Canons, numbers, and the Pythagorean table: an Italian method for composing

The canon is often described as an exercise in contrapuntal skill, a game of mirrors designed to display the composer’s technical mastery. Yet within the Italian tradition, it has been much more than that: a generative principle, a way of musical thinking grounded in proportion and transformation.

Seen from this perspective, the canon is neither an isolated artifice nor a puzzle for specialists, but an operational tool linked to compositional practice and musical training. Its logic is rooted in a proportional conception of music that spans centuries, from Renaissance masters to the Neapolitan pedagogy of partimenti.

The canon beyond the puzzle

In current manuals, however, the canon is often presented as an exercise in contrapuntal skill, a display of technical mastery, sometimes an elegant pastime for mathematical minds. Northern examples are frequently cited, while those of the Italian Trecento and the Stilnovo (in the musical sense proposed in our periodization) are overlooked; emphasis is placed on the combinatorial art of the Flemish school and its admired symmetry.

Reducing the canon to a game of mirrors, however, misunderstands its deeper nature. The canon is not merely strict imitation, but a generative principle — a structural machine capable of producing music from a numerical rule. It represents a Pythagorean way of thinking about composition.

Viewed in this way, the canon does not belong to a single historical season but moves across centuries and schools. In Italy, and particularly in the tradition that extends from seventeenth- and eighteenth-century masters to the Neapolitan environment, it takes on a distinctive character. It is not abstract speculation but a pedagogical tool and a creative exercise.

Italian musical culture never conceived counterpoint as pure theoretical speculation, since it always remained tied to practice and singability. Within this tradition, the canon is not an immobile monument but living material: adaptable, manipulable, even ironic.

This perspective helps explain the remarkable inventiveness of masters such as Nicola Sala. In one of his best-known examples, he constructs an augmented canon that appears destined to expand indefinitely, yet is structurally impossible to continue beyond a certain point. The rule is perfectly formulated; the effect is seductive, because it appears to be a true augmented canon, while in reality it is a “false” canon that eventually introduces free voices.

This is not an error but a conscious gesture, a sign of a culture that knows its rules so deeply that it can move beyond them. The canon thus becomes not only a demonstration of technical mastery but also a happy invention of form, time and proportion.

From this perspective, the canon is not an enigma to solve but a living laboratory of musical thought. The Italian tradition, too often described as merely a vehicle of a generic “European style,” reveals instead a remarkable capacity for structural invention.

From Renaissance polyphony to partimento pedagogy, the canon in Italy was never treated as a relic to be followed mechanically, but as an operational tool allowing wide creative freedom.

Canon and partimenti: a generative grammar

In the Italian tradition, the canon has its roots in the practice of partimenti.

In Neapolitan conservatories, students did not learn counterpoint as an abstract system of rules, but as the capacity to generate music from a structural framework made of numbers and governed by keys. The guided bass of the partimento forced students to think in terms of proportion, balance, and anticipation. It was a mental gymnasium, a testing ground for verifying the coherence of musical discourse. Writing a canon means demonstrating that a line made of numbers can sustain duplication, transformation, and temporal displacement without losing coherence.

In this sense, the canon shares deep affinities with the partimento, since both presuppose an underlying structure that must withstand transformation. If the partimento bass lacks numerical solidity, improvisation collapses. If the canonical line is not conceived with proportional rigor, the construction breaks.

The difference from certain modern manuals is evident. Today the canon is often presented as a typological category — canon at the unison, at the fifth, in inversion, augmented, diminished. Useful classifications, but insufficient. In Italian practice, the issue was not classification but making the mechanism work.

Here we see the difference between combinatorial artifice and musical thought. An artifice may be theoretically perfect yet musically sterile. Musical thought, by contrast, knows when to stop, when proportion ceases to be balance and becomes imbalance.

The Italian tradition consistently privileged coherence over riddles, structural functionality over mathematical virtuosity for its own sake.

For this reason, the Italian canon does not remain confined to treatises. It appears in sacred music, motets, fugues, and opera — a latent principle even when not explicitly declared. It is, in short, a way of thinking.

When everything is reduced to a generic “European koiné” or a chronological label, this dynamic element disappears. The canon becomes a chapter in a manual rather than a living formative device.

The Italian tradition instead uses it as an operational tool, which explains its extraordinary adaptability. Proportion in Italian practice is not rigid; it is plastic, adapting to context, genre, and function.

It is no coincidence that many Italian masters integrated contrapuntal rigor with theatrical naturalness, a hallmark that remains unmistakably their own.

A concrete example

This is not a theoretical abstraction. In the video below, you can hear a canon constructed according to the principle described in this article. The lines are generated from a handful of chords borrowed from a composition by Palestrina, transformed from vertical structures into diagonals that produce a real canon.
One can observe how the horizontal structure already contains vertical relationships and how the “diagonal” reading generates imitation. Nothing was added from outside: polyphony emerges from the system itself, like Venus from the sea.
Readers wishing to explore the detailed procedure — from matrix construction to numerical translation into intervals and the control of overlaps — can consult the full paper linked below, where the method is explained step by step.
The paper includes the matrix used for examples like this one, with number-to-interval conversion. The paper includes the scheme/matrix used for examples like this one, with the conversion numbers→intervals.

