2 September 2026
Using the language of typography, this piece asks a narrow question: before a migration system judges whether someone is credible, does its form even give them a way to describe their life honestly? I argue that the answer is often no — and that this failure happens earlier than most debates about biased algorithms acknowledge.
Picture a text box on a migration form. The instruction reads: “If you were not working, explain what you were doing.” Below it, the form offers examples: unemployed, studying, traveling, retired, in detention. You scroll through them looking for the months you spent hiding, moving between temporary homes, caring for a parent while your own papers lapsed, working informally, or struggling with depression at home. Well, too bad. You should not have been living through a period that does not fit one stable administrative category.
Even when none of the examples fit, the field is still required. Something must be entered: something compressed into language the institution knows how to process. This is where migration meets typography. In typography, simply put, legibility is whether individual characters can be recognized; readability is whether the text they form makes sense as a whole.
A font may be perfectly legible, letter by letter, while the paragraph it produces remains unreadable. Digital visa and residence-permit systems can create the same problem—only the material isn’t ink, but a life. Much existing research on digital borders rightly examines classification, automated risk assessment, bias, and migrant resistance—basically studying what happens after that life has been turned into data. But what happens one step earlier? Before an officer or automated system assesses the application, what about the moment a person sits down and must present a life that was never organized on institutional terms and fit—or force—it into an institutional framework?
While making the existing procedure faster, digitalization also redistributes this interpretive labor.
I call that work biographical kerning. In typesetting, kerning adjusts the space between pairs of characters, so a word reads as one coherent shape—badly kerned, “FLICK” can be misread as something else entirely. Its migration equivalent is the quiet labor of closing the gaps and untangling the overlaps in a life so it presents as one continuous, verifiable, and institutionally familiar profile.
Canada’s IMM 5669 personal-history form is a clean example: applicants must account for every period since turning eighteen, or the last ten years, whichever is more recent, with a warning that any unexplained gap may delay processing. Moreover, under section 40 of the Immigration and Refugee Protection Act, materially misrepresenting or withholding information in that timeline can have legal consequences.
The applicant therefore isn’t just reporting their life, but re-kerning it: deciding how to represent certain periods in a way a caseworker will read as continuous rather than suspicious.
But some experiences don’t just need re-spacing, because they don’t have a character to begin with. A missing glyph is what happens when a font simply has no symbol for something. In computing, you may have encountered that little placeholder box that shows up instead. Germany’s VIDEX system requires the use of Latin letters, and applications using other scripts cannot be saved or printed. A name in its original script can therefore become a missing glyph in that interface. The same failure happens one level up, at the level of biography rather than script. A field structured around “unemployed / studying / traveling / retired / in detention” may have no adequate space for “fled,” “was in hiding,” “moved repeatedly after displacement,” or “worked without papers to survive”—not merely hypothetical situations, but contemporary realities. The life is real. The field for it was never built. That’s the missing glyph in the title—not an inconvenience in a form, but an entire category of true statements that the interface has no character set for.
It’s tempting to read all of this as sorting people into types: the legible migrant and the illegible one. I want to resist that move, because it isn’t true to the evidence. The same person can be legible on one form and illegible on the next.
Design can therefore create risk, but it can also create routes for repair. Türkiye’s e-İkamet procedure requires extension applicants to check information supplied earlier and allows missing information to be provided within a deadline. Germany’s Consular Services Portal also demonstrates the value of proofing, because applications undergo preliminary review and applicants can be asked to make corrections before the appointment. Canada’s IMM 5669 does not provide the same built-in preliminary review, but applicants may attach a separate document when the form does not provide enough space.
Each of these can function as a mechanism for proofing: an opportunity to repair the administrative text before it produces negative consequences.
So there are at least four analytical positions a person can land in within a particular procedure: legible and readable; legible but unreadable; humanly understandable but administratively illegible; and the missing-glyph case, where no field is provided. Which one you land in has less to do with who you are or what happened than with which government’s form happens to be in front of you. Legibility is relational and procedural, not an inherent quality of a person.
This matters more now, not less, because the EU AI Act classifies certain AI uses in migration, asylum, and border control as high-risk, bringing risk-management obligations and, in certain cases, fundamental-rights impact assessments. Much of that scrutiny lands on the decision: was the algorithm’s output biased? Almost none of it lands on the input: was the applicant ever given a truthful way to write their own life before any algorithm touched it?
Biased outcomes may begin with constrained inputs: categories that exclude, timelines that cannot overlap, scripts that cannot be entered, or inconsistencies that cannot be explained.
If the form has no glyph for something that happened, no amount of downstream fairness testing will recover what was never allowed to be entered. Right now, accountability may start one step too late.
This suggests the need for a biographical readability audit of digital migration procedures—one that examines not only what information a portal collects, but what applicants can truthfully enter, explain, and correct. It would also ask whether public or nonprofit guidance is available, so that institutional readability does not depend mainly on money, professional assistance, or familiarity with bureaucratic language. None of this means that verification is bad. It’s an argument that digital verification quietly rewards the applicants who already have continuous, formal, institutionally familiar lives to describe (or the capital to get professional assistance) and may disadvantage the ones whose lives were shaped by the very displacement the form may exist to assess.
Verification is necessary, and preventing fraud remains a legitimate administrative goal. The central question is not only whether the border can read the data, but who is able to make a life readable to the border. A missing glyph does not mean that the character does not exist. It means that the chosen system has no way to display it. In the same way, an incomplete-looking migration file does not always indicate an incomplete or dishonest life. Sometimes the life is present, but the administrative font does not contain the character.
The views expressed in articles published on GAR Blog are solely those of the authors and do not necessarily reflect the views of the Association for Migration Research.
*Alvard Avagyan is an Armenian psychology student at Middle East Technical University (METU), where she studies as a Türkiye Scholarships recipient. Having lived, studied, and worked across Armenia, Türkiye, Spain, and the United States, she is interested in how public institutions, technology, and migration systems shape the way people’s lives are understood. Alongside her studies, she has contributed to nonprofit and international education initiatives focused on equality, youth engagement, and community development.
