The creation of a musical work, or the final form of a musical work, as it is being performed. It may involve the work's immediate composition by its performers, or the elaboration or adjustment of an existing framework, or anything in between. To some extent every performance involves elements of improvisation, although its degree varies according to period and place, and to some extent every improvisation rests on a series of conventions or implicit rules. The term ‘extemporization’ is used more or less interchangeably with ‘improvisation’. By its very nature – in that improvisation is essentially evanescent – it is one of the subjects least amenable to historical research.
BRUNO NETTL (I) ROB C. WEGMAN (II, 1) IMOGENE HORSLEY (II, 2) MICHAEL COLLINS/STEWART A. CARTER (II, 3(i, ii)) MICHAEL COLLINS/GREER GARDEN (II, 3(iii)) MICHAEL COLLINS/ROBERT E. SELETSKY (II, 3(iv–ix)) ROBERT D. LEVIN (II, 4(i)) WILL CRUTCHFIELD (II, 4(ii), 5(ii)) JOHN RINK (II, 5(i)) PAUL GRIFFITHS (II, 6), BARRY KERNFELD (III)
In virtually all musical cultures there is music that is improvised. Societies differ, however, in several ways: the degree to which improvisation is distinguished from pre-composition; the nature and extent of the musical material which improvisers use as a point of departure or inspiration; the kinds and amounts of preparation required of improvisers, either in their musical training or in relation to individual performances; the relationship of written to oral transmission; and the relative social and musical value assigned to improvisations, compositions and the musicians who practise them. For further discussion of specific traditions see entries on individual countries.
2. Improvisation in musical cultures.
3. Models or points of departure.
Improvisation, §I: Concepts and practices
The term ‘improvisation’, in suggesting a failure to plan ahead or making do with whatever means are available, may have negative implications. However, in many of the world’s musical cultures the ability to improvise is often highly valued. In societies such as those of the Middle East and North India the improvised portions of a performance carry the most prestige.
The relationship between pre-composition and improvisation may be intricate. In Karnatak music the formal techniques of both composed songs and improvisations include repetition, variation, melodic sequences and returns to the point of departure. Similarly, in 19th-century European organ improvisation one purpose was to produce well-crafted fugues that might not be distinguishable from the composed canon. Elsewhere composition and improvisation represent opposite ends of the musical spectrum, as in 19th-century piano music where the concept of improvisation, in composed genres such as impromptu and fantasia, was drawn upon to explain or justify departure from formal norms.
One of the typical components of improvisation is that of risk: that is, the need to make musical decisions on the spur of the moment, or moving into unexplored musical territory with the knowledge that some form of melodic, harmonic or ensemble closure will be required. While risk is always present, its character varies greatly. In the improvisation of a fugue the difficulty is in adhering to the predetermined form; in the kalpana svara of Karnatak music it is the juxtaposition of rhythmic patterns that depart from but return to the tāla; in Iranian music it is the maintenance of a balance between quoting memorized material and moving too far beyond it; in South Slavonic epics it is of keeping to a textual line structure while alternating memorized themes with commentary. In most instances audiences evaluate improvisations by their balancing of obligatory features against imaginative departures from them. They also appreciate exceptional virtuosity, either technical or intellectual.
Even in societies in which improvisation is not recognized it may play a role in the conception of music. Some Amerindian music is said to be created in a moment of ecstasy, with a suddenness analogous to improvisation. This is the case in songs learnt in visions or dreams by Amerindian peoples of the North American Plains who say that in the dream the song was sung to them by an animal guardian spirit only once. The visionary, however, rehearses the song before singing it for other humans. In Plains culture this way of composing is contrasted with another, in which someone may sing through the songs they know and will consciously combine sections to create a new work.
The concept of improvisation has also been dealt with variously in the history of Western musical scholarship. In the study of Western music history improvisation has principally attracted scholars interested in historically informed performing practice and was associated with the early music movement of the second half of the 20th century. Earlier it interested music educators who used it to enhance music learning, and it has continued to play a role in music education in Europe and North America. Nevertheless, before the 1970s the field of musicology tended to treat improvisation as a ‘craft’, in contrast to the ‘art’ of composition. Case studies of improvisation began in ethnomusicology in the 1960s, concentrating on three repertories: jazz, Indian art music and Iranian music. To a substantial extent approaches to the study of improvisation in other cultures have been informed by the types of studies suggested by these three repertories. However, since the mid-1970s the distinctions between improvisation and other forms of music-making have been investigated in research that deals with concepts of risk, competence, dealing with unexpected situations and making positive use of mistakes.
Improvisation, §I: Concepts and practices
The improvising musician may play a special role in a culture’s conceptualization of musicianship. In Western culture the musics that are most dependent on improvisation, such as jazz, have traditionally been regarded as inferior to art music, in which pre-composition is considered paramount. The conception of musics that live in oral traditions as something composed with the use of improvisatory techniques separates them from the higher-standing works that use notation.
By contrast, in West Asian societies improvised music (and music that gives the impression of being improvised) has been the ideal. These cultures associate it with the concept of freedom, with the ability of a musician to make his or her own decisions and with the absence of restriction such as metre. The respect for individual decision-making is extended to evaluating musicians, thus, for example, privileging the learned amateur who needs only to follow his or her own inspiration over the professional who is obliged to perform when and as directed. Improvised genres are regarded as central and composed ones as peripheral. Non-metric genres, which are always improvised, have the highest prestige.
Karnatak musicians are judged both on their knowledge of the repertory of composed songs and by their ability to improvise particularly the unmetred ālāpanam. However, it is the performance of rāgam-tānam-pallavi, the South Indian genre that relies most on improvisation, that is considered the greatest test of a musician’s skill. According to Ki Mantle Hood (1964), in the group improvisations of Javanese gamelan, the performers on those instruments that hold the ensemble together (and thus depart from the model) as well as those that move most from the model, are the most highly esteemed.
Although some degree of improvisation may be said to be present in all musical performance, improvisation should be spoken of only when performances based on a model differ substantially or when a society distinguishes explicitly between the performance of a pre-composed piece and an improvisation on the basis of something given. Thus the many versions of an English folksong can be said to result from a personal interpretation of a local tradition passed on by a particular individual, rather like improvisation. Yet folksingers seemed traditionally to talk more about personal versions than improvisation. The concept of improvisation is readily accepted by the practitioners of West Asian music, although some musicians appear to memorize their improvisations and to perform these personal variants consistently. Japanese musicians ordinarily maintain that there is no improvisation in Japanese music. However, performances of the same piece of shakuhachi music may differ greatly in length and form, and to some extent in content.
Determining the presence or absence of improvisation in a particular culture depends to a great extent on the culture’s own taxonomy of music-making and on its assessing of the relationship between what is memorized or given and the performance. The prominence of improvisation varies greatly from culture to culture. For example, it characterizes the dominant genres in the musical cultures of South and West Asia, in Indonesia and in Africa, whereas in certain other societies it typifies individual, and perhaps exceptional, genres: jazz in the West, sanjo in Korea, the Philippine kulintang ensemble, and sections of the Cantonese opera from China.
Improvisation, §I: Concepts and practices
A common feature of improvised music is a point of departure used as the basis of performance. No improvised performance is totally without stylistic or compositional basis. The number and kinds of obligatory features (referred to here as the ‘model’) vary by culture and genre.
The most prominent model may be that of mode or a modal system. As a model these can be found in South, Central and West Asia, in North Africa and, in a somewhat different form, in Indonesia. In South Asia the predominant model is rāga. The definition of rāga is a subject of much discussion and dispute among South Asian musicians: it may, however, be described as a collection of pitches in a hierarchical relationship, from which are produced sets of typical, and often obligatory, melodic practices, motifs and ornaments. Rāga are the basis of both improvised and pre-composed genres in Hindustani and Karnatak music. In performance each item (either improvised or pre-composed) is ordinarily based on only one rāga. However, improvisations based on a series of rāga, with an emphasis on the elegant transition from one to the next, are occasionally heard.
In West Asian musics, improvisations are based on concepts similar to those of rāga, albeit with significant differences. Known as magām in Arabic traditions, makam in Turkey, mugam in Azerbaijan, makom in Uzbek culture and gushe in Iranian music, the West Asian modes are less complex and subject to fewer explicit requirements than rāga, and they are, in each culture, fewer in number. In contrast to performances of rāga, a West Asian performance makes use of several modes, moving from a principal one to secondary ones and back again.
In the gamelan music of Indonesia a fundamental melody is varied in a different way by each instrument in the ensemble. There is relatively little improvisation in Balinese gamelan, but it is more definitively present in Javanese gamelan music, in a strictly controlled form, as the model comprises the mentioned skeletal melody, of a pathet (‘mode’), and, for each improvising instrument, a specific pace and density.
The numerous musics of sub-Saharan African societies exhibit a great variety of improvisational practices. Prominent among them, in all parts of the continent, is the use of improvised variation. In this a vocal or instrumental soloist repeats a short phrase many times, varying it slightly each time but maintaining a consistent length and rhythmic framework. Similarly, a call-and-response form may consist of a refrain that alternates with a soloist’s variations of a theme. The model for improvisation may also include a set of styles through which the improviser should pass, devoting an unspecified amount of time to each. Percussion ensembles in West Africa may consist of a number of performers, each of whom presents a single repeated rhythmic pattern, while an improvising master drummer selects from these patterns, juxtaposing and interweaving them, using them as the models. The xylophone orchestras of the Chopi of Mozambique and South Africa include the improvisation of simultaneous variations of a theme, in a somewhat similar fashion to the gamelan.
Jazz musicians use a variety of sources for improvisation. ‘Standards’, a repertory of popular songs, often from Broadway musicals or tunes composed specifically by or for jazz musicians, provide not only melodic material but also the (chord) ‘changes’, underlying harmonic progressions which form the basis for improvisations. Ordinarily a small band will play the ‘head’ (tune) in unison before the musicians take it in turns to improvise on it. Recorded solos by other musicians, memorized and sometimes transcribed, may also serve as models for improvisation. A further technique is the inclusion of quotations from other pieces or solos in the improvisation.
European traditional music displays a variety of improvisational techniques. The South Slavonic tradition of epic singing consists of combinations of themes and motifs dealing with the historical deeds of military and royal figures. These are juxtaposed with a small number of melodic lines (and their numerous variations) and with melodic motifs performed solo on an accompanying instrument, the one-string, bowed gusle. The model for improvisation includes textual and musical content, but also stylistic elements. A line of text must consist of ten syllables, with a word boundary between the fourth and fifth. These materials, passed on orally, are then manipulated and varied to provide a performance (sometimes lasting several days) that is improvised but also predictable.
A further example comes from Genoa. There, groups of four to six sailors sing in ensembles in which each member assumes a stylistic and musically functional role, e.g. la donna (a falsetto obbligato) or la guitarra (a vocal imitation of strumming a guitar), the entire structure using simple chord progressions for guidance. It has been argued that most European traditional music repertories consist of tune families, each a body of variants of one parent tune that has been developed through oral transmission with some improvisatory behaviour; explicit improvisation is not common. When it does occur it is found predominantly in southern, eastern and Celtic regions of Europe, and more in instrumental than in vocal music.
By comparison with non-Western and vernacular music, Western art music, in which improvisation plays a small role, uses a number of contrasting models. For example, forms such as fugue and specific themes were and still are occasionally the basis for improvisation by keyboard players. Themes from a concerto movement were the basis for classical cadenza improvisation. The rules and options for the use of a vocabulary of ornaments formed the model in Baroque and early Classical music. In contrast, some of the music of the second half of the 20th century has used non-specific models. A general style or sound provided guidance in improvisatory ensembles. A set of directions for volume and frequency served as the model for John Cage’s Imaginary Landscape no.4 for 12 radios. Here the actual sounds are ‘improvised’ by the natural and cultural environment.
Related to the question of model is the issue of learning improvisation, in which, too, societies differ greatly. South Indian musicians learn a series of exercises intended to help them juxtapose rhythmic and melodic structures with the melodic grammar of rāga. Iranian musicians are told that memorization of the radif, a repertory of 250–300 short pieces, will automatically teach them the techniques of improvisation. Jazz musicians have a variety of learning techniques, including the notation and memorizing of outstanding solos.
Improvisation, §I: Concepts and practices
The genres that follow are characterized by being from art music repertories and are contrasted by the relative prominence of the model, the density of the obligatory features, and what is added creatively by the performer.
The genre of ālāp, jor, jhālā, gat, as played by North Indian instrumentalists like Ravi Shankar, Ali Akbar Khan and Bismillah Khan, is the first genre considered here. A full performance of a rāga consists of two sections. In the first, ālāp, jor, jhālā, the performer explores the rāga without pre-composed material as a model. The first subdivision, ālāp, presents the constituent parts of the rāga without a metrical structure, in what might be termed ‘free rhythm’. It brings in the characteristic motifs of the rāga, moves from the low tonic to the octave and beyond and eventually descends again. The second, jor, introduces a non-metrical pulse, and the third, jhālā, greatly increases the tempo. The second main section, gat, introduces the tāla (rhythmic cycle) and the accompanying tablā. The gat proper is a short composition which, once stated, becomes the basis for further improvisation, now within the constraints of the tāla. Improvisation here may consist of variations on the gat theme alternating with freely composed lines. The tablā may also improvise, and there is quickening of the tempo which leads to a climax ending the performance.
The most elaborate form in Karnatak music is rāgam-tānam-pallavi. This is a predominantly vocal genre but may also be instrumental. In the three sections the performer moves gradually from improvising on a general model (the rāga) to material that is substantially restricted. The first section, rāgam, is akin to the Hindustani ālāp, albeit differing stylistically. This is followed by the tānam (analogous to jor), sung to syllables such as tā nam, ta ka nam, ā nan dam. The model for the rest of the piece, which is now sung in tāla with drum accompaniment, is the pallavi. This is a composed line, often technically difficult, which is followed by niraval (variations on the composition) and svara kalpana (improvised passages which lead back to the given line). These may include techniques such as augmentation, diminution and imitation. The piece is concluded by a restatement of the pallavi. Hindustani and Karnatak genres, including dhrupad and khayāl in the North and South Indian nāgasvaram performances, have models that move from less to more specific forms and from non-metric to metric structures.
Āvāz is the central genre of Iranian art music and, like the South Asian genres, is cast in a complex standard form. Based on the memorized radif, an āvāz uses the material of one dastgāh (modal group). Non-metric (although occasional metrical structures occur), it is intended as vocal music but it is frequently performed on string instruments or flute. Since a dastgāh consists of a number of constituent gushe, each with a characteristic collection of notes and at least one characteristic motif, the performer, in advance or during the performance, selects which gushe will be used and in what order. It is the degree of creative improvisation as against quotation and variation that distinguishes performances and performers. Those gushe that have metre or memorable motifs are less subject to elaborate improvisation than those whose content is melodically and rhythmically less specific. While improvisers are free to select from the available gushe and determine their treatment in performance, there are discernible characteristic patterns.
In performances of the dastgāh Chahārgāh, which has some 15 available gushe, the great majority use only Darāmad, Zābol, Mokhālef and Hesar, most frequently in that order. A large number of gushe that appear in the radif are virtually never used in performance. Several schools of improvisation may be distinguished, including those musicians who believe that an avaz should follow the radif strictly and contain improvisations which are variants of the gushe in the prescribed order, in contrast to others who eschew close adherence to the radif and others who emphasize the mixing of gushe and the elegant transition from one to another.
The most prominent improvisatory genre in the various schools of Arabic and Turkish art music is the taqsīm. Ordinarily non-metric, it is exclusively instrumental and sometimes used as an introductory piece for vocal performances. It is most frequently performed on one of the West Asian lutes (‘ūd, bozuk, tār) or on a flute (nāy). A taqsīm may have a consistent pulse which is relieved occasionally but non-metric cadential formulae. Cast in one principal maqām, a taqsīm ordinarily moves to others built on any of the scale degrees of the primary mode, to which it eventually returns. While an improviser may theoretically move from any maqām to any other, there are, as in Iranian music, clear patterns. A taqsīm in the maqām of Nahawand is likely to move to Rast and Bayati before returning. Other maqāmāt such as ‘Ajam, Hijāz and Sabā appear only occasionally and others rarely or not at all. Geographic regions may be distinguished by improvisatory customs, and individual musicians develop personal styles that are easily recognized.
