The Composition That Refuses to Become Repertoire

There is a point in the life of a composition when something dangerous can happen.

It becomes repertoire.

Usually this is considered success. The piece survives its premiere. Another performer learns it. A recording circulates. Difficult passages acquire accepted solutions. Ambiguities become traditions. Eventually the work stops being approached as a problem and starts being approached as something people know how to play.

That is when the piece begins to stabilize.

Repertoire is not simply a collection of works that survive. It is a system for converting uncertainty into repeatability. What once seemed strange becomes stylistic. What once seemed impossible becomes technical. What once demanded interpretation becomes precedent.

For some music, that is a form of neutralization.

I am increasingly interested in compositions that resist this process.

Not works that avoid repeated performance. That would be easy. The more interesting possibility is a piece that invites repetition while refusing convergence.

The first performance should not clarify the work. It should complicate the second.

The second should not improve on the first. It should reveal another contradiction.

After ten performances, we should not possess a clearer model. We should have ten incompatible pieces of evidence.

This requires a different relationship between score and history. Every performance normally leaves behind useful residue: recordings, annotations, fingerings, rehearsal methods, practical solutions. Over time, these become tradition.

A repertoire-resistant work would make that knowledge unreliable.

Previous solutions might not transfer. One performer's answer might become another performer's obstacle. Instructions might conflict. Decisions might be irreversible. The piece might change according to who performs it, where it is performed, or what has already happened.

The goal is not variation.

The goal is to prevent interpretive convergence.

Difficulty alone cannot do this. Difficult music eventually produces specialists. Specialists create technique. Technique becomes pedagogy. Pedagogy creates repertoire.

The more radical problem is to create a work that cannot be mastered simply through accumulated expertise.

The performer might become intimate with the piece without ever possessing it.

The work retains leverage.

Recording creates another danger. A recording can manufacture authority. Tempo, articulation, balance, character, even mistakes begin to harden into expectations simply because one interpretation circulates more widely than another.

For a composition that refuses repertoire, there should be no definitive recording.

A recording should be evidence of one event. Nothing more.

Ideally, each recording would make the others less authoritative.

This also changes the question of identity. Perhaps a composition does not need to remain identical through its sound.

Its identity might reside in a conflict.

A procedure.

A constraint.

A particular crisis imposed upon the performer.

The surface can change while the underlying problem remains.

I am also interested in works that are altered by being performed. A score might accumulate annotations, lose sections, acquire new instructions, or change according to previous realizations.

The piece would acquire scars.

Normally the score survives and performances disappear.

Here, performance changes the supposedly permanent object.

Repetition consumes it.

Such a work would not be retrieved from storage and played again. It would have to be reactivated.

That is a different model of musical survival.

A composition does not have to preserve itself by becoming standardized.

It might survive by continually resisting standardization.

It could accumulate history without becoming governed by history.

It could attract performers without allowing them to inherit solutions.

It could remain recognizable without becoming predictable.

Repertoire preserves a work by stabilizing it.

Another kind of composition might preserve itself by refusing stability.

It would not ask to be remembered correctly.

It would ask to remain unresolved.