What's in a Note?

Author's note: This essay grew among others from my recent years of experience in playing and singing in a Tango Orchestra, which required me to learn much more about music notation than I had previously needed. I learnt a lot, had a lot of thoughts about what I learnt, and really wanted to write them down. While writing, I used OpenAI's ChatGPT (GPT-5.6 Sol) as a sparring partner to get the facts right (as much as possible at least) and the structure sound, as well as do a lot of the hard work of actually writing.

On notation, interpretation, and what music asks of musicians

A note on a page seems like a precise thing. It tells us what pitch to play and roughly how long that pitch should last in relation to the notes around it. We can add markings for dynamics, articulation, tempo, phrasing, bowing, fingering and increasingly detailed technical instructions. The more information a score contains, the more tempting it becomes to think that the music itself has now been completely described.

But imagine that Frank Zappa had lived before sound recording. During a concert he improvises one of his guitar solos and someone carefully transcribes it note for note. The performance disappears, but the paper survives. Two hundred years later, guitarists study that same solo. Conservatories teach it, musicologists compare editions and playing a wrong note can cost you points in a competition. Something about this immediately feels strange because we know what improvisation is. Zappa might have played a completely different solo the following evening. The transcription could therefore be perfectly accurate while still giving us a misleading idea of what kind of musical object it preserves. It tells us exactly what someone played at one particular moment, but not necessarily that somebody else should later play exactly the same thing.

That makes an apparently simple question surprisingly interesting: what is actually contained in a note?

When an improvisation becomes a score

With Keith Jarrett, we do not even need to invent the thought experiment. His famous The Köln Concert from 1975 was improvised. The music came into existence while he was playing and survived because the performance was recorded. An authorized transcription was published later.

That reverses the direction of the creative process. For Jarrett, the notes were the result of improvisation. For the pianist who later buys the transcription, those same notes become the starting point. What was a musical decision for one musician becomes an instruction for another. Jarrett himself pointed to the limitations of this process. In connection with the authorized transcription, he made clear that the recording should remain the ultimate reference, precisely because some aspects of what happened in Cologne resist complete representation in conventional notation.

That does not make the transcription less valuable. It allows us to study and reconstruct an extraordinary improvisation. But it does show that the same black dot on five lines can represent different kinds of information. Sometimes it means something close to “this should be played.” Sometimes it means “this is what somebody played on this occasion.” The note itself does not tell us which one it is.

What did not need to be written down

Musical notation is not a universal language that appeared in finished form. It developed over centuries. Alongside pitch and rhythm, musicians gradually acquired more ways to indicate dynamics, articulation, tempo, pedaling, bowing and other details.

That does not mean musicians had not been doing those things before they were routinely notated. Often something did not need to be written because an experienced musician within a particular tradition already knew what was expected. The score worked together with knowledge that existed outside the score.

Even familiar symbols have not always been as unambiguous as they appear in modern teaching. Take the hairpins < and >. A student today will usually learn these as crescendo and diminuendo. Research into nineteenth-century performance practice, including David Hyun-Su Kim’s work on Brahms, suggests that in some contexts such markings belonged to a broader expressive gesture in which timing could also play a role. This does not mean that < simply meant “speed up” and > meant “slow down.” It means that a symbol we now associate strongly with one parameter could evoke a richer set of performance conventions for musicians within another culture.

Mozart presents the same problem in a different form. Improvisation was part of the professional skill set of an eighteenth-century keyboard player, and Mozart was renowned for it. In his piano concertos, cadenzas and shorter Eingänge could provide places where the soloist supplied material. Ornamentation was also shaped by conventions that a competent musician was expected to understand. Historical fidelity can therefore mean two things that overlap strongly without being identical: reproducing as precisely as possible what Mozart wrote, and understanding as well as possible what Mozart expected a good musician to do with what he wrote. The score is essential evidence, but that evidence existed within a performance practice.

