Saying “Thank You” for Quality Closed Captions: A Promising Shift in Inviting Access
Dire "Merci" pour des sous-titrages codés de qualité : un changement prometteur pour favoriser l'accès

Cheryl Green, MFA, MS

Independent Media Artist and Media Access Specialist

chrl [dot] grn [at] gmail [dot] com

WhoAmIToStopIt.com

Abstract

Even as deaf and hard of hearing filmmakers and activists repeatedly call for quality captions on all video content, many non-deaf filmmakers have managed to remain unaware of the need for and purpose of captions. Implicit biases drive many filmmakers to exclude access from their budgets and their films. These biases include a notion that caption users are not a viable audience, concerns that captions will threaten the beauty of video images by covering part of the screen, and an audist attitude that any level and quality of transcription of spoken dialogue must be adequate. The author is a hearing captioner and filmmaker. In this essay, she reflects on how she advocates for film accessibility through captions. She describes her strategy, how she frames “onscreen real estate,” and responses from filmmakers for captions, including the hopeful way that some say thank you. Quality captions are contrasted against woefully inadequate captions—or “craptions”—provided automatically by YouTube and by companies with cut-rate services. The author considers a focus on inviting access rather than waiting for a compliance-based method of only captioning a film when the filmmaker learns it is required.

Résume

Malgré les demandes constantes des cinéastes et des militant·es sourd·es et malentendant·es d’inclure des sous-titres de qualité dans tout contenu vidéo, de nombreux cinéastes entendant·es continuent d’ignorer le besoin et le but d’ajouter des sous-titres. Pour plusieurs d’entre eux, des biais implicites les poussent à exclure l’accessibilité de leurs budgets et de leurs films. Ces biais incluent l’idée que le nombre d’utilisateurs et d’utilisatrices de sous-titres n’est pas suffisant pour que ce soit une option viable, des craintes que les sous-titres menacent l’esthétisme des images vidéos en couvrant une partie de l’écran et l’attitude audiste selon laquelle le niveau et la qualité de la transcription du dialogue parlé doivent être adéquats. L’auteure est sous-titreuse et cinéaste entendante. Dans cet essai, elle réfléchit à la façon dont elle milite pour l’accessibilité des films par les sous-titres. Elle décrit sa stratégie, son cadrage de la « zone utile de l’écran » et les réponses des cinéastes au sujet des sous-titres, y compris les remerciements reçus de la part de certain·es et qui donnent de l’espoir pour l’avenir. Les sous-titres de qualité sont mis en contraste avec les terribles sous-titres générés automatiquement par YouTube et par les entreprises proposant des services à prix réduit. L’auteure envisage une stratégie qui se concentrerait sur l’invitation à l’accès plutôt qu’une stratégie axée sur la conformité où un film est sous-titré uniquement lorsque le cinéaste apprend qu’il doit se conformer à cette exigence en matière d’accessibilité.

Keywords: closed captions, accessible media, implicit bias, deaf and hard-of-hearing



Let me introduce myself. I am a white, hearing and sighted person earning a living making captions, subtitles, and audio description for film and TV, and transcripts for podcasts. I know only a couple dozen ASL signs, including for “captions” and “captioner.” Formerly a fluent speaker of Spanish, I sometimes work bilingually on subtitles. I work mostly for independent filmmakers who did not consider they might need captions when they started making their films because frankly, that is most independent filmmakers. In addition to providing access, I am also an independent filmmaker and audio producer, and I have oriented myself to think about accessibility as part of the art, not as a compliance-based task for the end of a project. While my clients understand what I do, some in my family believe I only create access for films featuring disabled people. This bias about what should be accessible runs deep for creators, audiences, and people who present content. I believe all content should be made as accessible, and therefore inviting, as possible.

