Musical score by Sedje Hémon (1923–2011), recoloured in VectorQ

Does improvisation improve piano skills?

The crown of all our study and the highest reward of our long labours is the power of improvisation.
Quintilian, Institutio Oratoria

They played the song many times, and with every repetition the song was involuntarily enriched with embellishments and variations.
Joseph Knecht’s first trial encounter with the Music Master in Hermann Hesse’s Glass Bead Game took the form of an improvisation

Sometimes it happened that the chosen theme accidentally yielded, in embryo, out of its folds, a line or a rhythm whose developments could be treated according to a more independent and less scholastic system. It was then a question of which young disciple would be the first to dare break the fetters, pry open the door to the ideal, and leave with his eyes blinded by such a brilliant light.
Léon Kreutzer’s Diary, on the improvisational composition classes of Abbé Vogler (1749–1814), from Fantasies of Improvisation by Dana Gooley (2018)

Musical improvisation

There are many competing definitions of musical improvisation. In a comparative review of musical improvisation in different cultures, Nettl (1974) noted that specifically or implicitly accepted in all definitions is the suddenness of the creative impulse: ‘musical improvisation is when a performer makes creative choices during a performance.’ [i]

Would we expect a child to develop proficiency and a full range of language skills by teaching them only to read and recite from a text, and never to try to listen and speak? Doesn’t a child learn to speak, in practice, by experimenting with all kinds of oral sounds and nonsense noises before discovering those with meaning? Musical improvisation as a means of expression is analogous to speaking in language, and the analogy should carry across between learning to make music and learning to speak.

You can learn to recite from memory something in Czech or Chinese, for example, without speaking the language at all, and with more work you could learn to use the written script as a visual aid. But if you are conversationally fluent in the language it will be much easier to learn and more authentic to recite. Music is another means of self-expression, and the analogy holds true that improvisational ability in a particular genre can make performance from memory or notated music so much easier and more genuine. In musician and writer Giorgio Sanguinetti’s words, ‘to “speak” eighteenth-century musical language [for example], we need something going much deeper, down to the level of what we call “musical instinct.” This was the exact purpose of the rigorous partimento training the Neapolitan maestri imposed on their novices: to make music a native language.” The same can be said of any genre, whether classical, traditional, or popular, from French Baroque to Early Modern, Balkan folk to Delta Blues: developing a native language capability will facilitate learning and authenticity.

In this essay we will begin by reviewing the long history of improvisation in the western classical music tradition, and see that improvisation used to be a pillar of music education and performance until the late nineteenth century. We will see how modern general theories of education champion properly-supported pupil-led learning, how training in other performance disciplines such as theatre, ballet, and oratory includes improvisation, and how prominent recent music educators have researched and stressed the importance of improvisation, and yet current common practice in individual instrumental tuition in classical music is an outlier in rarely using improvisation at any level of training. We will look at four recent piano courses for beginners which do use improvisation, consider how improvisation can be built into lesson plans around other course materials and objectives, weigh some pros and cons, and summarise some empirical research on the effect of including improvisation in music lessons.

Improvisation in the western classical music tradition

Improvisation was once an important, if not essential, component in the classical tradition of music pedagogy and performance in Europe.

Aspects of what we would now call improvisation featured in the music of medieval times, and continued into the Middle Ages (Treitler, 1991) and early Renaissance, at which time musicians rarely worked from fully written out scores, instrumentalists would turn a single bass line annotated with numbers into a full harmonic foundation, and improvisation was a standard aspect of both musical training and performance. (Gould, 2000)

J.S.Bach’s first early biographer, Johann Nikolaus Forkel, tells us that ‘the organ compositions of J.S.Bach are great, but his improvisations on this instrument were even more fervent and solemn, soothing and sublime.’ (quoted in Jaques-Dalcroze, 1932) A contemporary report from Bach’s Leipzig years describes him beginning performances with a written composition to ‘set his powers of imagination in motion’ before proceeding to improvisation. (Schulenberg, 2020)

Handel’s music education ‘included the development of improvisational skills requiring the memorisation of musical formulas that he could retrieve at the moment of performance’ (Sung, 2016) and he too was renowned for his organ improvisations.

Later, C.P.E. Bach was especially famed for his exceptional power of improvisation, which at the time was widespread among professionals and amateurs alike (Richards, 2001). His expressive, stirring style is associated with Haydn (Gooley, 2018), for whom keyboard improvisation was ‘a way of arriving at ideas’ (Clark, 2019), and also with Beethoven, who was famed at the time for his improvisation. Heinrich Schenker, an early twentieth century theorist, praises the ‘miracle of improvisation’ to which he credits the unity and synthesis in compositions of Haydn and Beethoven: ‘only invention in the spirit of improvisation is granted the unity of composing-out.’ (Bent, 1986)

For improvisation on a single theme, Czerny credited Beethoven as being unsurpassed: ‘through the entire improvisation there ran through the middle voice like a thread the notes he found on the accidentally opened page of Pleyel’s quartet.’ (Sitton, 1987) Sometimes these improvisations later formed the basis of notated works, for example the Choral Fantasy which began with an improvisatory piano solo that Beethoven only later fixed in notation when the work was published. After 1808, following his retirement as a performing pianist due to his deafness, he seized greater control of cadenzas in his written works and foreclosed the opportunity of performers to improvise, even while continuing to develop the improvisatory quality of the written work. (Kinderman, 2009)

Another line of improvisational practice has been proposed around Mozart and Clementi (Gooley, 2018), characterized by patterned figuration idiomatic to the piano, variations on binary phrase melodies, free form slow fantasies and introductions, and capriccios. Gooley groups a further line of practice around Abbe Vogler, an organist and pianist who in his time was perhaps even more famed than Beethoven for his improvisations, and who prominently included improvisation in his teaching of gifted and musically progressive improvisers and composers among his pupils, such as Weber and Meyerbeer.

Schubert is said to have delighted in improvising waltzes at the piano for informal dance gatherings (Aldrich, 1997), and his compositional approach is supposed to have been improvisatory in nature (Nettl, 1974).

