I’ve never been comfortable with the idea that music is somehow fixed in a definitive notated version – an Urtext that represents the piece fully and eternally, like a beautiful butterfly dried and pinned in a display case. I imagine music as a living, breathing entity, set free in space and constantly adapting to its surroundings and context. An evolving spirit, rather than a resuscitated corpse.
Urtext and the tyranny of 'tradition'
But we have inherited a culture of conformity in the performance of classical music. Narrow limits for the ‘appropriate’ way a piece should be interpreted – an almost fanatical replication of every detail of the score and a pious reverence for reference recordings. Why is this?
Strangely enough, this conformity doesn’t come from the composers or the performers of their day, as Anna Scott’s work on early performances of Brahms’s music testifies, but something far more potent and dangerous – tradition. Tradition provides the template for assessment, and perhaps this is where part of the problem lies. By establishing narrow norms, it is much easier to assess and critique performance in exams, competitions, concerts, and recordings. If the score and/or approved recording is our model answer, then the assessment criteria write themselves. However, the spiritual figure, philosopher, and writer, Jiddu Krishnamurti warns us that ‘tradition becomes our security, and when the mind is secure, it is in decay.’
This approach starts early in our musical experience. Several years ago, I recorded some of the guitar grade syllabus exam pieces. I had to stick closely to the given metronome mark, I wasn’t permitted to add or change dynamics, articulation, rubato, or stylistic features like ornaments or notes inégales. My ‘interpretation’ became a tightrope walk through an obstacle course. The message was clear: ‘copy the recording if you want a good mark’. Heartbreaking.
Should the published Urtext have the last word?
Are published scores reliable? Let me take the example of composer/performer collaborations. There are some commonly held assumptions: the published score represents the final and correct version of the music; the collaboration was happy and egalitarian; there was plenty of time to get the details right and finely polished; the collaborating performer’s interpretation represents the style and intentions of the composer. Recent research reveals that these assumptions are highly misleading.
For instance, composer Julian Anderson and guitarist Laura Snowden had to produce the final version of the score for Anderson’s Catalan Peasant with Guitar a month before the premiere so that the publisher could get the score ready for the first performance at Wigmore Hall. But the piece was still evolving in the lead up to the premiere in that final month. So when the first edition of the score was published it was already out of date.
When composer Per Nørgård worked with guitarist Stefan Östersjö on a recording of his complete guitar music, they made several significant changes to scores published decades earlier. A published score might only be a snapshot in time, rather than something eternal. The French poet and philosopher, Paul Valery reminds us that ‘art is never finished, only abandoned’.
Is a composition ever really finished?
In his doctoral thesis, Irish guitarist Morgan Buckley, closely analysed several collaborations in real time. A key element in the success of any of these collaborations was the performer and composer having enough time together to build trust, to explore the instrument’s idiom, and for the performer to make positive interventions. If there was a clear hierarchy between composer and performer, the collaborations were less fruitful.
Rather than providing all the answers, it may well be the case that most published scores are unstable texts, ripe for more imaginative interpretation, and perhaps even revision or completion.
The Urtext... and creative approaches to interpretation
In recent years, many artists have shown a more creative approach to interpretation, such as Gilles Apap and Patricia Kopatchinskaja with their unorthodox concerto cadenzas; Max Baillie’s work with ZRI and the Lodestar Trio; and Gennaro Desiderio’s anarchic Piazzolla. These performers are invested in performance as a creative act, not a recreative one. Robert Fripp takes this idea one stage further when speaking about his own music – ‘it is no concern to me that even one note of the original is sounded, only that the spirit of the piece is set free to enter the space’. And composer Ferruccio Busoni declared that ‘the spirit of music can never be captured by signs and lines. The written note is at best a hieroglyph; its essence must be divined’.
For me, the very finest performers are alchemists, transforming base metal into gold through their proactive interpretations, communicating the music via their artistry to the audience. I think that music is something that performers should inhabit, interpret, and own, rather than worshiping the score like a relic. Interpretation is subjective and plural, not objective and singular.
'From Honey to Ashes', a double-disc of music by Stephen Goss will be released on Deux-Elles on 20 March 2026 performed by duo Francisco Correa (guitar) and Emily Andrews (flautist/mezzo). The works feature a wide range of Goss’s compositional styles including several arrangements of traditional folk songs.