When music became a sellable object on paper

There is another historical reason why written music acquired such unusual authority. For centuries, notation was one of the few practical ways to detach a musical creation from the physical presence of the musician, reproduce it at scale, transport it over distance and sell it repeatedly.

Before sound recording, a performance itself could not be copied. A musician could perform again, teach, work for a court or church, or write for a patron, but the sound of one particular evening could not be packaged and sold thousands of times. Written and, eventually, printed music could. With the development of music printing from the early sixteenth century onward, notation became not only a mnemonic and compositional technology but a commercial one. Printed music could circulate through distribution networks similar to those used for books, and by the eighteenth century sheet music had become a substantial commercial product in European cities.

A score was not a recording in the modern sense. It preserved no sound. But before the phonograph, record, tape and digital audio, it was one of the few scalable ways in which a specific musical creation could be separated from the people who made it and reproduced elsewhere.

That matters because media do not preserve all musical information equally well. Pitch relationships, rhythmic structure and harmony transfer relatively well to paper. Microtiming, tone quality, improvisational decisions, physical technique and interaction between musicians do not. The aspects of music that notation could preserve efficiently therefore had an unusually good chance of surviving historically, circulating commercially and becoming objects of study. Printing did not single-handedly create score-centered musical culture, but it created a powerful selection effect: for several centuries, the most portable, reproducible and saleable representation of music was the written one.

Sound recording fundamentally changed that situation. For the first time, one particular performance could itself be reproduced. With Zappa, Jarrett, Troilo, Hendrix, Charlie Parker or Stéphane Grappelli, we can therefore do something impossible with Mozart, Paganini or Liszt: place the written representation beside the actual sounding behavior of the musician and compare the two directly.

When technique came before the note

Liszt and Paganini add another dimension. They were not only composers but exceptional instrumentalists, and their compositions emerged partly from what they had physically discovered on their instruments.

A virtuoso spends years developing a personal technique. Certain movements become easy, new sounds are discovered and new combinations emerge under the hands. Those possibilities become part of the musician’s language and eventually find their way into compositions. For the next generation, the process runs in the opposite direction. The student first receives the end result on paper and then has to develop the technique required to reproduce it.

Paganini makes this especially tangible. His violin music employs harmonics, multiple stops, left-hand pizzicato and unusual bowing techniques. He was also an accomplished guitarist, and researchers have explored guitar-like idioms in his violin writing. For Paganini, a musical idea could therefore grow out of something he had discovered with his hands. A violinist two hundred years later first encounters the notes and is then asked to develop the hands necessary to execute Paganini’s solution.

This is enormously valuable. Difficult music forces us to acquire techniques we might never have discovered ourselves. But it also shows that a written note can be the end product of years of physical development. Notation preserves the result much more easily than the process that produced it.

With Jimi Hendrix, this seems almost obvious. A detailed transcription can preserve pitches, rhythms, bends and articulations, but his musical language emerged from the interaction between his hands, the electric guitar, amplifier, distortion, feedback, vibrato and tremolo system. The written pitch is a trace of a much larger physical and technological event.

The same symbol, a different instrument

Even when musicians read exactly the same symbol, it does not necessarily describe the same physical action. A pianist cannot make a single piano tone grow louder after the hammer has struck the string. A violinist can increase the intensity of a sustained note, while a bandoneon player can shape it through the bellows. A crescendo across a long note therefore requires fundamentally different physical solutions on these instruments.

The same is true of legato and staccato. A violinist can physically connect several notes within one bow stroke. A pianist creates the impression of connection through attack, release, overlap and sometimes pedal. On bandoneon, fingers, articulation and bellows movement interact in yet another way. The symbol can describe a musical intention without completely specifying its realization.

Tango sometimes makes this very concrete. The marcato has developed into a highly distinctive articulation: often extremely short and percussive, with a dry attack that can feel almost like striking a match. On bandoneon, that character is produced not only with the fingers but through the bellows and the body of the instrument itself. Players commonly use a sharp bellows impulse supported by a bounce on the leg, producing an accent that is difficult to reproduce with quite the same physical mechanism and attack on an accordion.