I am not a savior for doing this work; filmmakers do not have bad intentions for not thinking of captions. But the very idea of “I didn’t think about captions” when starting a film points to some insidious ideas: beliefs that people who rely on captions are not a real audience or worth considering; quality standards for captions must not exist or do not matter; or the cost of adding accessibility is not justified the way the cost of professional cinematography and editing is. Captions are just an add-on, after all. They are not part of the content. Quite often—especially outside of films by or about politicized disability and Deaf community—captions are added because a distributor, streaming platform, or audience says captions are needed or required. I sometimes encounter a bit of righteous indignation from creators who are being forced to do something they had not anticipated as part of their final vision for the work. Being forced to comply with a law is never inviting, and it is often accompanied by a tight deadline and the stress of feeling left without guidance. If we could get more filmmakers to “think of captions” from the earliest stage planning a film, they could relieve themselves of this tension.

I would argue—and I do, and often—that accessibility, even if it is not part of an artistic practice or aesthetic, can still be part of the art. I have a few repeat clients who no longer wait for me to comment on where they’ve placed onscreen text such as speakers’ names. They now put that text midway up the screen to make room for captions at the bottom. They have told me it works just as well as the standard spot for onscreen text in the lower third of the screen. And it does. I have one client who no longer complains when I ask her to turn down the music in her work when I, as a captioner with auditory processing difficulties, struggle to identify the spoken words. I worry that other people will have to work too hard to watch the videos, even with captions. After all, not all viewers can read the language presented or read fast enough to keep up. This client first told me that she thought deaf people used my captions, and “regular” audiences would be just fine without them. She now lowers the volume a touch when I request it, acknowledging that all potential audiences’ access needs are varied and valid.

Subtitles, which translate spoken dialogue into a different language, are common in film and television separate from deaf and disability access. Captions, however, include identification of who is speaking (or when speakers change), music, ambient sounds, and sound effects. Descriptions of paralanguage qualities such as whispering, sighing, or yelling could be included when possible. Captions should create an accurate approximation of a rich aural environment while balancing timing and readability (DCMP, n.d.). Captions often have an opaque background to make them clearly readable against any type of image in a film while subtitles typically have a dark outline or drop shadow to the text but no dark background.

There are two types of captions, Closed and Open. “Closed” refers to those that individual users turn on to display and turn off so that they do not display. “Open” refers to captions displayed onscreen for all users. Because Open Captions are essentially video graphics, they can, and do, take advantage of a range of visual styles around font size and type and colors; changing onscreen placement; and animation and movement. The Dodo distributes adorable animal videos online that provide salient examples of aesthetically-driven, creative Open Captions.[1] This paper will focus exclusively on Closed Captions for two reasons. Firstly, the bulk of my captioning work is Closed Captions for independent film distributed via DVD, streaming, and on television through PBS (the Public Broadcasting Service). Secondly, while many topics explored here apply to both Open and Closed Captions, the variety of visual choices that can be made in Open Captions warrants its own paper.

Despite having Closed Captioning on television in the US since the 1970s (NCI, n.d.), despite federal laws stating that certain content must be captioned (FCC, n.d.), despite the growing body of films from Deaf filmmakers like Jade Bryan—some of whom caption or subtitle their work themselves—despite new social media-driven research that a huge number of videos and ads on Facebook are watched without sound (cha-ching) (Patel, 2016), and despite research that captions support deaf and hard of hearing people’s access to information as well as people gaining literacy or learning a new language or who want support for auditory processing (Linder, 2016), resistance to adding them persists, especially online (Butler, 2019).[2]

This could be resistance to paying for captions to be made or resistance to paying for corrections when I point out that their $1/minute captions from Rev.com contain, at minimum, dozens of errors. At best, these errors make the captions hard to read. The middle range of awful is that they deliver literally a different film from the one intended because so many words in the captions do not match what is in the film. The worst is that they provide racist and other culturally harmful text. Well, I do not know what is the worst. Is it the Rev.com misspellings, the ethnocentric reliance on “speaks foreign language” even when we know what language people are using because the filmmaker has made it clear, or captions that are 700 words per minute and have a super-long top line and a one-word bottom line, forcing eyes to dart quickly and making caption users work harder than they should have to?

