Tables and—

 

[Image description: A color video still of an Asian woman with dark, sharply cut bangs and wire-rimmed glasses, wearing an ice-blue long-sleeved top and black pants, centered against a plain gray wall. She looks directly at the camera with a sharp, almost skeptical expression, her nose crinkled, her lips upturned in a slight but hard-edged smile. With arms neatly bent, she tucks her hands behind her back. The pink-toned arms of a second person, wearing rolled-up white sleeves, push through the crooks of the woman’s elbows. This right arm is lifted to the height of her collar bone, pointer and middle fingers pinched together against thumb, while this left arm hangs downward, its hand out of frame.][1]

Christine Sun Kim and Thomas Mader produced the two-channel video Tables and Windows in 2016. I came across the work nearly a decade later. As installed, the two, wall-mounted flatscreens are slightly offset, the lower lefthand one accounting for tables (“Weathered Beer Garden Table with Splintery Surface,” “Concrete Outdoor Table with a Built-in Chessboard”), the higher righthand one accounting for windows (“Small Cracked Attic Triangle Window,” “Bus Window with a Smudge of Greasy Hair”). Such phrases appear in a sans-serif font—white, all-caps, briefly laid over the moving-image contents. In both cases, these text screens introduce, and separate, brief episodes in an improvisational comedy game known as “Helping Hands,” whereby one person stands behind another, lending their arms and hands to the person in front. The same game, yet the two screens repeat distinct loops: At left, Kim’s face, shoulders, torso, literally re-armed by Mader, describe each kind of table in American Sign Language; at right, Mader’s front, Kim’s arms, describe each kind of window. The same game yet split again: Kim is a native signer, while Mader is not. Nor are the two screens meant to remain in any kind of formal lockstep, either, for the loops run marginally different lengths—fifteen tables in just over nine minutes, fifteen windows in just over ten.

There is, too, a slight misalignment between tables and windows to start with. I keep wanting to type “tables and chairs,” “doors and windows,” or “walls and windows.” Tables are horizontal surfaces for support, windows are openings for letting light and air in, though some of Kim and Mader’s tables seem wobbly, some of their windows grimy, or resolutely shut.[2] To a certain extent, each of their written English phrases is descriptive enough, on its own, to extrapolate that much. More accurately, there is a specificity to each—materials stated, environments and contexts implied, but not described. Kim and Mader do not translate these several written English words at a time into ASL. Rather, they set and spatialize a scene for every table or window, rendering description palpably narrative even for someone, like me, who does not know sign language. This next distinction, then, is fundamental to the work, and active within both its channels: Sign language teacher Andreas Costreau has observed how non-Deaf students struggle to describe, in detail, rooms and the objects within them, in part because ASL is an embodied language, both manually and facially expressive.[3] Together, hand signs and face movements (especially of the eyebrows and lips) communicate, for instance, an object type, its size and shape, as well as where and how it might sit; how and across what distance a person might approach and use it; its tactile qualities.

Once, in graduate school, a professor advised that to describe a thing is not the same as to describe the experience of a thing—an admonishment that has haunted me ever since. Or, I took it to heart, but now it haunts me. Watching Tables and Windows, I grow increasingly aware of distinctions and misalignments, of splits in knowledge and understanding, but also I see a thing and an experience of that thing described simultaneously. To offset my own formulation from my professor’s (their table, my window?): Description and experience are ideally imbricated, and thoroughly co-implicated.

 

Not-a-Glossary

 

Image description and audio description share a joint purpose, to verbally communicate visual artworks, performances, or events for audiences who are blind and low-vision. An image description might appear online, to be scanned by a screen reader, while an audio description might be a spoken recording of that image description, as for a museum audio guide stop; or interpolated into a television program’s breaks in dialogue; or provided live in a theater or gallery setting. These forms might also be called simply verbal description, or visual description. “Alt text” is a short visual description for web images specifically, often limited to 125 or so characters—for people who use screen readers, for people with unstable Internet connections, for when images fail to load, and for search engines to “read” as well. Purposeful forms yet under-used, under-imagined: Georgina Kleege has invoked a future in which audio description escapes “its current status as a segregated accommodation outside the general public’s awareness and launch[es] into…a literary/interpretive form with infinite possibilities.”[4]

Captions transcribe spoken language and (though far less frequently, far less adequately) other sounds—in a work of video art, a movie, a play, a talk or meeting—to be read on a screen by audiences who are d/Deaf and hard-of-hearing. Closed captions, when available at all, can be turned on and off. Open captions are a fixed feature, permanently embedded. A number of studies point to the rising use of closed captions, in particular, among younger audiences. “People both with and without sensory disabilities consume media in multiple modalities,” Kleege confirms.[5]

Creative captioning refers to captioned content—which, in the most comprehensive cases, includes lines of dialogue, sound descriptions, and the transcribed words of an audio describer—that has been formally, affectively, thoughtfully integrated into the field of a given video screen or stage set. Elaborated by artists, creative captioning diverges wildly, too. As Mara Mills and Neta Alexander recently put it, “Freed from its status as a device for ‘mere’ transcription of preexisting sound, this type of captioning now often references conjectural sound or acts as a spur to thought and performance.” They go on to elucidate “captioning” as an umbrella term (alongside or as synonymous with an umbrella term like “disability text”), and one that signals both an ethic and an aesthetic: “In disability art, captioning—whether literal or figurative, even calling the use of any text ‘captioning’—signifies disability and disability aesthetics…as opposed to ‘inclusion.’” In sum, “Intimacy, justice, or eventfulness supplant access as the intention of the captioning.”[6]