Or consider a percussive violin effect such as tambor. Its identity lies at least as much in the physical gesture and resulting timbre as in the written pitch. The effect is produced by making the string strike against a stopped finger, creating a dry, drum-like sound. At the same time, the result is not acoustically pitchless: changing where the gesture is produced also changes its resonance and pitch content.

The useful question is therefore not whether the written note may simply be ignored. It is what that note is specifying. Is absolute pitch essential? Is the notation partly indicating a conventional physical way of producing the sound? Is it describing an effect or a function? Or is ordinary pitch notation simply the most convenient available way to represent something more complex? At that point, the question shifts from what is written to why it is written.

Jazz: less on the page, more in the musician

Jazz offers an illuminating comparison because it developed a very different relationship with notation. A jazz ensemble can construct an extraordinarily sophisticated performance from remarkably little written information. A lead sheet may contain melody, chord symbols and form. In many contemporary playing situations, a musician may have little more than a chord chart on a tablet, more often than not these days with an app on a phone or a tablet, allowing them to change the key quickly if the other musicians require it.

Yet the music is not correspondingly simple. Much of its information has simply moved somewhere else: into the musicians.

A chord symbol does not prescribe a complete piano part. It describes a harmonic situation within which the pianist chooses voicing, register, rhythm, density and color. The bassist constructs a line that supports harmony and pulse. The drummer uses a learned rhythmic language while responding continuously to the ensemble. A soloist creates new melodic material within a form that everyone must keep internally.

This is why a Charlie Parker transcription occupies such an interesting position. A saxophonist may spend months learning one of Parker’s solos note for note. That can be an extraordinary way to study articulation, harmony, rhythm and melodic construction. But nobody concludes that every future performance of the tune should contain Parker’s exact solo. The transcription is evidence of how one exceptional musician solved a musical problem. The student reproduces the solution in order to learn the language behind it.

Jazz therefore provides a useful counterexample to the idea that precision necessarily requires more written information. A performance can be highly disciplined while leaving enormous freedom to the musicians, provided they share enough harmonic, rhythmic and stylistic knowledge to make meaningful decisions.

Jazz and tango: two modern urban languages

Jazz and tango also make an interesting historical pair. Both developed as urban popular musics around the turn of the twentieth century and underwent enormous transformations during the era of records, radio, dance halls and expanding international mass culture.

They were not isolated from each other. Jazz was heard in Buenos Aires, tango travelled internationally, and musicians operated in a world in which different forms of dance music circulated rapidly. Research on the early recording industry shows that American jazz, tango and other popular genres shared commercial and cultural networks, while tango’s modernization during the Guardia Nueva unfolded in a world increasingly familiar with jazz harmony, instrumentation and orchestration. Osvaldo Fresedo provides a particularly tangible example of these connections. He had encountered American popular music during an early visit to the United States and incorporated instruments and colors associated with jazz into his own orchestral experiments. Then, in 1956, the connection became quite literal: during Dizzy Gillespie’s visit to Buenos Aires, Gillespie joined Fresedo’s orchestra at the Rendez Vous nightclub. A recording survives of Fresedo’s famous tango Vida mía with Gillespie improvising on trumpet over the orchestra — an almost perfect meeting, in sound, of the two traditions discussed here.

It would nevertheless be too simple to say that the tango variación developed from the jazz solo. Their musical functions are different. In jazz, improvisation became a central mechanism through which an individual musician could create new melodic material over an existing harmonic and formal structure. In tango, a variación is more specifically a transformation or intensification of thematic material, often highly virtuosic and frequently associated with the bandoneons. It may be composed, arranged or shaped through performance practice rather than freely improvised.

