Beyond Improvisation



The experience with the madrigal of Vicentino was the start of my research on how to produce a resemblance with a primary work divergent from interpretation. In its on-stage enactment, I was confronted with the relationship between the phonographically inscribed soundtrack and the live actions of the performers. I decided to play back the soundtrack and to integrate it with semi-improvisatory actions by a viola player and by myself at the guitar. These live parts were a sort of gestural recapitulation of the sounds of the track. They were constituted by very simple textural actions, and were in no need for a notational codification in order to be remembered and performed. This combination proved especially interesting for the ambiguity generated between live performance and playback. For the next experience, I wanted to achieve a similar result without the aid of a pre-recorded soundtrack. The fundamental problem at this stage was how to achieve at the same time a certain amount of clarity (the primary work should keep a degree of recognisability, even if it was still enigmatic to what extent and according to which principles) and unclarity (such recognisability could not take place through semiotic categories).

While reflecting upon the moulding effect of notation on performance, it became clear that semiotic categories do not belong only to notation sensu stricto, but to any form of pre-constituted component of a musical system – be it the instrumental interface, a pitch system, rhythm, metric division, even human anatomy, perception, consciousness, or linear temporal organization. Notation is more than the writing tool that is used to transcribe the semiotic units of sound onto a rhythmic-diastematic score. Without a cultural and mental approach to sound already as notation, it would not be possible to transfer it into musical writing, that is, to reduce it into pre-constituted semiotic categories. Writing, in its broad sense, can thus be regarded as the established territory of music, the accumulation of stratifications that constitute musical structures, and through which it is possible to concieve of an identity between performance and a given work. But, in the interpretative vision, in which the connection between coding and materiality follows stable rules, where is the “original” sound that performance is reproducing? What is the unspoken sonic paradigm with which resemblance is established? In other words, given that symbolic codification and performative event have no conformity whatsoever with each other, how can a performer think of his or her activity in terms of conformity and identity?


These questions remained suspended as I proceeded in my practice. I decided to “use” a score as starting material for live improvisation, so that an underlying structure might still be “felt” in a performance divergent from the text as such. I transformed a score into a visual diagram that could account for a level of the musical text exceeding its semiotic codification. I chose to work on the song Amarilli mia bella (1602) by Giulio Caccini (~1550–1618), a piece of music whose composition displays an extremely forceful rhetorical organisation, which constituted the main material for the diagram. Soon after that, improvisation presented me with operative problems that interfered with the purpose of my research. From this, it emerged that improvisation is likely to present two opposite risks: 1) excessive semiotic clarity: while improvising it is easy for a performer to fall back into predetermined categories that, operatively speaking, do not differ from those of a fixed textuality. I experienced that pre-constructed semiotic categories can operate in improvisation with even more strength than in the interpretation of written music. In the first place, the anatomical and instrumental interfaces are much more likely to impose their fixidity upon the sounding result, moulding it accordingly. Moreover, internalised formal structures, together with the will and habits of the performer, are likely to come to the fore much more forcefully. 2) Excessive unclarity: in the improvisatory situation where the musician tries to abandon such structures in order to open the performative dimension to the unexpected, it is very difficult to adhere to the task of trying to produce a resemblance with a given primary work.

If I wanted to jeopardise the primacy of semiotic structures, which are conveyed not only through notation, this could not be achieved by means of improvisation alone. When abandoning the territories of notation, a performer enters, willingly or not, other territories, with other codes, stratifications, and constraints. Operatively speaking, improvisation and composition do not differ so much from each other. The problem does not simply lie in the abandonment of text as authority. The problem for my musical practice was rather how to problematise what constitutes musical identity in interpretation altogether. From the perspective of the “of” connecting “performance” and “written music”, the practice of improvisation shifted completely the balance of interpretation towards the former. Textuality had been eliminated, but in a “false” way, in that its (supposed) objectivity had been obliterated in favour of the complete (supposed) objectivity of the performer. Yet, the locus of the “of”, through whose transformation I wanted to bypass both subjective and objective approaches, had been eluded.

In my research, I had been asking myself where the sound that an interpreter “reproduces” lies. This unsuccessful experience with improvisation helped in my realisation that the interpretation of written music necessarily entails the belief in the existence of an original sound, present before performance, and ideally located before/behind the score. For a composition, this is the fictitious sound originated in the composer’s mind, supposedly in dialogue with itself. In approaching a score, the interpreter has to penetrate a depth to discover a hidden dimension that needs to be brought back to life in performance. However strange this might sound, in the case of improvisation, the problem of this “original” is not eliminated, it is just shifted to a different position. If the interpreter imitates a sound fictitiously present in someone else’s mind, then the improviser imitates what is fictitiously present in his or her own mind, the sounds produced must necessarily reflect those originating in such a dialogue.


P O W E R S   O F   D I V E R G E N C E

n(Amarilli-1), after Giulio Caccini’s song Amarilli mia bella

Composer: Giulio Caccini (~1550-1618)
Primary work: Amarilli mia bella
Published in Le nuove musiche (1602)
link to webpage

n(Amarilli-1) is a series of performances after Giulio Caccini’s song Amarilli mia bella. The starting point for each performance was the forceful rhetoric structure of the song. The score of the primary work was first transcribed into a graph, where three lines stand for the extremes in the melodic range: the red line is the body singing voice (G note), the blue line the body of Amarilli (high D note), the black line corresponds to the visceral interiority of the singing voice (low D note). Vertical space represents frequency, while horizontal space represents time, as in a standardly notated score. Pitches are represented by rings, connected to each other in order to emphasise the profile of melodic line. A bigger ring is used whenever the melody hits one of the lines that represents the three main pitches. The graph is divided into six vectors, each corresponding to a rhetoric situation and to a bodily scenario of the song. The six vectors provided the subtext for the trans-codification of Caccini’s song into each new performance.

B E Y O N D   I M P R O V I S A T I O N

T H E   V E C T O R S   O F   T H E   B O D Y

c a s e   s t u d y :   g i u l i o   c a c c i n i

Two possible sonic re-enactment of Amarilli mia bella. The markers show at which position the vectors intersect with each other

vector 1

vector 1

vector 2

vector 3

vector 4

vector 2

vector 3

vector 5

vector 4

The two notes enclosing the melodic range, symbolised by the red and blue lines, describe a physical space: the D (blue) is the shining, yet distant, world of the beloved one, of Amarilli; the G (red) is the opaque world of the singing persona, its appearance often coinciding with the words “mia”, “mio” [my] etc. The D, first element of difference in the harmonic series, is uncompromisingly other than the G, yet it belongs to some extent to its projection. In the first two phrases, the singing voice is reaching for Amarilli, trying to “throw” melodic phrases at it and to drag her closer to himself. Three times the melody lingers on the D and then it is dissolved into a different descending melody, each with its own varied affective nuance – the first time twisted, as if tormented, in the harmonic contour; lightening up on a C the second time, with the word “credi” [believe], as if illuminating the lover’s sincerity; lingering on the Bb, as if to rejoice Amarilli’s beauty the third time.

V E C T O R   1   –   d i s t a n c e

"Amarilli, mia bella,
Non credi, o del mio cor dolce desio,"


"Amaryllis, my lovely one,
do you not believe, o my heart's sweet desire,"

vector 5

vector 6

V E C T O R   2   –   d e s i r e

"D'esser tu l'amor mio?"


"That you are my love?"

On the fourth phrase a stronger gesture is accomplished: the pitches presented at the beginning are sung in a descending line, from D to G: with the words “d’esser tu l’amor mio” [that you are my love] the singer is trying to materialise his wish to draw closer two worlds previously presented as distant – the incredulous beloved, the passionate lover.

V E C T O R   3   –   s u s p i c i o n

"Credilo pur: e se timor t'assale,
Prendi questo mio strale.


"Believe it thus: and if fear assails you,
Take this arrow of mine.

In the middle section, the repeated A conglomerates into a static membrane lying between Amarilli and her lover: the membrane of suspicion, against which the singer continuously directs his voice in the attempt to undermine it with his sincerity. The verbal text itself becomes corrugated by harsh consonant sounds (‘cr-‘, ‘pur’, ‘-mor’, ‘pr-‘, ‘str-‘).

V E C T O R   4   –   l a c e r a t i o n

"Aprimi il petto e vedrai scritto in core:"


"Open my breast and see written on my heart:"

In a recalling of the successful attempt to draw Amarilli close to himself (what happened in vector 2), in the climax of the section the singer starts from the D pitch – the open sound ‘a’ emphasising the pathos of the moment – to descend and break the membrane of doubt. But what happens is the unexpected break not only of it, but of the interiority of the singer himself: the “proto-leading tone” F sharp weakens into a F flat, and the melody leads towards the diatesseron low D. In a “moment of profound erotic surrender” (McClary, 2007, 99), the singer is opening his breast to show the inscription on his own heart. The gesture is of incredible violence: Amarilli has taken the blade of a dart to lacerate the body of the lover, and now she is contemplating his open chest, where the D, perhaps the sign of the possibility of similarity and union between lovers, pulsates.

V E C T O R   5   –   p e n e t r a t i o n

"Amarilli, Amarilli,
Amarilli è il mio amore.


"Amaryllis, Amaryllis,
Amaryllis is my beloved.

Straight after that, the singer starts protruding again towards the distant and beautiful Amarilli. The leading tone that introduces to his own interiority is the starting point of an ascent that happens through three slow and painful fits, the physical effort of which is also underscored by the use of secondary dominants; until he reaches the D passing through the C sharp as if it were a key of access to the interiority of Amarilli.

V E C T O R   6   –   r a p t u r e

"Amarilli è il mio amore!"


"Amaryllis is my beloved!"

The ending figuration, a sort of melisma carrying the liberating force of a conclusive amen in a sacred chant, sees the access to the interior world of Amarilli: for the first time we have a pitch above the diapente, the E, suggesting erotic rapture.