Or is the worst the Rev.com ads that Facebook delivers to me guaranteeing typo-free captions? Is it the filmmakers who I cannot convince to let me remake their Rev.com captions to be as meticulous, thoughtful, and beautiful as their films? Their films do not contain 10+ errors per minute, but their captions do. Or, no, is the worst all the filmmakers who tell me, “My captions look fine,” when I have told them that their captions have dozens to hundreds of errors and sometimes are actually only subtitles and not even captions? Because subtitles are only dialogue, usually with no background color, they presuppose a hearing audience who can read text placed atop any color or texture of image. I saw subtitles in one film that did “look fine” as far as font, but the white text was often over the chest and shoulders of the film’s main character who was wearing a white shirt. This rendered them completely illegible, white against white. In this case, they did not look fine when played.

I sometimes find myself fighting back boredom when I correct poor Rev.com captions and come across the umpteenth use of the same adjective to describe different qualities of sounds. I consider it imperative for captioners to spend time regularly using a dictionary and thesaurus, refreshing themselves on punctuation conventions, and reading about culture, pop culture, news, art, and communities to which we do not belong—especially those that may be under-represented in the media—to inform us for making excellent, informed captions (Reeb, 2019; Zdenek, 2015). This is to prevent such egregious lack of effort demonstrated in things like “speaks foreign language.” Or the deeply microaggressive “exotic music plays” that I found in a film whose captions I repaired. It was a documentary by an Indian-American woman made in India, featuring Indian people. At no time was anything remotely “exotic” in this film. It is not only a matter of avoiding typos but of demonstrating respect, knowledge, and care.

As I mentioned, I not only create access post-production. I am a filmmaker and Member-Owner at New Day Films, a distribution cooperative for social issue documentaries. This means I am intimately familiar with the pre-production, production, editing, and distribution phases of filmmaking. I know what it can cost to make and distribute content and to make it more accessible. I am aware that accessibility can be cost-prohibitive for some filmmakers, especially for emerging and/or marginalized creators who have less-ready access to funding or who are so busy making ends meet that there is no time to start a non-profit, run a Kickstarter, secure a fiscal sponsor, or apply for grants to pay for everything that they would wish to include. I extend a discount to nearly anyone who asks for one, and for some filmmakers who do not ask but who demonstrate their care for access and regret when they cannot afford it, I usually simply offer the discount.

As New Day Films is a cooperative, all Member-Owners do a volunteer job toward running the business.[3] We are an innovative lot. A year after I joined, I created my own job of captioning the trailers on our YouTube channel, because when I joined, I was startled to discover that very few trailers representing our collection had captions or subtitles. All of our films are captioned or subtitled. I argued that librarians and professors who rely on captions exist and should be able to watch our trailers. The vote was unanimous to have me spend my job hours on captioning. I had started preaching about quality captions when I joined in 2016 and had already captioned several new and old New Day titles by the time I suggested captioning our trailers. New Day has several proactive initiatives toward improving access for under-represented communities in the cooperative (in terms of film subjects, filmmakers, and the audiences we serve), and the coop understands that initiatives improving representation do not apply only to race and ethnicity, gender, sexual orientation, and the other identity markers quite often discussed in the film industry. They must also apply to people who use accommodations to engage meaningfully with content.

Because New Day is a cooperative, there are perks too. After I caption a trailer, I give the caption file to each filmmaker to use on their own YouTube channels. The response is mixed, from no response at all (most common) to emails filled with exclamation points thanking me for the awesome service and their almost disbelief that they can just have the captions at no cost. The ones who write with the exclamation points fuel my hope for making it a trend to build in access by default; plan, design, and budget for it from the beginning; and not have to race at the end to meet compliance. We could choose to celebrate the beauty of access, whether we personally utilize accommodations or not.

