The previous three chapters discussed uses of pre-recorded and live spoken text where that text was kept relatively intact semantically, according to Kane’s source text. This chapter moves on to ‘tape-cutting’ techniques, where bits of text are cut up, fragmented and treated more like musical material from which a sonic texture can be made.
As discussed in Chapter 1 – Background, the inspiration for this work comes from techniques of tape-cutting and vinyl scratching found in electronic music, with analogies in the theatre such as Beckett’s stream-of-consciousness piece Not I. In this chapter, I outline research in two earlier works, Unleashed and Illusions, and go on to explain how these were developed in Scene 5B of 4.48 Psychosis and then tested in two rounds of workshops. I will discuss how learning points from these workshops and the earlier works informed a U-turn in compositional approach in writing this scene, whilst striving to retain the original concept and spirit of the scene.
Unleashed (2012) is a short music theatre piece based on verbatim transcripts of interviews with gay men about their sex lives. The first scene of Unleashed illustrates my early experimentations with tape-cutting techniques, mimicking those of Cassetteboy, The Avalanches and Beckett. However, Unleashed did not work with recorded material but rather applied similar principles to a score: a cut-and-spliced text collage to be performed live by a group of actors and one singer. To illustrate this, here is an excerpt from the original verbatim text from one of the source interviews:
I’ve had this with friends before and I have like a sad I don’t have one memory where I’m like yes that’s like my pinnacle, I don’t think I’ve had my pinnacle, or i hope not. I hope I’ve not reached that point where it’s all downhill, I’m hoping for something slightly better that I’ve not had yet.
Many ‘samples’ (very short excerpts) were taken from this text and re-ordered/repeated to create a ‘stream of consciousness’:
I’ve had this with friends I’ve had this I’ve had this with friends I’ve had this with friends before and I have like a sad I don’t have one memory sad sad a sad I don’t have one memory I’ve had this with friends where I’m like yes that’s like my pinnacle, and I have like a sad I don’t have one I’ve had this with friends before and I have like a sad I don’t have one memory where I’m like yes yes yes yes yes yes yes that’s my pinnacle. Yes that’s like my pinnacle I don’t think I’ve had my pinnacle, I don’t think I don’t think I don’t think I’ve had my pinnacle or I hope not. I hope I’ve not reached that point where it’s all I’ve had this I’ve had this with friends I’ve had this I don’t have one memory where that’s my pinnacle with friends yes with friends yes sad yes sad I have like a sad memory where I’m like yes yes yes that’s my pinnacle. I don’t think I’ve had my pinnacle. or I hope not I hope not. I hope I’ve not reached that point where it’s all downhill all downhill sad yes, I’m hoping for something slightly better that I’ve not had yet.
This text was separated between a singer and a speaker, so that a singer and two viola players memorised musical motifs for the words I’ve/I, like, and (see Figure 30), and sung/played only these words from the text, highlighted above in red. The black text was spoken by an actor, the pace of which dictated the tempo and flow of the music, without pulse or barlines. Underlined text indicates the words that should be spoken by the whole chorus of actors. Figure 31 shows an excerpt from the performing score with the sung text raised onto a line above for ease of reading; the upper line is the singer’s part, the lower line the actors’. Notated fragments for the violas are only given where they differ from the memorised key.
A score of Scene 1 of Unleashed can be seen here and a video of this passage from a work-in-progress presentation can be seen here:
The rehearsal process for this scene exposed the following problems:
- Intense rehearsal was required to maintain a constant flow without hesitation or mistakes by the performers
- Memorising the stream-of-consciousness text was extremely difficult, almost impossible.
- It was not robust: if any of the vocal performers made a mistake, the performance would likely fall apart because there were so few cue or anchor points on which to reunite.
