TELLEF ØGRIM

Why Free Improvisation Doesn’t Seem to Fit Music History and why that Matters

Published in Tempo: A Quarterly Review of New Music, July 2026

(Subscribe here)

By Tellef Øgrim and Mia Zabelka

Free improvisation is often treated as marginal or ahistorical, even in serious accounts of twentieth-century music. This article argues that the problem is not the music, but the way music history is usually written: in many narratives, free improvisation is treated less as a practice than as a symptom. It is not sufficient, however, to describe freely improvised music as an outburst, a reaction, a moment of rupture, something that happens after styles have broken down rather than something that carries its own historical weight. Taking The Rest Is Noise by Alex Ross as a reference point, we suggest another way of hearing improvisation: not as an exception to
history, but as a practice that challenges how historical authority itself is organised.¹

Even in thoughtful and wide-ranging histories such as The Rest Is Noise, improvised music tends to remain at the edge of the story. It is acknowledged, sometimes vividly, but rarely allowed to reorganise how the story itself is told. This is not because free improvisation lacks depth or influence, it is because it does not behave in ways that conventional music history finds easy or convenient to describe. Many listeners will recognise this tension, even if they have never thought about it in these terms. Free improvisation often feels historically present when you hear it, yet historically absent when you read about it. The gap between those two experiences is where this article begins.

From early vocal and percussive practices to baroque performance, from Indian classical music to jazz and contemporary electronic forms, improvisation has been a continuous and central mode of musical expression. Yet in dominant accounts of Western music history, it appears as a supplement, something secondary to more formalised methods of composition and notation. Free improvisation is particularly difficult to place within music history because it resists the categories on which most historical narratives depend. It does not centre on stable works, fixed repertoires or stylistic succession. It unfolds as practice rather than product, as event rather than object. In histories that are built around composers, scores and movements, this creates friction.

It is important to note that such histories are concerned almost exclusively with Western art music, just one historically specific musical culture among many. From a global perspective, music is very often created, transmitted and transformed without written notation. In many cultures musical practice is fundamentally oral, embodied and situational; improvisation is not a specialised activity but the normal condition of music-making. What Western discourse labels as ‘improvised music’ is, in much of the world, simply music.

Before going further, it is worth being clear about what The Rest Is Noise set out to do and when it did it. First published in 2007, the book is a broad cultural history of twentieth-century art music, not an ethnography of musical practice. Its focus is on how composed music responded to war, ideology, power and trauma and, from that perspective, it remains persuasive and influential. Given the book’s position as a widely read account ^[1]
of twentieth-century art music, it is still relevant to ask what lies in the background of such a narrative.

As it resists standardised organisation around works, composers and stylistic succession, free improvisation fits poorly into such narratives. But that difficulty should not be mistaken for marginality. It may instead point to a central musical practice of the twentieth century that demands a different way of being written into history. Ross does acknowledge free jazz and free improvisation, but as the precursor to composed music, as in a section of the ‘Beethoven was wrong’ chapter which takes us from Charlie Parker, through John Coltrane and Ornette Coleman and eventually to ‘younger classical composers looking for a way out of Schoenberg’s maze’, such as Steve Reich.^[2] Improvisation is a transitory phase, something one arrives at, encounters and then goes beyond. It becomes part of a sequence rather than a practice with its own internal logic and continuity. What is missing in this framing is not respect, but depth.

If one actually listens—across performances, ensembles and decades— the idea that free improvisation exists outside tradition quickly collapses. One hears jazz phrasing and time-feel, extended techniques from postwar contemporary art music, noise practices, folk and instrumental memory, ritual gesture, call-and-response. These elements are not denied or erased; they are made available. Yet it is not unusual to encounter descriptions of freely improvised music that describe it as somehow freed from the rules of genres and styles, drawn from the moment in which it is performed. Such descriptions may be incorrect, but they do at least offer inspired attempts at branding, at a sort of anti-categorisation.

In this context it may be fruitful to look back on the legendary improviser Derek Bailey’s concept of ‘Non-Idiomatic Improvisation’.^[3] Since he coined the term go years ago it has often been treated too literally and become somewhat superficial. We believe that the key to understanding Bailey’s concept is his statement that free improvisation has no stylistic commitment: it contains style but avoids being ruled by it. Ten years after Bailey’s book, another central player from British improvised music, Edwin Eddie Prevost, joined the discourse, in an answer to a question about whether this music had rules or not.