Around 2014, Jay Dolmage proposed all supposedly access-motivated language materials—“ASL interpretation, visual description, alt tags within websites, and other ‘accommodations’”—as kinds of ekphrasis. (I will add that etymology could support a particularly strong link between ekphrasis and audio description—in the ancient Greek, “to speak out,” describing an object for a hearer.) Ekphrasis “reveal(s) the essential differentiation and overlap of sensory engagements.” It “add[s] artistic and rhetorical value,” and, Dolmage concludes, “it is the spirit of ekphrasis that matters more than its accuracy.”[7]

This isn’t to devalue accuracy as such. An accurate description’s value is in the usable information it gives. As the title, alone, of Joseph Grigely’s ongoing series Craptions connotes, a bad transcription can be worth less (resulting artwork excepted) than no transcription at all. I would say the same for description, partly because eliding “description” and “transcription” can appear to bolster the former.

[ID: “A screenshot of the YouTube web browser shows the face of a male singer. He has short black hair that is receding at his forehead, and a full beard. The collar of his shirt is white. He sings into two gray microphones. His mouth is wide open and his eyebrows slightly furrowed. He looks ahead. His face, microphones, and collar are lit. Everything else is dark. On the lower right of the video is a label in white. It reads: ‘subtitles/closed captions unavailable.’ On-screen text below the video reads: ‘Caruso bracket the most powerful end bracket.’”[8]]

But Dolmage’s aim, like mine, is to revalue description—not least during this period of its radical revaluation as computational data. Shortly after the publication of Dolmage’s book, Mills made plain the brief, potent history of “visually descriptive text…on the web” being used to train machines in recognizing, and modeling, correspondent image structures. “Ekphrasis,” she writes, has been “streamlined and operationalized.”[9]

Description is central to the labor of art history yet regularly dismissed as “merely” formal. Not all description is ekphrasis, but this is another elision I’m keen to lean into. Dolmage uses the word within the broader context of rhetoric and its tools. I use it for its ostensible exaltation, its rarefied status. After all, ekphrasis is assumed to mean a literary description of an artwork, a flexing of verbal powers in response to, or in the face of, visual ones; formal in its effects and by design. Other assumptions include: Either ekphrasis translates one medium into another, or it inevitably, crucially, fails to do so, in which case it highlights instead what is untranslate-able, the fact of untranslate-ability, between artwork and word.[10] 

Failure is, paradoxically, key to any revaluing. Description not only can but should prove a means, or at least a measure, of both accessibility and of inaccessibility. In this way—and pursuant to the artworks in this essay, as well as the scholars cited—I see “ekphrasis as a mode of disability justice,” precisely because of the systemic failures that demand it.[11] To be sure, so-called accessibility accommodations hardly amount to disability justice, a movement that affirms intersectionality and equity as necessary to combatting ableism. Disability justice also contravenes productivity as the best (capitalist) metric for a life lived. The potential for failure that is core to ekphrasis, and its frequent depreciation as device rather than method, are apropos, where description’s criticality and subjectivity can dispel expectations of its sheer productivity, too.

 

“A blue”

 

[ID: A vivid blue rectangle, horizontally oriented, with two text captions of identical length positioned near the bottom edge of the frame. The upper caption, in a simple, yellow-tinted font, reads, “A blue of implausible density, dragging the regard.” The lower caption, in a simple white font, reads, “For Blue, there are no boundaries… OR solutions.”][12]

Blue (1993) is arguably Derek Jarman’s best-known work, as well as his last, made in the months before his death from AIDS-related illness. For the duration of its 76 filmic minutes, the screen comprises a blue monochrome, populated by audio: narration, voiced by Nigel Terry, John Quentin, Derek Jarman, and Tilda Swinton; threads of music, composed by Simon Fisher Turner; ambient as well as contextual sounds, from hospital-room machinery to clicking footsteps. To the extent that the narrator narrates, this is the story of Jarman’s hospitalizations on and off during 1993 with cytomegalovirus retinitis, which affected his vision. But this is also the story of Blue, personified by Jarman as a character, and it is an elegy for Jarman’s beloved friends who died in earlier years of the AIDS pandemic. Either gradual or sudden shifts in tone are paramount—from clinical accounting to lament to screed, from observational to dreamy to sorrowing to frantic, from mundane to surreal, and back again. Blue is the constant, and Jarman refers to the color repeatedly in universal or oceanic terms, sweeping, grandiose, yet always resolutely unfixed.

Commonly described as an imageless feature-length film, Blue proffers, on the contrary, nothing like a blank or an empty screen. Above all, it eschews any single or standardized format, from the start a polyvalent work, motivated by its variety of sensorial registers. Jarman published the script as a book chapter; it was simulcast on BBC Radio 3 and Channel 4 television; radio audiences could send away in advance for a blue postcard. Commonly vaunted, as well, within queer history, Blue is also “a foundational Crip work,” as Christopher Robert Jones asserts, “foreground[ing] an intersectional analysis of a sick/disabled experience.” While most writing on Blue connects its infamous blue monochrome to loss or lack—of sight, of image—Jones argues instead for “Blue as a site [homonym perhaps intended?] wherein Jarman developed a visuality that is expansive and transcendent,…deeply embodied and imperfect.”[13]