The comparison is more interesting without forcing a direct genealogy. Jazz and tango were two living musical cultures confronting some of the same possibilities: individual musicians, shared stylistic languages, ensemble interaction, increasingly sophisticated arrangements and, crucially, the new possibility of preserving performances through recording. That last point distinguishes both traditions from Mozart’s world. We do not have to infer how Charlie Parker placed a phrase, how Troilo attacked a note or how Grappelli shaped a violin line solely from notation; we can listen.

Tango: when a small amount of notation carries a lot of information

The relationship between written information and internalized knowledge is particularly visible in Argentine tango. I once played with a quartet from Argentina that used only a few letters in the parts to indicate different ways a measure should be played. I no longer remember the exact abbreviations, but for the musicians those brief indications were sufficient. Behind a few letters lay an entire performance practice they already knew.

Tango has a shared vocabulary of rhythmic models and playing conventions, including marcato, síncopa, arrastre, yumba and many instrument-specific effects. Musicians who know that language do not need every physical action written out. A short indication can invoke an entire performance model.

In this respect, the tango marking and the jazz chord symbol perform a similar informational trick: they compress a large amount of musical knowledge into a small amount of notation. The amount of information in a score therefore cannot be measured simply by counting symbols. A few letters can tell an experienced musician a great deal and an outsider almost nothing. The missing information has not disappeared; it exists in the performance tradition.

The same notes can become a different orchestra

The great tango orchestras demonstrate how much different music can emerge from broadly similar raw materials. D’Arienzo, Di Sarli, Troilo, Pugliese, Fresedo and other bandleaders worked with the same basic instrumental families, a shared rhythmic vocabulary and often the same repertoire. Yet their orchestras can be distinguished within a few measures.

Bassist, producer and tango educator Ignacio Varchausky has done extensive work analyzing these differences. His approach looks beyond which notes an orchestra plays to rhythmic models, articulation, dynamics, texture, melodic treatment, instrumental functions and the personalities of individual players. In his work on Di Sarli, he explicitly emphasizes the difference between what is played and how it is played.

That “how” cannot be separated entirely from the people in the orchestra. Carlos Di Sarli and Osvaldo Pugliese were both pianists, and the piano acquired an important organizing function in both orchestras, but with very different results. Di Sarli’s broad melodic treatment and remarkable string sound are central to his orchestral identity. Pugliese and his musicians developed a much more elastic language of rhythm, dynamics, rubato and tension. Juan D’Arienzo was a violinist, yet the distinctive rhythmic identity of his orchestra also depended heavily on musicians around him, including pianists such as Rodolfo Biagi.

An orchestral style is therefore better understood as a system of musical choices than as a collection of prescribed notes. An arrangement is also a system of choices. It records one solution to questions of balance, rhythm, register, instrumentation, texture and musical hierarchy. Understanding those choices makes it possible not only to reproduce them but also to reconsider them when the musicians or circumstances change.

Fraseo: when the correct notes are still not enough

One important tango concept is fraseo. Simply translating it as “phrasing” does not quite capture it. A tango melody can play systematically with time. Notes and groups of notes may be stretched, shortened or displaced relative to the underlying pulse while their relationship to that pulse remains musically intelligible.

Researchers have measured such timing differences in historical tango performances. They are not simply sloppy rhythm or vague expression. The deviations can be systematically related to melody, form and style. You can therefore play all the correct pitches and written note values and still sound very little like Troilo. A significant part of the musical information lies in where the notes actually occur in time.

Here Jarrett, jazz and tango meet. Metric notation can be accurate without being exhaustive.

When a variación becomes fixed

The tango variación provides an almost ironic example of what can happen when living musical decisions become repertoire. In many instrumental tangos, a variación provides a virtuosic transformation or intensification of thematic material, often in rapid sixteenth notes and frequently played by the bandoneons. It also has a practical significance on the dance floor. Because the variación so often appears towards the end of a tango, its sudden increase in rhythmic density and energy has become a familiar cue to dancers that the end is approaching. Experienced dancers can hear the orchestra moving into its final stretch and prepare the shape and energy of their dancing accordingly. This is not a fixed rule — variaciones can also occur elsewhere in a piece — but it is another example of musical information that is immediately meaningful to someone who knows the language, even though nothing in the individual notes says “the end is coming.”