I’m saying that there is a set of rules. It’s no good Derek (Bailey) saying he doesn’t have any rules. Well, he can say that but it doesn’t necessarily mean that it’s true. The very fact that I can recognise his playing from one occasion to another indicates to me that there is a set of rules. If it’s coherent there must be rules. There are rules; it’s a different set of rules. And a different set of rules relates to a different worldview. What we are proposing, whether we are doing it consciously or not, is a different worldview—and there has to be rules in a worldview. I’m not proposing anarchy, I don’t even believe that anarchy exists, there’s nothing in nature, which is anarchic, it always gravitates towards form eventually. The question is, ‘What form?’ and what we’re proposing, consciously or otherwise is a form, which in essence is completely different in its political and social implications from the form which classical music has perpetuated. And let’s have no illusions about it classical music is not apolitical, it’s very political indeed, and so is pop music.^[4]

Prevost’s concept of ‘meta-music’ is another oft-quoted attempt to grasp the core of free improvisation. The concept may not be entirely defined but it fleshes out some essential demands for musicians taking part in collective improvisation, such as a profound adherence to their ‘sonic contribution’, even when that contribution is silence. According to Prevost, no sound, including the absence of sound, is truly innocent and the people listening to an improvisation contribute to its identity just as much as the musicians and the sounds they produce, although this is only partially preserved in a recording.

This could also be said for many other musical genres and styles. The emphasis on discovery, both relating to excavating sounds from a concert situation or discovering sides of one’s self is however one aspect of Prevost’s metamusic concept that is unique to free improvisation: ‘You can never be certain of your place within the whole. But, in my opinion, part of the reason to practice this music is to try and discover something more about yourself. This is in relation to the materials you use and in relation to the musicians you perform with’.^[5] Musical coherence emerges not from hierarchy but from coordination: a shared negotiation of musical space in
which no single participant controls the outcome. Improvisation models a non-hierarchical form of organization that is both aesthetic and social. The ensemble dimension complicates authorship.

This explains why free improvisation cannot be reduced to a niche practice. Its principles—interaction, contingency and openness—are not confined to experimental music but appear across genres, including popular music. Late works by David Bowie or the experimental strategies of Bjork demonstrate how improvisational thinking can permeate structured forms. At the same time, improvisation challenges one of the central assumptions of Western art music: the idea of the work

more central than modern narratives suggest. Cadenzas in concertos were improvised, not notated, and eighteenth-century keyboard practice assumed improvisational fluency as a basic skill. Composers such as Mozart and Bach often developed their works at the keyboard, meaning that music later fixed in notation frequently originated in improvisatory exploration. Opera composers such as Verdi shaped arias collaboratively with singers during rehearsals, drawing on performers’ spontaneous contributions. The sharp opposition between composition and improvisation is a comparatively recent construction rather than a stable feature of Western musical life.

If free improvisation is historically saturated, it nevertheless treats history as material rather than as law. When improvisers say that their music has nothing to do with genre or style, they are rarely making a literal historical claim. They are making a normative one: no inherited form has authority in advance and no tradition dictates what must happen next. This becomes clear in performance. Improvisers comment on, rewrite, mesh, parody or invert traditions in real time. Styles appear briefly, collide, dissolve and return altered. The past is not quoted as a reference point but activated as a resource. Meaning is produced through interaction rather than through adherence to form.

Anyone who has tried to write promotional material for a festival, a concert or a recording will know that it is not easy and that often the words seem to fall short of describing the music. When asked to categorise their
music, musicians often point out that categorisations make it difficult for both them and their public to keep their ears open to the music: mapping the music within a system of genres and styles limits creativity. Such categorisations also restrict festival programming and too many contemporary music festivals are evidently reluctant to include improvised music within their concept of contemporary art music. We challenge festival directors to revitalise their festival programmes in dialogue with improvising musicians.

After a recent Zabelka/Lie/0grim trio concert in Bergen, an audience member approached us and gave a remarkably detailed account of the performance from her own listening perspective. Considering the music’s eclectic character—violin (Mia Zabelka), electric guitar (Tellef Øgrim) and turntable (Håkon Lie)—the clarity of her recollection was striking. What is often described as chaotic or impenetrable had evidently registered as a coherent and memorable sequence of events. A practice that hands over as much as possible to the intuition of the moment creates problems, however, for conventional music history. Most historical writing depends on stable works, named movements and the continuity of institutions such as conservatories and orchestras.

Free improvisation undermines such constructions. It prioritises events over objects, relations over artifacts, listening over notation. History is no longer stored primarily in scores or canons but in skills learned through practice, shared habits and collective memory. That does not make it ahistorical. It makes it difficult to archive. Historical studies of postwar experimental music reinforce this view. In his account of the Association for the Advancement of Creative Musicians, for example, George E. Lewis demonstrates that improvisation was neither anarchic nor anti-intellectual, but a highly structured practice grounded in collective responsibility, self-determination and alternative models of musical authority.

