Abhiman

• Language: an Indo-Aryan word attested across Bengali (spelled/pronounced obhiman, অভিমান), Odia (abhiman, ଅଭିମାନ), and Hindi (abhimān, अभिमान), all descending from Sanskrit abhimāna (अभिमान). This project's first case study from Odia specifically — a language distinct from, though closely related to, the Hindi/Urdu already covered by Jugaad — and its richest Bengali source material to date. Source; Source
• Popular gloss: "a degree of upset/grudge held against a loved one... a 'softer' form of anger, disappointment or hurt" (Urban Dictionary) — a relationship-specific emotion, directed only at someone close, combining hurt, sulking, and an unspoken hope that the other person will apologize. Source; Source
• Status: this project's twenty-fourth word case study. Its complication is a documented historical semantic split, independently corroborated by three separate sources: the same word's Sanskrit root and its modern Hindi reflex mean straightforwardly "pride, arrogance, ego" (a directly translatable, unremarkable concept), while the word's modern Bengali and Odia usage names a second, later-developed sense — hurt/sulk directed at a loved one — that a peer-reviewed corpus study of published literary translations shows genuinely resists any single, consistent English rendering. Unusually for this project, the "untranslatable" half of the claim is not just plausible but quantitatively measured against real professional translation practice. Primary-dictionary confirmation now covers all three of the word's attested languages (session 53 closed Odia, the last): three independent Hindi dictionaries and a 19th-century Marathi dictionary plus four aggregated modern Marathi sources all show only the pride/ego sense; a Bengali government dictionary lists the relational-hurt sense first, ahead of pride; and Praharaj's 1931–1940 Odia dictionary gives the relational sense as a separate, distinctly-tagged sub-entry ("conjugal sulks; silent conjugal displeasure due to jealousy") alongside a mostly pride/self-conceit Sanskrit-tagged sub-entry — the first source of any kind to treat the two senses as etymologically distinct rather than one sense drifting into another. Session 54 closed out the basic reference-work checks: confirmed absent from Collins and Oxford Learner's (via the r.jina.ai proxy, both direct 403s otherwise); no evidence found of coverage in Cassin's Dictionary of Untranslatables; the 2010 Telegraph India article remains unreachable after a fourth attempt (now a settled fence, not a quick win); and the Praharaj dictionary's abbreviations key was tracked down but turned out to be the wrong document (a source-citation key, not the grammatical-tag key), with OCR of the scan proving too unreliable to use. Session 55 closed the Cambridge gap (a dictionary-search-endpoint route, not the entry-page URL, gets past the silent-homepage failure and shows a genuine confirmed absence for both spellings) and found this project's first mainstream (non-crowdsourced) popular listicle for the word: a 2026-08-25 The Marginalian article by Maria Popova, drawing on Tiffany Watt Smith's The Book of Human Emotions (2015), glosses abhiman as "Sanskrit" and — notably — blends both documented senses into one undifferentiated definition (opens with "self-pride" but describes the pain/betrayal-by-a-loved-one sense throughout). The Praharaj grammatical-tag key remains unfound: this session ruled out two further candidate documents (the dictionary's own short bibliographic blurb, and its 33-page volume-1 front matter — visually sampled directly, not OCR'd, and found to be an English introduction, a patron photo gallery, and the compiler's own Odia-language autobiographical preface, no abbreviations table).