The young Schumann, attempting to follow Hummel’s Klavierschule in preparation for a performance of the Hummel piano concerto, in 1829 confessed his general habit in piano practice was to do much improvising and play little from the score (MacDonald, 2002). His improvising made a great impact in the salons, and Schumann’s companion Anton Topken records how, ‘out of a single thought, which he made appear in all different guises, everything streamed and poured forth as if from within itself.’ As late as 1839 his compositional process was still dependent on improvising, though in his late career Schumann distanced himself from the improvisatory approach to performance and composition. (Gooley, 2018)

From an early age Chopin was a prodigious improviser: ‘From the most tender age he astonished by the richness of his improvisation. He took good care however not to parade it; but the few chosen ones who have heard him improvising for hours on end, in the most marvellous manner, without ever recalling a phrase from any other composer, without even touching on any of his own works — those people will not contradict us if we suggest that his finest compositions are only the reflections and echoes of his improvisation.’ (This is according to Chopin’s contemporary Julian Fontana, commenting on Chopin’s posthumous pieces in 1855, cited by Kallberg, 2001) Chopin improvised regularly in public before 1830, but after that only in private. (Gooley, 2018)

Liszt’s relationship with improvisation was ‘up and down. Free improvisations were central to his reputation as a child prodigy, yet the more he sought recognition as a composer, the more compelled he felt to rein it in for fear of being judged as superficial.’ (Gooley, 2018)

Ignaz Moscheles was famed for his improvisation, and his pupil Mendelssohn was one of the last of the great improvisers of the period. In 1837, Hiller wrote about a concert where Mendelssohn ‘improvised, combining in the cleverest way a theme of Bach’s with his own well-known “Song without Words” in E (no.1 bk.1) — thus uniting past and present into something new and difficult to describe.’ (Hiller, 1874) Another renowned later Romantic improviser was Busoni, who improvised early in his concert career, but not in his mature years. (Gooley, 2018)

There were many purposes served by improvisation — most prosaically, the performer’s need to test an unknown piano, or alerting the audience in a salon or concert hall to the start of a performance, as well as more aesthetic reasons such as attuning the audience to the mood and key of a forthcoming piece, or directly displaying the virtuosity and creativity of the performer. Nevertheless, by the end of the 19th century, the vogue for piano improvisation had passed, along with its routine inclusion in musical education and performance. Gooley lists the causes as ‘the growing authority of the “work” concept among composers and audiences, the emergence of a guilty conscience about improvisation among piano virtuosos, and the growth of a large, general listening audience insufficiently informed to understand the nuances of improvisation.’ Beyond musical considerations, he explains the emerging anxiety over improvisation in terms of a reaction to ‘certain anti-economic impulses — a dilated sense of temporal unfolding, a strenuous type of performer training, and a risk of inefficacious communication — that ran counter to bourgeois ethical codes such as the containment of excess of waste and the rational ordering of available resources.’

In the context of church organ playing, the tradition of improvisation has continued unabated, through noted improvisers such as Saint-Saens and Bruckner, to the present day where church organists are expected to be able to improvise to ‘fill the gaps’ as required in the moment between parts of the ceremony. (Gooley)

Whatever the cause of the decline in classical piano improvisation, it is clear that there is a discontinuity between the earlier and later western classical music tradition regarding improvisation, and its place in training and performance within that tradition.

General theories of cognitive development and their application to student-led learning

A music lesson which includes improvisation is an example of student-led learning in general: it is the improvising student, and not a teacher or musical score, which determines the actual detailed content of the activity.

John Dewey, Jean Piaget and Lev Vygotsky each contributed to the development of the social-constructivist theory of learning, an influential movement which holds that learners construct knowledge for themselves, individually (and socially) constructing meaning as they learn. It has not completely replaced the ‘behavioural’ method of teaching by rote-learning and teacher-led instruction, but most modern curriculums in the West at least include elements of student-led learning, and in some curriculums it is the dominant learning strategy (e.g. International Baccalaureate (IB), Steiner, Montessori, Reggio).

‘When students have freedom to explore and construct their own understandings, the possibilities for student growth improves. [This pedagogical movement] worked to combat the rote recitation that was common in the latter part of the nineteenth century. These reform efforts appear to create a balance between doing and receiving, which Dewey notes is vital for quality experiences.’ (Cohen, 2000) Music was part of the Dewey system of education, but rather than being focused on developing performance competence on a particular musical instrument, it was more as a kind of communal activity, a means of reaching transcendence through aesthetic appreciation, which is held to be a part of realizing a healthy and holistic life experience.

Jean Piaget’s stages of learning are also general to all types of cognitive development, but a particular application to the teaching of ballet to young learners is considered below, and this can be reapplied in the specific context of music education.

The extent that direction in improvisation may be given by the teacher is a matter of degree. It was Vygotsky who drew a clear distinction between the development of individuals left entirely to their own devices, and learners who develop within a framework provided by more able peers or teachers. This framework was shown to be important in getting the most benefit from student-led learning. The supporting framework must allow scope for the learner to apply their creativity, and examples of how this can be done in practice in the context of music students are considered below.

Use of improvisation in training for other performance disciplines

Improvisation is a training method for actors, and is one strand of ‘The Method’ based on Konstantin Stanislavski’s approach of being an ‘active representative’ in a sequence of dramatic situations which are improvised in rehearsal, with the goal that actors should really feel what they portray at every performance, be it the first or the thousandth time. Viera (1990) describes improvisation in acting classes as designed to ‘to find ways for the actor to draw upon personal experiences to create honest reactions to new situations.’ She notes it is also used as a rehearsal technique ‘to work on dead places, places in the script that are not working… to get more intense, more inventive, less predictable performances… to get to the polished, perfected performance.’

Interviewed about the role of improvisation in preparatory work for theatre, Peter Brook says: ‘Improvisation is a fantastic learning activity for those who are looking for a greater degree of freedom in their work, but it has to be remembered that, if it is conducted without any sense of discipline or rigour, anything goes and it means nothing at all.’ He draws an analogy (without source) to Japanese archery to illustrate his view of the role of improvisation in theatre performance: ‘The archer improvises every time he draws and fires his bow: it’s a new event, nothing is given, nothing is fixed. The only real event occurs when the arrow strikes the target. If the archer improvises in a clumsy or insensitive way, the event misfires, as it does for an actor who wastes his lines. But with good improvisation and a certain grace, an archer can fire with his eyes closed.’ (Brook, 1992)

Classical ballet training uses improvisation with young learners, using wands, streamers, scarves, or simply rhythmic body movement to encourage body awareness and coordination. However, it’s soon phased out in favour of rigorously disciplined routines and repetitions.