In the large Javanese gamelan composed melodies are played in slow notes by the saron (bronze metallophone) or sung, while instruments such as gongs mark subdivisions of the metric cycle. The drummer regulates tempo and dynamics. The other instruments such as members of the gender family, the family of bonang (sets of gong-chimes) and the gambang (wooden xylophone) embellish the balungan in various ways, the highest ones providing the greatest density. The rebab (two-string spike fiddle), the melodic leader of the gamelan, and the gender barung (bronze metallophone with resonators) play the main improvising roles. In this complex set of interactions the improvisation may be compared to the performance of a set of variations on a theme, rendered simultaneously.
Improvisation, §I: Concepts and practices
R. Haas: Aufführungspraxis der Musik (Potsdam, 1931/R)
E.T. Ferand: Die Improvisation in der Musik (Zürich, 1938)
Kh. Khatschi: Der Dastgah (Regensburg, 1962)
D. Rycroft: ‘The Guitar Improvisations of Mwenda Jean Bosco’, AfM, iii/1 (1962), 86–102
E. Gerson-Kiwi: The Persian Doctrine of Dastga-Composition (Tel-Aviv, 1963)
M. Hood: ‘Improvisation as a Discipline in Javanese Music’, Music Educators Journal, l/4 (1963–4), 34–8
A.B. Lord: The Singer of Tales (New York, 1965)
N. Caron and D. Safvate: Iran: les traditions musicales (Paris, 1966)
C. McPhee: Music in Bali (New Haven, CT, and London, 1966)
E. Wilkens: Künstler und Amateure im persischen Santurspiel (Regensburg, 1967)
R. Katz: ‘The Singing of Baqqashot by Aleppo Jews’, AcM, xl (1968), 65–85
M.T. Massoudieh: Awâz-e Šur (Regensburg, 1968)
H.H. Touma: Der Maqam Bayati im arabischen Taqsim (Berlin, 1968)
J. Becker: ‘The Anatomy of a Mode’, EthM, xiii (1969), 267–79
G. Tsuge: ‘Rhythmic Aspects of the Avaz in Persian Music’, EthM, xiv (1970), 205–27
M. Hood: The Ethnomusicologist (New York, 1971 )
N.A. Jairazbhoy: The Rags of North Indian Music (London and Middletown, CT, 1971)
J. Reichow: Die Entfaltung eines Melodiemodells im Genus Sikah (Regensburg, 1971)
H.H. Touma: ‘Maqam Phenomenon: an Improvisation Technique in the Music of the Middle East’, EthM, xv (1971), 38–48
B. Nettl and B. Foltin: Daramad of Chahargah: a Study in the Performance Practice of Persian Music (Detroit, 1972)
E. Zonis: Classical Persian Music: an Introduction (Cambridge, 1973)
B. Nettl: ‘Thoughts on Improvisation: a Comparative Approach’, MQ, lx (1974), 1–19
B. Nettl and R. Riddle: ‘Taqsim Nahawand: an Analysis of Sixteen Performances by Jihad Racy’, YICTM, v (1974), 11–50
F. Tirro: ‘Constructive Elements in Jazz Improvisation’, JAMS, xxvii (1974), 285–305
M. Hood: ‘Improvisation in the Stratified Ensembles of Southern Asia’, Selected Reports in Ethnomusicology, ii/2 (1975), 25–33
W.P. Malm: ‘Shoden, a Study in Tokyo Festival Music: When is Variation an Improvisation?’, YIFMC, vii (1975), 44–66
T. Viswanathan: ‘The Analysis of Raga Alapana in South Indian Music’, AsM, ix (1977), 13–71
P. Berliner: The Soul of Mbira (Berkeley, 1978)
S. Elkholy: The Tradition of Improvisation in Arab Music (Giza, 1978)
D. Sudnow: Ways of the Hand: the Organization of Improvised Conduct (Cambridge, MA, 1978)
M.J. Kartomi: ‘Childlikeness in Play Songs: a Case Study among the Pitjantjara at Yalata, South Australia’, MMA, xi (1980), 172–214
P. Alperson: ‘On Musical Improvisation’, Journal of Aesthetics and Art Criticism, lxiii (1984–5), 17–30
J. During: La musique iranienne: tradition et évolution (Paris, 1984)
J. Pressing: ‘Cognitive Processes in Improvisation’, Cognitive Processes in the Perception of Art, ed. W.R. Crozier and A.J. Chapman (Amsterdam, 1984), 345–63
V. Erlmann: ‘Model, Variation and Performance: Fulbe Praise Song in Northern Cameroon’, YTM, xvii (1985), 88–112
B.-S. Song: The Sanjo Tradition of Korean Komun’go Music (Seoul, 1986)
A. Kaeppler: ‘Spontaneous Choreography: Improvisation in Polynesian Dance’, YTM, xix (1987), 13–22
G.R. Kassebaum: ‘Improvisation in Alapana Performance: a Comparative View of Raga Shankarabharana’, YTM, xix (1987), 45–64
B. Lortat-Jacob, ed.: L’improvisation dans les musiques de tradition orale (Paris, 1987)
S. Slawek: Sitar Technique in Nibaddh Forms (Delhi, 1987)
H. Susilo: ‘Improvisation in Wayangh Wong Panggung: Creativity within Cultural Constraints’, YTM, xix (1987), 1–12
R.A. Sutton: ‘Variation and Composition in Java’, YTM, xix (1987), 65–96
R.D. Trimillos: ‘Time-Distance and Melodic Models in Improvisation among the Tausug of the Southern Philippines’, YTM, xix (1987), 23–36
E.F. Clark: ‘Generative Principles in Music Performance’, Generative Processes in Music: the Psychology of Performance, Improvisation, and Composition, ed. J.A. Sloboda (Oxford, 1988), 1–26
J. Kippen: The Tabla of Lucknow: a Cultural Analysis of a Musical Tradition (Cambridge,1988)
J. Pressing: ‘Improvisation: Methods and Models’, Generative Processes in Music: the Psychology of Performance, Improvisation, and Composition, ed. J.A. Sloboda (Oxford, 1988), 129–78
A. Jung: Quellen der traditionellen Kunstmusik der Usbeken und Tadschiken Mittelasiens: Untersuchungen zur Entstehung und Entwicklung des Sasmaqam (Hamburg, 1989)
H. Farhat: The Dastgâh Concept in Persian Music (Cambridge, 1990)
S.Y. Chan: Improvisation in a Ritual Context: the Music of Cantonese Opera (Hong Kong, 1991)
M.J. Kartomi: ‘Musical Improvisations by Children at Play’, World of Music, xxxiii/3 (1991), 53–65
I. Monson: Musical Interaction in Modern Jazz: an Ethnomusicological Perspective (diss., New York U., 1991)
A.J. Racy: ‘Creativity and Ambience: an Ecstatic Feedback Model from Arab Music’, World of Music, xxxiii/3 (1991), 7–28
G. Smith: ‘In Quest of a New Perspective on Improvised Jazz: a View from the Balkans’, World of Music, xxxiii/3 (1991), 29–52
L. Treitler: ‘Medieval Improvisation’, World of Music, xxxiii/3 (1991), 66–91
D. Bailey: Improvisation: its Nature and Practice in Music (New York, 1992)
E.T. Hall: ‘Improvisation as an Acquired, Multi-Level Process’, EthM, xxxvi (1992), 223–36
S.L. Marcus: ‘Modulation in Arab Music: Documenting Oral Concepts, Performance Rules and Strategies’, EthM, xxxvi (1992), 171–96
B. Nettl, ed.: The Radif of Persian Music: Studies of Structure and Cultural Context (Champaign, IL, 1992)
P.F. Berliner: Thinking in Jazz: the Infinite Art of Improvisation (Chicago, 1994)
B. Nettl: ‘Persian Classical Music and the Twentieth Century: the Changing Value of Improvisation’, To the Four Corners: a Festschrift in Honor of Rose Brandel, ed. E.C. Leichtman (Warren, MI, 1994), 57–66
C.D. Georgescu: Improvisation in der traditionellen rumänischen Tanzmusik (Eisenach, 1995)
H.H. Touma: The Music of the Arabs (Portland, OR, 1996)
B. Nettl, ed.: In the Course of Performance: Studies in the World of Musical Improvisation (Chicago, 1998)
Improvisation, §II: Western art music.
The concept of improvisation has been current in the West since the late 15th century to designate any type, or aspect, of musical performance that is not expressive of the concept of the fixed musical work. Its precise definition depends on the stability and perceived identity of the ‘fixed musical work’, which varies widely according to musical culture and historical period (see Composition). If the performance is of a work in the form of a notated composition, ‘improvisation’ tends to refer to departures from the text that would have been notationally available but were not actually written out, often for reasons of notational economy, and which rely on the existence of well-known, implied conventions of performance (such improvisations may therefore be recognized by the composer – perhaps within limits – as well as the performer as essential to the complete performance of the work). This definition normally excludes choices of tempo, but does include ornamentation and other kinds of melodic elaboration, as well as cadenzas in solo concertos. Looser definitions of the fixed musical work tend to allow much greater scope for improvisation. This is true of memorized harmonic schemes, as in 16th-century dance music or in jazz. Such schemes are usually identified by the performers as ‘pieces’ and may circulate under popular titles, even though the actual performances may obscure the identity almost beyond audible recognition. Even in musical cultures where there is no actual concept of the ‘musical work’, there may still be some perception of musical identity between one performance and the next, in terms of which ‘improvisation’ may also be identified. Typically, the definition of musical identity depends on the nature of the notational system or on methods of memorization; often it embraces versions that would be heard as different by Western listeners.
Musical traditions that do not rely on a strong conception of the fixed musical work tend not to have a concept of ‘improvisation’, but rather qualify performances in terms of a musical idiom or a set of performative conventions. For instance, polyphonic improvisation over plainchant melodies was designated as discant or contrapunctus throughout the late Middle Ages, until the conceptualization of the musical work (Res facta) in the late 15th century necessitated the identification of ‘singing on the book’ (cantare super librum) or ‘counterpoint made in the mind’ (and, in the 16th century, such terms as Sortisatio and singing ex tempore, subita, abrupta, improvisa, repente, alla mente). In such cases it may be inappropriate to speak of ‘improvisation’ as though it described the objective state of affairs in that musical tradition, and scholars often prefer to use terms less obviously premised on modern ‘work’ concepts (for instance ‘oral’ or ‘idiomatic’). As this example suggests, since the 15th century the Western classical music tradition has developed an acute sense of what constitutes improvisation, though musicologists have become increasingly circumspect about projecting that sense on non-Western or popular musical traditions, or remote historical periods. It is now considered essential for historical and ethnographic research to make explicit whether the concept of ‘improvisation’ is being applied as an ‘emic’ (from within the culture) or an ‘etic’ (from the ‘outsider's’ point of view) concept.
This distinction is important also for another reason. Once the written – improvised distinction had become identified in the West, it allowed two interesting permutations to emerge: compositions written in the style of improvisations, and improvisations shaped with the distinguishing properties of musical works. In such cases, to insist on an etic, technical definition of improvisation – involving the absence of, or departures during performance from the texts of, written or memorized works – would be to overlook compositions that, although written, do shed light on the emic understanding of improvisatory style. One such piece, for example, is Josquin's motet Stabat mater which, according to Joachim Thuringus (Opusculum bipartitum de primordiis musicis, 2/1625), was fashioned ‘in imitation of sortisatio’, and is indeed virtually unique having a cantus firmus without rests, an essential feature of polyphonic improvisations. A composition like this is conceptually quite different from the many earlier compositions that were indistinguishable from improvised discant simply because they implied no distinction to begin with (an obvious example being the discant settings in the Old Hall choirbook). The same may be true of such improvisatory keyboard genres as the 16th-century fantasia and ricercare, of the 19th-century prelude and rhapsody. Conversely, the purpose of improvised fugues, variations, and fantasias on given material, in the 18th century, was surely that listeners should evaluate the performer's skill on the terms of written compositions. In this sense, improvisation and composition can also be viewed, over and above the strictly technical distinction between them, as musical styles distinguished by the degree to which they give the appearance of performative spontaneity or authorial planning. (This distinction is essential, for instance, to any understanding of the history of piano music in the early 19th century.)
For information about particular aspects of improvisation see Aleatory; Cadenza; Continuo; Division; Jazz; Notation; Ornaments; Performing practice; Prélude non mesuré; and Singing.
Improvisation, §II: Western art music.
After the breakdown of Greco-Roman civilization, music in western Europe was preserved by rote memory, and new music was presumably worked out in performance or created spontaneously in improvisation. Since our knowledge of this music is limited to traditional liturgical chants written down in imprecise notation long after they were created, it is difficult to draw concrete conclusions about any improvisatory techniques used in their creation. The most reliable extant evidence relates to the spontaneous improvisation of the Jubilus, a melismatic flourish found on the last syllable of certain alleluias preserved in the early Christian liturgy. Clear references to this type of improvisation appear in the writings of the early Christian Fathers. St Augustine (354–430) described this jubilus as the musical outpouring of ‘a certain sense of joy without words … the expression of a mind poured forth in joy’. In its melismatic, virtuoso style it is not unlike the vocal cadenza added centuries later at the cadence of a Baroque aria.
A second, more controlled, improvisatory technique is hinted at in the structure of a number of surviving chant melodies. The chants in a particular mode, such as the Dorian, often use the same, or a similar, vocabulary of melodic motifs. This is taken as indicating an improvisatory practice in which the modes, like the rāga of India, included thematic materials as well as a roster of notes in their identities.
(iii) Improvisation on ‘perfect instruments’.
Improvisation, §II, 2: Western art music: History to 1600
Although melodic improvisation remained a factor in Western culture, it is indicative of its later development that the earliest substantial information about improvisation appears in treatises instructing the singer how to add another line to a liturgical chant as it was being performed. This is no doubt due to the fact that, although this early organum may have been derived from folk practice, the problem of improvising a melody to fit with a given chant required a technical knowledge of vertical consonance and dissonance and of the melodic materials available in the diatonic system. Furthermore, while at first the improvising singer may have relied on his memory of the chant to which he was adding a counter-melody, the improvisers eventually saw this chant in some sort of visual notation so that they could anticipate its notes. Thus the first manuals on improvisation are those concerned with the beginning of contrapuntal theory and practice and with the development of mensural staff notation.
Starting in the 9th century with the anonymous Musica enchiriadis and Scolica enchiriadis, which tell how to double a chant at the perfect intervals and to make oblique motion at the beginning and end of the chant, there was a steady growth and refining of the technique of organum. By the 11th century, when Guido of Arezzo described in his Micrologus the use of contrary motion and of rudimentary cadence formulae, there are also, in one of the Winchester tropers, a number of two-voice organa written down in staffless neumes that apparently use these same devices. From this time on, many written-down organa are found outside theoretical treatises, but the precise relation of these to improvised style is not defined. Two- and three-voice organa which introduce the use of several notes against one in the cantus firmus, such as those found in 12th-century manuscripts at St Martial, Limoges, and Santiago de Compostela, have the appearance of written-down improvisations, and it is unlikely that a composition was yet worked out by writing it down. Musical forms of the 13th century – discant, organum, motet etc. – were created by adding one line at a time to a previously worked-out melody, each of the added lines agreeing with this melody but not necessarily with each other. The interrelation of the parts was the same as when two singers improvised on a cantus firmus. By the 14th century, when a precise visual notation for music was established, complex structures, such as the isorhythmic motet, that could be worked out only by being written down, had also developed. Thenceforward the dependence on notation for composing music as well as for preserving it became one of the distinguishing features of Western musical culture, and in due course composed music, precisely notated, became the primary basis for performance.