Meaning and etymology — one root, two documented senses

• Sanskrit abhimāna combines abhi- (toward) and māna (thinking/measuring/pride), and is glossed by a Sanskrit-usage dictionary as "conception," "pride," "false prestige," "desire," "an impression" — all senses drawn from classical Vaishnava devotional texts (the Chaitanya Charitamrita, the Srimad Bhagavatam). Source
• In the Hindu philosophical tradition (per Wikipedia's tertiary synthesis of specialist monographs), abhimāna denotes the ego's false identification of the self with "I" and "mine" — something to be transcended for spiritual liberation, though treated as a virtue in some worldly and aesthetic-theory contexts. Source
• A 2018 peer-reviewed corpus study independently confirms this "pride" sense is still alive in Bengali literary usage: analyzing 36 instances of obhiman across Rabindranath Tagore's short stories, novels, and dramas (in both native-English and native-Bengali translators' published renderings), Mukherjee & Das found "pride" used in 44% of English-translator instances and 27% of Bengali-translator instances — with English and Bengali translators of the same passage agreeing on "pride" in 9 of 10 such cases. The paper concludes "the script of pride and the original meaning of obhiman are very similar." Source
• The same study documents a second, later sense with no stable English equivalent: for the majority of instances not translated as "pride," translators reached for a wide, inconsistent array of words — wounded/injured/hurt pride, hurt, resentment, sulking, self-respect, grievance, offense, sensitive, indignant, petulance, disappointment — with English and Bengali translators of the identical original sentence frequently choosing different words from this list. In 60% of these non-"pride" cases, a given English word was used only once across the whole sample. The paper's own conclusion: "no English word appears to have a script close to the current usage of obhiman." Source
• A 2010 Telegraph India article, quoted secondhand inside the 2018 paper (the original URL is unreachable to this project — see Gaps), states the split explicitly: "The original meaning was pride, but the current usage — anger, or something close to it, at being upset with a loved one." The same article quotes Bengali poet Sankha Ghosh: "Try to translate it. Abhiman is not anger or sorrow. It is a temporary expression of a feeling about a loved one that he or she is supposed to address. People speaking in other languages must also be feeling the sentiment. I wonder how they express it." Source

The modern relational sense, in depth — a 2024 Odia study

• A 2024 qualitative study in the International Journal of Indian Psychology interviewed 13 Odia literature experts, writers, and folklore researchers about abhiman, explicitly filing it as an Odia word "we could not find [an] exact English equivalent" for. Its informants independently confirm the two-sense split found for Bengali: several stated that in Hindi, the word means only "pride," while in Odia its everyday and literary use is the relational-hurt sense. Source
• The study's seven identified themes converge on a single emotional shape: abhiman arises only toward a loved one with unmet expectations; it is expressed through sulking (Odia rushiba — withdrawal, silence, reduced eye contact, changed facial expression) rather than direct confrontation; it carries no intent to harm, only a hope that the other person will notice and apologize; and if well resolved, it can leave the relationship closer than before. One informant called it "a soft weapon"; another, "a soft anger" (madhura krodha, literally "sweet anger"). Source
• The study's own illustrative example is drawn from Indian mythology: when Krishna breaks a promise to meet Radha, she does not confront him with open anger but withdraws — refusing to dress up or see him — until he apologizes, which the paper offers as a canonical instance of abhiman in action. Source
• Independent, non-academic corroboration: a crowd-submitted Urban Dictionary entry for the Bengali spelling ("obhimaan") describes an unrelated-author gloss strikingly close to both academic studies' findings — "a degree of upset/grudge you hold against a loved one... a 'softer' form of anger, disappointment or hurt" — converging from a third, wholly independent source. Source

