I’ve read and heard two rival etymologies before. The first is that ‘drunk as a Lord’ became ‘drunk as a blood’ (from the idea that a ‘blue blood’ is a Lord or noble) which became ‘bloody drunk’ and then it later started to be used as a general intensifier. The second theory is that it’s from ‘by our lady’, referring to Mary, mother of Jesus. The first of these seems more plausible but, in any case, I doubt it’s a contraction of ‘by lady’ without the word ‘our’ in between ‘by’ and ‘lady’.Overlordnat1 (talk)08:03, 1 October 2025 (UTC)Reply
That does not account for my username,bloody beginner.
I did a spit-take once seeing that Danish phonology contains ablødt d,blød(“soft”), from*blautaz, because it sounds likeblöd(“dull, silly”), from*blauþaz, same meaning but from a notably different root, which is closer toμαλάκας.
I imagine that some connotation of weak, small, and effiminate could eventually give rise to the b'lady theory. As for the allusion toblue, seefair, eventuallymy fair lady.bloody mary, now that I can see refer to its viscosity and color, naturally.Blutiger Anfänger (talk)21:30, 1 October 2025 (UTC)Reply
Latest comment:1 month ago2 comments2 people in discussion
Is this Ukrainian term attested in the 1800s at all? Because then I can possibly determine whether thePannonian Rusynдїєслово(djijeslovo) was modelled directly off of that, or borrowed from an earlier Carpathian Rusyn cognate of the Ukrainian term.Insaneguy1083 (talk)08:52, 2 October 2025 (UTC)Reply
That's been there since April of 2019, and the reason seems to be that thereis no word in Proto-Austronesian, as asserted byCoconut on Wikipedia.Wikipedia. The genetic evidence seems to show that the type associated with the Austronesian-speaking peoples was domesticated in the Philippines by the early Malayo-Polynesian-speakers after they left Taiwan/Formosa.Chuck Entz (talk)04:49, 3 October 2025 (UTC)Reply
Latest comment:5 days ago14 comments7 people in discussion
Seeأَبَو(ʔabáw, ʔabaww). It's used primarily among Armenians from what I can tell, to the point where one of the quotes I added claims it meanswow effect "in Armenian", but an Armenian friend online tells me the "aw" sequence is foreign to Armenian (even Western). Any ideas?Still, when you think about it (talk)19:52, 3 October 2025 (UTC)Reply
Huh I asked an Arab friend from Lebanon. He doenst recognize that word. I guess it’s an Arabic word that Lebanese Armenians use in their speech, perhaps because of community ties with Syrian Armenians. But I’d still classify it as an Arabic word and not an Armenian word.Hovsepig (talk)20:31, 3 October 2025 (UTC)Reply
Thanks! Very confusing. It has no apparent Arabic etymology either. The only hint so far is that one of the quotes I added refers to it as an Aleppine term without making reference to Armenians specifically, but given that Aleppo has the largest Armenian population in Syria that still doesn't isolate anything...Still, when you think about it (talk)21:29, 3 October 2025 (UTC)Reply
Something like /waʊ/ or /waw/ is found in a lot of languages as onomatopoeia, with its contrast of high-rounded and low unrounded sounds. The current form would seem to be the result of this passing between languages with mutually-incompatible phonotactics- Arabic in general has no problem with initial "w" sounds, but Armenian seems to convert them into "v". Would the local Arabic lect convert that into a "b"?.Chuck Entz (talk)23:20, 3 October 2025 (UTC)Reply
I was talking about pronunciations. Orthographic ⟨b⟩ and ⟨v⟩ both are pronounced with the phoneme/b/. The North Levantine Arabic pronunciation with a/b/ might be explained by borrowing from Castilian or Occitan instead of from Italian. ‑‑Lambiam20:03, 8 October 2025 (UTC)Reply
I wondered whether there could be some connection to the Arabic expletive "ʔabu", but maybe the semantics don't really make sense...Wakuran (talk)22:10, 8 October 2025 (UTC)Reply
This isn't a bad idea at all either.أَبُوه(ʔabū́, literally“his father”) actually feels really intuitive to me to use as an interjection, and it would traditionally be pronouncedʔabáw in much of rural Lebanon. The only problem would be determining whether the areas that historically exhibit this-ū(C) >-aw(C) diphthongization (off the top of my head Zgharta, the entire(?) Beqaa, and the Shiite south) are the same areas where "abaw" is attested or where there are Armenians, and honestly I don't know if there's a lot of overlap there. There are many Armenians in the Beqaa, but there are also many in Beirut (where this diphthongization hasn't been attested), and the term is also found in Aleppo (where there are again a lot of Armenians but no traditional diphthongization).Still, when you think about it (talk)22:21, 8 October 2025 (UTC)Reply
┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Maybe there's a chance it's from Ottoman Turkish, which Vahag showed me has an interjection of similar force with variants likeabov andabav, exported in modern times as Swedishabow and apparently identically into colloquial German as well. Compare Tatarabaw. I'm wary of proposing an exact loanword origin for a sound that could totally just have arisen on its own like Chuck Entz was saying (to that point, check out the unrelated Hanunooabaw), but I'm drawn to this Ottoman Turkish idea specifically because the Arabic term is best known among Armenians and Aleppines and little-used elsewhere (I had never heard of it, for example).Still, when you think about it (talk ·stalk)23:59, 31 October 2025 (UTC)Reply
According to Sevan Nişanyan, it can be attested as early as the 14th century in the Kipchak dictionary/grammar guideet-Tuhfetu'z-Zekiyye fi'l-Lugati't-Türkiyye as two separate exclamations transcribed asabaw for surprise/shock andabo/abû for fear/panic. It's no surprise it's present in Tatar and presumably other Turkic languages in the area. In particular, Armeno-Kipchak once spoken in Crimea is a likely candidate for the medium for the borrowing, although I don't know enough about it to make an educated guess as to which direction.
Anecdotally, I can say that it's associated with the Turkish dialects spoken in Cilicia, which makes me slightly more inclined to believe it was borrowed from Armenian.~2025-34469-84 (talk)02:03, 18 November 2025 (UTC)Reply
well, /e/ to /je/ was a pretty common sound change, so i assume because of the sound change, people assume it underwent changes while evolving from latin instead of having been borrowed, leading to the belief that it's inheritedTooSimilarT0DaFollowingUsername (talk)18:48, 4 October 2025 (UTC)Reply
In Spanish, the reason for saying that -ente is borrowed is because the regular inherited outcome of Latin "e" in stressed syllables is "ie" (aside from certain exceptions), so a word like "contenente" should have been "conteniente" if it was inherited: compare "conteniendo". However, there are a few complications that would be good to clarify in regard to-ente,-iente,Category:Spanish terms suffixed with -ente andCategory:Spanish terms suffixed with -iente.
First, "ye" is regularly used in place of "ie" after a vowel (compare incluyendo, atrayendo). So in my opinion, forms likeincluyente andatrayente, currently categorized as ending in -ente, should actually be marked as ending in the suffix -iente. Do others agree?
Second, verbs likedormir (from Latin dormīre) could be considered to have a stem ending in -i-, and so words likedurmiente (compare Latindormientem) could theoretically be divided as durmi-ente. But I think it is easier and feels more consistent to categorize these as ending in -iente.
As for Italian, while it also had the e > ie change in open syllables (at least in some words), I think it did not generally occur in closed syllables like it did in Spanish (compareviento andvento,tierra andterra,ciento andcento). So in terms of form, I don't know of a reason why Italian-ente could not be inherited.--Urszag (talk)20:12, 4 October 2025 (UTC)Reply
I can't say for sure whether it makes more sense to interpret verbs likeincluir as having an underlying stem ending in -u- or -uy- (likewiseleer, etc.). If we look at cases liketraer/atraer, where tray- is found only in the gerund, that suggests thatatrayente at least is probably best analyzed as atra- + -iente. Furthermore, in the gerund, I don't believe we ever see -endo after consonants except for -y-: assuming we analyze the -y- as stem-final, that suggests that the sequence -yie- (which seems to not really be allowed according to Spanish phonotactics) is simplified to -ye- in verb inflection: compare also the preterites of verbs like comer/comió, recibir/recibió, incluir/incluyó (not incluyió). So regardless of whether the first component of incluyente is inclu- or incluy-, I think the second element can be -iente, and I think this inherited suffix is more likely than the borrowed suffix -ente because we know words ending in -yente were not borrowed as a whole from Latin forms: Latin doesn't use -y- as a consonant.--Urszag (talk)00:50, 9 October 2025 (UTC)Reply
you mean -endo?