Not every historical variación was freely improvised. Some were composed, some arranged and others were shaped around particular players or performance practices. What becomes interesting is what happens afterwards. A passage whose very purpose is to vary existing material can become famous, be written down and then be learned note for note by later generations. A particular solution becomes repertoire in its own right.

Compare that with Charlie Parker. His improvised solo can also be transcribed and reproduced exactly. But jazz pedagogy generally treats that reproduction as a route towards understanding Parker’s language rather than as the required solo for every future performance.

The comparison suggests a useful question for tango pedagogy as well. Learning a historical variación exactly can be enormously valuable technically and stylistically. But after learning it, we can also ask what the passage is doing. Is it increasing tension, intensifying the rhythm, transforming the theme, exploiting register or driving towards a climax? Once that function is understood, the same musical problem can be approached with different material.

A note also has a role

Within an ensemble, a note acquires another property: it has a function in relation to the other voices. Tango makes this particularly easy to hear. The melody may move from one instrument or section to another within a single arrangement. At first the violin may carry the melody while piano and bandoneons accompany it. A few measures later the bandoneons take over and the violin becomes a second voice, an answer or part of the rhythmic texture.

Dynamics are therefore relational. Forte is not a fixed number of decibels. Four violins against two bandoneons create a different balance from one violin against four bandoneons, even if the parts contain identical dynamic markings. But numbers alone are not enough either. If one violin has the melody while four bandoneons accompany it, four instruments may collectively need to make room for one. A few measures later the bandoneons may take over the melody and the relationship reverses.

A musician who concentrates exclusively on their own part can therefore play everything correctly while producing the wrong musical result. The player needs to understand who currently carries the melody, who answers it, who has a second voice, who supports the harmony, who establishes the pulse and who should build tension or create space. When the musical roles move, the relationships between the players must move with them.

The piano: working backwards from notes to harmony

The piano reveals another side of the same idea. A tango piano part can be written out completely note for note, but much of what appears on the page can also be understood through the underlying chords, the function of the part and familiar rhythmic and technical conventions.

A pianist who learns only the written part knows what is played in this particular version. A pianist who also understands the harmonic structure, accompaniment model and reasons behind particular voicings, bass lines and rhythmic figures understands where those notes came from.

That difference matters when the situation changes. If the double bass takes over the bass function more strongly, perhaps the pianist can play less. If the bandoneons are already supplying substantial rhythmic information, the piano can create space. If something disappears from the ensemble, the pianist may take on more responsibility.

The written part can therefore be understood as one implementation of an underlying musical model. If you know the implementation, you can reproduce it. If you understand the model, you can adapt it.

The singer makes the problem visible

We tend to expect something different from a tango singer than from the instrumentalists. The singer is generally expected to have internalized the words and melody. A singer continually reading from a score during a performance would seem unusual in many tango contexts, while instrumentalists sitting behind stands with fully written parts are easily accepted.

Yet the singer is precisely the person to whom the orchestra needs to respond. A singer may delay words, push phrases forward, take space and develop an individual fraseo. A good accompanying orchestra must be able to follow and support that.

Even the key becomes a musical choice. For a singer, the issue is not simply the highest and lowest note. Tessitura matters: where does most of the melody lie, where are the climaxes and how do they relate to the register transitions of this particular voice?

The ideal key for the singer, however, is not automatically the easiest key for the orchestra. On violin, open strings, resonance and fingerings change. On piano, patterns fall differently under the hands. On bandoneon, transposition can have significant consequences for fingering, register and bellows organization. A passage that feels natural in one key may become considerably more awkward a semitone or tone away.