Improvised music, however, remained an important element in art music for several centuries, and the interaction of the two types was fruitful. A well-known instance of the influence of improvised style on composed music is the introduction of the fauxbourdon style into the music of Burgundian composers, such as Du Fay and Binchois, in the 15th century. In Britain in the late 14th and early 15th centuries, the practice of improvising above a ‘sighted’ chant took the forms of parallel 3rds (gymel) and parallel 3rds and 6ths together (English discant). Although scholars still differ as to the exact historical process and the meanings of certain terms, the musical effect of these techniques on styles of composition is clear. In Britain this resulted in compositions that were structurally based on a series of 3rds or 6-3 chords that were freely ornamented and sometimes interspersed with other combinations of intervals. In Burgundian music, sections of parallel 6-3 chords (fauxbourdon) are inserted into the common styles and are found frequently at cadence points.
It was also in the 15th century that theorists of counterpoint first made a distinction between improvised and written-down styles. In 1412 Prosdocimus de Beldemandis simply mentioned the existence of two types of counterpoint – the ‘sung’ or ‘performed’ and the written – but in 1477 Tinctoris clearly spelt out the difference between the two in his Liber de arte contrapuncti. Because each improviser could make his line agree with the cantus firmus alone, dissonances and awkward part-writing could not be avoided. Moreover, an improvised piece was looser in construction, whereas the composer could order such features as cadences and rhythmic motion, thus producing a finished work (res facta) with a distinctive character.
The difference between the two styles became even greater in the 16th century, when composition on a cantus firmus became less a normative and more a specialized technique. Composed works appeared in a number of new styles and forms, while singers still improvised over a cantus firmus in long notes, often repeating a single figure as long as it fitted. At this period, improvisation was widely practised in Italian churches, where it was normally used over the chants of the introits in the Proper of the Mass, as well as over the hymns, antiphons and graduals. Since motets and the Ordinary of the Mass were set as artfully worked-out compositions, listeners will have apprehended two distinct usages.
It was not until 1553, when Vicente Lusitano’s Introdutione facilissima was published, that a methodical procedure for learning to improvise on a cantus firmus was made available. His ‘secrets’ seem very simple, as he was presenting a basic method by which the technique could be learnt; it may well have been based on his own way of improvising. He first gave a number of mechanical patterns in long notes that fit over the different intervals found in plainchant, such as those in ex.1a, where the syncopated line makes a series of 3rds, 5ths and 6ths over a succession of horizontal 3rds in the cantus firmus. Once the basic pattern was learnt, the singer could fill in the long notes with florid passages. Lusitano also included a number of single melodies in florid counterpoint above a cantus firmus (ex.1b). This exemplifies another of his suggestions: if a passage will fit more than once, the singer should continue to use it. He gave a number of florid examples above the same cantus firmus, and when two of these are combined, as they would be in performance, a number of dissonances and parallel 5ths appear between them.
Two years after Lusitano’s treatise appeared, Nicola Vicentino, in L’antica musica ridotta alla moderna prattica, condemned these devices as old-fashioned and recommended more modern ways, such as having the imitating voices imitate each other rather than the cantus firmus. Zarlino, in the third edition of his Istitutioni harmoniche (1573), gave, for the first time, instructions for more sophisticated devices, such as the improvisation of strict two-voice canons on a cantus firmus and two- and three-part canons without a cantus firmus. These techniques signalled a great change in improvised counterpoint. Both require more technical skill on the part of the performers. The leading singer must know all the possible combinations at specific times over precise pitches, and those following him must have good ears and memories. But this kind of planning also makes correct part-writing and dissonance treatment possible in improvisation. Zarlino also introduced rules for creating invertible counterpoint and for adding a third part in the performance of an already composed duet. During the last quarter of the 16th century his successors expanded the possibilities inherent in these new devices and also continued with the old. The art of improvised vocal counterpoint came to a final climax among the theorists and practitioners of counterpoint in the prima pratica style.
Although there are many early visual and literary references to instrumental ensemble music, no direct discussion of improvisation by instrumental groups has been found. We can only surmise that when contrapuntal – as opposed to heterophonic – improvisation took place in instrumental performance it was by players who had been trained in the vocal practice of improvising on a cantus firmus. The first sure evidence of such a practice is in the improvisation of the music for the bassadanza and saltarello danced in 15th-century Italy. Surviving collections of bassadanza tenors in long notes, along with pictures showing two high instruments presumably improvising on a tenor played by a sackbut, indicate that the music accompanying these dances may well have been produced by just such an improvising group. Later compositions on one of these bassadanza tenors, La Spagna, in which the tenor in long, even notes is accompanied by florid melodies in the upper parts, add weight to this conclusion.
It is significant too that the only book in this period giving examples of ensemble improvisation (for violone and harpsichord), Diego Ortiz’s Trattado de glosas (1553), still used the old La Spagna tenor when illustrating the technique of improvising on a cantus firmus. The beginning of one such improvisation (ex.2a), when contrasted with another (ex.2b) from the same book, shows the archaic nature of this improvisation. The one shown in ex.2b has a 16th-century Italian dance bass, which acts as a series of roots for triads, and the improvised melody is shaped by the notes of each chord and is organized motivically. The bass also gives the rhythm of the dance and is organized in phrases that are multiples of four bars. It is also short and is repeated several times, showing a series of improvised variations – a form and style that were to be used in improvisation for several centuries to come.
Improvisation, §II, 2: Western art music: History to 1600
A less difficult type of ensemble improvisation occurred in the 16th century when florid passages were added to a single line of a composed work while it was being performed. These ornaments were called diminutions, since they reduced the longer notes of a piece into a number of shorter notes; this practice was also referred to as the ‘breaking’ of a melodic line. Skill in diminutions belonged to the performer rather than to the improvising composer, and it required little theoretical knowledge, since the performer needed only to fit florid patterns into the longer notes of an already composed piece. More care had to be taken by the singer or instrumentalist who embellished the single line of a polyphonic work of which he saw only his own part than by the organist who had a complete work under his control, but the procedures taught by the various published manuals on diminution were designed to help him avoid these difficulties. There were lists of the numerous melodic patterns that fitted into each melodic interval and note length commonly found in music of the time, and these patterns could be transferred directly to the melodic line that the soloist wished to ornament.
Three general procedures were followed in creating these embellishments. The simplest was to have the substitute passage begin and end on the pitch of the notes being ornamented and then move immediately to the next note in the melody, a procedure shown in ex.3a. This was considered the safest, for it preserved the original contrapuntal movement of the work. The second way, shown in ex.3b, was to start on the original note but, instead of ending on it, to move on and arrive at the next note by conjunct motion. Although this way might produce contrapuntal errors such as unauthorized dissonances or parallel 5ths, it was assumed that they would not be noticed by the listener because the notes were so short. The third manner was simply to be freer, perhaps encompassing a longer segment of the original line in the embellishment or replacing one of the main melodic notes with a pattern not touching on it. While it was not approved, this technique can be seen in the ornamented works that are given in the diminution manuals; it often involved motivic or sequential patterns, as seen in ex.3c. It could be successful, however, only if the performer knew what was going on in the other parts.
This type of improvised ornamentation was usually applied to only one voice of a polyphonic work, but when a madrigal or motet was performed by soloists, each improvising diminutions on his part, care was taken to agree in advance the order in which they would add ornaments, to avoid contrapuntal confusion and dissonant clashes that might result from simultaneous ornamentation. Care was also taken by the performer of the bass part not to let his ornaments go above the tenor part and to limit his embellishments, so that the overall structure of the supporting bass line was retained throughout.
The first manual teaching the art of improvising diminutions for a solo singer or wind or string player in ensemble performances of polyphonic works was the Fontegara of Sylvestro di Ganassi, published in Venice in 1535. It is believed, however, that this practice must have appeared early in the development of polyphonic music, since the earliest known keyboard tablature, the Robertsbridge Codex (GB-Lbl Add.28550, c1360), contains elaborately ornamented versions of contemporary motets, and it is generally considered that the published manuals of the 16th century were a late attempt to codify and make available to all musicians the ‘secrets’ of this technique.
Improvised diminutions had a definite influence on composed music, for they introduced elements into the performance of Renaissance music that became an integral part of Baroque style. The ornamentation of a single line of a polyphonic work by a solo instrument while the entire work was played on a keyboard instrument, as seen in the second book of Ortiz’s Trattado de glosas, anticipated the solo instrumental writing of the early Baroque period. In the same manner, the ornamented cantus parts of frottolas such as the anonymous Aime sospiri printed in Petrucci’s sixth book of frottolas (1506), and the embellished top voices of selected four-part madrigals by Rore included in Girolamo dalla Casa’s Il vero modo di diminuir (1584), which were sung while the other parts were played on the lute, were forerunners of early 17th-century monody.
Improvisation, §II, 2: Western art music: History to 1600
For the Renaissance musician, the ‘perfect instrument’ was one such as the organ or lute, on which a single performer could play all the parts of a polyphonic composition. Starting with the Robertsbridge Codex and the Faenza Codex (I-FZc 117, c1400), elaborately ornamented versions of polyphonic motets and secular works appeared as a constant part of the repertory and can doubtless be seen as written-down examples of a common improvisatory procedure. Highly embellished intabulations of polyphonic vocal works continued to appear in keyboard manuscripts into the 16th century, when, with the development of music printing, a great many such arrangements for lute and vihuela, as well as for keyboard, were published. From the 14th century on, keyboard tablatures also included sacred chants and secular songs used as cantus firmi with florid countermelodies. The fact that a number of 15th-century manuscripts, such as the Fundamentum organisandi (1452) of Conrad Paumann, give practical instructions for adding keyboard-style countermelodies to fit with the intervals commonly found in such pre-existing melodies tends to confirm that contrapuntal improvisation on a chant in church or on a popular song in secular music-making was a common practice with the professional keyboard virtuoso. While these techniques were still important in 16th-century keyboard improvisation, the gradual abandonment of the central cantus firmus and the use of free forms based on fugal imitation in vocal music are reflected in the inclusion of canonic and fugal devices in Hans Buchner’s Fundamentum (c1520) and the Arte de tañer fantasía (1565) of Tomás de Santa María.
A new form, the set of variations on a popular tune or dance bass already familiar to the listener, also became a major element in improvisatory practice in the 16th century. These variations used both the older technique of cantus firmus and the new one mentioned above (see ex.2b), which has a set of chords as the ‘theme’. In some cases both the melody and the harmony of a popular song form the basis of a set of variations. This practice is paralleled in vocal music, where poet-composers, and less sophisticated figures too, improvised both words in fixed forms such as terza rima and ottavas, and vocal embellishments, over standard melodies, such as the romanesca and Ruggiero, and their attendant harmonies.
A special genre associated with the keyboard was the prelude or intonation, a free improvisation meant to establish the mode for a vocal or instrumental piece that followed it. The earliest written-out examples of this type are found in the keyboard tablature of Adam Ileborgh (1448). This genre was characterized from the beginning by idiomatic virtuosity, rhythmic freedom and loose thematic construction – features that listeners have always considered the true hallmarks of extemporaneous improvisation.
The line between improvising and composing was less clearly drawn in solo improvising because here the player normally had his own repertory, playing from memory, improvising and often changing his compositions and re-using materials from earlier improvisations. The fact that there were so many famous keyboard composers was no doubt due to this practice. The large number of collections of printed works for keyboard, lute and similar instruments in the 16th century, bringing music from the repertory and inventions of the professional performer into the hands of the amateur, led to a great change. The works of the professional virtuoso, whether devised through his own improvisations, worked out on his instrument in playing or first created in written notation, became in published form the repertory for the amateur, and improvisation became associated with the professional virtuoso.
Improvisation, §II: Western art music.
(i) Early 17th-century Italian practice.
(ii) 17th-century English practice.
(iv) Later italianate embellishments.
Improvisation, §II, 3: Western art music: Baroque period
All the modes of improvisation practised in Italy during the Renaissance continued into the early Baroque. The two principal types were the embellishment of an existing part and the creation of an entirely new part or parts. Though fundamentally different in theory, the two types are not always separable in practice.
Important modifications were introduced to the practice of improvised embellishment. Composers, perhaps from a greater concern for the text, began to exercise more control over ornamentation, in two ways: by writing out embellishments in some instances, and by introducing symbols or abbreviations for some ornamental patterns. The written-out embellishments in ‘Possente spirto’, the great aria in Act 3 of Monteverdi’s Orfeo – the words of which are, significantly, in terza rima – represent an extreme example, now in early Baroque style, of the type of embellishment earlier improvised by poet-musicians and referred to in §2(iii) above. Stylistic modifications of the old, melodically orientated embellishment occurred about 1600 through the establishment of a true basso continuo and as a consequence too of the new emphasis on emotional qualities in singing. The basso continuo, with its firm bass line and improvised chords, emphasized vertical rather than linear aspects, and embellishment gradually attained more harmonic implications, adding the spice of dissonance to the written notes. The new emotional style in vocal music caused two modifications in the melodic lines: the smoothly flowing notes of 16th-century passaggi were sometimes alternately dotted to form either trochaic or iambic figures that might emphasize sobbing or sighing qualities; and a new vocabulary of short embellishments was invented for use on notes sung to the accented syllables of emotive words in the text. The new style of embellishment is first seen in G.B. Bovicelli’s Regole, passaggi di musica (1594), but the aforementioned modifications came about largely through the influence of the Florentine Camerata and its insistence on correct and emotive declamation of the text. Passaggi were for the most part relegated to penultimate syllables of verses, where they did not obscure the meaning of the words; they can thus be seen as early forerunners of the later cadenza.
The preface to Caccini’s Le nuove musiche (1601/2), the most celebrated expression of the passionate attitude of the men who invented the new style, contains much discussion of vocal embellishments. Some of these are illustrated in ex.4: Caccini’s cadential trillo and gruppo are shown in ex.4a, and the early Baroque preference for dotted figures is seen in his illustrations of the desirable way of performing the phrases shown in ex.4b. That the embellishment of penultimate (and sometimes other) syllables often reached cadenza-like proportions is demonstrated in several songs in Le nuove musiche, notably Fortunato augellino, and in those of other Italian monodists of the period.
Le nuove musiche also contains the only surviving portions of Caccini’s opera Il rapimento di Cefalo (1600), one of which, the concluding chorus, is shown with embellished solo interpolations. In his accompanying remarks Caccini reiterates his position regarding the liberties that may be taken with the rules of counterpoint in making passaggi. The ornamented solos bear the names of three of the greatest singers of the age, Melchior Palantrotti, Jacopo Peri and Francesco Rasi. While Caccini apparently wrote all the embellishments printed here, according to the rubrics, Palantrotti sang them as written, Peri substituted ‘different passaggi, according to his own style’ and Rasi sang ‘some of the passaggi as given and some according to his own taste’ (trans. H.W. Hitchcock, RRMBE, ix, 1970).
The new kinds of embellishment became known in due course in Germany, where descriptions are found in such works as Syntagma musicum, iii (1618) by Michael Praetorius and Musica moderna prattica (1653) by J.A. Herbst. In addition to the intonatio, exlamatio, trillo and gruppo, discussed by Caccini, Praetorius illustrates the tremulo (in two forms, ascendens and descendens; ex.5a and ex.5b) and the tirata (ex.5c) and also provides numerous illustrations of the accentus, applied to various intervals. Four examples of the accentus, as applied to the ascending 2nd, are shown in ex.5d.
Agostino Agazzari (Del sonare sopra ’l basso, 1607) discussed the role of instruments in concerted music, classifying instruments in two groups, those of foundation and those of ornament. He counselled restraint, decorum and the judicious enhancement of the written notes and outlined the roles appropriate to the various instruments (translation from StrunkSR1):
He who plays the lute … must play it nobly, with much invention and variety, not as is done by those who, because they have a ready hand, do nothing but play runs and make divisions from beginning to end, especially when playing with other instruments which do the same, in all of which nothing is heard but babel and confusion, displeasing and disagreeable to the listener. Sometimes, therefore, he must use gentle strokes and repercussions, sometimes slow passages, sometimes rapid and repeated ones, sometimes something played on the bass strings, sometimes beautiful vyings and conceits, repeating and bringing out these figures at different pitches and in different places; he must, in short, so weave the voices together with long groups, trills, and accents, each in its turn, that he gives grace to the consort and enjoyment and delight to the listeners, judiciously preventing these embellishments from conflicting with one another and allowing time to each. … The violin requires beautiful passages, distinct and long, with playful figures and little echoes and imitations repeated in several places, passionate accents, mute strokes of the bow, groups, trills, etc. The violone, as lowest part, proceeds with gravity, supporting the harmony of the other parts with soft resonance, dwelling as much as possible on the heavier strings, frequently touching the lowest ones. The theorbo, with its full and gentle consonances, reinforces the melody greatly, restriking and lightly passing over the bass strings, its special excellence, with trills and mute accents played with the left hand. The arpa doppia, which is everywhere useful, as much so in the soprano as in the bass, explores its entire range with gentle plucked notes, echoes of the two hands, trills, etc.; in short, it aims at good counterpoint. The cithern, whether the common cither or the ceterone, is used with the other instruments in a playful way, making counterpoints upon the part. But all this must be done prudently; if the instruments are alone in the consort, they must lead it and do everything; if they play in company, each must regard the other, giving it room and not conflicting with it; if there are many, they must each await their turn and not, chirping all at once like sparrows, try to shout one another down.