i believe that -yente feels like a variant of -iente because of the gerund doing the same thing with -yendo and -iendo
the adjective "estupendo" came from a latin participle, but that isn't grammatically used as a gerund, so we can assume it's borrowed (plus, in spanish, gerunds don't decline, yet estupendo does decline)
plus, there is no *estupir or *estuper that we can model this off of anyway, so considering that there isn't even an infinitive of the verb the adjective is based off of, and that there are other borrowed derivations of the verb (i.e. estúpido), we can say that it's borrowed.TooSimilarT0DaFollowingUsername (talk)15:55, 9 October 2025 (UTC)Reply
Latest comment:1 month ago1 comment1 person in discussion
Hobo is short for hoe boy. It refers to an agricultural worker that travels by foot carrying a hoe, which is his main work implement. They were often seen carrying a small bundle of items which were tied to the hoe end of the tool handle and slung over the shoulder. This accounts for the oft mentioned relation to the term bindlestiff.74.254.4.12216:52, 7 October 2025 (UTC)Reply
Our entry for Arabicجَدْوَل(jadwal) says that Malayjadual and Indonesianjadual,jadwal are descendants, so ifjudul is also from that word, they should be marked as{{doublet}}s. It would also be good to explain why the“title, heading” word has different vowels from the“schedule” word if they're both borrowed from the same Arabic word. —Mahāgaja ·talk17:53, 8 October 2025 (UTC)Reply
For the wordtì in Vietnamese and its meaning "dollar, buck" (I'm not even sure if it have broader meaning)
Latest comment:1 month ago1 comment1 person in discussion
I seen the using of the wordtì in the translation of the novelThe Catcher in The Rye, publish first time in 1964-1965, with a typical old writing style and a strong Southern Vietnamese vibe at the time. I can't find a single definition for this meaning of the word in any dictionary, from the one I had in 1980s to current online dictionary.Queen Duck שא (talk)03:03, 8 October 2025 (UTC)Reply
Latest comment:1 month ago1 comment1 person in discussion
@DDG9912 Regarding "arkib", I think it's probable that it was borrowed from English "archive" because it's not unheard of for Malay to adapt English words with the "price" vowel as /i/ or /e/ as we can see in words like "lésén" (from "licence"), "Palestin" (spelled before as ڤاليستأين (Paléstain)) and "prébet" (from "private"), and adaptation of English final "-ve" as "-b" in Malay can be found in other words too in words like "rizab" (from English "reserve").
In fact, this adaptation of that "price" vowel as /i/ or /e/ can be found in loanwords from other languages as well such as the Arabic names "Husin" and "Syuib" which came from "Husayn" and "Shu'ayb".
(Note: In the first paragraph I'm only referring to English loanwords that entered Malay before around 1972 when English loanwords started being spelt to reflect their etymology rather than based on how Malay speakers actually pronounced them)
Latest comment:1 month ago3 comments1 person in discussion
This one has puzzled me for a while. Is it some other language's word for pinky, or for coffins, or for cutting someone off? All I've been able to find is a 1973 Lebanese play by the nameسنكف سنكف(“Singof Singof”), but the title uses it as an already-established term and I don't have access to the script to see if there are any hints. There's also a reasonable chance it was originally with/k/, as in*sinkuf sinkuf or*sinkif sinkif, if those forms look at all usable... also, this transcription convention is somewhat opaque, and it might be more familiarly spelledsengof orsingof or the like.Still, when you think about it (talk)19:45, 8 October 2025 (UTC)Reply
Forgot to clarify that I'm wondering about coffins because of the rhymeسنكف سنكف عالتابوت(singuf singuf ʕa t-tābūt, literally“singof singof on a/the coffin”). Any relation to zinc as a coffin liner...?
A forum thread mentions that the associated pinky gesture is found outside of the Levant, just with the Arabic verbحَارَب(ḥārab) instead ofسَنْكَف(sangaf), which to me says it must be a new loan applied to an old practice.
Considering new loans, the only crackpot connection I can think of is to Russianцинков(cinkov), which is a match in form but a stretch semantically. (In light of the military sense listed at Russianцинк(cink), I suppose it's worth noting that Russian military terms are well-known from the days of the civil war in Lebanon --كْلَاشِنكوف(klāšinkōv, klāšinkōf,which is entrenched enough to have a familiar abbreviation I forget...klāšin?) -- but I don't know if that helps.)Still, when you think about it (talk)19:54, 8 October 2025 (UTC)Reply
Like many of the contributions of this editor, this shows a great deal of knowledge and research, but odd lapses in details and no sources. The Arabic term here is a redlink because it was created as a Latin-script entry with no headword and no sources that might be consulted to fill in necessary details. I hope they're not using AI to create entries in languages they don't know.Chuck Entz (talk)14:51, 9 October 2025 (UTC)Reply
Latest comment:1 month ago14 comments3 people in discussion
The page for Turkishşınav (push-up) says it comes from Englishchin-up, however I can find only one source[2] for this and several (Nişanyan[3], EtimolojiTürkçe[4], Ak Sözlük[5]) that says it comes from Persianشناب orشناو meaningsidestroke. I'm inclined to trust the first source (the Turkish Language Association) on most things, but <ç> to <ş> feels unlikely and there are mutiple sources that contradict it. There's an IP edit from 2018 saying that chin-up "fits neither semantically nor phonetically and is almost certainly nonsense", which I agree with, though "sidestroke" to "push-up" also feels somewhat strange to me. Unless anyone objects or gives other sources I'm going to replace the current etymology with the one from Persian.Wreaderick (talk)13:41, 10 October 2025 (UTC)Reply
Why would Ottoman Turkishشناو(şinav,“the act of swimming”) become modern Turkishşınav(“push-up”)? As far as I can tell, the former does not survive in modern Turkish, and the latter pops up in 1980s.Vahag (talk)14:17, 10 October 2025 (UTC)Reply
I don't know how to do rigorous etymological research, but looking up "şınav etimoloji" only gives sources supporting the Persian origin. Where did you find that the modern meaning arose in the 1980s? I do agree that the semantic change is tenuous, but I feel you could say the same for the English origin too. Also just Googling "شناو" gives me images of push-ups, nothing about swimming, though Ak Sözlük says the <av> is from Persianآب (water). Do you have any sources?Wreaderick (talk)14:52, 10 October 2025 (UTC)Reply
I searched on Google Books. To me, the etymological proposals have no value until someone does a rigorous philology: where and when doesşınav(“push-up”) first appear? The Persian borrowingşinav would not lay dormant and suddenly come back to life in 1980s in a different form and meaning.Vahag (talk)17:20, 10 October 2025 (UTC)Reply
It is just Classical Persianشنا(šinā,“swimming”), see its many alternative forms, one of which includesشناو(šināw), without prejudice to more specific meanings such as specific swimming moves, and related terms some of which we added. Arabicشَنّ(šann,“waterskin”), to be added because related terms sections generally only allow the same language, and a whole article by Buyaner 2006 cited on both leads you farther but I don't have it (1980–2007 ofthe journal missing …) and only used the journal piece from previews.
Why does Googling شنا give images related to swimming but شناو only gives push-ups? I don't see how the two meanings are related. It seems that the meaning didn't originate in Turkish but rather Persian. I would have found it strange for a word that meant swimming in Turkish to suddenly take on a whole different meaning in recent decades with such little documentation. Maybe there's some nuance I'm unaware of in Persian that could lead to the alternation.Wreaderick (talk)17:29, 10 October 2025 (UTC)Reply
I did not say anything contrary to this, it was only about the etymological identification, where ever semantics developed. Some of the alt-forms developed alternative meanings in Persian, and exercise names, together with the regulization of their sports such as swimming, altogether were standardized only in recent decades, probably after the Ottoman era though even in this case, and the base-meanings, by reason of which they are in the alternative forms section, belong to Classical Persian – like 1½ century ago thewoodlouse had dozens of names across Britain and now all are unrecognizable and thus do not suit image searches.Fay Freak (talk)17:54, 10 October 2025 (UTC)Reply
@Wreaderick I added the earlies quote (1986) I could find toşınav. Please check the spelling and translation, I used OCR and machine translation. As you can see, the term was unusual enough in 1986 that the author had to explain it in parentheses.Vahag (talk)19:42, 10 October 2025 (UTC)Reply
This of course means Ottoman Turkishشناو(şinav,“the act of swimming”) did not become Turkishşınav(“push-up”) since we did not have Turkishşınav(“**swimming”) at that1970–1980s calisthenics craze. The same word was reborrowed again from Iran, where the form with the meaning “swimming” was not wholly dead everywhere, since an origin in Englishchin-up would not only be phonologically, but also semantically too ignorant an invention to be possible in reality (of the 1980s, with the education even of meatheads), even more so thanpull-up, given the supinated grip of a chin-up which cannot be mirrored in push-up position (and the actual mirror of a push-up, working the same muscles flipping the body around, is a bench press).Fay Freak (talk)20:06, 10 October 2025 (UTC)Reply
Hmm. Is a Persian borrowing that late possible? Perhaps it is from the Kurdish cognate of the Persian word. I have ordered big, fat Kurdish dictionaries from Turkey. Will check when they arrive.Vahag (talk)20:22, 10 October 2025 (UTC)Reply
Slyly I wrote without prejudice to the language. What I know is that you Caucasians get your anabolic steroids from Iran. I have had some swole internet acquaintances who when emigrating from Russia in 2022 choose Armenia or Georgia for this consideration. This scene of course leaves little trace in written history, or where are the Turkish muscle magazines?Fay Freak (talk)20:47, 10 October 2025 (UTC)Reply
Hey, first of all I'd like to thank you for your addition of the quote, I had found the same quote when you mentioned you had looked at Google Books. The translation looked fine, I only had to fix a minor OCR typo.
