“A Cognitive Listening”

Abstract

This paper proposes a theory of text on-screen as “unvoiceover.” It addresses both the case for captioning as social good and the affordances (aesthetic, affective) of writing in or over the moving image. Advancing an argument informed by perspectives from d/Deaf Studies, Critical Disability Studies and Digital Interface Studies, and applying modes of analysis from literary criticism alongside those proper to the study of moving image and sound, it examines the idiosyncrasies of text-in-motion as non-sonorous, fugitive counterpart to the traditional, troublesome “voiceover.” To develop a poetics of the unsounding voice on-screen, the paper focalizes its argument through multimedia artist Liza Sylvestre’s Captioned series: a body of moving image work that is itself, paradoxically, uncaptioned. Framing Sylvestre’s lyrical “unvoiceover” as a reimagining of the lost roles of film explainers and literary intertitles, I argue that the artist’s takeover of the caption track intervenes critically in contemporary debates about the ethics of audio-visual translation, situated description and access as public ethos rather than private concern. Posing the artist’s personal-and-political writing as suggestive of a lower-case analogue to Deaf Gain, I show how Sylvestre’s “unvoiceover” educates its “receivers” in the purpose and functioning of captions. By reading Sylvestre’s writing on-screen more closely than its fugitive form seems to invite, I show how the unvoiceover encultures its own demands on its readers and elicits its own habits of reading. By scrutinizing how Sylvestre’s series makes the case for captioning, this paper makes the case for a new appraisal of the aesthetic, affective and political affordances of the unvoiceover as writing on the run.

Keywords:

• captioning
• voiceover
• access
• description
• d/Deafness
• Liza Sylvestre

Daffodils by Nataliia Kutykhina.

In the early 1980s, the Caption Center at the American public broadcasting group, WGBH, collaborated with the US National Association for the Deaf on a range of resources designed to “build captioning awareness.” They sought to address a viewership that had proven disappointingly slow to respond to the advent, in 1980, of closed captioning (Downey 209). Whilst some of these (manuals, decision trees) addressed the adults who were envisioned either as potential purchasers of the station’s closed caption decoder sets, or as the teachers who might make use of captioned materials in their classrooms, others were directed at children. The manuals for adults were accompanied by a videotape and workbook for kids, produced in response to research indicating that d/Deaf children confronted with the unfamiliar apparition of text on television were “often confused by the words on the screen and sometimes fail to connect them with spoken dialogue” (US Department of Health 30). Renowned Deaf actor Bernard Bragg was enlisted to lead a remote, recorded lesson for prospective caption-users. To deliver this training, Bragg assumed the character of Captain Marvelous: a superhero alter-ego dressed in spangled trousers and hat, dedicated to spreading the captioning Good News. The task entrusted to Captain Marvelous was that of explaining captioning to a young d/Deaf American TV public of future media (and advertising) consumers. Captain Marvelous’ mission was to demonstrate how captioning worked and why captions were worth reading.¹

Some four decades later, the Captioned (2017–) series of video works by artist Liza Sylvestre fulfils a cognate, though self-directed pedagogical brief – addressing itself not to TV viewers but to art publics. And just as Captain Marvelous pursued a patient, two-phase approach, so Sylvestre speaks (through text on-screen) to a readership of art viewers as well as to the institutions they frequent. This article sets out to read Sylvestre’s writing on-screen as an “unvoiceover”: a form of time-based writing that variously replaces the role and refracts the capacities of its more familiar sonorous counterpart. In what follows, I will frame Sylvestre’s Captioned series as a body of work whose “unvoiceover” educates its “receivers” in the purpose and functioning of captions. I will propose that it does so in content by articulating the effects of the near-ubiquitous absence of captioning, and in form by mimicking the appearance and rhythm of an open caption track overlaid on an extant moving image. The Captioned series prompts engagement with episodes from the histories of early cinema and of captioning in TV. It explores the pragmatics of lip-reading and cochlear implant as mutually co-imbricated and differently fraught everyday technologies of interpretation. By reading the unvoiceover more closely than its form ostensibly encourages, I will develop an account of how Sylvestre’s work challenges the presumed verbocentricity of contemporary film-viewers, hierarchies of sonic significance, and the status and sensory apperception of text on-screen – the under-examined idiosyncrasies of writing on the run. Borrowing and reorienting a construction from the subtitle of Georgina Kleege’s 2018 book, what this article proposes is to develop an account of what captioning brings to moving image art, via an analysis of what Liza Sylvestre’s unvoiceover brings to the cultural understanding of captioning.²

I come to captioning and the idea of unvoiceover via “voiceworks”: audio-visual works that derive a major part of their motive force from their deployment of voice.³ Overlapping in part with the categories of artists’ films or videos, voiceworks are artworks that are primarily animated by voiced writing. But while voice has been centred in film sound since its introduction, the application of the term within artists’ moving image is intended to solicit a voco-tropic rebalancing of attention. It announces an orientation towards spoken language as medium and material: material which, owing to its apparent resemblance to quotidian speech, is often not apprehended as written, as potentially poetic and, in various sorts of ways, performative.

Many major voiceworks bring sonorous and non-sonorous voices into relation, into compellingly cross-modal conversation. In Promised Lands, Emma Wolukau-Wanambwa deploys text on-screen as emphatic, insistent rebuttal to the sonorous revoicing of colonial propaganda.⁴ In Tony Cokes’ The Book of Love, the artist’s mother Dorothy R. Cokes recounts (resoundingly) her experiences of work – paid and unpaid – and the artist’s unvoiceover (silently) bears witness to Cokes’ tortured contemplation of his own artistic labour as “work.” Laure Prouvost’s early videos make copious use of text on-screen and the conjunction of audible and legible language to open a space for thinking about slippages between words as sound, and as sense, particularly as they pass across linguistic borders.⁵ As my research on voiceworks developed, I became increasingly concerned, first with the interplay between audible voices and legible ones that manifested as text on-screen, and then with captioning.⁶

Via encounters with work by d/Deaf and disabled artists using access, politically, as primary material – artists such as Liza Sylvestre, Carolyn Lazard, Christine Sun Kim and Jordan Lord – making work “from / the conditions of debility or difference / not translated for debility or difference” – I became belatedly conscious of the near-absolute absence of captioning in artists’ moving image (Lazard, Recipe). While curating the 2020 exhibition Many Voices, All of them Loved, I came to see that the voiceworks that so intrigued me were designed or at least designated for the hearing. Consequently, and in large part through engagement with Liza Sylvestre’s Captioned series, I came to recognize that the publicly funded installation and screening of captionless moving image depends upon what Tanya Titchkosky terms “the common-sense understanding of disability as excludable” (Question of Access 6).⁷

Today, while consumer resistance to text on-screen is, anecdotally at least, much less vehement than was once supposed, and Gen Z’s liking for open captions perturbs their progenitors,⁸ many cinemas, terrestrial broadcasters and moving image artists remain averse to sullying the sanctity of the moving image with the means to make its accompanying audio intelligible to those who don’t apprehend it aurally (whether fully or partially). The much pronounced-upon pivot online performed by art institutions in the early phases of this ongoing pandemic induced some museums and galleries to reappraise certain of their access responsibilities. Unfortunately, these are too often understood to extend only to online presentations, with on-site installations still deemed immune to any such expectation for captioning or audio description. And regrettably, as San Alland and others have pointed out, the realizations and reappraisals occasioned by pandemic closures turned out, very often, to be reversed. Instead of a great reset, the artworld has undergone a great return to ableist form.⁹ At this juncture, what’s needed is a full and sustained reckoning, a media justice engagement with access in art as a project with benefits that are ethical and social as well as aesthetic. Before that can happen, we need to recognize how certain kinds of media access are understood as givens, and others are conceived as additional and optional liable to be granted only on petition, or withheld. Liza Sylvestre’s serially unspooling unvoiceover is an extraordinary device through which to consider these ideas. And so, in writing this essay, I wanted to think both about the case for captioning as social good and about the affordances (aesthetic, affective) of writing in or over moving image.

The attention I will attempt to bring to the text that appears on-screen and instream, is solicited in part by Craig Dworkin’s recent remarks on “the amnesia surrounding the materiality of language more generally,” and informed by Lori Emerson’s explicit focus on the interface by which writing is encountered (Dworkin 24; Emerson). Writing on new ecologies of reading instantiated by digital media, Robert Clowes calls (via Lambros Malafouris’ Material Engagement Theory) for awareness of “the complex dynamics by which material culture does not just disturb but makes minds” to be applied to the analysis of this ecology “on its own terms” (706). Clowes’ concept of screen reading is one exemplified by the skimming, annotating reader of multiple pdfs – a model closer, in one sense, to a “traditional” reader of printed text than to a reader who engages with fugitive text in motion. Nonetheless, I want to respond to his prompt for a more materially conscious reading of equally (if differently) unorthodox new “events of reading,” by attending throughout to the particular affordances of text-in-motion and weighing these in relation to the capabilities and effects of the sonorous voiceover (Cayley 213). Curiously, even as technotext theorists stressed continuities between traditional page-bound materialities and those things they labelled cybertexts, hypertexts, e-lit, etc., they jumped over the particular (if less transfiguring) affordances of writing on-screen that is not subject to the participating reader’s intervention. Their leap from page to participation leaves a strange void in the record of responses to new modes of reading.¹⁰ Approaching from an oblique angle, this essay will try, retrospectively, to supply this gap. Just as Doug Downs notes that “[long-form] print codices enculture a particular attention and habits for meaning making,” so I want to show how writing that pops on and off screen in the guise of open captions encultures its own demands on its readers, and elicits its own habits of reading (205).

In recent years, artists including Christine Sun Kim, Alison O’Daniel, Seo Hye Lee, Carolyn Lazard and Jordan Lord have directed a new and welcome attention to artists’ innovations in captioning, some presenting institutions have begun to commit to captioning (percentages of) new work and networks dedicated to expanding the availability and creativity of captions in art contexts have formed.¹¹ In an essay that first appeared in the art magazine Mousse and later in extended format in Future Anterior, art critic and curator Emily Watlington made a compelling call for the retrospective captioning of historic works of video art and projects to undertake such retrospective access-making are underway (“Critical Creative Corrective”; “Radical Accessibility”). Meanwhile, recent accelerations in AI speech-to-text transcription technologies, increased provision of captioning on streaming sites, media coverage of the new appetite for captions among younger hearing media-audiences and public campaigns such as those led by actor Marlee Matlin and Rikki Poynter have brought captioning to public consciousness in a new way.¹² This essay responds to Emerson’s proposition, via McLuhan, that

The point at which a technology saturates a culture is the point at which writers and artists, whose craft is utterly informed by a sensitivity to their tools, begin to break apart that same technology to once again draw attention to the way in which it offers certain limits and possibilities to thought and expression. (50)

What follows addresses captioning as a technology in just such a generative state; it responds to the extraordinary expansion in how captioning is conceived and practised by artists by close reading a series that, paradoxically, is itself entirely without captions.

