Answers are grounded in this paper and cite their sources.
Cross-catalogue analysis reveals widespread absence of original language metadata across millions of books, highlighting systemic information gaps for minority languages.
A reader of German or Italian who wants to know whether a book exists in their language asks once and gets an answer. A reader of Ladin can ask twenty times and reach only silence — unable to tell whether the book was never translated, or was translated and nobody wrote it down. That silence is not indifference. Nobody is obliged to record which language a book was translated from, and what nobody is obliged to record gets recorded only where somebody needs it. Working across more than fifty open catalogues, we find that about 43% of works — 5.7 million as of 14 August 2026 — carry no attested original language at all. When we measured how much of that was our own failure to collect, it came to 0.2%. The rest was never written down.
No takes yet. Share an insight, caveat, or question.
Ausiàs Tsel (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: