The Editor’s Dictatorship

Art in edition and steps to democratise editorial processes

This essay builds upon ideas that were first tentatively put forward in my BA Notation Portfolio.

The role of an editor is that of an organiser, a decision–maker. No music can be communicated fully–formed, nor can it be transmitted from its conception to its reception completely intact. Suffice to say, as a form of music is transmitted, it is transformed and “corrupted”. Corruption is the process by which musical information is inevitably lost during its transmission, similar to how the meaning of a phrase differs when it is translated from one language to another.

For musical information to be transmitted, it must first be codified into a language. Music is not a language but to be transmitted it must first be codified into one. Any kind of communication cannot occur without some kind of codification, i.e. a set of rules being agreed upon that govern the significance and representation of a pattern of signals through media. An idea is never conceived in language as mental conception is beyond language—anyone who has ever attempted to express a complex thought in language knows this well—but to transmit an idea, the idea must first be serialised and codified into a language that represents a serialisation of the idea, at least in part. A musical idea is no different to any other idea: to be transmitted it must be formed into a language and the codified information transmitted in lieu of the idea itself. This is done in the hope that the receiver of this information understands the language and can interpret the information into an idea that could be represented in this way.

All “musicking”, a term coined by Christopher Small that includes all actions that contribute to the making of music, fall into this category of musical codification and transmission because multiple readings of the musical idea must be transmitted to various musical agents to create music. This extends not just to interpersonal musical transmission between, say, a composer and a performer, but also to musical transmission within oneself such as the storage and recollection of a musical idea which requires an act of serialisation as the music is played in the mind’s ear. Musical agents within and without a person require the idea to be transmitted and collaborated upon and as such require a codification, a language with which to represent the ideas of music.

However, any musical idea cannot be represented completely in a codified form. If music could be completely codified then it could simply be replaced by its codification which is far easier to manage and protect against corruption than the musical idea itself. By way of analogy, a painting is codified onto a canvas but the canvas and the paint act only as the media for the art. Moreover, the art is not contained in the canvas or the paint, nor is it in the arrangement of the paint onto the canvas, for the arrangement is only a codification of the art. For example, a replication of the painting is not a replication of the art but another work of art that shares a codification of paint on canvas with the first artwork. As the art is replicated it is transformed, perhaps imperceptibly, but transformed nonetheless. The art occurs only at the painting’s observation and the transmission of the paint–arrangement and art codification to the viewer. The same is true for music. There is no music contained in the paper nor music in the arrangement of ink on the page. These are only representations of a musical idea. The representations can only become music once they are “observed” i.e. transmitted to another musical agent.

We can see that as music has not been replaced, then it is clear that the representation of a musical idea is inferior and less accessible than a musical idea itself. It is simultaneously a blessing and a curse that all art must be corrupted on its journey from one person to another: cursed in its ambiguity and the frustration of endless generations of artists who must find ways to deal with this corruption and blessed in the knowledge that art cannot exist without first being transmitted. It is in this second axiom that music must lie as it is its transmission that we would most readily call “art”. Returning to the analogy of the painting, visual art can be appreciated at a glance or pored over for hours. Its transmission is frozen, save from alteration, destruction or defacement, and instantaneously accessible by observation of the art information. It does not require an external impetus to be transmitted and can be received passively by an observer as long as there is light, therefore a painting in light can be classed as “self–transmitting” art. A physical representation of music such as musical notation can also be self–transmitting: the reader only needs to read the language to receive the musical information contained therein. Music, unlike visual art, is portrayed most concisely in sound and so in this form cannot be self–transmitting like a visual representation. A sonic representation of music, i.e. the performative “end goal” of most musical endeavours must be actively transmitted and be communication between musical agents.

