.

---

5. Exact Finite-Window Return

5.1 The claim

The raw orbit $\{n/\tau\}$ never repeats. But a specific coding of it returns to itself with pointwise identity after renormalization — not approximately, but exactly — because the residual error stays below every carry boundary in a bounded window.

5.2 The renormalized return map

For $n \in \{0, \ldots, 6389\}$, define:

$$y_q(n) = \left\lfloor \frac{n + 6390q}{\tau} \right\rfloor - 113 \left\lfloor \frac{n + 6390q}{710} \right\rfloor.$$

This subtracts the dominant linear drift (113 per 710-block), isolating residual modular structure.

5.3 The carry mechanism

Let $\varepsilon = 6390/\tau - 1017 \approx 8.636 \times 10^{-5}$. Decompose the accumulated drift:

$$q\varepsilon = m_q + \eta_q, \qquad m_q = \lfloor q\varepsilon \rfloor, \qquad \eta_q = \{q\varepsilon\}.$$

Then:

$$y_q(n) = y_0(n) + m_q + \widetilde{C}(n, q),$$

where the residual carry function is

$$\widetilde{C}(n, q) = \left\lfloor \left\{ \frac{n}{\tau} \right\} + \eta_q \right\rfloor.$$

The window returns exactly (up to integer shift $m_q$) when $\widetilde{C}(n, q) = 0$ for all $n$, which occurs when $\eta_q < \delta$, where $\delta = 1 - \max_{0 \le n < 6390}\{n/\tau\}$ is the minimum gap to a carry boundary.

5.4 Exact return at $q_2 = 1{,}308{,}519$

Theorem 3. $q_2\varepsilon \approx 113.0000516$, so $m_{q_2} = 113$ and $\eta_{q_2} \approx 5.164 \times 10^{-5}$. The minimum carry gap is $\delta \approx 1.327 \times 10^{-3}$. Since $\eta_{q_2} < \delta$:

$$\widetilde{C}(n, q_2) = 0 \quad \forall\, 0 \le n < 6390, \qquad \text{therefore} \quad y_{q_2}(n) - 113 = y_0(n).$$

This is pointwise identity, not approximation. Verified for all 6390 values.

5.5 Persistence and breakdown

Corollary. The return persists for $a = 1, \ldots, 25$. At $a = 26$, three indices fail:

$$n \in \{4593, \; 5303, \; 6013\}, \qquad \text{spaced by } 710.$$

These are the three values of $n$ in $[0, 6389]$ whose fractional phase $\{n/\tau\}$ is closest to 1. Verified by direct computation.

---

6. The 102,240-Cycle Drift: Unification

6.1 Empirical discovery

The research notes record that within the 102,240-cycle ($= 16 \times 6390$), the block sequence drifts from the base pattern at specific indices. Comparing $f(n)$ with $f(n + 102{,}240)$ over $n \in [0, 6389]$ reveals exactly 6 mismatches:

$$n \in \{2463, \; 3173, \; 3883, \; 4593, \; 5303, \; 6013\}, \qquad \text{all spaced by } 710.$$

This was discovered empirically. The notes further document that the drift continues into subsequent 6390-windows: at $n = 6723, 7433, 8143, \ldots$, still spaced by 710, with one new failure added per 710-step advance.

6.2 The carry-threshold explanation

Since $102{,}240 = 16 \times 6390$, the accumulated drift at this scale is $16\varepsilon \approx 0.001382$. Compare with the carry threshold $\delta \approx 0.001327$:

$$16\varepsilon \approx 0.001382 > \delta \approx 0.001327.$$

So carries fire wherever $\{n/\tau\} > 1 - 16\varepsilon$. The 6 values of $n$ satisfying this in $[0, 6389]$ are exactly $\{2463, 3173, 3883, 4593, 5303, 6013\}$, verified computationally. These are the 6 positions in the base window whose fractional phase is closest to 1, and they are spaced by 710 because the irrational rotation $\{n/\tau\}$ revisits the near-1 region at the 710-approximant scale.

