The music
From the outset of this research, I decided that the experiments I would carry out as part of it would exclusively feature music composed by myself, and not any already existing pieces. The reason for this was that I wanted to have as complete a control as possible over the different components that were going to be part of these performances, which was also the reason why I decided to perform in the experiments myself. In addition to this, I wanted to experience how the creation 'from scratch' of each of these components would influence my creative process when working on the other ones (more on this can be read in the section Mediating elements during the creative process).
Once I decided that I would compose the music, the first question I had to address was what instrumentation I would use for the experiments. I had already decided that I would be performing myself, and chose to write the music for traverso instead of modern flute or 8-keyed Classical flute because I had wanted to explore the timbral possibilities of this instrument in a new music context for a while. I also felt attracted by the intimacy that the sound of the traverso would necessarily impart to the experiments, given its limited dynamic range, and was interested in how this intimacy might be reflected in the food that I would be serving. Apart from the traverso, I decided to include the baroque violin as the other instrument in the experiments, because I wanted the real-life performances that would derive from this research to be performed by my ensemble Celeritas, which consists of traverso and violin in its most basic form. I felt attracted to this duet formation due to its limited range, which would, I hoped, encourage me to delve into the kind of detailed timbral exploration that I consider to be one of the defining features of my music in general. For the performance Four Bites of Autumn, I decided to also include a viola da gamba, since it was to be substantially longer than the experiments and I felt that the duet formation might not provide enough variety. For this performance, I decided to also include 18th-century repertoire, apart from music that I would compose myself.
I knew quite early on that I wanted to serve a few different small dishes during the experiments, and that each of these would have its own music. Initially, I thought about composing a miniature for each of these bites (a musical ‘bite’ of sorts), but I became increasingly interested in the idea of ‘modulating’ between one bite and the next. This would allow me to create a longer musical tension arc that would match not only the individual dishes, but also the overall structure of the meal, and would give the whole performance a greater sense of unity and focus. This in turn generated the question of how to cue the moments of eating to the audience so there would be as precise a synchronisation as possible between what they were eating and what they were hearing. I decided that the cue would be musical, as a visual or verbal one would have disrupted the flow I was aiming for. The cues for the two experiments and Four Bites of Autumn are described below, as is their musical importance within the overall structure of the pieces.
Of particular importance in all the different pieces is the pitch D, and more specifically the pitch D4. This is the lowest note on the traverso and I consider it to be the foundation of its sound, with a full and rich tone that blends wonderfully with the same note on the baroque violin, either when played as an open string or as a stopped note on the G string. It is also a very stable pitch on my instrument in particular (a copy of a G.A. Rottenburgh traverso from the Brüggen Collection, made in 2020 by Roberto Bando in The Hague), which I always use to tune with the violin. The music for the two experiments and Four Bites of Autumn generates from the note D for this reason.
The rhythmical writing does in general fall into two categories. On the one hand, there are flexible sections that allow the performers to stretch them for as long as needed (depending, for instance, on how long the audience is taking to eat a particular dish). In these, the instruments are not exactly aligned with each other, and there is either no pulse or each instrument follows a different pulse. On the other hand, there are sections in which the different parts are exactly aligned, generally with a stable common pulse. The first type of writing is meant to convey a feeling of stasis without too much development in terms of intensity or affect, while the second type aims to be more teleological and develops the material towards the next section.
In terms of interpretation, there might be small discrepancies between the scores and the recordings, which might be due to last-minute revisions or improvised additions. Being Early Music practitioners used to improvising ornaments, I considered that allowing myself and the other performers to add spontaneous little elements in these pieces would keep the performances fresh.