Another example, this time Kyle Gann characterising the group AMM:

AMM is an improvisation group, but not improvisation in the dictionary sense of “to fabricate out of what is conveniently on hand”, and still less in the jazz sense of playing around within a preset structure of melody and harmony. The aims of AMM, and the style of its music, have changed from time to time, and its members are sometimes hard put to describe their working methods; but it is clear that, for each one, improvisation is the embodiment of a philosophy, not a technique. Each AMM improvisation is a search for something freed from the limits of past experience, something that did not exist until its
realization. The members of AMM are uninterested in flaunting their virtuosity or expertise, and indifferent to their past achievements, and even to the concepts of success or failure. Yet despite this Zen-like uninvolvement with conventional standards of good and bad, AMM has garnered an enormous amount of critical praise and built up a wildly enthusiastic European following, and several of their albums have been best-sellers.⁶

Again, there is a suggestion that free improvisation is freed from past experience. Yet to ‘search for something freed from the limits of past experience’ is not the same as finding it: the search constitutes a central part of this form of music, even if past experiences might sometimes pop out.⁷ Such practices do not fit easily into narratives organised around composers, scores and stylistic succession: not because they lack history, but because they embody it differently.

Free improvisation does not usually present explicit political messages. Instead it performs social relations. Authority is negotiated rather than fixed. Leadership is temporary. Responsibility is shared. There may be virtuosity, but it does not automatically confer control. These are enacted conditions, not symbolic gestures, and, in this sense, free improvisation does more than represent democratic or egalitarian ideas: it practices them, however imperfectly, in sound and time. This explains why it sits uneasily within narratives of music history. The Rest Is Noise is primarily concerned with how music responds to war, ideology, power and trauma. Free improvisation often does something else. It does not reflect history so much as interfere with it. It shifts attention away from works and toward relationships, away from fixed meaning and toward negotiated action.

To describe this as music outside history is therefore misleading. What it resists is not history itself, but a particular way of organising authority around established traditions, granting the past an automatic right to govern the present. Free improvisation does not reject the past: it refuses its authority and music history’s body of evidence must be widened if it is to account for this practice more fully. Not only scores and institutions, but listening practices, shared ways of responding and the responsibilities performers assume when making sound together need to count as historical material.

As form-orthodoxy has waned somewhat in contemporary art music, improvisation becomes more important in so-called ‘score-music’. George Stevenson, for example, is a composer who says that ‘improvisation remains at the core of my music, driven by the pure excitement of seeking out harmonic objects and musical structures that are new and fascinating to me’.^[6] Just as we do not insist that free improvisation take over the music world and that notated music should be forbidden, we applaud such signs that improvisation is gaining artistic ground again.

Human beings improvise in art, as we do in our everyday life. Improvisation and experimentation have been fundamental components of music since its beginnings. From singing and drumming to electronic music, humans have continuously explored new sounds and forms of expression and with the advent of new technologies and studio techniques, the possibilities for improvisation have expanded further still. Electronic music now allows artists to manipulate sounds in real time and thus to improvise. Consequently, free improvisation is better understood as a method than a genre. It does not prescribe a stylistic identity, but establishes conditions: attention, interaction and responsiveness in real time.

This has several consequences. First, improvisation operates as a dynamic system, in which small gestures can have disproportionate effects on the unfolding musical form. The music is not ‘built’ to a plan but develops through feedback within the ensemble. Second, it depends on what practitioners describe as ‘automatic playing’: a state in which accumulated experience is activated without conscious pre-structuring. This is not the absence of knowledge but its redistribution, from planning to action. Third, improvisation is a form of real-time composition, requiring continuous decision-making under conditions of uncertainty. What is often described as freedom is, in practice, a highly demanding mode of attention.

Away of listening to the improvisational process is to adapt the so-called ‘insist-adjust pendulum’, a state of heightened communication in which the players shift between proposing motifs and building new motifs on suggestions from their co-players. If and how the insist-adjust pendulum plays itself out in a certain session is dependent on the fundamental preferences and background of the musicians taking part in the session but, if the music is truly democratic, with no hierarchical system in operation, the musicians are equally important to the output. For listeners, following this shifting back and forth between insisting and adjusting can enable an unconditional connection with what is going on on stage.

Free improvisation does not stand outside music history but rather exposes its limits: the limitation of The Rest Is Noise reveals not a failure of insight but a structural boundary. Alex Ross offers a powerful account of twentieth-century music’s response to history but free improvisation asks a different question: how might history itself be handled, reworked or refused in the act of making sound together? Free improvisation is not an escape from tradition, but tradition held open, questioned, tested and reconfigured in the present tense. Recognising this music as central rather than marginal has consequences not only for historiography but for how contemporary music is taught, programmed and evaluated.