The Tonnetz maps pitch classes as a hexagonal lattice where proximity represents voice-leading efficiency. Navigating the Tonnetz with minimal distance reveals optimal triadic paths preserving common tones. This spatial representation explains voice-leading logic in functional harmony and triadic post-tonal music.
Plot triadic progressions on a Tonnetz diagram and measure voice-leading distance. Compare efficient Tonnetz paths to voice-leading smoothness in scores; analyze how composers balance efficiency against other structural goals.
From your study of Tonnetz pitch space and neo-Riemannian operations (P, L, R), you know the Tonnetz as a hexagonal lattice where each node is a pitch class, and each triangle represents a major or minor triad. Adjacent triangles share an edge, which means they share two common tones — and the P, L, R transformations are exactly the moves that flip between adjacent triangles. The key insight for *navigation* is that every sequence of triadic transformations traces a *path* through this lattice, and the length of the path — measured in number of transformation steps or common tones lost — is a precise measure of voice-leading efficiency.
Voice-leading efficiency between two triads is minimized when the individual voices move by small intervals (ideally semitones or common tones). On the Tonnetz, this maps to spatial proximity: two triads that are close on the lattice share more common tones and require smaller voice movements. The P transformation (parallel, e.g., C major ↔ C minor) shares two common tones and moves one voice by semitone — the most efficient possible transformation, appearing as a flip across an edge. The L transformation (leading-tone exchange, e.g., C major ↔ E minor) also shares two common tones. The R transformation (relative, e.g., C major ↔ A minor) likewise shares two tones. Composite transformations like PL or PLPL trace longer paths with more cumulative voice movement.
From your graph theory prerequisites, you can now read Tonnetz navigation as a shortest path problem. Given two triads as start and end nodes, the neo-Riemannian shortest path (the sequence of P, L, R operations with fewest steps) corresponds to the most parsimonious voice-leading route. This is the parsimony principle: the voice leading between two triads is as efficient as their Tonnetz distance allows. Schubert's "Drei Klavierstücke" and Brahms's late piano music are rich with Tonnetz paths that traverse regions of the lattice far from any tonal center, but with impeccably smooth voice movement — chains of L and R operations that slide through remote harmonic territory without a single large melodic leap.
Importantly, the Tonnetz makes visible why certain harmonic progressions that seem distant in tonal theory are actually voice-leading neighbors. C major to A♭ major is not a standard functional progression (it's not a chord built on a diatonic scale degree), but on the Tonnetz it's just two steps — sharing the pitch class C and E♭ — with one voice moving by semitone. This explains the characteristic sound of chromatic mediant relations in Romantic music: the smooth voice leading keeps the progression from sounding disruptive even though functional grammar doesn't sanction it. The Tonnetz reveals the *geometric logic* behind these progressions.
The limitation you must keep in mind is that Tonnetz proximity describes voice-leading efficiency, not musical meaning or function. A composer may choose an inefficient path for expressive reasons — a sudden distant modulation can create a dramatic rupture precisely because it violates parsimony. And Tonnetz analysis applies most naturally to triadic, harmonically saturated music; it doesn't straightforwardly extend to seventh chords, non-triadic sonorities, or pitch-class set contexts where the lattice structure breaks down. Used within its domain, however, the Tonnetz is a powerful tool for showing that voice leading has geometric structure — that musical space can be mapped, and that composers navigate it with measurable efficiency.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.