^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:1 month 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:1 month 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:1 month 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:29 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:1 month 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:15 days ago22 comments6 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
(1) The Greek word was probably borrowed directly from Aramaicיְרוּשְׁלֶם(yərūšəlem) (ending in/em/), not from Hebrewיְרוּשָׁלַיִם(yerushaláyim) (ending in/ajim/). (2) By the time of the borrowing, Greek was in the process of losing/h/ anyway. (3) In loanwords,/j/ was rendered withι and/ʃ/ withσ, so/ierusaˈle̝m/ is pretty much exactly what Aramaicיְרוּשְׁלֶם would yield as a "mere borrowing". What's folk etymology is spelling it with the rough breathing instead of the smooth breathing to make it look likeἱερός(hierós). There's absolutely no PSM happening, and even the folk etymology is purely orthographic. —Mahāgaja ·talk19:34, 27 November 2025 (UTC)Reply
@Mahagaja If you're going to claim it requires actual phonetic changes in the realisation (which very arguably are the case here anyway), I advise you take that up with the Japanese editors, as there are going to be quite a few issues. Stop being awkward.Theknightwho (talk)21:32, 28 November 2025 (UTC)Reply
Per Wikipedia, PSM means "the word's non-native quality is hidden by replacing it with phonetically and semantically similar words or roots from the adopting language. Thus the approximate sound and meaning of the original expression in the source language are preserved, though the new expression (the PSM – the phono-semantic match) in the target language may sound native" and is "camouflaged borrowing ... [as] 'simultaneous substitution and importation'". Simply changing the spelling of a loanword because one mistakenly thinks it's related to a native word is not what PSM is. —Mahāgaja ·talk21:39, 28 November 2025 (UTC)Reply
Salem was traditionally identified with Jerusalem, primarily as a way of claiming a positive connection between Jerusalem and the people of Abraham going back to the days of the Abraham himself. It would be perfectly normal for early Jews or Christians to associate "Yerushalayim" with "Hiereus" + "Shalem", especially since early written Greek didn't have diacritics to show the "h" sound.
As for phonosemantic matching: I always understood it to be a way of "language-washing" foreign words to seem less foreign by replacing pieces of the foreign term with the phonetically-closest thing that could be found in the borrowing language whichsort of meant something like the meaning of the foreign term. It's entirely possible for the same borrowing to be interpreted either as PSM or folk etymology. This one seems to me just Christians thinking thatἱερεὺς(hiereùs) andΣαλήμ(Salḗm) are in the same semantic neighborhood, so it would just be natural for them to go together in forming the word for a synonym of Salem that's associated with priests.Chuck Entz (talk)19:47, 27 November 2025 (UTC)Reply
If indeed it should be difficult to distinguishphono-semantic matching fromfolk etymology, for the sake of accesibility we should avoid the former term in favour of the more common latter term.
You may well be right about the influence of those Bible passages. I'd argue however that this is a clear-cut case of PSM:
To echo FB, a partial disguise of a foreign word is not a disguise at all.Ἱερουσαλήμ(Hierousalḗm) is still obviously foreign. Kind of like a meat dish with added vegetables is still a meat dish.
PSM can't apply to names because they are normally understood to not have a meaning. You could disguise the loan, but it wouldn't be PSM.
