1 Questioning improvisation (first phase of the research)

1.1 Introduction

From my own practice, I would define improvising as inventing (or: composing) and performing music at the same time. This definition focuses on the one moment in which the improvisation actually takes place. The real timeline however will have an antecedent as well as a consequent. The way of preparing and setting goals (in advance), and the way of evaluating (afterwards), will to a large degree determine what the improvisation sounds like. Definitions of improvisation when put into words generally do not differ that much from one another. The musical outcome however can differ greatly.

Let’s compare two writings on improvisation to examine how the word “improvisation” is used: the chapter Improvisation from Das Meisterwerk in der Musik by Heinrich Schenker and Anto Pett’s teaching system by Anto Pett and Etienne Rolin. Both authors do not mention any definition of improvisation. They apparently take this word as commonly understood. Although their ways of proceeding are completely different, as is the artistic outcome, they seem to have the same understanding of the word Improvisation indeed. The differences lie not in their conception of improvisation, but in their ways of preparing, setting goals and choosing criteria for evaluation. In the chapter Improvisation by Heinrich Schenker, for example, we find very sophisticated theoretical knowledge about chord progressions, voice leading and the art of diminution. In contrast to this, in Anto Pett’s teaching system there is nothing about chords or any notated musical material at all. Instead, he describes “the long thinking line” and moods like “furious destroying thunder”.

The aim of this research is to broaden the possible outcomes of improvisation by broadening the view on exactly these issues in which the differences are found.

1.2 Practical situation, beginning and development goal

The practical situation in the first phase of the research was the music theory classroom. In this room the main goal is set already: to achieve knowledge of music theory. But what exactly is knowledge? In music it is obvious that knowledge can be a practical issue. Knowing how to determine or write a certain note in a certain chord is knowledge, but playing this note on the instrument as a reaction to perceived sound at exactly the right moment certainly is knowledge too, and may be a more useful kind of knowledge. To go one step further we could even argue that being able to come up with a musical expression of your own is knowledge, both in interpretation and in creation of new music.

To stimulate and integrate all these kinds of knowledge I used improvisation as a learning tool in music theory class. Learning goals were to integrate theoretical knowledge, musical imagination and auditive and instrumental skills. (At the time I took this step several colleagues at various institutions made the same switch. At the Royal Conservatoire of The Hague improvisation became a standard subject of the music theory curriculum within only a couple of years.)

At the beginning two strategies for improvisation were available: improvising on a given chord scheme, and improvising after instructions in words in a non-tonal idiom.

The following audio examples show these possibilities:

Audio 1

Audio 2

The goal for development in the first phase was to invent strategies for improvisation that would combine the qualities of handling pitch and harmony, and in the second, complete improvisatory freedom.

1.3 Relating to the history of improvisation

In music theory courses, history is a part of the subject. Students study music theory for several reasons, among which an important one is the understanding of the musical repertoire from the past. Internalizing this repertoire becomes more natural for a person who understands its language, and it is this language that can be learned by inventing music in it by yourself. Nowadays, knowledge on how to improvise in historical styles has been made more and more available in rapid tempo. Below, an overview of available techniques from the past and an example of a source is given:

Vocal improvisation in the Renaissance:
Available techniques: Cantare super Librum (singing upon the book), that is: creating a counterpoint to a cantus firmus; canon (=stretto fuga)

Instrumental improvisation in the Renaissance and early Baroque:
Available techniques: performing chord progressions, improvising melodies and figurations on it.

Baroque: figured bass, diminution, embellishment
Available techniques: Creating an accompaniment out of a bass line; embellishing a simple melodic line according to the accompanying harmonies.

Classical period and later:
Available techniques: creating and embellishing chord progressions, creating forms such as a prelude or a fantasy.

Romantic era: Clara Schumann
Available technique: creating preludes to pieces that would be performed right after in a concert program.

Although many different approaches were gathered through the study of historical improvisation, not many of them suited the situation in music theory class. First of all, theory lessons are group lessons. Nearly half of the above mentioned methods are for solo keyboard improvisation. This does not match the needs of a group. The situation in a group is essentially different from a solo situation, since the group’s members have to react to each other’s decisions during their improvisation. How to react and which material to define in advance in order to be able to play together are essential questions in this research, which are not important in the same way to a keyboard player who is improvising solo.

From all historical approaches, the most useful to do in the classroom was improvising on historical chord schemes as given in 50 renaissance & baroque standards. This material is very understandable and flexible. The chosen chords are mostly the most common ones in many styles besides Renaissance and Baroque music, such as modern pop or film music, and are therefore experienced as flexible in style. They can be used in a jazzy mood, in more than one tempo, in a historical way or otherwise. They also can have any instrumentation.

Working with canon was also very productive. It trains the ear, has many possible outcomes, can be adjusted to any level of performing and is very flexible in style.

