In the spring of 1991, Canada's Immigration and Refugee Board tested 370 of its working interpreters against a new accreditation standard. 40% failed (Barsky, 1993). That is the clearest evidence there is that testing catches something real, people who were already inside a life-or-death system without the skills to do the job properly. The UK has never run that test. Nobody knows what its equivalent number would be.
The Evidence
Canada requires interpreters to pass a three-part accreditation test before they can work for the IRB:
- sight testing
- auditory testing
- general language ability
This is set out directly in the IRB's own Interpreter Handbook: “To provide interpretation services to the IRB, you must successfully pass an accreditation test.” The system now covers over 260 languages across 40,000–60,000 procedures a year, with random performance audits built in.
The UK has no equivalent requirement. There are qualifications available, the Chartered Institute of Linguists (CIOL) offers several, and the Diploma in Public Service Interpreting (DPSI) is the most advanced of them, described by the CIOL itself as a “driving test” for interpreting: proof that someone can handle the cognitive load and linguistic precision the job actually demands. But none of these qualifications are required. An interpreter can work a UK asylum hearing having never sat that test at all.
This matters because of what happens when interpretation goes wrong. Williamson (2024), in an MPhil thesis for the University of Glasgow, describes the Home Office using every word of an applicant's testimony to try to “undermine the applicant's credibility.” When the interpreter isn't qualified, the inconsistency that shows up in the record might not be the applicant's at all. It might be the interpreter's. But it's the applicant who pays for it.
The Explanation
Interpreters give a voice to people who don't speak English in the UK asylum system. That's the entire point of having them — to make sure everyone's account can be heard on equal terms, regardless of what language they arrived speaking. An unqualified interpreter can quietly undo that. An ICIBI inspection report found interpreter quality lacking across several measures:
- interpreters' grasp of English
- their ability to interpret accurately at speed
- failures to render what claimants actually said
Every one of those failures becomes part of the official record. And the official record is what decides a person's case, not what they actually said, not what they meant, but what an unqualified interpreter, working in real time and under pressure, managed to get down. A small inaccuracy is not a small thing here. It can be the difference between a case that holds together and one that doesn't.
Barsky's 1993 paper on Canadian refugee hearings, published two years after the accreditation test was introduced, documents cases of interpretation going wrong in exactly this way. It's the earliest record of what happens when the gap between what someone says and what gets written down is left unchecked.
The Reform
This is what Translational Justice's Reform 1 asks for: mandatory qualification for interpreters. The DPSI already exists. It already tests linguistic competence, professional ethics, and legal knowledge. It isn't a new invention, or a burden dreamed up to slow the system down. Interpreters working under Home Office contracts are never required to hold it.
Canada has run its version of this system since 1991. The UK could start tomorrow.
Sources: Immigration and Refugee Board of Canada, Interpreter Handbook. Barsky, R.F. (1993), “The Interpreter and the Canadian Convention Refugee Hearing,” TTR: Traduction, Terminologie, Rédaction, 6(2), 131–146 — an academic source, not a Government of Canada publication. Williamson, A.J. (2024), MPhil thesis, University of Glasgow.