Caption users themselves have been advocating and being activists and fighters for their rights to access information and entertainment for years (Butler, 2019). Thanks to them, I have a career I enjoy and feel proud of. Thank you, caption users. I see you and your constant fight to address the implicit and explicit biases against your needs and rights in how media is delivered. Rikki Poynter, Deaf YouTuber, I own the t-shirt you designed with “Caption Your Videos” on the back and “#NoMoreCRAPtions” on the front (Wong, 2017). I was once approached by a caption user behind me in a grocery store line who saw my shirt and wanted to give a hearty thanks for sharing the message so publicly. Once he got my attention and saw the front of the shirt, he had countless stories of his run-ins with craptions.

Craptions are what we call YouTube’s auto-generated “captions.” It is a more meaningful descriptor than “captions.” I say that “craptions” is more meaningful because one, the auto-generated text is actually subtitles (text, but not who said it or anything beyond the spoken word). And two, they are sometimes worse than Rev.com captions because automatic speech recognition software cannot be as accurate as human captioners can be (Khamis, 2019). If there is background music or someone with a speech disability speaking, for example, the craptions become what I might refer to as a dumpster fire as far as caption users’ needs go. (Full disclosure: I. Do. Not. Get. This. Rev.com did a feature article on Rikki’s No More CRAPtions campaign. Good for them, but I am not linking to it. I cannot tolerate the ironic tokenization of Rikki’s hard work by a company that churns out consistently low-quality captions.)

I still have the rare clients who unabashedly complain to me, of all people, that they are being forced to add captions, that it is crucial I respect that their footage is gorgeous, and that actually, they do not want their images to be covered with words. (Words that represent what they themselves put in the film.) I reassure them that of course their footage is gorgeous. I expect no less! And that I am honoured to make it so they can have a larger audience thanks to captions. I do not convince all of them with this charm, but I have won over a couple. The only client I gave up on is the one who repeatedly insisted I “bleep” cussing from her film because it was being shown in schools. I asked how she felt about students’ auditory access to “f*ck” and “sh*t,” and she said it was fine. I told her it is unethical for me to editorialize in the captions and to purposefully fail to represent the audio as accurately as I can. She insisted on no f*cks in the text, and I acquiesced. She then demanded I find a more child-friendly word for “sex” when someone in the film spoke that word. How can I get more child-friendly? How can I keep the captions timed accurately if I were to change “sex” to “doing what grown-ups do in private?” How would she or I answer to a viewer who can hear enough to parse “sex” and wonders why they were given something different to read? (Ratcliff, 2018). That is when I fired her.

I want to thank filmmakers who allow me to advocate directly to them. They write to me with, “Your captions are covering my onscreen text.” I write back, “Your onscreen text is in the caption area.”[4] Although it is usually too late for them to move their text, I am impressed that they do not argue when I reframe the question of onscreen real estate. Some now ask me to check their layout before finalizing the film, which makes my heart sing. I gladly do that at no charge. I want to also thank the filmmakers who email me with the exclamation points about receiving captions and who comment how gracefully captions can flow with voices and sounds and how fun it is to see how many adjectives I can fit into one trailer or film. They recognize captions as value added. Because they are.

When I presented at the 2018 VIBE Symposium at Concordia University, I screened a short film that I made called In My Home about the institutional bias we have in the United States toward funding the institutionalization of disabled people over funding home and community-based care.[5] I got my captions translated into French and displayed both English and French Open Captions onscreen throughout the English-language film (with English Audio Description). I had no way to know if there might be French readers in the audience, but it is part of my disability aesthetic to offer access features without anyone asking (Green, 2019; Carter-Long, 2019). That does not mean all people are always included, not by a long shot. But it means that what I do have, I will offer. Though I was presenting as a filmmaker, the Q and A focused mostly on accessibility, demonstrating to me what hunger people have for knowing how to do it, and what hunger people have for using accommodations.