Agazzari’s advice is applicable in a variety of musical contexts, but it has particular relevance for the theatre. In early Italian operas, instrumental ritornellos and sinfonias are frequently indicated, but sometimes written either in skeletal form (bass line alone, or bass with treble) or not at all. Luigi Rossi’s Il palazzo incantato (1642) begins with a sinfonia, for which only a bass line is given. Cavalli’s Didone (1641) at one point has the rubric ‘all the instruments enter’, although no parts are written, and at another point, ‘aria with all the instruments’, accompanied by continuo only. There are many instances in operas of the Contarini Collection (I-Vnm) in which a dance is indicated in the score, but no music is supplied; in other cases a bass line only is given (Rose, 1965). The passacaglia seems to have originated as an improvised ritornello in theatrical productions (Hudson, 1981).
The quotation from Agazzari demonstrates a close affinity between improvisation over a cantus firmus and realization of a basso continuo. The continuo part replaced the cantus firmus, and the improvised parts became more harmonically based, partly because of the harmonic implications of the bass line itself.
For organists, improvisation was a functional task as much as an artistic one. During Mass or services of the Office, organists improvised versets over liturgical cantus firmi, in alternation with the choir. This practice has clear relationships to basso continuo technique, and in fact some of Banchieri’s organ basses in L’organo suonarino (3/1622) contain occasional thoroughbass figures. Frescobaldi’s organ masses offer some indication of the function of improvised music within the mass (e.g. Tocata per le levatione). Improvised toccatas, preludes and intonazioni frequently served to establish the mode or pitch (or both) for the singers. Some of the earliest keyboard toccatas (by Andrea Gabrieli, Merulo and others) are based on psalm tone cantus firmi, and undoubtedly reflect improvisatory origins (Bradshaw, 1972). Maugars wrote of Frescobaldi that while ‘his printed works render sufficient evidence of his skill, to judge his profound knowledge adequately you must hear him as he improvises toccatas full of refinement and admirable inventions’ (trans. C. MacClintock, 1979).
A distinctively Italian type of improvisation was the viola bastarda technique, which appears in several Italian sources during the period 1580–1630. Taking a polyphonic composition as a model, the viola bastarda roams around the texture, embellishing now one voice, now another, and at times creating an entirely new voice (Paras, 1986).
The venerable practice of improvised vocal counterpoint over a cantus firmus (contrappunto alla mente) is mentioned by several 17th-century theorists. Its mastery was a requirement for members of the papal choir, who according to G.B. Doni (1647) were sometimes guilty of abusing it, insistently repeating a musical figure that did not always agree with the cantus firmus (Ferand, ‘Improvised Vocal Counterpoint’ 1956). It may have been a common feature of the training of singers, even in the stile recitativo (Hill, 1994). When two or more parts were improvised over a cantus firmus dissonances could result; this was tolerated and even prized by some commentators. Banchieri (Cartella musicale, 1614) states that the effect of contrappunto alla mente can be obtained even in a written composition if each voice is composed separately with reference to the bass only. He comments that there may even be hundreds of singers, and that although none of them knows what the others are doing, the result will be pleasing (Ferand, op. cit., 1956). A subspecies of contrappunto alla mente in which two or more singers improvise over a cantus firmus without contrapuntal errors is called contraponto in concerto. Vicente Lusitano describes it in Introdutione facilissima (1553), as does Pedro Cerone in El melopeo y maestro (1613)
Scipione Cerreto (Della prattica musica, 1601) describes canonic improvisation over a cantus firmus. His illustrations are designed to give the singer practice in improvising over common intervallic patterns (conjunct motion, rising 3rds, etc.) in the bass (ex.6). Improvising singers in the 17th century, like jazz musicians of more recent times, probably committed to memory a repertory of melodic figures for use in specific contexts.
Contrappunto alla mente and contraponto in concerto were applied principally in the context of sacred music. On the secular side were performers who sang vernacular lyrics, accompanying themselves, often on a plucked string instrument, to stock musical formulae. These singers ranged from purveyors of medicines and potions (ciurmadori), popular actors and street entertainers, to trained professional singers who intoned the works of leading poets. The formulae, each customarily consisting of a simple melody with bass line such as the Aria del gran duca and Ruggiero, were frequently subjected to strophic variation. Giovanni Stefani’s collection Affetti amorosi (1618) contains an aria per cantar sonetti and an aria per cantar ottave, designed to accommodate the singing of any verses in their respective genres. The simplicity of both settings suggests that they are not finished compositions, but frameworks suitable for elaboration. Similarly, in several Italian publications for Spanish guitar (e.g. Millioni, Abbatessa), the performer was expected to sing standard poetic forms (strombotti, ottave rime, etc.) to an improvised or familiar melody over the chords indicated by the alfabeto tablature (Cavallini, 1989).
Improvisation, §II, 3: Western art music: Baroque period
In just the manner described by Agazzari, there was a flourishing school of instrumental improvisation in England in the 17th century. Variation was a process inherited from the virginalists, and English instrumentalists were fond of improvising diminutions or variations on popular songs such as Greensleeves, itself based on the passamezzo antico. From this practice it was only a simple step to composing an original ground and repeating it ad libitum while another instrumentalist or two improvised divisions on it. A valuable source for this practice is Christopher Simpson’s Division Violist (1659);ex.7 shows excerpts from a set of divisions from it for solo bass viol. The vocal lines of English continuo songs were also subjected to extensive embellishments of various kinds, and several manuscript sources are largely devoted to such florid songs (see Till, 1975).
Thomas Mace (Musick’s Monument, 1676) is one of the most cogent witnesses to improvisatory instrumental practices in England. For him the Praelude is
a Piece of Confused-wild-shapeless-kind of Intricate Play … in which no perfect Form, Shape, or Uniformity can be perceived, but a Random-Business, Pottering and Grooping, up and down, from one Stop, or Key, to another; And generally, so performed to make Tryal, whether the Instrument be well in Tune, or not … After they have Compleated Their Tuning, They … fall into some kind of Voluntary, or Fanciful Play, more Intelligible; which … is a way, whereby He may more Fully, and Plainly shew His Excellency, and Ability, than by any other kind of undertaking; and has an unlimited, and unbounded Liberty.
Mace also writes of improvising an interlude in order to make a transition from a ‘suit[e] of lessons’ in one key to one in another key: ‘They do not Abruptly, and Suddenly Begin, such New Lessons, without some Neat, and Handsom Interluding-Voluntary-like-Playing; which may, by Degrees, (as it were) Steal into That New, and Intended Key’. In addition he offers an unusual account of the creation of the lesson entitled ‘The Authors Mistress’, published in Musick’s Monument. In his description of playing his lute at random when alone he provides a rare account of ‘spontaneous’ improvisation, with no functional purpose nor any reference to a pre-existing framework.
Thomas Morley devoted a substantial proportion of his Plaine and Easie Introduction to Practicall Musicke (1597) to improvised vocal counterpoint, stating that ‘when a man talketh of a Descanter it must be understood of one that can, extempore, sing a part upon a plainsong’. His examples of descant reveal sophisticated techniques: imitation, canon, double counterpoint and inversion. Morley stated that this art, though still very much in use in churches elsewhere, had declined in England by the end of the 16th century – yet Elway Bevin discussed it as late as 1631 in A Briefe and Short Instruction (see Ferand, op. cit., 1956).
Improvisation, §II, 3: Western art music: Baroque period
While the influence of Caccini's expressive, text-orientated style of ornamentation can be seen in Guédron's declamatory récits for solo voice, notated diminutions in both polyphonic and solo forms of the early 17th-century air de cour show that, in general, performers imitated the purely melodic formulae of earlier Italian practice. ‘Passages are especially pleasing when they are sustained, and last a long time’, wrote Mersenne in 1637. Slow-moving songs for solo voice on melancholy subjects received the most profuse ornamentation. A specifically French practice arose in the air de cour as sung by a soloist, whereby florid diminutions were concentrated in the second and later verses, to give the effect of a musical variation on the ‘simple’ melody presented in the first verse. A melody thus embellished was termed a diminution in the air de cour and a double (or second couplet) in the later air and air sérieux. Ex.8 shows an air de cour by Antoine de Boësset (taken from Mersenne) with Henry Le Bailly's diminution for verse 2. Moulinié's lighter ornamentation for verse 1 features the port de voix that anticipates the beat, the appoggiatura type most favoured in France before the 18th century. According to Mersenne, French airs were set apart from those of Italy, Germany and Spain because good singers like Le Bailly added trills extempore in diminutions and at cadences, ‘the more so as at those moments the voice is greatly softened and doubles its charming motion’. The signs for ornaments in ex.8, namely a cross for cadences (trills) with two to four repercussions, and the letter m for longer cadences, were not printed in the original text; Mersenne added them later by hand to his personal copy of his own treatise. Other graces besides trills and dynamic shadings regularly left to the performer's discretion included the plainte and the accent.
Diminution practice based purely on melodic and metrical considerations is still the focus of La belle méthode, ou L'art de bien chanter (1666) by the Besançon musician Jean Millet. However, from Mersenne onwards this approach had been censured for failing to take the text into account. Bénigne de Bacilly's Remarques curieuses sur l'art de bien chanter (1668) details the art of ornamentation typical of a new generation of singers, whose roulades were now tailored to the importance and length of syllables. Bacilly's exemplar was Michel Lambert, the greatest singer of Louis XIV's reign. While Bacilly's list of specific graces is much longer than Mersenne's, most were still left to the discretion of the singer to supply. Lambert committed many of his doubles to print, yet remarked in the preface to his Airs of 1660: ‘I would have dearly liked to be able to mark in my score all the ornaments [graces] and subtleties [petites recherches] that I try to bring to the performance of my airs, but these are things no-one has discovered how to write down’. Lully apparently disliked the practice of doubles for dramatic reasons, and later theorists devoted relatively little space to diminution techniques, but written-out examples in the songbooks show that doubles in the style of Lambert were sung well into the following century.
Improvised embellishment in instrumental music tended to reflect vocal practice. Jean Rousseau dedicated his Méthode claire, certaine et facile pour apprendre à chanter la musique (1678) to Michel Lambert, and in his Traité de la viole (1687) he set out to transfer Lambert's style to the viol. It contains the illustrations seen in ex.9, together with elaborate rules for the placing of the port de voix and tremblement in performance. Anticipating the beat by taking value from the preceding note was a technique not generally used in keyboard music, although there are rare examples in the récit of Nivers' Suite du 1er ton from his third Livre d'orgue (1675). All the ornaments in Couperin's L'art de toucher le clavecin (1716) begin on the beat.
Notated agréments are often lacking in 17th-century instrumental music as well as in vocal music, and must be added by the performer. The music of Louis Couperin, for example, contains few written ornament signs. In the later Baroque period, by which time written symbols were in general use, the degree of precision with which François Couperin and Marin Marais set them down is nevertheless unique. Montéclair's section on ornaments in his Principes de musique (1736), a noteworthy attempt to standardize ornament names and signs, makes it clear that graces were still often left to the performer's ‘taste and experience’ to supply. Most important in his array of 18 agréments (‘those who perform it badly will never sing pleasingly’) was the tremblement (trill), which he subdivided into four types. Expressive ornaments include the son enflé et diminué (crescendo and decrescendo), for which he claimed the invention of the modern ‘hairpin’ signs. Montéclair's examples, and the evidence of his own music, show that he did not object to virtuoso passage-work as such, but he was very much against free diminution (his term is passage) that obscured the melodic line. In his opinion, instrumentalists used it to excess out of their desire ‘to imitate the style of the Italians’.
Du Mont advertised his petits motets of 1652 with continuo as being the first of their kind to appear in France, but, as in Italy, a tradition of adding instruments impromptu to accompany (or replace) voices preceded the advent of the continuo. Untexted bass notes, found from time to time in polyphonic airs de cour from Guédron's second book of 1612 onwards, indicate the need for an accompanying instrument. A ‘basse continue pour le luth’ is mentioned by name in Moulinié's third book (1629), and a ‘basse continue pour les instruments’ in Boësset's seventh book (1630).
Other typically French aspects of improvisatory practice concern rhythm. Stylish performance of French Baroque music in general called for a knowledge of conventions of overdotting and notes inégales, which for the sake of flexibility and subtlety were not expressed in the notation. Players of the harpsichord, lute and viol were expected to improvise the entire rhythmic fabric of many preludes, which were printed only in skeletal form in long notes and with very few ornament signs. (See Dotted rhythms; Notes inégales; and Prélude non mesuré.)
Improvisation, §II, 3: Western art music: Baroque period
In the early 18th century, French composers like François Couperin, and some German masters like J.S. Bach, included their desired ornamentation in the text of the music, either with symbols for single-note graces (Fr. agréments; It. abellimenti), or with actual florid notation for passages. Their specificity, unusual in its day, was, however, to continue to the present, eventually eclipsing the more prevalent 18th-century trend, favoured especially by Italian composers and those working in their styles, of leaving embellishment to the performer. It was generally felt that, with less specific notation, the music served as something of a blueprint, and could be constantly refreshed and kept current by the idiomatic addition of improvised graces. While the greatest practitioners of this lavish, expressive improvisatory vocabulary were the solo violinists and singers, all soloists, vocal and instrumental, were expected to enter into this kind of creative collaboration with the composer. Although the practice is frequently seen as Italian, by the middle of the 18th century as impressive a contribution in this area was being made by French and Germanic musicians, especially violinists. Quantz, recognizing the basic difference between what he saw as French and Italian aesthetic (Versuch, 1752; chap.10, no.115), referred to the frequently notated single-note ornaments as ‘wesentliche Manieren’ (necessary ornaments) and the more elaborate personal improvised graces as ‘willkürliche Veränderungen’. Burney remarked later of Italian practice that ‘an adagio in a song or solo is generally little more than an outline left to the performer’s abilities to colour’. The best-known examples of improvisation in Italian adagios appear in Estienne Roger’s edition of Corelli’s Sonatas for violin and violone or harpsichord op.5 (Amsterdam, 1710), which the publisher prefaced by a declaration that the adagios were graced by Corelli himself: ‘Composez par Mr A. Corelli comme il les joue’; ex.10 presents the opening bars of the third sonata. Further, in Roger’s 1716 reprint the injunction has changed to ‘comme M. Corelli veut qu’on les joue’. These are significant samples to be emulated by the performer in his or her own manner, in addition to their value as documents of Corelli’s playing; Corelli, like any other 17th- or 18th-century musician, would have varied his graces at every performance.