6.3 Connection to Theorem 3

The Theorem 3 failures at $a = 26$ are $\{4593, 5303, 6013\}$ — a strict subset of the 102,240-drift positions. The mechanism is the same: accumulated error exceeding the carry threshold. The difference is quantitative:

Scale	Accumulated drift	Exceeds $\delta$ by	Failures in $[0, 6389]$
$a = 25$ ($25\eta_{q_2}$)	$\approx 0.001291$	does not exceed	0
$a = 26$ ($26\eta_{q_2}$)	$\approx 0.001343$	$\approx 1.6 \times 10^{-5}$	3
102,240 cycle ($16\varepsilon$)	$\approx 0.001382$	$\approx 5.5 \times 10^{-5}$	6

The 3 failures at $a = 26$ are the 3 positions with the largest fractional phases; the 6 failures at the 102,240 scale are the 6 positions above a slightly lower threshold. Both are governed by the same carry function, the same 710-lattice, and the same underlying approximation quality.

6.4 Significance

This is the paper's strongest structural result: an empirically discovered pattern-drift phenomenon and an algebraically derived carry-threshold mechanism converge on the same indices. The 102,240-cycle "drift" is not mysterious — it is the carry function firing at exactly the positions predicted by the rational approximation error. The 710-spacing of the failures is inherited from the 710-approximant structure. Two independent analyses, carried out months apart with different methods, produce the same list of integers.

---

7. Primes at Scale

Status: computationally verified observations.

7.1 The prime chain in $710/\tau$

$$710/\tau = 113.00000959524569\ldots$$

Reading left to right: $11$ is prime, $113$ is prime, $113{,}000{,}009 = \lfloor 710 \times 10^6/\tau \rfloor$ is prime. More precisely, $\lfloor 71 \times 10^7/\tau \rfloor = 113{,}000{,}009$ is prime (since $710 \times 10^6 = 71 \times 10^7$).

7.2 Primes from $71 \times 10^n/\tau$

$\lfloor 71 \times 10^n/\tau \rfloor$ is prime for $n = 7, 262, 3933$. At each of these scales, $710 \times 10^{n-1}/\tau$ is near an integer whose integer part is prime.

7.3 The meta-return prime

$\lfloor 73{,}996{,}200/\tau \rfloor = 11{,}776{,}861$ is prime. Furthermore, $1{,}177{,}686{,}100{,}001$ is prime. The function $\lfloor 73{,}996{,}200 \times 10^k/\tau \rfloor$ is prime at $k = 0, 5, 74, 193, 282, 775$.

7.4 Repeating decimal tails

The decimal expansion of $710 \times 10^y/\tau$ for large $y$ eventually settles into repeating tails. At $y = 6$, the tail is $\overline{16901408450704225352112676056338028}$. At $y = 261$ and $y = 3932$, different tails appear. The ratios between the lengths of the non-repeating prefixes at these three scales are remarkably clean: $4/11$, $33/14$, and $7/6$ — all ratios of small integers.

---

8. The Tau-Wave System

Status: computationally verified observation providing independent confirmation of the 73,996,200 return.

8.1 Setup

Define wave functions with two slightly different frequencies:

$$W_a = r_a \sin(x \cdot r_a^{-1}), \quad W_b = r_b \sin(x \cdot r_b^{-1}),$$

where $r_a = 113\tau/710 \approx 1 - 8.5 \times 10^{-8}$ (the $k$-constant from Section 4.5) and $r_b = \tau^{-1} \approx 0.15915$. Phase-shifted versions $W_c, W_d$ use an offset of $6390 \cdot h$. The interference pattern $(W_a - W_b) - (W_c - W_d)$ measures how the slightly detuned frequencies drift relative to each other as $h$ increases.

8.2 Return at $h \approx 11{,}580$

The wave interference returns to its initial state when $6390 \times h \approx 73{,}996{,}200$, giving $h \approx 11{,}580$. This provides a wave-based consistency check for the same meta-return scale — from continuous wave analysis rather than discrete block arithmetic.

Since $73{,}996{,}200 = 710 \times 104{,}220 = 6390 \times 11{,}580 = 25{,}560 \times 2895$, the return occurs simultaneously at every level of the hierarchy. The near-integer return $73{,}996{,}200/\tau \approx 11{,}776{,}861.0000165$ places the meta-cycle scale within $\sim 10^{-5}$ of an integer — and $11{,}776{,}861$ is prime.