[ID: A photograph of a darkened movie theater, looking toward a large projection screen. This screen is a vivid blue rectangle with two rows of text captions near its bottom edge. It is flanked by, on the left, the small, fuzzy green light of an egress sign, and on the right, two spotlit white people backed by a black curtain. The two people face outward, angled toward an audience that must be, in turn, facing the screen. Their elbows are bent, arms raised in front of them, fingers outspread. The taller, righthand person has their mouth open, stretched into an O. On the screen, the upper caption, in brackets and pale, grayish font, reads, “[deep catacomb organ, two-toned, insistent, repeating].” The lower caption, in a thicker white font, reads “As they set out for the indigo shore, under this jet-black sky.”][14]

In 2024, artists Liza Sylvestre and Christopher Robert Jones, along with scholar Sarah Hayden, debuted the Blue Description Project, a collaborative reimagining of the “immensity” of Blue—a collaborative re-inscription of that earlier work’s sensorial capaciousness. Although Blue appeared from the start in published, text-only forms, its screen-based forms were not captioned. Therefore to watch Blue meant also to listen to it but not to read it (at least until its eventual digitization began to activate certain closed-captioning options). Moreover, Jarman’s script transcribes the work’s narrative track but does not include nonverbal sounds, such that to read Blue did not capture either its visual or its fuller aural aspects. Beginning in 2023, Sylvestre and Jones commissioned, gathered, edited, and ultimately interwove what can be called, collectively, creative captions for the feature-film version of Jarman’s work. Screenings also include sign language interpreters. Yet the Blue Description Project is not simply Blue-rendered-accessible to listeners who are blind or low-vision, readers who are d/Deaf or hard-of-hearing.

The artists worked closely with audio describer Elaine Lillian Joseph to organize and refine their understanding of how much cumulative time they had to work with—how they could best operate within Jarman’s sonic breaks, so as not to pause or extend his film nor alter its particularities of flow. Corvyn Dostie wrote captions for Jarman’s narration, not lifted directly from the published script but rather from attending to his own experience of the film version, including indications of tone or quality of voice, volume, emphasis.[15] Joseph initiated evocative sound descriptions for the music and effects in Jarman’s film work (“welcoming wobbling guitars amble in and spread to fill the room”; “the twanging of a tropical vacation going stale and cool”), also clarifying an alliance between the occasional character of Blue and a recurrent chime. As for visual descriptions: Sylvestre and Jones solicited submissions from dozens of people—friends, students, workshop attendees—requesting that they “please describe the visuality of Blue.” They then collated, ordered, and shuffled the submissions thematically, selecting locations for them as both text captions and recorded descriptions. These captions appear in yellow (distinct from the white captions of the narrator, and the pale blue, bracketed captions of the descriptions for nonverbal sound), and each begins in the same way:

 

A blue like a velvety night blanket.

A blue, bitter pill. An open mouth.

A blue that feels stable and alive, like wood.

 

At the start of Blue Description Project, a two-minute captioned introduction (against a black screen) explains the work’s structure, and how it will unfold. This introduction also defines visual description as “a gesture”—meaning, the conversion of visual experience into text—“that creates slippages, or gaps, and inequivalences between the primary document and the text description.” Sylvestre and Jones’s artwork tends to (still more, emerges from) those gaps, and in keeping, Sylvestre has illuminated a three-dimensional conception of the project: Jarman’s steady blue “gave us a floor to stand on for thinking about description”; his Blue is “generous—not a closed room” but an open one.[16]

 

*

 

So, what can I describe, or in some cases effectively transcribe? As a sighted and hearing person, watching Blue Description Project on my laptop, with sound on: I felt entranced, sunk in, carried simultaneously by Jarman’s variable pacing and by Sylvestre and Jones’s incisive cuing to and enhancing of that pacing. I felt also, at moments, overwhelmed by how much I wanted to take in—wanting, I mean, to read the sound and visual descriptions and the narrator’s captions as they appeared, while also listening along to all three. Now and then the captioning slipped off screen before I could finish reading. Despite Sylvestre and Jones’s conviction about working within the constraints of Jarman’s 76 minutes, I did pause my video link from time to time, in order to jot down notes or take a screenshot—for while I could not travel to a public screening of the work, through my art-world network I gained the privilege of repeated access from home.[17] My notes and screenshots accrued:

* Instances of Jarman’s own variously associative or descriptive language for Blue:

“I step into a blue funk.”

“Blue Bottle buzzing.”

“Like a blue frost, it caught them.”

* Moments when, for example, the newly collected descriptions augmented or countervailed that frame’s sonic track to especially keen effect:

“[music box returns in force, overlapping with itself, ever louder]”

“A blue that starts to hurt my eyes — I want to look away! Or close my eyes completely!”

“[round and round teacup spin]”

“A blue that fills the screen with a subtle movement on loop, in the corner, like a mesmerizing fire, or the repetitive movements of waves.”

* Instances of the new descriptions ascribing polarities, this blue as both ends, or both/and:

“A blue like a voice coming from another dimension, full of grief…and also faggy, foggy delight!”

 “A blue through eyes open… and closed.”

* Moments when Sylvestre and Jones’s choices diverged from Jarman’s language, or showed the new descriptions contradicting one another:

After Jarman’s narrator invokes “a timeless ocean,” this yellow caption appears: “A blue that is not an ocean blue. It lacks the green of phytoplankton and the brown of sediment.”

