Part 2: Creative Journey

As I stated above, the potential of employing such long-range harmonic techniques to control Pearl’s structure in tandem with a tropological-topical approach to narrative pictorialism was part of my thinking from a relatively early stage in the creative process. As I noted in my diary (written, at this point, with a view to including the work in my doctoral portfolio):

Diary Entry 12 November 19

[C]urrently looking at word painting on structural-narrative level. Thinking about how sinking is inherent feature of early verses in the real world (mourning) and now rising, energetic material can be used to lead up to the narrator’s surging through his dream landscape in excitement to find Pearl. Perhaps have a look at the email from [music psychologist, Prof.] John Sloboda to your student […] about how these gestures actually mimic the sounds people make when they’re sad / excited etc and link with topic theory?…

Moreover, many textual sketches show my attempts to divide the piece up into something like narrative isotopies (with some topical descriptions). Beyond the annotated poem (Figure 2) I later created the textual overview shown below in Figure 10.

Figure 10. i. Sketched narrative overview (undated); and ii. transcription for reference.

i.

ii.

Formal / texture / harmonic overview

However, I recognize that sharing the foregoing analysis (written and assembled post-hoc, after all, but with reference to my notes and sketches) runs the risk of suggesting that the structure of the work was conceived fully (or near-fully) formed, with all of these interconnecting factors in mind. Indeed, I am well aware that to readers, perhaps especially fellow composers, it may all seem too ‘tidy’, and the piece implausibly considered in its conception. Indeed, colleagues (myself included when listening back now) have also commented on the ‘smoothness’ of the work.

Image Gallery: Some(!) of the author’s many sketches for the work, unearthed for this research publication

On one level, I believe this reflects my conscious response to the qualities of the poem: its easeful alliteration and technique of ‘concatenation’, and arguably also the remarkably flowing quality of Armitage’s translation. (Armitage discusses the effects of its poetic devices and his approach in his foreword; 2016: xi–xvii.) However, in a more down-to-earth sense, completing the piece involved (as undoubtedly familiar to most fellow practitioners), a very messy mix between conscious and intuitive decision-making, in which my analytical understanding of the work emerged gradually through its (unplanned) long gestation, right up until the very point of completion and in some aspects, even after its premiere. Failure was (as always for a composer) a real possibility, and given different circumstances (e.g. a 2020 premiere without the intervening pandemic), I would certainly have arrived at a different version of the piece. Owing to circumstances, it had an unusually long period of development, and looking back over my diary in its entirety now, I am very struck by the sheer number of distressed depictions of being ‘lost’, in a ‘funk’, at a ‘low point’, ‘burnt out’, and a palpable sense of alarm at the experience of having to repeatedly ‘chisel’ away at the materials and structure more than I had ever done before (all quotations from my diary, 2019–20). Indeed, I am struck by how little of it seems to be about creative discoveries and solutions; reading between the lines I suspect I was naively awaiting the kinds of epiphanies I had experienced elsewhere (cf. Kaner 2022a; 2024), and when, for whatever reason, these never seemed to occur, I resorted to very slow and gradual realization of my creative aims that involved not only just ‘chiselling’ but even ‘smashing through walls’ (diary entry, 1 October 2020), as part of a creatively difficult and uncomfortable new experience for me.

On a day-to-day basis, while I recognized the nature and structural role of some passages relatively easily as I wrote them (and in turn used these findings to guide their further development), I experienced much difficulty in analysing others. I could often sense in vague, intuitive terms how these passages ‘needed’ to work but did not know how or why this should be achieved:

Diary Entry 1 August 2019

it’s a bit depressing when I realize some stuff [...] works well locally but the larger arc doesn’t feel right. Sometimes it’s too much of the same material cycling around, sometimes the harmonic pace feels inconsistent, etc. etc.

Diary Entry 15 August 2019

Back in London [after a residency in Aldeburgh, Suffolk], and going quite hard at the harmonic outline. It’s pretty tricky as there are so many small decisions involved to really fit everything together convincingly… Making progress though, it’s just frustratingly slow. But I have so much material now and I need a way to see how it all fits together to allow me to start putting it together properly[.]

Some months after the comments above (November 2019) I attempted, and failed (again), to create a harmonic outline of the entire work so far, including (as shown in Figure 11) thinking hierarchically about the role of the bass G at the opening, which I beamed with a white notehead as if on a Schenkerian voice-leading graph.

Figure 11. Early harmonic voice-leading graph, showing early attempt to define hierarchical role of G in the bass (highlighted).

Looking back now, though this structural sketch was simply abandoned after one page (taking it only as far as bar 66), I suspect it nevertheless represents an early step in my journey towards the final harmonic structure of the work.

Crucially, not long before what would have been the deadline for the vocal score (May 2020, if Covid had not intervened) I had a pair of doctoral supervisions with Dr Richard Baker, who alerted me to what he perceived as a lack of sufficient harmonic tension, in my original draft, during the first orchestral interlude (following the lines ‘and I mourn now, with a broken heart, | for my priceless pearl without a spot’). I reflected in this shortly afterwards in my diary:

Diary Entry 3 April 2020

had a lesson with RB last week and he really said what I needed to hear. I’d just not thought enough about the role of tension and darkness near the beginning. I need(ed) to sow seeds so that it can bloom into the foreground later, and this is really making much more sense. For example, in the first choral passage, the harmony was original so sweet—and without tension it just seems twee almost. So, I’ve been thinking about how to subtly introduce these sounds earlier, so that they just seem to appear unexpectedly but inevitably later on.