Despite the rehearsal difficulties, the performances demonstrated that this tape-cutting technique brought out a sense of anger, confusion and uncertainty. The circumlocutions, repeated hesitations and superfluous conjunctive words brought out the vulnerability of the source text, and the hocketing between performers gave a sense of ‘common experience’. One personal story was being told but across a number of performers, indicating that this was a shared story, a common experience, possibly happening in multiple, similar iterations in many different lives.
This sense of anger and confusion was important to bear in mind when it came to dealing with 4.48 Psychosis. The techniques developed in Unleashed were used developed further in Scene 5B, as I will discuss later. But first it is important to illustrate another collage technique that laid the groundwork for Scene 5B, this time using recorded rather than live material, like the work of The Avalanches and Cassetteboy. A good example is found in Illusions, written a year before 4.48 Psychosis.
In March 2015, whilst sketching 4.48 Psychosis, I wrote Illusions, a short collaboration with performance artist David Hoyle for live ensemble with video projection, commissioned by the London Sinfonietta as a ‘Note to the new Government’ and to be premiered at the Southbank Centre two days after the UK General Election 2015.
We filmed David’s Hoyle’s ‘stream of consciousness’ improvised rant based on our chosen themes of democracy and gender. This provided two hours of audio-visual source material to which I applied the same sampling techniques as in Unleashed, except this time sampling recorded video material rather than text transcriptions. Around 3% of the footage was used for the seven minute piece. The resulting, re-assembled video was put on a click track, and the ensemble music composed in such a way as to mimic the sampling and stuttering techniques of the video material: chords were repeated where words were repeated, as if the live musical soundtrack had been cut and spliced at the same points as the video. To illustrate, four discontiguous excerpts from the unedited footage provided material for a passage of the final piece:
Gender is an illusion. Yeah. Some of us wee-wee out of a fleshy tube, some of us sit down. I do not believe that genitals mirror a gender, because I don’t believe in male and female, masculine feminine.
Fleshy tube or no fleshy tube. To me, yes, there may be biological differences between two people. So what! The idea that you incubate a very separate way of thinking based on one’s genitals. The idea that genitals dictate gender, when there is no such thing as gender. You’re either a cunt or you’re not. The fleshy tube. The idea that the humanoids who wee-wee out of a fleshy tube should have more power than those that don’t.
It’s cunt eat cunt. It’s dog eat dog. It’s vile. It’s nuclear ridden. It’s full of deceit. And we’re all complicit in that.
Underlining indicates samples that were cut from this material and reassembled to form the final collage (See the score of Illusions, bars 123–168):
fleshy fleshy fleshy fleshy fleshy fleshy fleshy fleshy fleshy fleshy fleshy fleshy fleshy tube so what fleshy tube fleshy tube or no no fleshy tube some of us wee-wee out of a fleshy tube y-y-y-y-yes some of us sit down so what y-y-yes the idea that the humanoids who wee-wee out of a fleshy tube should have more power than those that don’t it’s vile vile vile vile vile gender it’s vile let’s get real gender masculine is an illusion we’re all gonna die feminine so what take acid I don’t believe vile vile yes there may be biological differences so what you’re gonna die fleshy fleshy fleshy fleshy fleshy fleshy tube
(Underlined ‘fleshy‘ indicates a different sample of the word than the non-underlined sample.)
See a video of this excerpt here:
See the complete video of Illusions here:
The rehearsal of Illusions, in comparison to the difficulties of Unleashed, was much more straightforward. The live ensemble music matched the video easily and required little rehearsal, because everything was absolutely repeatable and synchronised using the click track. All of the effects of tape-cutting techniques such as juxtaposition, stuttering, rapid repetition, were far more effective using recorded material than attempting to recreate these in live performance.
The live approach in Unleashed and the studio approach in Illusions provide the background to discussing how both of these approaches were tried and tested in the writing, workshopping and rehearsal of Scene 5B of the opera.
