Courts record everything. The date the application was filed. The time the hearing started. The case reference number. The documents submitted. The name of the judge. The name of the legal representative.
They do not record the quality of the interpreter.
This is a structural choice with structural consequences. Asylum proceedings generate an official record that is treated as a neutral transcript of what was said. That record passes through an interpreter before it exists. The interpreter's competence, their familiarity with dialect, their knowledge of specialist terminology, their ability to render meaning at speed under institutional pressure determines what the record contains. None of that is tracked. None of it appears in the official file.
Consider what this would mean in another context.
A clinical trial measures every variable that could affect the outcome. Dosage, duration, patient history, methodology, the conditions under which results were collected. The integrity of the finding depends on knowing what produced it. If a variable that affects outcomes is left unmeasured, the result is noise. The scientific and medical communities understand this. The courtroom, apparently, does not.
Or consider engineering. A bridge is tested under known loads, with known materials, under recorded conditions. When something fails, the failure can be traced. Accountability exists because measurement exists. You can only investigate a collapse if you know what was holding it up.
Asylum proceedings make credibility determinations about people whose lives depend on those determinations. The primary mechanism for establishing what they said is interpretation. That interpretation is produced by someone selected through a procurement chain running from the Home Office to a contractor to a sub-contracted agency selected primarily by cost, with no mandatory accreditation requirement, no minimum qualification threshold, no verification available to the applicant or their legal representative.
And then: nothing. No record of who the interpreter was, beyond a name that may appear somewhere in the file. No record of their qualifications. No record of whether they held any qualifications at all. No record of their dialect competency. No record of their familiarity with the specific terminology the applicant's account required.
The system then uses the output of this unrecorded variable to assess credibility.
Inconsistency in testimony is a credibility finding. In cases where corroborating evidence is scarce by definition because people fleeing persecution rarely carry documentation, credibility is often the whole case. If the inconsistency originated with the interpreter rather than the applicant, there is no mechanism to identify this. The official record does not show what was said in the source language. It shows what the interpreter produced. Those are not the same thing, and the gap between them is not recorded.
This is not how rigorous systems should handle consequential decisions.
Medicine tracks the variables that affect patient outcomes. Science tracks the variables that affect experimental results. Engineering tracks the variables that affect structural integrity. These fields understand that an untracked variable is a source of error that cannot be corrected because it cannot be seen.
Asylum proceedings track the applicant's account for inconsistency. They do not track whether the account was accurately rendered in the first place.
The minimum standard would be to treat the interpreter as what they are: a variable. Record their qualifications and their dialect competency. Record whether they hold accreditation and under what body. Give this information to the applicant and their legal representative before the hearing begins. Build a mechanism for raising concerns about interpreter quality that does not require the applicant to understand a system they have no access to.
Until then, the official record is a record of what the procurement chain produced. The system calls these the same thing.