In her analysis of dance education, Stinson (1985) includes various teaching techniques using improvisation, corresponding to Piaget’s stages of development.

• At the earliest ‘sensori-motor’ stage from birth to two years, before they can respond to instruction, simply allowing a child to improvise ‘while the adult responds by watching intently, smiling, imitating the child, and naming the movements and/or the body parts performing them,’ to enhance positive feelings about the self and movement.
• At the next ‘pre-’ stage, through about age 7, teachers may invite imitation of objects or events, at early ages pretending to be something (e.g. ‘pretend to be thunder and lightning’), and later like something (e.g. ‘cats go lightly on their feet; can you touch lightly?’).
• At the stage of ‘concrete operations,’ between around age 8 to 11, improved language skills and motor skills allow for richer exchanges of information between teacher and pupil, allowing a passage to be selected ‘to try out different variations of the sequence according to the teacher’s directions, perhaps doing it faster, then slower.’ Piaget would recommend that teaching for adolescent and adult beginners be pitched at this level.
• The final stage of ‘formal operations’ may be reached by students with a suitable depth of interest and experience, at which point, alongside the capacity for logical and systematic approaches to performance practice, they have the capacity to imagine and realise all sorts of possibilities and experimentation.

Stinson notes, however, that avant-garde creativity does not always flourish in dance classes, and that lack of an appropriate curriculum addressing improvisational techniques may be responsible: ‘Far too often in the dance curriculum, stimulating the imagination is seen as significant for young children only: creative dance classes stop at age eight (or even earlier), when dance technique begins. As dance educators, we must encourage an integration of creative development with development of body skills throughout the educational process.’

Dance theorist and psychologist Philippa Ziegenhardt encourages adult ballet dancers to improvise to gain a greater capacity for expressiveness, to balance the rigidity of classical technique training to achieve freedom of movement and an appearance of effortlessness, and to improve mindfulness and creativity. (Ziegenhardt, 2019)

On the other hand, van Delinder (2000) stresses the importance of technique as a foundation for successful free expression: ‘the rigorous discipline of ballet, once mastered, sets the artist free to pursue the emotional expression that is a result of dancing,’ citing dance theorist Anna Aalten’s conclusion that women in ballet can ‘experience, even for a moment, the synchronizing of physicality, willpower and emotionality.’

The influential 19th century writer on piano education and technique, Carl Czerny, compares a performer improvising on the piano in front of a large audience to an orator ‘who strives to develop a subject as clearly and as exhaustively as possible on the spur of the moment.’ The arts of rhetoric and music were two of the seven liberal arts which formed the basis of liberal arts education in Western society from the time of Plato to the Middle Ages, the two performative disciplines share many fundamental concepts and technical terms, and educated musicians would have been deeply familiar with rhetorical concepts.

Improvisation is explicitly valued in the art of rhetoric and oratory as more credible, more engaging, and more authentic than speech meticulously prepared in advance, and is considered the mode of performance to which all oratory should aspire. The classical Roman author Quintilian rejects the sufficiency of natural ability to support successful improvisation in oratory, parodies artless attempts at extemporisation, and promotes his own educational program based on study, training, and art. For Quintilian, improvisation is the mode of performance to which all oratory should aspire. (Holcomb, 2001)

In Institutio Oratoria, Quintilian suggests methods to develop improvisation skills include acquiring ‘a store of the best words and phrases on lines that I have already laid down, while our style must be formed by continuous and conscientious practice in writing, so that even our improvisations may reproduce the tone of our writing, and after writing much, we must give ourselves frequent practice in speaking. For facility is mainly the result of habit and exercise and, if it be lost only for a brief time, the result will be not merely that we fall short of the requisite rapidity, but that our lips will become clogged and slow to open.’ Once this facility is achieved, he notes the necessity of developing a certain ‘mechanical knack’ which must be developed ‘by gradual stages from small beginnings, until we have reached that perfection which can only be produced and maintained by practice.’

Many of these contrasting themes from theatre, ballet training and oratory — discipline and rigour versus honesty and intensity; routines and repetitions versus freedom and effortlessness; study, training, and art versus credibility and authenticity — are paralleled by the mutual relationship between training in piano technique and inspiration in performance, as pursued by piano educators from the early 19th century to the present day. Yet oratory, classical ballet and theatre all accommodate improvisation in their standard training methods. In contrast, even though improvisation is often used especially in early years music classrooms where its efficacy is acknowledged, it exposes current common practice in individual instrumental tuition in classical music as an outlier in rarely using improvisation at any level of training.

Improvisation in prominent methods of general music education

Music educators have made use of improvisation in support of various objectives, including development of skills in group dynamics and social behaviour which, like Dewey’s wider holistic and societal goals, are not explicitly musical. While some of these innovations relate to the music classroom group context, there are aspects which are also applicable to individual teaching of instrumental performance.

Emile Jaques-Dalcroze (1865–1950) developed a system for teaching music which integrates training in body motion with training in pitch and music theory. The system culminates in improvisation, which is considered ‘the proof of the pudding of the learning skills that have been accomplished and provides a base-line for further study and experience.’ (Abramson, 1980) Acquisition of instrumental technical skills is part of the kinetic training in this method, and musical awareness is engendered by ear training, sight singing and study of musical theory. But improvisation is not only the apex of a learning triangle, it is also taught as an integrated and indispensable part of the method. Jaques-Dalcroze stated that ‘improvisation is the study of the direct relations between cerebral commands and muscular interpretations in order to express one’s own musical feelings. Performance is propelled by developing the students’ powers of sensation, imagination, and memory. It is not based on direct imitation of the teacher’s performance.’ (cited in Abramson, 1980) As Covington (1997) notes: ‘Jacques-Dalcroze was amazed that improvisation received so little attention in applied instruction, for he believed that it “established direct communications between the spirit that pulses, the brain which represents and co-ordinates, and the arms and hands which put into execution.”’

