1. INTRODUCTION: ‘TREVURR’
‘Trevurr’ is how Trevor Synge-Perrin, a key interlocutor in my doctoral research, signs his emails. He moved from Bristol, where he grew up, to St Just, a small town in Cornwall, in the 2000s; ‘Trevurr’ is his transcription of a West Country English accent. I got to know Trevor in 2018 over several sessions with a dementia-friendly walking group of which he is a member, before spending seven days with him over the course of 2019 at his house and in his local area. The conversations we had over this time took in various themes, including uncertainties around short-term and spatial memory, his musicking, hyperacusis and awareness of it as an impairment among his social groups. The composition we made together, Trevurr, documents the entirety of this research process (Sound Example 1).Footnote 1 It also explores some of Trevor’s lived experience of sound through phonography, using a variety of methods that became increasingly dialogic over time. My intention was for this process to arrive at a kind of negotiated, ‘co-aesthetic’ production (Cox and Wright Reference Cox, Wright, Fardon and Gledhill2012: 125), which I tentatively referred to as ‘soundscape co-composition’ but as I shall discuss later, it may be better described as a dialogic creative research practice.
Before discussing some of the background to this praxis and providing an account of fieldwork and composition events, it is important to note some of the ethical procedures involved here. Trevor’s participation grew organically from our conversations at the group meetings, with formal informed consent and release forms only signed once I was confident that Trevor understood the project’s aims and methodology, and once the composition was finished, respectively. These forms, as with the wider project, had been approved by the departmental research committee. Furthermore, no potential issues (such as medical concerns, previous traumas, or social impacts) were foreseen by Trevor or those close to him whom I spoke to over the course of fieldwork. Ongoing consent is essential in dialogic research processes and underlies the collaborative impulsion of the project. I have thus opted to describe Trevor’s role using the term ‘interlocutor’ over ‘participant’, ‘informant’, ‘subject’ or the like, as it most accurately reflects the dialogic nature and collaborative intention of the project. ‘Collaborator’, on the other hand, would imply a symmetry of responsibilities that was not possible to achieve here. Raising awareness of the everyday challenges of hearing impairments motivated both Trevor and I in this composition work.
2. BACKGROUND
2.1. The hearing of others
In his essay on hearing in Keywords in Sound, Jonathan Sterne identifies an assumption that can often be found in discussions around sound phenomenology: that through its role in rational discourse, hearing is a faculty that is essential to intelligence and (inter)subjective experience (Sterne Reference Sterne, Novak and Sakakeeny2015: 66). Sterne points out an unacceptable implication of this assumption: that being d/Deaf or hard of hearing through age or hearing damage means being ‘doomed to receding relations’ to these aspects of human experience (ibid.: 73). Psychoacoustics and audiometric research for applications such as telephony, data compression and injury compensation reconceptualised hearing as a measurable aptitude, separate from listening (Sterne Reference Sterne2012; Tkaczyk, Mills, and Hui Reference Tkaczyk, Mills, Hui, Tkaczyk, Mills and Hui2020). Quantifications of frequency range and acuity level established expectations of ability, contributing to a characterisation of hearing as a supposedly ‘natural’ or universal faculty. The resulting biomedical construction of hearing is abstract, with relatively little basis in everyday experiences of sound. Sterne calls thus for the study of a broader plurality of auditory abilities, asking:
[w]hat would sound studies become if we began without the automatic assumptions that we have direct, full access to our own hearing, or through our hearing, direct access to the sonic world, or through the sonic world, intersubjectivity with others? … If no sound is possible without hearing, then sound studies – but also many forms of politics – begins with hearing the hearing of others. (Sterne Reference Sterne2012: 74)
There is now a growing literature developing an auraldiverse, body- or ability-relative (or indeed ‘auralrelative’ (Drever, Reference Drever, Grant, Matthias and Priorforthcoming)) approach to hearing, offering a critique of assumed invariability of hearing in musicology and media studies (e.g., Drever Reference Drever2015, Reference Drever2019; Ellcessor Reference Ellcessor2016; Holmes Reference Holmes2017). Sterne’s ideal of ‘hearing the hearing of others’ is, of course, not possible in a direct or disintermediated sense – hearing makes no soundFootnote 2 – but the phrase might be read as a call for more empathetic or perspective-sharing orientations in sound studies, emphasising hearing abilities and the situated, intersectional sonic experiences they produce.