I do find a Kurdish borrowing quite unlikely; Turkish for some reason hasvery few Kurdish borrowings, most of them are place names or clearly Kurdish-context words (one of the ones that isn't I even added myself). Should I expand the etymology with the information we have now while making it clear that it is uncertain? I also find Persian borrowings at such a recent date unusual. Are there any Persian-speaking editors you know of who would be available to ask about the meaning of شناو and if it has any meaning related to swimming? ThanksWreaderick (talk)09:45, 11 October 2025 (UTC)Reply
@Wreaderick: This is fromBurhān-i Qāṭiʕ copied into Vullers already linked in the entry and under thesame entry inDihxudā with quotes; the Prophet himself is ascribed the wordsبیاموزید فرزندان را تیراندازی و شناو.(“Teach your children archery and swimming!”). I don't know why you cast doubt upon it, when I already told you that image searches are slanted in disfavour of historical usage, though in turn they are good positive indicators that a term is known. We always watch for a reliable chain of transmission.Fay Freak (talk)12:00, 11 October 2025 (UTC)Reply
Latest comment:1 month ago2 comments2 people in discussion
The etymology claims the word was inherited from Maghrebi Arabic, which is a dialect continuum. Furthermore, Libyan Arabic is part of this dialect continuum. I can't understand how a word could be inherited from a dialect continuum, perhaps the editor meant that the word traces its root to the proto-language that the dialect continuum descends from, but that claim seems completely unverifiable as Libyan Arabic (and most other dialects in the dialect continuum) are solely vernacular dialects with no written history. The word in question (fakruna) has various cognates (often with the same meaning) in a lot of the dialects in the dialect continuum. I fins this claim in the etymology questionable/incorrect, or perhaps I might just be struggling to understand the claim. Any clarification/guidance would be much appreciated :)CasualDudeeee (talk)15:32, 13 October 2025 (UTC)Reply
It spread around at some point of course. At that point, inMiddle Arabic if you want, spreading around Maghrebi Arabic was easier and was not felt like borrowing (it felt like intralanguage borrowing of idiolects and familects rather than like interlanguage borrowing) because regiolects had not had the time to develop apart so much as they are now.
Then, editors that are not acquainted with any Berber language are avoidant of noting any forms, they could miswrite, and they can't add the Arabic dialects in the descendant section—what they actually want to do—of any specific Berber language header ever entered, since the situation of Medieval Berber is even more muddled.Fay Freak (talk)23:26, 13 October 2025 (UTC)Reply
I assume it's supposed to be pronounced serk-horal, although it's easy to parse it as serch-oral. Not the most intuitive orthography, arguably.Wakuran (talk)00:37, 15 October 2025 (UTC)Reply
Latest comment:1 month ago1 comment1 person in discussion
RFV of the etymology. The OGP etymon is attested, so Portuguese and Galician surely come from it, but the Latin part is unsourced and differs between entries: Portuguese mentions Latinmarculātus, frommarculus(“hammer”), which doesn't seem to be attested; Galician only mentionsmarculus and cites DCECH, but it doesn't have this etymology; and Spanish goes for a completely different origin.Trooper57 (talk)19:05, 15 October 2025 (UTC)Reply
Latest comment:1 month ago2 comments1 person in discussion
Kloekhorst, Beekes, Kortlandt, Martirosyan, and Kroonen all prefer to reconstruct the term with an initial*h₃ laryngeal. This allows for the term to be connected with a root*h₃erǵʰ-, which itself could be the source for*argaz. However, the reconstruction*h₁órǵʰis allows for a connection to the root*h₁erǵʰ-, whenceAncient Greekὀρχέομαι(orkhéomai).
Kroonen reconstructs a root*h₃erǵʰ-(“tocopulate”), which may be the source ofProto-Germanic*argaz,Lithuanianeržùs,aržùs(“ardent, voluptuous, stubborn”), andHittite[script needed](arkari,“to copulate”). Kloekhorst connects the Hittite verb to a hypothetical pre-form*h₃ṛǵʰ-o ~ h₃órǵʰ-ey, itself from the same root. However, LIV connects the term to root*h₁erǵʰ-, whence also perhapsAncient Greekὀρχέομαι(orkhéomai). Beekes claims that the wordὀρχέομαι(orkhéomai) primarily means "tomount" in inscriptions, although it is also attested with the meaning of "to dance." LSJ make no mention of this sense of "to mount," which seems strange given how comprehensive their dictionary is otherwise. Apparently, the linguist Calvin Watkins published an article claiming that the word had a more sexual meaning in inscriptions, although I cannot read it because it is in French.[1] Nevertheless, Beekes reconstructs a root*h₁erǵʰ-(“tomount”) forὀρχέομαι(orkhéomai), although he doubts the connection withὄρχις(órkhis) as he prefers the reconstruction*h₃orǵʰis for the latter.
Semantically, a connection between the words*h₁órǵʰis and*h₁erǵʰ-(“tomount”) seems possible, although perhaps somewhat dubious. Interestingly, the possible Hittite cognate[script needed](arkari,“tomount (sexually),copulate”) shares the same meaning of "to mount." Moreover, a semantic development from "to mount" to "to copulate" seems conceivable, especially considering that it happened in English. I also find it interesting thatLatvianērzelis(“stallion”) may be related to*h₁órǵʰis, and a semantic relationship between a word meaning "to mount" and a word for "horse" also seems quite possible. It still seems strange for the wordὀρχέομαι(orkhéomai) to have also developed the meaning "to dance;" it makes me suspect that maybe the sense of "to dance" has a different origin than the sexual meanings, assuming those senses actually exist. I'm also not sure whether the root forὀρχέομαι(orkhéomai) has to be reconstructed as*h₁erǵʰ-. Based on my knowledge of the development of the Ancient Greek language, a hypothetical pre-form*h₃orǵʰ-éye-tor would also produceὀρχέομαι(orkhéomai). In fact, the Hittite cognate mentioned by LIV—[script needed](arkatta)—is actually derived from*h₃ṛǵʰ-o ~ h₃órǵʰ-ey by Kloekhorst. Perhaps it is possible to unify all of the forms under a single root*h₃erǵʰ-? This would leaveAlbanianerdha,Sanskritṛghāyáti, andOld Irishregaid—the future oftéit—unexplained.Graearms (talk)22:52, 16 October 2025 (UTC)Reply
^Watkins, Calvert.1975. La famille indo-européenne de grec ὄρχις:linguistique, poétique et mythologie. Bulletin de la Société de Linguistique de Paris 70/1:11–26.
Latest comment:1 month ago3 comments3 people in discussion
I am assuming this word for "east" is related toזריחה (zrikha) meaning "sunrise", as is common in a lot of languages. Can someone please verify this? ThanksSkuld (talk)15:23, 17 October 2025 (UTC)Reply
Latest comment:30 days ago4 comments3 people in discussion
The word adab on the Persian page suggests that the Arabic word adab is a borrowing from the Persian dab, ultimately from Sumerian. However, there is no evidence that the Arabic word adab comes from the Persian word dab, which has no proven existence LoL.John daar (talk)23:18, 17 October 2025 (UTC)Reply
That's why we don't duplicate difficult etymologies like that (former one) of Arabic ʔadab on known descendants like Persian adab. I never edited the Persian page, but now I have removed that duplication. On the talk page of the Arabic page, one editor made a good reference for the native origin of the Arabic word—although the frequent claims of alleged “Persian” or otherwise Iranian origins are as well true.Fay Freak (talk)10:43, 18 October 2025 (UTC)Reply
Yeah, if only because there are some steps left out for such a far-reaching idea of continuity.ادب(adab,“letters”) +ـستان(-estân) seems easy enough, and @ZxxZxxZ apparentlycopied the end etymon from an etymology at the Arabic entry we later softened, only to avoid an insinuation that this is all the same Arabic word.Fay Freak (talk)14:58, 23 October 2025 (UTC)Reply
With some effort I extracted from GBS snippet views[6][7] the following:
Nun entstand span.carajo „männliches Glied“ gewiſs mit kat.carall dass. aus lat. *characulum „kleiner Pfahl“, dem Diminutiv des durchcharacātus „mit Pfählen versehen“ Columellas für das Volkslatein Hispaniens gesicherten *charax,-acis „Pfahl“ griech. Ursprungs (Spitzer, BAR. II, 1, 158 unten, zu S. 35: Brüch, BAR. II, 3, 36 oben; Meyer-Lübke 1672b);
I did not find a bibliography section with a clearer reference and do not know the meaning ofBAR. It might be the abbreviation of a journal title, presumably published in German. ‑‑Lambiam13:35, 19 October 2025 (UTC)Reply
Latest comment:1 month ago7 comments3 people in discussion
The breaking seen in Old Englishgriellan (late WSgryllan) and the umlauted past tense "grylde" of the latter suggest that this goes back to *grallijan, with original geminate *-ll-, not gralljan with secondarily geminated -l(l)j-. Compare*wallijan,*fallijan > Early West Saxon OEwiellan,fiellan vs.*salljan,*talljan > Early West Saxon OEsellan,tellan. Hogg and Fulk ("A Grammar of Old English", Volume 2) says on page 275 that the etymology of griellan is uncertain. Do any of the non-English descendants here or at*grulljan point to *-llj-* as opposed to *-llij-?Urszag (talk)16:34, 20 October 2025 (UTC)Reply
I missed one piece of evidence:rattig (“in heat”). Some German Low German dialects have rr for tt, so the comparison ofrut androar makes sense. To extend the argument to ll is wanting, I admit, but then it is onomatopoetic, call it a blend if you will, or vulgar because I project Latin loans onto what you might call PWG,imitative in any event. Whereasrally also indicatesligo, theintelligent observer will note how meaningful this is for the comparison torallen, I think.RacketRalf (talk)22:12, 20 October 2025 (UTC)Reply
The vocative ending in -u would then appear to be analogical, based on the i-stems (such as *gasti).