Carl Orff (1895–1982) acknowledged the influence of Jaques-Dalcroze in his Schulwerk form of musical learning through creative play, using the special smaller scale Orff xylophones of wood and metal commonly found in classrooms, which allow removal of notes to allow euphonious free play with drones, pentatonic and modal scales. An Orff-influenced music curriculum ‘needs to reflect the commitment to learn about music not only by imitating musical examples, but by exploring and improvising as well. […] As students acquire skills to play, sing, move, read and write with each step of their learning they also develop the potential to generate their own ideas through improvisation and composition. Ideally, each lesson should contain an element of this highest level of understanding of even the simplest music concept.’ (Steen, 1992)

The popular Shinichi Suzuki (1898–1998) method’s rote-before-note learning is done under the close direction of the instructor, and is not a type of improvisation by our definition. Equally, the aspect of Suzuki teaching known as ‘Mother Tongue Method’ which holds that repeated hearings of the music to be learned is the most important facet in music learning, much as young children learn speech (Shehan, 1986) is focused on the faithful copying of an existing model, rather than production of anything original. In the analogous context of learning speech, this would be to try to learn language strictly only by mimicry, rather than ever attempting to engage in conversation and original speech production and learning from the resulting feedback.

Edwin Gordon’s (1927–2015) influential empirical research into learning of music in early childhood led him to his conviction that ‘it is prudent to introduce students to improvisation as soon as possible, long before they are burdened with formal learning of music notation and music theory.’ (Gordon, 2011) Drawing a comparison between young children learning music and language, and perhaps with reference to the Suzuki method of teaching music, he states rhetorically that we should stop speaking to newborns and just play recordings of persons speaking English: ‘If that sounds like a ridiculous way to learn a native language, why then apply such inanity to learning music.’ Equally, would we expect a child to develop proficiency and a full range of language skills by teaching them only to read and recite from a text, and never to try to speak?

Improvisation in published methods of piano education

In the early nineteenth century, the golden age of piano improvisation, methods of piano instruction were a popular vehicle for the burgeoning music publishing industry. Authors such as Gretry, Czerny and Kalkbrenner published treatises on the subject, but were not renowned improvisers. On the other hand, admired improvisers such as Vogler, Hummel, Moscheles and Liszt published little on the subject. (Gooley, 2018) Hummel’s comments on improvisation are relegated to the last chapter of his popular three volume piano method, perhaps reflecting a view similar to Jaques-Dalcroze that improvisation itself is a culmination of musical technique and understanding, from which it emerges and flourishes with exploration and discovery.

The Greek word Quintilian uses for improvisation in oratory is φαντάσιαι (fantasia), which gives Czerny the German word he uses in the title of his systematic introduction to improvisation on the pianoforte (Systematische Anleitung zum Fantasieren auf dem Pianoforte, 1829), which covers different kinds of classical piano improvisation, from prefacing a performance with a short prelude, to extemporisation of a full fantasia developed over many sections. He starts by giving an example of a brief sequence of chords, and showing how it can be ‘spun out into innumerable preludes, both melodious and brilliant.’

In the first example of developing the theme shown here, he does little more than arpeggiate the chords, while in the second he interpolates related passing chords.

To make use of such examples in improvising, he adds: ‘one must transpose these and similar examples into all keys, alternate the passagework with other suitable sections, and know how to execute everything with such ease and lack of restraint that the preludes maintain the character of the momentary fancy. For nothing is more disturbing to the effect than the recognition that it has been drilled into the performer.’

Later published piano methods have addressed improvisation, in spite of its general decline as a mainstay of classical piano teaching and performance. For example, Carl Faelten of the New England Conservatory published a treatise on keyboard improvisation in 1908. However, the majority of published classical piano methods and manuals for instrumental teachers published in the twentieth century have concentrated on learning fixed pieces from notation, without any mention or attempt to introduce improvisation.

Meanwhile, a number of instructional publications have addressed jazz keyboard improvisation for beginners (e.g. ‘Jerry Coker’s Jazz Keyboard’, 1984) and more advanced learners (e.g. a series of workbooks by Jamey Aebersold), alongside a range of compositions and transcriptions of performances by jazz pianists which can be learned by students as fixed pieces.

In very recent years, there has been a revival of interest in classical music improvisation and some performers are seeking to differentiate themselves through introducing extended improvisation into performances. Gjerdingen (2019) has examined a wide repertoire of works in the ‘galant style’ and codified it into basic building blocks. He then presents these systematically to the learner, along with a grammar by which they may be joined into larger units, to reuse and recombine in an improvisatory context. All examples are presented as excerpts which can be realised on the keyboard. Mortensen (2020) is somewhat similar in providing a vocabulary, drawn from a wider range of models (including a few from Gjerdingen), which can be used to build larger structures, while also encouraging the learner to gather and develop their own materials into a zibaldone — a ‘heap of things’ or musical notebook intended to provide useful materials for improvisation, in the same manner suggested by Quintilian in the context of oratorical improvisation.

Improvisation in modern piano primers

Most modern piano primers do not address improvisation in any way, so teachers interested in including improvisation in their approach must build it into lesson plans around available materials. Ruiter-Feenstra (2011) cites four relatively recent piano methods for beginners which include improvisation in their approach, and invite further attention from interested teachers.

Music Pathways (Olson et al, 1983) has a ‘you the composer’ exercise in every section of the entry level Piano Discoveries books. The suggestions (e.g. developing an improvisation from a particular note, interval, or rhythm, up to using ‘anything you like’ in the final exercise of Book A) are intended to ‘serve as springboards for the student’s own ideas. Because these creative pieces may involve tonalities, rhythms and elements of expression for which the student has no notation, it is suggested that the compositions not be written.’ The improvisations are unaccompanied.

The Music Tree (Clark et al, 1973) includes improvisations and extempore compositions (‘Composer’s Corner’) throughout the elementary library for piano students. The introductory Time to Begin primer begins by asking the young pupil to use the groups of two black keys to ‘make up some pieces of your own … about a frog jumping, rain falling, an elevator, or whatever you like.’ Some improvisations are accompanied by the teacher, requiring the pupil to listen and fit the improvised melody to the pulse of the accompaniment, rearranging and trying notes and rhythms in a different order. Having introduced notation using a ‘landmark’ approach (i.e. around middle C, left hand F and right hand G), by the end of the Part 1 book the pupil is being asked to complete an initial idea for a simple piece given in full piano score with ‘an idea that is as different as possible’ from the initial idea, and is given an overall structure to make a longer piece in ternary form with six sections, presumably reproducing the self-composed section from memory. By Part 3, the early intermediate stage in the series, the pupil is given, for example, an arpeggiated pattern in D, and asked to ‘complete the piece by playing the given pattern on IV, V and I,’ or is given the first and last two bars of a piece with an arpeggio-style bass and simple melody, and asked to ‘continue the RH melody in your own way by choosing triad tones of the chords [notated in jazz chords] in measures 3 through 6.’