In a media context within which the terms “captioning” and “subtitling” are differently understood in different anglophone regions, and an Internet environment that, as Elizabeth Ellcessor identifies, “[embraces] a neoliberal drive for profit that ignores the public interest at stake in media access,” media access terminology gets murky (“Captions On, Off” 341). Definitional ambiguities are compounded when tech platforms mislabel speech-to-text products intended to increase SEO (search engine optimization) capture, deploying what Mara Mills calls an “assistive pretext” (“Deaf Jam” 39).¹³ In such instances, initiatives introduced to increase advertising profit margins are marketed under a premise that is also a false access promise (Ellcessor, “Captions On, Off” 336). Because accurate communication of provision parameters is a prerequisite for access, and for the avoidance of any confusion, I will make three prefatory clarifications: first, by captioning, I mean the entextualization of both (a) verbalized content, dialogue or narration and (b) the rich description of all other sound – incidental, atmospheric, musical, phatic, etc. – that makes up the audio side of the audio-visual medium.¹⁴ In the context of media art, such captions aim to produce an equivalence of experience for d/Deaf and hearing receivers alike. Conversely subtitling, whether inter- or intra-lingual, would encompass only portion (a). Second, when I refer to d/Deafness, I’m intending for that term to carry with both upper and lower-case initial letters. I use the d/D that encompasses signing communities of Deaf culture, deaf, hard-of-hearing and late-deafened individuals: a slash that serves to imply a coalition of shared concerns, while acknowledging very different experiences.¹⁵ Finally, here, as elsewhere, I will refer to a “receiver” more often than a “viewer” (who must see) or “audience” (who must hear). Borrowing the noun from the title of a 2019 film by the Irish artist Jenny Brady, I do this in an attempt to keep unsettling my own and others’ expectations about the sensory channels by which particular media are to be encountered or “received.”¹⁶

In her ongoing Captioned series, artist Liza Sylvestre intervenes upon familiar, historically uncaptioned filmic artefacts. But instead of supplying the caption tracks without which these works were made and without which they are still being screened, the artist writes into that space instead. Her writing work inserts itself in a gap that ought not to be – in part, to make the case for what ought already to be there. In the Captioned series, existing films are inscribed by the artist with something equivalent (in length, purpose and complexity) to an essay, but received as a faux real-time drip-feed of text that is read from the screen.

Sylvestre’s unvoiceover is written in the first person, in a lyric register. Its text-stream manifests on-screen at a speed and cadence familiar from the captioning of live sonorous speech. And just as the artist-author’s first-person voiceover typically, Cavarero-style, aggregates sonic particularity with first-person subjectivity, so the first-person writing on-screen presumes to borrow this, too.¹⁷ Like Silverman’s classic (Hollywood male) voiceover, “it ‘hovers’ above the image track, in an invisible spatial register, it occupies the same temporal register as the other characters, and often comments upon events as they occur” (48). Ruth Finnegan writes that “sound dies leaving nothing behind. It enables secrecy, flexibility and informal interactions that would be different if articulated into fixed visible form” (86). But what happens when the written form – as in highly ephemeral (1–10 second timeout) instant messaging – endures only an instant before erasing its own record? Sylvestre’s unvoiceover is writing that is “ephemeral, transient, incorporeal, ethereal” as voice in Dolar’s account (71). Like the voiceover so influentially characterized by Doane, it “speaks without mediation to the audience, by-passing the ‘characters’ and establishing a complicity between itself and the spectator” (42). Yet where the acousmatic voiceover builds mystique and authority by making itself unlocalizable and incorporeal, Sylvestre’s writing on-screen gives every impression of being the uncontested issue of the biographical being of the artist that is its author. Taking over functions more typically assigned to a sonorous voiceover, but un-encumbered with its acousmatic counterpart’s heavy, toxically coded legacy as a disembodied, authoritative voice, this text behaves as a non-sounding unvoiceover. Thus positioned, thus preserved from odious associations with the Voice of God, the unvoiceover is equipped to undertake particular kinds of political work in art. Sylvestre’s critical unvoiceover mobilizes the histories of voiceover and live commentary in film through the semblance of intra-lingual subtitling to make a media justice case for captioning.¹⁸

Each work in Sylvestre’s series intervenes upon a found source by writing over it. In Captioned: Twentieth Century (2018), the underlying work is Howard Hawks’ 1934 screwball comedy, Twentieth Century. As the title implies, Channel Surfing (2017) over-writes a sequence of short clips: Hollywood films and animated cartoons from different eras, the text on-screen suturing together what is otherwise as disparate as what could be summoned by the remote control in any hotel room. Third Space is a dual-screen installation, in which the artist’s unvoiceover appears over Sun Ra’s Space is the Place (1974) and Kubrick’s 2001: A Space Odyssey (1968). Claiming a space for itself within the image-frame, Sylvestre’s writing appears in a false caption-space situated below the frame of the former but above that of the latter, further complicating and physicalizing the reader-viewer’s ping-ponging engagement as they scan diagonally across projection spans.

Under Sylvestre’s intervention, moving image artefacts are made carriers for newly authored site-specific texts. This overlay essay-in-motion variously processes, reflects upon, reacts to, riffs around and ignores the film/video it frames. It evinces – and consequently effects in turn – a flickering, fluctuating engagement with the source-text that is its host/subject. As though proliferant marginal annotations had gone large and taken over the text, Sylvestre’s writing pulls focus from what it treats; it obscures even while it spotlights that over which it trespasses. Her unvoiceovers make the audio-visual artefacts they overwrite less legible, while determinedly reading what they simultaneously occlude and process. Because the unvoiceover is so compelling, and so persistently speaks across what it occupies, paratext becomes text, figure is made ground: a ground that is substantively, productively excavated by the caption-like text that, in transmission, transiently overwrites it. Text (film) and paratext (unvoiceover) are apprehended in one overextended instant. Each Captioned work’s imagetrack conscripts a reading that takes account of two kinds of legible material at once – but without benefitting from the mutually reinforcing doubling that (for the hearing) characterizes captioning. The film’s audio component is deprivileged as it becomes harder to focus on what, in the on-screen dialogue, is being “said.” Film footage becomes noise.¹⁹

For the hearing reader-viewer, the footage over which the textual unvoiceover “speaks” is already audio-visually replete. In prompting receivers to engage self-consciously with the workings (and failings) of their own perceptual processing capacities, it demands of the hearing receiver decisions to be made moment to moment as to which input to privilege. Co-located sounding image and text tracks compete on-screen for attention. When the text stream is especially busy, the texture of the work becomes tryingly thick – especially when, as in Third Space, dual channel installation (and consequently four-way text/image interactions) slice attention still more finely. Instantiating lo-fi cross-sensory overwhelm, they conscript a complex, shifting, compromising distribution of attention.

Once committed to following the unvoiceover, hearing viewers might be tempted to try to watch without sound. Or perhaps, if partiality perturbs, to watch twice, with the balance of concentration differently assigned each time. But in encounters with the long-form moving image screened in galleries, readerly autonomy is rarely accommodated. Receivers are granted neither remote control nor access to the knobs and buttons that could pause or rewind. And in Sylvestre’s Third Space, the two-channel installation means that both soundtracks (to Sun Ra’s Space is the Place and Kubrick’s 2001: A Space Odyssey) are audible at once, with, besides, two separate unvoiceover tracks to read. As a hearing, sighted receiver, I am practised in transcribing my own text versions of voicework audio. In attempting to do the usual here, I find myself constantly running up against the too-low limits of capacities formed out of a lifetime’s preferentially centred, hearing-and-seeing, media consumption. Temporarily thrown out of the habitual ease of inhabiting a media environment that anticipates a “normate” listener-viewer, in watching, I pause, rewind, double back, check that I have correctly understood.²⁰ And each time I bend to my notebook to write up a line of text from the screen, I must surrender to the probability of missing the next. The transcript comes out full of holes: my note-making a gappy record of partial capture. Even as the audiotrack is (for this author) always audible, I cannot afford to look away from the screen that tells its own story. The advantage ordinarily conferred upon the hearing viewer by the access abundance of audio-visual redundancy is denied and a moment’s equivalence is instantiated. For d/Deaf and hearing receivers alike, total capture is elusive.

Since video’s inception, artists have used it both to point up problems and to exercise the utopian possibilities of how screen media (historically television) meet their viewers.²¹ In an extension and further specialization of this form of medium-specific meta-critique, artists today are deploying access and “alt.” modalities as the means by which to point up the willed inaccessibility of media. Sylvestre’s deprivileging of the reader who simultaneously hears effects an instructive wrongfooting that recalls how, as Emily Watlington writes, Constantina Zavitsanos and Amalle Dublon’s April 4, 1980 (2018) “privileges deaf/Deaf viewers by adjusting the speed of the audio to privilege reading speed, rather than adjusting the speed to match the on-screen speakers (there is no image to which the audio needs to be synchronized)” (“Radical Accessibility” 117). In engaging with the Captioned series as a hearing viewer, properly concentrating on the artwork requires relinquishing (though to a fluctuating degree) any expectation of following the underlying, overwritten film. To have seen/received Captioned: Twentieth Century is not to have seen Twentieth Century and an additional artwork. It is to have been made party to Sylvestre’s intervention across (and in part against) Twentieth Century. Of course, as a stream of text that manifests only on-screen and lacks an audio equivalent, the unvoiceover will not be available to blind and low-vision receivers, for whom the Captioned film’s original audio will come through uninterrupted by the artist. Elsewhere in her work – in performance, in pedagogy, in sculptural practice – the artist is increasingly concerned with the creative possibilities of audio description, but Sylvestre’s intention here is to commit a focused, single-issue intervention.²²

Experienced as ephemeral, even though ineradicable from the digital work, the unvoiceover has curious implications for relations between writing and speaking. As legible text that refuses to sit still, that – though spatialized, resists being “language in persistent visuality” – the unvoiceover claims capacities from both sides of that gnarly binary (Cayley 216). No longer “thing-like, immobilized in visual space,” it divests itself of the responsibility to allow for the “backlooping” that Ong grants as the natural entail of writing which, in its natural habitat (in space rather than time), “establishes in the text a ‘line’ of continuity outside the mind” (39–40). Comparing the characteristics of print’s “typographic visuality [that] is static, fixed, although it may be spatialized in a number of ways so as to influence or inflect reading practices and strategies” with the “wide array of dynamic potentialities” introduced once “[t]ext in digital media can move and change,” Cayley invokes a degree of movement and change that exceed those that can be claimed of our unvoiceover (213). Nonetheless, it is not excessive to lay claim to his assertion that “The digitization of typography has given us new expressive structures for temporalities that have the potential to influence and change the fundamental events of language: our events of reading” (213). Though there is much to contest in Cayley’s insistence on the primary aurality of language, I am keen to pursue the implications of his proposition that “Time is arguably the most important, necessary, and most neglected property of textuality” – because the most significant characteristic of the unvoiceover is its transposition of text from materialization in space to instantiation in time (81).

