Translation does not merely transport a concept from one language into another; it gives the concept another intellectual home. Words carry histories, associations, political resonances and disciplinary conventions that rarely align perfectly across languages. A translation may preserve the practical mechanism of a concept while altering its rhetorical force; another may maintain etymological proximity while shifting contemporary associations; a third may require a phrase or explanatory construction where the original language permits a compact term. Exact equivalence is therefore the exception rather than the general rule, and this does not make translation a failure. It makes translation an explicit form of semantic transformation. A multilingual knowledge infrastructure should consequently record more than paired labels. It should be capable of indicating whether a translation is conventional, partial, functional, contested, historical or deliberately asymmetric. Directionality matters as well. A term may have a recognizable equivalent when moving from one disciplinary language into another while the reverse movement requires a longer explanation or loses an essential distinction. Such asymmetries contain intellectual information, and flattening them into standardized equivalence may improve database simplicity at the cost of conceptual precision. RegimeMutation occurs when a receiving linguistic or disciplinary environment changes not only the wording but the rules under which a concept becomes intelligible and usable. Translation also modifies provenance. Once a translated term begins circulating independently, later users may encounter it without knowing the genealogy of the source concept, and over time the translated version can acquire its own applications, associations and neighbouring ideas until it becomes partially autonomous. This need not represent degradation or error; it may constitute genuine conceptual development. What matters is preserving enough relational history to reconstruct the transformation. Benjamin’s account of translation as an afterlife of the original is particularly suggestive here: concepts survive through acts that simultaneously alter them. Digital systems now perform such acts at enormous scale, through machine translation, multilingual search and cross-lingual representations, making the fantasy of transparent equivalence especially seductive. A richer architecture would treat translation as another occurrence in the biography of a concept, connected to earlier formulations yet situated within a distinct linguistic ecology. One concept can therefore possess several linguistic homes without requiring those homes to become identical rooms. Translation is not simply a bridge between stable meanings; it is one of the places in which meaning continues to happen.
Benjamin, W. (1923) ‘Die Aufgabe des Übersetzers’.
Derrida, J. (1985) ‘Des Tours de Babel’, in Difference in Translation.
Eco, U. (2003) Mouse or Rat? Translation as Negotiation. Weidenfeld & Nicolson.
Venuti, L. (1995) The Translator’s Invisibility. Routledge.
Star, S. L. and Griesemer, J. R. (1989) ‘Institutional Ecology, “Translations” and Boundary Objects’, Social Studies of Science, 19(3), 387–420.