Music Moves for Piano (Lowe, 2004) is a radical piano method based on Edwin Gordon’s research. The meticulously programmed lesson plans follow a ‘sound to notation’ approach which initially prioritises ‘audiation’ (singing and playing by ear) over reading notation. Not until Book 3 will a student start to look at musical notation. Each lesson unit includes objectives around ‘Exploration/ Creativity/ Improvisation.’ In initial units the pupil develops single-note and single-rhythm ideas, and by the end of the first book improvises using tones from named major or minor triads, in contrasting binary and ternary forms. By the end of Book 4 the pupil is already performing more difficult solo repertoire, and improvising by changing tonality, meter, creating new music based on remembered tonal and rhythmic patterns, and creating melodies over a given chord progression. Examples from the end of Book 4 include: ‘improvise a G mixolydian melody in unusual meter in five, notate the melody, and make a melodic variation of this melody,’ and ‘improvise a B dorian melody in triple meter.’ The method is very different from the norm, and would be best suited to being followed consistently as an entire course, which wouldn’t initially transfer to, or comfortably sit alongside, other teaching methods. The supporting material the teacher would need to be familiar with to follow the method faithfully is very substantial, as might be expected from a method led by empirical research.

The Hal Leonard Student Piano Library (Kreader et al, 1996) includes accompanied improvisations and notated exercises, requiring the teacher to watch the pupil while playing to observe posture and hand position. It follows a ‘score to sound’ approach which introduces notation through a landmark approach, and recognition of steps, skips and other intervals. Improvisation begins at lesson one, and helps pupils to familiarize themselves with keyboard layout: ‘With your right and left hands, choose any groups of two black keys in the upper part of the piano. Listen and feel the pulse as your teacher plays the accompaniment below. When you are ready, play along and make up your own song. Have fun!’ Another early example uses the notes C D and E over a flowing accompaniment. This ‘My Own Song’ section continues through the series, with relatively few in Book 2 for some reason, until in Book 5, for example, pupils are asked to improvise sequences based on a motive using notes from a major or minor scale, and develop question and answer phrases using parallel and contrasting melodies. Scales, triads and arpeggios are introduced in Book 3, which by Book 4 are given in full score and jazz chord notation.

Also in the Hal Leonard Student Piano Library, the same authors have also produced Scales, Patterns and Improvs (2009) dedicated to improvisations using five-finger patterns, I-V7-I chords and arpeggios in all 24 major and minor keys. In each key, the pupil is asked to rehearse the five-finger scale, broken and blocked triads, a I-V7-I cadence in each hand, an arpeggio, and a simple five-finger melody which might feature mirror and parallel scalic passages, while the teacher accompaniments each have different textures and tempos over which the pupil is invited to improvise a melody playing hands separately or together. It’s not a standalone piano method, but can be used alongside the authors’ Piano Lessons books at about Book 3 level, which introduces triads, arpeggios and similar improvisation exercises involving five-finger patterns.

Setting up, developing and assessing improvisation in one-to-one piano lessons

General theories of education indicate that to get the maximum benefit from using improvisation in lessons, teachers should have a unified plan. ‘Teachers who use improvisation as a teaching technique must carefully plan all the variables of the activities instead of improvising their actions. Teaching improvisation involves designing, monitoring and evaluating. An initial aspect regards defining a context for improvising, providing a framework to the students and asking them to improvise within determined limits. It is a comfort zone in which students can experiment and discover the extent to which they can make music respecting these limits.’ (Biasutti, 2017)

The inclusion of improvisation in lessons in different ways will have a correspondingly different impact on the development of wider musical awareness and skills of the learner.

Bradshaw (1980) notes that improvisation can be stimulated by either words (e.g. the instruction of the teacher), sounds (i.e. listening to oneself or others), or reading from notated (or partially notated) music, and we shall later consider examples of all of these. But there is also the purely physical aspect of the instrument itself, its topology and the muscle memory used in playing it — what Schumann might have characterized as improvising or composing ‘with the fingers’ rather than ‘with the head’ (Gooley, 2018) — which emerges from the technical approaches to improvisation advocated by Czerny and Mortensen, for example. To make a pianist a good improviser, writes Jaques-Dalcroze (1932), there must be developed ‘automatic powers of the instinct, the product of countless repetitions of our sensations and feelings. The pianist must reach the state at which he plays without concerning himself about the way in which he plays, just as the child walks without counting his steps or exercising conscious control over feet, knees or thighs.’ This is the same apparent tension we saw in theatre, ballet and oratory, in that it is precisely the technical mastery and control which gives rise to the possibility of inspired performance. ‘Technique is freedom,’ as the dancer and choreographer Martha Graham famously said.

Teaching improvisation by working through Czerny’s op.200 technical exercises will have a very similar impact on finger dexterity as any other kind of technical etude. Conversely, practicing Hanon’s Virtuoso Pianist in Sixty Exercises, especially its sequences of scalic passages modulating directly into relative keys, and its completion of scales with a II-Ic-V-I sequence of cadences in the home key, or transposing basic exercises into other keys, will help to develop a harmonic understanding and technical facility which can be used in improvisation. Though this is not the stated purpose of Hanon’s exercises, practicing them as material to be used in improvisation may provide a more meaningful context for some learners to perceive the exercises as more than dry routine work. In this sense, ‘teaching improvisation’ is in many ways identical to teaching physical technique, though while it may have a similar outcome in motor skill acquisition, it will have a larger impact in musical awareness because the exercise is placed in a more meaningful musical context. As Azzara (1993) notes, ‘a common thread in many of the improvisation curricula for instrumental music education is the use of patterns and scales as resource material. Although this usage is prevalent, patterns and scales are often taught with little regard for an individual’s comprehension of tonality and meter.’

Exercises from Czerny or Hanon, or any pattern or excerpt from a piece being learned, can be taken as a basis for development in other keys, different rhythms, varying textures, and when committed to memory become part of the improviser’s stock repertoire which can be reapplied and adapted according to the context. For example, Mortensen’s Guide to Historic Improvisation (2020) takes the opening bars of the Prelude in C from Bach’s 48 as his first exploratory exercise, asking the learner to reduce it to block chords, transpose it into 24 major and minor keys, realise it with different chord positions and inversions, in three voices instead of five, in different patterns of figuration and so on.