Given the evidence of the examples above, the u-stem vocative in -au survived in PGmc long enough to become common in Gothic.This also raises the question of when the analogical vocative in -u originated, whether in PGmc or simply in Gothic.In any case, may I suggest an amendment to the PGmc u-stem declension template to reflect the inherited vocative in -au.
I agree that{{gem-decl-noun}} should be edited to indicate the vocative ending in-au rather than-u. As for when the analogical variant in-u showed up, it's probably impossible to say since both North Germanic and West Germanic have lost the vocative as a distinct category. No one can say whether the inherited vocative and the analogical vocative existed side by side in Proto-Germanic before being lost in Northwest Germanic or whether the analogical vocative didn't turn up until Gothic had split off. @Victar,Leasnam. —Mahāgaja ·talk06:29, 21 October 2025 (UTC)Reply
Latest comment:1 month ago2 comments2 people in discussion
Triggered by @Saam-andar’s ofRFV ofthe entryسروب(sarôb,“word”) which he deems but a transcription of a Middle Persian word.
Regarding the other purist termواژه(vâže,“word, term”), which @Samiollah1357 graciously quoted from 1599, this must outright be borrowed fromNorthwest Iranian after 1500 (Safavid dynasty rule) and no less introduced thanتکواژ(takvâž,“morpheme”) andواج(vâj,“phoneme”), though the introduction of such new technical terms presuppose acquaintance with the regiolectal stem. The unlearned reader is suggested by the current etymology formulation that this is inherited from Middle Persian, but bothژ(ž) and leadingواـ(vâ-) instead ofباـ(bâ-) speak against this.
A pro AI search I let be performed across both English-language academic treatments and available Persian corpora—she also found my reasoning valid, but often needs a nudge to elaborate upon her extrapolation process—, could found no newer quote, thoughwe know Iranians are most liberal with putting online even any copyrighted edition. Her abridged result formatted for you people is:
Perplexity Research:There is no attested use ofواژه in pre‑Safavid New Persian texts, such as the Shāhnāmeh, Vis o Rāmin, Sufi prose, or Khayyām’s works. Neither do the lexica of Asadī Ṭūsī or later dictionaries include it. The first known citation indeed appears with Bahāʾ al‑Dīn al‑ʿĀmilī around 1599 CE. (Because it was cited in the Wiktionary entry, proof we are the best in the West.)
Fay Freak:I want explicit locations indicated by Persian-language sources for pre-Safavid quotes ofواژه.
Perplexity Research:You are correct to demand this level of rigor. After searching Persian‑language scholarly and digital humanities sources (.ir domains, Ganjoor, university repositories), I have found no explicit citation ofواژه in any pre‑Safavid Persian text. The Iranian digital corpus—including: Ganjoor (comprehensive poetry archive with full‑text search of classical poets: […] University journals and linguistic studies on archaic lexicon […] Studies on Safavid‑era prose and lexicography […] No pre‑Safavid occurrences: Detailed lexicographical studies of rare or archaic words in classical Persian poetry and prose (e.g.,Zarīn Qabānāma, Bīhaqī'sTaʾrīkh, early dictionaries like Asadī Ṭūsī'sLughat‑i Furs) do not mentionواژه […] Silence of classical lexicons: Neither Dehkhodā's citations nor specialized Persian linguistic databases referenceواژه […] […]
(The link to the full version is blocked by the spam filter.) So I hereby strongly suggest our Persian and Iranian editors to do something about the etymologies.Fay Freak (talk)17:26, 21 October 2025 (UTC)Reply
Despite the claims of the artificial intelligence, I looked it up, and the poetry archive Ganjoor does in fact quote it as occurring in theShahnameh, in the line:
".جهاندار دهقان یزدانپرست \ چو بر واژه برسم بگیرد به دست"
Regarding the historical usage of the wordوَاژَه(wāža),Dehkhoda's dictionary mentions it as occurring in the terminology of the Zoroastrian priesthood, as well as a derived form occurring in the dialect of the village ofAhar. Ganjoor alsomentions it as having occurred in one of thefahlaviyat ofBaba Tahir, but this is only found on Ganjoor and other sources have it withوَاجَم(wājam) instead. Iranicamentions the verbvāž- ("to say, tell") as a feature of the fahlaviyat, soوَاجَم(wājam) is likely a 1st-person conjugation of this. As for the mentions ofوَاژَه(wāža) in the Kashkool of Sheikh Bahaei, these quoted portions were originally written in Arabic and more recently translated to Persian. The Kashkool originally contained a mixture of Persian and Arabic. You can find the original versions online, such asthis for the first quotation:
أقول: ما ذكره الزمخشري مشكل من جهة الصناعة، لأنه اذا كان المعنى ما ذكره يكون النهار معمول ابتغاؤكم و قد تقدم عليه و هو مصدر، و ذلك لا يجوز. ثم يلزم العطف على معمولي عاملين، فالتركيب لا يسوغ.
Latest comment:30 days ago3 comments2 people in discussion
A surname, apparently of Hungarian origin according tothis page. The-ски(-ski) component is obviously Slavic, but I wonder ifСалон-(Salon-) could be an alteration ofHungarianszálló. But I also found the Hungarian surnameSzalontai (possibly with some accents?), which survives in Pannonian asСалонтаї(Salontaji) as well, so I'm not sure.Insaneguy1083 (talk)09:30, 23 October 2025 (UTC)Reply
The Hungarian surnameSzalontai literally parses out to "someone from Szalonta", which city name is more commonly encountered in English contexts asw:Salonta. Looking at the#History section there, we see that the city name is probably derived fromHungarianszalonna(“bacon”), from the historical background of this area being big in raising swine for food production. In turn,szalonna is a borrowing from a Slavic language -- which might give a usable clue to the origins of Rusyn surnameСалонски.
@Lambiam: This claim wasadded in 2011 by Dbachmann (“Adopted in the 17th century, from RussianСибирь.”); we did not have a Latin entry at the time. I have edited the etymology in March to add the suffix, but the question whether it came via Latin remains.J3133 (talk)17:30, 24 October 2025 (UTC)Reply
The earliest use I found (using GBS) was English, and the earliest Latin use just 7 years later,from 1599. In view of the more limited coverage of Latin texts on Google Books, this is not conclusive. If the nameΣιβηρία is attested in pre-Modern Greek at all, it would be Neo-Ancient Greek. It is possible that the coiner modeled the name afterIberia. ‑‑Lambiam01:16, 25 October 2025 (UTC)Reply
There is a strong case for this, but it is not inconceivable that at some point a non-Latin use will be found that is earlier than 1586. Also, it is not obvious how to explain the transformation of the second Russianи into the Latin lettere. A route from English to Latin would go some way to explain this, with the phoneme/i/ represented in English orthography by the Latin lettere as a stepping stone.
Latest comment:16 hours ago3 comments3 people in discussion
The first etymological section at曲#Chinese traces this word back to Proto-Sino-Tibetan for the "bent" sense.
This word apparently also has a sense of "song". This sense also has a different tone in modern Mandarin (qǔ, third / low tone, compared toqū, first / high tone for the "bent" sense). The entry has nothing to say about how or why this separate tone evolved, nor how or why we got from "bent" to "song".
Latest comment:29 days ago3 comments2 people in discussion
Hey! I was looking at the page forgeronto- and under the "Derived Terms" section, there's a list labeled "English terms prefixed with geronto-", and within that list there's a link toProgeria, which only has "geronto-" as a mid-fix. It's definitely still derived from "geronto-", but it doesn't seem like it belongs.
What's the ideal pattern here? Is Progeria encoded improperly, getting picked up by this automated list when it shouldn't be there? Is the list coded wrong, picking up one element that shouldn't be in there? Are there terms derived from "Geronto-" that aren't showing up in the list, but should? Is everything actually working as intended and I simply haven't internalized the logic here?