Primary dictionary confirmation of the split — Hindi, Marathi, Bengali, and Odia

• Three independent Hindi dictionaries checked this session (Hindwi, Rekhta, HinKhoj) each give अभिमान/abhimaan exclusively pride/ego/arrogance/self-respect/vanity senses — none records any sense of sulking, hurt feelings toward a loved one, pique, or grievance. This is the first direct, primary-source confirmation of the Hindi side of the split (previously known only secondhand via both academic papers' own literature citations, per Gap below and NEXT.md). Source; Source; Source
• A Bengali-to-English dictionary published by the Government of Bangladesh gives অভিমান three senses in a single entry, in this order: (1) "offended state of mind; pique; sensitiveness... caused by undesirable conduct of a near and dear one," (2) "pride; conceit; egoism," (3) "sense of self-respect." This is the first primary dictionary record, for any of this word's three languages, to carry both senses under one headword — and its ordering, with the relational-hurt sense listed before pride, is consistent with (and strengthens) Mukherjee & Das's (2018) corpus finding that Bengali translators reach for "pride" in only 27% of instances, well under the English translators' 44%. Source
• Odia is now also primary-dictionary-confirmed, session 53: Praharaj's 1931–1940 Purnnachandra Odia Bhashakosha (queried directly via the same Digital South Asia Library query-URL pattern used for Marathi's Molesworth) gives ଅଭିମାନ as two separately tagged sub-entries under one headword — a Sanskrit-tagged block of twelve senses running mostly pride/self-conceit/haughtiness/arrogance (with "discontent" and "sensitiveness... to slight matters" already hinting at the relational sense within this block), and a second, distinctly tagged sub-entry (likely "deshaja," vernacular-derived — see source page for the inference caveat) giving "conjugal sulks; silent conjugal displeasure due to jealousy; wounded love or wounded vanity," "secret dissatisfaction," and "prestige; honour." This is a genuinely new structural nuance beyond Hindi/Marathi/Bengali: rather than one sense drifting into a second within a single entry, Praharaj's dictionary treats the two senses as etymologically distinct sub-entries — and narrows the relational sense specifically to a conjugal/spousal frame (jealousy between spouses/lovers), narrower than Bag & Dash's (2024) broader "loved one." The entry also supplies a literary usage instance from Krishna Singha's Odia Mahabharata (Vana/forest section), not yet cross-checked against any English translation. Source
• Marathi is now also primary-dictionary-confirmed, session 52: Molesworth's 1857 Marathi and English Dictionary (queried directly via the Digital South Asia Library's digitized edition) gives अभिमान only "pride, haughtiness, conceit, opinionativeness," a possessive/self-regarding sense ("apprehension or view of as self... maintaining or holding with engagedness of heart and soul"), and a claims/pretensions sense — closing with "corresponds well with our words Honor, proper pride, lofty sense of propriety, noble feeling." A second, independent aggregation (transliteral.org) reproduces three further modern Marathi sources under the same headword — a Marathi synonyms dictionary, Marathi WordNet, and the Maharashtra Shabdakosh — all likewise giving only pride/ego/self-regard/ownership senses. Across all four Marathi sources (one 19th-century, three modern), none records any sense of sulking, hurt feelings toward a loved one, pique, or grievance — Marathi patterns with Hindi, not with Bengali/Odia, strengthening rather than merely maintaining the split as a genuine Hindi–Marathi/Bengali–Odia divide rather than an idiosyncrasy of Hindi alone. Source; Source

Translator practice — this project's most rigorous data point to date

• Most of this project's prior translator-practice findings rest on one or two literary instances (see, e.g., Saudade, Lagom). Abhiman's case is different in kind: Mukherjee & Das's (2018) 36-instance, dual-translator-group corpus study is a quantified measurement of translation consistency, not an anecdote — and it directly supports the "two senses, one translatable and one not" claim with numbers rather than impression. Source
• The study's own interpretive proposal is that this is a genuine case of historical semantic drift observed through a translation-consistency lens: an original, cross-culturally ordinary sense (pride) gave rise, over time, to a second, culturally specific relational sense that native-English monolinguals apparently never needed a dedicated word for — paralleled, the authors argue, by English's own internal split of temper into temper and temperament, and the semantic narrowing of tantrum, both documented by Stearns & Stearns (1989) and cited secondhand in the paper. Source
• A companion word studied in the same paper, lajja (shame/embarrassment/shyness), acts as the paper's own internal control case: both translator groups converge heavily on "shame" (making it a translatable emotion word by the same measure), sharpening rather than diluting the contrast with obhiman's inconsistency. Source

Dictionary adoption

• No Merriam-Webster entry for "abhiman" or "obhiman" — confirmed by direct fetch, HTTP 404, this session.
• No English Wiktionary entry for either spelling — confirmed by direct fetch, HTTP 404, this session.
• Dictionary.com: neither spelling resolves to a dedicated entry (HTTP 301 redirect for "obhiman," the same "no entry" pattern this project has treated as a confirmed absence for other words) — checked this session.
• No confirmed OED entry: a direct search of the OED's own site via the standing r.jina.ai-proxy method returned no result for "obhiman" — treated, per this project's standing convention, as a non-exhaustive "no entry found," not a confirmed absence.
• No Collins entry: direct fetch returned HTTP 403, but the r.jina.ai proxy got through this session and shows a clear "Sorry, no results for 'abhiman' in the English Dictionary" message with spelling-suggestion alternatives (antiman, ashman, Ahriman, etc.) and an invitation to submit it as a new word — a genuine confirmed absence, not a silent wrong-page failure.
• No Oxford Learner's entry: direct fetch returned HTTP 403, but the r.jina.ai proxy got through and shows "Word not found in the dictionary" with spelling suggestions (subhuman, human, inhuman, airman) — a genuine confirmed absence.
• No Cambridge entry: direct fetch of the entry-page URL still silently redirects to the dictionary's generic homepage (the same failure mode documented sessions 51/54), but this session found that Cambridge's own search-endpoint URL, fetched via curl with a browser user-agent, redirects instead to a genuine spellcheck/"not found" page for both "abhiman" and "obhiman," listing spelling suggestions (adman, airman, bagman, human...) — a clean confirmed absence, closing the standing gap. Source