Finally, before I get into the detailed analysis of all the music I composed as part of this research, I would like to reflect on the table I include below. Drawing it was one of the first things I did when I started thinking about composing for the traverso and baroque violin, and it gives an oversight of the different extended techniques (the ones I was interested in, at least) for each instrument. The blue circles represent sounds that I classified as "equivalent" between the two instruments, while the violet triangles represent combined sounds that I thought might be interesting to explore. This table was extremely useful when composing the music for the two experiments, as will be described later. It is also interesting because of the visual adjacency it establishes between the sounds of the two instruments (further discussed in the section Mediating elements during the creative process).
The scores for all the pieces can be found in Appendix 2. In case any notation is not clear, the performance notes to all the pieces can also be found there.
Experiment 1
For my first experiment, I decided to focus on crossmodal correspondences between basic tastes and different musical parameters (namely relative pitch, degree of roughness, degree of discontinuity, degree of dissonance, and articulation) when thinking about the match between the music and the food (more information about this can be found in the chapter Mediating elements. Crossmodal correspondences). The analysis below will mainly focus on how the music was composed to function as a coherent whole, rather than on how the match between food and music was conceived, which is discussed in the different sections within the chapter Mediating elements.
Cue
When I began composing the music for this experiment, I started by defining the musical cue. I decided that pizzicato sounds from the violin would be a very recognisable sound for the audience to follow. To have particularly resonant pizzicati, I decided to use the open strings of the violin. However, since I was also going to use pizzicati for the musical material corresponding to saltiness and that was to be confined to the pitch space G4-F#5, I took the two lower strings of the violin up an octave by using octave harmonics instead of the regular open string. To decide what the flute should play during this cue, I checked the timbre table shown above to see what sounds would be compatible with the pizzicati of the violin, and decided to have long notes with flattement maintaining the lowest and highest notes of the violin utterance. I gave the cue a simple but characteristic rhythm so it would be easily recognisable.
Introduction
The idea of the Introduction was to present the four pitches of the musical cue while at the same time hinting at material that would be associated with specific basic tastes later on in the piece. Firstly, the pitch D is introduced in both instruments, then the G with the violin glissando, in what will constitute the musical material for the bitter taste. Immediately afterwards, the A is introduced as a partial of D in the harmonic glissando of the violin and then the whistle tones of the flute, both elements crystallising in the A harmonic of the violin. The flute plays an A5 with different articulations and lengths, hinting at the material corresponding to the acidic taste by playing one of them with flatterzunge. The remaining pitch, E, is then introduced by the flute, to be taken up by the violin in a harmonic double stop that will reappear later in the piece corresponding to the sweet taste. A traverso multiphonic is played within that double stop, which will also come back corresponding to the sweet taste. The cue is then played to indicate to the audience that they can start eating the first dish.
Dish 1
The musical material corresponding to the four basic tastes is introduced (a more detailed description of this can be found in the chapter Mediating elements. Crossmodal correspondences). The colours used in the score were initially chosen arbitrarily to make the reading easier for the performers, but happen to coincide with the crossmodal correspondences between colours and basic tastes pointed out in several studies (see Spence & Levitan, 2021).
The proportion between the amounts of time that each musical material occupies within this section is similar to the degree of presence of their corresponding basic taste within the dish (more details on this can be found in the section Mediating elements. Structure).
Since both the acidic and bitter tastes correspond with dissonant harmonic and melodic intervals, and the presence of these two tastes decreases from dish to dish, I decided to start with a fully chromatic selection of pitches that gradually become more diatonic.
Interlude 1-2
The bitter material appears in the violin one last time and becomes a harmonic double stop, corresponding to the sweet taste, a change that represents the decrease in bitterness and increase in sweetness of the second dish with respect to the first. The violin afterwards introduces new melodic material corresponding to the sweet taste.
Dish 2
The musical material corresponding to the bitter and acidic tastes is less prominent, and the presence of material corresponding to the sweet taste is increased.
The selection of chromatic pitches is reduced to 9.
Interlude 2-3
The flute's multiphonic dissolves into a single maintained pitch. The material corresponding to saltiness dissolves as well, and the selection of pitches is reduced yet again.
Dish 3
The musical material corresponding to the bitter taste is totally gone, as is the multiphonic part of the sweet music, which stays as a reminiscence in the harmonic double stop of the violin and the single F# with mesa di voce from the flute. The harmony has been simplified to pandiatonicism. Short fragments of the musical material corresponding to acidity and saltiness are still present, but much less prominent than the ones corresponding to sweetness.
Ending
The flute and violin gradually crystallise in a single high E.
Experiment 2
For the second experiment, I decided to focus on the semantic matching of music and food based on texture words. That is, I chose particular words that could describe both musical and culinary texture, and then paired food and music that could be described by the same texture word (more details about this can be found in the chapter Mediating elements. Semantic matching).
The colours in the score were chosen arbitrarily to facilitate its reading. The salmon coloured squares represent the beginning of a new section, the violet notes represent the musical cue, and the blue texts signal where to find the instructions of what to do when the audience finishes eating each of the dishes.
For this experiment, I decided to have the performers playing back to back, the audience sitting in a circle around them. The music allows for great flexibility so that the players can react to the setting and to each other without having to worry about being exactly together at any given time. This translated itself into rather lengthy verbal instructions in the score, which make for a less intuitive reading. However, this allowed us to memorise the instructions and, when performing, to only focus on the few notes that are written on the score, and on the coloured words as indications of when to move on to the next section. In the analysis, I included fewer examples from the score, as these are not easy to relate to the auditory result, which is what is most relevant here.
The music in this experiment takes into account the comments of the audience about the disturbance caused by their chewing sounds during the first experiment. The foam, which required no chewing, was matched with music with no stable pulse. The chip created a loud crunch in the first bite, a less loud one in the second, etc., which was matched with repeating gestures in diminuendo that start creating the feeling of a pulse. Finally, the paratha required more constant chewing, which was matched with a continuous pulse in the violin part. The sounds of this chewing were meant to interact with the music and become yet another layer of it. I would recommend to anyone trying something similar to inform the audience first about the fact that they should embrace the sounds they make when chewing, and accept them as part of the music, as even in this experiment some audience members mentioned that their experience of the music had been disturbed by their own chewing sounds.
I made the transitions in this experiment longer than those in the first, as a reaction to comments from the audience in this respect.
Introduction
The violin introduces the 'airy' D harmonic that will be suspended throughout the Foam part. The flute comes in with a series of notes played with air sound.
Foam
Over the suspended D harmonic, the flute begins to introduce a series of pitches that will reappear in later parts of the piece. These pitches were chosen to be comfortable on the traverso when changing from one to the next, and they hint at triadic harmonies. The violin, while maintaining the suspended D, plays the cue as a left hand pizzicato.
Once audience finishes eating the foam
The flute plays one last airy utterance with a crescendo that signals the end of the section. It then introduces the next musical gesture, a series of airy pizzicato utterances that become softer and softer, ending with an almost inaudible airy legato version of the same utterance. These utterances are made up from the note series from the Foam part. The flute then takes over the sustained D from the violin, which in turn plays a similar gesture to the flute, consisting of a col legno ricochet on three open strings plus one pitch from those that the flute has introduced previously. The suspended note is progressively left in favour of the newly introduced gestures of violin and flute, which start being superimposed over one another.
Chip
After the violin pizzicato cueing the eating of the audience, the utterances become more spaced out.
Once audience finishes eating the chip
The violin introduces the next gesture, which serves as a cue for the flute to gradually stop playing the previous material. The new gesture from the violin introduces a truly stable pulse, to which the flute adds a polyrhythmic counterpart, which aims to evoke the chewing speed of different audience members.
Once audience finishes eating the paratha
The initial harmonic D comes back in the violin, and all the previous material starts coming back together in preparation of the final dish, which the audience prepares on their plates.
Final dish
The combination of materials continues.