As an aside, I'm a software dev looking for little projects to contribute to Wiktionary; is there a already a version of the prefixed-by script for the other ways a word could be derived-by (suffixed, midfixed, just references it)? I just thought it might be nice to have more than the one list in the "derives" section.Vincent.Xavier.Zell (talk)15:30, 24 October 2025 (UTC)Reply
I looked up theOdd Couple episode mentioned, but it appears that it's likely just made-up nonsense;
So what do you want from me?I'm not a mind reader.(laughs)Excuse the jest, but...if I took my work too seriously...Nee-noo-nah-nah-noo-noo...I would go bananas.
Latest comment:28 days ago2 comments1 person in discussion
On the [Liver Attack|page for the chess opening called the Fried Liver Attack], it claims the phrase is a literal translation of the Italian “giuoco fegatello”, though the page displays a note asking for verification - and I think I’ve found something, but as you’ll see, I’m not sure. I’m completely new to posting here, so I’m hoping for a. advice on what to do and b. honestly just to show off my findings.
With an in-quotes Google search, 4 results pop up. Two wiktionary pages, and two pages from the New York Clipper, both from the late 19th century.
Thefirst is from the Illinois Digital Newspaper, is dated 1886 and features a chess game (third column, at the bottom, pg. 12) labeled “Giuoco Fegatello” which, translated into modern notation, begins:
1. e4 e52. Nf3 Nc63. Bc4 Nf64. d3
Thesecond is from the Cleveland Public Library Digital Gallery, is unhelpfully dated from 1884 to 1890, and features a chess game (third column, in the middle, pg. 87) also labeled “Giuoco Fegatello” which, once more translated into modern notation, begins:
1. e4 e52. Nf3 Nc63. Bc4 Nf64. Ng5 d45. exd5 Nb4
Neither of these are strictly the Fried Liver Attack, but do begin, until the first one’s fourth move by white and the second one’s fifth move by black, as a Fried Liver Attack would. Should I add this or no, given the dubious support of the etymology?TabaEnjoyer (talk)21:12, 25 October 2025 (UTC)Reply
I agree that-tor +-ium does not make much sense here (unlike inscriptorium). There is a Latin verbterreō from which an agent noun *territor could be formed, but no such noun, whose meaning would be something likebogeyman, is attested, andterritorium did not mean something like “the office of bogeyman”. We seeterri- forterra also in, e.g.,terricola andterrigena. I've changed the etymology section accordingly. ‑‑Lambiam06:33, 28 October 2025 (UTC)Reply
Latest comment:25 days ago2 comments2 people in discussion
The current etymology seems rather contradictory:
Inherited from Latin sīc (“thus, so, like, in this way”). Compare Aromanian shi, Megleno-Romanian și. Possibly related to Italian sia (originally a subjunctive form of essere “to be”) as in the construction sia … sia … (“both … and …”), which is equivalent to Romanian și … și ….
It's either descended from Latinsic, or from Latinsit, or a conflation of the two. I'd consider the derivation fromsit more likely, as the development from the conjunctive of the copula to a conjunction shows in Frenchsoit ... soit, Italiansia ... sia and Spanisho sea .2001:16B8:B1D2:6000:2C86:68B5:2F9B:5D6221:46, 26 October 2025 (UTC)Reply
Although you CAN say "repiä tekstiiliä", it should be translated as "rip textile" (tear textile), so the Finnish word "reps" has nothing to do with textiles in the context of internet slang.
I don't speak Finnish, but I think you're confused - the pagereps contains independent entries for multiple languages. The English word "reps", the French word "reps", and also the Finnish word "reps", with separate etymology sections. The Finnish entry doesn't mention English or French at all, it says "Shortened from repesin nauruun".
Links to other language wiktionaries are not translations, nor is there any requirement that they have the exact same content or refer to the exact same word - they are simply "this other wiktionary also has a page titled "reps"". Most of them don't even contain an entry for the Finnish term (because most other wiktionaries have far less content in languages that are not their main one).PhoenicianLetters (talk)19:25, 28 October 2025 (UTC)Reply
You are looking at the etymology of the English word "reps" - the big "English" header above does not mean "this is an English explanation", it means "this is an English word".
I believe the linking to different language wiktionaries is done automatically by the wiki software. And technically they often do have related content, but the important part is that they are "translations" of the dictionary entries, NOT of the word itself. Because of that, these links are far less useful than on Wikipedia.PhoenicianLetters (talk)20:08, 28 October 2025 (UTC)Reply
Even on Wikipedia, I guess a 1:1 correspondence often could be more of a preconceived ideal than practically realistic, since an English concept might correspond to two or three different concepts in another language, and vice versa.Wakuran (talk)16:06, 29 October 2025 (UTC)Reply
I guess most of the ordinary Wikipedia users don't know this:
Links to other language wiktionaries are not translations, nor is there any requirement that they have the exact same content or refer to the exact same word - they are simply "this other wiktionary also has a page titled "reps"".
The content is dependent on what the users add. Except for the setup that the first entry should belong to the main language of the Wiktionary edition, the editions cannot be directly based on each other.Wakuran (talk)19:59, 28 October 2025 (UTC)Reply
Wiktionary organizes its entries by written form (that's why there are many different "reps" words from different languages on one page). The English Wiktionary has English, Catalan, Finnish, French and Swedish. The French wiktionary has French, English and Swedish. The Finnish wiktionary has Finnish, English, Catalan, French and Swedish.
Ideally, when two different wiktionaries have an entry about Finnish "reps", they would contain the same information, but wiktionaries are largely independent from each other, with no control over what the other puts in their entry. But most importantly, there is no requirement that they have an entry on the Finnish word "reps" at all. Nobody on the French wiktionary added the Finnish word, so in this case it has no relevant information at all.
If you want to look up translations, these will always be based off the entry in the wiktionary's main language - English, in this case. So you would go toLOL#English, click on "Translations" there, and that will point you towards Finnish "reps" (among other things).
Translations are rarely one-to-one, there are synonyms and partially overlapping senses, it is simply not possible to have it work like Wikipedia where you "translate" the article by going to a different language wikipedia.PhoenicianLetters (talk)20:05, 28 October 2025 (UTC)Reply
Latest comment:12 days ago3 comments2 people in discussion
I note that the entry for PGmc *isōną lacks an etymology, and users are invited to contribute in this part of the online forum. On the evidence of PGmc *minnisōną 'to make less' I suggest:
I don't believe that it is possible the bypass the lack of stress on Pre-Germanic *-is- if we assume that Proto-Germanic *-isɔ̄janam is an old formation. However, I can propose an alternate theory.
There seems to have been a related -ī-/-jɔ̄- stem suffix *-(i/u)si in West Germanic, yielding words such as Old Englishmilts <milds < *mildisi,blīds <blīþs < *blīþisi, as well asæx <æcis < *akusi (with -u- from ablaut), approximately equivalent to -ness (except for the latter which has no accepted etymology). This indicates a Pre-Germanic *-ésī/-sjā́ > Proto-Germanic *-isī/-zjɔ̄ (see Ringe 2006) where the lack of Werner alteration originates from the generalisation of -s- to all declined forms in the late variety that gave rise to West Germanic. We may then derive *-isɔ̄janam by adding the denominative verbal suffix,e.g. *minnisī "lessness" + -*-ɔ̄janam > *minnisɔ̄janam "to make less".SériskaMarka-Lénínattijó (talk)07:53, 31 October 2025 (UTC)Reply
Latest comment:23 days ago2 comments2 people in discussion
I notice the etymology of this verbal suffix is missing, so I will state what seems clear. I suggest these verbs arose through rebracketing when *-ōną was added to nouns and adjectives ending in *-ulaz, *-ilaz, *-ilô and so forth.
The semantics of repetitive action were drawn from adjectives in*-ulaz 'prone to doing.'
Examples:
*bedulaz 'prone to asking' — *bedulōną 'to beg'
*wandilaz 'given to wandering' —*wandilōną 'to wander; to waver'
Examples that lack the semantics of repetitive action include:
*trundilaz 'hoop' —*trundilōną 'to trundle, roll'
*hwirbilaz 'circle' —*hwirbilōną 'to whirl (in circles)'
After rebracketing the new independent morpheme*-lōną was then applied in pairs like:*trampaną 'to step' —*trampulōną 'to trample'
I concur. I have always understood this to be the case, and have yet to come across a proposed etymology that makes more sense than this one.Leasnam (talk)17:45, 30 October 2025 (UTC)Reply
فَعِيل(faʕīl)-type stative-verb derived from theactive participle...