Five-voice canon

But how does one actually write a canon?

At this point, the question is inevitable: history, the Italian tradition, Sala’s inventiveness — all well and good. But concretely, how does one write a canon?

There is no magic formula, only a basic principle: proportionality.

A canon comes into being when a line is conceived in such a way that it can be duplicated according to a precise rule: temporal delay, fixed interval, inversion, augmentation, diminution. The initial line must contain, numerically within itself, the possibility of its own transformation.

This means the composer does not think only horizontally, but simultaneously in several directions. A melody is written, yet its future echo is already heard, its displaced entry is foreseen, its overlap is imagined.

And this is where the Pythagorean table — the one we learn in elementary school — becomes an operational tool.

Numbers, matrices, proportions

The Pythagorean table is not a symbolic object from a philosophy handbook. It is a numerical matrix. Every row and every column is generated by a multiplicative law. Every number is a relation.

If we translate this principle into music, every interval becomes a ratio — and it seems made for canons. Every melodic motion becomes proportion, read within the numbers of the table. Every overlap becomes numerical verification.

In a Pythagorean matrix, the horizontal already contains the vertical. Why?

Because every element of a row is simultaneously an element of a column. The system is not linear: it is bidimensional. And if we observe it closely, it is also tridimensional, because the diagonals introduce an additional level of relation.

What do we mean by “diagonal”?

Not only the canon in the strict sense. The diagonal is the path that relates distant elements through a coherent progression. It is the trajectory that turns succession into superposition.

In the canon, the diagonal is obvious: the second voice enters after a certain number of units, retracing the same path. But the diagonal is also present in other forms of polyphony: free imitation, sequential progressions, mirror structures.

The strength of the Pythagorean system is this: the horizontal structure automatically generates vertical and diagonal possibilities. There is no need to add complexity; complexity is already implicit in the matrix.

The Pythagorean table “sounds” for precisely this reason. It is not a static diagram. It is a network of proportions that can be translated into intervals, delays, and overlaps. If coherent durations and pitches are assigned to numerical relations, the system produces polyphony.

With such an framework one can write not only a canon, but as much polyphony as one wishes: fugues, imitations, augmented or diminished canons, multi-voice interweavings. The matrix does not impose a single outcome; it provides a coherent grid.

This also clarifies the structural limit observed in certain augmented canons: if proportion grows without compensation, the matrix loses balance. This is not a defect of the composer; it is a property of the system.

The Italian tradition never ignored these mechanisms. It practiced them, adapted them, transformed them — not as exercises in musical arithmetic, but as tools for sonic construction.

To write a canon therefore means this: to conceive a line that is already a matrix. To think of time as space. To think of succession as potential superposition. To think of melody as a number in motion.

And when number and sound coincide, proportion becomes music.

Whoever invented number invented polyphony

If the Pythagoreans represented sounds as numbers, they did not merely measure intervals: they laid the foundations of polyphony.

Because polyphony does not begin when someone, in the Middle Ages, decides to “add a second voice.” Polyphony is already implicit in the proportional system. The moment sound is defined as a numerical ratio, any ratio can be superposed on another ratio. Simultaneity is already possible. Verticality is already contained in horizontality.

As above, so below. And we might add: in diagonal as well.

The Pythagorean matrix is not a simple linear succession. It is a relational structure. Every element is linked to the others by multiplication, proportion, symmetry. If we assign pitches and durations to these relations, the structure begins to sound.

Who would have imagined that the elementary-school multiplication tables could become polyphony?

And yet, that is what happens. The horizontal generates melody. The vertical generates harmony. The diagonal generates canon, imitation, fugue. These are not later inventions: they are natural developments of a proportional matrix.

The Italian tradition was able to read this potential better than others. In partimenti, in canons, in fugues, proportion becomes a sonic gesture. It is not an abstract exercise: it is living practice.

Not convinced?

Then try it yourself. Take a simple sequence built on regular ratios. Shift it in time. Superpose it at a fixed distance. Check that the intervals remain coherent. You have written a canon.

You have done nothing more than activate a matrix.

This is exactly what Italian masters did: they did not start from chance, but from a structure. And within that structure they left room for imagination. Sala can “fake” an augmented canon; he can push proportion to the structural limit, because the system allows him to do so.

Polyphony is not an episodic invention of history. It is a property of number applied to sound.

When number enters music, music becomes multiple.

And from that moment, the canon is no longer an enigma to decipher, but a natural consequence.

Aneddoto

A compositional machine

“The Pythagorean table is not a schoolroom memory. It is a compositional machine.”

What for centuries was taught as a simple arithmetic tool becomes, in the hands of those who can listen, a sonic structure. Proportions do not remain on paper: they turn into intervals, overlaps, canons. The numerical matrix does not merely describe music. It generates it.