Once in stream rather than spatialized, text acquires what Bonnet terms the “irreducible fugacity” of sound (7). For Bonnet it is the impermanence of the sonic imprint that makes the ear “the organ of the ‘unproven’; ‘the unverifiable’” and, as a consequence, susceptible to domination (211). Does this mean that even the fleeting manifestation of text signals submission to scrutiny, a laying opening of the books? In a perturbing essay on “abusive” subtitling, Abé Mark Nornes frames the subtitle’s transience in freighted terms: writing of how as “the reader cannot stop and dwell on an interesting line; as the reader scans the text, the machine instantly obliterates it” (20). Under this account, the subtitle is not just sliding by but actively (even slyly) expunging itself from the record. Farocki maintained that printed text on film necessarily draws attention, prompting greater recognition, and more overt objection whereas “A spoken commentary often goes unnoticed, much like film music” (Farocki and Dziewlor 222). This implies that the unvoiceover is inherently more emphatic, or, casting back to the critique of documentary voiceover: less sneaky, less predisposed to insinuate itself – in part because in being read (post Ong’s still-oral early age of print), it does not presume to penetrate what Bonnet calls “the intimate confines of the auditor” (142).²³ Nor is the unvoiceover, like Nancy’s always “arrive”-ing, “attack”-ing sound “tendentially methexic (that is, having to do with participation, sharing, or contagion)” (10, 14–15, 2).²⁴

Commenting on the action on-screen, Sylvestre’s unvoiceover artifices an impression of being generated in real time. The co-viewing dispositive produced is one of watching along with a voice made visible. Although evidently also the product of re-watchings, spottings and synchings, the tense it inhabits is a sort of continuous present, albeit one that permits passages of retrospection.²⁵ In its format, the text mimics standard caption conventions. More importantly, it also complies with segmentation standards. This is to say that text appears in semantically coherent, screen-width-spanning units of one or two lines: prose-style, per Dworkin’s identification of the boustrophedonic imagination that produces “the illusion that the format is a continuous and uninterrupted procession of text rather than an assemblage of discrete and discontinuous lines” (44). Lines are broken wherever feels most enunciatively natural, in line with where a speaker might pause for breath – further underlining the text’s through-line to vocality. Chatty and responsive, Sylvestre’s continuous present is the tense of caption-writers and audio-describers.²⁶ The unvoiceover appropriates the habitual-occupational mode of the access worker whose descriptive “translation overlays” (whether live, pre-scripted or pre-recorded) all manifest a similar sort of real-time effect.²⁷ In Audio Description, this preference for the continuous present makes for a curiously propulsive concatenations of gerunds. In the Captioned series, it creates a paradoxically self-abnegating pretence of the describer’s co-presence. Present tenseness connotes presence: an effect conferred, again, by the unvoiceover’s borrowing of fleetingness, its “extremely elusive” fleetingness – as from that of a speaking voice: “just the vibrations of air which vanish as soon as they are produced, a pure passing, not something that could be fixed or something that one could hold on to” (Dolar 36).

Soundless but sustained, like a loquacious flatmate comfortably installed on the same couch, the unvoiceover delivers its moment-to-moment musings casually, unselfconsciously. The effect is amplified by the unvoiceover’s attentiveness to the films’ representations of race, and to what Davidson identifies as “the gender dynamics of the film as rendered in visual terms” (172), as the reader-viewer is interpellated by text that seems to presume/anticipate shared attitudes to co-experienced phenomena. There’s an intimacy to this style of continuous, ambient address, an effect amplified by the privacy effects that Ong identifies in the shift from “communal oral world” into print culture (128–29). In the two-channel installation Third Space, it remarks “I had forgotten we are on a spaceship”; in Captioned: Twentieth Century, mugs, “Seriously?”; and, when Kubrick’s camera cleaves in for a close-up of the Pan-Am anti-gravity-defying grip shoes, observes sardonically: “I’m obviously supposed to be impressed by those shoes” (Figure 1).²⁸

Fig. 1. Liza Sylvestre, still (detail) from Third Space, 2019, looped dual channel installation, courtesy of the artist.

When, in Captioned: Twentieth Century, the disguised John Barrymore character is unveiled, there’s a quasi-phatic (“Oh”) and when he’s subsequently found to have been housed (unknowingly) in the adjoining room to that of his estranged former lover, a surprised “Wow! The next room over??” Low-stakes in themselves, incidental exclamations on-screen develop the character of the unvoiceover as companionate co-viewer texting running commentary on the show you’re both watching. They also build a basis for more meaningful engagement (even soliciting of solidarity) later on.

Occasional minute typos – textual analogues to slips of the chatting tongue (“observes them throughly [sic]” in Channel Surfing, “smothly [sic]” in Captioned: Twentieth Century) – increase the impression of spontaneity. Errors typically imply the production of writing (as in the typing of an instant message) in the moment, on the fly, as though (as for speech) without recourse to editing. Of course, the notion of this kind of cosmetic editing – for correctness more so than for communication – summons the many ways in which language usage is policed, and the particularly sharp experience of this by d/Deaf and hearing speakers and writers whose pronunciation, spelling or speaking do not align with prescribed (and vociferously protected) norms. These errors might also put readers in mind of those that result when, as Hickman notes, real-time stenographers are denied rest. As a result, “The captions begin to lag in respect to the spoken content, parts of the sentences are dropped from real-time transcription, word conflicts appear on screen for the readers” (Hickman, “Willful Dictionaries” 334). In Distressing Language, Davidson writes of the “expressive force of ‘error’” and of its “generative potential,” especially where it relates to “individuals who, for varying reasons, live in a state of linguistic precarity” (164). Sylvestre’s non-correction of spelling “mistakes” functions politically as a refusal. Her typos refuse to surrender to the retroactive norm(ate)alization of copyediting. They make the less-visible figure of the lip-reading, CI-using, spoken-language-user legible in gallery space.²⁹

Seeming to manifest in real time, the unvoiceover stages textual ephemerality: vanishing almost in the moment that it appears (and without even allowing for recourse equivalent to a phone’s archive, a message history). Typos in the text stream resonate in a particular way when Sylvestre’s videos are shown on TV monitors rather than via projection. When on-screen rather than on a wall, they memorialize television’s atavistic associations with liveness – associations that, as Rachel Federman observes, endured long beyond the era of live broadcast as default. Even once the majority of TV broadcasts were pre-recorded, captioning was most commonly seen by the hearing public in news bulletins. When captioners respond live to presenters reporting live on events as they unfold, liveness becomes manifold. Besides signalling immediacy and the exigencies of real-time writing, these minute errors also serve to remind the reader-viewer that, as Hickman so neatly puts it: “the translation of speech-to-text only remains invisible when the human labor remains hidden, and the human labor only remains hidden when it is executed without error” (Access Workers 51). In the age of automatic captions, errors point up the relationality of access-making, they remind us that captioning is a social practice.

To be a d/Deaf artist writing back onto audio-visual media is to summon Lennard J. Davis’s account of how the advent of the novel and print technology “permitted the deaf, for the first time, to be part of the collective narrative tradition, thus reversing the overt interpretation of deafness” (108). In Enforcing Normalcy, Davis poses a compelling refutation of the ways in which writing has been, and continues to be, “unproblematically thought of as a process that involves hearing and vocalizing” (100). This ubiquitous tendency produces predictably devastating effects, as audist cultures presume to assume neat lines between access to acoustic sound and access to language and knowledge-making. Sylvestre’s first language is spoken English. As she writes in Captioned: Twentieth Century: “I am here because I have a romantic disability / I am a deaf white woman / You are watching this movie because / I have a certain amount of privilege.” Hers is not the upper-case Deaf culture that tends to be most grievously misrepresented as outside of language such that, as Davis writes, “blindness has been adopted by Western culture as a metaphor for insight, while deafness has been a signifier for the absence of language” (106). That said, there is yet a certain critical calibration of force to this artist’s specifically written interventions, and to the non-eradication therefrom of what would more typically be cast out as error, typo or mistake. The unvoiceover comes into new relief when considered in the ambit of Davis’s extraordinary accounts of how the availability of printed English in the eighteenth century enabled the imagination and acceptance of other forms of languaging. More directly, and maybe less speculatively, it also grounds the appearance of the work in the quotidian aesthetic reality of remote communication with d/Deaf friends and colleagues – via the screen-text-based interfaces of everyday communicative technologies.

The unvoiceover also opens a channel into the series’ recurring preoccupation with unsettling public/private dichotomies, via its description of a more particular (though not unique) experience of the “private sound” that is generated by Sylvestre’s cochlear implant: a sound made audible to no-one but the wearer (Figure 2):

My neighbor is back making more noise. / And my cochlear implant battery just beeped. / This means I’ll need to change the battery soon. / I think about this beeping sometimes. / It is a completely private sound. / Generated by the processor that lies against my skull, and reaches deep into my inner ear. / It is a sound that never exists in space. / And only I can hear it. / The difference between sound waves moving around me and the movement of a nerve that signals sound. / A cognitive listening. (Third Space)

Intensifying the sense of individual, particular, embodied and situated subjectivity, the unvoiceover’s evocation of cognitive listening denaturalizes the ear as the only place where hearing happens. It also summons contemplation of the innumerable invisible human–machine collaborations at play in the Speech-to-Text and Text-to-Speech transformations that proliferate where disability, d/Deafness, access workers and technologies (that are always already assistive) meet.

Fig. 2. Liza Sylvestre, still (detail) from Third Space, 2019, looped dual channel installation, courtesy of the artist.