Once audience finishes eating the final dish
The violin stays on its initial harmonic sound, the flute dissolves the pizzicato material, and ends on an airy utterance reminiscent of the one which marked the end of the Foam section, this time with a diminuendo instead of a crescendo.
Four Bites of Autumn
For the public performance Four Bites of Autumn, I decided to include 18th-century music, and not only my own, for several reasons. Firstly, because the length of the performance (about 45 minutes of music in total), which would have implied an enormous effort if I had had to compose all the music myself. Secondly, because I wanted to experiment with how I could apply the insights gained from the experiments to already existing music. Thirdly, since the performance was going to take place in the miller’s house of the windmill De Roos, which was finished in the mid-18th century, it seemed appropriate to have music from that time.
The 18th-century pieces chosen were:
Pietro Locatelli (born 1695 in Bergamo, Italy- died 1764 in Amsterdam, Netherlands)
Trio sonata for flute, violin, and , Op. 5 No. 1 (published in Amsterdam in 1736)
- Andante
- Largo Andante
- Allegro
Albertus Groneman (born 1711 in Hamm, Germany- died 1778 in Den Haag, Netherlands)
Trio sonata for flute, violin, and basso continuo, Op. 2 No. 6
- Adagio
- Tempo di minuetto
Arcangelo Corelli (born 1653 in Fusignano, Italy- died 1713 in Rome, Italy)
Concerto Grosso Op. 6 No. 9 (published in Amsterdam by Estienne Roger in 1714)
- Preludio Largo
- Allemanda Allegro
- Corrente Vivace
- Gavotta. Allegro
- Adagio
- Minuetto. Vivace
Michel de la Barre (born 1675 in Paris, France- died 1745, also in Paris)
Septiéme Suite a deux Flûtes traversieres Sans Basse, Cinquiéme Livre (published in Paris by Foucault in 1713, sold by Roger in Amsterdam)
- Prélude
- Allemande
- Gigue a la maniere Italienne*
- Les Ramages*
- Gavotte
*From Cinquième Concert à deux Flûtes Traversières sans basse, by Michel Pignolet de Montéclair (born 1667 in Andelot-Blancheville, France- died 1737 in Domont, France)
Le Sieur de Machy (active between ca. 1655-1700)
Suite de pièces de violle No. 1
- Prelude
- Sarabande
- Gigue
I decided to reduce my compositional contributions to musical preludes that would prepare the audience for the next bite, which would be eaten while the corresponding 18th-century piece was played. This meant that I would have to accommodate my musical language to that of the other pieces to some extent. I had already done something similar in A Year in Four Bites, the first performance I ever gave with a food and music pairing, for which I created pandiatonic soundscapes, the harmonies of which gradually melted into tonal patterns. For Four Bites of Autumn, I decided to start out with such soundscapes, then simplifying them towards the three pitches that would form the cue. The last chord of the cue would then be understood in the key of the next 18th-century piece and would lead to a small chorale that would act as a tonal prelude.
Only my preludes are analysed here, as an analysis of the 18th-century pieces would exceed the scope and focus of this research (some comments pertaining to the choice of 18th-century pieces are given in the chapters Mediating elements. Semantic matching and Mediating elements. Narrative).
Cue
As in the first experiment, I took the cue in this performance as an important generative element. I decided that it would be formed by the three pitches E, A, and D, the three common notes between the scales of G major, B minor, D major, F major, D minor, and A major, the keys of all the 18th-century pieces that would be played. The feeling of a chord with a suspension that these three notes give when played together in whatever disposition allowed me to treat them as a functional chord within the context of each of the previously mentioned keys. At the same time, they are not characteristic enough as a tonal chord to make a non-tonal arrival to them sound striking. How I resolved this chord in different ways can be seen and heard in the musical examples featured during the analysis.
