Improvisation is the activity of, to some extent, creating and constructing a piece of music in the same time as it is being performed. Improvisation in this wide sense is a necessity in all performed music whether it is called arrangement, interpretation, ornamentation, reading or something else.
Usually, the improvising takes place within some kind of given constraints or frames - sometimes guided by formalised rules - which are tied to some established tradition dictating how music should sound and be made according to a particular pattern (e.g., jazz, heavy metal, baroque music or flamenco). At other times the improvising does not take place within some specific tradition, but is still limited by certain rules which have been laid out in advance, such as directions from a composer or director or internalised rules about the proper techniques for playing on and the sound of musical instruments.
This limited form of improvising is seldom seen as strange or special, sometimes it is not even noticed by the musicians themselves. One "plays music" or "plays jazz" or "plays sitar" or "this or that particular piece by this or that composer" - the improvising is included as a matter of course, but is not particularly focused upon by the players.
Then there are a few who, like myself, cannot describe their music-making in any of the above mentioned ways. We are improvising, but, we claim, in contrast to most other improvisation our playing is FREE. There are also listeners who, even if they do not enjoy improvised music in general, like free improvised music. What are we talking about when we say such things?
One understanding of free improvised music is that this is also a tradition (or in the process of so developing) - an idiom with certain (strangely) sounding characteristic timbral properties, structure, idea of form, praxis for performance etc. This view of things - that free improvised music is characterised by its peculiar (and strange) sound - is typical of many reports and reviews of this music in the mass media.
This should come as no surprise. The media, as we all know, has a particularly hard time handling things which do not easily lend themselves to categorisation according to established patterns. But I think that also quite a few followers of free improvised music think that way: when they claim to enjoy free improvised music they report their liking of certain combinations of sounds. Moreover, people who dislike free improvised music often describe their feeling with the aid of words like "screeching", "cacophony", "no melodies" etc. Similarly, I can explain the way one free improvising musician handles timbres and rhythms, relates to co-players, and discuss the ideas of form which are unveiled in the music etc.
However, at the same time there is something which does not add up in this way of viewing the issues.
I remember an interview with the American jazz guitarist Pat Metheny, where he tells the story of how he, after three weeks of intense rehearsals and recordings with saxophonist Ornette Coleman, sat down to compose the music to a film. Coleman's very free attitude towards improvisation had such an impact on Metheny, that he started to search for notes on his guitar with the free state of mind suggested by his work with Coleman. Without thinking about the names of notes or anything like that he went on, and after a couple of hours he had found a sound that he liked. He wrote it down on paper note for note and did not until then realise - to his own amazement - that the sound was the simplest of c-major chords.
I have had similar experiences myself of playing or hearing free improvised music which I suddenly realise fits well into some established idiom. I am quite certain that this experience is not unique.
Another aspect of the same thing is that it is possible for a given piece of what we think of, and identifies as, free improvised music turns out to be to some extent composed in advance. In this case the music sounds the same as if it had been freely improvised, but that does not make it free improvised music.
Both of these examples hint at what is strange in the idea that free improvised music is characterised by how (strange) it sounds. The problem is that this conception aims at the sounding product, while what makes the free improvised music special rather is its history of creation. As Derek Bailey, pioneer and leading figure of European free improvised music has expressed it: "free improvisation is not a kind of music... it is a kind of music making".
Free improvisation hence is an artistic method, but is it a fruitful one? What is the point of making music in this way? Such questions call for a more precise account of the characteristics of the method in question.
The most basic element of the musical method of the free improviser is to be found in the attitudes of the latter towards musical traditions, idioms, genres etc. It has often been pointed out, and rightfully so, that free improvisation cannot amount to a total exclusion of traditional idioms. All of us were raised and fostered within some musical idiom. All of us have a history which has put its mark on us and which influences the decisions we take today - and this also applies when we make music. Sometimes, this gives rise to the reaction that free improvising is an impossibility. This for example seems to be the view of drummer and vocalist David Moss, though I hope to show in the following that this view is based on a misconception. The difference between one who is active within the borders of some particular idiom and the free improviser is instead to be found in the way of looking at this idiom.
All music-making require a limitation of available sound-possibilities. In order for a normal person to have any capability to form the music as he or she wants, some form of (formalised or implicit) principles for sorting and systematising the sounds available is needed. Otherwise these will be totally impossible to survey for the music-maker. Thereby there is a necessary selection of some available sound-possibilities at the expense of others. The chosen principles form an idiom, which is normally seen as a collection of rules for what is allowed, possible and suitable from a musical point of view. To choose to play (and compose), for example, bebop in this way involves a commitment to the rules and limitations - the conception of (good) music - of which this idiom consists. This kind of choice is the normal one among musicians.
The free improviser, however, refuses to make any binding choices of this kind. Certainly, it is trivially true that all music-making is idiomatic in the sense that it requires some kind of limitations. This, however, does not constitute a reason for committing oneself to a particular set of limitations. It is not prohibited to shape the music within the borders of some idiom, but neither is it necessary to keep to it. Particular idioms are no longer viewed as prerequisites for the music-making, but rather as tools which in every moment may be used or not used. Musical (idiomatic) rules are thus not considered to be valid in any other sense than that they, for the moment, are accepted by the improviser. However, in the next moment they may have been discarded in favour of some other point of view.