---

9. Complex Phase-Locking

9.1 The system

With correction normalization active:

$$S_n = \sin(n(\alpha + i\beta)), \qquad \alpha = \frac{\tau}{q}, \quad \beta = \alpha\tan(\pi T),$$

where $q = q_1\tau^{q_2}$. Two real controls: $\alpha$ governs rotation, $\beta$ governs exponential shear.

9.2 Spoke-count law

Proposition 3. If $q = a/b$ in lowest terms, the spoke count is $a$.

$q$	Reduced	Spokes
17	$17/1$	17
1.7	$17/10$	17
0.3	$3/10$	3
7.05	$141/20$	141 (not 705)

The decimal point is irrelevant; only the reduced form matters. The $q = 7.05$ case is the strongest falsification test.

9.3 The correction toggle

With correction on: $\alpha$ is fixed; changing $T$ only changes $\beta$. The locking skeleton (spoke count) is invariant while strand shapes deform. With correction off: $\alpha$ drifts with $T$, creating apparent tongue motion that is a parameterization artifact.

9.4 The reappearance of 710

Samples are taken at $n_x(x) = n + x \cdot 710$. The inter-block phase shift $710 \cdot \tau/q$ produces strong alignment when near an integer multiple of $\tau$ — the same condition governing the arithmetic system. The 710-lattice is not a parameter choice; it is the natural comparison scale from the continued fraction.

9.5 Asymptotic circularity

For large $n\beta$, $\sin(n(\alpha + i\beta)) \sim -e^{-in\alpha + n\beta}/(2i)$, whose normalized shape is $e^{-in\alpha}$ — a pure circle. The observed progression from elliptical to circular form is built into the complex sine asymptotics.

9.6 Gelfond's constant and the $n$-dependent system

A related system $\sin(n^{i\tau^k})$ with $k = n(Z\tau)$ exhibits different convergence behaviour: as $Z$ increases, all points eventually settle to $\sin(1) \approx 0.84147098$. At $Z = 0$, two critical crossing values appear:

$$N_{\text{firstCross}} = e^{\pi/2} \approx 4.81048, \qquad N_{\text{SecondCross}} = e^{\pi} \approx 23.14069,$$

where $e^{\pi}$ is Gelfond's constant. Note $N_{\text{SecondCross}} = N_{\text{firstCross}}^2$. The second crossing marks where the orbit first returns to the real axis; a further period completion occurs at $N_{\text{SecondCross}} \cdot Q$ where $Q \approx 111.318$. The appearance of $e^{\pi}$ — a known transcendental — as the natural scale of this $\tau$-indexed complex system is an open observation.

---

10. Symbolic Necessity

10.1 Definition

Definition 3. $P$ is symbolically necessary for system $\mathcal{S}$, written $\odot P$, when:

1. Operational Dependence. $\mathcal{S}$ requires $P$ for semantic completeness.
2. Constructive Inaccessibility. $P$ cannot be finitely constructed by $\mathcal{S}$'s operations.
3. Projective Exactness. Finite subsystems can generate exact local structure by referencing $P$ through bounded projections.

10.2 $\odot \tau$

1. Operational Dependence. Cyclic structure, Fourier analysis, $e^{i\theta}$, and the residue theorem all require a full-turn constant.
2. Constructive Inaccessibility. By Lindemann–Weierstrass (1882), $\tau$ is transcendental. Algebra cannot construct what it depends on.
3. Projective Exactness. This paper provides a worked mathematical example. The nonperiodic orbit $\{n/\tau\}$ produces exact block structure, exact column sums, and exact finite-window returns through bounded codings.

10.3 The observer equation

$$D = F(S, R_O):$$

$S = \{n/\tau\}$ (continuous source), $R_O$ (coding scheme), $D$ (discrete output). Multiple codings of the same source:

Coding $R_O$	Output $D$
$\lfloor n/\tau \rfloor \bmod 9$	Block values, 6/7 lengths
Renormalized 6390-window	Exact carry-stable return
Tau-Wave interference	73,996,200 phase return
Complex $\sin(n(\alpha+i\beta))$ on 710-lattice	Spoke counts

Recurrence belongs to the coding, not the source. "$\tau$ is irrational" and "the pattern recurs exactly" are not contradictions.