2.2. Representing experience: ineffable stories
In The Wounded Storyteller: Body, Illness, and Ethics, disability sociologist Arthur Frank characterises illness as the lived experience of disease: a particular case, encompassing all of the ways in which it affects and is woven into a person’s day-to-day life (Frank Reference Frank1995: 3–7). Sociologists can facilitate the telling of someone’s ‘illness story’ (ibid.: 53–9). These are a testimony of how the abstract and universalising concepts of biomedicine – usually discussed in specialised language not intended for non-expert patients – are lived mundanely by individuals. With much of the literature on specific auditory phenotypes written in this register, illness stories could afford a useful grounding for sound studies oriented towards auditory difference.Footnote 3
The prospect of illness stories about hearing raises a basic methodological issue, though: it is difficult to verbally describe auditory sensations and their affective dimensions (e.g., Vikman Reference Vikman, Järviluoma and Wagstaff2002: 148–54; Sterne Reference Sterne2003: 94, also 131–2). Discursive contexts tend to favour articulate interlocutors, such as those who are already skilled at producing narratives or who are versed in specialised terminologies from musicology, acoustics, or sound engineering. An instructive approach can be found in Yasmin Gunaratnam’s work, where she discusses how creative practice can serve to explore and convey ineffable aspects of illness experiences in research presentations (Gunaratnam Reference Gunaratnam, Chamberlayne and Smith2008: 21–6). The formal and semiotic flexibility of poetry enables the creation of intellectual spaces opening the audience up to difference and the unknowable. This medium, she argues, is more amenable in expressing the ambiguities and incoherences of illness experiences than academic discourse.
Creative practice also has precedence in ethnography as a result of the ‘crisis of representation’, spurred by James Clifford’s original 1986 introduction to Writing Culture (2010). Clifford forcefully questioned the adequacy of monographs for describing lived experience, framing ethnography as simultaneously a research method and a highly conventionalised form of output. This critique had aesthetic implications, gesturing towards non-textual forms of ethnographic representation: creative practice with an explicitly ethnographic orientation and communicative intention (Cox et al. Reference Cox, Irving, Wright, Cox, Irving and Wright2016: 2–3), which could be termed ‘creative research practice’. Clifford’s famous provocation, ‘[b]ut what of the ethnographic ear?’ (2010: 12), expanded upon contemporary criticisms of the ethnographic gaze, but also raised the possibility that non-textual representations might allow audiences to form their own sensory impressions of the worlds inhabited and experienced by others.
2.3. Representing others: power and interpretation
Clifford also observed that the people who are represented in ethnographic research often had little input in the interpretation of fieldwork events (2010: 14–19). Texts were not written for them but for an academic readership; they were not included in processes from defining research aims to writing and presenting research outputs. It is important, then, to find ways to try to flatten the hierarchy created by the privileged position of the academic, affording them the power to represent, as far as possible within a given research project. The inclusion of participants as collaborators, co-researchers, or co-authors is an attempt to address this, drawing on Mikhael Bakhtin’s characterisation of the dialogic, co-constructed nature of linguistic meaning: ‘[t]he word in language is half someone else’s’ (Bakhtin Reference Bakhtin and Holquist1981: 293). In anthropology, such praxis has become prevalent enough to be labelled ‘collaborative anthropology’ (Lassiter Reference Lassiter2005: 15–18; Holmes and Marcus Reference Holmes and Marcus2008: 81–3).
Some of these notions were developed by Steven Feld in his work with Kaluli people in Papua New Guinea. In ‘Dialogic Editing: How Kaluli Read Sound and Sentiment’ (Feld Reference Feld1987), he described reading parts of his monograph on Kaluli sound culture (Feld Reference Feld2012a) to some of the same interlocutors who appear in its pages. As he spoke of the concepts and events described and interpreted in the book, they would interject, filling in contextual details and recalling particular instances not present in the summarised form (1987: 197–8). They questioned why he wrote more about a single song or instance of ritual weeping than any of the environmental sounds that tell the weather, season or time of day (ibid.: 204–5). This led Feld to focus on recording these sounds. Following Jean Rouch’s technique of contredon audiovisuel – playback and feedback on recordings to the same individuals being represented (Rouch Reference Rouch and Feld2003, in Feld and Brenneis Reference Feld and Brenneis2004: 467) – he found ways to include imprints of interlocutors’ listening in his composition, Voices of the Rainforest (2011). Letting the tape recorder run, he would invite people to sit and listen, identifying sounds and discussing them. In another instance, having recorded Ulahi singing as she washed items in a stream, he set up separate recordings of different sound sources made there. Ulahi then adjusted level controls while monitoring in headphones to match her recollection of the soundscape as the tracks were mixed down. Feld describes these types of processes as ‘dialogic editing [and] auditing’ (Feld and Brenneis Reference Feld and Brenneis2004: 467; Feld Reference Feld2012b: 202).
