Discussions of translation and justice tend to assume a single direction of harm: a source language is flattened, simplified, or distorted as it passes into a dominant target language, and the injury runs one way. This essay argues that this picture is incomplete. Translational justice is not one-sided. When a term crosses from one language into another, both languages are put at risk of injustice, though the injustice each suffers is not the same in kind. The source language risks mistranslation: its meaning is altered or lost in the crossing. The target language risks something equally serious I call false representation, where a word is retained in its borrowed form but is quietly stripped of the connotations that gave it a stable meaning in the target language to begin with.
Translation Affects Both Directions
The starting premise is as follows: translation is not a one-way channel through which meaning is merely carried from a source into a target, unchanged in its effect on either side. Any act of translation reshapes two linguistic systems at once: the one it draws from, and the one it delivers into. Treating translation as a purely one-directional act of transmission obscures the fact that the receiving language is also altered by what it absorbs and can be misrepresented by how that absorption occurs.
A Case in Point: Asylum Terminology
A lawyer working in the UK asylum system described a concrete instance of exactly this problem. An Arabic term used by an asylum applicant, denoting an inside door, baab, was translated into English simply as "gate." A gate is a point of entry or exit at the edge of a boundary, approached from outside. The two terms carry distinct spatial and conceptual logics. "Inside door" frames the speaker as already within a building, navigating internal passage; "gate" frames them as outside a boundary, seeking entry. In an asylum interview, where a claimant's account of their situation is scrutinised for consistency and credibility, the conceptual frame supplied by a single mistranslated term can materially affect how their testimony is understood (Williamson A.J. (2024). "Defining the role of the UK asylum interpreter: expectations, realities and training needs." MPhil thesis, University of Glasgow).
The injustice here runs in both directions at once. To Arabic, the injustice is mistranslation: a specific, internally-oriented term is flattened into an external one that does not capture what the speaker meant. To English, the injustice is a false representation of the language itself. By rendering the term as "gate" and nothing else, the translation implies that English has no word for an inside door, only a gate, when in fact English has "door," "inner door," "threshold," and other terms perfectly capable of capturing the original sense. English is made to appear conceptually poorer than it is, misrepresented as lacking a distinction it in fact possesses.
The Independent Chief Inspector of Borders and Immigration's report on asylum casework, covering June to October 2023, found that interpreters' grasp of English was itself a documented concern within the Home Office's asylum process. This corroborates the claim that mistranslations of this kind are not isolated errors but part of a broader, systemic pattern, one with direct bearing on how asylum claims are processed, understood, and decided.
An instructive contrast comes from an unrelated but revealing case, one that led to my realisation about the injustice acting in both directions. A translator working on Niklas Luhmann's writings on the Zettelkasten explained a deliberate choice to render the German plural "Zettelkästen" as "Zettelkastens" in English, reasoning that the umlauted "ä" felt "too foreign and too disrespectful towards the English language to inject." Whatever one makes of the underlying linguistic judgment, the translator's reasoning shows an explicit awareness that translation choices can either respect or disrespect the receiving language, that the target language, too, has an integrity that can be honoured or violated by how foreign material is absorbed into it. The asylum case shows what happens when that awareness is absent: English is not respected by careful accommodation, as in the Zettelkasten example, but quietly misrepresented by a translation that erases a distinction it does not need to erase.
Two Distinct Injustices
By mistranslating the term in this way, an injustice is done to both languages involved, though not the same injustice. To Arabic-speaking asylum seekers and interpreters working from the original conceptual frame, the harm is a mistranslation. The term they know and use is rendered inaccurately, its meaning displaced.
To English, the harm is different. It is false representation. English absorbs "gate" as though it were an accurate equivalent, and English speakers, officials, and institutions come to treat the borrowed term as though it carried its apparent English meaning, when in fact it smuggles in a foreign conceptual frame under a false guise of familiarity. English is made to say something it does not actually mean.
This is to say that the source language is deprived of a faithful representation, and the target language holds that deprivation. That deprivation is what creates a hierarchical consequence, as I will discuss in the next section. But the fact that both languages receive an injustice in deprivation clearly illustrates that both languages are victims.
The source language is deprived of faithful representation. The target language carries that deprivation. The deprivation becomes institutional consequence.