10.4 Gödel complement

Gödel shows finite systems cannot internally verify all truths. The $\tau$-block system shows what finite systems can still achieve: exact projected law from a completion they cannot contain. The proposed complement:

Incompleteness does not merely limit finite systems. It also explains why local exactness can exist without global closure — because the local structure is a bounded projection of a richer whole.

10.5 Information at scale

At the scale of Theorem 3, nothing relevant to the verified recurrence is lost in the projection. The bounded coding captures the full recurrence structure of the coding on that window.

---

11. Discussion

11.1 Central thesis

A continuous, nonperiodic generator can produce exact finite structure once viewed through a bounded observer map.

11.2 Claims and non-claims

Proved: Sturmian spacing $(3,4)^8 \cdot 4 \cdot (3,4)^7$. Phrase architecture $8 + \text{break} + 7 + \text{break}$. Column sums $\{4068, 6390, 25560, 4068, 4068\}$. Exact return at $q_2$, persisting 25 times, failing at $a = 26$ at $\{4593, 5303, 6013\}$. Unification of 102,240-cycle drift with carry threshold at $\{2463, 3173, 3883, 4593, 5303, 6013\}$.

Observed: Primes at scale (Appendix B). Meta-return at 73,996,200. Tau-Wave confirmation. Repeating decimal tail ratios $4/11$, $33/14$, $7/6$ with product 1 (Appendix C). Gelfond's constant $e^{\pi}$ as natural crossing scale. $\sin(1)$ convergence.

Interpretive: Symbolic Necessity. Gödel complement.

Not claimed: That $\tau$ is uniquely privileged (open question). That these patterns have physical realisations. That Gödel is "overcome." That the primes are causally generated.

11.3 Falsification tests

1. Replace $\tau$ with $\phi$, $\sqrt{2}$, $e$. If equally coherent hierarchies appear, the principle is universal.
2. Change base. Replace mod 9 with mod 7 or mod 11.
3. Verify the exact return for all 6390 values and the three failure indices at $a = 26$.
4. Verify the 102,240 drift at exactly $\{2463, 3173, 3883, 4593, 5303, 6013\}$.
5. Straighten coordinates. Replot tongues in $\alpha = \tau/q$.

11.4 Open problems

1. Is the preserved per-unit error generic to all irrationals with good convergents?
2. Can the 8-and-7 phrase structure be rigorously derived from partial quotients?
3. What determines the phase-transition index at $n = 716$–$717$?
4. What is the precise observable for the 73,996,200 meta-return?
5. Are the arithmetic and wave projections functorially related?

---

12. Conclusion

We began at zero. The sequence $\lfloor n/\tau \rfloor \bmod 9$ produces blocks of length 6 and 7, distributed as a Sturmian word. The internal architecture — 8 phrases, a break, 7 phrases, a break — reflects the continued fraction $710/113 \approx \tau$. The column sums of the 6390-cycle converge to $4068$ and $25{,}560$, connected through $\tau$. The terminal indices are prime.

The central result is exact recurrence: a renormalized 6390-window returns with pointwise identity at $q = 1{,}308{,}519$, persists for 25 multiples, and breaks at the 26th at indices $\{4593, 5303, 6013\}$ — spaced by 710. These same indices appear independently in the empirically discovered 102,240-cycle drift, where $f(n) \ne f(n + 102{,}240)$ at $\{2463, 3173, 3883, 4593, 5303, 6013\}$. Both are instances of the carry-threshold mechanism operating on the same 710-lattice. The convergence of two independent analyses on the same list of integers is the strongest evidence that the mechanism is real.

The same architecture reappears in the Tau-Wave system (confirming the 73,996,200 return) and in the complex sine family (where 710 serves as the natural comparison lattice and rational locking produces spoke counts).

We call this Symbolic Necessity: the principle that finite systems can reference structures they cannot construct, and that the bounded projections can be exact.

Exactness need not come from global periodicity. It can come from bounded renormalized observation of an irrational source.