In his Compendium musicae signatoriae et modulatoriae vocalis (1689; chap.5, §19) W.C. Printz described some Italian ornaments – figure corte, messanze and salti (ex.11) – that are somewhat less florid than those encountered in adagios. Even Montéclair, who was against passages (see §(iii) above), admitted this type, which he called diminutions. The best way to see how they can be improvised is to strip the figure corte from a written-out piece, for which the aria ‘Singet dem Herrn’ from Buxtehude’s cantata of the same title (buxwv98) has been selected (ex.12). Bach’s cantatas are replete with examples of written-out figure corte. These are interesting because genuine Italian sources of the late 17th century are scarce. They reflect the earlier division type, related to some of the patterns from which Corelli’s ornamental style is ultimately derived, especially in anonymous 18th-century decorations for Corelli’s fast movements; Corelli’s graces for the adagios are largely elaborations of the gruppi, tremolos and other figuration typical in 17th-century ornamentation. It is therefore puzzling that Roger North should have referred contemptuously to Roger’s newly published Corelli graces as ‘so much vermin’, when they appear to be no more alien than familiar Italian patterns grouped under a single bowstroke. Beginning with the Corelli sonatas, the notation of florid ornamentation, however rhythmically approximate, appeared regularly in published and manuscript sources, examples of a composer’s or performer’s style or milieu. For Corelli’s op.5 sonatas alone there are over 20 known sources of preserved improvisation, each adding new forms of passage-work idiomatic to the period and provenance of the example. The density of notated embellishment increases noticeably throughout the century, even encompassing the chromatic inflections of the galant and Classical styles. The most ornate late example of a decorated movement for violin and bass may be seen in the fold-out of J.-B. Cartier’s 1798 compendium of 18th-century violin literature, L’art du violon, where a Tartini ‘Adagio’ is varied in 17 increasingly complex ways (see illustration). Throughout it one can still perceive the original text, however, and identify much of the figuration with its 17th-century roots. Other examples in the solo violin literature of the period include William Babell’s XII Solos … with Proper Graces (London, c1725), Franz Benda’s 33 sonatas for violin and continuo (D-Bsb Mus.ms.1315/15), Geminiani’s Sonatas for violin and bass (London, 1716, 2/1739 ‘carefully corrected and with the addition, for the sake of greater ease, of the embellishments for the adagios’), Telemann’s Sonates corellisantes (Hamburg, 1735) and Sonate metodiche (Hamburg, 1728) and Vivaldi’s Sonata in Arv29 (plain in Malipiero edition; second movement ornamented in D-Dlb, Pisendel collection).
Improvisation, §II, 3: Western art music: Baroque period
Until now, the discussion has centred on examples of improvised ornamentation in slow movements where improvised graces are substituted for a simpler notated text. In fast movements, virtually all in binary form, the varied reprise of each section was considered essential. Quantz wrote about it as follows (op. cit., from the translation Easy and Fundamental Instructions, c1790):
It
is a principal Rule with regard to Variations that they must have a just
reference to the plain Air, the variation is made upon … the first Note of the
Variation must for the most part be the same with the original or plain Note …
or any other Note may be chosen instead of it from the Harmony of the Bass,
provided the Principal or plain Note be heard immediately after it …
Brisk and lively Variations must not be introduced in an Air that is soft,
tender and mournful, unless the Performer knows how to render them more
suitable and agreeable in the manner of executing them.
Variations are only to be introduced after the simple Air has been heard first,
otherwise it will be impossible for the Hearer to distinguish the latter from
the former; nor does an Air, compos’d in a pleasing and graceful Stile, require
any such additions, unless one was sure to improve still more upon it, they
being used for no other end, than to render an Air in the cantabile Stile more
melodious, and Divisions in general more brilliant.
Those that consist in a continual series of swift Notes or quick Passages,
though ever so much admired by some, in general are not so pleasing as those of
the more simple kind, the latter being more capable of touching the Heart, a
Point that certainly is most to be aim’d at, and indeed at the same time the
most difficult Part in Music; for which reason a young Beginner is advis’d to
be cautious and moderate in the use of Embellishments and Graces.
However, it is quite clear that in many performances both playings were freely embellished. Extant Corelli material other than the Roger edition of 1710 contains heavily decorated examples of fast movements, some with more than one ornamental option. And the fact that the practice of embellishing both first statement and repeat is discouraged, in the name of good taste, by writers of the period proves how widespread the practice must have been.
Practical consideration occasionally led a composer to write out varied reprises, as C.P.E. Bach did in his Sechs Sonaten für Clavier mit veränderten Reprisen (1760). He wrote in the foreword (translation, 1961, from Ferand, Die Improvisation in Beispielen, 1956):
Variation upon repetition is indispensable today. It is expected of every performer. The public demands that practically every idea be repeatedly altered, sometimes without investigating whether the structure of the piece or the skill of the performer permits such alteration. It is this embellishing alone, especially if it is coupled with a long and sometimes bizarrely ornamented cadenza, that often squeezes the bravos out of most listeners. How lamentably are these two adornments of performance misused. One no longer has the patience to play the written notes the first time; the too long absence of bravos is unbearable. Often these untimely variations, contrary to the setting, contrary to the Affect, and contrary to the relationship between the ideas, are a disagreeable matter for many composers. Granted, however, that a performer has all the qualities necessary to vary a piece in the proper way; is he always ready to do so? Are not new difficulties raised thereby in unfamiliar pieces? However, aside from these difficulties and from misuse, good variations always retain their value … In writing these sonatas I have had in mind mainly beginners and such amateurs as … no longer have enough time and patience to practise especially assiduously. I have wanted to give them … the satisfaction of being heard playing variations without having either to invent them themselves or to have others write them down and then themselves learn them by heart with much effort. I am happy to be the first, so far as I know, to work in this manner for the use and the pleasure of his patrons and friends.
Ex.13 shows the beginning of no.1, with the embellished reprise below (the right-hand part is shown; there are only minor differences in the left).
The varied reprise is mandatory in da capo arias. P.F. Tosi gave good advice to singers on this matter in his Opinioni de’ cantori antichi e moderni (1723), here cited in J.E. Galliard’s translation of 1742 (chap.7, §4):
In the first [part] they require nothing but the simplest Ornaments, of a good Taste and few, that the Composition may remain simple, plain, and pure; in the second they expect, that to this purity some artful Graces be added, by which the Judicious may hear, that the Ability of the Singer is greater; and, in repeating the Air, he that does not vary it for the better, is no great Master.
Because the principle of embellishment in arias did not change radically during the 18th century, except perhaps in the ‘reform’ operas of Gluck, the rules given by J.A. Hiller in his Anweisung zum musikalisch-zierlichen Gesange (1780; see §4(ii) below) are equally applicable to Baroque music. He stated that an aria must first be performed as the composer wrote it, though the singer could add one or two small graces. The varied da capo must appear easy and pleasant, but should actually be difficult in order to give the singer an opportunity to demonstrate his skill. In slow arias it is best to introduce legato ornaments, in allegro, staccato ones. Passaggi and similar ornaments should never be sung twice in the same way, and generally speaking the same graces should not be used too close to one another or too often in succession.
Improvisation, §II, 3: Western art music: Baroque period
In addition to the free-form ornamentation described above, it seems to have been customary for shorter movements to receive added sets of figural, melodic variations following the playing of the movements that served as their ‘theme’, even if that theme already had its reprises decorated. Georg Muffat referred implicitly to this practice in the preface to his concerti grossi of 1701, saying that the ‘liveliest airs [are to be played] thrice (with all [their] repeats)’. By using a system of distinct characterized variations, a short movement of this type could gain added cohesive structure as well as length, while demonstrating the improvisatory skill of the performer. Once again, the best period examples are found in the music of Corelli, where gavottas and gigas are extant both in freely decorated versions and extended with sets of variations. Frequently, individual movements with added variations exist without reference to the entire sonatas that originally contained them – the basis for the later Classical theme and variations.
Improvisation, §II, 3: Western art music: Baroque period
A special instance of free improvisation is the cadenza (for a fuller account seeCadenza). In his section on cadences Quantz defined them as
those Embellishments commonly introduced on the last Note but one, mostly on the Fifth of the Key … the Productions of the momentary Invention of the Performer. Regular Time is seldom to be observ’d in Cadences … Those for Voice or Wind Instruments ought to be short and so manag’d that they may be perform’d in one Breath, but those for String Instruments are not limited, but the Performer has so much Latitude given him, as his own skill and fruitfulness of Invention will permit, but notwithstanding will gain more Applause from the Judicious by a moderate length than otherwise.
The short cadenza ‘within a breath’ and the extended solo fantasy are exemplified and contrasted in the works and performing practices of two famous early 18th-century violinists, the Roman Corelli and the Venetian Vivaldi. At the close of Corelli’s sonata and concerto grosso movements it is not uncommon to find, immediately before the final cadential trill, a flourish that recalls the slurred gruppi of the ‘graces’ in the Sonatas. By contrast, reports of Vivaldi’s pyrotechnic abilities, not always immediately evident in the texts of his concertos, are amply demonstrated in the lengthy manuscript cadenzas for the Concerto in Drv208 (‘il Grosso Mogul’), which came to light in the 1980s. Further, Pietro Locatelli, the most technically accomplished virtuoso before Paganini, published as Capricii the extended and difficult cadenzas for the concertos of his 1733 L’arte del violino op.3. C.P.E. Bach, in his Versuch über die wahre Art das Clavier zu spielen (1753; end of chap.2), also discussed cadenzas in terms of fermatas appearing at cadences as well as other places (trans. W.J. Mitchell, 1949):
Fermate are often employed with good effect … there are three places at which the fermata appears: over the next to the last, the last, or the rest after the last bass note … Fermate over rests appear most frequently in allegro movements and are not embellished. The two other kinds are usually found in slow, affettuoso movements and must be embellished if only to avoid artlessness. In any event elaborate decoration is more necessary here than in other parts of movements.
The cadenzas improvised in da capo arias by singers such as the famous castrato Farinelli often ran to inordinate length, notwithstanding the stricture that they should be sung in one breath. Tosi (op. cit., 128–9) criticized such cadenzas thus:
Every Air has (at least) three Cadences, that are all three final. Generally speaking, the Study of the Singers of the present Times consists in terminating the Cadence of the first Part with an overflowing of Passagges and Divisions at Pleasure, and the Orchestre waits; in that of the second the Dose is encreased, and the Orchestre grows tired; but on the last Cadence, the Throat is set a going, like a Weather-cock in a Whirlwind, and the Orchestre yawns.
The gigantic written-out cadenza for the harpsichord in the first movement of Bach’s Fifth Brandenburg Concerto is an excellent example of its type, though it may be considered too extensive for the usual extemporization and even to constitute an example of the soloist’s abuse of privilege. When the two outer movements of a Baroque concerto – or indeed any two movements in any other type of work – are separated only by two chords, usually constituting a Phrygian cadence, the first of them should be elaborated into an improvised cadenza; a familiar instance occurs in Bach’s Third Brandenburg Concerto.
Improvisation, §II, 3: Western art music: Baroque period
Developments in the realization of the thoroughbass kept pace with the elaboration of the melodic line in the late Baroque period (for a fuller account see Continuo). The most exhaustive treatment of this subject is by Heinichen in Der General-Bass in der Composition (1728). He declared that the simple three-part realizations of the 17th century were old-fashioned and advocated the doubling of chords in both hands and the introduction of improvised ornamentation. He provided (i, chap.6) three realizations of a single bass line (ex.14), first in a simple manner (a), which can be avoided by introducing ornamentation and a true melodic line in the right hand (b); or, to give even more freedom to the melodic line, the chords may be played entirely by the left hand (c).
Early 18th-century Italian continuo practice went far beyond Heinichen, especially in the accompaniment of recitatives. Gasparini’s L’armonico pratico al cimbalo (1708), Joseph de Torres y Martinez Bravo’s Reglas generales de acompañar, en organo, clavicordio y harpa (enlarged 2/1736), which cites Gasparini with clearer musical examples, and Alessandro Scarlatti’s Varie introduttioni per sonare e mettersi in tono delle compositioni (GB-Lbl Add.14244) give rules for the addition to triads of handfuls of acciaccaturas, to be played ‘quasi arpeggiando’. But other examples, now and later in the century – for instance in Niccolo Pasquali’s Thorough-Bass Made Easy (1757) – are more subdued.
There are first-hand accounts of Bach’s continuo improvisation. L.C. Mizler wrote (1738) of his accompanying ‘every thorough-bass to a solo so that one thinks it is a piece of concerted music and as if the melody he plays in the right hand were written beforehand’ (translation from Aldrich, 1949). C.P.E. Bach related that his father liked to extemporize a fourth part when accompanying a trio. Bach’s pupil J.C. Kittel described (Der angehende praktische Organist, iii, 1808, p.33) how Bach would become impatient with the inadequate accompaniment of a pupil and how ‘one had to be prepared to find Bach’s hands and fingers mingling with the hands and fingers of the player and, without further troubling the latter, adorning the accompaniment with masses of harmony, which were even more impressive than the unexpected proximity of the strict teacher’. Two of Bach’s accompaniments are thought to be examples of his own realizations in written-out form, since they are marked ‘cembalo obligato’ and are not in the same style as his usual composed harpsichord parts: they are those of the second aria in the solo cantata Amore traditore bwv203 (thought to be by Bach) and of the second movement of the Sonata in B minor for flute and harpsichordbwv1030.
Improvisation, §II, 3: Western art music: Baroque period
Improvisation of complete pieces of music was not new to the Baroque period, but some of the great composer-performers of the era achieved new heights of virtuosity. In the 17th century the organ improvisations of Sweelinck, Frescobaldi and Buxtehude won the admiration of crowds who were attracted from far and wide. Bach is known to have improvised a prelude and fugue, an organ trio in three obbligato parts, a chorale prelude and a final fugue, all on a single hymn tune. Forkel remarked on Wilhelm Friedemann’s impressions of his father’s improvisations at the organ, saying that his organ compositions were indeed
full of the expression of devotion, solemnity and dignity; but his unpremeditated organ playing, in which nothing was lost in the process of writing down but everything came directly to life out of his imagination, is said to have been still more devout, solemn, dignified and sublime.
A famous story that hardly needs repeating tells how in 1747, while visiting Frederick the Great, Bach extemporized a fugue on a ‘royal theme’ given him by the king that he later worked out in his Musical Offering. Two of the rare places in music by (or attributed to) Bach where the performer is allowed some freedom to improvise are seen in the bars of minim chords in the Chromatic Fantasia bwv903 and the semibreve chords that constitute the entire prelude of the keyboard fugue in A minor bwv895.
Handel was equally famous for his improvising, as is attested by Hawkins’s description of his playing of his own organ concertos:
His amazing command of the instrument, the grandeur and dignity of his style, the copiousness of his imagination, and the fertility of his invention were qualities that absorbed every inferior attainment. When he gave a concerto, his method in general was to introduce it with a voluntary movement on the diapasons, which stole on the ear in a slow and solemn progression; the harmony close wrought, and as full as could possibly be expressed; the passages concatenated with stupendous art, the whole at the same time being perfectly intelligible, and carrying the appearance of great simplicity. This kind of prelude was succeeded by the concerto itself, which he executed with a degree of spirit and firmness that no one ever pretended to equal.
Indeed, in the scores of the opp.4 and 7 organ concertos the soloist is frequently directed to play an ‘Adagio ad lib’ or ‘Fuga ad lib’. While Handel’s own improvisatory skills must have been prodigious, the publication of such instructions indicates that he expected no less from his colleagues.
A new type of improvisatory piece called a partimento arose in the Baroque period with the inception of the thoroughbass. In thoroughbass practice both the bass line and the melody are given, while in the partimento only the bass line with figures is given, over which it was the performer’s responsibility to improvise self-contained pieces, character-pieces often called toccatas, and even fugues. Practised mostly in Italy, the partimento is closely related to the English practice of making divisions on a ground (see §(ii) above). Partimento improvisations were cultivated extensively as pedagogical exercises in the later Baroque period, when numerous collections were published by Gaetano Greco, Francesco Durante, Carlo Contumacci, Gaetano Franzaroli and Giuseppe Saratelli.
The art of improvisation flourished to the very end of the Baroque era in the free fantasia. J.S. Petri, in his Anleitung zur practischen Musik (1767), claimed that the fantasia was ‘the highest degree of composition … where meditation and execution are directly bound up with one another’. The whole final chapter of C.P.E. Bach’s Versuch (1753) is devoted to the improvisation of fantasias. A free fantasia is given at the end both in written-out form and as a partimento. The absence of bar-lines emphasizes the freedom of such a composition.