Latest comment:22 days ago3 comments3 people in discussion
What does this phrasing mean? It's ina good amount of Arabic entries. I don't understand "stative verb" (or "stative-verb") and I definitely don't understand "active participle" -- if anything, to my understanding,فَعِيل(faʕīl) andفَعُول(faʕūl) behave like archaic passive participles and ones without verbal force at that. Can others besides me weigh in on whether we can/should update this phrasing?Still, when you think about it (talk ·stalk)18:49, 29 October 2025 (UTC)Reply
While I am not sure this form must be passive rather than active participle – like how canعَظِيم(ʕaẓīm) related toعَظِمَ(ʕaẓima,“to become great”) orوَسِيع(wasīʕ) toوَسُعَ(wasuʕa,“to become wide”) be understood passive? –, it may be faulty reasoning that it would be active only because of being from the active participle formفَاعِل(fāʕil). Even better, he actually wrote it when it was passive:حَبِيس(ḥabīs). Beyond that, I think it can be a general adjective-forming “having a property expressed by the root” pattern, as seen in repatternedطَيِّب(ṭayyib). Then we may doubt that the adjectives are from the verbs and not vice versa, likeجَمُلَ(jamula) is probably fromجَمِيل(jamīl), and everything else is just medieval logics, by which the base stem, 3rd personm sg, is the root of all, due to having the least additions to the root, filled up withفَتْحَة(fatḥa) =alpha.Fay Freak (talk)11:53, 30 October 2025 (UTC)Reply
Latest comment:22 days ago1 comment1 person in discussion
Meaning "all", as cited on many modern sources (Lobel, 2016; many facebook posts). Niemann also cited in his glossary that the word is translated as "all; once; ever; again". Possibly from Proto-Philippine*minsan ("at one time"), via*kon minsan ("OBL one.time")? SeeTagalogminsan(“sometimes”),Maranaominsan(“although; kinship of one degree”).
Note: Ponosakan is quite niche compared to many other languages here, so if I found someone with the same interest who's quite knowledgeable in it, then I'll be one lucky son of a gun:)Udaradingin (talk)10:30, 31 October 2025 (UTC)Reply
^Mayrhofer, Manfred (1992), “divyá-”, inEtymologisches Wörterbuch des Altindoarischen [Etymological Dictionary of Old Indo-Aryan][1] (in German), volume 1, Heidelberg: Carl Winter Universitätsverlag, page727
Now that the EWAia is no longer available online, it's impossible for me to check the reference to figure out what was really meant, and whether it makes any sense in the context of the rest of the etymology. Can someone fix this? Thanks!Chuck Entz (talk)01:13, 1 November 2025 (UTC)Reply
Latest comment:20 days ago2 comments2 people in discussion
Almost certainly ultimately derived from Arabicمَحْجور(maḥjūr), as suggested by the presence of the Arabic definite article in the plural form. It was probably borrowed through Moroccan Arabic, though I have not been able to find this word with the sense “orphan”. Can anyone confirm its use with that meaning?Lankdadank (talk)21:11, 1 November 2025 (UTC)Reply
Latest comment:18 days ago4 comments3 people in discussion
In the etymology, the gloss of heilige as "holy man" seems to me to have the wrong connotation in English; it brings to mind monks, ascetics, priests, etc. Would it still have the correct meaning if the gloss was changed to "saint" (a word that is already frequently used metaphorically in English for exactly this type of thing)?TooManyFingers (talk)16:33, 2 November 2025 (UTC)Reply
Dutch happily forms noun–noun compounds, and soschijn +heilige is theoretically possible, and perhaps even found in the wild, similar toschijnhuwelijk(“sham marriage”),schijnoffer(“fake sacrifice”) andschijnproces(“show process”). However, this (theoretically possible) noun–noun compound can be distinguished from the nominalization of the compound adjectiveschijn +heilig (“hypocritical”) by an audible difference in stress. Dutch compound nouns have word stress on the first component:schijnhuwelijk,schijnoffer,schijnproces,beschermheilige(“patron saint”),patroonheilige(“patron saint”),pilaarheilige(“stylite”). Dutch compound adjectives are stressed on the second component:schijnheilig, and this stress pattern is conserved in the nominalizations. ‑‑Lambiam18:33, 3 November 2025 (UTC)Reply
The compoundschijn +heilige, meaning as much as “fake saint”, can be found in the wildhere (in a book, in the phrasedeschijnheilige Sint Magher van Gecxhuysen) and alsohere (in a blog). I suppose these uses are too rare to meet our CFI. I’ve fixed the etymology. ‑‑Lambiam12:07, 4 November 2025 (UTC)Reply
Latest comment:18 days ago3 comments2 people in discussion
A bit late since it’s already FWOTD, but the etymology seems to be wrong. Japanese wikipedia has an extensive section on the history of the word in the tokusatsu genre, which makes it clear that it was originally a wearable (着 ki-) puppet (縫いぐるみ nuigurumi). I don’t think any Japanese speaker understands ぐるみ as being literally “to wrap up” in this word.209.35.66.5019:03, 3 November 2025 (UTC)Reply
Agreed that JA sources clarify this as basically "a kind of縫い包み(nuigurumi) that you wear (着る(kiru))":
There isn't any attestation date given for this "wearable" sense forkigurumi, butFWIW the same "wearable" sense fornuigurumi is dated to at least 1703 in kabuki contexts, so this concept has been around a while.
That said, these sources also list a sense forkigurumi of "what someone is wearing and their whole appearance", based on verb着る(kiru,“to wear”) and suffixぐるみ(-gurumi,“including [suffixed word], [suffixed word] and all, [suffixed word] and everything”), which latter entry we don't yet have. Our entry at着ぐるみ could do with some updating. ‑‑ Eiríkr Útlendi │Tala við mig01:28, 4 November 2025 (UTC)Reply
It has never been a term used colloquially, which makes it hard to assess its current status. I suspect it is still used in such texts as contracts and commercial correspondence, provided they are drawn up in Dutch, which seems to be increasingly rare.Here is a use from 1942, andhere one from 2012 – although it refers to a historical situation, the text itself is recent. ‑‑Lambiam23:12, 6 November 2025 (UTC)Reply
I don’t think there is an easy test – the same modifier can be adverbial in some hyphenated compounds (half-baked) and adjectival in others (half-brained). Cases likeabove-mentioned,far-gone andwell-done are easy, because they clearly correspond to the qualified participle constructionsmentioned above,gone far anddone well, in which the second term is an adverb. Perhapsready-made can be compared tofull-made, in which I tend to analyzefull as an (archaic) adverb, just like infull-grown. An argument against classifyingready inready-cooked as an adverb is that it is hard to find examples in whichready as an isolated word clearly functions as an adverb modifying a verb or an adjective – we don’t say *“it isready available” or *“it workedready for more than an hour”. ‑‑Lambiam13:52, 8 November 2025 (UTC)Reply
On historical grounds: I pointed at the fact that 'ready' as "already" dies out or had already died out when prepositive 'ready' seems to have started appearing (around 1600), see quotes. In these early attestation I also found it hard to translate them as "already-". I don't know if these two "facts" prove anything anyway.
Your tests (which work because 'ready' alone is obviously no longer an adverb, the whole point I think is that the obsolete sense "already" for some would have been crystalized in these prepositive formation) also go that way; as I already said, I think 'ready' is mostly like 'hard' inhard-boiled, proleptic, "boiled to be hard" (or "boiled (as to be) hard") not "boiled hardly".
Should we keep 'ready' as an adverb, make it an adjective or even remove the thing and treat each case separately? I see thathalf only mentions this use in the usage notes,above the adverb doesn't bother but the adjective does.
Also, I asked Benwing2 about this but got no response: we have a label and gloss at the glossary for 'postpositive', but neither of them for 'prepositive', would you do that please? Making a category would be nice as well, postpositives likegalore have one.Saumache (talk)09:14, 9 November 2025 (UTC)Reply
Prepositiveready seems to carry the sense ofalready, which may explain how it could be considered an adverb.
On the other hand,made seems to be super-flexible about what can be prepended:tailor-made,hand-made,custom-made. Agreed: It's hard to say whetherready is an adverb or adjective.Latvvot (talk)08:14, 9 November 2025 (UTC)Reply
We call “rebar” ablend ofreinforcing +bar. We do not have an entry for a noun sense ofdebar, which appears to be synonymous withdeformed rebar. The deformations are ribs, as can be seen in the imageshere. I have a hard time finding sources in durably archived sources in which the sense of the term is clear. ‑‑Lambiam18:32, 8 November 2025 (UTC)Reply
Latest comment:7 days ago3 comments2 people in discussion
RFV for this comment: "The post-Classical sense of 'having the form of a line' is likely a back-formation from the adverbial form līneāriter (literally 'using a line or lines')"
Can we get a citation for:
(1) That the sense oflinearis of 'having the form of a line' is post-Classical? Wasn't that already one of the senses in Classical Latin?
(2) That this sense is a backformation fromlīneāriter? I find this hard to believe.
@Latvvot I will have a look in more depth, but none of the major dictionaries have that as a Classical sense. The back-formation argument is as follows:
The Classical meaning is closer to "line-based", sopictura linearis(“line-based drawing”),probatio linearis(“line-based [geometric] proof”), and so on.
The adverblīneāriter was regularly derived from this with the meaning "using a line" or "in the manner of a line".
As it was commonly used with verbs of motion (e.g. see two cites in the DMLBShere), the original sense of "[moving] using a line" was reanalysed as "[moving] like a line", as they're essentially indistinguishable in that context.
The adjectival sense of "like a line; having the form of a line" was then back-formed from that reanalysis.
OK, thank you for reply. Maybe we could reword this distinction a bit, to make clear what sense 4 allows, but sense 1 does not allow?