Some fifteen minutes later, as assertively: “A blue that is like the ocean – not quite the surface, where the water is warm, but a bit deeper… where the light still touches, but it’s darker.”

There is such care-full precision in both of these, technically opposed, descriptions. Color largely aside, the language and sounds of water do abound in Blue, and in the Blue Description Project, from Jarman’s invocations of a “cobalt river” or ocean crossings or sub-marine sequences, to soundtrack elements of crashing or lapping waves, foghorn-like calls, whale-like songs, “[ribbiting-rippling, peaceful, full moon lake]” and “[oceanic distortion],” to “A blue watching the darkest blue waves, from the shore.” In the visual descriptions, blue operates as multiply, however, as it also does for Jarman—as liquid, as light, as cloth; as warming or cooling (“A blue that LOOKS like a cold night when you’re well-dressed FEELS.”); as hyper-specific (the blue of accessibility signage, or of Parliament-brand cigarette packages) or as shapeless, possibly generic (“just blue”).

Throughout, variation is grounded by reiteration. A fair number of the visual descriptions suggest movement, a flickering, a wavering, a pulsing. Even conspicuous repetition evinces no definitive, objective experience of this blue, but rather a sense of collective potential, that description here—and maybe elsewhere—is a joint or shared, as well as profoundly subjective, endeavor. This same sense of being part of a community of describers prevailed when I caught myself shaking my head not only in appreciation but also disagreement. I was tickled by certain natural-world comparisons (“A blue dappling poison dart frogs!”), yet personally unpersuaded by this blue as either sky-like or ocean-like in its hue.

“A blue that is not Dell, NOT Facebook, not Twitter, not Vimeo, not IBM, not HP, not Skype, not PayPal…none of those,” an early caption reads. Yet, “A blue on the phone in a car, in the dark, in the rain.” For me: A blue inseparable from screen, from glow, from the “blue light” of the computer on which I have viewed it. As it happens, the only other time I had seen visual frames from Jarman’s Blue before this was within Moyra Davey’s video work Notes on Blue (2015), in part of which the artist watches a closed-captioned Blue on her phone while riding the subway.

[ID: A color video still, camera zoomed in closely to capture two open, white palms and fingertips slung beneath a black-cased iPhone, held horizontally. (It feels easy to extrapolate that the person represented by these hands sits hunched over their phone, looking at it, while a camera peers over their right shoulder to look at it, too.) The phone’s screen shows a vivid blue rectangle framed by black. Near the bottom of the blue rectangle, two lines of white text captions are too blurry for me to make out, though I can strain to see that the last words read “dead friends.”][18]

Davey’s video demonstrates her shifting relationship to Blue over time—remembering her previous viewing of Jarman’s work twenty years before, then rewatching it in her present tense of 2015. Meanwhile the Blue Description Project makes possible an encounter with description (with other people’s descriptions, with other people’s encounters with Blue and blue, with the vagaries available within or through a single color) that is far from secondary either to a thing or to the experience of that thing. Far from secondary, not even primary, but in fact constitutive. The closing credits for the Blue Description Project list one anonymous and thirty-five named describers. Many more collected descriptions didn’t find their way in: “We could make a whole other film,” Sylvestre says. “There’s no end to the descriptions. We would never finish describing blue” [Blue? Blue?].[19]

“A blue, without landing.”

“A blue, as… leeway, permission, without a gatekeeper.”

[ID: A screenshot of a vivid blue, horizontally oriented rectangle, about the size of an index card, with two text captions near its bottom edge. The upper caption, set between squared brackets in a plain, pale blue font, reads, “[the sound of Blue re-beginning, across itself],” while the lower caption, in a plain, yellow font, reads, “A blue… afterimage.”][20]

 

Before-images, Afterimages, part 1

 

“In a culture overwhelmed by (mostly digital) photographs,” David Horvitz states, “I decided to start erasing my photographic archive.”[21] Since 2018 the artist has erased thousands of his personal snapshots, sometimes documenting these deletions—under the project name Nostalgia—in book form. Each black page includes, in white sans-serif font, a short description of the image, framed by the combination of letters and numbers that once named the digital image file and by its date of capture.[22] Unlike the insistent, immersive blue field of both Blue and Blue Description Project, Horvitz’s black fields are literally image-less, his photographs deliberately displaced by visual descriptions. Bent toward conclusion, I have gathered a few formally consonant works that nevertheless posit an array of relational tensions between monochrome and text elements, each of them either loading or suspending the temporal sequence of what comes first and what comes after.

That said, temporality always matters. At the time of writing, in summer 2025, it feels impossible to think about visual description without reference to NO-PHOTO, an anonymous international group of activists and artists. They do not delete photographs so much as redact them, and specifically photographs, taken by Palestinians, of Israel’s genocide in Gaza. On the one hand, ours is a “culture overwhelmed by (mostly digital) photographs.” On the other, Western media outlets have largely censored photojournalism from Gaza by persistent acts of omission. Rather than proliferate those images themselves, NO-PHOTO’s own persistent action—in Instagram posts, physical wheat-pasted posters, and a newsprint publication—has been otherwise: placing a black rectangle side by side with a description of the image thus blacked-out. Both posters in these diptychs include the name of the Palestinian photographer, and the descriptions, as Aruna D’Souza writes of them, “hover…between ekphrasis and poetry. ‘A baby’s hand, their tiny fingernails peeking through grey dust, like pearls,” reads part of one.’”[23]

The political potency of ekphrastic description made manifest. Even so, description, ekphrasis, the “merely” formal or “merely” rhetorical, their pleasures as well as the appreciation of their failures, might seem (maybe are) insufficient to the task of meaning-making while systems of violence render pleasure—creativity, imagination, let alone dailiness—impossible for so many. There is little way, now, to smooth a transition into the remainder of my essay, so I must let it be jarring. As we have seen, the investment of time that both the processing of description and the act of describing entail gauges disjunction as well, and time is one thing that I can currently give.

