On Musical Imagination as part of my working method

I have been thinking a lot about the role that musical imagination plays in my compositional working method. To inform this reflection, I have been immersing myself in literature on the topic, most recently a book chapter by Freya Bailes and Laura Bishop (Musical Imagery in the Creative Process[1]). What follows here is a reflection on examples from my own recent practice, and I will consider these within frameworks proposed by Bailes and Bishop.

I have long been aware that I do my most vital creative thinking away from my desk, piano and/or computer. My most vivid ideas for material emerge in my ‘inner ear’ when I am either drifting off to sleep, in the shower, cooking, walking outside, or driving. In all of these cases, I am by myself, or at least, not engaged in a particularly cerebral activity, like having a conversation. My mind is free to invent, but my body is in most cases occupied doing something else.

What is it about being away from my designated composing space (namely a home office with a digital piano, desk and laptop) that enables optimal functioning of my musical imagination?  Is it that I am not poised at a manuscript, not feeling immediate pressure for a formed idea to flow out onto the page? I know I am not alone in this experience.

In my designated composing workspace

When ideas for material come to mind, I often note down a verbal description of what I am hearing (in the optimistic hope that I will be able to reconstruct something resembling the idea at my piano in the morning). If the idea is something like a gesture or a melody, a scale-run, or some other short pattern that I have some hope of writing out in a short sitting, then I will often race to the piano to hammer out what I think I’m hearing, refining it to as close as what I think I heard. There are also shades between these two extremes, for example, I might not get as far as working out precise pitches, but still notate a precise rhythm with an indication of contour. Tempo markings are also crucial to maximise the chances of my sketches serving as useful memory aids.

Just yesterday I felt moved to start working on a short but virtuosic piece for soprano recorder. I had bought recorders for my two young sons as well as one for myself, and I had been listening to a number of contemporary recorder pieces earlier that day. It was while drifting to sleep that a vivid idea spontaneously came my mind. I should point out that it seems only to be on days/periods when I have been doing some sort of composition-related activity or task, even just exploring new repertoire, that my musical imagination ignites – and even then it will only happen at points in the day when conditions are right (ie I’m alone, but engaged in another non-verbal type of activity).  

Other times, for something that is larger-scale or more complex, the process of translating from my mind’s ear to page is much more complicated and time-consuming, to the point that I have little hope of transcribing in a single sitting anything approaching a precise representation of the idea I am ‘holding’ in my mind. The aural image is fleeting and transient, and if I am distracted, or move onto to another task before doing something to document the image, then it will disappear. This is certainly the case for material that I am thinking about for the first time. In cases where I have meditated upon material more extensively (ie a piece that I’ve been working on for some period of weeks or months), I am more able to recall, reconstruct and rehearse what seem like variations of the initial musical image.

A case in point is my work on a commission for amateur SATB choir from the Glasgow School of Art Choir.

In the early stages of thinking about this piece I explored existing repertoire for choir, including pieces by Nico Muhly, Anne Boyd, Charlotte Bray, David Lang, Kerry Andrew, Katy Cooper, among others.

Just on this point: some of my composer peers I know think differently from me on this, but I strongly endorse the benefits of exposing myself to existing repertoire as a source of inspiration – to develop my awareness of new techniques, resourceful approaches to instrumentation and approaches to notation. I remember many years ago a peer at a music festival commenting that while he was in the midst of a piece he would go to great lengths to avoid listening to any other music. But equally I can point to composers who espouse the same viewpoint as me.

Alongside this stage of immersion in related repertoire, I thought a lot about the parameters of the commission brief, specifically that I aimed to write music that I felt was representative of my style, but was simple and direct in its notation. This posed an interesting creative constraint, one that I have grappled with over the last two years, to the point that I have produced numerous versions of the opening few minutes using a variety of notational approaches, ranging from very precise to indeterminate notation.