Suggestion: Maybe mention the sorts of nouns that might be modified? I'm guessing that there must be one or more genres of nouns (iter?via?) that cannot be modified bylinearis in sense 1 but that can be modified in sense 4.
To state the matter in another way: It's hard to imagine (for me, anyway) that classical Latin had sense 1 but not sense 4. Maybe some specific examples might make this clear.
For example: Is it true thatlinearis ratio was allowed in Classical Latin, but thatlineare iter was not? I'm not sure I understand. Thank you very much.Latvvot (talk)23:59, 15 November 2025 (UTC)Reply
@J3133 Even ifrematch can have that sense, I would argue that speakers might casually just stickre- ontomatchmaker, rather than explicitly thinking "this person is the maker of a rematch!". I think you are overly optimistic in your etymology. Oh well.2A00:23C5:FE1C:3701:7082:78C2:9876:E5CA 22:09, 9 November 2025 (UTC)
Latest comment:12 days ago2 comments2 people in discussion
I suspect the first part of this word isقارا(qara), presumably because the wind is so dust-laden it blackens in the sky, and the second part is the Uyghur cognate ofбуран(buran). Can anyone confirm or correct?- -sche(discuss)22:08, 10 November 2025 (UTC)Reply
I'm sure you're right that it's from those Turkic roots, but proving that it's Uyghur as opposed to another Turkic language might be difficult. —Mahāgaja ·talk22:52, 10 November 2025 (UTC)Reply
Latest comment:7 days ago8 comments5 people in discussion
Theknightwho insists that the Greek borrowing from Hebrew involves phono-semantic matching rather than a simple folk-etymological re-interpretation of the beginning of the Hebrew name. But calling this adaptation phono-semantic matching would only work if the adapted name looked fully Greek, both the first and the second part, but the second part (ending in mu) still looks very obviously foreign. I already gave my rationale in an edit summary but the user keeps ignoring it and insists on their understanding of phono-semantic matching as correct. --Florian Blaschke (talk)14:51, 11 November 2025 (UTC)Reply
The entry currently states "partial phono-semantic matching". How is that incompatible with your argument that the word is only partially affected?
The only "rationale" you gave was an edit summary where you saidw:Phono-semantic matching has a meaning in linguistics; it's not this. Not good enough, quite frankly.
This does not look like phonosemantic matching to me at all. PSM in my understanding is an instrument of language policy that is quite consciously applied. It is especially prevalent in languages like Icelandic and Chinese that hold a language ideology of linguistic purism. The above word shows the effect of folk-etymology. I think the fact the h-less variants exist is an extra testament to that. —Caoimhin ceallach (talk)02:24, 14 November 2025 (UTC)Reply
Latest comment:9 days ago6 comments5 people in discussion
The entry currently claims that it might be derived from the root*skek-, though I can't find any actual source for this etymology and all of the other resources discussing this root do not mention the term. In fact, the root is generally considered to be confined to Celtic, Germanic, and Balto-Slavic.Graearms (talk)16:27, 11 November 2025 (UTC)Reply
Having dealt with Irman, I will say that the wording is unfortunate- but more for its affect on innocent third parties than what it says about this person. They were willing to make up all kinds of lame nonsense in order to pretend that everything important came from Persian, and resorted to all kinds of block evasion, sockpuppetry and other dishonesty to sneak it into every etymology they could think of. Fortunately, they were too incompetent to pull it off for long.Chuck Entz (talk)03:32, 13 November 2025 (UTC)Reply
I've gone and removed the current etymology from the page. That still does leave the problem of what the actual etymology is.
It turned out there was a pretty obvious source I've missed. I foundthis article which, in the abstract, seems to claim that it derives from Old Persian*fraxsa, itself from some rootraxsa-(“to slip”). This same abstract provides etymologies for serval other words, includingسریدن(soridan), which also seems to currently have an etymology added by Irman. I can't actually read the source as it is in Persian. Otherwise, there isthis paper, which purports to analyze certain Classical Persian words in the Quran. It also apparently describes saxsidan on page 5, though I can't read it because it is also in Persian.Graearms (talk)04:10, 13 November 2025 (UTC)Reply
Latest comment:11 days ago1 comment1 person in discussion
I do believe it ultimately either comes from the proposed KʷeHn- meaning canid/wolf. Or alternatively (what I think to be more likely) ʔušən which also carries this root definition of canid/wolf, but also lines up better to the tʃ sound of [cce] in uccen.However I cannot at this moment find a decent source for either term, I have tried.2601:603:600:FE10:B91C:2949:4965:96F518:47, 11 November 2025 (UTC)Reply
There's nothing in the Proto-Indo-European*ǵʰer- entry to explain why it's *ǵʰ and not *gʰ. Even the linked-to entry for Proto-Indo-European*ǵʰórtos and the linked-toCategory:Terms derived from the Proto-Indo-European root *ǵʰer- (enclose) have onlycentum descendants. You have to go to the Pokorny reference to see anything to support it. Then there's the link to Proto-Indo-European*gʰerdʰ-, which looks suspiciously like a form of the same root. It's been almost four decades since my last class in Indo-European studies, so I don't feel comfortable working on the entry myself.Chuck Entz (talk)23:53, 14 November 2025 (UTC)Reply
Have there been other etymologies proposed? How about a phrasing like "A connection to PIE *ǵʰer- (enclose) is tempting, but hard to fit on phonetic grounds." ?Wakuran (talk)14:01, 19 November 2025 (UTC)Reply
It's not just "hard to fit", it's impossible. Not only does *ǵʰ not match thegh, but the vowel ofghera has to go back to PII *ay < PIE *ey or *oy, and*ǵʰer- doesn't have ay in it. The two forms have nothing in common but ther. —Mahāgaja ·talk14:45, 19 November 2025 (UTC)Reply
Latest comment:6 days ago4 comments2 people in discussion
I think Latintonus needs to be split into two etymologies. Let me explain:
Checking three major dictionaries (Lewis & Short, Gaffiot and Georges), they all say more-or-less the same thing: that it's a borrowing of Ancient Greekτόνος(tónos) (from Proto-Indo-European*ten-(“stretch”)) with the following senses:
tension in a string/cord/rope etc [the literal meaning].
the tone produced by an instrument [i.e. the sound with a specific pitch; a transferred sense from the use of tension in stringed instruments].
the tone of colour [presumably by analogy with the previous sense, but in the domain of light rather than sound].
Despite what seem like fairly drastic differences on the surface, it's fairly clear that these senses do indeed share one etymology. However, L&S and Georges also mention another sense, which they both treat as a figurative development of the second sense:
Tonitrua nos pluraliter dicimus ; antiqui autemtonitruum dixerunt auttonum.
We saytonitrua ["thunder"] in the plural, but the ancients said [singular]tonitruum, ortonum.
tonitruum (andtonitrus) both derive fromtonō(“to thunder”), from Proto-Indo-European*(s)tenh₂-(“thunder”) (i.e. a different PIE root). Does it not seem more likely that this was the origin of ancienttonus instead (perhaps a fourth-declension noun?), as opposed to some doubly-figurative use of a Greek borrowing?Pinging @Urszag, who will likely be interested.Theknightwho (talk)00:00, 15 November 2025 (UTC)Reply
Gaffiot also gives the sensetonnerre,[10] citingCaecin.d.Sen.Nat 2, 56, 1, where Seneca writes further:
Hoc apud Caecinam invenio, facundum virum et qui habuisset aliquando in eloquentia nomen, nisi illum Ciceronis umbra pressisset.
I find this in Caecina, an eloquent man, who would even have been renowned for his eloquence, had not Cicero overshadowed him.
“Caecina” refers toAulus Caecina, a contemporary of Cicero, just one century before Seneca, so hardly an “ancient”. So does Seneca mean to say that he has, on the authority of Caecina, that the termtonus was used by “the ancients”? It seems that in this sense it is a hapax, and, at that, not a use but a mention of a mention. If Seneca was correct in his interpretation of Caecina, and Caecina was also right (although we do not know his sources), it appears indeed unlikely that this “ancient” Latin term was borrowed from Ancient Greek. However, derivingtonus fromtonō, with a stemtona-, is IMO implausible. Something I found at the entry “tonō, tonāre” inde Vaan is of interest in this connection:
The origin oftonāre is disputed: an original causative *(s)tonh2eie- (Eichner 1974: 58; but the meaning oftonāre is not causative), a derivative from the nountonus (Schrijver 1991: 396, as one of the possibilities; buttonus does not mean ‘thunder’, and is attested too recently), or an iterative *(s)tonh2eie/o-, as Skt.stanáya- would suggest.