Beyond the feats of the 17th- and 18th-century virtuosos, improvisation was certainly a customary skill among competent, less renowned musicians. Extemporizing over ground basses, for instance, was an essential part of any musician’s training, and the extemporization of dances, suites and even more complex forms was not considered unusual. The incomplete scoring of 17th-century Italian operas seems to indicate that much of the instrumental music was improvised: at many junctures, one finds simply a bass line with two to four cleffed blank staves above it. It is clear that five-part texture (either two violins and two violas with bass, or one violin and three violas of differing sizes with bass) was usual in music throughout Europe, yet few viola parts or viola lines in scores can be found. There is little doubt that these were improvised, particularly given the more aural and less page-orientated training of most musicians of the period; unquestionably musical literacy was not a critical requirement for a practising musician of the day.
Improvisation, §II: Western art music.
Improvisation, §II, 4: Western art music: The Classical period
Performers and composers of the Classical period perpetuated three types of Baroque improvisation: embellishment, free fantasies and cadenzas. (The present article treats the two former; for discussion of the latter see Cadenza.) Although all three practices gradually declined, the universality of their use is confirmed in contemporary reports of concert performances (Burney), in treatises (e.g. by C.P.E. Bach, Manfredini and Türk – who in turn drew from earlier treatises e.g. by Marpurg and Quantz), and in composers' written-out embellishments of their own and others' musical texts. It seems likely also that soloists performing keyboard concertos improvised during orchestral ritornellos.
Reports and histories of the period inveigh against excessive amount and length of improvisation and embellishment. Such polemics confirm that these practices were widespread. Their aesthetic prescriptions are best evaluated in comparison with written-out embellishments, for example those by Haydn and Mozart, and the pseudo-improvised modulating preludes Mozart wrote for the use of his sister Nannerl; such documents show what was actually done (as opposed to what some observers would like to have heard). Of the authors whose treatises contain detailed descriptions of the use and methods of improvisation, C.P.E. Bach is the only one whose music elicits the same respect as his writings; the disparity in proficiency between composers and performers of Haydn's or Beethoven's level and those of, for example, Türk's cannot be overlooked. Nor can the fact that exhaustive treatises often tend to the categorical and pedantic. While writers and composers of the period may have agreed generally on the decisive role of taste, that of the master Classical composers is not necessarily identical with that of their writer contemporaries.
Treatises affirm that competence in prepared or improvised embellishments is dependent on thorough knowledge of the basic types (‘Manieren’) and their execution. The sources give sample embellishments but cannot clarify what proportion was prepared and what was improvised in 18th-century practice, although they do show that there were discrepancies between original text and actual performance. Composers wrote out embellishments for the benefit of amateurs or students, who, unlike the composer or virtuosos, were not expected to have mastered the art of improvisation. Mosel (1813), following in the footsteps of Gluck, Leopold Mozart, Quantz and C.P.E. Bach, specifically mentioned Mozart as a composer who wrote out precisely the ornaments he desired. Bach (1753–62, p.165) warned:
many variants of melodies introduced by executants in the belief that they honour a piece, actually occurred to the composer, who, however, selected and wrote down the original because he considered it the best of its kind.
He elaborated his reservations elsewhere:
Today
varied reprises are indispensable, being expected to every performer. A friend
of mine takes every last pain to play pieces as written, purely and in accord
with the rules of good performance. Can applause be rightfully denied him?
Another, often driven by necessity, hides under bold variations his inability
to express the notes as written. Nevertheless, the public holds him above the
former. Performers want to vary every detail without stopping to ask whether
such variation is permitted by their ability and the construction of the piece.
Often it is simply the varying, especially when it is allied with long and much
too singularly decorated cadenzas, that elicits the loudest acclaim from the
audience … Often these untimely variations are contrary to the construction,
the affect, and the inner relationship of the ideas – an unpleasant matter for
many composers. Assuming that the performer is capable of varying properly, is
he always in the proper mood? Do not many new problems arise with unfamiliar
works? Is not the most important consideration in varying, that the performer
do honour to the piece? … Yet, regardless of these difficulties and abuses,
good variation always retains its value.
The locus classicus of improvised instrumental ornamentation was the restatement of the principal theme, particularly in slow movements and rondos. Composers did not always notate such restatements, but signalled them in the manuscript with da capo signs. Thus the literal reprinting of the theme in modern editions creates an implication – not found in the sources – that the composer desired a note-for-note repetition of the opening music. Composers themselves often provided embellishments of principal themes immediately before performances by pupils or before publication. The differences between the texts Mozart used himself and those he presented to the general public may be seen by comparing autograph manuscripts and first editions of the second movement of the Piano Sonata in F k332/300k (ex.15) and the third movement of that in D k284/205b, variation xi. By the 1790s composers were writing elaborate embellishments into thematic reprises, having expropriated embellishment from the domain of improvisation (ex.16 shows an example by Beethoven).
Improvisation, §II, 4: Western art music: The Classical period
Little or no ornamentation is advised for plain recitative; Hiller (1780) stated that ornaments should be confined to occasional mordents and Pralltriller, though he accepted the need for more in scenas in accompanied recitative. J.F. Schubert advocated ‘appropriate free embellishment’ for fermatas but cautioned that this should be avoided where the word to which it would be sung made it inappropriate. In slow arias, or the slower sections of two-part rondò arias, the use of portamento was advocated by Corri, with messa di voce on longer notes, and the line might be highly graced, with appoggiaturas (single or compound) and acciaccaturas as the most common ornaments; little running passages might bridge leaps or fill out long notes, and syncopation, echo effects and division-like passages might be used. The surviving examples embellished by Mozart show the use of such devices (ex.21). Allegro arias or sections were less subject to decorative ornament, largely because of their greater speed and stricter tempo. Passing notes, appoggiaturas and the like are found, but in less profusion, and unmeasured flurries of quick notes are rare. Staccato and syncopation are sometimes used as ornamental devices, and running passages of semiquavers may be constructed on the outlines of melodies in longer note values. Hiller (1780) explained the distinction: ‘in slow and pathetic arias, slurred and drawn-out ornaments are the most appropriate, just as thrusting ones belong more to the Allegro’. In both styles, ornamentation generally involves adding notes of quicker rhythmic denomination rather than rerouting existing semiquavers into another region of the voice or simply recomposing the melodies (a practice often followed in modern revivals). Some degree of thematic variation, however, was expected in the case of an aria with a recurring theme as it reappeared.
Many complete arias survive with ornamentation attributable to specific composers or singers. The Czech composer Václav Pichl noted, in Milan in 1792, the variants sung by Luigi Marchesi in different performances of Zingarelli’s Pirro (ex.22); its opening line gives some indication of what theorists meant when they decried ornamentation that overwhelmed the original. At least eight examples were published of the decorations sung by Angelica Catalani, which were equally florid. Mozart’s preferred style of ornamentation, ascertainable from ex.21 and from his elaboration of an aria from his Lucio Silla (1772), shows a number of features: (a) the use of passing notes and other small ornaments in the first statement, increased in the repeat; (b) the standard use of appoggiaturas on feminine line endings (and often elsewhere; the speed of their resolution should be noted); (c) the variety of pace in the passage-work, with the prevailing semiquavers often enlivened by a burst of demisemiquavers or uneven groups of quick notes; (d) the tendency to use embellishment to increase the complexity and speed of figuration (as opposed to altering melodic shape or tessitura: the idea of adding ‘excitement’ through high notes seems to play no part); (e) the use of syncopation and phrasing to vary the line, and (f) the increase in elaboration as cadences are approached. In the Lucio Silla aria it is worth noting, additionally, that the wide leaps in long notes are left unornamented.
Duets, trios, quartets etc. must be sung as they are written, and though it is permissible to vary this or that in the solos, in the remainder it is necessary to proceed with unanimity, and to pay close attention to forte, piano and pianissimo; to smooth out, connect and separate … in concerted pieces, … appoggiaturas, trills and mordents are still permitted, but always with moderation.
Others emphasized that cadenzas for two voices or for voice with obbligato instrument (which was expected when the accompaniment featured one) must be prepared in advance and were often written out by composers. Many examples survive, among them several for the duet for Susanna and Countess Almaviva in Le nozze di Figaro and for ‘Ah perdona’ from La clemenza di Tito.
Improvisation, §II: Western art music.
Improvisation, §II, 5: Western art music: The 19th century
As for fantasies committed to paper, these reveal the evolution in
the models used by improvisers even if they lack the spontaneity of live
improvisation. Although dubbed a capriccio by Czerny, Beethoven's Fantasia
op.77 (published 1810) amounts to a ‘Fantasy-Prelude’ preparing a
theme-and-variations set with coda, this formal succession corresponding to a
background bass motion (ex.23) from B
/A
(‘Fantasy-Prelude’) through B
(theme and variations) and C (final variation) back
to B (coda). This
simple turning shape – recalling the structural foundation in Mozart's Fantasiak475 (see Rink, 1993) – may be typical of tonally conceived
early 19th-century improvisatory models, likewise the starkly juxtaposed keys
(f–f–f) in Schubert's
Fantasia for piano duet d940 (1828; ex.24); the extended chain of 3rds (f–A–c–E–G–[B]–G–b–D–f–A) spanning Chopin's Fantaisie op.49 (1841); and, in
Chopin's Polonaise-Fantaisie op.61 (1845–6), the linear bass ascent A–B–C–D–E uniting seemingly disparate thematic and virtuoso
episodes (see Rink, 1993). In contrast, Liszt's Réminiscences de Don Juan
(1841) derives structural logic through a reordering of Mozart's operatic plot:
after a Grave opening intoning the Commendatore's music from Act 2 (D
major) follows a lengthy theme-and-variations treatment of ‘Là ci darem la
mano’ (A major), then a brilliant finale in B based on ‘Fin ch'han dal vino’ (typical of the
‘rousing, dazzling conclusions’ recommended by Czerny in longer fantasies).
Liszt's dramatic conceit – whereby dissolution triumphs over virtue – controls
the Fantasy, not an underlying tonal framework inherited from earlier
improvisatory traditions. In contrast, his B minor Sonata (quasi una
fantasia?) directly exploits tonally defined improvisatory models, a simple
key progression b–D–F–b/B
stabilizing the alternation between thematic and bravura passages.
Improvisation, §II, 5: Western art music: The 19th century
In orchestrally accompanied recitative, cantabile, arioso elements and ornamental vocalization continued to be important; composers increasingly wrote the ornaments into the score. Ex.25 shows part of Tancredi’s opening recitative, as scored and then as realized by Rossini for a singer. Examination of this, along with recitative realizations by Rubini, Garcia and others, shows that the greater floridity and greater variety of note values observed in Verdi’s recitatives reflect an increase not so much in elaborateness of recitative style as in the specificity of notating it – indeed, the elaboration of execution probably decreased somewhat, in contrast to the notational practice.
The ‘aria cantabile’ continued to be the principal locus for melodic embellishment. Garcia, who is more explicit even than Domenico Corri, gives details of phrasing, dynamics and expression as well; ex.26 shows a section from one of his examples. A decrease in density of figuration across the first half of the century is easy to discern; extending the examples backwards to, say, Catalani, at the beginning of the century, and forward to the artists heard on early recordings, would show this to be a steady, continuous process.
In Rossini, Bellini and Donizetti, the coda is based on the same chord sequence observed at the end of the cantabile, but in tempo and usually with accelerating harmonic rhythm (ex.27); this portion was elaborately and freely varied. Garcia, Lanza and others give tables showing the multiple possibilities, and examples by Rossini and many singers exist. Cabalettas continued to be embellished in early Verdi (Giulia Grisi’s variations for I due Foscari survive in Rome). By the time he stopped writing them – his last orthodox solo cabaletta is in the 1863 version of La forza del destino – the repeat was commonly omitted in performance. It seems likely that cabalettas were ornamented only as long as they were repeated.
Up to, and probably throughout, Rossini’s time, the end of the coda seems to have been taken without rallentando and with a burst of florid virtuosity, often concluded in the low register. But as the excitement of the held top note gradually assumed a greater role in vocal expression, it became customary to add one just before the final note of the cabaletta while the orchestra pauses or sustains. This is sometimes found in Verdi’s scores, and occasionally earlier, although it became a universal practice only later. Increasingly, towards the end of the century, much or all of the coda was omitted; the singer proceeded from the final fermata of the strophe directly to the last few bars of coda or to the postlude. The conclusions in ex.27 clearly envisage an ending in tempo. Ex.28 shows two more cabaletta conclusions: one by Naldi, clearly ending in tempo, and one by Viardot from about 1880, showing a hold before the final note, now taken in the high register. Many sets of cabaletta variants break off at this point, suggesting a conclusion more or less as in the score or with conventionalized elaboration, and many sources that are specific about accompaniment practice (above all Garcia) show the conclusions of their cabalettas with no hint that the tempo was to be retarded.
Improvisation, §II: Western art music.
Nevertheless, his influence may have contributed to the re-emergence of improvisation in composed music during the 1960s – along with other factors including the growth of live electronic music (and hence of unpredictable situations), the development of jazz to a point where it could embrace almost everything in the contemporary classical tradition (so that the definition of a performance as ‘free jazz’ or, from the classical standpoint, ‘free improvisation’ became arbitrary), the arrival of performers who wanted to go their own ways, composers’ growing knowledge of non-European music, their extension of Aleatory procedures, and a general movement in Western culture towards democratization and universal self-expression. Yet that movement did not extend so far, or for so long. Improvisation was rapidly reaffirmed as secondary to composition, in that improvising artists (except those also known as composers, such as Vinko Globokar, La Monte Young or Terry Riley) gained no broad platform, and composers (Stockhausen, for example) soon retracted the freedom they had permitted. Once the 1960s had passed, the old division between creative and performing musicians was restored.
Improvisation, §II: Western art music.
Improvisation, §II: Western art music: Bibliography
MersenneHU
V. Lusitano: Introducione facilissima et novissima de canto fermo (Rome, 1553, 3/1561/R)
G.B. Bovicelli: Regole, passaggi di musica (Venice, 1594/R); ed. in DM, 1st ser., Druckschriften-Faksimiles, xii (1957)
T. Morley: A Plaine and Easie Introduction to Practicall Musicke (London, 1597/R); ed. R.A. Harman (London, 1952, 2/1963/R)
S. Cerreto: Della prattica musicale vocale, et strumentale (Naples, 1601/R, 2/1611)
G. Caccini: Le nuove musiche (Florence, 1601/2/R); ed. in RRMBE, ix (1970)
A. Banchieri: L’organo suonarino (Venice, 1605/R1969 with selections from later edns, 4/1638)
A. Agazzari: Del sonare sopra ’l basso (Siena, 1607/R; Eng. trans. in StrunkSR1); ed. V. Gibelli (Milan, 1979)
P. Cerone: El melopeo y maestro (Naples, 1613/R)
A. Banchieri: Cartella musicale nel canto figurato fermo et contrapunto (Venice, 1614/R1968 in BMB, section 2, xxvi)
C. Simpson: The Division Violist (London, 1659, 2/1667/R as Chelys: minuritionum artificio exornata/The Division-Viol, 3/1712)
J. Millet: La belle méthode, ou L’art de bien chanter (Lyons, 1666/R1973 with introduction by A. Cohen)
B. de Bacilly: Remarques curieuses sur l’art de bien chanter (Paris, 1668, 3/1679/R; Eng. trans., 1968, as A Commentary upon the Art of Proper Singing)
J. Rousseau: Méthode claire, certaine et facile, pour apprendre à chanter la musique (Paris, 1678, 5/c1710/R)
J. Rousseau: Traité de la viole (Paris, 1687/R); Eng. trans. in The Consort, no.34 (1978), 302–11; no.36 (1980), 365–70; no.37 (1981), 402–11; no.38 (1982), 463–7
J. Mattheson: Exemplarische Organisten-Probe im Artikel vom General-Bass (Hamburg, 1719, 2/1731/R as Grosse General-Bass-Schule)
J. Mattheson: Kleine General-Bass-Schule (Hamburg, 1735/R)
M.P. de Montéclair: Principes de musique (Paris, 1736/R); Eng. trans. of section on ornamentation, RRMBE, xxix–xxx (1978)
F. Geminiani: The Art of Playing on the Violin (London, 1751/R1952 with introduction by D.D. Boyden)
J.J. Quantz: Versuch einer Anweisung die Flöte traversiere zu spielen (Berlin, 1752/R, 3/1789/R; Eng. trans., 1966, 2/1985, as On Playing the Flute)
C.P.E. Bach: Versuch über die wahre Art das Clavier zu spielen, i (Berlin, 1753/R, 3/1787/R); ii (Berlin, 1762/R, 2/1797/R); Eng. trans. (1949)
F.W. Marpurg: Anleitung zum Clavierspielen (Berlin, 1755, 2/1765/R)
L. Mozart: Versuch einer gründlichen Violinschule (Augsburg, 1756/R, 3/1787/R; Eng. trans., 1948, as A Treatise on the Fundamental Principles of Violin Playing, 2/1951/R)
J.F. Agricola: Anleitung zur Singekunst (Berlin, 1757/R) [trans., with addns, of P.F. Tosi: Opinioni de’ cantori antichi e moderni (Bologna, 1723/R)]; Eng. trans., ed. J.C. Baird (Cambridge, 1995)
J. Adlung: Anleitung zu der musikalischen Gelahrtheit (Erfurt, 1758/R, 2/1783)
J. Lacassagne: Traité général des élémens du chant (Paris, 1766/R)
J.S. Petri: Anleitung zur practischen Musik (Lauban, 1767, enlarged 2/1782/R)
G. Tartini: Regole per arrivare a saper ben suonar il violino (MS compiled by G.F. Nicolai, I-Vc); facs. of 1771 Fr. version, with commentary by E.R. Jacobi (Celle, 1961) [incl. Eng. and Ger. trans.]