Arranging therefore becomes a search for an optimum. Perhaps the orchestra accepts a more difficult key because the singer sounds dramatically better in it. Perhaps several keys work vocally and the musicians choose the one that works best instrumentally. Perhaps some voicings and accompaniment figures are redesigned rather than mechanically transposed.

A rigid orchestra forces the singer to adapt to the arrangement. A flexible orchestra can investigate how the arrangement can best support this singer.

A la parrilla: when musicians distribute the functions themselves

With tocar a la parrilla, much of this becomes explicit. Musicians play without a fully written arrangement, perhaps using no written music at all or working from a simple lead sheet containing little more than melody, chords and form. They draw on a shared tango language of melodic treatment, rhythmic models, articulations, fills and familiar instrumental gestures. The written information may therefore be remarkably sparse, but that does not mean the arrangement is absent. Much of it is being created by the musicians themselves, using knowledge they already share.

This is not simply tango’s version of a jazz jam session. The languages and conventions are different. But both practices demonstrate how an ensemble can function when the musicians possess enough shared knowledge to make decisions that the notation does not make for them.

Who has the melody? Who supports the harmony? Where does the pulse come from? Is there room for a countermelody? Is the texture becoming too crowded? Who takes over a function when somebody else stops playing? A written arrangement can answer many of these questions in advance. A la parrilla forces the musicians to answer them while playing.

What two musicians can preserve from an orchestra

The duo Los Suplentes, guitarist Emiliano Faryna and bandoneonist Hugo Satorre, provides a useful modern example. They translate characteristic sounds and articulations of historical dance orchestras into music for just guitar and bandoneon. Their arrangements are developed orally, without written scores, and leave room for improvisation.

Two musicians cannot reproduce every part of an orquesta típica. Choosing becomes unavoidable. At one moment the guitar may suggest rhythm, bass and harmony while the bandoneon carries the melody. Later those functions can shift. Some inner voices disappear completely, while a characteristic rhythmic gesture may prove essential.

The result does not have to sound like an incomplete orchestra. Through the act of choosing, a distinct ensemble sound emerges. This is one of the clearest examples of the difference between reduction and realization. The question is not how many orchestral notes two people can retain, but what musical information is needed for the tango to work convincingly with two musicians.

What happens when the score also becomes the curriculum?

A developing musician can understand much more musically than their hands are capable of executing. On violin this is painfully obvious: even producing a beautiful sustained note requires considerable coordination. Add shifts, double stops, difficult bow strokes and ornamentation, and years may pass before a particular written realization becomes technically comfortable.

The obvious response is to practise, and rightly so. If you cannot execute a bow stroke, chord or shift, that is a good reason to learn it. Difficult repertoire is valuable precisely because it forces us to acquire possibilities we might never have discovered ourselves.

But that does not have to be the only response. At the same time, we can ask what we can already make of the music today. Perhaps we play the melody without every double stop. Perhaps we simplify a run but preserve its harmonic destination and arrive at the important next note in time. Perhaps we remove inner voices while retaining melody, pulse, harmony and character.

Six months later, more technique is available and the realization can become richer. Eventually we may be able to play the original version exactly. That can itself be a form of homage: we have developed our technique far enough to understand another musician’s solution literally with our own hands.

Adaptation and technical development are not opposites. One allows us to make music with the abilities we possess today; the other increases the possibilities we will possess tomorrow.

Not less reading, but more understanding

None of this is an argument against reading music well. Fluent sight-reading is an enormous skill. It allows musicians to understand new music quickly, learn repertoire faster and sometimes join an ensemble immediately without first spending days studying the piece.

The more interesting distinction is between the score as information and the score as complete instruction. A versatile musician can read that information quickly while also understanding what it is doing musically. They can reproduce a part exactly when that is required, but they can also hear when another voice is more important, notice that the instrumentation has changed, simplify a passage without losing its structure, respond to a singer and adapt when the musical situation demands it.