Cassin's dictionary

• No evidence found that Cassin's Dictionary of Untranslatables covers this word. A direct web search for "abhiman" alongside the dictionary's title returned nothing; the dictionary's own documented scope (Franco-German-English-centered philosophical vocabulary, with a review noting a few individual extensions to Russian, Portuguese, and Italian terms — see the Cassin concept page) makes coverage of a South Asian relational-emotion word unlikely on its face, but this is an inference from scope, not a direct check of the dictionary's own entry list or index — which remains unconsulted for this word, as for every other word on file (see standing item in NEXT.md Fences).

Popular "untranslatable words" genre presence

• Found in one dedicated crowdsourced "untranslatable words" database (Eunoia), tagged Bengali, with a gloss closely matching — likely descending from — the same 2010 Telegraph India quote reproduced inside the academic literature. This is a thinner popular-genre footprint than this project's headline cases (hygge, saudade), closer in shape to Sankofa's single-source genre presence, though here the academic backing is far stronger than sankofa's. Source
• This project's first mainstream (non-crowdsourced) listicle for the word, found session 55: The Marginalian (Maria Popova), 2026-08-25, "The Vocabulary of Vulnerability: 10 Wonderfully Untranslatable Words," drawing on Tiffany Watt Smith's 2015 book The Book of Human Emotions. Tags the word simply "Sanskrit" (unlike every other source on file, which specifies Bengali or Odia) and defines it as a blend of both documented senses at once — opening with "self-pride" but describing "the pain and anger caused when someone we love... hurts us," framed as a betrayal of "unspoken contracts of love and respect" — without acknowledging these as two historically distinct senses. The article postdates every prior session's search (published roughly two weeks before this session), which likely explains why it wasn't found earlier. Source