 

Before-images, Afterimages, part 2

 

[ID: A photograph of a white-walled gallery with a parquet floor. The camera angle centers on a corner of the room, and artworks line both walls, three on the left, two on the right. These artworks are identically sized horizontal rectangles, entirely flush with the wall, each in a different solid shade of brown or gray. Every color field centers a shorter or longer block of text in a contrasting color. The work nearest to the camera is most legible. In pale gray against a reddish, earthy brown, its words read, “Closeup faces of happy multiethnic children embracing each other and smiling at camera. Team of smiling kids embracing together in a circle. Portrait of young boy and pretty girls looking at camera.”][24]

During the first fall and winter of the COVID-19 pandemic, a conducive gallery at the University of Hawai‘i at Mānoa rotated its exhibitions within a three-sided space that has a fourth wall of plate glass, artworks fully visible from outside, whether or not the gallery could open. For one of these, Stephanie Syjuco produced her series Diversity Pictures (2020), sourcing images and their available captions from stock photography websites via the search term “diversity.”[25] Each artwork features the text of one caption, seemingly in place of its image. Yet the image is not entirely absent, for the singular shade that now backs each text—a warm gray, a medium brown—is in fact a Photoshop amalgamation of all the original photograph’s colors. In the instance above, then, this means the reddish, earthy brown merges not only every skin tone and hair color of the “happy multiethnic children” once pictured individually, but also the colors of their clothing, and of any background or surrounding environment, none of which the as-found caption indicates, in any of its three stand-alone descriptions.

Thus Syjuco’s monochrome field performs the same gloss as does the caption, but it does so pointedly. Put another way, by combining differential visual data into a single visual note, she processes an image of a supposed “melting pot.” I’ll try a third time, myself, too: Syjuco’s monochromes are not just backdrops for text. Instead, they enact, and activate, the blithe genericism of their captions, which themselves are powered by presumption rather than by actual, descriptive information. Moreover, by “captions,” for Diversity Pictures, I mean something different from elsewhere in this essay: These are not transcriptions of anything aural, nor are they alt-text, or another type of “disability text,” so much as they are keywords with marketing and search-engines in mind. Syjuco’s focus on them draws out all manner of latent tonalities, their language at once careless, absurdist, sinister. They appear authorless—“stock” like the photographs they once accompanied—yet the extent to which they inspire mental pictures, for me, of exactly such stock photography, brings home how much I have absorbed both kinds of language, verbal and visual; how much I have internalized them.

 

*

 

[ID: A photograph of a pristine, light gray wall, taken head-on as if to emphasize the wall’s flat expanse. In the lower right corner of the image, a blocky shelf, also painted gray, pushes out from the wall, supporting black headphones perched on a black stand. The shelf casts fan-like shadows on the wall below it, and a black cord trails downward vertically from it, leading out of frame. What might be a black remote or volume control rests beside the headphones. On the other side of the image, a small white rectangle floats near the lefthand edge: a museum wall label that is only long enough to include a title, medium, and dimensions, though none of those text details are readable here.][26]

For his 2025 exhibition at Illinois State University, Andy Slater expanded his Invisible Ink series, installing eleven works from it in all. To install each one means: on one side of an apparently blank patch of wall, to wire and mount headphones through which a hearing visitor could listen to a recorded audio description; on the other side of the apparently blank patch of wall, to print and affix an object label with “tombstone” information for the audio-described artwork. As for the blank patch in between, in previous iterations it has been just that—undifferentiated wall. At Illinois State, Slater opted to paint a rectangle in the dimensions each audio description and object label supply, but he did so using a high-gloss version of the gallery’s standard wall color. The resulting area of sheen was visible to sighted visitors from certain angles of light and viewer position. The same, resulting area of smoother, more resinous paint was discernible to anyone who touched it.

Despite the pretense of the tombstone details for each, which claim them to be materialized more or less traditionally—“oil on canvas,” “gel pen and stencil on paper”—these are audio works first, available as well in written form, for those who need or prefer.[27] Which is to say, the relative blankness of the wall, the physical gap between headphones and object label, is not the work. Slater’s texts describe figural compositions in rich detail, two self-portraits and an assembly of complex, enigmatic sci-fi vistas or scenes. Read aloud, in a steady, resonant voice by Solon Kelleher, the shortest audio description runs about 90 seconds, the longest about five and a half minutes. If description is this artist’s medium, then it is not so only materially, categorically, structurally, but also metaphorically: Slater’s mind is evidently rife with images (also objects—paintings, drawings, mixed-media works), and in this series, he seeks to transfer or transmit them to each auditor or reader, one by one, one on one. Yet of course no such transferal occurs one to one. While either listening to or reading Slater’s descriptions, my mind can only construct the image best available to it, based on my own associations, my own proclivities, my own pre-existing image bank.