2.4. Dialogic creative research practice
Echoing Clifford’s criticisms of the production of ethnographic representations, Grant Kester draws attention to the power dynamics involved in making participatory or collaborative art (Kester Reference Kester2011: 1–10). Is the artist instigating and controlling the production with limited input from others, or ceding control to collaborators in a more symmetrical distribution of power and responsibility? As David Bell observes in an analysis of their conflicting arguments, Kester broadly relates ‘participatory’ art to top-down practices and ‘collaborative’ art to bottom-up ones, while Claire Bishop instead contrasts provocative, artistically focused projects with activist, predictable or even moralistic ones (Bishop Reference Bishop2006, in Bell Reference Bell2017: 73–4). Similar to her earlier criticism (Bishop Reference Bishop2004) of relational aesthetics (Bourriaud Reference Bourriaud, Pleasance and Woods2002: 14–18), Bishop argues that participatory/collaborative art is not concerned with the types or quality of relationships produced. Its success is determined rather by what the art does socially – promoting inclusion, fostering aspiration, reducing isolation – while otherwise leaving the ‘structural inequalities of society uninterrogated’ (2006: 179).
It would seem reasonable to suggest that the kind of practice exemplified by Voices of the Rainforest is participatory (in Kester’s usage) due to its creative direction by Feld, but the relationships involved in its production are not only artistic but also rooted in the research dynamic. Indeed, it is unlikely that many creative projects undertaken for research would meet Kester’s specification of collaborative practice due to the differing institutional responsibilities and positioning of those most commonly involved. Feld’s dialogic interactions with his interlocutors prompted him to make the recordings, and he arranged them with the intention for listeners to ‘hear what inspires the [Kaluli] composers and singers, what they’re singing with and to and about’ (Feld and Brenneis Reference Feld and Brenneis2004: 467). Counter, then, to Kester’s ideal of collaborative production, Kris Rutten, An van Dienderen and Ronald Soetaert’s argument that the ‘immense complexity of the relations between artist/researcher and subject … [means this] relationship will, inevitably, always be unequal’ (Rutten, van Dienderen and Soetaert Reference Rutten, van Dienderen and Soetaert2013: 470). They point instead to Arnd Schneider’s argument, following Trinh T. Minh-ha (Trinh and Chen Reference Trinh and Chen1992), that ‘in ethnographic representations we cannot [unproblematically] speak about or for the other … but at best can speak nearby’ (Schneider Reference Schneider2013: 525). Feld’s practice is one based on inescapably different roles, and is thus better understood as dialogic rather than either participatory or collaborative.
Feld has described this approach to social relationships in his practice as ‘ethnoaesthetic negotiation’ (Feld and Brenneis Reference Feld and Brenneis2004: 467), while Rupert Cox and Christopher Wright refer to it as ‘the development of a co-aesthetic’ (Cox and Wright Reference Cox, Wright, Fardon and Gledhill2012: 125).Footnote 4 Developing co-aestheses with interlocutors may be one way of responding to Lennard Davis’s observation that ‘those who pay attention to art and cultural production have really thought very little about the way in which such endeavours are based on normative practices that imply a normative body and normative communication’ (Davis Reference Davis1995: 124–5). Any attempt to hear the hearing of others through research or creative practice will always need to navigate a series of ethical issues that are structural, discursive, representational and praxial in nature. But, as Sterne and Drever make clear, sound studies should engage more directly with the experienced realities of bodily difference as a key modulator of intersubjective exchange through sound, and following Gunaratnam and Feld, creative research practice might be considered a viable way to explore this. As I shall now outline, this methodology can be adapted to a context of creative research practice exploring illnesses and related sensory experiences.
3. FIELDWORK
3.1. Group fieldwork: first interactions with Trevor
The Sensory Trust is a charity based in Cornwall, organising accessible, nature-based activities for people with dementia and their carers. I spent much of 2018 accompanying its walking groups on their regular excursions, and it was in the ‘Wednesday Wanderers’ (St Ives) group that I met Trevor. I attended three walks and one session with an arts group in Paul, which Trevor is also a member of, before we began working together. The first was in Hayle. Passing dogs frequently yapped at my furry microphone windshield, making it serve an unforeseen purpose in sparking conversations about the research. The interactions of hearing and dementias were relatively undiscussed in the group, and in a cafe afterwards, I gave a short talk about their potential implications for acoustic ecology.