As Czerny observed: ‘The art of extemporaneously spinning out a theme into an entire piece in an interesting and orderly fashion appears to be more difficult than it actually is, if one observes the following: every theme, without exception, and even if it were to last for only two arbitrary tones, can serve by means of several modifications in meter and rhythm as the opening of all species of compositions that exist in the realm of music.’

In fact, any musical learning objective can be explored further through the use of improvisation:

• A new rhythm can be clapped, and used as a basis for clapping improvisation through variation, speeding up, slowing down, or used as a rhythmic foundation for more inventive variations.
• A new chord or harmony to be studied can be revoiced, arpeggiated in different ways, transposed into different keys or ending on different cadence.
• An ostinato pattern such as an Alberti bass, or waltz pattern in the left hand, can be isolated and repeated, and then the right hand can explore the appropriate scales through independent figuration.
• A bass figure can be isolated as a ground bass and used as a foundation for exploring its harmonisation in various ways. J.S.Bach himself taught his students using this ‘doctrine of variation.’ (Schulenberg, 1995)
• Duets, with the teacher supplying an ostinato, or a ground bass or harmonic sequence, can also be a way of establishing confidence in improvised figuration above a given figure in the bass.

Many teachers may use such learning techniques as a teacher-led activity, and the difference may be only in the extent to which variation is allowed to be student-led, and deliberately and routinely planned to be included in each lesson. [ii]

Other ways of framing improvisation exercises around study of notated repertoire include inviting the student to improvise a prelude to a piece being studied, perhaps drawing on material from the piece being studied. In this case, it may be evaluated by the teacher and student on criteria such as being idiomatic.

In a freer style of improvisation for young learners in Piaget’s earlier ‘pre-operational’ stages, or more free-form improvisation with learners at later stages, students can be invited to improvise to a direction, mood, or title, in the manner used by students in ballet and theatre. For example, Abramson (1980) suggests mimetic piano improvisations, in the manner of Jaques-Dalcroze, which find sounds for movements, and movements for sounds, which ‘allows the invention of many original keyboard techniques’:

‘The teacher makes a gesture, then the students make a new one.
Teacher: Sticks out tongue, then plays one hard, tight, staccato dissonant cluster.
Student: Shrugs shoulders, places palms of hand on piano and shimmies hands upward and downward softly and legato.’

Abramson adds: ‘It is important for the teacher to elicit and project variety in expression of tempo, dynamics, duration, direction, and articulation. Otherwise the improvisation will be dull and unmusical.’

As Cohen (2000) remarks when commenting on Della Pietra’s (1997) research, his teacher-led feedback misses an opportunity for the students themselves to reflect and connect with each other’s performances from their own perspectives, and introducing meaningful connections to their own lives, which in the context of rhythm might be the temporal rhythms of nature and the human body.

At the earliest ‘sensori-motor’ stage, the same ‘loose play’ approach as used in ballet can be used in music, allowing the child to freely explore sounds while the adult describes the sounds, objects and gestures producing them.

Positive and negative impacts of including improvisation in lessons

There will surely be trade-offs from including improvisation in lessons — there are only so many minutes available in a lesson, and more time spent on improvisation is less time spent elsewhere.

The positive impacts of including improvisation in lessons include the following:

• Children love to improvise (Jaques-Dalcroze, 1932), which increases the enjoyment of the lesson and the musical experience.
• Improved ability to read and perform from musical notation, as indicated by empirical research (see below). [iii]
• The ability to improvise may additionally help with:
  - acquisition of a new piece, as reading comprehension is assisted by being able to ‘speak the language’ of the written text.
  - sight reading, from being able to anticipate what may come next (Azzara, 1993).
  - memorization of a piece of music, from being able to identify and commit more easily to memory a recognised musical sequence, pattern, or larger scale organizing principles rather than each detail of a piece.
• Aural skills by definition are built on listening skills. Pauline Oliveros (1999) lists several positive learning outcomes from improvisation, and the first three all involve development of more effective listening skills.
• For students with limited formal repertoire, improvisation can be a means to extend practice time with new material. As Joan Last (1972) notes: ‘the teacher must do what they can to keep variety.’
• An important part of effective improvisation is to develop a sense of genre, mood and idiom. You can’t put a four beat bar into a waltz, or a three beat bar into a march, or play a toccata in a slow, strict tempo. [iv]
• Improvisation around technical exercises can give a meaningful harmonic context and variety to practice of scales and other drills.
• The ability to include student’s favourites in lessons: when students pick their own popular music favourites for improvisational activities (e.g. playing by ear), they may bring more enthusiasm and open-mindedness to the task. (Whitcomb, 2013)
• Just as playing without notation allows performance technique to develop at an independent rate than music reading (Knerr, 2008), then improvising additionally frees technical development from the bounds of what has been memorized.

Negative impacts of including improvisation in lessons may include the following:

• Additional technical development comes only from ad hoc creativity rather than systematic extension of existing technique.
• Reduced time practicing hand/eye coordination in score reading.
• Trying to fit too much into a lesson — to help to manage this, the improvisation should ideally support the lesson’s existing learning objectives incrementally, rather than be totally independent.

In all of the above, improvisation techniques introduced to students in lessons may also be continued in new contexts in their own home practice. If this is the case, care should be taken to always retain balance, and avoid Schumann’s complaint that he spent more time in improvising than practicing from the score.

Academic studies of the relationship of improvisation to musical awareness and skill

Some academic studies have been carried out to assess the impact of including improvisation in lessons on musical awareness and skill acquisition, including performance of prescribed repertoire, scales and arpeggios, sight reading, and aural skills. [v]

A study of 101 Australian high school instrumentalists at grades 3 to 6 tested for correlations between skills in sight reading, performing rehearsed music, playing from memory, playing by ear, and improvising. (McPherson, 1995) The results were positive, and identified a significant and substantial correlation between an ability to play by ear and to improvise, and also between sight-reading and performing rehearsed music, and sight-reading and improvising. [vi]

Another study by Azzara (1993) examines how the impact of including improvisation in grade 5 music lessons contributed to the improvement of instrumental performance achievement. The results suggested to the researchers that ‘improvisation skills contribute to more accurate student instrumental performances when reading from notation,’ that students developed ‘an increased understanding of harmonic progression through the mental practice and physical performance of tonal and rhythm patterns with purpose and meaning,’ and that ‘improvisation ability appears to transfer to a student’s clearer comprehension of the tonal, rhythmic, and expressive elements of music in an instrumental performance from notation.’ [vii]

Della Pietra (1997) conducted a study focused on rhythmic perception and reproduction. A significant improvement in perception of meter was noted in the group which had been exposed to improvisation, indicating a causality between exposure to rhythmic improvisation, and rhythmic perception. [viii]

Conclusion

Recent general theories of cognition and education do support the inclusion of student-led activity in effective learning, and several prominent systems of classroom music education (as well as training in other performance arts, and even individual instrumental music teaching in earlier times) specifically support inclusion of improvisation in lessons, as a constant component which should be well planned, and subjected to careful and collaborative reflection.