Gaps

• The 2010 Telegraph India article this project's two academic sources both draw on is still unreachable — now confirmed across four attempts spanning two sessions. Session 51 tried a direct fetch (failed outright) and the r.jina.ai proxy (silently returned the site's live homepage instead of the 2010 article). This session repeated the r.jina.ai proxy attempt and got the identical failure (the site's September 2026 homepage, not the article, not an error) and separately searched for a syndicated mirror or independent quotation of the article by name/URL/content — the only material found was the same secondhand quotation already filed via Mukherjee & Das (2018), including one apparent new lead (a ResearchGate hit for "Emotion without a Word: An Analysis of Bengali Emotions and Their English Translation") that turned out, on inspection, to be another mirror of the very same Mukherjee & Das (2018) paper already on file (confirmed by matching authors, journal, and findings), not a new source — its ResearchGate page and direct PDF-asset URL both returned the standing CAPTCHA/403 wall regardless. Its "original meaning was pride... current usage... anger" framing and the Sankha Ghosh quote remain filed here as secondhand via Mukherjee & Das (2018), not independently confirmed against the original news article. A future session should try a different route entirely (a newspaper-archive/library database, or a search-engine cached-page viewer) rather than repeating direct fetch or r.jina.ai against this specific URL again.
• No standard scholarly Sanskrit lexicon (Monier-Williams, Apte) was successfully consulted directly — three different access routes (an INRIA-hosted digitized edition, wisdomlib.org, and an ibiblio.org page-image mirror) all failed this session (see NEXT.md Fences). The etymology given here rests on a smaller usage-based aggregator site instead.
• Hindi is now independently confirmed via three primary dictionaries (see above) — closed, no longer a gap. Bengali is now independently confirmed via one primary government dictionary (see above) — substantially narrowed.
• Marathi is now closed, session 52 (see above): a primary 19th-century dictionary (Molesworth, via DSAL) and three aggregated modern sources all confirm only the pride/ego sense, no relational-hurt sense — matching Hindi, not Bengali/Odia.
• Odia is now closed, session 53 (see above): Praharaj's 1931–1940 dictionary, directly queried, gives the relational-hurt sense explicitly as a separately tagged sub-entry ("conjugal sulks"). All three of the word's attested languages are now primary-dictionary-confirmed on both sides of the split; the only dictionary-side gap remaining is whether a modern Odia dictionary/aggregator (of the kind transliteral.org supplied for Marathi) corroborates this one 1931–1940 source (see Praharaj source page's own Gaps).
• The Praharaj dictionary's own abbreviations key was located and fetched session 54 but turned out to be the wrong document: frontmatter/abbreviations.pdf is a list of source dictionaries/works and their citation abbreviations (e.g., the kind of key that would confirm "M.W." = Monier-Williams, already used inline in this entry's senses) — not the short grammatical/etymological category tags (ସଂ., ଦେ.) actually used before each sense. This page's reading of ସଂ./ଦେ. as "Sanskrit"/"Deshaja (vernacular)" therefore remains a project inference, unconfirmed. Two further candidate documents ruled out this session (55): the dictionary's separate "About"/frontmatter/praharaj.pdf document is a two-paragraph bibliographic blurb (from Andrew Dalby's A Guide to World Language Dictionaries), not an abbreviations key; and frontmatter/frontmatter1.pdf (the dictionary's own 33-page volume-1 front matter, linked from new_frontmatter.html alongside six further per-volume front-matter PDFs, frontmatter2.pdf–frontmatter7.pdf, all still untried) is a scanned, image-only PDF — read directly via pdftoppm-rendered page images and visual inspection (not OCR, avoiding the standing OCR-reliability caution) across roughly a third of its pages including the entire closing section — containing only an English-language introduction to Oriya/Indo-Aryan linguistic history, a gallery of patron/subscriber portrait photos, and the compiler Gopal Chandra Praharaj's own Odia-language autobiographical preface (his career, the dictionary's compilation history, subscriber and honors lists) — no grammatical/abbreviation table anywhere in the sampled pages. If the tag key exists as a distinct document at all, it is not among the three most likely candidates now checked; the six other per-volume frontmatter PDFs are the only remaining untried lead, but there is no strong reason to expect a volume-specific document to carry a key that logically belongs once, in general front matter — so this lead should be treated as low-yield unless a future session finds a more specific pointer to it. The Krishna Singha Mahabharata quotation remains an unverified single literary instance — no dedicated new search for an English translation of this specific rendering was made this session (see queue).
• Cassin's dictionary: searched this session, no evidence found (see Cassin's dictionary section above) — a non-exhaustive negative result, since the dictionary's own full entry list/index remains unconsulted directly, as for every other word on file.
• No dedicated search was made this session for a mainstream (non-crowdsourced) "untranslatable words" listicle featuring this word specifically.

See also

• Jugaad — this project's only other Indo-Aryan/South Asian case (Hindi/Urdu); together the two words show that "South Asian" is not a single case shape: jugaad's complication is a live public contronym (virtuous ingenuity vs. corruption euphemism) argued out in the same language community, while abhiman's is a historical semantic split between one Indo-Aryan language's usage (Hindi, retaining the older sense) and two others' (Bengali, Odia, carrying the newer one).
• Toska — this project's other case resting partly on a single celebrated author's own literary corpus (there, Chekhov and Nabokov's gloss of him; here, Tagore) as the primary evidence base for translator practice.
• Mianzi — this project's other case where an academic source directly measures translation/rendering choices against real bilingual usage data, rather than relying on impressionistic reports.
• Litost — like abhiman, a word whose popular "untranslatable" reputation and its actual dictionary/lexical history turn out to be two different things worth separating (there, a novelist's own definitional prose against a plainer dictionary sense; here, a documented historical drift from one plain sense to a second, harder one).
• Oodal — this project's first Dravidian-language case study, prompted directly by this page's own Note 3 (the abhiman source paper naming Tamil as a language with a related concept); oodal's relational-sulking territory overlaps abhiman's Bengali/Odia sense, but oodal is a named technical category inside a formal classical poetic system rather than a plain, cross-language-contested dictionary sense.