---

Appendix A: Computational Verification

import math
tau = 2 * math.pi

def build_blocks(N):
    blocks, val, start = [], math.floor(0/tau) % 9, 0
    for n in range(1, N + 1):
        v = math.floor(n/tau) % 9
        if v != val:
            blocks.append({'length': n-start, 'value': val, 'start': start, 'stop': n-1})
            val, start = v, n
    if start < N:
        blocks.append({'length': N-start, 'value': val, 'start': start, 'stop': N-1})
    return [b for b in blocks if b['length'] > 0]

# Proposition 1
b710 = build_blocks(710)
assert len(b710) == 113
assert sum(1 for b in b710 if b['length']==6) == 81
assert sum(1 for b in b710 if b['length']==7) == 32
lengths = [b['length'] for b in b710]
phrase, brk = [7,6,6,7,6,6,6], [7,6,6,6]
i, p, k = 0, 0, 0
while i < len(lengths):
    if lengths[i:i+7] == phrase: p += 1; i += 7
    elif lengths[i:i+4] == brk: k += 1; i += 4
    else: raise ValueError(f"Unexpected at {i}")
assert (p, k) == (15, 2)

# Theorem 1
seven_pos = [i for i,b in enumerate(b710) if b['length']==7]
spacings = [seven_pos[i+1]-seven_pos[i] for i in range(len(seven_pos)-1)]
assert spacings == [3,4]*8 + [4] + [3,4]*7

# Theorem 2
b6390 = build_blocks(6390)
assert len(b6390) == 1017
assert sum(b['value'] for b in b6390) == 4068
assert sum(b['value']*b['length'] for b in b6390) == 25560
assert sum(b['start']%9 for b in b6390) == 4068
assert sum(b['stop']%9 for b in b6390) == 4068

# Theorem 3
eps = 6390/tau - 1017; q2 = 1308519
def y(n, q): return math.floor((n+6390*q)/tau) - 113*math.floor((n+6390*q)/710)
for n in range(6390): assert y(n,q2)-113 == y(n,0)
for a in range(1,26):
    for n in range(6390): assert y(n,a*q2)-113*a == y(n,0)
f26 = [n for n in range(6390) if y(n,26*q2)-113*26 != y(n,0)]
assert f26 == [4593, 5303, 6013]
assert f26[1]-f26[0] == f26[2]-f26[1] == 710

# Section 6: 102,240-cycle drift
drift = [n for n in range(6390)
         if math.floor(n/tau)%9 != math.floor((n+102240)/tau)%9]
assert drift == [2463, 3173, 3883, 4593, 5303, 6013]
assert all(drift[i+1]-drift[i]==710 for i in range(len(drift)-1))
assert set(f26).issubset(set(drift))

---

Appendix B: Complete Prime Catalogue

The following is a systematic record of all prime candidates found by evaluating $\lfloor c \times 10^n / \tau \rfloor$ and related expressions, with exponents tested from $n = 1$ to $9999$.

B.1 Primes from $\lfloor 1/\tau \times 10^n \rfloor$

Exponent $n$	Expression
71	$\lfloor 10^{71}/\tau \rfloor$
90	$\lfloor 10^{90}/\tau \rfloor$
155	$\lfloor 10^{155}/\tau \rfloor$

B.2 Primes from $\lfloor \tau \times 10^n \rfloor$

Exponent $n$	Expression
344	$\lfloor \tau \times 10^{344} \rfloor$
382	$\lfloor \tau \times 10^{382} \rfloor$
521	$\lfloor \tau \times 10^{521} \rfloor$
3779	$\lfloor \tau \times 10^{3779} \rfloor$
5754	$\lfloor \tau \times 10^{5754} \rfloor$

B.3 Primes from $\lfloor 71 \times 10^n / \tau \rfloor$

Exponent $n$	Value	Notes
7	113,000,009	$= \lfloor 710 \times 10^6/\tau \rfloor$
262	(large prime)
3933	(large prime)

The exponent gaps are 255 and 3671.

B.4 Primes from $\lfloor 73{,}996{,}200 \times 10^n / \tau \rfloor$

Exponent $n$	Notes
0	$11{,}776{,}861$ (prime)
5
74
193
282
775

Equivalently expressed as $\lfloor 739{,}962 \times 10^{n+2}/\tau \rfloor$ at $n+2 = 2, 7, 76, 195, 284, 777$.

B.5 The prime chain in $710/\tau$

$$710/\tau = 113.00000959524569\ldots$$

Reading the integer part left to right:

Truncation	Value	Prime?
1 digit	1	unit
2 digits	11	prime
3 digits	113	prime
9 digits	113,000,009	prime

Additionally, $\lfloor 73{,}996{,}200/\tau \rfloor = 11{,}776{,}861$ is prime, and $1{,}177{,}686{,}100{,}001$ is prime.