Trevor remembered this talk when I spoke to him for the first time on the next walk, along the river in St Erth. We spoke about his time at art college, music education and instruments, and his early music group. He was also outspoken about his dislike of Op Art and drum kits. The session ended with the usual cafe visit, this time to a very small one in the back of a Post Office. Plates clanked, chairs scraped against the floor, and voices and laughter crescendoed as energetic conversations carried on around the small tables; Trevor seemed withdrawn, waiting in the corner for his ride home. My third meeting with him was at the Paul arts group, in a church. Talking with him about the importance of musicking for his social life, he made a thought-provoking comment: although he knew his hearing had changed in recent years, he could still hear which local piano tuner had tuned the pianos in the various churches where he performs. He thought it was important to acknowledge what people are able to do with their listening, rather than focusing solely on impairments. Later, with the Wednesday Wanderers, the crashing of water over a weir in one of Zennor’s streams brought our conversation onto Foley sound and location recording in radio dramas.
Trevor is an avid listener of radio dramas, describing how he found the sound of bedsheets in The Archers distracting: they sound ‘too bright’ to him, more ‘like the crinkling of a plastic bag than [the] swoosh’ of linen. He then told me of his curiosity about certain sonic effects that would be possible in studio composition: what would it sound like if one layered many recordings of ‘the same oboe playing exactly the same note’? Or if one held a piano chord and compensated for the decay by increasing the level? These effects would be simple to implement, and as Trevor had a piano, we arranged to meet again in our own time.
3.2. One-to-one fieldwork: listening activities
Trevor’s home is a typical Cornish terraced cottage, with stone walls and small windows. He wants to move because it is too dark for him, being registered blind due to macular degeneration. We recorded a held piano chord and several notes of his oboe, and I would usually bring acoustics demonstration recordings or recordings of particular sounds one of us had mentioned earlier for us to listen to. Mostly, though, we talked. Trevor seemed stimulated by the possibility that this project might somehow be useful for others with hearing issues.
My conversations with Trevor often centred on listening activities. The sequencing of these was largely based on topics we had discussed in the previous session or over email, but was also contingent on circumstances; for example, the weather. Similarly to Jayne Wallace et al. in their work collaborating on digital jewellery design with people who have a dementia, it was important to be sensitive to variations in Trevor’s enthusiasm, mood and energy level on different days, being ready to switch, postpone, or drop intended activities (Wallace, Wright, McCarthy, Green, Thomas and Olivier Reference Wallace, Wright, McCarthy, Green, Thomas and Olivier2013). I usually had specific questions I wanted to ask or listening activities to suggest, but would only do so when I thought Trevor would enjoy it. Participant consent is necessarily an ongoing negotiation based on openness and checking for understanding, and it was also important that my relationship with Trevor was not defined solely by my need to conduct research.
Listening activities served as ways of drawing attention to aspects of hearing and listening for discussion and explaining terms that came up in conversation, such as the harmonic series, filtering or just intonation. They included completing online hearing tests, useful not as measurements but as prompts for discussing the highest-pitched sounds that Trevor can hear (e.g., grasshoppers and chaffinches), listening to psychoacoustics demonstration recordings (e.g., Risset tones and reverberation parameters) and playing back recordings from the walks to determine whether we could discern what people were saying.
In situ activities included listing all the sounds we could hear in Trevor’s garden, and walking around St Just, stopping to listen in to the soundscape on a track through nearby fields. This activity was similar to a ‘sensory memory walk’ (Järviluoma and Vikman Reference Järviluoma, Vikman, Richardson, Gorbman and Vernallis2013), as Trevor spoke of his memories of places along the route: a spot near the recycling bins where people would leave surplus furniture; a former priest in the ‘hideous’ church; the quiet winter roads. Differences in our hearing were contrasted on listening to a recording of bats I made in a cave overlooking the Dead Sea: while I could hear the quiet fluttering of wings and some high chirps, Trevor could only hear the louder squeaks. This was, of course, not a straightforward comparison, as I had listened to the recording multiple times before and vividly remember experiencing the malodorant location in person.
Trevor had suggestions for activities and sounds that we could record, too. The layered oboe experiment discussed earlier was confounded by unsteady pitch as Trevor had not played for years, but resulted in our listening to recordings of György Ligeti’s Lontano (Ligeti Reference Ligeti2002) and Niels Lyhne Løkkegaard’s Music for 8 Recorders: Alto Recorder (Løkkegaard Reference Løkkegaard2014). Having recorded a blackbird and humming bees in his garden, we travelled to Porthcurno Beach to record shell fragments underfoot: a sound Trevor recalled fondly in response to a question about sounds of locations further afield. This was audible to us both insitu but difficult to record due to strong winds and omnidirectional microphones.Footnote 5 Trevor later suggested recording in Penzance. He talked about the sea wall at Battery Rocks, where he goes swimming most weeks, having noticed how people’s voices are intelligible much further out from the shore when they reflect off the sections of wall behind a path down to the rocks. Over email towards the end of the year, he wrote out a list of his favourite sounds, which we decided to sound out at the end of the piece: cats purring, blackbirds, cricket bats, oboes, pianos. I think of these activities and events as exercises in companion listening, or listening together as part of a dialogue: they helped to familiarise us to each other’s hearing over the course of the encounter.