J.A. Hiller: Anweisung zum musikalisch-richtigen Gesange (Leipzig, 1774, 2/1798)
J.A. Hiller: Exempel-Buch der Anweisen zum Singen (Leipzig, 1774)
V. Manfredini: Regole armoniche (Venice, 1775/R, 2/1797)
J.A. Hiller: Anweisung zum musikalisch-zierlichen Gesange (Leipzig, 1780/R)
H.C. Koch: Versuch einer Anleitung zur Composition, i (Rudolstadt, 1782); ii–iii (Leipzig, 1787–93/R)
J.C. Bach and F.P.Ricci: Méthode ou recueil de connaisances élémentaires pour le forte-piano ou clavecin (Paris, c1786/R) [attribution to Bach doubtful]
D.G. Türk: Clavierschule (Leipzig and Halle, 1789, enlarged 2/1802/R; Eng. trans., 1982)
F. Galeazzi: Elementi teorico-pratici di musica con un saggio sopra l’arte di suonare il violino analizzata, ed a dimostrabili principi ridotta, i (Rome, 1791, enlarged 2/1817); Eng. trans., ed. A. Franscarelli (DMA diss., U. of Rochester, NY, 1968); ii (Rome, 1796); Eng. trans. of pt 4, section 2, ed. G.W. Harwood (MA thesis, Brigham Young U., 1980)
J.G. Albrechtsberger: Kurzgefasste Methode den Generalbass zu erlernen (Vienna, c1791, enlarged 2/1792; Eng. trans., 1815)
J.A. Hiller: Anweisung zum Violinspielen (Leipzig, 1792)
N.J. Hüllmandel: Principles of Music, Chiefly Calculated for the Piano Forte or Harpsichord with Progressive Lessons (London, 1796)
J.P. Milchmeyer: Die wahre Art das Pianoforte zu spielen (Dresden, 1797)
C. Gervasoni: La scuola della musica (Piacenza, 1800)
C.F. Ebers: Vollständige Singschule (Mainz, c1800)
M. Clementi: Introduction to the Art of Playing on the Piano Forte (London, 1801, 11/1826)
J.P. Milchmeyer: Kleine Pianoforte-Schule für Kinder, Anfänger und Liebhaber (Dresden, 1801)
A.E.M. Grétry: Méthode simple pour apprendre à préluder en peu de temps avec toutes les ressources de l’harmonie (Paris, 1802/R)
J.F. Schubert: Neue Singe-Schule, oder Gründliche und vollständige Anweisung zur Singkunst (Leipzig, 1804)
D. Corri: The Singer’s Preceptor (London, 1810)
A.M. Pellegrini Celoni: Grammatica, o siano Regole di ben cantare (Rome, 1810, 2/1817)
P.A. Corri: Original System of Preluding (London, c1813)
J.N. Hummel: Ausführliche theoretisch-practische Anweisung zum Piano-Forte-Spiel (Vienna, 1828, 2/1838; Eng. trans., 1829)
L. Lablache: Méthode de chant (Paris, ?1829; Eng. trans., 1840)
C. Czerny: Systematische Anleitung zum Fantasieren auf dem Pianoforte, op.200 (Vienna, 1829/R); Eng. trans., ed. A.L. Mitchell (New York, 1983)
A. Calegari: Modi generali del canto (Milan, 1836)
F.-J. Fétis and I.Moscheles: Méthode des méthodes de piano (Paris, 1840/R)
M. Garcia: Traité complet de l’art du chant (Paris, 1840, 2/1851)
G.L. Duprez: L’art du chant (Paris, 1845, 2/c1850)
F. Kalkbrenner: Traité d’harmonie du pianiste: principes rationnels de la modulation pour apprendre à préluder et à improviser, op.185 (Paris, 1849; Eng. trans., 1849)
F. Sieber: Vollständiges Lehrbuch der Gesangkunst (Magdeburg, 1858, 2/1878; Eng. trans., 1872)
F. Lamperti: Guida teorico-pratica-elementare per lo studio del canto (Milan, 1865; Eng. trans., 1868)
J.-B. Faure: La voix et le chant (Paris, 1870)
E. Delle Sedie: Arte e fisiologia di canto (Milan, 1876)
E. Delle Sedie: Estetica del canto e dell’arte melodrammatico (Livorno, 1885)
Improvisation, §II: Western art music: Bibliography
BurneyFI
BurneyGN
BurneyH
HawkinsH
ReeseMMA
SchillingE
A. Maugars: Response faite à un curieux sur le sentiment de la musique d’Italie (Paris, c1640; Eng. trans. in MacClintock, 1979, and also1993); ed. J. Heuillon (Paris, 1991, 2/1992)
T. Mace: Musick’s Monument (London, 1676/R)
J.-J. Rousseau: Dictionnaire de musique (Paris, 1768/R; Eng. trans., 1771, 2/1779/R)
S. Arteaga: Le rivoluzioni del teatro musicale italiano dalla sua origine fino al presente (Bologna, 1783–8/R, 2/1785)
C. Ditters von Dittersdorf: Lebensbeschreibung (Leipzig, 1801; Eng. trans., 1896/R); ed. N. Miller (Munich, 1967)
H.C. Koch: Musikalisches Lexikon (Frankfurt, 1802/R, rev. 3/1865 by A. Dommer)
I.F. von Mosel: Versuch einer Ästhetik des musikalischen Tonsatzes (Vienna, 1813, 2/1910)
E.D. Wagner: Musikalische Ornamentik (Berlin, 1869/R)
E. Dannreuther: Musical Ornamentation (London, 1893–5)
E. Bücken: Die Musik des Rokokos und der Klassik (Potsdam, 1927)
R. Haas: Aufführungspraxis der Musik (Potsdam, 1931/R)
E.T. Ferand: Die Improvisation in der Musik (Zürich, 1938)
H. Stubington: Practical Extemporization (London, 1940)
P.C. Aldrich: The Principal Agréments of the Seventeenth and Eighteenth Centuries (diss., Harvard U., 1942)
R. Fasano: Storia degli abbellimenti musicali dal canto gregoriano a Verdi (Rome, 1949 [recte 1947])
H.-P. Schmitz: Die Kunst der Verzierung im 18. Jahrhundert (Kassel, 1955, 4/1983)
R. Steglich: ‘Das Auszierungswesen in der Musik W.A. Mozarts’, MJb 1955, 181–237
E.T. Ferand: Die Improvisation in Beispielen aus neun Jahrhunderten abendländischer Musik, Mw, xi (1956, 2/1961; Eng. trans., 1961)
E.T. Ferand: ‘What is res facta?’, JAMS, x (1957), 141–50
W. Georgii: Die Verzierung in der Musik: Theorie und Praxis (Zürich and Freiburg, 1957)
W. Tappolet: ‘Einige prinzipielle Bemerkungen zur Frage der Improvisation’, IMSCR VII: Cologne 1958, 287–8
F.-H. Neumann: Die Ästhetik des Rezitativs (Strasbourg, 1962)
R. Donington: The Interpretation of Early Music (London, 1963, 4/1989)
A. Geoffroy-Dechaume: Les ‘secrets’ de la musique ancienne (Paris, 1964/R)
E.T. Ferand: ‘Didactic Embellishment Literature in the Late Renaissance: a Survey of Sources’, Aspects of Medieval and Renaissance Music: a Birthday Offering to Gustave Reese, ed. J. LaRue and others (New York, 1966/R), 154–72
W. Dürr: foreword to Franz Schubert: Lieder, Neue Ausgabe sämtlicher Werke, iv/1–2 (Kassel, 1968)
A. Geoffroy-Dechaume: ‘Du problème actuel de l'appogiature ancienne’, L’interprétation de la musique française aux XVIIème et XVIIIème siècles: Paris 1969, 87–105
E. Melkus: ‘Die Entwicklung der freien Auszierung im 18. Jahrhundert’, Der junge Haydn: Graz 1970, 147–67
M. Cyr: ‘A Seventeenth-Century Source of Ornamentation for Voice and Viol: British Museum MS. Egerton 1971’, RMARC, no.9 (1971), 53–72
C. Rosen: The Classical Style: Haydn, Mozart, Beethoven (New York, 1971, enlarged 3/1997) [incl. CD]
H.M. Brown: Embellishing Sixteenth-Century Music (London, 1976)
E. Derr: ‘Zur Zierpraxis im späten 18. Jahrhundert’, ÖMz, xxxii (1977), 8–16
J.C. Veilhan: Les règles de l’interprétation musicale à l’époque baroque (Paris, 1977; Eng. trans., 1979)
M. Morrow: ‘Musical Performance and Authenticity’, EMc, vi (1978), 233–46
F. Neumann: Ornamentation in Baroque and Post-Baroque Music (Princeton, NJ, 1978, 3/1983)
J.E. Smiles: ‘Directions for Improvised Ornamentation in Italian Method Books of the Late Eighteenth Century’, JAMS, xxxi (1978), 495–509
C. MacClintock, ed.: Readings in the History of Music in Performance (Bloomington, IN, 1979)
R. Donington: Baroque Music: Style and Performance (London, 1982)
F. Neumann: Essays in Performance Practice (Ann Arbor, 1982)
P. Alperson: ‘On Musical Improvisation’, Journal of Aesthetics and Art Criticism, xliii (1984–5), 17–30
F. Neumann: Ornamentation and Improvisation in Mozart (Princeton, NJ, 1986)
J. Spitzer and N. Zaslaw: ‘Improvised Ornamentation in Eighteenth-Century Orchestras’, JAMS, xxxix (1986), 524–77
P. Lescat: Méthodes et traités musicaux en France 1600–1800 (Paris, 1991)
M. Cyr: Performing Baroque Music (Portland, OR, 1992)
B. Dickey: ‘Ornamentation in Early-Seventeenth-Century Italian Music’, A Performer’s Guide to Seventeenth-Century Music, ed. S. Carter (New York, 1997), 245–68
S. Feisst: Der Begriff Improvisation in der neuen Musik (Sinzig, 1997)
Improvisation, §II: Western art music: Bibliography
ApelG
F. Liszt: ‘Lettres d'un bachelier ès musique’, RGMP, iv (1837), 97–110
R. Schumann: Gesammelte Schriften über Musik und Musiker (Leipzig, 1854/R, rev. 5/1914/R by M. Kreisig; Eng. trans., 1977–80; selective Eng. trans., 1946/R)
C. Moscheles, ed.: Aus Moscheles’ Leben (Leipzig, 1872–3; Eng. trans., 1873)
‘Johann Nepomuk Hummel on Extemporaneous Performance’, MMR, xi (1881), 214–15
C. Reinecke: Zur Wiederbelebung der Mozart’schen Clavier-Concerte (Leipzig, 1891)
H. Schenker: Ein Beitrag zur Ornamentik als Einführung zu Ph. Em. Bachs Klavierwerken (Vienna, 1903, 2/1908/R; Eng. trans. in Music Forum, iv, 1976, 11–140)
A. Schering: ‘Zur instrumentalen Verzierungskunst im 18. Jahrhundert’, SIMG, vii (1905–6), 365–85
J. Chantavoine, ed.: F. Liszt: Pages romantiques (Paris, 1912)
J.P. Dunn: Ornamentation in the Works of Frederick Chopin (London, 1921/R)
A.M. Richardson: Extempore Playing (New York, 1922)
M. Pincherle: ‘De l'ornementation des sonates de Corelli’ Feuillets d'histoire du violon (Paris, 1927), 133–43; (2/1935)
F.T. Arnold: The Art of Accompaniment from a Thorough-Bass (London, 1931/R)
M. Rinaldi: Il problema degli abbellimenti nell'op. V di Corelli (Siena, 1947)
P. Aldrich: ‘Bach's Technique of Transcription and Improvised Ornamentation’, MQ, xxxv (1949), 26–35
P. Aldrich: Ornamentation in J.S. Bach’s Organ Works (New York, 1950/R)
R. Wangermée: ‘L'improvisation pianistique au début du XIXe siècle’, Miscellanea musicologica Floris van der Mueren (Ghent, 1950), 227–53
A. Kreutz: ‘Ornamentation in J.S. Bach's Keyboard Works’, HMYB, vii (1952), 358–79
W. Emery: Bach’s Ornaments (London, 1953/R)
E. Badura-Skoda: ‘Über die Anbringung von Auszierungen in den Klavierwerken Mozarts’, MJb 1957, 186–98
E. and P. Badura-Skoda: Mozart-Interpretation (Vienna, 1957; Eng. trans., 1962/R, as Interpreting Mozart on the Keyboard)
A. Gottron: ‘Wie spielte Mozart die Adagios seiner Klavierkonzerte?’, Mf, xiii (1960), 334 only
I. Horsley: ‘The Sixteenth-Century Variation and Baroque Counterpoint’, MD, xiv (1960), 159–65
P. Aldrich: ‘On the Interpretation of Bach's Trills’, MQ, xlix (1963), 289–310
P. Badura-Skoda, ed.: Carl Czerny: Über den richtigen Vortrag der sämtlichen Beethoven’schen Klavierwerke (Vienna, 1963; Eng. trans., 1970)
F. Neumann: ‘A New Look at Bach's Ornamentation’, ML, xlvi (1965), 4–15, 126–33
G. Rose: ‘Agazzari and the Improvising Orchestra’, JAMS, xviii (1965), 382–93
P. Badura-Skoda: Kadenzen, Eingänge und Auszierungen zu Klavierkonzerten von Wolfgang Amadeus Mozart (Kassel, 1967)
S. Jeans: ‘English Ornamentation of the 16th to 18th Centuries (Keyboard Music)’, Music antiqua: Brno II 1967, 128–36
K. Polk: Flemish Wind Bands in the Late Middle Ages: a Study of Improvisatory Instrumental Practices (diss., U. of California, Berkeley, 1968)
P. Schleuning: Die Fantasie, Mw, xlii–xliii (1971; Eng. trans., 1971)
D. Boyden: ‘Corelli's Solo Violin Sonatas “Grac'd” by Dubourg’, Festskrift Jens Peter Larsen, ed. N. Schiørring, H. Glahn and C.E. Hatting (Copenhagen, 1972), 113–25
M.C. Bradshaw: The Origin of the Toccata, MSD, xxviii (1972)
M. Collins: ‘In Defense of the French Trill’, JAMS, xxvi (1973), 405–39
C.R. Suttoni: Piano and Opera: a Study of the Piano Fantasies Written on Opera Themes in the Romantic Era (diss., New York U., 1973)
Zu Fragen des Instrumentariums, der Besetzung und Improvisation in der ersten Hälfte des 18. Jahrhunderts: Blankenburg, Harz, 1975
H.J. Marx: ‘Some Unknown Embellishments of Corelli's Violin Sonatas’, MQ, lxi (1975), 65–76
H.G. Mishkin: ‘Incomplete Notation in Mozart's Piano Concertos’, MQ, lxi (1975), 345–59
D.A. Lee: ‘Some Embellished Versions of Sonatas by Franz Benda’, MQ, lxii (1976), 58–71
B.B. Mather and D.R.G.Lasocki: Free Ornamentaion in Woodwind Music, 1700–1775 (New York, 1976)
C. Wolff: ‘Zur Chronologie der Klavierkonzert-Kadenzen Mozarts’, Mozart und seine Umwelt: Salzburg 1976 [MJb 1978–9], 235–46
C. Pond: ‘Ornamental Style and the Virtuoso: Solo Bass Viol Music in France c1680–1740’, EMc, vi (1978), 512–18
Zu Fragen der Improvisation in der Instrumentalmusik der ersten Hälfte des 18. Jahrhunderts: Blankenburg, Harz, 1979
C. Dahlhaus: ‘Was heisst Improvisation?’, Improvisation und neue Musik: Darmstadt 1979, 9–23
T. Mulhern: ‘Improvisational Avant-Garde Guitar: its History, its Proponents, its Future’, Guitar Player, xiii (1979), 36–8, 118 only, 120–22
I. Poniatowska: ‘Improwizacja fortepianowa w okresie romantyzmu’ [Piano improvisation in the Romantic age], Szkice o kulturze muzycznej XIX. wieku, ed. Z. Chechlińska, iv (Warsaw, 1980), 7–28
D. Carew: An Examination of the Composer/Performer Relationship in the Piano Style of J.N. Hummel (diss., U. of Leicester, 1981)
R. Hudson: Passacaglio and Ciaccona: from Guitar Music to Italian Keyboard Variations in the 17th Century (Ann Arbor, 1981)