This distinction matters because it resists a common assumption in translation ethics that only the source language and its speakers can be wronged by translation, while the target language is a neutral, unharmed receiver. The asylum-system case shows the opposite. Interpreters working in English lack the very conceptual apparatus that would let them recognise the substitution has taken place, precisely because the false representation is seamless. It does not look like an error from within English.
A Response to Hierarchical Models of Translation
This account departs from an existing model of translation and injustice: Angermeyer's discussion of translating "up" into a dominant language and translating "down" into a subordinated one, a distinction he draws from Simon (2006). Angermeyer's model is hierarchical throughout. Translation is treated as power moving in one direction between two unequally positioned languages, and injustice is located in that directional relationship itself, from the moment the information is transmitted to the moment it lands on the person affected. Even where Di Martino, responding to Angermeyer, shows that translating "up" can be discriminatory, as in her own experience of punitive translation, where a speaker is forced to reformulate their words into a more "refined" register of the dominant language, the dominant language itself remains exempt from injury throughout. It is always the mechanism of harm, never its victim.
I do not reject Angermeyer's model outright. I think it is right about one thing and wrong about another, and the two belong to different moments in the process.
At the moment of transmission, when the information first crosses from one language to another, the injustice is lateral, not hierarchical. Neither language sits above the other as the source of harm. The source language is deprived of a faithful representation, and the target language bears a meaning it did not honestly produce. Arabic is not wronged by English, nor English by Arabic. Both are wronged independently, at the same moment, by the act of translation itself.
It is only once translation moves beyond transmission and into consequence that a hierarchy appears, and here I agree with Angermeyer. The linguistic harm does not stay at the level of language. It lands on the asylum seeker, whose testimony is filtered through the mistranslated term and assessed by institutions operating in English. The institution conducting the interview absorbs no cost from the error, and may even benefit from the ambiguity it introduces, while the asylum seeker bears the full institutional consequence of a testimony altered without their knowledge. This is the hierarchy Angermeyer identifies, and it is real, but it belongs to the institutional stage, not the transmission stage. The injustice does not start out hierarchical. It becomes hierarchical once it reaches the institution.
This is why interpreter accreditation matters as a matter of justice, not administrative quality control. If translational injustice were hierarchical from the start, the remedy would only need to protect the subordinated language and its speakers. But because the first injustice is lateral, English is wronged too, quietly misrepresented rather than merely used to harm Arabic, and accreditation reform must also stop institutions relying on a false confidence in a target language that has been emptied of distinctions it actually has.
Conclusion
Translational justice, then, cannot be adequately addressed by asking only whether a source language has been faithfully rendered. It must also ask what the target language has been made to believe about itself, what borrowed terms it has accepted as native, and what conceptual frames have entered undetected inside familiar-looking words. Both languages receive an injustice in the act of translation; recognising that both wrongs exist, and that they take different forms, is a precondition for any translation practice, particularly in high-stakes institutional contexts like asylum systems, that aims to do right by everyone it touches. This requires a reform in which interpreters are qualified and tested in English in the UK to prevent the injustice against English.
That is why Reform 1 now requires English accreditation as part of asylum interpreter accreditation. The target language cannot be treated as automatically safe just because it is dominant.
References
- Independent Chief Inspector of Borders and Immigration (ICIBI). An Inspection of Asylum Casework, June to October 2023. David Neal.
- Angermeyer, Philipp Sebastian (2023). "From Punitive Multilingualism and Forensic Translation Towards Linguistic Justice." Language in Society 52(5): 901-908.
- Simon, Sherry (2006). Translating Montreal: Episodes in the Life of a Divided City. Montreal: McGill-Queen's University Press.
- Di Martino, Emilia (2023). Commentary in "From Punitive Multilingualism and Forensic Translation Towards Linguistic Justice: Commentaries." Language in Society 52(5).
- "Improved Translation of 'Communications with Zettelkästen.'" Zettelkasten Method. zettelkasten.de/communications-with-zettelkastens/#comments-on-the-translation.
- Williamson, A.J. (2024). "Defining the role of the UK asylum interpreter: expectations, realities and training needs." MPhil thesis, University of Glasgow.