4. MCI AND TREVOR’S HEARING
Trevor’s diagnosis is one of mild cognitive impairment (MCI), a syndrome with many similarities to early stage Alzheimer’s disease but which does not follow the same rate of neurological change (Idrizbegovic, Hederstierna, Dahlquist, Nordström, Jelic and Rosenhall Reference Idrizbegovic, Hederstierna, Dahlquist, Nordström, Jelic and Rosenhall2011). MCI is considered either to be an intermediate form of dementia diagnosis or a separate diagnosis involving greatly increased risk of Alzheimer’s. This discrepancy occurs because it is difficult for memory clinicians to distinguish between MCI and Alzheimer’s early on (Golob, Irimajiri and Starr Reference Golob, Irimajiri and Starr2007). Such was not the case for Trevor, though. He was diagnosed first with Alzheimer’s around 2010, but a few years later rediagnosed with MCI, apparently because scans did not show the rate of atrophy that would be expected in Alzheimer’s disease. The rediagnosis had practical implications, as MCI diagnoses do not afford the same level of care assistance from local services in Cornwall, so his visits from community nurses and access to assisted transport stopped.
MCI has a similar auditory phenotype to Alzheimer’s disease (Idrizbegovic et al. Reference Idrizbegovic, Hederstierna, Dahlquist, Nordström, Jelic and Rosenhall2011). Auditory disorientation – impaired ability to track and locate sound sources in busy acoustic environments (Hardy et al. Reference Hardy, Marshall, Golden, Clark, Mummery and Griffiths2016) – is a processing disorder common to both. Trevor has notable difficulty with cocktail party listening scenarios, as was the case in the cafe stops after the walks. These situations were a frequent topic in our conversations, and one of the clearest examples of where his particular hearing abilities and sociality intersect. He described the experience of physical pain and isolation these situations can cause as ‘profoundly disabling’.
Beyond cocktail party listening, other abilities relating to auditory scene analysis are also impacted by presbycusis (age-related hearing loss). Now in his late sixties, Trevor has noticed the characteristic decrease in the upper limit of his audible frequency range. His difficulty following conversations in the cafes could be a result of either the presbycusis or MCI. He described how his ability to localise sound has ‘definitely got worse’. This is most noticeable to him when he goes sailing, which he does through a local charity. Diagnostic tests would be required to differentiate between these issues with either peripheral hearing or central auditory processing systems; if it were to be the latter due to MCI, then hearing aids would have little effect. There appears to be a complex interrelationship between progressive hearing loss and cognitive impairment. Hypotheses here state that cognitive impairment may be compounded by presbycusis as higher-order brain functions are overloaded in compensating for the reduction in available sensory information, that presbycusis is observed in tonal audiometry results as cognitive impairment affects sound comprehension, or that lesions happen to affect both cognitive and hearing functions simultaneously due to brain topography (Costa Lopes, Magaldi, Gândara, Barros Reis and Jacob-Filho Reference Costa Lopes, Magaldi, Gândara, Barros Reis and Jacob-Filho2007).
Trevor also has issues with hyperacusis (increased sound sensitivity resulting in a lower sound pain threshold) and this is a source of emotional distress for him. He told me of how a trombonist in one of his ensembles a few years prior had, while warming up, played a note as loud as possible while he was nearby. It made him feel immediately nauseous and left him with a headache, but others present suggested that he was overreacting. Hair cell recruitment accompanies hearing loss and can easily be mistaken for hyperacusis, but this did not appear to be the case for Trevor. Recruitment occurs when the sound energy in an affected frequency range (i.e., around non-functioning hair cells) rises above a particular threshold, causing nearby hair cells to become activated (‘recruited’) by distortion of the membrane below. This causes those frequencies to suddenly become painfully loud, having been previously undetectable. Trevor did not recognise my explanation of this characteristic jump in loudness as part of his experience, however, and there were no instances of his being unable to hear me in conversation before suddenly telling me I was speaking too loud. Hyperacusis is very common among people with dementia, where it is usually accompanied by tinnitus and an aversion to certain sounds. It is generally reported by people with MCI, Alzheimer’s disease and frontotemporal dementias (Hardy et al. Reference Hardy, Marshall, Golden, Clark, Mummery and Griffiths2016). These symptoms appear to be present in Trevor’s case: his sensitivity leads to apprehension around potentially loud situations and sharp memories of particular loud sounds, such as the trombone incident or when encountering amplified drums. When I asked Trevor if he had tinnitus, he answered in the affirmative but added that he did not wish to think more about it as he knew this could make it worse, and he had ‘enough problems with [his] hearing already’.