As Jaipreet Virdi observes in Deafness Cures, “[f]rom the outset, the debate over cochlear implants has resided in highly emotive territory, centred on the transforming meaning of deafness” (253). In the Captioned series Sylvestre vividly illustrates the complex in/visibilities of what Brenda Brueggemann has called “the cochlear implant cyborg,” historically and for highly charged reasons, the less-visualized face of Deafness (17). As James L. Cherney observes, the figure of the cyborg has long featured in the fraught debates about cochlear implants, and is often invoked problematically in sci-fi depictions of cyborg existence as “not for every body […] not an evolutionary advance for all humanity, but rather a solution for the unfortunate condition of being disabled” (26). Within ableist, audist cultures prone to conceiving of technological fixes that “reify and affirm notions of normalcy that disable people who do not conform to the norm,” such framings come loaded with an ineradicable weight of troubling prior associations (22). As Kafer shows in Feminist, Queer, Crip, in spite of Haraway’s intention for the figure to “critique dualistic understandings of nature and culture or of human and machine, too often it serves only to reify such binary logic” (108).³⁰ Kafer’s meticulous cripping of the cyborg pulls it out of the binarism that comprehends the relation between (only) the disabled body and (an artificially circumscribed concept of) “assistive technology” as cyborgian, “cast as modeling a cyborged existence that nondisabled people have yet to achieve” (118). Going far beyond the entry-level prosthesis-preoccupied plane of direct machine/body interface, Kafer reinvests the cyborg with the much more disparate, economic, political, medico-social entanglements proper to it as (via Steve Kurzman) subject position rather than identity (121). In Third Space, Sylvestre starts by describing how

My body is combined with the computer / My own body is also part computer / Five coils of computer are pressed into small crevices of my inner ear. / Bone and muscle have fused around it. / Sound passes through it like cells. / I’m not certain where the computer ends and my own body begins (Third Space)

but then goes much further. Rather than dwelling on the classically cyborgian concept of human/machine interdependent integration, in what follows, I want to show how Sylvestre enacts and extends the politicized bent of Kafer’s “critical crip engagement” with the cyborg figure.

In Third Space, Sylvestre retools the cautionary tale at the core of 2001: A Space Odyssey and so much of the science fiction that preceded and followed it: parables of technological “advance” inevitably trailing unanticipated effects and total reliance carrying risks of system collapse. Throughout, Sylvestre’s commentary on the imagetrack directs receivers to notice how the film stages the threat to humankind from the machines they build, observing “The spaceship reminds me of a backbone / which adds to the feeling of incredible vulnerability.”³¹ When the astronauts hide in an EVA (extravehicular activity) pod to discuss disconnecting HAL, Sylvestre’s unvoiceover commentates in real time on the anthropocentric ableism that delivers them to catastrophe:

They should stop looking at the computer / They assume the computer cannot read their lips / As someone who can read lips in profile, I wouldn’t be so stupid / But this is a movie made by hearing people, and I shouldn’t assume the worst.

Soon after, as the camera’s tennis-style back-and-forth motion indicates that HAL is indeed capable of following their conversation without relying on sound, its vindication is savoured: “Ha! I knew it / Welcome to my experience of conversation” (Figure 3).

Fig. 3. Liza Sylvestre, still (detail) from Third Space, 2019, looped dual channel installation, courtesy of the artist.

Sylvestre holds them up for ridicule – these men of science so set in their thinking that they presume to extrapolate from hearing human onto a super-computer that, as it turns out, capitalizes instead on the affordances of d/Deafness.³² It is not (as per the sci-fi creed) hubris but audist ableism that causes the catastrophes that befall the astronauts who are humanity’s representatives in outer space. Sylvestre’s tone is light but the point being made (however comedic in tone) is no mere digression.

Instead, Kubrick’s iconic rendering of science fiction’s favourite technophobic warning – beware over-reliance on artificial intelligence – is appropriated by the artist as suggestively apposite background over which to think about the implications of becoming physically and financially and socially entangled with Advanced Bionics: the company manufacturing her implant and its batteries.³³ Notwithstanding the intense surface-to-surface proximities instantiated between implant and user, “there is an incredible distance between things” (Sylvestre, Third Space). Though it bypasses the ear to deliver something that might be conceived as intensely direct, the sensory experience produced is by no means unmediated. What is transmitted must be interpreted. While the popular miracle prosthesis trope positions implant as “‘cure’ – restoring a deaf person to ‘normalcy,’” and wholeness, the unvoiceover emphatically aligns it with partiality (Virdi 252). Translating sound into electrical signals that are then conveyed by radio waves directly to the auditory nerve, the implant “actually allows me to hear the full spectrum of sound” (Sylvestre, Third Space). But “the human ear is capable of detecting 40,000 different pitches, from high to low / My implant only gives me access to 16 different pitches.” Technology is taken into the body but its efficacy is not raised on a plinth, as “transformative.”³⁴

Contra the claims made since the inception of cochlear implant technologies – their promotion (as Virdi observes) in Time, Washington Post and Scientific American articles of the 1980s as “‘bionic ears’ […] ‘tiny computers’ […] ‘a functional replacement of the ear,’” and notwithstanding the technological advances of recent years, cochlear implants still “do not replicate the natural sense of sound” (250). And as the unvoiceover relates, “things are shifted, condensed, lost” as a consequence of the device’s functional limitations. Learning to hear across the implant’s imperfections takes work: “I once heard listening with a cochlear implant described as listening to a hologram. / And I have not yet come up with a better description.” Countering [hearing reporters’/commentators’] fascination with “gleaming high-tech prosthetics,” Sylvestre points up imperfections (Kafer 108). The unvoiceover portrays the implant as incorporated machine that ages, fails, breaks down and is doomed to eventually die: “My cochlear implant is beeping / I need to change my battery soon” (Sylvestre, Third Space). Its insufficiencies must be accommodated and worked around by the implant-users who train their brains to read its translational renderings of sound. Far from transcending bodily susceptibility to ageing and decay, the implant shares in it, degrading with time: “I am down to 2 rechargeable batteries and they hardly get me through a full day. / These batteries cost $300 and are easy to loose [sic].”³⁵ The second time this line is spoken, the fix is figured as much more complex than the mere swapping of old battery for new, as the unvoiceover continues: “I need to purchase a new cochlear implant battery, actually.”

As Kurzman observes, the term cyborg is useful much less as symbiont identity than it is as articulation of an individual’s position “vis-à-vis the relationships of production, delivery, and use surrounding [their] prosthesis than [their] actual physical interface with it” (qtd in Kafer 121). And whereas popular prosthesis narratives either avoid any mention of cost or glorify the rare bequest by a benevolent donor, Sylvestre acknowledges the embeddedness of these everyday dependencies on technologies that circulate within biomedical economies, requiring purchase by individuals whose access to such devices and the extensive medical care that fits and maintains them will be determined by income, health insurance, state policies, postcode lotteries. From the exclamatory observation that “new batteries last up to twenty hours” (and particularly the ad-echoing usage of that indemnifying marketer’s fudge “up to”) the receiver is reminded that, far from being partitioned off in the “assistive bracket,” these technologies are as susceptible as any consumer tech to planned obsolescence: “I’ve been living with my cochlear implant for sixteen years / Which means that the computer that has merged with my scull and muscles and sensory experience / is sixteen years old.” When the unvoiceover alludes to the implications of machine mortality, there’s a stagey theatricality to its staggering of this disclosure across two lines: a canny withholding of the punchline to ensure that the reading receiver is afforded time to contemplate what it means to be surgically implanted with a device that is already over halfway through its second decade. Careful timing prepares reader-viewer to be receptive and ready to speculate in response to the unvoiceover’s semi-rhetorical “What if it stops working? / A sixteen year old computer is my only connection to the world.”

Running over the imagetrack and pulling the reader’s attention away from the action slowly unfolding on-screen, the unvoiceover primes the receiver for the parable in which this extended lyric passage concludes: Sylvestre explains that when parts of the external processor breaks, she has to “figure out how to replace them” – usually by e-mailing Advanced Bionics, the company who manufacture the cochlear implant. And yet, even though “In those moments, I will have absolutely no access to sound / because of a broken component of my implant / […] the company who supplies my implant will ask me to call them on the phone” (Sylvestre, Third Space). Led to this point by a weave of persuasive commentary and personal anecdote, the receiver is then delivered to a rousing declamatory climax that recapitulates all of the principal points in its critique (exploitative industry practices, corporate ableism, unnecessary impracticality):

I resent that I am reliant on something that requires me to walk around with hundreds of dollars of rechargeable batteries that can be easily lost / I resent that the company who sets the high prices for these components / has the nerve to ask me to call them on the phone. / I haven’t been able to hear well enough to talk on the phone for twenty years / And my cochlear implant, even when it is working properly, does not allow me to talk on the phone. / Ever. (Sylvestre, Third Space)

The passage ends here. Though the unvoiceover’s conversational demeanour prevails, ostensibly casual chat has slid into a rousing declaration of revolt – not against deafness, but against intransigent audism.

Sylvestre’s unvoiceover goes in disguise as “open captioning”: “captions that are openly displayed and cannot be turned off” (Butler 680). Conversely, closed captions are publicly transmitted but accessed privately. Secreted away – historically in line 21 of the analogue broadcast signal – these are made available on an individual basis only when the caption-user deploys extra technology (e.g., by acquiring a once-expensive set top box), toggles a slider or clicks a button to decode them for display (Downey 4). Turning the medium inside-out, Sylvestre’s open caption-mimicking unvoiceover is intimate in tone and content – but is made publicly accessible via the open stream on-screen. Personal experience and political polemic are plaited into this single spool of text that, being burnt into the artwork’s digital file, is indelibly part of the picture. Greg Downey gives a rich account of how the “open v. closed” issue played out in America when the rights of d/Deaf viewers to access information or entertainment were set against the putative rights of hearing viewers to watch TV without having to “accommodate” the sight of open caption text they may find visually (or more honestly, associatively) unpalatable (82–102).³⁶ In art, as in TV and cinema, arguments about closed and open captions are arguments about access as social responsibility or access as private (frequently consumerized) work-around. The provider’s decision is one Soledad Zárate neatly identifies as a decision between accessibility with and without inclusivity (92). It communicates how the creator or broadcaster delimits their intended public. Privatizing access provision – requiring potential users to self-identify at a front desk, make special requests at reception or come back another day for a one-off captioned showing – consolidates an image (and a resulting reality) of the expected “public/receivership” as a homogenous mass composed of identically sensing beings.³⁷ Likewise, making captioned versions available to viewers (on inevitably inferior screens) at home while the gallery/cinema version is screened without (e.g., via an exhibition website) reinforces a toxic expectation that d/Deaf and disabled publics belong at home, and have no stake in the public spaces of culture. Whether read/used or not, open captions signal; their very visibility stakes a claim for disability visibility.³⁸

Occupying the channel that would have delivered this discreet content, Sylvestre’s unquiet (chatty, fluent, irreverent) unvoiceover fills the gap left by the captioner who was never commissioned. In doing so, it alters the site of its intervention. In the speech-to-text professions at large, even more so than in inter-lingual subtitling, erasure of the translator is customary. What is delivered “displayed” on-screen is seen to emanate from out of nowhere, such that as Downey observes, it is “rarely considered to be ‘authorship’ itself” (284). Due to their unlocatability the describer is made “unable to be called into account” (Haraway 583). Hickman’s feminist angle on the invisibility of access work in real-time captioning develops Downey’s account of the speech-to-text professions’ “visibility paradox” (Access Workers 4–7). Hickman’s subtle analysis traces the gendered development of norms dictating that stenographers are “expected to fulfill their role quietly,” and (like sign language interpreters called upon in emergency planning situations) are supposed to conceive of their work as a civic responsibility (47–48).³⁹ As Hickman also notes, this quietness and near-invisibility makes the profession especially vulnerable to the cost-cutting of austerity politics (Access Workers 4) and automation of their labour (“Willful Dictionaries” 334). In Hickman and Finnegan Shannon’s 2020 short film, Captioning on Captioning, this labour is made visible in the display on-screen of the CART captioner’s real-time generation of steno briefs (a kind of phonetic shorthand produced with Case Catalyst software) from speaker-and-discipline-specific dictionaries. To the right of the screen, Zoom thumbnails of the artists are accompanied alternately by the image of the captioner (Jennifer)’s face and of her shorthand typing hands.⁴⁰ More typically, as is handwritten on a slide that flashes up within the film: “Access work is hidden.”