(“Schrijver 1991” is the monograph: Schrijver, Peter.The reflexes of the PIE laryngeals in Latin. Amsterdam – Atlanta: Rodopi.) I can’t check what Schrijver wrote, but de Vaan is apparently unaware of this passage by Seneca. ‑‑Lambiam09:27, 15 November 2025 (UTC)Reply
@Lambiam It's not completely implausible, if we accept what is stated on the entry that it used to be in the third conjugation and underwent a sound shift. There's potentially some kind of parallel withdomō anddomus (though I appreciate that the question of how - or whether - they're related is not fully worked out).Theknightwho (talk)20:29, 15 November 2025 (UTC)Reply
I misinterpreted the antecedent of “this” in your question, “Does it not seem more likely thatthis was the origin of ancienttonus ... ?” – I readthis as referring to the earlier “tonō”, but now I understand you meant to refer to “*(s)tenh₂-”. I agree that it is far more plausible that the verb and the noun in this sense derive from the same PIE root, reconstructed as*(s)tenh₂-, than that the noun has a very stretched sense of a Greek loan (which, according to de Vaan, is attested “too recently” – perhaps also too recently for Caecina’santiqui). The shortest path through which they are genetically linked remains unclear; Schrijver 1991 actually suggests a rather short path, raising the question whether PIE (or some later ancestor of Latin) may have had a noun with the sense “thunder” from whichtonus is descended. ‑‑Lambiam09:07, 16 November 2025 (UTC)Reply
Latest comment:4 days ago4 comments3 people in discussion
This word has a very interesting etymology, which is the subject of tons of popular science articles, youtube videos, and reddit posts. Most etymological dictionaries give a derivation from Arabicعَوَارِيَّة(ʕawāriyya,“damaged goods”), but the OED disagrees, calling the latter “a modern Arabic translation and adaptation of the western term in its latest sense”. Can anyone with knowledge of Arabic chime in? In particular wasعَوَارِيَّة(ʕawāriyya) actually used before the 12th century and if so in what sense? —Caoimhin ceallach (talk)01:49, 18 November 2025 (UTC)Reply
It wasn't, I never found it anywhere; even for Modern Standard Arabic it is difficult, so it must be an occasional phono-semantic matching. In sum I don't know the origin of the nautical termHavarie and must oppose the majority references, but was not emboldened to voice it earlier. Is it formally and semantically logical to derive (Old) Italianavaria fromavere?Fay Freak (talk)13:36, 18 November 2025 (UTC)Reply
Interesting. I thought that etymology was fishy but it's literally everywhere. The derivation in Old Italian is also problematic though. But the OED says Spanish and Catalanaveria (from 13th century) withe works better, but I don't know the details. —Caoimhin ceallach (talk)16:28, 18 November 2025 (UTC)Reply
Latest comment:2 days ago5 comments4 people in discussion
what language isla maes abhal? welsh? gaelic? in either case, our one cite is so old that the source language may have been different then.Lollipop(an alt account ofSoap) —talk04:34, 19 November 2025 (UTC)Reply
Well in Irish, "the day of the produce of the apple tree" could be renderedlámeasaabhla, and it would probably be something similar in Scottish Gaelic. But whether that's actually the etymology of this word, I couldn't tell you. —Mahāgaja ·talk07:24, 19 November 2025 (UTC)Reply
The first mention I can find is inCharles Vallancey'sCollectanea de Rebus Hibernicis which contains a lot of his crackpot theories about how the Irish were somehow connected to all of ancient history, but also says that
"The first day of November was dedicated to the angel presiding over fruits, seeds, &c. and was therefore namedla mas ubhal, that is, the day of the apple fruit, and being pronouncedlamasool, the English have corrupted the name tolambswool, a name they give to a composition made on this eve of roasted apples sugar and ale. This festival of the fruit was also of oriental origin as will be explained hereafter."
Somehow this ended up in theChambers Encyclopaedia with the wording paraphrased in our etymology. Oddly enough,ubhal isScottish Gaelic. InIrish Gaelic, the words quoted would belá(“day”)meas(“fruit”)abhaill(“apple tree”) /abhall. It looks like somewhere between Vallancey and Century, the Irish was (sort of) corrected, and someone realized that there were no Irish in ancient England, so they changed "Irish" to "British"Chuck Entz (talk)08:30, 19 November 2025 (UTC)Reply
Interesting. Thanks to everyone for the surprisingly quick and thorough research. I was actually thinking it might have been Welsh, and that Welsh might have been spelled with-bh- and similar digraphs at the time. I didnt think to look upmeas formaes though.—Soap—14:01, 19 November 2025 (UTC)Reply
Maes threw me off too, becauseae is a rather rare digraph in Irish (occurring inGael and its derivatives likeGaeilge but only in a handful of other words) and (to the best of my knowledge) not occurring in Scottish Gaelic at all. I only got there by searchingtheFoclóir Gailge-Béarla for the English word "fruit". As for the "apple" word, it's spelledúll in Irish today, but up until the spelling reforms of the mid-20th century it was spelledubhall, so Vallancey wasn't that far off in his spellingubhal. —Mahāgaja ·talk16:28, 20 November 2025 (UTC)Reply
Latest comment:1 day ago5 comments4 people in discussion
Our entry claims this is calledbaker's chocolate because it's used in baking. My understanding (cf. Wikipedia) is that it's actually called that because it was produced by people with the surname Baker who operated Baker's Chocolate company.- -sche(discuss)21:50, 19 November 2025 (UTC)Reply
The high relative frequency atGoogle NGrams of bothBaker's chocolate andBaker's Chocolate overbaker's chocolate andbakers' chocolate strongly suggests thatBacker's is the original orthography and the etymon. OTOH,baking chocolate has since 1960 become the most common spelling. Moreover it seems to have always been more common thanbaker's chocolate andbakers' chocolate.DCDuring (talk) 16:17, 20 November 2025 (UTC)DCDuring (talk)16:34, 20 November 2025 (UTC)Reply
At this point, it's probably really both. There is a brand called Baker's Chocolate, which produces chocolate for baking, i.e.baking chocolate. Genericization of the trademark would certainly have been facilitated by reinterpretation of Baker's Chocolate asbaker's chocolate (chocolate used by bakers, much asbrewer's yeast is yeast used by brewers, not a brand of yeast produced by a family named Brewer). —Mahāgaja ·talk16:37, 20 November 2025 (UTC)Reply
It seems as if there has been suggested a connection to Latinsingulus and thus, PIE*gʰe (intensifying/ distributing particle). Then, I guess it might also just be some kind of phonetic assimilation.Wakuran (talk)21:45, 21 November 2025 (UTC)Reply
Latest comment:14 hours ago3 comments2 people in discussion
Apparently this term means "to destroy, to cut into pieces", by extension "to scrap" as one would cut up an old car for parts and materials.
The etymology currently just says "Fromesguazar." Presumably then these should be antonyms, consideringSpanishdes- being cognate and largely synonymous withEnglishdis-.
Any details on sense development? Rather thatSpanishesguazar means "to ford / cross a river" and etymonItaliansguazzare means "to splash about", the semantic connection to "to destroy, to cut into pieces" is far from obvious.
Latest comment:16 hours ago2 comments2 people in discussion
Not sure if this is the right place, but I couldn't find a more suitable one. If anyone is aware of a more suitable one, my comment can just be moved there.
I see that the template for IPA of Ancient Greek now allows the restriction to only certain periods, making it possible for names unattested before a certain time to be left without a pronunciation according to the norms of earlier periods. And this option is indeed used, so that, for instance, names first borrowed from Latin or Aramaic in the Hellenistic period are not given a Classical Attic pronunciation. I would like to object to the very idea of such a restriction. The point of giving a pronunciation for an ancient language is not only to give an idea of how it probably sounded in reality, but also of how it should be pronounced within a certain convention. Many, if not most, people use the pronunciation of only one period on a regular basis, regardless of the period of the text they are reading, for reasons of practicality: it is difficult to constantly think of the period of a text and to juggle between multiple different phonological systems; indeed, it isimpossible to consistently reflect each and every change that may have occurred within a given century or region, so even the use of several separate systems, regardless of how many they are, will always be an abstraction. Therefore, people should be informed of how the word ought to sound in the system of pronunciation of Ancient Greek that they do use.~2025-33239-81 (talk)11:37, 22 November 2025 (UTC)Reply
I think it's very rare for people to read a text in an older pronunciation than the text itself. I can easily imagine reading Homer or Aristophanes using Koine pronunciation, but reading the New Testament using Homeric pronunciation would just be weird, and I can't imagine it actually happens very often. Also, from early Koine on, vowel length is no longer significant, and for a lot of loanwords in Koine and Medieval Greek, it's difficult to know what the vowel length would have been in an earlier stage of Greek. It's much easier to just write|period=koi1 or|period=byz2 or whatever rather than trying to reconstruct a hypothetical vowel length for a word that didn't even exist in Greek at the time when vowel length was contrastive. —Mahāgaja ·talk11:48, 22 November 2025 (UTC)Reply
Latest comment:1 hour ago2 comments1 person in discussion
Currently, the pagemerja claims that it is derived fromProto-Germanic*marzijaną, a claim which is supported by Kroonen and Oryol. However, the entry for the Icelandic termmerja contradicts this information, instead deriving the term fromProto-Germanic*marjaną. Moreover, there appears to be third etymology present in the literature—it is described as a derivative of a causative formation to the root*merh₂- in the LIV.[1]Graearms (talk)02:22, 23 November 2025 (UTC)Reply