5. PRACTICE AND DOCUMENTATION
The first section of the piece represents my participation in the Wednesday Wanderers walking group and the Paul arts group, focusing increasingly on Trevor’s and my interactions (Sound Example 1, 0′00″–13′43″). The recording of this section was a solo effort intended to document and contextualise the fieldwork process. I recorded entire walks initially so as to have a reference of events and quotations, but it later became clear as we talked about the form that the emergent meaning of the piece relied on showing how our relationship developed over time. After everyone had dispersed at the end of the walks, I would go back along the same route to record ambient sounds for laying underneath the voices. These location recordings are intended to help the listener imagine the events insitu, to help reduce the distracting impact of later editing and to better stabilise the stereo image.
The first time Trevor heard any of these earlier recordings was at the Wednesday Wanderers’s group visit to an exhibition of photographs they had taken for the Arts Council-funded project Observations in Nature (2019), organised by the Sensory Trust. I had made some short documentary sound pieces for both this group and the others that I had been visiting. The visit took place just before my last session with the Wednesday Wanderers. In what felt disastrous at the time, I failed to press the record button a second time to actually start the documentation recording and thus hurried to make field notes immediately after. I later used these notes to return to some of the same topics in later meetings with Trevor: they included the recordings’ fidelity, not being able to understand speech among other voices, his experience of doing a hearing test, how soundproof his house is and his dislike of loud, sudden sounds.
In the middle section of the piece (Sound Example 1, 13′57″–43′13″), derived mostly from conversations I had with Trevor in his front room, the montage is less chronological. Instead, this section is organised thematically: Trevor’s music career, spatial and short-term memory issues in his music practice, and sound sensitivity and its relation to environmental and attitudinal barriers. These sessions were recorded in their entirety, resulting in more editing but meaning that we could converse freely, without needing to repeat discussions. I would go through the raw recordings and drop named markers every 30 seconds or so to produce a granular labelling. Having split all the raw recordings at the marker points into separate clips, I numbered the clips chronologically and assigned colour-coded tags. These tags indicated basic topical categories: biographical information, music, memory, sensitivity, hearing loss, disability and environmental sounds. The colour coding was useful for separating hours of unstructured, recorded conversation into the preceding themes, which I was eventually able to discuss with Trevor as distinct categories, which in turn Trevor suggested we map into sections.
I initially played the first section, representing the group sessions, in near-final form to Trevor on the fourth day at his house. We had previously used his hi-fi system to listen to sounds, but on this occasion, I brought two pairs of closed-back headphones. I had mostly used the same headphones to work on the composition at this point, and having Trevor listen with these would allow me to gauge how he was hearing the piece more reliably, as the hi-fi was optimised for music playback. As we listened, Trevor chuckled at Marilyn’s (one of the volunteers, and now a good friend of Trevor’s) comment that his eyesight is ‘dreadful’. He grimaced as we listened to the part in the cafe at the end of the second walk, so I immediately turned his headphones down for the rest of the part: although I had pointed out the volume knob on the headphone amplifier, it turned out that Trevor couldn’t see it well enough to use it. He also gestured to me to pause the playback during the part recorded in the church (Sound Example 1, 6′07″–8′01″), asking what it was he had been saying there. This part is intelligible to me, although I do have to strain somewhat to decipher quieter words due to the large amount of reverberation, while Trevor cannot, although he did recognise his own voice. After we had finished listening, at the point where the final walk ends, Trevor remarked that he had found hearing one of the group members speak in one of the earlier parts poignant, as they had become noticeably less vocal over the course of time since. We then discussed the degree to which we were both able to understand some, but not all, of the speech in this part of the piece. I could hear more of it, of course, but it is not possible to determine how much this is due simply to our different degrees of hearing loss or to cognitive differences, such as auditory scene analysis abilities, which could be affected by previous ear training or MCI. This playback session, however, certainly helped to direct Trevor’s thoughts towards opportunities for a more collaborative approach to the later parts of the piece. It was at this point that I was able to ask him what he wanted people to know about what it is like to have his particular hearing and listening profile.