Where the sonorous voiceover carries with it still a certain (compromised but unextinguished) gravitas – an “impression of objectivity and unproblematic truth” – transcoded text that appears on-screen in the black bar or matte of the caption track is coded as desiccated vocal remainder (Rangan 283). In an environment that is “not media-agnostic,” the reception of language art in digital media is highly sensitive to its temporal entanglements.⁴¹ Pitched away from the literary, text that pops on and off a screen in speech-time is aligned with orality. Though received as more “authentic”ally immediate, it is less likely to be deemed worthy of real reading – even when it entirely circumvents sonorization by an orally emitted voice. In a strange twisting of phonocentric logic, the low prestige accorded to captioning work carries across, even when writing only looks caption-like. However written (however artful) this writing, its associations with mere speech and visual resemblance to something now usually presumed to have been generated automatically deflects sustained critical engagement. This makes Sylvestre’s decision to write in caption-like unvoiceover form a risky, if ultimately rewarding, tactic.

Behind the disdain for the text that translates sound is a meanly limited conception of cross-modal translation (captioning, alt-text, audio description) as dealing only in “objective” translational truths. In their Alt-Text as Poetry workbook, Bojana Coklyat and Finnegan Shannon identify the power dynamics that have typically been in play when one homogenous position has claimed objectivity as its property, “while marking other perspectives as ‘subjective’” (35). Objectivity’s promise of transcendent “vision from everywhere and nowhere” is, as Donna Haraway much earlier observed, “an illusion, a god trick” (582). Contra the endurance of Joel Snyder’s WYSIWYS (what you see is what you say) mantra as prime directive in Audio Description, Georgina Kleege has proposed that the “pretext of objectivity” be abandoned (121) and specialist audio-describers like Elaine Lillian Joseph are devising new approaches to embedding politics and positionality within anti-illusionistic, aesthetically attuned descriptions of artists’ moving image.⁴²

In their 2016 Representations special issue on Description Across Disciplines, Sharon Marcus et al. recommend acceptance of “the basic critique of objectivity as impossible and undesirable,” advocating instead for “forms of description that embrace subjectivity, uncertainty, incompleteness, and partiality” (13–14). They propose replacing the emptied ideal of a “universal account” with a situated form of objectivity that would “[try] to get something right by attending to it closely and accurately describing how you see it – you in particular, you as a member of a class or group, you as a person with a particular kind of knowledge, training, or values” (13–14). This idea was already being explored in description-driven arts practices that frame cross-modal translation as a relational act taking place between people, just as disability itself is also inherently social and relational (Titchkosky, Question of Access 4–5).⁴³ These works exemplify Kleege and Wallin’s point that “the act of describing is itself an aesthetic performance that generates its own meanings,” without losing track of the crucial responsibility to seek after accuracy (n. pag.).⁴⁴

By channelling a personal essay via a customarily de-subjectified, ostensibly “authorless” form, Sylvestre’s unvoiceover undoes the association of captioning with neutral objectivity. She demonstrates that situating description “allows us to become answerable for what we learn how to see” (Haraway 583). The friction that results between the inherited, historically subjectivity-denying, unlocated, unbodied medium of the caption track and Sylvestre’s hyper-first-personal content prompts the reader-viewer to reappraise the origins and givens of words on-screen. By giving over much of the concertedly first-person unvoiceover to novel interpretations of the film’s sound and imagetracks, Sylvestre points up the inadmissible possibility of the captions writer as (situated) writer. In the process, she prepares the ground for the reframing of text-on-screen as something other/more than “a mechanical process of unreflective transcription” (Zdenek xiii). Off-screen and outside of the diegesis, the acousmatic voiceover, as Silverman observes, “is privileged to the degree that it transcends the body,” and as Rangan notes, “forcefully drew attention away from its own source, materiality, and corporeal particularity” (Silverman 49; Rangan 283). Sylvestre’s unvoiceover exerts a contrary, corporeality-centring intention.

Doane writes of the voiceover that “it censors the questions ‘Who is speaking?,’ ‘Where?,’ ‘In what time?,’ and ‘For whom?’”; the unvoiceover anticipates all of these questions and answers them prophylactically and personally (Doane 42). Whereas standard captions traditionally disclose no trace of their production by any writer, the unvoiceover vividly situates its production as resulting from the effortful labour of an embodied (and so bodily bothered) individual (made thirsty by an air vent, sore of back by bad typing posture) who is, in turn, situated in a real (particular and problematic), geographic location:

I resent the idea of divine artistic insight. / I resent the idea of the artist as god-like. / My own ideas are always grounded in my life. / There is nothing transcendent about them. / I am here watching and typing and my back hurts. / And the air vent above me is blowing dry air, and it makes me thirsty. / And if I’m honest, I am tired of always pushing and pushing. / And there is so much work ahead. (Sylvestre, Third Space)

In the “Crip Technoscience Manifesto,” Hamraie and Fritsch recover the etymological origins of access in notions of attack (11). Sylvestre’s access-oriented intervention, her way of “[using] technology as a friction against an inaccessible environment,” takes up this idea in two ways at once (Hamraie and Fritsch 11). Her intervention can be considered an act of access aggression – even graffiti-like vandalism. By writing over the film, she also attacks the complacency of the hearing viewer who presumes always to have the luxury of switching modes of engagement as fits their fancy. With the Captioned series, Sylvestre offers a corrective to the co-opting of captioning within “curb-cut” discussions: contexts within which claims are made for the benefits of captions for hearing audiences. One of the advantages most prominently evoked (alongside benefits for language learning, literacy) is the possibility for exploiting redundancy.⁴⁵ For the hearing and sighted receivership, the co-presence of audio and captions typically indicates options. A sighted, hearing receiver might choose to engage via either sight or sound, or to toggle between them depending on the environment (e.g., accessing a broadcast on mute via captions in a public space) or listening only to the audio of a TV show while washing up. For the hearing viewer, Sylvestre’s Captioned films undermine that entitlement, requiring instead a supernormal – and initially startling – degree of attention to the fleeting text. Sylvestre sets up a viewing dispositive within which the eyes must be held unusually, uncomfortably open. With barely any overlap between what is audible and what is legible, Captioned works demand an augmentation of attention among those ordinarily most centred and made most comfortable. Instead of facilitating the distracted, semi-present engagement of hearing viewers, they compel concentration. The advantage ordinarily conferred upon the hearing viewer by audio-visual redundancy is denied. This overworking of, and ultimate beleaguering of, the receiver achieves a didactic purpose. It sets up the embodied, sensorially overextended, semantically frustrated experience within which the receiver can begin to imagine a sensorium beyond their own.

Having thus primed them to think about their own audio-visual engagement in terms of effort, decision-making and partiality, Sylvestre’s unvoiceover cues its hearing receivers to consider the enormous extra effort required when, in the absence of captions, lip-reading is the sole means by which verbal content can be accessed. Referring to the energy expended in “the strain of trying to comprehend” uncaptioned audio as “wasted energy” (Sylvestre, Captioned: Twentieth Century), it points up the opportunity cost of time spent “groping after threads of information” (Sylvestre, Third Space). In Captioned: Twentieth Century, Sylvestre writes of how she has “redirected that energy here [into the writing of the unvoiceover commentary] so that it feels less wasteful.”⁴⁶

By way of an ambient refresher in d/Deaf-inclusive communication, the screentext observes: “I appreciate people who talk with their hands / Clear, even voices. Eye contact” (Sylvestre, Channel Surfing). In illustration of the benefits of this approach, speech that is delivered to camera, by faces framed in close-up in the film on-screen is relayed in rare perfect-match form by the unvoiceover. The reader-viewer is trained to recognize what effective communication looks like. The text on-screen momentarily functions to caption spoken words. The unvoiceover also underscores the contingencies that further complicate engaging with screen media as a d/Deaf person in an ableist, captionless culture: “Beards cover mouths, accents do too” (Channel Surfing). Over an animated passage, she writes: “It is impossible to read cartoon lips.” In place of an indexically intelligible relation between visible oral movement and sound there is instead only the hollow simultaneity of “A flicker of lips. A flutter of words” (Channel Surfing). As well as drawing reader-viewers’ attention to the frequency with which crucial facial parts are obscured, the artist also raises awareness about less obvious obstacles out in the receiver’s real world. Writing over footage of a 2001 space-station crisis meeting in Third Space, the unvoiceover points out that:

Even if I was attending this meeting in person it would be almost impossible to decipher his words. / I know those faux windows would flatten the shapes of everyone and make it hard to see the details of their faces. (Third Space)

Sylvestre’s running commentary also draws attention to the centrality of contextual information and pattern recognition in lip-reading, as in all translational and interpretive practice. On recognizing an utterance of the phrase “and my life depended on it,” the unvoiceover remarks: “This is the perfect example of how context is necessary for me when communicating verbally. / Because that is a common phrase it’s like there is a small part of my brain that is always expecting it” (Sylvestre, Captioned: Twentieth Century). The lesson being taught spills beyond the frame speech-interpretation in d/Deafness to inflect thinking about communicative acts of all sorts. In analysing these instances in the text on-screen, the point is to move beyond hiving off the insights to be gleaned here as pertaining only to communication in the context of d/Deafness, “as if there were bright lines” between disabled and non-disabled experience (Kafer 11).

Watching along with Liza – for the unvoiceover constructs a cosy sort of first-name-terms companionship – it becomes clear that Sylvestre’s interpretation of film even (or maybe especially) as it approaches its object obliquely is not impoverished by her hearing status. Instead, the unvoiceover can be read as an accumulative essayistic enactment of a phenomenon relatable to, but not collapsible into, Bauman and Murray’s theory of Deaf Gain.⁴⁷ As Michael Davidson observes, “Deaf gain involves reordering of priorities not around sound but around sight, of signed rather than spoken or written language, of communication written on the body, not about it” (8). Davidson acknowledges that the move from a discourse of hearing loss to one of Deaf gain “has been a positive step forward in claiming deafness for diversity.” However, he goes on to argue that “this value is weakened when it excludes large portions of the non- or partially hearing public who occupy various positions on what Christopher Krentz has called ‘the hearing line’” (8–9).

