Photo by Estrogin via Creative Commons.

Three quarters of the way down the 2019 Unicode Document Registry, somewhere below a proposal for the Iranian symbol Faravahar (فروهر ) and just above proposals for an accordion emoji and a cockroach emoji, is L2/19-080: the Proposal for Transgender Flag Emoji. Submitted on March 14, 2019, the proposal source is coded “Ted Eytan et al,” shorthand for a team that includes a physician and several well-known transgender activists, as well as designers and software engineers from Google and Microsoft. It’s the result of nearly two years of writing and revising, building out a team of experts, shoring up support, and harnessing the energy of Tweet storms and Pride Parades into a very precisely formatted, data-heavy document.

It’s not the first time the proposal has been submitted—this particular document has gone through two previous iterations, and several members of the current team submitted their own proposals before joining forces. But it is the most thorough and strongly argued version thus far, with everything from a section on “image distinctiveness” to Google trend charts that show a significant increase in searches for “transgender,” to the vocal support for a transgender emoji playing out online. Eytan, a physician and family medicine specialist who began the document in 2017 with activist Bianca Rey, has been keeping careful documentation of the two-year proposal process on his website, revealing the massive effort that can go into creating one of these tiny glyphs—particularly one that’s weighted with so much meaning.

Tweet from Ted Eytan about the resubmitted Proposal for a Transgender Flag Emoji on March 4, 2019.

It also shows that getting an emoji approved through Unicode is more of a science than an art, with a process not entirely intuitive to the average person. That’s perhaps especially true for activists who are used to advocating for a cause by appealing to people’s compassion and sense of empathy. In other words, a deep belief in transgender rights and equal representation—even a lifetime of fighting for those things—will not necessarily sway the committee tasked with approving the encoded characters we know as emoji. In order to create the language needed for visibility in the digital realm—historically a unifying space for the transgender community—the team has to prove, with hard evidence, that a transgender flag emoji has cultural relevance.

This is by necessity, explains Jennifer Daniel, the creative director of emoji at Google and a member of the Unicode Emoji Subcommittee, who offered her support on the proposal. She compares the Unicode approval process—adding an emoji to the Unicode language, which makes it compatible with widely used software and platforms—to adding words to a dictionary. “You don’t see the Oxford English Dictionary adding words that don’t already ‘exist,’’’ she writes in an email. “No, they observe how people are using language and formally recognize words that are already in use by the masses.” In that vein, anyone can submit a proposal for a new emoji. The Emoji Subcommittee then evaluates them based on a “prototype of the image, an explanation of how and why people would use it, and an argument for how the addition would improve the greater emoji ecosystem.”  

Screenshot from Proposal for Transgender Flag Emoji, submitted March 13, 2019.

It was those last two points that proved to be much more technical than the team behind the transgender flag emoji originally predicted. “At first, we didn’t have a lot of justification [in the proposal], because we just thought, ‘Of course we should have an emoji, this is basically erasure. How are you meant to have a language where you can’t represent an entire group of people?’” says Tea Uglow, creative director of Google’s Creative Lab in Sydney, and part of the team submitting the proposal. She first submitted a proposal for a transgender pride emoji in 2017, and after not receiving a response, was connected to Eytan and Rey via Twitter. They also joined forces with activist and author Charlie Craggs, who was taking a more populist approach with her #clawsoutfortrans petition. With the addition of software consultant Alda Vigdís Skarphéðinsdóttir, and activists Hannah Simpson and Chad Cipiti, this is a veritable transgender flag emoji super-team, composed of some of the most outspoken transgender activists. But being well-versed in the human language necessary for rallying, representing, and making visible the transgender community was only getting them so far.

“At this point, there’s not even an agreed upon symbol, because the language has been evolving.” 

Bringing on Daniel, as well as Seb Grubb from Google and Olli Jones from Microsoft, as advisors went a long way toward understanding the very meta technical language behind emoji, and gathering the data for an effective proposal. The most recent proposal is much more thorough, precise, and data-driven than the previous attempts. “Developing the proposal is like doing a design document where the formatting needs to be exactly right,” says Uglow. “The subcommittee is highly pragmatic. All of the justifications need to be supported, and supported in the right form with the right data in the right place. It’s a slow system, but it does mean there’s an extraordinary rigor in that process.”

Screenshot from Proposal for Transgender Flag Emoji, submitted March 13, 2019.

First, there’s the design of the emoji itself to consider—a pink, blue, and white striped flag that would likely be no more technically complex to create than the rainbow flag. The proposed design is a combination of the white flag emoji 🏳 + five pastel bands—pastel Blue (RGB #5BCEFA), pastel Pink (RGB #F5A9B8) and white—repeating horizontally. Like all emoji, it needs to be built on a universal code (thus the portmanteau, Unicode), which gives each letter, digit, or symbol a unique numeric value that can be rendered across different platforms and programs. One of the problems the team ran into was that within the Unicode language, there are already three symbols for transgender and transexual. “At this point, there’s not even an agreed upon symbol, because the language has been evolving,” says Uglow.