7.3. Scene 5B
Scene 5B retrospectively tells a story about the main character’s experiences of institutionalised mental health treatment. It is a first person, past tense, montage sequence about many visits to doctors and hospitals over a long period of time. The tone is a barbed, bitter retort to one particular doctor, with whom the main character develops a special relationship through the opera. Here is an example of Kane’s text from this scene (Kane, 2000: 7):
Inscrutable doctors, sensible doctors, way-out doctors, doctors you’d think were fucking patients if you weren’t shown proof otherwise, ask the same questions, put words in my mouth, offer chemical cures for congenital anguish and cover each other’s arses until I want to scream for you, the only doctor who ever touched me voluntarily, who looked me in the eye, who laughed at my gallows humour spoken in the voice from the newly-dug grave, who took the piss when I shaved my head, who lied and said it was nice to see me.
We conducted test readings of this scene during auditions in November 2014 – March 2015. Although many different interpretations of this scene can be justified, the readings that most struck me were those that were uncontrolled, angry, fast-paced, confused. For this reason, I felt that the white-hot anger could best be expressed using techniques similar to those I used in Unleashed and Illusions – both of which successfully manifested anger and/or confusion. Therefore this stream-of-consciousness tape-cutting approach was decided as the general concept for Scene 5B. However, the techniques developed in Unleashed and Illusions would need to be expanded to a much larger ensemble of six voices and twelve players.
Given that the rehearsal issues experienced in Unleashed would be magnified in a larger ensemble setting, a series of workshops in the first year of the residency were designed to do some more exploratory problem-solving around the rehearsal issues with using tape-cutting techniques in live performance. These workshops used some originally-composed material and some leftover material from Unleashed to explore how live performance might be made more practical, such as:
- Having one performer perform both sung and spoken parts together.
- Sampling sung fragments and allowing a performer to trigger these samples while only performing spoken text live.
- Using live electronics to make one instrumentalist or singer sound more like a whole ensemble. For example, adding octave doublings with a digital pitch shift unit, or using digital chorus effects and reverb to enlarge the sound.
All these approaches yielded limited results. Method 1 was not possible because such rapid and continuous switching between spoken and sung voice is very difficult and tiring, and in a continuous stream-of-consciousness, allows no time for breathing. Method 2 was open to performer error in the sample triggering, since vocal performers are not trained with this skill. The development of this dextrous skill took just as long as it did to rehearse the performer to perform the material live, without the triggered samples. Method 3 did achieve a larger sound, but there was significant latency in an acoustic where the audience received as much live sound direct from the performer as they did processed sound from the speakers. On instruments, octave doublings to the lower octave tended to have a slightly granular sound, whereas pitch-shifting on voices was successful within a range of around a third or fourth up or down, but became more obviously artificial outside of that range.
Although these workshops did not produce satisfactory solutions to the problems posed by Unleashed, I continued to write a first draft of this scene for a completely live performance, since there was an opportunity to workshop a full draft in the Year Three Workshops. It was important to focus on the musical intention before being too concerned with practical solutions, which could be addressed after another round of workshops on a completed draft.
7.3.1. First draft – a live version
The text was given the same fragmentation and repetition treatment as Unleashed and Illusions. To illustrate this, the excerpt of Kane’s text shown above was cut up and spliced to create the following stream-of-consciousness, to be spoken by a solo speaker:
Inscrutable doctors, sensible doctors, way-out doctors, doctors you’d think were fucking fucking patients if you Inscrutable sensible you you inscrutable doctors fucking patients you’d think think you’d think were fucking patients if you weren’t shown proof otherwise anguish ask the same same anguish same ask the same questions, put w w w words in my mouth w words my mouth, offer chemical you’d think were fucking the same questions offer chemical cures for con-genital my mouth genital anguish cover each other’s arses until I want I I want to con con doctors and cover each other’s arses until I I want to scream for you the only doctor you the only you touched touched only who ever touched ever touched me voluntarily you who looked me in the eye, who laughed at my my gallows humour spoken spoken in the voice from the newly-dug grave, who took the piss when I I shaved who took the piss when I shaved my head, who lied and said and said it was nice to Who lied. And said it was nice to see me.