The gain evidenced in Sylvestre’s deaf reading of uncaptioned film is not the same as the Deaf Gain that was proposed by Aaron Williamson, theorized by Bauman and Murray and has since been elaborated and celebrated. Nonetheless, the advantages communicated do originate in, or are made possible through, the artist’s small-d deafness. Noting that Williamson, “who inspired the phrase Deaf gain, is a successful deaf performance artist whose work about being deaf is everywhere and yet who does not himself sign and who reads lips,” and later that he [the author] himself has “certainly gained” from progressive deafness, Davidson suggests that “the claims [of Deaf gain] need to include a broader constituency of persons who live close to the hearing line and who support the political goals of Deaf World” (9). Davidson goes on to underline the specific boons of his “slow and gradual ability to use ASL” (10) and, importantly, is careful not to undermine the particular weight and meaning of Deaf gain within Deaf World. In invoking a relation to Deaf gain here, I mean to suggest that Sylvestre’s work might be read as an articulation of gain amongst that “broader [d/Deaf] constituency” – in this case, as experienced by a lip-reading, ASL-learning cochlear implant-user: “someone who is not culturally deaf, even though I am medically deaf” (Sylvestre and Hamraie). With Bauman and Murray’s claim that “Deaf Gain is itself diverse. We do not posit a universal Deaf Gain” in mind and with Davidson’s call for broadening the constituency as cue, I want to propose that Sylvestre’s unvoiceover delivers, among many other things, an assertion of “the unique cognitive, creative, and cultural gains manifested through deaf ways of being in the world” (Bauman and Murray xv).

As the unvoiceover demonstrates, not compulsively following the dialogue to the exclusion of almost everything else leaves time and energy to ponder more abstract phenomenological aspects of the mise-en-scène and cinematography: “How can that red color feel both calming and anxiety inducing?” (Sylvestre, Third Space). The Captioned series makes its receivers aware of the degree to which their film-viewing is circumscribed by the habits of verbocentric attentiveness into which they have been culturally trained. Via the unvoiceover, Sylvestre attunes the receiver to perform forms of image analysis and character study not available (not accessible) to one obsessively following every exchange in a conversational set piece. Access is not always available. Absent the accompanying audio, the moving image must be read more searchingly. Shortcuts are not so easily, unconsciously taken. Sylvestre’s visually hyper-attentive film-reading practice slows down and separates stages of apprehension, comprehension, interpretation, etc.: calling the receiver to notice not just the presumed “meaning” of a particular facial expression but the sequence of muscular and architectural moves that make it up, when: “Her mouth and eyes stretch open” (Channel Surfing). The unvoiceover’s deconstructions of what adds up to a particular facial expression recalls contemporary debates among audio description professionals about the probity of presuming to decode such facial expressions for a blind and low vision listenership.⁴⁸ It re-aestheticizes film-making as art.

In implicit rebuke to historic injunctions against text on-screen as distraction from, and sullying of, the image, Sylvestre uses the caption track to expose how typically distracted the hearing cinema audience usually is – and, how narrowed and impoverished thereby their engagement with film. The neat symmetry of this reversal is discernible in the observation of how, when speech on-screen becomes shouting, “And then the volume of their voices takes over. / Distracting me from their movements” (Sylvestre, Captioned: Twentieth Century). When, in an emphatically line-heavy film such as Twentieth Century, near-incessant dialogue is uncaptioned, while an unvoice speaks distinctly and directly from the caption track, otherwise audible dialogue is marginalized and the hearing receiver’s attention is accordingly loosed to alight elsewhere in the moving image, to do other sorts of reading. Sylvestre’s Captioned series poses a convincing corrective to image purity preciousness. Via the voice made visible, we are primed to see otherwise, by other ways. In cinema at large, as Liz Mills observes of engagement with voice in dramatic performance: “As soon as the actor speaks, sound as vibration, sound as energy, and sound as texture are eclipsed by the presence of language” (394). Sylvestre’s series trains reader-viewers to understand that it is not text on the screen but instead the dialogue in the film that distracts from what is going on elsewhere – in the image, and/or elsewhere in the rest of the soundtrack. Guiding the receiver towards new ways of seeing, Sylvestre’s work answers Alison Kafer’s call, in Feminist, Queer, Crip for “the narratives of people whose bodies and minds cause them to interact with [culture] in nonnormative ways” – transposing the terrain from nature to film (135). As attentional triumphs accumulate, a wry playfulness is evinced – as when, in the same video, the unvoiceover points out that none of the constantly chatting characters in a train carriage scene are swaying with the train’s movement, and suggests “I think IMDB should hire hard of hearing people to catch those trivia bloopers” (Sylvestre, Captioned: Twentieth Century).⁴⁹

The rich, startling visual descriptions that populate the unvoiceover upend the prejudiced presumption that it is the d/Deaf experience of film that is impoverished and partial, and demonstrates a lower-case gain in action: “I suppose if I were more drawn into the details of the conversation / I wouldn’t notice these things as much” (Sylvestre, Captioned: Twentieth Century). Throughout the series, body language is explicitly analysed, and faces are read for clues as to characters’ intentions and relations. Under the tutelage of the expert lip-reader’s unvoiceover, facial mobility is made newly noticeable as legible; facial structure is loosed from being the personal property of a particular actor and becomes newly appreciable as integral, dynamic architectural element of a cinematic composition. The lip-reading artist enjoins the receiver to notice the “Deep lines and rounded arches” of a face that is shifting from surprise to amusement (Sylvestre, Channel Surfing).

As Marcus et al. observe, “description can make us more attentive […] description can allow us to see more and to look more attentively, more fully, and more selectively” (14). As a result, and as Georgina Kleege and Scott Wallin have argued, audio description has massive pedagogical potential for teaching visual analysis and critical thinking. Much more widely, the arguably more distributed practice of writing alt-text immediately elicits a new kind of visual attention from those who find themselves learning, for the first time, to describe what they might at first presume to be a simple, straightforward image. In their reaching, experimental formulations, Sylvestre’s descriptions exemplify what Marcus et al. celebrate as “the essential generosity that can attach to description as a practice when it attends not only to its objects but also to the collective, uncertain, and ongoing activity of trying to get a handle on the world” (4). Like a helpful teacher, Sylvestre’s unvoiceover cues the reader-viewer to notice what might otherwise pass them by. Via the artist’s interventions, the aesthetics of the films she over-writes come to seem richer and more complex – in part due to the synaesthetic bent of the unvoiceover’s analysis. The textures of three-dimensional objects already translated via celluloid into two-dimensional images are described in terms that – through sound – summon an irretrievable hapticity. Meanwhile the soundtrack is described in terms borrowed from visual rubrics: music “grows stronger and more in focus” and “voices grow sharper as they approach” as though increasing proximity pulled blurred soundwaves into tighter definition (Sylvestre, Third Space). Watching along with the voice made visible retrains the reader-viewer to engage with film as much more than the vehicle or wrapping for a script, or story. In the process, preoccupation with plot cedes to attention to form, rhythm. In this delivery from narrative, “describing and descriptions can produce pleasure – granular, slow, compressed, attentive, appreciative” (Marcus et al. 14). The reader is delivered into a more aesthetically acute encounter.

The unvoiceover also serves to expand the receiver’s engagement with sound, with sonic events described suggestively, as “An electronic woozing,” or “A grunting knowing” (Sylvestre, Third Space). Where sound studies likes to make grand, sometimes gruesome, claims about the bodily basis of vocal production, Sylvestre’s commentary explicitly, concretely conceives of voice as vibration.⁵⁰ Vocal utterances register as much more than mere sound-borne signification. Within the sensorium instantiated by this work, sound is manifestly massy, material, relational and bodily; it has mass, depth and mobility. Sounds can “spill onto one another” (Third Space). One character’s voice “rests low in his face. Almost like he is talking from his chin” (Third Space). A voice that is “moving slowly right now […] drags slowly over [a character’s] thoughts”; another “flies up smothly [sic] and then twists in little rapid sounds” (Sylvestre, Captioned: Twentieth Century). The unvoiceover visualizes the process of sound production and makes freshly sensible the often-remote reality of a soundwave as a physical phenomenon. A change in tone can be ascertained visually when “The dark parts of his face are moving in new shapes” (Sylvestre, Channel Surfing). Lip-reading leads to comprehension only when a visually legible pattern forms: “His voice just has to ‘fall into the shape’ and suddenly I comprehend” (Sylvestre, Captioned: Twentieth Century) (Figure 4). When singing voice and instrumentation interact, “[t]he music parts and makes room for the shape of this voice” (Sylvestre, Third Space).

Fig. 4. Liza Sylvestre, still from Captioned: Twentieth Century, 2018, single channel video, courtesy of the artist.

Attending to sound in this more material, non-acoustic way also denaturalizes immaterial cultural boundaries between sound and music; it hollows out absolute distinctions between intended and ambient sonic registrations. In Third Space, the unvoiceover asks “Is that sound laughter or music”; a voice that “moves in quick shapes that I can’t follow” is “more like noise and music than voice” (Sylvestre, Third Space). This flattening of hierarchies can be imagined spatially, as the collapse of distance between perspectival sonic locations that, as Theo van Leeuwen summarizes, all share a “three-stage plan”: Beeby’s “‘Immediate,’ ‘Support’ and ‘Background,’” Murch’s “‘foreground,’ ‘mid-ground’ and ‘background,’” Murray Schafer’s “‘figure, ground and field’” (16). The unvoiceover describes an unplaceable sound that is maybe soundtrack, background noise, or “my neighbor playing obnoxious music” (Sylvestre, Third Space). When suspended in moments of non-recognition, it can be possible to be taken by surprise, taken raw, apprehended by a sensory percept, before it is fully, comprehendingly categorized. These articulations of uncertainty facilitate realizations about the ways in which hearing is always conducted through culture and categories, as well as through air, bodies, spaces and other materials. No mere rhetorical gimmick, these translations of percepts across and between the senses reinscribe the point of Sylvestre’s captioning project and of media access strategies for devising alternative formats: generating experiential equivalences and dismantling presumed “givens” of sensory experience more generally.

Sylvestre’s takeover of the space that could carry a caption track, and pointed mis-titling of the unvoiceover as captioning summons reflection on the imparities that can result when, in the absence of universal open captioning, gatekeepers determine what is necessary, appropriate or morally justifiable to caption (Downey 238–40). It also invokes ongoing conflicts about whether d/Deaf receivers ought to be granted access to verbatim transcripts of dialogue or, for rationales that never fail to sound paternalistic, provided instead with edited versions thereof. Reading the Captioned series with these unresolved histories in mind reveals the unvoiceover as rebuke to caption-(less) culture. Masquerading as transcript while pursuing its own agenda, the unvoiceover makes a covert claim for media justice.