6. CO-COMPOSITION
6.1. Co-composed techniques
Although there had been some discussion about our collaboration on the composition throughout my one-to-one fieldwork with Trevor, there was an instance from shortly after this first playback session that was especially productive. Sitting outside Porthcurno Beach’s cafe, we discussed the potential of phonography for representing experiences such as his hyperacusis. We were both keen for there to be moments where the auraltypical listener may get some sense of what this might be like. For me, this related to hearing the hearing of others, and for us both, it was a chance to explore and communicate how his hearing impairments impact on his daily life, with the possibility of this leading to discussions in the groups of how different hearing abilities might be accommodated. We would try simulating the experience by increasing the level of a recording of a typically problematic sound to one bordering on uncomfortable for a non-hyperacusic person (in this case, me), although this would of course need to be carefully calibrated so as not to be dangerously loud. Trevor suggested a particularly piercing scream that small children sometimes do, as this sound, unlike the trombone, is one that is already commonly found to be unpleasant, and so would not need to be as loud to be effective. Back at the house, Trevor sketched out with lines and annotations on a slip of paper how the sound seems to continue in his ears after the child has stopped screaming, and returns at varying levels over the following minutes and hours as he remembers it and the sensation of sound pain seems to continue. I hurriedly produced a basic representation of this continuation effect using a scream taken from our recordings at Battery Rocks earlier and Jean-Francois Charles’s FFT-based spectral accumulation patch in Max (Charles Reference Charles2008), but this did not match Trevor’s auralisation of the sound very closely: the scream was not piercing enough, and the patch’s output was too stable. Another scream sample downloaded from FreesoundFootnote 6 was suitably piercing, and I used Trevor’s sketch to arrange copies of the clip processed with convolution reverb. ‘That looks about right’, Trevor noted while looking at the arranged clips. This particular simulation required more work to produce an uncomfortable sensation for me than the cafe recordings did in the earlier sections; those were left unmodified as they were already effective.
Intelligibility in cocktail party scenarios was another hearing difference that we determined to be characteristically challenging for Trevor. This is also an issue I have a degree of experience with, as I have some noise-induced hearing loss and a bilateral tinnitus that sounds like a (roughly) 13kHz sine tone. I found that I had to strain to hear what people were saying in the cafe recordings and this is still harder with loudspeakers; unlike Trevor, though, I do not need to actually reduce their levels to avoid sound pain. The sense of intelligibility ebbing and flowing was thus an effect we both wished to represent. It evokes the way in which cognitive capacity for speech comprehension and auditory scene analysis seems to be variable hour by hour for people with dementia, which is likely to be related to comorbid presbycusis (there is more missing spectral information to ‘fill in’). We did this by overlapping and gradually fading speech clips, a technique that can also convey a sense of conversational duration passing, which Trevor recalled from radio dramas. There are several edit points where this contrapuntal arrangement technique was useful (e.g., Sound Example 1, 14′34″–14′59″, 60′00″–60′21″). Again echoing radio dramas, we also used montage to superimpose extra-diegetic sounds; for example, in the representations of Trevor’s favourite sounds at the end of the composition (Sound Example 1, 60′53″–62′23″). There are a number of sounds that are not present despite intentions, too: we both wished to include some of the recorder music Trevor had written, but he could not find the sheet music despite looking repeatedly.
6.2. Practical challenges
The use of phonography and sound design in a research context such as this presents practical challenges. There are numerous areas of practical knowledge necessary for its creation, from sound recording and cataloguing to audio editing and signal processing. Trevor’s input in the multitude of minor decisions involved in these processes was relatively limited beyond suggesting specific sounds and asking whether they had recorded clearly (‘Will it work if you get closer?’, ‘Why don’t we get some of the church bells? I never hear the church bells!’), reliant more on verbal descriptions. Giving feedback on versions of a composition being realised elsewhere places the research participant in a reactive position, only able to intervene at certain points when it would, of course, be preferable for them to have the ability to work on the composition directly. That said, as documentation of this composition process forms a substantive part of the composition’s duration, we hope that listeners will gain some awareness of how the roles played out in our interactions.