An obvious consequence of this way of seeing things is that it becomes much more difficult to uphold any sharp distinction between composers and musicians, between those who make the music and those who perform it, when it comes to ultimate responsibility for the emergence of a particular piece of music. Of course, a musician may chose to follow the directions of a composer, but this choice is always open to revision. Such a viewpoint easily makes the world of music turn upside-down. The slightly bizarre idolatry of the composer which is so common, at least within Western written art-music (or "serious" music) suddenly appears without warrant. However, just as unwarranted appears also the subservient role which is taken by many musicians. Certainly, the composer has created an interesting theoretical construction, an idiom, but the performing musician is, according to the viewpoint at hand, in every moment solely responsible for whether the music is to be formed according to this idiom or not. Thus, improvisation turns into a necessity: to perform is to compose and viceversa.
Another consequence concerns the role of free improvised music in relation to the avant-garde - with which it has often been linked. During an interview with the German percussionist, drummer and free improviser Paul Lovens, he told me about the music created by Alexander von Schlippenbach's trio, a group which has been working for many years. Lovens claimed that the group had a certain devotion to the avant-garde in the early years and that their music then presumably was innovative. This state faded, however, as a consequence of the members developing their respective ideals for how they wanted to play together. The group's method still is free improvisation, but, Lovens said, "it is no longer innovative or revolutionary in the true senses of these words".
Thus, in the same way that the starting-point of free improvised music contains a refusal to commit to any particular tradition or idiom, it no more favours any experimental or innovative attitude towards music (other than in the trivial sense that nothing is prohibited and that the music always is a product of the musician's own and, in practice, always unique choices). Certainly, free improvisation has been used as a means to musical experiments and the creation of new idioms, but this is, if not a coincidence so at least no necessity. An experimental or innovative attitude can of course be useful for the free improviser, but this does not lessen the fact that such an attitude is a tool which may also be discarded. Free improvised music is experimental or innovative only in as much as it is made by persons whose ambitions are to innovate or experiment.
This way of looking at innovation, traditions, idioms and rules, I believe to be the key to free improvised music. The free improviser sees things as they are: in every moment he or she chooses and constructs the components of the music which then is being created as well as the idiom which in the same moment is used for making this music. In the light of the resulting sounds and the improviser's continuous evaluation of these, the improvising continues as a series of choices on both these levels. As the free improviser sees it, this is a true description of all music and all musicians, the difference being that the free improviser consciously views music-making in this way.
This conception of what characterises free improvised music is not uncontroversial. It reflects an outlook according to which the artistic work is the ultimate end of the artistic process of creation (given that the artist is pleased with it). However, this view contrasts sharply against a common basic idea of (dare I say traditional?) modernist thinking. According to that idea, the artistic work is essentially a means for consciously changing reality in some way or other. The desired changes may be political, personal or mass-psychological.
Within free improvised music, thoughts of this kind have been voiced by Anthony Braxton, whose goal is hard for me to comprehend but seems to be of a religious kind. In Sweden, the thoughts have been visible in the work of Dror Feiler, whose goal primarily seems to be political, but have, as I see it, got their clearest expression in sayings of saxophone player Johan Petri. Petri seems to mean that improvised music cannot be considered to be free unless it is the product of a psychological process which has affected the player so that he or she has been freed from influences of idiomatic codes, traditions etc. "It is necessary", Petri writes, "to ascribe the designation 'free' a total limitlessness and accordingly reject every idiomatic norm, every gesture sprung from anything that is not personal and revolutionary".
Free improvisation here turns into a kind of psychoanalytic therapeutic method, meant to uncover those factors in our personal histories which influence our choices and likes and ultimately aiming at making us capable of rejecting (or at least reflecting on) those factors and become ourselves, so to speak.
For my own part, however, I have some trouble with understanding why it is "necessary" to see free improvisation in this way. Even granted that it would be desirable if the choices of people in general and musicians in particular were less influenced by their past, this seems to be no reason to make such a goal into a defining property of free improvised music. Moreover, since the goal seems impossible to reach, it appears to be plainly insensible to make such a move. Such a conception would mean that free improvisation is impossible. Some improvising musicians - as mentioned earlier, David Moss - admittedly seem to accept such a consequence. For myself, however, this rather constitutes an argument in favour of the claim that Petri's view of free improvisation is unfruitful. A discussion of how free improvisation is to be best explained obviously rests on the presupposition that there is such a thing, or at least can be. If that presupposition is denied, the discussion ends even before it has begun - why, according to one of the parties there is nothing to discuss.