To ask how the inclusion of improvisation in lessons could impact on the development of the learner’s musical awareness and skill acquisition is to stand Jaques-Dalcroze’s triangle of musical achievement on its head. For Jaques-Dalcroze, technical skills and musical awareness are at the base of the triangle, and together form the necessary preconditions which give the opportunity for the student to reach self-expression through musical improvisation.

Here we are asking the opposite: what is the impact of improvisation on musical awareness and skills acquisition. From this perspective, including improvisation in lessons supports achievement of learning objectives towards a range of musical skills and awareness for successful achievement of competence in instrumental performance. Anecdotal evidence of positive impacts is backed up by empirical research which has shown correlations between improvisation skills and other aspects of achievement such as sight reading and performing rehearsed music, and causality has also been shown between training in improvisation and more accurate instrumental performances, clearer comprehension of tonal, rhythmic, and expressive elements of performance from notation, and rhythmic perception.

If some students will enjoy improvisation, and it will encourage them to persist in their lessons and practice more, routinely including it as an element of lessons (and home practice) will deliver wider rewards in their musical skills and awareness.

It will also allow them to approach what may be the crown of all their study and highest reward of their long labours: the power of improvisation itself.

Bibliography

Abramson, R.M., ‘Dalcroze-based improvisation.’ Music Educators Journal, vol.66 no.5, 1980

Aldrich, E., ‘Social dancing in Schubert’s world.’ in Erickson, R., (ed), Schubert’s Vienna,1997

Azzara, C.D., ‘Audiation-based improvisation techniques and elementary instrumental students’ music achievement.’ Journal of Research in Music Education, vol.41 no.4, 1993

Bent, I., ‘Heinrich Schenker, Chopin and Domenico Scarlatti.’ Music Analysis, vol.5 no2/3, 1986

Biasutti, M., ‘Teaching improvisation through processes. Applications in music education and implications for general education.’ Frontiers in Psychology, 8:911. Doi: 10.3389/fpsyg.2017.00911, 2017 (accessed online March 2021)

Bradshaw, M., ‘Improvisation and comprehensive musicianship.’ Music Educators Journal, vol.66 no.5, 1980

Brook, P., ‘Any Event Stems from Combustion: Actors, Audiences, and Theatrical Energy (Peter Brook in conversation with Jean Kalman).’ New Theatre Quarterly 30: Volume 8, Part 2, 1992

Clark, C., ‘The Cambridge Haydn Encyclopedia.’ 2019

Clark, F., ‘The Music Tree.’ Parts 1 to 4. Part of the Frances Clark Library for Piano Students.. Alfred Publishing, 1973

Cohen, M.L., ‘A constructivist approach to elementary school music learning experiences with reference to the ideas of John Dewey.’ MEd Thesis, 2000 (PDF accessed online March 2021)

Covington, K., Improvisation in the aural curriculum: An imperative.’ College Music Symposium, vol.37, 1997

Czerny, C., ‘A systematic introduction to improvisation on the pianoforte: opus 200.’ translated and edited by Mitchell, A.L., 1983 (first publication 1829)

Della Pietra, C., ‘The effects of a three-phase constructivist instructional model for improvisation on high school students perception and reproduction of musical rhythm.’ PhD dissertation, University of Washington, 1997

Gjerdingen, R., ‘Music in the Galant style.’ 2019

Goldstein, M., ‘Sounding the Full Circle (concerning music improvisation and other matters).’ 1988

Gooley, D., ‘Fantasies of improvisation: free playing in nineteenth century music.’ 2018

Gordon, E.E., ‘Roots of Music Learning Theory and Audiation.’ Chicago: GIA publications, 2011 (PDF accessed online 10 March 2021)

Gould, C.S. and Keaton, K., ‘The essential role of improvisation in musical performance.’ The Journal of Aesthetics and Art Criticism, vol.58 no.2, 2000

Hiller, F., ‘Felix Mendelssohn, Letters and Recollections.’ 1874

Holbrook, C., ‘The crown of all our study’” Improvisation in Quintilian’s Institutio Oratoria, Rhetoric Society Quarterly, Vol. 31, №3, 2001

Jaques-Dalcroze, E., ‘Rhythmics and pianoforte improvisation.’ Music & Letters, vol.13 no.4, 1932

Kallberg, J., ‘Chopin and the aesthetic of the sketch: a new prelude in E flat minor?’ Early Music, vol.29 no.3, 2001

Kinderman, W., ‘Genetic criticism and Beethoven’s creative process.’ Acta Musicologica, vol.81, no.1, 2009

Knerr, J., ‘The pros and cons of method books.’ The Official Journal of the Missouri Music Teachers Association, vol.XXXIX no.2, 2008

Kreader, B., et al, ‘Piano Lessons.’ Books 1 to 5. Part of the Hal Leonard Student Piano Library, Hal Leonard Corporation, 2009

Kreader, B., et al, ‘Scales, Patterns, and Improvs.’ Part of the Hal Leonard Student Piano Library, Hal Leonard Corporation, 2009

Last, J., ‘The Young Pianist.’ 1972

Lowe, M., in cooperation with Edwin E. Gordon, ‘Music Moves for Piano.’ Books 1 to 5. Music Moves LLC, 2004

MacDonald, C., ‘Schumann’s Piano Practice: Technical Mastery and Artistic Ideal.’ The Journal of Musicology, vol.19 no.4, 2002

McPherson, G.E., ‘Five aspects of musical performance and their correlates.’ Bulletin of the Council for Research in Music Education, no.127,1995/1996

