Source: Cassin's Dictionary of Untranslatables — publisher's own scope description

• Author: Princeton University Press (publisher's own book page); Internet Archive (hosting record for a physical-copy scan)
• Publication: publisher's descriptive copy for Dictionary of Untranslatables: A Philosophical Lexicon, ed. Barbara Cassin (Princeton UP, English edition 2014); Internet Archive catalog record for the same book
• Date accessed: 2026-09-07
• URL: https://press.princeton.edu/books/hardcover/9780691138701/dictionary-of-untranslatables ; https://archive.org/details/dictionaryofuntr0000unse
• Type: secondary (publisher's own marketing/scope description, and a library catalog record) — not the dictionary's own index or table of contents, which remains unread (see Gaps). Neutral in stance.

Extract

• Directly fetched and read (not WebSearch-synthesized) from Princeton UP's own page: "This is an encyclopedic dictionary of close to 400 important philosophical, literary, and political terms and concepts that defy easy—or any—translation." "Drawn from more than a dozen languages, terms such as Dasein (German), pravda (Russian), saudade (Portuguese), and stato (Italian) are thoroughly examined in all their cross-linguistic and cross-cultural complexities." "The dictionary also includes essays on the special characteristics of particular languages—English, French, German, Greek, Italian, Portuguese, Russian, and Spanish."
• The Internet Archive catalog page for a scanned physical copy (dictionaryofuntr0000unse) independently gives the same "close to 400 terms," "more than a dozen languages" framing in its own jacket-copy description, and separately confirms the same eight dedicated-essay languages (English, French, German, Greek, Italian, Portuguese, Russian, Spanish) — corroborating the publisher's own page rather than adding a new claim.
• Neither source names Inuktitut, Eskimo-Aleut, any Arctic-Indigenous language, or any Indigenous-American language among the dictionary's covered languages or dedicated-essay languages. Both sources' example terms and named languages are exclusively European (with the dictionary's broader academic reputation, corroborated elsewhere on this project's file via tcs-mercier-cassin-review.md, being built specifically around the European philosophical tradition).
• A separate WebSearch pass returned a longer, more specific list of covered languages (adding Arabic, Basque, Catalan, Danish, Hebrew, Hungarian, Latin, Polish, Romanian) — this list is not filed as confirmed, per this project's standing caution against trusting WebSearch's own synthesized text without directly reading the underlying page: no single fetched source was found that states this exact expanded list, so it is recorded here only as an unconfirmed lead, not added to the two directly-read extracts above.

Gaps

• Neither the publisher's page nor the archive.org catalog record is the dictionary's own index or table of contents. The Internet Archive scan of the physical book itself carries the same access-restricted-item Controlled Digital Lending flag already documented for Thibert's dictionary and Sanders' book (see internetarchive-thibert-1997-access-restricted.md) — its full index remains unread by this project's tools, for the same reason.
• This source supports a strong inference (a dictionary scoped to "more than a dozen" named European languages, with worked examples and dedicated essays exclusively in European languages, is very unlikely to carry an Eskimo-Aleut headword) but does not amount to a direct, page-by-page index check. Treat as narrowing, not closing, the iktsuarpok/Cassin's-dictionary lead.

Used in

• Iktsuarpok
• Oodal — reused session 61 to narrow (not close) the same open Cassin's-dictionary-coverage question for a second word: Tamil/Dravidian is likewise absent from this dictionary's documented scope (a fixed list of European languages and dedicated-essay languages), so the same inference against inclusion applies. Not a fresh fetch — the underlying scope facts (covered/essay languages) are word-independent, so citing the existing extract rather than re-fetching is a legitimate reuse, not a new claim about oodal specifically.