The amount of editing used in this piece is quite extensive compared with other sonic ethnography works (e.g., The Last Boat Builders of Belfast (Milligan Reference Milligan2014) or Heard Laboratories (Karel Reference Karel2010)), due to the constant recording and how much of this contains (intelligible) speech, clothes rustling and footstep-induced microphonic bumps. While the use of editing raises ethical issues of veracity and positionality as described by Clifford, the editing here is rooted in discussions about representational intentions and techniques, and while effort has been made to make the edits less distracting, they are nonetheless clearly audible. As a researcher, I obviously have an agenda in how Trevor’s hearing is represented: I want to make its difference to my own hearing (or to more auraltypical hearing generally) clear to an academic audience, and to highlight some of the implications this has for Trevor’s wellbeing. I speak less in the piece than Trevor and this is a result of my having done the editing; I am (at least personally) more interested in what he has to say. The exploration of Trevor’s thoughts on his own hearing is the motivation for this work, but we did try to ensure that a sense of the piece being grounded in social interaction remains.
Trevor’s long-standing enjoyment of radio dramas meant that he already had a good degree of familiarity with phonographic montage and representational techniques, such as the layering of ambient sounds to construct a locality where speech events are heard to happen, or the superimposition of ‘non-diegetic’ sounds to perform the narrative function of referencing other times and places. As Daniel Makagon and Mark Neumann state in relation to non-fiction and documentary forms of ethnographic sound practice, the ‘editing and mixing of gathered sounds, interview voices, and ambience all come to bear on the composition of acoustic environments … [Such forms] provide their own version of a sonic theatre by composing an arena of sound that creates a sense (and illusion) of presence and immediacy’ (Makagon and Neumann Reference Makagon and Neumann2009: xi). Trevor’s ready understanding of these possibilities for working in sound was very useful, as it meant that we could discuss the montage of the various different types of sound recordings without my needing to make examples to return to later, when the moment would have been lost. It also reduced the potential that such demonstrations would have had to bias Trevor’s interpretation of the statements being made with sound design in the piece. Discussions are situated alongside environmental and otherwise incidental sounds to provide a document of collective hearing.
7. CONCLUDING REMARKS: CO-COMPOSITION OF TREVURR
As mentioned earlier, the categorisation of a composition such as Trevurr as either participatory or collaborative is made difficult by its being part of a research project. Kester’s distinction, based on the power dynamics involved in creative practice with others and the question of who benefits from it, was not intended to refer to scenarios of practice research. My intention was for the creative process here to be as collaborative as possible but due to its only having taken place because of the research context, it is more accurate to describe it as having been dialogic. This has previously been a relatively effective way for researchers to reduce the impact of doxic bias in interpretive accounts of fieldwork encounters and to try to address inequalities inevitably engendered by ethnographic research. This is not to say that the whole composition is dialogic, though. It only becomes so as it progresses beyond the earlier sections, where Trevor’s thoughts on his hearing start to become apparent as we discuss the intersections between his hearing and both the physical and the social environments he is a part of. This forms a kind of sonically oriented and conveyed testimony or illness story. Further, some later episodes in the piece are also not dialogic due to necessity: as I mentioned earlier, Trevor cannot listen to the cafe or scream episodes at full volume as these are supposed to verge on thresholds of discomfort for non-hyperacusic listeners.
‘It takes time’, as I recall a collaborative design researcher once memorably putting it, ‘to get warm to people’, to find the right way to convey the aims of a research project to a particular participant (Pitsillides Reference Pitsillides2017). This research is predicated on interlocutors being as free as possible to choose how to describe their sound experiences both in words and through phonography. As Kester argues, social interactions can be more significant than an autonomous art object resulting from collaborative practice (2011: 9). My reliance on conversation over formal interview methods is intended, in this spirit, to facilitate interlocutors in producing descriptions or narratives. Conversation is one of the most equal, or least technology-dependent, parts of this research process. The mix of methods described here – of participating in group walks, doing listening activities or ‘dialogic auditing’ and having related conversations – has differed somewhat from the ideal of a co-aesthetic or co-compositional approach envisaged when I started out. While I had intended to create situations for making cogent, co-created works able to be autonomously meaningful, and supporting these with separate versions of the pieces documenting their development processes, Trevor and I inferred that this kind of decontextualised output might make little sense outside of the wider context of social interaction from which it came. The work of making the creative process as dialogic as possible has been more insightful as an overall narrative than the directly co-aesthetic episodes. The story of the piece – the relations it formed by necessity and by the agendas negotiated – is indeed a more evocative way to hear Trevor’s hearing.
Acknowledgements
I am very grateful to Trevor Synge-Perrin for his time, thoughtfulness and enthusiasm in contributing to this project; for his permission to reproduce sound examples from our composition here; for his comments on an earlier version of this article; and for persevering with emails to provide those comments due to the COVID-19 pandemic. I would also like to thank the two anonymous reviewers for their insightful feedback, and CHASE-AHRC/Goldsmiths for funding this research.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit https://doi.org/10.1017/S1355771821000273