This was scored out for all six singers, using similar principles to Unleashed: free rhythm without barlines, the flow and tempo dictated by the fast-spoken text. Suzy was given the solo spoken text and the other cast members had the sung material (in bold red) and spoken chorus (underlined) distributed between them, all hanging off Suzy’s central, spoken thread. An example of the score is shown in Figure 33; the cue line shown in the score was included in every instrumental and vocal part, and the bold and underlined key system was kept in the score to add more information to the cue line. Individual vocal parts were made for this version of Scene 5B, due to the fast pace and visual confusion of the full vocal score. The score of this draft is here, along with a sample vocal part and instrumental part.
Using similar notational practices developed in Unleashed and in the Year 1 workshops, pitches were sustained with extended beams through passages of text. The instrumental ensemble (tutti except percussion) worked primarily in chordal blocks of quartal and diatonic chords, aligned with particular sung words. Some short text phrases were given sung choral gestures in rhythmic notation, barred out, and to which the suggested tempo at the beginning of the scene applied. For example, I gape in horror at the world, shown in Figure 32.
As in Unleashed, pitches for the short sung words (I, and, me, etc) were generally fixed through each section, and raised by semitone steps at various intervals, the structural pacing and location of these transpositions determined intuitively. These short sung words were associated with particular chords in the instrumental ensemble, for example the E minor chord on you’d/you (see Figure 33). The instrumental harmony was also transposed by semitone steps on the same basis as the voices.
Sung words were alternated freely between singers, with a broad rationale of giving particular words to particular singers (e.g. Jen sings and; Clare and Lucy sing me/my), although this was by no means dogmatic. The word ‘I’ was distributed through the cast to make clear to the audience the dramaturgy of the hive-mind; every performer is the first person. Suzy was given all the solo spoken text to provide a character thread through the scene for staging reasons, and to allow for easier rehearsal and memorisation. The significant burden of memorising this scene was balanced against the relatively lighter load that the role of Suzy has through the rest of the opera.
The whole scene was treated in a similar way according to these basic principles illustrated above. With this draft score completed, the next step was to test it in the Year 3 Workshops.
7.3.2. Year 3 workshops
During the read-through workshops in Year 3 we rehearsed an excerpt of Scene 5B between rehearsal figures 24–27 (See the score of this version here, pages 14-19). Singers were rehearsed first alone, then tutti with the band. The workshops exposed the following difficulties:
- Singers found it difficult to anticipate their entries, in the absence of a pulse or rhythmic notation. In addition, speech rhythms and timings varied between several run-throughs, so repeated rehearsal did not necessarily improve the accuracy of performers’ entries.
- Similar problems were experienced by the instrumental ensemble, leading to great difficulty in maintaining rhythmic precision, both between instruments and singers and between individual players in the band.
- Memorisation of the text was extremely difficult because of the circumlocutions in the text and lack of patterns in the repetitions.
These problems were, predictably, similar to those encountered in Unleashed and in the exploratory Year 1 workshops. Although more rehearsal would have significantly improved performance, as it had for Unleashed, the creative team concluded that this version of Scene 5B would be unfeasible in the context of the production constraints of this opera:
- This scene would require so much rehearsal time that it would compromise the performance standards and staging of the rest of the opera. Significant rehearsal time would also be needed with the band, which would not be available.
- This material would not be resilient to mistakes in performance, risking the overall performance quality and consistency expected by paying audience, and placing huge stresses on the performers.
In light of these considerations, I took the decision to change approach from a live performance version to the method that I had used successfully in Illusions – a tape version. A recording of all of the vocal material in the ‘live’ score would be made, and then edited into a collage using the same tape-cutting techniques as Illusions. Just like in that piece, the instrumental music would be re-written entirely and mapped onto this tape part, synchronised with a click track, so the final version of Scene 5B would consist of a recorded vocal collage with live instrumental music.