 

*

 

Speaking of mediums: As with all but one of the works in this essay, my experience of Slater’s Invisible Ink series did not take place in person, in a gallery, but was mediated by my computer. I will confess, too, that I read his descriptions first, before listening to their recorded versions. The last of these works, Your Turn (2024), rather than replicating one descriptive form in another, offers (is, I would argue, about) divergent experiences for reader and for listener. Namely, the audio description for Your Turn is among Slater’s longest, just shy of five minutes, while its written text version is by far his shortest. “Stencil on wood,” twenty-two by eleven inches. “On a muddy brown panel, bright purple text set in Sloan font reads:

 

‘DOYOUSEEWHATISAYTHATIHEARIFSOTHENTELLTHESKIESBEFORETHEYREDARK.’”

 

The audio recording delivers this text all the way through twice, adding (like a wink) the word AGAIN to the end of those two spaceless, all-caps sentences. Kelleher’s voice also discharges those two spaceless sentences letter by letter, as if spelling out a single, interminable word, the first time through with a marked pause between each letter, the second time through with a bit more speed, for that is how most screen reader technologies register all-caps writing—as an acronym, of sorts, instead of as a sentence.

 

*

 

[ID: A photograph of a dimly lit room, with dark gray walls and a gently scuffed, rubbery looking black floor. Two wooden benches with seat and back cushions bracket the space, one on the left side in the foreground of the image, the other on the right side in the background. The bench on the left faces a trio of blocky black television monitors placed directly on the floor in what appears to be the center of the room. Their screens are black, nearly the same color as the floor, except for where each casts a bluish-white pool of light downward onto it. A white caption sits at the bottom of each of the three black monitors, and each is a slightly different length, so presumably these are three different texts—though they are too blurred in this photograph to make out. Beyond the monitors and benches, five vertical, light gray acoustic panels occupy the long back wall, evenly spaced, large but not distracting.][28]

Carolyn Lazard has included descriptive texts, among other kinds of creative captioning, in their moving-image works for a number of years. Long Take (2022) is the first installation in which Lazard tenders visual description without corresponding imagery: The video monitors remain largely dark (though also aglow), only text captions shifting from frame to frame. Lazard shared a movement score with Jerron Herman, Herman interpreted the score for a video camera, and Joselia Rebekah Hughes wrote audio descriptions of that performance. In the gallery, a hearing visitor could listen to the original movement instructions, the sounds of Herman’s breath and body as he danced, and the audio descriptions of him dancing; while a reading visitor could follow the monitor-based text captions for each of those layers.

Lazard has spoken about Long Take, among other works, in terms of a “Black krip ethic or methodology”—the captions that are central to their installation providing both “access and opacity,” both “visibility and invisibility”; the technical withholding of visual imagery an enhancement not only of the captioning overall but also of “opacity” and “invisibility” in particular.[29] Importantly: Description here does not reverse or interfere with obfuscation, operating instead hand in hand with it—for description’s prospective inadequacy is its non-transparency, its discrete materiality, or materialities. Nor does Lazard’s interest in withholding rebut sustained engagement. Indeed, certain other descriptions extend Long Take: no longer installed, yet available online via an audio description written and read aloud by Elaine Lillian Joseph; or via an extensive Word document describing the exhibition experience spatially and sensorially, including full transcripts of the layered audio tracks.[30]

 

*

 

This essay could be endless—a lovingly compiled running list of artworks. There are further striking examples of artist-driven description to describe, and clearly the works themselves beget still more description. Describing artworks is typically how I begin a piece of writing, though rarely where I feel able to end, given the conventional mandate that would separate description from argument. Then again, description alone can accomplish an astonishing amount of conceptual and interpretive labor, as all of these artworks have shown.[31] But to enforce an end now demands reflecting on the ekphrastic experience, by which I mean both the description-led experiences empowered by this diverse group of artwork examples and the experience of being a describer. It is through the latter that these different works speak to each other, for they impel a deep attention to the range of forms and formats description arrives in, and the sensory implications of each.

They impel deep attention, as well, to the felicities and inefficacies of describing artworks to readers, and of describing—or at least making explicit, offering up—my inherently subjective experiences of them. As mentioned already, I have encountered most all these works on my laptop; certainly I have written about them on my laptop; presumably you are reading these words on your own electronic device. I am accustomed to declaring a preference for in-person artwork encounters, nor do I quite disavow that position. Yet properly accounting for works that center “disability text,” applied as widely as possible, requires me to take description seriously as a first-hand encounter in itself, and really one can be “in person” with description from anywhere.

Regarding “persons”: These closing paragraphs are first-person heavy, especially for an essay aiming to focus on the artworks, and often the ekphrastic acts, of others. Disparate though they are, the works allow—insist on—a describer’s and a receiver’s subjectivity because they likewise allow descriptive practices as collective, collaborative, relational (even when resistant). I suppose, once more, the artists and thinkers I have followed throughout help me embrace ekphrastic action and ekphrastic experience as two sides of a coin. And I suppose what I hope to test is the applicability of these fresh ekphrastic circuits within art-historical writing more broadly. Finally, I suppose I am resisting an ending, too, because I know that were I to begin this essay all over again (“now” as opposed to “then”), I’d describe these works some decisive bit differently—ever closely, ever insufficiently, every time. 

 

---

 

The author would like to acknowledge the contribution of Georgina Kleege, who served as access-oriented consultant on an early draft of this text. Kleege is Professor Emerita of English at UC Berkeley. Her most recent book is More Than Meets the Eye: What Blindness Brings to Art.