I do not for a moment want to deny that there may be good reasons for free improvisers (and others) to adopt some kind of modernist outlook (for example, Petri's). What I contest is the move to define the notion of free improvisation in terms of such an outlook. For the free improviser to continuously contest the ground for his or her doings or to strive for political effects through the music, may certainly at many occasions be a fruitful tool for making music (especially if the music-maker's evaluation of the music is done from such effects). But outlooks of this kind are also nothing more than tools, which certainly may be put to one side if that is the wish of the person who uses them.
What is the point of free improvisation, then? Is it a fruitful method for making music? My answer is that this depends on who the performer is and which interests he or she has.
The advantage of using the method of free improvisation is that it is always possible to revise the music one is in the process of creating as well as the vantage-points of this music. At every moment the free improviser may throw the tools used hitherto overboard and change the direction of his or her music-making. If one is a person whose musical goals are in a constant flux or, at least, have a continuous need for new means of expression, free improvisation thus appears to be a fruitful method. If, instead, one is clear over exactly what it is one wants to do and how one is to do it and if this want is stable over time (the latter is important), it appears as more fruitful to commit oneself to an idiom, either an available one or one which is constructed by oneself. Of course, such a person may also reach his or her goals by using free improvisation, in that case he or she is simply making the same choice of idiom over and over again. But at the same time it seems unnecessary and perhaps distracting to keep a constant lookout for alternative idiomatic prerequisites.
I would like to conclude with a totally unwarranted accusation. I believe that most people who perceive themselves as having stable aesthetic ideals and convictions regarding how these ideals are to be met live within a false picture of things. If they would adopt the viewpoint of the free improviser, their ideals would become much more dizzy and the uncertainty as to which way that really suits them best would increase. On the other hand, however, that judgement may very well be an expression of some wishful thinking on my part.
 The same observation has been made by Derek Bailey, Improvisation: Its Nature and Practice in Music, second revised edition, London 1992: The British Library National Sound Archive, p. xii.
 The interview was conducted by Art Lange in connection with the release of the record which Metheny recorded with Coleman. It was published in Down Beat, no. 6, 1986 with the title "Pat Metheny & Ornette Coleman - Songs of Innocence and Experience".
 This statement was included in the first edition of Bailey`s book Improvisation (Ashbourne 1981: Mooreland Publ., p. 151), but has apparently been deleted from the second edition of the book (quoted in note 1) and has not been replaced by any similarly explicit expression of the same thought. However, several statements hinting at more or less the same idea can be found on pp. 83-85 & 142.
 Expressed during an interview I conducted with David Moss in 1990, which was published in the Swedish magazine for jazz, Orkesterjournalen, no. 12, 1990, as "Musikalisk mångfald hindrar förstelning" (Musical variety prevents rigidity).
 The interview was conducted by me, and published as "Improvisationen är en musikalisk metod som följer sina egna lagar" (Improvisation is a musical method which follows its own laws), in Orkesterjournalen, no. 5, 1990.
 For the person who has committed to an ambition to be innovative, the attitude of the free improviser may thereby be seen as a tool which is used in as much as it promotes the aim of the avant-garde.
 The last mentioned goal seems to have been embraced by John Cage, at least as to be judged by a TV-interview made quite late in his life. Cage said that the goal for his creative activity was to change people's conception of and relation to the sound surrounding them.
 See Graham Lock, Forces in Motion - Anthony Braxton and the Meta-Reality of Creative Music, London 1988: Quartet Books. Braxton is an American multiinstrumentalist, composer and a prominent figure within the particular type of improvised music which has grown out of the jazz tradition.
 Dror Feiler: "Improvisation and resistance", in Sounds, accompanying booklet to the double-LP SOUNDS - Contemporary Free Improvised Music in Sweden, Blue Tower records BTLP 01/02, 1990, p. 32. Feiler is a reed player and composer as well as leader of the groups Lokomotiv Konkret and The Too Much Too Soon Orchestra.
 Johan Petri: "Unica", in Sounds, accompanying booklet to the double-LP SOUNDS - Contemporary Free Improvised Music in Sweden, Blue Tower records BTLP 01/02, 1990, p. 12. Petri is member of the group Så Vidare.
 To support this claim in a minimally thorough way would lead much too much astray. Let me just point to one thing. Even if we, in our music-making, do (believe ourselves to do) our best to free ourselves from influences of internalised idiomatic codes, this very aspiration has most certainly sprung from such a code - a way of seeing art and music which we have been made to internalise. Hence, even if we would succeed, the result reached would most adequately have to be described as the outcome of a successfully applied traditional or idiomatic norm.
 Originally written in Swedish, this text was first published as "Vad är fri improvisation?" in the Swedish magazine for contemporary art-music, Nutida Musik, no. 2, 1992, pp. 12-15. The thoughts expressed in it are anticipated in two earlier texts: "Free Improvised music in Sweden", in Sounds, accompanying booklet to the double-LP SOUNDS - Contemporary Free Improvised Music in Sweden, Blue Tower Records BTLP 01/02, 1990, pp. 4-10; and "From things to sounds ... from sounds to things", linear-notes to a CD by the Swedish free improvising group GUSH, From Things to Sounds..., Dragon Records DRCD 204, 1991.
Originally published in