For the trans flag to be a white flag + a transgender symbol, one of those three symbols needs to be emojified. (Grubb, from Google, says the subcommittee doesn’t usually ask for the code that the emoji will take, instead deciding on it’s validity first, but this was still included in the proposal.) The decision to use the transgender flag, designed by Monica Helm in 1999, as the emoji representing the transgender community was not without its own considerations. Not everyone likes the use of the traditional pink and blue gender colors, but it has become the global symbol of trans pride, flown at protests, displayed on clothing, and used as a banner on government buildings and institutions like the Smithsonian. “Just like the rainbow colors on the queer pride flag represents the queer community, those colors and the flag itself has been really what all of transgender people and gender nonconforming people identify as part of the queer community space,” says Rey.

What makes emoji like an avocado 🥑, a toboggan 🛷 and a smiling turd 💩 easier adds to the Unicode language than a transgender flag?

In addition to the prototype of the image, the proposal needed to show that this particular emoji representation of the transgender community was the correct choice. In order to do that, it had to prove the flag’s cultural relevance—another area that benefitted from the expertise of the engineers and designers consulting with the team. In the latest proposal, this evidence came in the form of search data, which compares data for “trans person” and “transgender person” to “construction worker person” (an identity that currently has emoji representation). The searches for the former totally eclipses that of the latter: on Google, for example, there were 550,000,000 searches for trans person in Feb/March 2019, as compared to 11,300,000 for construction worker person.

General search data for various terms, as displayed in the Proposal for Transgender Flag Emoji, March 13, 2019. All data are from February and March 2019.

The proposal also collects data from social media and Google trend data from the last five years that shows “transgender” as compared to “LGBT” and “Pride,” in an attempt to show that the transgender flag is necessary in addition to the LGBT flag. According to the proposal, the rainbow flag emoji is “one of the most successful flag emoji” (which are subject to special criteria), even with the same-sex family emoji and same-sex couples emoji providing various options. The proposal uses the rainbow flag to argue for a transgender flag’s potential popularity, while also pointing out that the former is insufficient to represent the transgender community.

Comparing the Google trend topic for “transgender” against reference topics. From Proposal for Transgender Flag Emoji, submitted March 13, 2019.

Using the “construction worker” emoji as a point of comparison had a specific purpose: it was meant to show that in addition to community affiliation (represented by a flag), the transgender emoji would also play an important role in personal identity (much like representing one’s profession). But the comparison also brings up a point many who are fighting for the transgender flag have been been making for years. If emoji can be specific enough and diverse enough to represent lots of different professions—now for male and female, and with different skin tones, too—surely it can include one that represents transgender identity. Even more puzzling, what makes emoji like an avocado 🥑, a toboggan 🛷 and a smiling turd 💩 easier adds to the Unicode language than a transgender flag?

“It’s different when you’re dealing with a group of people,” says Uglow, noting what she’s learned from the proposal process. In other words, the fact that the lobster has been Unicode-approved and emojified is not necessarily because the subcommittee believes it’s more necessary or more significant (“The exclusion of the lobster emoji from the current emoji list places an additional burden on communication for everyone who consumes seafood, works in the fishing industry, or lives in a coastal region,” reads the lobster emoji proposal), but perhaps simply because it was easier and quicker to approve. Less rigor is needed when there’s less at stake. Daniel puts it like this: “The Emoji Subcommittee isn’t about blocking proposals, but about making them as good as they can be—and for good reason—each emoji addition is scrutinized by people around the world and we want to avoid criticisms and mischaracterization of emojis serving as some sort of ‘propaganda.’”

“If you want a truly inclusive internet, you have to give people words.”

Still, Uglow maintains that its important to lift the veil on some of this decision-making process. One of the most interesting things about the transparency this team has brought to the process is how it reveals a chicken-and-the-egg problem for building a more inclusive internet. The online world has long been thought of as a space that has the potential to bring visibility to people who have not previously been represented. But digital systems haven’t always proven to be more open and flexible than physical and historical systems, especially when it comes to social change.

Even outside of emoji, “this is a massive problem for marginalized communities,” says Uglow, “which is that if the data does not exist, you do not exist.” The double bind made apparent through the transgender emoji approval process is that in order to create language that gives transgender people visibility, the community essentially has to prove that it exists, that they are here. “We have these very, very normative approaches to data: If something is common, then it is valid. But if you want a truly inclusive internet, you have to give people words.”