My thanks as well to Hallie Ringle, Troy Sherman, Andy Slater, Liza Sylvestre, and Gan Uyeda for past and recent discussion or correspondence about many of the artworks included. Sara Greenberger Rafferty has also been a meaningful interlocutor, for a few years now, on the topic of image descriptions as a form of art-historical analysis in themselves.

 

---

 
[1] For the image described, see the second frame in the carousel at <https://www.sfmoma.org/read/christine-sun-kim-and-thomas-mader/>.

[2] Each signed description concludes in the same way for all of the tables—pointing to where the horizontal surface might be, then folding arms as if to rest upon it—and for all of the windows—pointing to where the vertical pane would be, then sliding stiffened, parallel hands up and down as if against it; or as if the lower hand were the sill, the upper hand the sash.

[3] This reference to Costreau appears in numerous didactic texts on Tables and Windows, and emerged from personal discussion between him, Kim, and Mader.

[4] These are the closing lines of Kleege’s chapter 7, “Audio Description Described,” in her book More Than Meets the Eye: What Blindness Brings to Art (Oxford: Oxford University Press, 2018), 108. Related for its dedication to literary and interpretive expansiveness, one of my own earliest engagements with any of these descriptive forms was through the workbook and website Alt-Text as Poetry, by artists Bojana Coklyat and Finnegan Shannon. See <https://alt-text-as-poetry.net/>.

[5] Georgina Kleege, “Fiction Podcasts Model Description by Design,” in Mara Mills and Rebecca Sanchez, eds., Crip Authorship: Disability as Method (New York: New York University Press, 2023), 324.

[6] Mara Mills and Neta Alexander, “Scores: Carolyn Lazard’s Crip Minimalism,” in Film Quarterly (2022) 76 (2): 39–40, <https://online.ucpress.edu/fq/article/76/2/39/194938/ScoresCarolyn-Lazard-s-Crip-Minimalism>. In addition to Mills and Alexander’s concise account of “captioning” and related terms, see as well Emily Watlington’s excellent essay “Critical Creative Corrective Cacophonous Comical: Closed Captions,” in Mousse 68 (6 June 2019): <https://www.moussemagazine.it/magazine/critical-creative-corrective-cacophonous-comical-closed-captions-emily-watlington-2019/>; and the Activating Captions online magazine produced by ARGOS centre for audiovisual arts in Brussels as part of their 2021 exhibition, <https://visit.argosarts.org/activatingcaptions/magazine/caption-culture>.

[7] Jay Timothy Dolmage, Disability Rhetoric (Syracuse: Syracuse University Press, 2014), all p. 140, except for the last quotation, which is from p. 139. In 2012, Susan Schweik led “a creative writing/poetry workshop…exploring the relations and differences between traditional modes of ekphrasis and contemporary modes of audio description,” at the University of California Humanities Research Institute. See Catherine Kudlick and Susan Schweik, “Collision and Collusion: Artists, Academics, and Activists in Dialogue with the University of California and Critical Disability Studies,” in Disability Studies Quarterly 34, no. 2 (2014): <http://dsq-sds.org/article/view/4251/3609>.

[8] Grigely showed this and three other, similarly screenshot frames as the multi-part work Craptions, each printed and framed in black to hang on the wall as part of his 2023 exhibition In What Way Wham! (White Noise and Other Works, 1996–2023) at MASS Moca. Whereas the other bracketed image descriptions in this essay are my own, here I have quoted the description of one of those Craption shots from the terrific texts provided for the works in that show—compiled and edited by Andy Slater from drafts by Tressa Slater, Kat Germain, Iris Xiu, Meesh Fradkin, Denise Markonish, Tess Davey, and Pierre Von-Ow. The full transcript and audio description for Craptions, and all descriptions for the exhibition, remain available online at <https://massmoca.org/grigely-descriptions/>. Grigely occasionally posts new “craptions” to his Instagram and published a number of those posts as “Craptions: Instagram notes from Joseph Grigely,” in a special issue on “Translation,” Refract: An Open Access Visual Studies Journal, vol. 2, issue 1 (Fall 2019): 17–30, <https://escholarship.org/content/qt0qk2z1c2/qt0qk2z1c2.pdf?t=qb3syo>.

[9] Mara Mills, “Listening to images: audio description, the translation overlay, and image retrieval,” in The Cine-Files 8 (Spring 2015), <https://www.thecine-files.com/listening-to-images-audio-description-the-translation-overlay-and-image-retrieval/>.

[10] While my focus here is on what may be un-translate-able, I’d like to point to Amanda Cachia’s recent book, the first chapter of which proposes a conflation between “description” and “translation,” as terms and as ideas. See Cachia’s The Agency of Access: Contemporary Disability Art & Institutional Critique (Philadelphia: Temple University Press, 2025), pp. 29–66.

[11] I borrow this phrasing—“ekphrasis as disability justice”—from Jennifer Nelson’s syllabus for the fall 2024 seminar Writing about Art at University of Delaware. It was a pleasure to speak with their students during the session so titled. Sins Invalid delivers a valuable list of “10 Principles of Disability Justice”—a term originated by activists Patty Berne, Mia Mingus, and Stacey Milbern in 2005—at their website, <https://sinsinvalid.org/10-principles-of-disability-justice/>.

[12] For the image described, see <https://whitney.org/events/blue-description>.

[13] Christopher Robert Jones, “Blue Cripistemologies: In and Around Derek Jarman,” Art Papers (2021), from a theme issue on “Speculative Masculinities”: <https://www.artpapers.org/blue-cripistemologies-in-and-around-derek-jarman/>.