As well as training its receivers to engage in a new way with film, Sylvestre’s series also reorients their (our) conception of episodes from film history. The artist’s inaudible, legible unvoiceover reprises roles previously fulfilled by (legible) intertitles and in the film lecturers or explainers who, in early film screenings (as before in magic lantern shows) “mediated between the screen and the audience, introducing, commenting on, and so implicitly interpreting the film” (Nagels 370). Its aura of liveness and provisionality replicates the conditions of labour under which film lectures were often given. As Kozloff explains, “qualified lecturers were hard to find and cost money, and since they didn’t always get a chance to view the films before they were shown […] their commentaries were not always well prepared” (24). The unvoiceover slides between interpreting the film itself, explaining how access is routinely denied to d/Deaf publics/receivers and narrating, in first-person form, her experiences of processing inaccessible media. Similarly, film explainers performed numerous functions in cinemas – as hinted at by the dual circulation in Francophonic film contexts of the terms “bonimenteur” (translated as yarn-spinner) and “conferencier” (lecturer). Meanwhile, the prowess of their Japanese equivalents, the “benshis,” made some “stars in their own right,” celebrity describers who “lasted into the early sound years” (O’Sullivan and Cornu 16). Sylvestre’s unvoice on-screen encompasses each of these roles, and more besides. As Louise Fryer observes, these explainers “did not actually explain films but made them accessible to audiences who were not sufficiently literate to understand the ‘language of cinema’” (15). And indeed, it is possible to align the film explainer’s alignment with a kind of access work – particularly in view of the fact that, as O’Sullivan and Cornu note, once title cards were introduced, these film lecturers would “read out or explain their content for the benefit of illiterate audience members” (16). Kozloff explains that although “in the United States, lecturers seem to have been employed primarily for clarity and convenience, the fact that Japanese lecturers were more popular with audiences than movie stars demonstrates the affective potential of filmic narration” (24). This spectrum from minimum box-ticking functionality to fully elucidative cross-modal translation – and its indexation to resource-provision and preparation-time – mirrors the spectrum discernible in the production (and resultant use and appreciation) of captions and audio description. Ultimately, as Kozloff notes, lecturing was “reborn in the sound era” by voiceover narration (26). Sylvestre’s unvoiceover recuperates and reimagines all of these ostensibly obsolete, chronologically distributed roles – the intertitle, the explainer, the lecturer, even the chorus, the banished voiceover – transposing them from audible to legible form.

A further model for the chatty plenitude of Sylvestre’s unvoiceover might be located in the “literary intertitles” most famously associated with the “deliberately conspicuous,” fourth-wall demolishing extra-diegetic commentaries of Anita Loos: a precociously early proponent of expanded description (Nagels 376). Still more strikingly, Kozloff writes of how D.W. Griffith “habitually took advantage of his titles to judge his characters, make personal asides, or draw parallels between the screen action and current events” (25). The line between the titles of the 1920s film and Sylvestre’s series is made still more explicitly by Kozloff’s suggestion that the film-maker did so “to open up a direct line of communication between himself and his audience and to suggest a personal tone of voice” (25). In 1926, Loos’ contemporary, the film critic Iris Barry compared the function of intertitles to that of the chorus in Greek and Japanese theatre (78) and Sylvestre’s unvoiceover – with its mix of commentary and interpretation, and stable standpoint – also does much of the work of a chorus. Like the intertitle in Barry’s paean to what she (now rather aptly) called the “sub-title,” the artist’s use of the caption-like unvoiceover “might well be held to be a new form of literary style” (82).

Staying in this early film era, antecedents for contemporary resistance to captioning can be found in early twentieth-century responses to the introduction of intertitles in film. While some celebrated title cards as evidence of the medium’s increasing sophistication,⁵¹ others decried them as either evidence of aesthetic weakness, or “the intrusion of the literary into what should be an essentially pictorial realm” (Nagels 373–75; Marcus 290). Between the poles of this debate about the place of text within the frame of an always already replete visual medium, many held on to a purist ideal of cinema’s visual language as one that ought not to need the additional explication provided by text title cards, whilst pragmatically acknowledging the benefits for some, if not (as they saw it) all viewers and films. Under this view, titles were “necessary, if perhaps not desirable,” as Nagels concludes (375). Those of more utopian bent grounded their opposition in allegiance to a belief in the internationalism of film as a border-transcending visual language (Hillman 382) that was “supposed to be the esperanto that would unite a world divided by misunderstandings and narrow nationalism, and ultimately wracked by two world wars” (Kozloff 10).

In the bad faith debates of later twentieth-century sub-dub wars, resistance to text-on-screen was often purveyed as motivated by a defensive antipathy to visual distraction from captions understood to “distort the visual purity of the image and misrepresent the fullness or subtlety of its sounds while requiring audiences to break modes of identification and suture in order to read rather than watch and listen” (Dwyer 28). Or, as Bosley Crowther put it in a 1960 article on why “Subtitles Must Go” for the New York Times, as all of “wrong,” “thoroughly inartistic” and “obsolete” (qtd in Dwyer 20). Dubbers maintained that reading text alongside the image would inevitably detract from the viewer’s appreciation of the visuality of the medium – producing thereby an “audience/viewership” of bad viewers. Casual scholarly references to text on-screen can still disclose almost comically uncritical disgust at how subtitles corrupt what is otherwise pure. One subtitling handbook remarks that subtitles:

of course [italics mine] encumber the screen. The time it takes to read them prevents us from watching the visuals properly, and can seriously distract from the visuals. In standard film interpretation a hierarchy of the senses is in operation, with the visual having clear primacy, so that such schizophrenia of the senses, and a division within the visual area, can create irritation. (Hillman 385)

While the author here goes on to acknowledge the further exacerbating effect of the linguistic imperialism of English in the international film industry, before concluding that “the US dislike of subtitles is readily understood,” the handbook-writer’s own apprehension of subtitles as aesthetically abhorrent obstruction to understanding is equally easily discerned.

Approached retrospectively, historic arguments about intertitles as obstacles disrupting the egalitarian mutual, simultaneous comprehension of viewers representing multiple linguistic communities exert a certain (if qualified) charm. But anti-captioning sentiments circulating today lack any such idealistic associations. Rather, I want to propose that hostility to open captions mirrors the phobic quality of the ecologically skinned (surfaced) outrage Alison Kafer identifies in resistance to initiatives designed to make nature trails more accessible. In both instances, the ableist opposition evinced is predicated upon “the ways in which nondisabled access is made invisible while disabled access is made hypervisible” (Kafer 138). In Feminist, Queer, Crip, Kafer elucidates how efforts to make hiking trails wheelchair-accessible are frequently construed as “inherently more damaging” than interventions made to enable access by walkers. Transposing Kafer’s analysis from the hike trail to the gallery allows me to suggest that envisioning a media context in which each presentation of moving image is made “as accessible as possible” requires that screenings and installations of those works “are no longer designed only for one single body, and that decisions about [access] are recognized as decisions, ones that can be changed, extended, modified” (139). Kafer’s reframing of steps as “themselves an accommodation” (138) allows us to recognize the equivalent, equally artificial and arbitrary distinctions that are drawn in screen media contexts between what is considered necessary and natural access provision (the installation of a sound system powerful enough to make dialogue on-screen audible to hearing viewers scattered across a large space) and access provision as intrusive, expensive and additional (captioning, audio description).

Reflecting on the phases of the introduction, testing and ultimate abandonment of open captioning by American broadcasters of the late twentieth century, Greg Downey’s history of closed captioning discloses a note of regret as it marks a juncture when media accessibility might have taken a different direction:

Perhaps the potential solution of open captioning – artistically more adaptable, economically more palatable, and arguably more democratic – was written off too soon by deaf activists and network executives alike. Perhaps hearing audiences would have embraced text-on-television in at least some contexts – sports, news, education – providing a simpler and broader starting point for the ultimate goal of universal broadcast accessibility. Today’s littering of American television screens with textual channel logos, program advertisements, and news “tickers” makes one wonder what might have been.⁵² (101–02)

In 2008, Downey wonders. Decades later, our infinitely more cluttered screen media contexts demand a retrial. A new reckoning with open captioning as screen media default is overdue. In certain limited contexts, it is already underway.

Just as Kafer highlights how anti-ramp furore in New Hampshire’s White Mountains is predicated upon ecocritical assumptions that “some kinds of experiences can be interpreted as more valid than others, as granting a more accurate, intense, and authentic understanding of nature,” so Downey’s “what might have been” points up the TV industry’s presumptive anticipation of public opposition to captioning as derived from something other than mere aesthetic preference. As Emily Watlington argued, we need to interrogate this “aesthetic distaste” to the translation of sound on-screen, and to consider “whom it privileges” (Watlington and Lehrer 48). Resistance to open captions is not about disrupting how moving image art looks. It is, rather, about disrupting ableist, audist cultural expectations of who should be looking at the moving image. This opposition which passes itself off as aesthetic, transcendent, immune to critique resistance to the presence (coded, pace Kafer’s example) as an improper, wrong-headed, naive intrusion of d/Deafness and/or disability on a terrain that is accustomed to rendering itself accessible only to “one single body,” one single sensory system.

Sylvestre’s decision to work within the constraints of conventional caption-form is crucial to the realization of her mission. Encountered unflagged in a gallery, it might take a hearing receiver a moment to make sense of the mismatch between the words on-screen and those they hear. An appreciation of the politics of media access is taught via a participatory exercise in frustration by form. Under cover of captioning, the unvoiceover – her subtle, serial, sub rosa essay – can pass incognito, without the palaver of Captain Marvelous in his hat. Textual trojan horse, it passes itself off as only the faithful entextualization of the film’s voicetrack. In the process, access is made material and the case for captioning is made. So, too, is the case for exploring the potentiality of the unvoiceover as time-based textual medium.

Thanks to Liza Sylvestre and Louise Hickman for their reading, writing and thinking, and to Hannah Wallis for wisdom. Thanks are also due to those who offered responses on early versions of some of this material, at the UEA Modern and Contemporary Seminar, and at the University of Hamburg. This work was supported by the AHRC under Grant AH/V006096/1.

disclosure statement

The author reports that there are no competing interests to declare.

Notes

Thanks to Liza Sylvestre and Louise Hickman for their reading, writing and thinking, and to Hannah Wallis for wisdom. Thanks are also due to those who offered responses on early versions of some of this material, at the UEA Modern and Contemporary Seminar, and at the University of Hamburg. This work was supported by the AHRC under Grant AH/V006096/1.

1 See also Hickman on the campaign by the National Captions Institute to frame captioning as “The New Reading Teacher.” Whereas Captain Marvelous addressed Deaf children as users, the mission here (the “assistive pretext” in play) was to engage hearing children and their pedagogically minded parents as a means to develop majority support for captioning (Hickman, Access Workers 58–59).

2 The title of Kleege’s book, from which I borrow this construction, is More than Meets the Eye: What Blindness Brings to Art.

