Extensions

The parameter n

Every definition and theorem in the formal development is stated for G_n = ℤ/nℤ. Setting n = 12 gives the chromatic universe; the 223 TI-classes, the Forte numbers, and the familiar chords and scales are consequences of that particular choice. But the structural theorems are n-agnostic:

• The finite Wiener–Khinchin correspondence holds for any n.
• Periodic gap words have sparse Fourier support regardless of the ambient group size.
• The layered symplectic form ω_n,k is exact on ℂⁿ⁻¹ × ℂ^k for every (n, k).
• Transposition and cyclic reindexing remain Hamiltonian; inversion remains anti-symplectic.
• The total space is stratified by cardinality for any n.

What does change with n is the combinatorial landscape: the number of TI-classes, the density of the quotient graph, the fiber sizes, and the symmetry-order distribution all depend on the arithmetic of n.

Microtonal and alternative equal temperaments

Several equal temperaments beyond 12-TET have significant historical and practical use:

• n = 19 — 19-TET approximates major thirds and minor thirds more closely than 12-TET and has been used by composers from Costeley to Sethares.
• n = 24 — Quarter-tone tuning, widely used in Arabic maqam theory and 20th-century Western composition (Hába, Ives, Wyschnegradsky).
• n = 31 — 31-TET gives excellent approximations to 5-limit just intervals; Huygens and Fokker championed it. Its TI-class count and quotient graph are substantially richer than those of 12-TET.
• n = 53 — 53-TET approximates both Pythagorean and just major thirds to within a fraction of a cent, making it a theoretical touchstone for tuning theory.

In each case the layered construction applies without modification: compute the DFT on ℤ_n, enumerate TI-classes by Burnside, build rooted cyclic fibers, and thicken into the symplectic ambient space. The Forte-style classification counts change — but the structural theorems are identical. This connects directly to Quinn's "General Equal-Tempered Harmony," which develops a DFT-based framework that is already n-parametric.

Beyond equal temperament

The cyclic group ℤ_n is the load-bearing assumption. Departing from equal temperament means departing from finite cyclic symmetry, and with it the finite Fourier machinery that drives the entire content layer. Several important tuning systems fall outside this scope:

• Just intonation — Pitch classes are rational frequency ratios, not equally spaced points on a circle. The relevant group structure is a free abelian group generated by primes (2, 3, 5, …), not a cyclic group. A Fourier-based content layer would require harmonic analysis on ℤ^d or on the p-adic integers rather than on ℤ/nℤ.
• Non-octave-repeating scales — The Bohlen–Pierce scale divides the tritave (3:1) into 13 steps; Wendy Carlos's alpha and beta scales use irrational step sizes. These require replacing the cyclic group with other discrete subgroups of ℝ/ℤ or with quasicrystalline models.
• Continuous pitch — In the limit of arbitrarily fine resolution, the discrete group ℤ_n gives way to the circle group ℝ/ℤ, and the finite DFT becomes the classical Fourier series. The symplectic thickening generalizes to infinite-dimensional Hilbert-space settings, but the combinatorial fiber structure (finite permutations) would need a fundamentally different treatment.

Each of these extensions is mathematically natural but nontrivial. The present framework handles them only to the extent that they can be approximated by a sufficiently large n-TET.

Rhythm, time, and the second circle

Music has two fundamental circles: pitch (the octave) and time (the metric cycle). The same layered content/order architecture could potentially apply to rhythmic patterns on ℤ_n, where n now indexes time points modulo a metric period rather than pitch classes modulo the octave.

This is not merely speculative. Yust's Organized Time (already cited in GAMUT's bibliography) applies Fourier analysis to rhythm and meter, treating rhythmic patterns as subsets of a time circle and analyzing their spectral content. The layered extension would add an order layer — rooted cyclic orderings of onsets, gap words measuring inter-onset intervals, and a fiber bundle over rhythmic content classes — and then ask whether the resulting space also admits a natural symplectic structure. This direction is unexplored but structurally natural.

A still more ambitious extension would consider pitch-rhythm interaction: a product of two cyclic groups, ℤ_m × ℤ_n, encoding simultaneous pitch and onset information. The resulting layered space would have both content and order layers in each factor, with symplectic forms combining across dimensions.

Connections to other fields

Crystallography and phase retrieval

Homometric sets — two distinct subsets of ℤ_n with identical autocorrelation — are the Z-related pairs of music theory. In crystallography, the same phenomenon appears as the phase problem in X-ray diffraction: the diffraction pattern records only the squared magnitudes of the Fourier transform of the electron density, losing phase information. Two crystal structures can produce identical diffraction patterns yet have different atomic arrangements. The mathematical core is the same: autocorrelation determines the power spectrum but not, in general, the underlying structure.

Signal processing and engineering

The Wiener–Khinchin theorem and cyclic autocorrelation are standard tools in signal processing, communications engineering, and radar. Duncan's original paper appeared in the Journal of the Audio Engineering Society, not a music-theory journal — reflecting the fact that the mathematical content lives at the intersection of discrete harmonic analysis and applied engineering. GAMUT's symplectic layer adds a geometric interpretation that is new in the musical context but whose individual ingredients (action-angle variables, moment maps, Hamiltonian symmetries) are well-established in physics and engineering.

Combinatorics and group theory

The fiber construction is a special case of studying orbits and stabilizers under group actions on finite sets. The content layer is the orbit space of the dihedral group acting on subsets; the order layer is the orbit space of the cyclic group acting on permutations of a fixed subset. The Burnside counts, the stabilizer orders, and the stratification by cardinality are all instances of standard combinatorial group theory. What is new is the particular combination — content orbits, rooted cyclic order fibers, Fourier lifts of both — and the observation that this combination lands in symplectic territory.

Open problems and future directions

• Registral information and voice leading. The current framework is octave-equivalent: it does not distinguish between a C4 and a C5. Adding registral data would connect GAMUT to the orbifold chord spaces of Tymoczko and Callender–Quinn–Tymoczko, where voice-leading distance is the primary geometric structure. The challenge is to integrate this with the symplectic layer without losing the clean action-angle decomposition.
• Musically motivated Hamiltonians. The kinematics — the symplectic form, the symmetries, the reduction — are proved. But the framework does not yet contain a dynamics: there is no Hamiltonian whose flow models compositional preference, voice-leading cost, or harmonic tension. Defining such Hamiltonians, calibrating them against real repertoire, and studying the resulting flows is the principal open direction.
• Nonlinear spectral embeddings. The current content renderings use multidimensional scaling or linear projections. Diffusion geometry (Coifman–Lafon) and other nonlinear spectral methods could produce lower-dimensional embeddings that better preserve the graph topology of the content quotient.
• Computational scaling. Exact fiber enumeration is feasible through cardinality 5 (24 orderings per fiber); by cardinality 12 the fiber has nearly 40 million elements. The current implementation samples at most 96 per fiber. More sophisticated sampling, approximation, or algebraic strategies could extend the computational reach.
• Empirical validation. The geometric framework makes structural predictions (e.g., about the topology of fiber neighborhoods, the distribution of entropy across orderings, the spectral signatures of stylistically coherent pattern families). Testing these predictions against corpora of real compositions would ground the theory in musicological evidence.