[14] For the image described, see <https://aemi.ie/accessibility-artists-film-the-blue-description-project-reinterpreting-derek-jarmans-final-film/>. The photograph’s caption on the web page names the two people in it: “Lianne Quigley (Di) and Elzbieta Cichocka interpreting Blue through ISL [Irish Sign Language] at the IFI [Irish Film Institute] on June 5th.”

[15] Funnily enough, the most readily available copy of Jarman’s script for Blue (though again, vocally undifferentiated, and without sound descriptions) appears as part of “The Ekphrasis Series,” published by David Zwirner Books. See Derek Jarman, Blue (New York: David Zwirner Books, 2023).

[16] Conversation with the artist, 17 July 2025. The two-dimensional page (or screen) remains meaningful to Sylvestre and Jones as well. This fitting Jarman line from the chapter on “Blue” in his book Chroma in fact appears on the opening (initially black) screen of the Blue Description Project: “If I have overlooked something you hold precious—write it in the margin” (London: Vintage Classics, 1995), 42.

[17] I am grateful to Megan Heuer, Director of Public Programs and Academic Engagement at the Whitney Museum of American Art, for introducing me to Liza Sylvestre, and to the artist for sharing a viewing link.

[18] For the image described, see the second embedded still at <https://walkerart.org/magazine/moyra-daveys-notes-on-blue-a-post-screening-meditation/>.

[19] Conversation with the artist, 17 July 2025.

[20] In this case, the image described is my own screenshot, of the final (pre-credits) frame of Blue Description Project.

[21] Horvitz quoted at <https://jeankentagauthier.com/en/expositions/presentation/144/nostalgia>.

[22] For the book’s third edition, with some page spreads imaged, see <https://www.printedmatter.org/catalog/62490/>.

[23] Aruna D’Souza, “NO-PHOTO 2025, at Les Rencontres de la Photographie 2025, Arles, France, through October 5, 2025,” in “Critic’s Picks: Exhibitions of the Summer,” 4 Columns (8 Aug 2025), <https://4columns.org/columns-4/critics-picks-exhibitions-of-the-summer>. Similar language appears in another account of NO-PHOTO’s project, too, Sunil Shah describing each description as “…an ekphrasis of sorts but really a rendition of a poetic alt-text” (see Shah, “No Photo 2025,” American Suburb X [21 July 2025], <https://americansuburbx.com/2025/07/no-photo-2025.html>).

[24] For the image described, see <https://stephaniesyjuco.com/content/7-projects/15-2020/2-diversity-pictures/syjuco_s21_commons-7_web.jpg>.

[25] This installation ran from January 17–February 28, 2021, as the third “chapter” in a three-part exhibition called Out of the Camera: Beyond Photography, curated by Maika Pollack.

[26] For the image described, see <https://www.flickr.com/photos/52024644@N05/54281618843/in/album-72177720323335462>.

[27] In the gallery, a QR code for each work led visitors to the written transcript, as needed or desired. The audio recordings and their transcripts for all eleven Invisible Ink works remain available at <https://galleries.illinoisstate.edu/exhibitions/2025/andy-slater/accessible-page.php>. The exhibition itself—Andy Slater: Paintings & Sculptures, curated by Troy Sherman—ran from January 14–March 26, 2025.

[28] For the image described, see <https://walkerart.org/calendar/2022/carolyn-lazard/>.

[29] Carolyn Lazard interviewed by Edna Bonhomme, in “Carolyn Lazard on Illness, Intimacy, and the Aesthetics of Access,” Frieze, issue 225 (28 February 2022): <https://www.frieze.com/article/carolyn-lazard-edna-bonhomme-interview-2022>.

[30] The short audio description—which clarifies, unlike my own image description above, that Lazard’s installation actually comprises three separate but interdependent artworks—is available on the Nottingham Contemporary website, <https://www.nottinghamcontemporary.org/record/carolyn-lazard-long-take-audio-description/>. The “exhibition experience” description remains downloadable at the bottom of the ICA Philadelphia web page for Long Take, <https://icaphila.org/exhibitions/carolyn-lazard-long-take/>. Carolyn Lazard: Long Take was a co-commission, curated by Pavel Pyś, Olivia Aherne, and Meg Onli, and ran at the Walker Art Center, Minneapolis, February 12–December 11, 2022; Nottingham Contemporary, UK, February 11–May 7, 2023; and the Institute of Contemporary Art, Philadelphia, March 10–July 9, 2023.

[31] On the complexity of both ending and beginning visual descriptions—as well as the meaningful subjectivity it requires throughout—I highly recommend [voices surface]: An Audio Documentary about Accessing Handsworth Songs by Hannah Kemp-Welch, in which Sarah Hayden interviews Elaine Lillian Joseph about her research, writing, and recording processes while she crafted the audio descriptions for slow emergency siren ongoing: Accessing Handsworth Songs: <https://lux.org.uk/voices-surface-an-audio-documentary-about-accessing-handsworth-songs/>. See also Joseph’s “Audio Description Script” and “Audio Describer’s Foreword,” <https://slowemergencysiren.org.uk/audio-description-script-by-elaine-lillian-joseph>, as well as Hayden’s own essay, about ongoing-ness, “Audio Description as Verbal Art,” <https://slowemergencysiren.org.uk/audio-description-as-verbal-art-by-sarah-hayden>.