Mortensen, J.J., ‘The pianist’s guide to historic improvisation.’ 2020

Nettl, B., ‘Thoughts on improvisation: a comparative approach.’ The Musical Quarterly, vol.60 no.1, 1974

Oliveros, P., ‘Quantum Improvisation: The Cybernetic Presence.’ Keynote address at the Improvisation across Borders conference, 1999

Olson, L.F., et al, ‘Piano Discoveries.’ Book A. Part of Music Pathways: A course for Piano Study. Carl Fischer Inc., 1983

Quintilian, ‘Institutio Oratoria.’ Loeb Classical Library edition, 1920

Richards, A., ‘The Free Fantasia and the Musical Picturesque.’ 2001

Ruiter-Feenstra, P., ‘Bach and the art of improvisation: volume 1.’ 2011

Sanguinetti, G., ‘Partimento and incomplete notations in eighteenth-century. keyboard music.’ in Guido, M., ‘Studies in Historical Improvisation From Cantare super Librum to Partimenti.’ 2017

Schulenberg, D., ‘Composition and improvisation in the school of J.S.Bach.’ in Stinson, R. (ed), ‘Bach Perspectives: Volume One.’ 1995

Schulenberg, D., Bach (Master Musicians Series). 2020

Shehan, P.K., ‘Major approaches to music education: an account of method.’ Music Educators Journal, vol.72 no.6, 1986

Sitton, M.R., ‘Beethoven’s Opus 77 Fantasy: an improvisational document?’ American Music Teacher, vol.36 no.6, 1987

Solis, G., Nettl, B., ‘Musical improvisation: art, education, and society.’ 2009

Steen, A., ‘Exploring Orff: A teacher’s guide.’ 1992

Stinson, S.W., ‘Piaget for Dance Educators: A theoretical study.’ Dance Research Journal, vol.17 no.1, 1985

Sung, H., ‘Handel’s Organo ad Libitum: a study of adagios in his organ concertos.’ PhD thesis, Moores School of Music, University of Houston, 2016

Treitler, L., ‘Medieval improvisation.’ The World of Music, vol.33 no.3, 1991

Van Delinder, J., ‘Dance as Experience: Pragmatism and Classical Ballet.’ Social Thought & Research, vol.23, no.1&2, 2000

Viera, M., The work of John Cassavetes: script, performance style, and improvisation.’ Journal of Film and Video, vol.42 no.3, 1990

Whitcomb, R., Teaching improvisation in elementary general music: facing fears and fostering creativity.’ Music Educators Journal, vol.99 no.3, 2013

Ziegenhardt, P., ’10 reasons improvisation should be an integral part of all vocational ballet training.’ www.stageminded.com/improvisation-ballet-training, 2019 (accessed 20 February 2021)

[i] This definition is intended to include the following as examples of improvisation:

• The greater or lesser extent to which any performance (novice or virtuoso, classical or jazz) is permitted to vary in an unplanned (albeit sometimes controlled) way by the performer, freely or from a range of options decided by the performer during the performance
• In first learning a work from the score and sight reading, the exercising of musical judgements not explicitly marked on the score
• Creative realisations of figured bass and other partially notated music
• Extemporisations of preludes, fantasies, cadenzas and other performances which are not fully prepared beforehand

The definition is also intended to exclude the following:

• Playing a learned piece of music, whether from notation or by memory, in the same way every time
• A continuo keyboard part played from figured bass in the same way each time
• ‘Preludising’ a composed piece with a composed prelude (e.g. Clara Schumann’s composed and written down prelude to Robert Schumann’s Des Abends)
• Performances of a composed cadenza, in the same way every time
• Extemporisations and embellishments rehearsed and repeated in the same way on every occasion
• Mistakes and performances of composed music which may approach but do not match the actual intent of the performer
• Mechanically randomised performance techniques
• Exercises in transposition at the keyboard (though they may develop some of the same skills which support improvisation)

[ii] Such uses of improvisation as a supplemental teaching technique are to be found in Steen’s Orff-inspired lesson plans, and the second phase of Della Pietra’s three phase lessons (see below section on academic studies), which emphasise the exploratory and student-led aspect of the activity of variation.

[iii] ‘Speaking and conversing serve as readiness for reading language and enhance the understanding of the written word. Similarly, improvisation enhances the performance of notated music.’ For early learners of music, teachers may ‘consider the role of improvisation as a readiness for learning how to comprehend music notation.’ (Azzara, 1993) ‘To improvise is to speak the musical language of motion and pitch, without a text, but clearly, expressively, and memorably. A living language is constantly being improvised in every conversation and every composition and is recreated in every live performance.’ (Abramson, 1980) How effective is a recitation in a language which the reciter is not able to speak?

[iv] As one researcher felt compelled to offer as feedback in a classroom improvisation: ‘No Susan, you don’t end a nocturne with a smash’ — though perhaps the point could have been made better in a student-led manner. This type of knowledge of idiom can be acquired without improvisation, but it is certainly very effectively practiced and tested through application in improvisation.

[v] Any wider effects of acquiring improvisation skills on the development of the learner’s awareness and ability (e.g. skills in free or extended improvisation either solo or in groups, or advancing of wider social and interactive skills) are not discussed here.

[vi] The study also showed that the correlations strengthen with increasing experience and musical maturity. It would be wrong to conclude from this that training in improvisation will improve abilities in other skills — correlation does not imply causation — but neither does it rule out the possibility. McPherson’s paper doesn’t make it clear whether, and how, any of the students had been instructed in improvisation before, or as part of the study — an important omission, because teaching different aspects of improvisation using various methods will have a different impact on the development of wider musical awareness and skills of the learner.

[vii] 66 instrumental music students in the US took part in the study, in which one group had an element of improvisation included in their lessons, and the other did not. Improvisation activities included (a) learning a selected repertoire of songs by ear, (b) developing a vocabulary of tonal syllables and rhythm syllables, © improvising with their voices and instruments tonic, dominant, and sub-dominant tonal patters, and (d) improvising rhythm patterns. At the conclusion of the study, the result was assessed by having each student perform three etudes: one prepared without teacher assistance, one with the help of the teacher, and one read by sight.

[viii] In Della Pietra’s research over fifteen lessons, the first phase of the lesson was teacher-led, the second phase provided a framework for students to improvise, and a third phase involved performance and feedback from the teacher, while a control group took no part in any of these exercises. Della Pietra does not state what the alternative programme of the control group was.