I wish to conclude with an invitation here for high-quality access that is based on best practices developed by caption users and captioners in collaboration (Butler, 2019; DCMP, n.d.). Sure, if you have no money, use YouTube’s caption editor to tweak the craptions for your online video to be accurate(er) or use the very powerful freeware Subtitle Edit on a PC. (I don’t yet have a good recommendation for freeware for Mac.) But be aware that captions are not just about the spoken words. They are available to describe dialects, tone of voice, mood of the music, and more to create a textual version of the aural landscape.

My invitation starts with addressing the idea that creating quality captions is a worthy endeavour for worthy audiences. This invitation is based on shifts such as moving the consideration of captions from post-production to pre-production. When producers and directors consider captions an absolutely essential part of a film, the transcription and captioning processes can receive line items in a film’s budget. I invite cinematographers to consider framing shots to accommodate captions so that an ultra-close up of someone’s face is not changed by captions covering their mouth as they reveal something highly emotional. For film editors, rather than ask the captioner to move captions around the screen so they do not cover crucial text or images onscreen, find the places, where possible, to keep the bottom center of the screen clear so captions are consistent in their usual spot. Editors can sometimes add brief pauses to allow captions to be displayed long enough to be comfortably read, especially in the case of captioning someone who speaks very quickly. If a fast speaker were followed by one second of silence rather than a fast cut to something else requiring captioning, a caption could linger that moment longer to be savoured. And high quality, culturally-sensitive, creative captions that accurately represent a film are most assuredly something to savour.

Endnotes

  1. https://www.thedodo.com/: The Dodo uses animation particularly to represent animal vocalizations. Rather than “[barking]” or “[whining]” as captions might typically display, The Dodo overlays moving lines emanating from an animal’s mouth. The size and thickness of the lines and the speed and distance the lines travel vary to indicate different volume, urgency, or emotiveness in the animals’ sounds. Human speech is indicated at the bottom of the screen, in the standard caption area.
  2. Captions For Literacy has a wealth of links to primary research on the positive effects of captions for children and adults, especially those who are deaf, come from economic disadvantage, are learning a new language, or who struggle to attain literacy. The organization also provides advocate resources and lists of where to find captioned materials and how to turn captions on at http://captionsforliteracy.org/learn.php. Additional research can be found in peer reviewed journals from a variety of disciplines, including Journal of Deaf Studies and Deaf Education, Journal of Research in Reading, Policy Insights from the Behavioral and Brain Sciences, and Rhetoric Review. Caption advocates regularly publish blog posts and have write-ups on their grassroots work in local newspapers that can be found online.
  3. https://www.newday.com/
  4. For DVD, Blu-ray, TV broadcast, and some streaming platforms, closed captions can be moved so that other text and graphics are not covered. However, even if a captioner puts in coding to move the captions, YouTube and Vimeo will ignore that coding and put the captions in the default position: centered at the bottom of the screen. Per FCC guidelines, streaming media players must allow individual users to be able to move captions on playback, but I would never consider it appropriate to ask users to keep their hand on their mouse or keyboard and be ready to move captions back and forth during a film with no notice when the overlapping captions and onscreen text are coming. Captions that move around (by the user or by the captioner) can be jarring. But at the same time, they can be moved to enhance meaning or salience or draw a viewer’s gaze deliberately to a particular part of the screen.
  5. http://cdrnys.org/free-our-people-film-contest/

Bio:

Cheryl Green, MFA, MS is a captioner, audio describer, multi-media digital artist, a 2017 AIR New Voices Scholar, 2020 DOC NYC Documentary New Leader, and Digital Operations Lead and a Member-Owner at New Day Films. She brings her lived experience with multiple invisible disabilities to creating accessible media that explores politically- and culturally-engaged stories from cross-disability communities. She reported and produced one episode for the Peabody-nominated Season 2 of 70 Million. Find her TBI-related short films on BrainLine.org. She captions and audio describes films for Kinetic Light, Superfest, and Cinema Touching Disability, among others, and leads workshops for artists and museums on arts accessibility. Visit her website at www.WhoAmIToStopIt.com.

References