“Singing upon the book”, as well as the baroque art of embellishment, requires a more developed historical consciousness. The judgement of whether a note is correct or not is therefore often experienced as coming from some place far away. Why this judgement should be made, in other words, why one note is more correct than the other even though for the modern ears of the student both sound good good, requires explanation on the technique of a historical style. It is the choice to be made if this fits in the main goal of the lesson.

For my lessons, studying historical styles was never the only learning goal. Creativity was equally important, as well as the development of the musical personality of young musicians. Improvisation as a historical exercise, that is, with the aim of reproducing something that is correct in a well-defined historical style, thus turned out to be less productive than the other course of action: invent your music through imagination without evaluating if the outcome is somehow “correct”, but instead consider whether you have expressed what you wanted to say in music. In group sessions, of course the last comment becomes even more interesting because of the presence of more than one opinion in a group. Towards the end of the research, creativity became the center of all improvisation activities.

The experienced difference between improvisation with the aim of gaining knowledge on a historical style and improvisation with the aim of developing creativity means that the stylistic outcome of an improvisation lesson will not be historical by itself. Also, spontaneous evaluation by students usually differs from historical choices. Therefore historical styles turned out to differ from the natural outcome in expression of most sessions.

Reproducing historical styles and developing creativity are different learning goals. How to proceed is a choice: if there is a difference in experiencing tonality as used in historical styles and tonality as used in a contemporary way, one could choose to develop the consciousness of the historical by taking intuitive tonal knowledge as starting point, or to develop personal styles of improvisation using creativity. The creativity has won out in my courses more and more frequently as the research proceeded. In fact, in the final phase of the research, historical styles do not play any role anymore and the focus is completely on providing tools for everybody to become able not only to perform improvisation after instructions but also to consciously create these instructions (i.e. the open form score) themself.

1.4 Development in the first phase: five new approaches

As an answer to the question how to broaden the perspective on improvisation strategies five new approaches were found:

1
Change a pedal note after a signal

Short description

Improvisation in a group:

one person plays a low sustained tone: a pedal note.

All others play melodies freely. The melodies can fit the tonality of which the pedal note is the tonic.

The pedal will change after a signal. This signal can be decided on by the group and can be anything. Suggestion: the signal is an emphasized melodic note that fits the scale. As soon as this note is emphasized, other players will play it too, and as soon as it is very much on the foreground, the player playing the pedal will change to a different note as pedal. This will define a new scale on which the other players will invent new melodies.

2
Create your own improvisation model using counterpoint 

Short description

Improvisation for 2 instruments: one melody player and one accompaniment player.

This exercise consists of two steps. 

Step one: Both players agree on a time signature and tempo. The melody player starts playing one note on the first beat of the first bar. In real time, the second player invents a bass note to it that is a consonant and does not cause a disturbing parallel and plays this note on a previously defined moment in the bar. The next bar is created the same way, till 8 bars are created. These 8 bars are to be memorized by both players and serve as a model in step 2.

In step 2, the model is used as improvisation model like a chord scheme.

3
Melody rules over harmony

Short description

Improvisation for one melody player and one accompaniment player.

In this exercise, the melodic player determines the harmonic progressions. This is done by omitting the harmony on every first beat. On the first beat of each bar, the accompaniment player just plays a rest or a fixed pedal note. The melody player is free to play any note. The accompaniment player has to react to the choice of the melody player. A list of possible chords to each melodic note can serve as a help.

4
Melodic motifs as a signal

Short description

Improvisation for a melody player and an accompaniment player.

This exercise can be considered an advanced stage of the previous exercise. Instead of listening to the melody players choice on the first beat of each bar, the accompaniment player can predict the chosen note that will appear on the first beat when the melody player uses motives.

A melodic motif will be agreed on in advance by both players.

5
Alterations as signal

Short description

Improvisation for one melody instrument and one accompaniment instrument.

Again, like in exercises 3 and 4, the player of the chord instrument has to listen and react to choices made by the player of the melody instrument. One new principle is added: the alteration. In advance both players will agree on which alteration will occur and what the consequences for the harmonic progressions will be. 

In the following audio examples is shown how two students (violin and piano) proceed in approach 2. The first example shows the phase of creating the model after the given instruction and the second shows the improvisation on this model.

Audio ex. 3

Audio ex. 4

The created model, on which the improvisation is made, in musical notation:

Ex. 1

1.5 Conclusions in the first phase

Although five new approaches were formulated and a clearer idea about stylistic outcome evolved, the results were in one respect not sufficiently satisfying: they were not suitable to present in a concert. Many of the learning goals were achieved. The slight renewal of the approaches formulated as five new approaches broadened the possible outcomes in a way that was productive in the music theory classroom in the sense that they were useful exercises. But only doing exercises leaves us with the question of what activity the exercises are meant for. In other words, the wish evolved to create something that would enable a stage performance in which the achieved abilities were needed.

The answer to this desire became the next step in the research. It was to create an open form score that would provide a framework for improvisation and thus enable improvising musicians to improvise on stage without losing the presence of structure and coherence in the musical work.