This idea allows us to draw four interesting conclusions, the first of which being the medium–agnostic nature of art, that nowhere in this system does the medium play a part in limiting the transmission of the art. If musical information can be written down in part by musical notation and commuted between media, then similarly it must be possible for a sound to represent a painting or, famously, an architectural plan to be communicated via interpretative dance. If a language can be decided upon to communicate these artistic ideas, then such events are possible. An art teacher can describe to a student exactly how he would like the student to paint without ever demonstrating the brushstrokes required by using verbal language to communicate his intent. This simple ad hoc protocol tunnels artistic idea through verbal language. However, the teacher cannot expect the student to reproduce his artistic idea exactly because, as we have seen, language is limited and cannot fully communicate the idea.

The second conclusion we must make is that what we could call the transmission of art is art, and art is therefore a process of transformation that transforms a representation of art into another representation of art. Therefore art is not singular but in fact infinitely replicable: transforming a representation of art copies the artistic idea in part into another representation and creates art at that moment of transformation. The art, lying dormant in whatever codified form it inhabits, is released, transmitted and transformed. Once the transmission is aborted—for it can never be completed—then the art ceases to exist because the moment of transmission was the moment of art. The Mona Lisa on display at the Louvre Museum transmits its artistic idea—its art—thousands of times each day, but an unknown painting kept in a cupboard never transmits its art. This idea is usually more complex with music but the principle is the same. For example, the original autograph score of Beethoven’s Ninth Symphony held in the Berlin State Library rarely transmits its musical idea except to scholars, but editions and copies of the musical representation transmit similar—but crucially not identical—musical ideas to thousands of musicians every day and each performance transmits copies of these musical ideas to audiences via sound.

The third conclusion we must make is that some understand of the language of art is required for a person to appreciate the art that is being experienced. Those who are are inexperienced in the language of painting may not be able to decode as much information from an artwork as a well–versed critic and those who are not musically trained may not be able to figure out as much meaning from a performance as a well–trained musician. These are obvious points but it is important to stress that those who are not as trained are just as deserving of the art but may not have the knowledge required to appreciate some of the finer points. As musical art is created in a performance, for example, the art is transmitted and transformed from both the notation and the memory of the performers into performance. The elusive musical idea is interpreted by the performers and then in turn interpreted by the listeners. Furthermore, each listener with their broad and fleeting states of mind and knowledge will interpret the performance in their own distinct way and gain a different concept of the musical idea from the musical information that they have heard.

Finally, as art transmission is our definition of art, our definition of the the actions that create art must broaden significantly to encompass all musicking actions, those actions that transform artistic representations. Art is created each time an artistic representation is transformed into another representation and so each of these actions create art. Transformations that transmit the musical idea without much corruption will create art that is much more similar to the original art than a transformation that corrupts the idea to a greater extent or introduces external influences and ideas into the art.

Flows of transformation in music

Any act that transforms musical representations is an act that creates art. Musical transformations and therefore art–creating acts could be anything: improvising, composing, altering, notating, editing, transcribing, transposing, dictating, practising, destroying, parodying, reproducing, typesetting, listening, thinking, writing, reversing, extending, bowdlerising, perverting, critiquing, studying etc. At each stage, new art is created from the transformation of a representation of old art into another representation and when the transformation stops, the art stops.

It is often assumed that in most circumstances, the flow of music is a simple one–way movement of musical information from composer through editor to performance (Figure 1). However the reality is much more complex. Flows of transformation work both ways and in feedback loops, shuttling back and forth through the system until they come to an end (Figure 2). In this fictional and simplified example the musical idea, which for simplicity’s sake originates at the composer, can travel through the system to any one of three endpoints. Each node represents a state of musical codification and each arrow a possible transformation of the music from one state to another. The idea could travel straight to the bin or loop hopelessly around the composer’s mind and through multiple drafts, never achieving a performance. Any one transformation is always effected by an external force onto the system, for example a desire by a musician to improve the music in their view. In each transformation, the musical idea travels through mental, oral and literal mediums. However, it is not fair to say that the art ends at performance. In reality the musical idea in the listener’s mind may go on to influence and create new art.