3 Jarring though it might be, this shift into a more lyrical, quasi-confessional “I” voice is purposive: a gestural acknowledgement of this author’s personal shift in attitude. This “I” that acknowledges its oversights as well as outlining the structure of its argument registers a belated shift inspired by an artwork that speaks, itself, in a constructed, constructive first-person, from a situated position.

4 See Emma Wolukau-Wanambwa’s Promised Lands (2015), Hayden, “Emma Wolukau-Wanambwa.”

5 See Prouvost’s You Are the Only One (2008), OWT (2007) and We Will Go Far (2015).

6 See Hayden, “Liquid Citizenship” and “When Attitudes Become Platitudes.”

7 As Titchkosky writes elsewhere: “We are educated to sense the appearance of disability as the excludable” (“Educated Sensorium” 284).

8 See, for example, recent web pieces by Kehe, Kanter and Youngs.

9 See, for example, San Alland on the disappearance of “Online events, which were suddenly possible when non-disabled people couldn’t go out” in “Writing from the Groin,” forthcoming as a commission in text, captioned video “footnotes,” plain language, alt text, audio and BSL from Birds of Paradise Theatre Company.

10 “Technotexts” is the neologism proposed by N. Katherine Hayles in 2002, one she offers as “a term that connects the technology that produces texts to the texts’ verbal constructions” (25–26).

11 See, for example, the Art of Captioning research group I co-led with artist, curator and d/Deaf activist Hannah Wallis at Wysing Art Centre under the auspices of the British Art Network (2022), our Caption-Conscious Ecology programme with Nottingham Contemporary (2021) and my slow emergency siren, ongoing project with LUX (2022).

12 See, for example, Ellcessor, “One Tweet” and Hussain.

13 On Google offering “essentially a subtitling service masquerading as a captioning service,” see Zdenek 107.

14 My usage of “entextualization” here follows that of Sean Zdenek in [reading sounds], which comes in turn from Mary Bucholtz’s work on the “entextualization and recontextualization of speech via transcription [as] a fundamental methodology of discourse analysis.” See Bucholtz 784 and Zdenek 250.

15 Padden and Humphries’ definition remains helpful:

We use the lowercase deaf when referring to the audiological condition of not hearing, and the uppercase Deaf when referring to a particular group of deaf people who share a language – American Sign Language (ASL) [or other sign language] – and a culture. The members of this group have inherited their sign language, use it as a primary means of communication among themselves, and hold a set of beliefs about themselves and their connection to the larger society. We distinguish them from, for example, those who find themselves losing their hearing because of illness, trauma or age; although these people share the condition of not hearing, they do not have access to the knowledge, beliefs, and practices that make up the culture of Deaf people. (2)

16 See Hayden, “Caps On/Caps Off.”

17 See, for example, Cavarero’s assertion that the voice is indexed to “the vital and unrepeatable uniqueness of every human being” and claim that “When the human voice vibrates, there is someone in flesh and bone who emits it […] this corporeal root of uniqueness” (4–5).

18 On accessible cultural commons and media justice, see Ellcessor’s Restricted Access.

19 As Michael Davidson observes, these and other “recaptioning” works can “[complicate] the idea of captions as textual adjuncts to something else” (177).

20 The term “normate” was coined first by Rosemarie Garland-Thomson.

21 See, for example, Federman, Antin and Spigel.

22 See, for example, Sylvestre’s 2018 Audio Description project and 2022 video, asweetsea: a paean to description as everyday access-making. For the artist’s 2022 solo exhibition by that name, Sylvestre made a pair of captioned audio-visual audio-tours: one of which provided detailed visual description of the exhibition contents and arrangements; the other of which offered a lyrical, poetic entry route.

23 See Ong on how “Hearing rather than sight had dominated the older noetic world in significant ways, even long after writing was deeply interiorized” (117) and how “whereas we feel reading as a visual activity cueing in sounds for us, the early age of print still felt it as primarily a listening process” (117–19).

24 Conversely, the fleetingness of text-on-screen might arguably exceed even that of a sonorous voiceover, if we follow Nancy’s claims for the capacity of sound to outweigh, enlarge and thicken form, such that while “The visual persists until its disappearance; the sonorous appears and fades away into its permanence” (2).

25 Determining line-break-like divisions and timing the appearance and disappearance of the captions on-screen takes time.

26 See Zárate 72 on the prevalence of “simple present tense or present participle to subtitle” sound in captioning professional practice.

27 The useful umbrella term “translation overlay” is proposed by Mara Mills in “Listening to Images.”

28 Here, and throughout, quotations from voiceworks are marked by parenthetical citations indicating the title of the work from which the line is transcribed. The convention of indicating line breaks in quotations from poetry with [/] is borrowed to indicate breaks between image-frames in voiceworks.

29 A parallel can be drawn here to Hickman’s call for the cultivation of “a feminist practice that allows for and forgives human-made mistakes in captioning” to counter the hiding of “the diversity of those assembled in its standardization” (“Willful Dictionaries” 336–37). See also Hickman’s “The Abundance and Conflict of On-Demand Writing.”

30 Jillian Weise’s “Common Cyborg” essay poses a persuasive alternative critique to Haraway’s disability-erasing use of the cyborg as universal metaphor.

31 On HAL as, with Star Trek’s USS Enterprise, the twentieth-century “‘parents’” of the talking computers that became ubiquitous in the twenty-first century, see Faber 25–56.

32 In The Computer’s Voice: From Star Trek to Siri, Liz Faber quotes Philip Kuberski’s claim that in Kubrick’s films, “technology is never rid of its corporeal designers: It carries within a kind of transmuted human psychology and intrinsic waywardness” (31). Following this through might suggest that HAL was the work of lip-reading d/Deaf designers.

33 On the complex and enduring physical and (federally differentiated) financial demands of cochlear implant use as knotty counter to the smooth fictions of cochlear implant as “deafness cure,” see Virdi 252–54.

34 This adjective abounds in the context of cochlear implant stories (e.g., “Cochlear Implants to Benefit More People with Hearing Loss” on BBC News Online), and is the first line on the Advanced Bionics “Cochlear Implants for You” page.

35 The apparent typo – “loose” for “lose” – registers as something between a Freudian slip and a poetic play on the experience of easily mislaid possessions as, in their capacity to vanish, more creaturely than thingly: tiny autonomous bugs loosed to roam a home.

36 In/visibility haunts at the point of event/screening promotion too, where the use value of any access provision is voided in the absence of comprehensive, accurate, advance communication about the form that provision is to take. See, for example, Carolyn Lazard’s influential and useful “Accessibility in the Arts”: “The only thing more inaccessible than an inaccessible space is not providing information about how the space is structured. Listing access information requires that institutions address what they can accommodate and especially, what they can’t accommodate.”

37 As Waldo observes:

From the perspective of people with hearing loss, open captions can be attractive because there is no need to self-identify, no equipment to check out, and no need to focus back and forth from the caption-display device to the screen. (1057)

38 Readers are referred to the Disability Visibility Project.

39 This surrender of subjectivity in civil service is pointed up in Hickman’s more recent work on wilful CART dictionaries, wherein she notes the function of the CART captioner’s “hybrid Zoom name (a mix of their proper name and the name of the CART agency)” to further set them outside of an online conversation, as neutral, quasi-corporate observers (“Willful Dictionaries” 332).

40 Within the film, Hickman makes a point of asking the CART captioner if she can be identified as “Jennifer” within the film. Jennifer’s response – “Yeah, there’s enough of us […] There’s a lot of Jennifers” points up the presumed (illusory) mutual replaceability of individual access workers. What the film makes clear is the degree to which, as another slide puts it: “Access work is shaped by relationships.”

41 Cayley suggests that though “embraced and appreciated as aesthetically, culturally valuable at the highest levels,” “[p]erformative, time-based linguistic practices […] may not be accepted as unequivocally within the domain of literature” (211).

42 In illustration, Kleege compares the presentation of the three types/forms of audio-guides MoMA (Museum of Modern Art) makes available. While the version for sighted adults is usually voiced by and attributed to a specific museum curator, those “for blind people and children are delivered without attribution, as if the words are indisputable facts” (Kleege 111). An annotated version of Joseph’s script for Black Audio Film Collective’s Handsworth Songs, is published at slowemergencysiren.org.uk.

43 Jordan Lord explores “the performative capacity of telling” as a counter to the presumed primacy of simply showing (After … After … Access). In their videos, Lord’s mother and grandmother, Deborah and Mimi Lord, frequently join or replace the artist in delivering visual descriptions from a position that whether off-screen or on-screen is emphatically proximal (intimate) and familial – even when, as in Expressions, it is additionally mediated by telephone.

44 On the risk of enthusiasm for universal design easily tipping into the sidelining of the disabled people for whom access provisions were originally developed, see Williamson and Ellcessor, “Blurred Lines.” See also Coklyat and Shannon’s insistence on centring the experience of alt-text users (11).

45 See, for example, Perego 115–16. On the historic use of comparisons between captioning and “sidewalk cuts” in pro-caption advocacy by the President of Gallaudet University testifying before congress, see Downey 242.

46 Sylvestre also discusses the time and energy expended at school on deciphering inaccessible content in the artist’s Contra* podcast interview with Aimi Hamraie.

47 This theory was first proposed by Bauman and Murray via ASL presentations at Gallaudet University in 2009 and published in 2010. See Bauman and Murray.

48 On these debates among description professionals, see Hutchinson and Eardley 45 and Perego 120.

49 On a contemporary analogue, see Bauman and Murray’s discussion of the decision by the government in Oaxaca, Mexico, to invest “$4.4 million to install 230 security cameras that are monitored by deaf and hard-of-hearing police officers” (xxiv–xxv).

50 See, for example, Christof Migone’s discussion of “The mouth [as] the meeting place of the sacred and the profane; sacred texts are salivated by the mouth’s viscosity; the Word is born in a cavity that tears, chews, licks, spits” (67), or Barthes’ description of the grain of the voice as “something which is in the cantor’s body, brought to your ears in one and the same movement from deep down in the cavities, the muscles, the membranes, the cartilages” (Migone 181–82).

51 In his 1926 history of the motion picture, Terry Ramsaye presents a teleology of film’s sophistication through its incorporation of printed titles that had previously featured in magic lantern shows, writing that: “As one follows the evolution of the screen art the increasing importance of the relation of the word becomes apparent. Only the primitive pictures required no titles. Pictures remained primitive until they got titles” (qtd in Nagels 367). On the prior familiarity of titles, see Sopocy.

52 See, for example, Downey’s discussion of how the Ninth Circuit District Court in November 1981

ruled that because of the high cost of closed captioning decoders, and the (statistical) lower economic resources of the hearing impaired, only open captions met the definition of “equal access.” This striking judgment did not last, however, and was overturned entirely exactly two years later (with further appeals declined by the Supreme Court). (204–05)