The point that I most want to make here is that throughout this two (plus)-year period, I have worked with ideas that nevertheless bear close resemblance to those which I initially imagined. I recall very vividly hearing a texture, initially harmonically static, built around the note E, then scale descending scale steps to C# in the key of C# minor. In fact, I outlined this idea in a video blog I made for this project in May 2020. Normally I eschew tonality, but it felt compelled (in a positive way) to use a key for this piece, not just to make my music more easily performable, but I felt that consonance would lend itself to the immersive and resonant atmosphere that I was aiming for.

It was only after coming up with some initial musical ideas that I identified the text for this piece – a poem by Judith Bishop entitled 14 Weeks. The poem is all about the early stages of pregnancy, and so this very much fed into the aesthetic aim to create a piece based on a gradual accumulation of textural density, possibly to evoke the idea of cell division in the course of a growing foetus, and definitely to suggest something of the enveloped, womb-like environment of a developing foetus. In other words I sought to write this piece from the perspective of a foetus in the womb.

I remember clearly some of the occasions early on when I mentally rehearsed ideas for this piece away from my composing desk/piano. Typical situations included times when I was out walking for exercise/leisure, or cooking at home. I relied considerably on the voice recorder in my mobile phone to capture both sung fragments as well as verbal descriptions for melodic patterns formal trajectories. For example, in a voice memo from 12 September 2019 I sang a call-response type of idea, consisting of an ascending C-D-E pattern followed by a descending C-B-A-G phrase. Then on 14 October 2019 I noted that I wanted to maximise harmonic tension within a tonal framework. On 3 January 2020 I noted that I wanted to aim for a single-spanning “block” type of form, where a single idea unfolds and evolves gradually.

Listening back to my voice memos, and also looking back at countless score drafts (handwritten and typeset), it’s possible to track a very clear progression and evolution through time of my material. For example, I’ve realised that fragments of scale patterns have been features of my ideas for this piece, really since its inception.

At no point in this process have I felt as though the whole piece could flow freely from my mind to the page. The process of transferring what I imagine to something notated concrete in staff notation is labour-intensive and time consuming. It’s far from an automatic procedure (how I wish I could just plug my brain into a computer and download it that way, or even to capture an audio recording of what I hear internally!). Every iteration of drafted material has required significant effort to support the mental improvisation demanded by this project. I became even more dependent on my imagination to “audition” material as it developed due to my decision to embrace indeterminate notation, specifically the presentation of short melodic cells that singers repeat freely and independently until cued to change. Once I’d made the decision to pursue this approach as a means to simplify the look of my score, I instantly increased the difficulty for me as composer since I could no longer rely on the playback feature of my notation software. 

Fig 1:  A page excerpted from current version (in draft) using indeterminate notation.

The indeterminate nature of this piece, specifically the difficulty entailed in hearing an increasingly complex and multifaceted texture in my mind’s ear, led me to embrace a digital audio workstation (GarageBand), to assist me with the process of testing out textures and forms built from layered, repeating melodic cells. The idea to use GarageBand for this was suggested to me by Katy Cooper who is a choral director and frequently uses this as a tool herself. I have found this to be an immensely useful tool, but I do acknowledge that creating a concrete aural draft in this way comes at a cost of limiting freedom. Specifically, once you pin down, even in a provisional way, a version of something which you then play over and over to yourself, it feels difficult to ‘unhear’ that version. This brings to mind a somewhat related point made by Bailes and Bishop a number of composers have commented that that improvising mentally (ie in your mind’s ear) provides a composer with greater freedom than improvising at an instrument; conversely, improvising at an instrument can have a restricting effect.[2]

I have spent some time considering my recollections of musical imaging within a theoretical framework that proposes four different orders of imaging. Adapted by Bailes and Bishop from Ainsworth-Land,[3] this is a called a “Stage Development Model”. In its original form, it was intended for use in the context of visual art research, and so Bailes and Bishop have considered how imagining by composers might applied to this framework. In Ainsworth-Land’s original version, First Order Imaging refers to images that are “spontaneous”, “direct representation” and “realistic”.[4] Such images are created “out of need or survival”. Second Order Imaging features images that are still “comfortable” and “predictable” but show some evidence of an “ability to manipulate” as well as being “goal-directed”. Third Order imaging  is “abstract and symbolic”, showing signs of “innovation” and crucially projecting “integrated synthesis  of old and new abstractions”. Finally, Fourth Order Imaging consists of a “renunciation of control”, is “chaotic”, “illuminating”, and showing a “receptivity to unconscious material”[5]

Bailes and Bishop explore how this model can be applied to the context of imaging in musical composition.  They skip over the First Order since it entails the most basic form of “re-presentation of sensory information”[6]. I understand this to refer to a scenario of someone re-playing a song they have heard in their mind’s ear without any sort of alteration or modification. 

It is in Second-Order Imaging where manipulation of a pre-existing sonic artefact takes place: “Generating and mentally rehearsing  a personal interpretation of a music performance is an obvious example.”[7] This form of imaging is certainly something that I experience myself engaging with as part of my process, such as when I am reviewing material I have notated, something that can’t be performed readily ono the piano or played back using music notation software (as in the case of my indeterminate material for choir). I have observed myself consciously re-playing ideas in my mind over and over, intervening mentally with variations as a means to test ways to extend an idea (ie. “where should I go next?”…”Does this phrase flow well into the next phrase?”…. ).

In Third-order imaging a composer will be able to form images, unconstrained by the limits of existing musical instruments (eg. the semitonal tuning of a standard piano is an example of a constraint). Bailes and Bishop explain that this order “necessitates a renunciation  of familiar patterns, and a susceptibility to thinking and imagining in a new way.”[8] A distinction between this and Seond-Order Imaging is that Third-Order involves a capacity to “spontaneously ‘receive’ unconscious material.” Certainly this accounts for experiences I have had when an idea seems to ‘hit’ me all of a sudden, something I consider to be novel and therefore haven’t ‘heard’ before.

I would like to think that my imaging reaches into Fourth-Order, at least some of the time. Bailes and Bishop write that this order is also characterised by “spontaneity”, but a defining characteristic is that it tends to be associated with what we refer to as “inspiration”.[9] In setting this out, Bailes and Bishop dispel the myth that imaging at this level requires that music flows “in a sequential manner from mind to paper”.[10]

A key point I take away from this is that Fourth-order Imaging is not necessarily superior to orders 1-3. All are necessary in the course of developing a creative output. I remember times, early on in the process of writing 14 Weeks, when through my mental improvising I stumbled upon some melodic patterning that resembled a passage from Benjamin Britten’s   The Company of Heaven. I recalled an except for a performance of this piece in the course of thinking about similarities with my material, but this didn’t involve any sort of manipulation or extension of Britten’s piece. As mentioned above, I regularly engage with Second-Order Imaging in the course of developing and refining material. Conversely, if I were to operate exclusively at the level of Fourth-Order Imaging, I expect that I would never actually finish a piece. I am reminded of the saying “1% inspiration, 99% perspiration”, and it’s true – much of the mental imaging I engage with is at the levels of Second and Third Orders. More broadly though, thinking this through makes me more appreciative, but also more critically aware, of the different ways that I process ideas mentally in the service of my creative practice. This might even help me to adjust my working method in order to strike an optimal balance between composing at and away from my usual work space.


[1] Freya Bailes and Laura Bishop, ‘Musical Imagery in the Creative Process’. In Dave Collins (ed), The Act of Musical Composition: Studies in the Creative Process, (Routledge, 2016), 53-78.

[2] Bailes and Bishop, ‘Musical Imagery in the Creative Process’, 56-7.

[3] Bailes and Bishop, ‘Musical Imagery in the Creative Process’, 60.

[4] Vaune Ainsworth-Land, ‘Imagery and creativity: an integrating perspective’, the Journal of Creative Behaviours. 16/1 (1982), 12.

[5] Vaune Ainsworth-Land, ‘Imagery and creativity: an integrating perspective’, the Journal of Creative Behaviours. 16/1 (1982), 5-28.   

[6] Bailes and Bishop, ‘Musical Imagery in the Creative Process’, 59.

[7] ibid

[8] ibid

[9] ibid

[10] Bailes and Bishop, ‘Musical Imagery in the Creative Process’, 60.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s