Figure 1: The common model of edition flow

Figure 2: A real–world model of edition flow

The music that is conceived in the composer’s mind has undergone such a radical transformation by the time that it is, say, performed that the musical idea received by the composer at the première may seem alien to his mind and not strictly a work of his making. The composer would be right to make this assertion: as soon as the information contained in the representation of the music (say, the composer’s manuscript) left his hands for the editor, it was instantly transformed to a different but related music by the interpretation of the music by another person. Similarly, when the final copy is interpreted by the performer, the conception of the music in the performer’s mind will be different to that of the composer’s conception.

Along each stage of this process, corruption occurs to varying degrees. The example mentioned above introduce a relatively large amount of corruption as the music repeatedly bridges the gap to and from marks on paper and musical idea. In this, the codification of music drastically alters the musical idea in the same way that an idea is altered by its explanation in verbal language. It is true however that music must be codified in order to be transmitted. A system is required to codify music for transmission. It is not fair to say that the codification is music, rather that it is a code that represents a selection of possible musics. Indeed, some codifications introduce very little corruption into a musical idea. An high–fidelity audio recording, for example, can capture and replicate a musical performance fairly accurately, but is is important to note that even the best recording has peculiarities of timbre and depends on how the microphones in the room were mixed together. Besides, no–one except snake oil hi–fi manufacturers would claim that a sound system playing a recording can perfectly represent the original sound source.

As corruption is inevitable, any method that transmits music or its representation must transform and corrupt it. It is important to note that these corrupting transformations do not mar the “pure” musical idea that was first conceived in the mind of the composer, on the contrary these stages are necessary to add to the musical idea. The earliest idea, although uncorrupted from edition, is only a small kernel of musical thought and requires the external input that comes with multiple iterations of transforming processes in order to be improved, augmented and to become a larger musical idea. As the music is transformed, it is simultaneously corrupted by the process as well as improved and altered by the external inputs.


The process of editing transforms previously corrupted representations of musical acts into other, even more corrupted representations. To argue that a recent edition is less corrupted than its source material is to argue against the second law of thermodynamics. Even if there are only two steps

Steps from A→B→C

the corruptions that occurred on the path from A→B cannot be reversed in the path from B→C without knowledge of A. If C has knowledge of A, then there can be one step from A→C with one step’s worth of corruption. Without any more input to the system, with each step information is lost and corrupted as it is transformed. The amount of corruption which occurs in each step varies depending on the purpose, quality and medium of the edition but it can never be undone.

Counterintuitively, the editor’s aim is not solely to attenuate this effect as one might expect but to manage corruption along with other considerations, sometimes placing great emphasis on minimising corruption that is introduced and at other times artificially inflating the corruption introduced in order to solve other issues but accepting that regardless of the decision, overall corruption will increase. The editor, decision–maker and organiser, must make decisions in the creation of his art, the edition, to organise and balance the extending and corrupting influences that may make their way into the edition. In this way, the editor is often a point of dictating control between musical ideas which ultimately affects hot the musical ideas communicated onwards to performance and the interpretation in the minds of the listeners. This is a problem because one bad editor or bad edition is a weak point in the transforming processes that could adversely affect downstream musics.

Democratic editions

Ideally musical edition would have routes to collaboration, like performance and composition, but edition–making is currently a task for one person. This is because editing is seen largely as an academic exercise. However, as we have seen above, editing is musical because in dealing with music, it creates art.

My endeavours recently have been on creating the infrastructure to begin to democratise the creation of editions by splitting the edition–making process in two: the “traditional” editor edits the music for a wider range of possibilities in a dynamic format that allows for a narrowing of scope further downstream. When the edition is required to be narrowed, the possibilities of edition are narrowed by the user to create a singular notation. This broadening of scope allows the editor to shift the burden of balancing corruption further downstream to the musicians who will be most affected by it, allowing them to choose how to transform and corrupt the musical representation best to suit their needs.