B.6 The prime-factor scaffold

All cycle-length and block-count numbers decompose into the same small set of primes:

Cycle lengths	Factorisations	Block counts	Factorisations
710	$2 \times 5 \times 71$	113	prime
6,390	$2 \times 3^2 \times 5 \times 71$	1,017	$3^2 \times 113$
25,560	$2^3 \times 3^2 \times 5 \times 71$	4,068	$2^2 \times 3^2 \times 113$
102,240	$2^5 \times 3^2 \times 5 \times 71$	16,272	$2^4 \times 3^2 \times 113$

The observation that 1017, 4068, and 16272 are composed exclusively of the primes $\{2, 3, 113\}$, while 6390, 25560, and 102240 are composed of $\{2, 3, 5, 71\}$, means the entire hierarchy rests on four primes: $\{2, 3, 5, 71\}$ and $\{113\}$.

---

Appendix C: Repeating Decimal Structure

C.1 Setup

The decimal expansion of $710 \times 10^y / \tau$ for large $y$ eventually settles into a non-repeating prefix followed by a repeating tail. Three cases were examined:

$y$	Non-repeating prefix	Repeating tail (35 digits)
6	$\ldots 15915494225352112676056338028$	$\overline{16901408450704225352112676056338028}$
261	(268 digits, begins $15915494309\ldots$)	$\overline{46478873239436619718309859154929577}$
3932	(3939 digits, begins $15915494309\ldots$)	$\overline{19718309859154929577464788732394366}$

C.2 The tail ratios

Let $a$, $b$, $c$ denote the three 35-digit repeating tails read as integers. Then:

$$\frac{a}{b} = \frac{4}{11}, \qquad \frac{b}{c} = \frac{33}{14}, \qquad \frac{c}{a} = \frac{7}{6}.$$

These are exact rational ratios of small integers. Their product:

$$\frac{4}{11} \times \frac{33}{14} \times \frac{7}{6} = \frac{4 \times 33 \times 7}{11 \times 14 \times 6} = \frac{924}{924} = 1.$$

This is required by the cyclic structure ($a/b \times b/c \times c/a = 1$), but the individual ratios being clean fractions of small integers is not forced. The prime content of the ratios is:

$$\frac{4}{11} = \frac{2^2}{11}, \qquad \frac{33}{14} = \frac{3 \times 11}{2 \times 7}, \qquad \frac{7}{6} = \frac{7}{2 \times 3}.$$

Every prime appearing in the numerator of one ratio appears in the denominator of another. The primes involved are $\{2, 3, 7, 11\}$ — all small, and $7$ and $11$ echo the $71$ and $113$ that structure the block hierarchy.

C.3 The exponent spacing

The three $y$-values at which these tails were observed are $6, 261, 3932$, with gaps $255$ and $3671$. These are the same exponents at which $\lfloor 71 \times 10^{y+1}/\tau \rfloor$ yields primes (Section B.3: $n = 7, 262, 3933$). The repeating-tail structure and the prime-generating property occur at the same scales.

---

References

1. Steinhaus, H. (1957). One Hundred Problems in Elementary Mathematics. Problem 57.
2. Morse, M. and Hedlund, G. A. (1940). "Symbolic dynamics II: Sturmian trajectories." Amer. J. Math., 62(1), 1–42.
3. Lothaire, M. (2002). Algebraic Combinatorics on Words. Cambridge.
4. Hardy, G. H. and Wright, E. M. (1979). An Introduction to the Theory of Numbers. 5th ed. Oxford.
5. Gödel, K. (1931). "Über formal unentscheidbare Sätze." Monatshefte Math. Phys., 38, 173–198.
6. Lindemann, F. (1882). "Über die Zahl $\pi$." Math. Annalen, 20, 213–225.
7. Khinchin, A. Ya. (1964). Continued Fractions. U. Chicago Press.
8. Arnol'd, V. I. (1961). "Small denominators. I." Izvestiya AN SSSR, 25(1), 21–86.

---

Correspondence: i.am@timothysolomon.com

Developed through human reseach and perseverance AI-collaborative writing and code generation (2024–2026), with all mathematical claims verified by computation.