It’s something that I’ve known since an UG—I remember CW talking about this technique in Schubert, and GB made a big fuss about the idea of an initial source of tension in my music—climaxes and tension have to grow naturally they can’t just be shoved in. He gave example of La Valse. It’s so important. And now things feel like they’re clicking into place.

The two versions can be compared in Figure 12.

Figure 12. Comparison of draft and final version of first orchestral interlude in Pearl

i. Early draft shown to Richard Baker

ii. Final version [note the addition of new material, on a rumbling topic, in the extreme bass].

This discussion proved pivotal for me creatively, crucially, reminding me of a discussion of my work as a Fellow at Tanglewood with Prof. George Benjamin, and his comments on an orchestral work I had written for the LSO’s Panufnik workshop in 2011 [Chasm, subsequently withdrawn]. Benjamin pointed out that I had simply written a climax without any real source of harmonic tension to set the musical forces in motion for such a build-up. He stressed, citing several examples from the repertory, that such tension should be somehow inherent in the material to be convincing. In my attempts to solve this, I referred myself back to Christopher Wintle’s chapters ‘Idea as Steigerung [Intensification]’ (2006: 89–93) and ‘On Intensification’ (101–10) (both of which I read as an undergraduate in Wintle’s analysis classes) for their insightful analyses of the means with which Britten and Chopin achieve such compelling climaxes in their respective Nocturnes for solo piano.

All the same, certain aspects of the piece’s structural design continued to elude me until much later, and I continued to feel a high degree of tension between my intuitive impulses and analytical knowledge:

Diary Entry 26 July 2020

A lot of it is just sitting with the notes and listening to them incredibly carefully, making slight adjustments until they ‘sit’ right with my ears. Not easy to articulate why this might be during the process—but that is how I work. Always guided by what I hear. When I ignore these instincts (usually because of time pressure) I always feel some ambivalence about the results, and it’s always frustrating when I don’t realize his until after a piece is finished...!

So, it may seem mad but it’s a process of having ideas, revising and refining them and endless tweaking. Of course, sometimes the material might ‘want’ to go a certain way, and I might follow it, but then realize that takes it in the wrong direction within the wider shape or trajectory of that passage within the piece as a whole.

This type of struggle was more familiar to me as practitioner and teacher of other composers. (It is described and analysed in relation to another composer’s working process as ‘problem accumulation’ by Pohjannoro, for example (2016: 224).) Yet in previous works my experience had been somewhat different, with such creative difficulties often resolving themselves in a continuous burst of activity towards the end of the composition process that might be considered a ‘flow state’ (Harmat, de Manzano, and Ullén 2021).

Instead, as noted, the creative journey in this work was characterized, apart from a few more intense bursts of activity in its early stages, by a slow realization that the whole piece needed to be complete in a provisional state before the large-scale structural concerns could truly be solved. As I wrote in my diary in November 2019:

Diary Entry 6 November 2019

Today quite comfortably came up with more material later in the piece (ending ideas and elsewhere) which in fact is probably exactly what I need to do be doing to solve earlier problems—the difficult thing with writing those passages is that I don’t know enough of what happens later to know where I need to take the harmony / textures structurally.

Indeed, with passages that did ‘flow’ quickly, it was not immediately obvious how I should integrate them with their surrounding materials, perhaps because certain crucial harmonic details were yet to solidify. This problem became acutely apparent when I realized, returning to complete the unfinished draft in 2021, after a great deal of agonizing, that a significant and lengthy passage (bars 387–435, leading up to the moment before Pearl appears) would need to be transposed down a major second, because of its premature arrival on an earlier version of the ‘maximal shimmer’ chord, which seemed poorly paced and insufficiently dramatic as a result. Extracts from both versions are shown in Figure 13.

Figure 13. Comparison of draft (July 2020) and completed versions of Pearl and transposition/reworking of passage leading to ‘maximal shimmer’ chord. (Note: the bar numbers do not correspond as other changes were also made earlier in the piece.)

i. Original untransposed draft of bass melody below ‘shimmering’ strings

ii. Final version, a tone lower

iii. Earlier arrival of ‘maximal shimmer’ chord in draft version (highlighted in blue)

iv. Completed version, with ‘maximal shimmer’ chord respaced and delayed (blue)

Perhaps most disconcertingly for me, the precise chord progression leading to the moment of ‘maximal shimmer’ ([LL] to [MM] in Figure 13d) underwent many revisions, right until the very day before the submission of the vocal score in February 2022. With hindsight, I believe this was because I had not fully understood its role as ‘pre-climactic’. (An understanding that I perhaps only fully reached, and/or accepted, when revisiting the work and analysing it for this paper.)

Slideshow 1: A small selection of sketches for the climax of Pearl at various stages

Indeed, the sheer number of iterations of this harmonic progression in my sketches (a selection are shown in Slideshow 1) speaks to the degree of fine-tuning required to satisfy the complex harmonic and voice-leading demands of the passage (the ‘maximal shimmer’ chord occupies nearly the full tessitura of the orchestra from the double basses low C to a very top A-flat), not to mention the narrative weight of this highly dramatic and uncanny moment in the poem, which had profound (or, given the profile of the premiere, somewhat terrifying) implications for ‘getting it wrong’.

← Previous: Integrated Narrative, Storyworld and Topical Analysis

Next: Epilogue: Reflections and Responses →