7.3.3. Making a tape version
Given the decision to now make a tape version of this scene, using the ‘live’ version as a template score, there were now other potential advantages that could be exploited:
- The tape could take advantage of surround-sound dispersion in the hall to give a more immersive sonic experience to the audience.
- Recorded voice samples could be accurately treated to a variety of sound processing such as reverb, distortion and EQ.
- Clarity and balance of text and music could be more carefully controlled, and there would also be no conflicts between amplified sound and live in the auditorium. Close studio-recording of spoken text would give maximum clarity.
- Hocketing between solo and chorus, sung and spoken text could be more rapid and intricate on tape than in live performance, making the texture more energetic, disrupted, stuttered.
- A much more intricate instrumental texture could be written, because the band would now be written in barred, regular metre.
- The memorisation load on the singers would be significantly lightened, thus saving rehearsal time and performance stress.
For these reasons, it was not satisfactory simply to make a pre-recorded version of the live score. To take full advantage of these new possibilities, the scene needed to be completely re-composed, using only the live score as a rough framework. Whilst a tape version afforded many advantages such as clarity and repeatability, it might also make performances ‘flat’, lacking in the energy that a live vocal performance normally has on stage.
The recording of vocal material took place in the first week of the rehearsal period. The material was recorded exactly as indicated in the live score, stem by stem, resulting in a huge palette of materials that were then assembled intuitively in editing software, using the live score only as a very rough guide. Rhythms were changed, scratching/stuttering techniques were used more extensively and generally a denser, more polyphonic texture was created. Sung material was significantly reduced in the new version compared with the live score. For example, many of the measured, sung phrases were cut from the live score entirely and replaced by spoken text. This left mostly only the short words being sung (I, me/my, and, you/your/you’d), although even the frequency of repetitions of these was greatly reduced compared to the live score, with most changed to spoken/shouted words.
The shift away from recorded sung material towards spoken material helped with the following issues:
- It kept a fast ‘machine gun’-like stream of text flowing.
- It improved text comprehension. Sung words tended to be less intelligible on recording in the tessitura in which they had been written.
- It felt more angry and less controlled, closer to the initial audition readings of Kane’s text that had informed my original concept for this scene.
- The finished tape parts were then tested in the rehearsal room, alongside MIDI mock-ups of the instrumental music that had been written to match. As a result of these tests, the Director had concerns about text clarity in particular passages of the tape. As a result, new mixes of each tape part were made with the following adjustments. These can be seen by looking at the screenshots from Logic Pro X below:
- passages of many overlapping layers of text were thinned out by muting many of the layered tracks (visible in the screenshots as muted (greyed out) segments);
- left-right and front-rear panning was used more extensively to spatially separate different voices;
- more of the sung layers were muted to reduce obscuration of the spoken text;
- levels were adjusted to push non-important layers into the background;
- ‘cassette’ distortion effects and EQ filters were reduced in some passages.
The final version of the tape part alone for scene 5B can be heard here:
To summarize, the treatment of Scene 5B was a development of techniques first developed in two earlier works, one using a live approach to tape-cutting and one a recorded approach.
The live approach was first applied to this scene, but many practical problems forced a change to the recorded approach. However, the change of approach afforded many benefits, both musical and practical. In rehearsal the new, tape version of the scene was straightforward to put together, and in performances it was effective in immersing the audience in loud, clear, fast-moving text – perhaps best summed up by a reviewer’s comment that Kane’s text “seems to emanate from everywhere” (Financial Times) (see Chapter 9.1 – Critical and public response).
The significant use of pre-recorded material and the complex and intuitive editing stage that was required to make these tape parts throws up issues for future productions with different casts. These questions will be addressed in the Evaluation, Chapter 9.2.1 – Different casts.
Scene 5B from 4.48 Psychosis (28th May 2016): (video available on request)