Jazz makes the same point from the opposite direction. A player may have only a chord symbol in front of them, but that does not make the task easier. It may require more harmonic, rhythmic and stylistic information to be carried internally.

The ideal is therefore neither dependence on notation nor rejection of notation. It is to have as much information available as possible — on the page, in the ear, in the hands and in our understanding of the music — and to know how to use it.

How can we practise this?

These skills can be trained deliberately without replacing technical study, repertoire work or sight-reading.

Remove information. Take a familiar piece and play only the melody. Add the bass, then only the most important harmonic events. Deliberately remove material while keeping the structure and character recognizable. This reveals which information is essential and which belongs to one particular realization.

Simplify without stopping. Play a technically difficult passage in tempo and, if necessary, remove double stops, inner voices or intermediate notes. Preserve the pulse, important melodic points and harmonic destinations.

Learn a difficult passage in two directions. First find a way to realize it convincingly with the technique you have today. Then identify precisely which technical ability prevents you from playing the written version and practise that separately.

Work backwards from the chords. Take a fully written accompaniment and analyze its harmony, bass function and rhythmic model. Put the part away and build a new accompaniment from that information. Then compare your solution with the written arrangement.

Exchange functions. In an ensemble, deliberately move the melody from one instrument to another. The other players must adjust dynamics, articulation and density without receiving newly written parts.

Build an arrangement from incomplete information. Give an ensemble only melody, harmony and form. Let the musicians decide who supplies melody, bass, rhythm, second voice and fills. Then remove a player and keep the music working.

Reduce an orchestra to a duo. Take a short passage from an orchestral arrangement and make it convincing with two instruments. You cannot preserve everything, so decide which functions are essential.

Make your own variación**.** Take a melodic fragment and use your current technique to increase rhythmic density, transform the melody or move towards a climax. Then learn a historical variación and compare the solutions.

Treat a transcription as a lesson, not a command. Learn a Parker solo, a Grappelli phrase, a Troilo passage, a Hendrix gesture or a Zappa solo as accurately as possible. Study its timing, articulation, harmony, sound and technique. Then put the transcription away and create something different using principles you discovered in it.

Practise playing less. Remove increasing amounts of material from an arrangement. When does the music become clearer? When does something essential disappear? Learn to distinguish a place where you can play from a place where the music actually needs you to play.

So what is in a note?

A note can contain a surprising amount. It can be an instruction, carry a harmonic or melodic function, preserve an improvisatory decision, or record the result of a technique that a musician spent years developing. At the same time, a note can only contain what notation is capable of containing. Timing, sound, physical technique, interaction and stylistic knowledge can be represented on paper only partially.

None of this makes the score less important. On the contrary, it can be music to perform, historical evidence to investigate, and a record of musical decisions from which we can learn. But once we recognize that the notes in front of us are sometimes the outcome of a creative process, we can also try to read that process backwards. We can ask not only what somebody played, but why it worked.

The question “what is written?” therefore remains essential. It simply does not have to be the final question. We can follow it with another: what does the music need from the musicians who are playing it now?

Sources and further reading

Keith Jarrett — The Köln Concert**, authorized transcription, Schott Music.** Useful for the unusual case of a fully improvised performance later becoming an authorized score, and for Jarrett’s own reservations about the limits of transcription. https://www.schott-music.com/en/the-koeln-concert-no38222.html

Julián Graciano — “Tango and Jazz: Cross-Genre Relations in History and Practice,” in The Cambridge Companion to Tango**.** Particularly relevant to the historical relationship between jazz and tango, improvisation, spontaneity and the use of reduced notation such as lead sheets. https://www.cambridge.org/core/books/abs/cambridge-companion-to-tango/tango-and-jazz-crossgenre-relations-in-history-and-practice/86A6D96F1A5A13FBFE4F00AEAB3E78B8