Prelude to Dish 1
The beginning is meant to generate the chord that will form the musical cue. Everyone starts on a D4, a common pitch to all the instruments, from which the other pitches from the cue, E and A, emerge,
The last note of the the cue in the flute part (E) is held, the gamba adds a D and finally the violin adds the A in a new configuration of the three pitches. The violin changes to a G, identifying the chord as a 4/2 (in basso continuo notation), which resolves in a stepwise descending bass motion to a 6 chord, a gesture that is spun out in a typical 18th-century "moto del basso". A three-part chorale functions as a tonal prelude to the trio sonata by Locatelli.
Prelude to Dish 2
The prelude starts with a pandiatonic soundscape with the notes of G major, which gradually become reduced to the notes of the cue. As with the first prelude, the flute maintains the last note of the cue, a new configuration of the chord is presented. The violin raises its A to an A#, then the gamba lowers its D to a C#, creating a #6 chord which resolves to B minor. A short chorale-like prelude ends on a Phrygian cadence in B minor, as an introduction to the trio sonata by Groneman.
Prelude to Dish 3
The gamba and violin play harmonic double stops on the three pitches of the cue plus G (an additional common note between D major and F major, the keys in which the Groneman piece ends and the Corelli piece starts, respectively). Meanwhile, the flute improvises, first with the pitches of D major, then with those of F major. After ending on a chord with the same pitches as the cue, the cue itself is played. The usual procedure occurs, then the flute moves to an F, determining a D minor chord which becomes a natural 6 over the maintained D of the bass. This harmony resolves to a C major chord with a 4th suspension. Again, a chorale-like prelude follows to introduce the Corelli concerto.
The final part of the performance took place in the windmill itself, which meant that the audience had to climb a rather steep staircase. In the interests of accessibility, we decided to have two options for the audience. They could either stay downstairs in the old miller's house and listen to music played on the viola da gamba, or join the violinist and me upstairs in the windmill, where they would listen to music for traverso and violin. Below are both the preludes for the (same) final dish of the performance.
Prelude to Dish 4 (version for gamba solo)
A bass line that explores the lowest notes of the gamba (first the descending tetrachord from F to C, then another one from D to A, making it feel like a half cadence in D minor). Interspersed between the notes of this bass line, short double stops are played in a higher register, confounding the very clear harmonic implications of the bass line.
The cue is adapted in pizzicati. It becomes a pizzicato arpeggiated prelude in D minor.
Prelude to Dish 4 (version for traverso and violin)
Given the long time that passed between the previous dish and this one during the performance (we moved to a different room for this part), I did not consider it necessary to start in the previous key. Therefore, the prelude begins directly in a pandiatonic texture in A major. After the violin introduces an arpeggiated A major chord, somewhat hidden by the dissonances the flute plays against it, a progression begins in which the violin plays double stops with slowly ascending glissandi, and the flute expands on the triadic arpeggio idea of the violin. The two instruments end on a new chordal constellation of the three notes of the cue.
The cue is presented in a duet version, an octave higher than previously. The chorale-like prelude is presented with a bariolage accompaniment from the violin, given the fact that the Prélude from the De la Barre suite it introduces is already a slowly moving piece, and I thought that a bit more movement would be a welcome addition to introduce it.
Cue
In this experiment, I decided to use an extremely simple cue, two pizzicato notes played on open strings from the violin. I decided for this simplicity because I did not want the cue to interrupt the gradual transitions between textures, and orchestrating a melodic fragment in different ways depending on the moment of the piece did not appear to be a good solution, as it would have been less recognisable. Since pizzicato would not be used anywhere else in the piece, I decided that it would be recognisable enough. While this was indeed the case, it did not feel like this cue had the same impact on the audience the more characteristic ones I wrote for the first experiment and Four Bites of Autumn had (this is further discussed in the section Mediating elements. Ritual). It was more apologetic in the sense that the texture that was sounding continued under it, instead of taking a moment to just state the cue, which would have integrated it in the music more. In other words, I would recommend to make such a cue an indispensable part of the music.