L.F. Ferguson: ‘Col Basso’ and ‘Generalbass’ in Mozart’s Keyboard Concertos: Notation, Performance, Theory, and Practice (diss., Princeton U., 1983)
J. Paras: The Music for Viola Bastarda (Bloomington, IN, 1986)
J. Pressing: ‘Improvisation: Methods and Models’, Generative Processes in Music: the Psychology of Performance, Improvisation, and Composition, ed. J.A. Sloboda (Oxford, 1988), 129–78
T. Szász: ‘Figured Bass in Beethoven's “Emperor” Concerto: Basso Continuo or Orchestral Cues?’, Early Keyboard Journal, vi–vii (1988–9), 5–71
I. Cavallini: ‘Sugli improvvisatori del Cinque–Seicento: persistenze, nuovi repertori e qualche riconoscimento’, Recercare, i (1989), 23–40
J.S. Rink: The Evolution of Chopin’s ‘Structural Style’ and its Relation to Improvisation (diss., U. of Cambridge, 1989)
R.E. Seletsky: Improvised Variation Sets for Short Dance Movements, circa 1680–1800, Exemplified in Period Sources for Corelli’s Violin Sonatas, opus 5 (DMA diss., Cornell U., 1989) [see also idem, EMc, xxiv (1996), 119–30]
J.S. Rink: ‘Schenker and Improvisation’, JMT, xxxvii (1993), 1–54
T. Szász: ‘Beethoven's Basso Continuo: Notation and Performance’, Performing Beethoven, ed. R. Stowell (Cambridge, 1994), 1–22
V.W. Goertzen: ‘By Way of Introduction: Preluding by 18th- and Early 19th-Century Pianists’, JM, xiv (1996), 299–337
N. Zaslaw: ‘Ornaments for Corelli's Violin Sonatas, opus 5’, EMc, xxiv (1996), 95–115
H.D. Johnstone: ‘Yet More Ornaments for Corelli's Violin Sonatas, op.5’, EMc, xxiv (1996), 623–33
Improvisation, §II: Western art music: Bibliography
GroveO (‘Ornamentation’; A.V. Jones, W. Crutchfield) [incl. further bibliography]
P.F. Tosi: Opinioni de’ cantori antichi e moderni (Bologna, 1723/R; Eng. trans., 1742, 2/1743/R, as Observations on the Florid Song)
G. Mancini: Pensieri e riflessioni pratiche sopra il canto figurato (Vienna, 1774, 2/1777; Eng. trans., 1967)
T. Lemaire and H.Lavoix: Le chant: ses principes et son histoire (Paris, 1881)
A.B. Bach: On Musical Education and Voice Culture (Edinburgh, 1883, 5/1898)
M. Kuhn: Die Verzierungs-Kunst in der Gesangs-Musik des 16.–17. Jahrhunderts (1535–1650) (Leipzig, 1902/R)
H. Goldschmidt: Die Lehre von der vokalen Ornamentik (Berlin-Charlottenburg, 1907/R)
A. Della Corte, ed.: Canto e bel canto (Turin, 1933)
L. Ricci: Variazioni, cadenze, tradizioni per canto (Milan, 1937)
I. Horsley: ‘Improvised Embellishment in the Performance of Renaissance Polyphonic Music’, JAMS, iv (1951), 3–19
E.T. Ferand: ‘Improvised Vocal Counterpoint in the Late Renaissance and Early Baroque’, AnnM, iv (1956), 129–74
A. Heriot: The Castrati in Opera (London, 1956/R)
V. Duckles: ‘Florid Embellishment in English Song of the Late 16th and Early 17th Centuries’, AnnM, v (1957), 329–45
E.F. Schmid: ‘Joseph Haydn und die vokale Zierpraxis seiner Zeit, dargestellt an einer Arie seines Tobias-Oratoriums’, Konferenz zum Andenken Joseph Haydns: Budapest 1959, 117–30
D. Bartha and L. Somfai, eds.: Haydn als Opernkapellmeister (Budapest, 1960); corrected and updated in New Looks at Italian Opera: Essays in Honor of Donald J. Grout, ed. W.W. Austin (Ithaca, NY, 1968), 172–219
C. Mackerras: ‘Sense about the Appoggiatura’, Opera, xiv (1963), 669–78
A.B. Caswell: The Development of Seventeenth-Century French Vocal Ornamentation and its Influence upon Late Baroque Ornamentation Practice (diss., U. of Minnesota, 1964)
K. Wichmann: Der Ziergesang und die Ausführung der Appoggiatura (Leipzig, 1966)
E. Melkus: ‘Zur Auszierung der Da-capo-Arien in Mozarts Werken’, MJb 1968–70, 159–85
W. Dean: ‘Vocal Embellishment in a Handel Aria’, Studies in Eighteenth-Century Music: a Tribute to Karl Geiringer, ed. H.C.R. Landon and R.E. Chapman (New York and London, 1970), 151–9; rev. in W. Dean: Essays on Opera (Oxford, 1990), 22–9
H.C. Wolff: Die Oper, ii, Mw, xxxix (1971; Eng. trans., 1971)
H.C. Wolff: Originale Gesangsimprovisation des 16. bis 18. Jahrhunderts, Mw, xli (1972; Eng. trans., 1972)
A.B. Caswell: ‘Mme Cinti-Damoreau and the Embellishment of Italian Opera in Paris, 1820–1845’, JAMS, xxviii (1975), 459–92
D.H. Till: English Vocal Ornamentation, 1600–1660 (diss., U. of Oxford, 1975)
G. Buelow: ‘A Lesson in Operatic Performance Practice by Madame Faustina Bordoni’, A Musical Offering: Essays in Honor of Martin Bernstein, ed. E.H. Clinkscale and C. Brook (New York, 1977), 79–96
A. Ransome: ‘Towards an Authentic Vocal Style and Technique in Late Baroque Performance’, EMc, vi (1978), 417–18
W. Dürr: ‘Schubert and Johann Michael Vogl: a Reappraisal’, 19CM, iii (1979–80), 126–40
M. Cyr: ‘Eighteenth-Century French and Italian Singing: Rameau's Writing for the Voice’, ML, lxi (1980), 318–37
F. Neumann: ‘Vorschlag und Appoggiatur in Mozarts Rezitativ’, MJb 1980–83, 363–84
M. Cyr: ‘Performing Rameau's Cantatas’, EMc, xi (1983), 480–89
W. Crutchfield: ‘Vocal Ornamentation in Verdi: the Phonographic Evidence’, 19CM, vii (1983–4), 3–54
G. Durosoir: L’air de cour en France, 1571–1655 (Liège, 1991)
J.W. Hill: ‘Training a Singer for Musica Recitativa in Early Seventeenth-Century Italy: the Case of Baldassare’, Musicologia humana: Studies in Honor of Warren and Ursula Kirkendale, ed. S. Gmeinwieser, D. Hiley and J. Riedlbauer (Florence, 1994), 345–57
E. van Tassel: ‘“Something Utterly New”: Listening to Schubert's Lieder, i: Vogl and the Declamatory Style’, EMc, xxv (1997), 702–14
For further bibliography see Cadenza; Continuo; and Performing practice.
2. Solo and collective improvisation.
It is, however, demonstrably untrue that all jazz must involve improvisation. Many pieces that are unquestionably classifiable as jazz are entirely composed before a performance, and take the form of an arrangement, either fixed in notation or thoroughly memorized by the players; this approach to jazz is characteristic of much music for big band, notably that of Duke Ellington, extended works that combine elements of jazz and Western art music (see Progressive jazz and Third stream), and much jazz-rock.
In some pieces motivic procedures are applied not to freely invented material but to a motif or series of motifs drawn from a theme stated at the outset; this subcategory of motivic improvisation may be termed thematic improvisation, though the derivation of a motif from the theme is generally incidental and merely convenient rather than structurally significant. Thematic improvisation is regularly mentioned in jazz literature in connection with the music of Sonny Rollins, but it has scarcely any meaning for Rollins's work (for further discussion see Rollins, Sonny). It is a more appropriate concept in some free jazz, where musicians develop fragments of thematic material in ways that cannot be construed as melodic paraphrase. Examples include Albert Ayler’s deconstruction of the theme in early versions of Ghosts recorded in 1964 (on the albums Spiritual Unity, ESP 1002, and Ghosts, Debut 144) and Don Cherry’s and Gato Barbieri’s improvisations on Cherry’s album Complete Communion (1965, BN 84226).
Before the late 1950s, motivic improvisation occurred in jazz far less often than either paraphrase or formulaic improvisation. The reasons are clear: until that time a jazz improvisation was expected to accord with an underlying theme; the given theme usually involves a functional progression, which moves at the rate of one, two or four chords per bar; the improvisation itself often moves along quickly. Given these conditions it is extremely difficult to develop a motif systematically without stumbling. Hence among the greatest improvisers in early, swing and bop styles, perhaps only three players consistently utilized motivic techniques: Benny Carter, Count Basie and Thelonious Monk (see also Lewis, John).
The defining characteristic of modal improvisation is that it explores the melodic and harmonic possibilities of a collection of pitches, often corresponding to one of the ecclesiastical modes or to a non-diatonic scale from traditional or non-Western music. The mode is expressed harmonically through drones or through two or more chords that oscillate beneath melodic lines using the same pitches; a typical feature of modal improvisation is therefore harmonic stasis and consequently an absence of incident and progression in the short term. Modal improvisation is not conterminous with Modal jazz, a style in which improvisers regularly select pitches in a loose, perhaps free, perhaps chromatically complex relation to underlying modes. It is much more likely to be found in jazz-rock and other fusions, which not only involve a simple, static harmonic underpinning, but in which the soloist is expected to improvise in close accord with such an underpinning.
GroveJ (B. Kernfeld) [incl. further discussion and examples]
J.-E. Berendt: Das Jazzbuch: Entwicklung und Bedeutung der Jazzmusik (Frankfurt, 1953, 2/1959 as Das neue Jazzbuch, Eng. trans., 1962; enlarged 5/1981 as Das grosse Jazzbuch: von New Orleans bis Jazz Rock, Eng. trans., 1982, as The Jazz Book: from New Orleans to Fusion and Beyond)
A. Hodeir: Hommes et problèmes du jazz, suivi de La religion du jazz (Paris, 1954; Eng. trans., rev., 1956/R, as Jazz: its Evolution and Essence)
G. Schuller: Early Jazz: its Roots and Musical Development (New York, 1968)
A.M. Dauer: ‘Improvisation: zur Technik der spontanen Gestaltung im Jazz’, Jazzforschung, i (1969), 113–32
F. Waidacher: ‘Freiheit in der Beschränkung: zur schöpferischen Arbeit am Grazer Jazz-Institut’, Jazzforschung, i (1969), 140–47
M.L. Stewart: ‘Structural Development in the Jazz Improvisational Technique of Clifford Brown’, Jazzforschung, vi–vii (1974–5), 141–273
E. Jost: Free Jazz (Graz, 1974)
T. Owens: Charlie Parker: Techniques of Improvisation (diss., UCLA, 1974)
L.O. Koch: ‘Ornithology: a Study of Charlie Parker's Music’, Journal of Jazz Studies, ii (1974–5), no.1, pp.61–87; no.2, pp.61–95
J. Patrick: ‘Charlie Parker and Harmonic Sources of Bebop Composition: Thoughts on the Repertory of New Jazz in the 1940s’, Journal of Jazz Studies, ii/2 (1974–5), 3–23
R. Byrnside: ‘The Performer as Creator: Jazz Improvisation’, in C. Hamm, B. Nettl and R. Byrnside: Contemporary Music and Music Cultures (Englewood Cliffs, NJ, 1975), 223–51
L. Gushee: ‘Lester Young's “Shoeshine Boy”’, IMSCR XII: Berkeley 1977, 151–69
D.J. Noll: Zur Improvisation im deutschen Free Jazz: Untersuchungen zur Ästhetik frei improvisierter Klangflächen (Hamburg, 1977)
M.L. Stewart: ‘Some Characteristics of Clifford Brown’s Improvisational Style’, Jazzforschung/Jazz Research, xi (1979), 135–64
D. Bailey: Improvisation: its Nature and Practice in Music (Ashbourne, 1980, 2/1991)
D.B. Zinn: The Structure and Analysis of the Modern Improvised Line, i: Theory (New York and Bryn Mawr, PA, 1981)
M. Berger, E. Berger and J. Patrick: Benny Carter: a Life in American Music (Metuchen, NJ, 1982)
J. Pressing: ‘Pitch Class Set Structures in Contemporary Jazz’, Jazzforschung, xiv (1982), 133–72
W.A. Fraser: Jazzology: a Study of the Tradition in which Jazz Musicians Learn to Improvise (diss., U. of Pennsylvania, 1983)
B. Kernfeld: ‘Two Coltranes’, Annual Review of Jazz Studies, ii (1983), 7–61
L. Porter: John Coltrane’s Music of 1960 through 1967: Jazz Improvisation as Composition (diss., Brandeis U., 1983)
P. Rinzler: ‘McCoy Tyner: Style and Syntax’, Annual Review of Jazz Studies, ii (1983), 109–49
D.L. Moorman: An Analytic Study of Jazz Improvisation: with Suggestions for Performance (diss., New York U., 1984)
L. Porter: ‘John Coltrane’s A Love Supreme: Jazz Improvisation as Composition’, JAMS, xxxviii (1985), 593–621
L. Porter: Lester Young (Boston, 1985)
R.T. Dean: New Structures in Jazz and Improvised Music since 1960 (Milton Keynes, 1992)
H. Martin: Charlie Parker and Thematic Improvisation (Lanham, MD, and London, 1996)
R.T. Dean and H. Smith: Improvisation, Hypermedia and the Arts since 1945 (Amsterdam, 1997)
J. Mehegan: Jazz Improvisation (New York, 1959–65)
J. Coker: Improvising Jazz (Englewood Cliffs, NJ, 1964)
J. LaPorta: A Guide to Improvisation (Boston, 1968)
D. Baker: Jazz Improvisation: a Comprehensive Method of Study for all Players (Chicago, 1969, 2/1983)
D. Baker: Advanced Improvisation (Chicago, 1971)
A. Jaffe: Jazz Theory (Dubuque, IA, 1983)
B. Benward and J. Wildman: Jazz Improvisation in Theory and Practice (Dubuque, IA, 1984)
P.F. Berliner: Thinking in Jazz: the Infinite Art of Improvisation (Chicago and London, 1994)
B. Kernfeld: What to Listen for in Jazz (New Haven, CT, and London, 1995)
I.T. Monson: Saying Something: Jazz Improvisation and Interaction (Chicago, 1996)