Ignacio Varchausky & Adam Tully — “Orchestral Rhythmic Designs and Performance Practices: Juan D’Arienzo and Aníbal Troilo,” in The Cambridge Companion to Tango**.** A useful scholarly treatment of rhythmic language and performance practice in two major tango orchestras. https://www.cambridge.org/core/books/cambridge-companion-to-tango/3BC7F58C563B0BEA422B3BEB71DE1065

Ignacio Varchausky — Los Estilos Fundamentales del Tango**.** Varchausky’s educational material on Di Sarli, D’Arienzo, Troilo and Pugliese is especially useful for hearing style as a combination of rhythmic models, instrumental functions, articulation, texture, dynamics and individual musicians rather than as notation alone. https://www.ignaciovarchausky.com/

Andrés Serafini — “Pichuco y su ‘goma de borrar’: diferencias entre texto y ejecución en un tango arreglado por Astor Piazzolla para la orquesta típica de Aníbal Troilo,” Revista Argentina de Musicología 19. A particularly concrete case study comparing a surviving written arrangement with Troilo’s recorded execution. https://ojs.aamusicologia.ar/index.php/ram/es/article/view/270

Todo Tango — “Listening to Tango Dance Music: A Beginner’s Guide.” A practical listening introduction to the different layers of tango performance, including rhythm, melody and contrasting uses of the bandoneon. https://www.todotango.com/english/history/chronicle/448/Listening-to-tango-dance-music-A-beginners-guide/

Los Suplentes — Emiliano Faryna & Hugo Satorre. Guitar and bandoneon provide a useful practical example of reducing the resources of the orquesta típica while preserving and reinterpreting its musical functions. Their own descriptions and live performances are particularly valuable because the duo develops arrangements orally and leaves room for improvisation. https://hugosatorre.com.ar/

Jessica Dauterive, Matthew B. Karush & Michael O’Malley — “Hearing the Americas: Understanding the Early Recording Industry with Digital Tools,” Journal of the Gilded Age and Progressive Era (2023). Useful historical context for the early recording industry and the international circulation of popular music in the Americas. https://www.cambridge.org/core/journals/journal-of-the-gilded-age-and-progressive-era/article/hearing-the-americas-understanding-the-early-recording-industry-with-digital-tools/32F846C800B73D93ACC78B01949F1FE4

David Hyun-Su Kim — “The Brahmsian Hairpin,” 19th-Century Music 36, no. 1 (2012), 46–57. An illuminating study of nineteenth-century hairpins and the possibility that such markings communicated a broader expressive practice than a purely modern dynamic reading suggests. https://online.ucpress.edu/ncm/article-abstract/36/1/46/69638/The-Brahmsian-Hairpin

Frederick Neumann — Ornamentation and Improvisation in Mozart**.** A substantial study of ornamentation, unwritten additions and improvisatory performance practice in Mozart. https://www.degruyter.com/document/doi/10.1515/9780691194684/html

Dana Gooley — Fantasies of Improvisation: Free Playing in Nineteenth-Century Music (Oxford University Press, 2018). Useful for understanding the changing relationship between virtuoso, improviser, composer, performer and the concept of the musical work in the nineteenth century. https://academic.oup.com/book/25754

Tim Carter — “Music-printing in late sixteenth- and early seventeenth-century Florence,” Early Music History**.** Useful background on the commercialization of music printing and the development of printed music as a marketable product. https://www.cambridge.org/core/journals/early-music-history/article/abs/musicprinting-in-late-sixteenth-and-early-seventeenthcentury-florence-giorgio-marescotti-cristofano-marescotti-and-zanobi-pignoni/CFA08F6021E7D7D338736EDDF43AD565

Rupert Ridgewell — “Inside a Viennese Kunsthandlung: Artaria in 1784,” in Consuming Music**.** A useful view of the commercial world of printed music in Mozart-era Vienna. https://www.cambridge.org/core/books/consuming-music/inside-a-viennese-kunsthandlung-artaria-in-1784/D4CA01B497A5953DD477E1E0BD435FB1