By definition, a PSM is intended to hide the foreignness of the word by making it look like a native word.Ἱερουσαλήμ(Hierousalḗm), being an indeclinable proper noun ending inμ, still looks very, very foreign, and the-υσαλήμ portion doesn't mean anything in Greek. Spelling it with a rough breathing instead of a smooth is in no way hiding its foreignness. The alternative nameἹεροσόλυμα(Hierosóluma) comes a little bit closer because at least it looks like itcould be a Greek word, though theΣόλυμα(Sóluma) portion still doesn't carry any meaning beyond being adifferent foreign place name. But in that case you can at least imagine people thinking "there's regular Solyma in Lycia, and Holy Solyma in Judaea", which is a weak case for PSM. But in the case ofἹερουσαλήμ(Hierousalḗm) there's simply no plausible argument for PSM whatsoever. —Mahāgaja ·talk10:06, 29 November 2025 (UTC)Reply
@Mahagaja My issue with the argument that it's intended to hide the foreignness of the word is that that would undermine the concept of partial PSMs altogether, so it cannot be a requirement.Theknightwho (talk)16:17, 29 November 2025 (UTC)Reply
I don't see how it would do that. None of the etymologies on wiktionary mentioning "partial phono-semantic matching" are comparable. As far as I can tell (the glossing in these etymologies is poor) it's the semantic match which is partial. None of them only partially hide the foreignness. —Caoimhin ceallach (talk)19:35, 29 November 2025 (UTC)Reply
Latest comment:1 month 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:1 month 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:29 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,[5] 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 ago5 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:25 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:23 days 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
De Vaan dismisses the "older etymology as*sem-ǵno- to the root*ǵenh₁- ‘to be born’" without explanation. His alternative involving*gʰe precludes connection with non-Italic cognates (it would require PG*ainagulaz or similar). I would like to know why he considers*sem-ǵno- so unlikely. I actually like it. It would allow connecting Proto-Germanic*ainakulaz(“single”), Latinsingulī(“one apiece”), Sanskritअकज(ekajá-,“born alone, not a twin”), and Ancient Greekἵγγια(híngia,“one(?)”), with various dissimilations and substitutions of language-specific words for “one”.[1]Caoimhin ceallach (talk)18:40, 29 November 2025 (UTC)Reply
I don't understand why EWN[2] states there are no Germanic cognates when there are Frisian and Low German cognates. Or are they from Dutch? Gothic𐌰𐌹𐌽𐌰𐌺𐌻𐍃(ainakls) and Old Norseekkill have slightly different suffixes, but I wonder if they could be argued to be cognate anyway. —Caoimhin ceallach (talk)22:02, 29 November 2025 (UTC)Reply
Latest comment:19 days ago4 comments3 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.
It is not a prefixdes- but a variant ofesguazar with epenthesisd-. According to Corominas&Pascual, it is akin to the old Italian sense. The Spanish obsolete sense "to smooth a piece of wood with an axe" was used by the navy as "to dismantle a ship", then "to dismantle something, to undo into pieces".Vriullop (talk)09:22, 26 November 2025 (UTC)Reply
Latest comment:23 days 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:16 days ago7 comments3 people 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
^De Vaan, Michiel (2008),Etymological Dictionary of Latin and the other Italic Languages (Leiden Indo-European Etymological Dictionary Series; 7), Leiden, Boston: Brill,→ISBN,page390
ᄇᆡᆨ셩의게 was used in 텬로력뎡, 주교요지, 셩경젼셔,The Independentment, and 어제유원춘도영동영서대소사민윤음. Since ᄇᆡᆨ셩 is an outdated word for백성, I think "의게" is an archaic form of에게.~2025-35981-93 (talk)
Latest comment:20 days ago2 comments2 people in discussion
We have adverbs likeгореводом(horevodom,“upstream”) where, even with the "wrong" instrumental suffix, we can figure out that the root word for*водом(*vodom) isвода(voda). I have absolutely no idea what native Slavic word starts withbzd- and is followed by a vowel of some sort. There certainly aren't any such words still attested in Pannonian Rusyn.
There also exists an synonymous adverb, that beingгорездном(horezdnom). No idea what the root word for this would be either. Feel free to check out the Old Slovak dictionary at{{R:sk:HSSJ}}, if you think you can find it there. I certainly couldn't.Dijacz (talk)21:03, 24 November 2025 (UTC)Reply
There are no sources given, and the online dictionareis I find don't seem to trace it further back than the 1300s (piehus). The connection to the similar Latin word is unclear, as both the English and Latin word aren't attested until the medieval period.Wakuran (talk)12:35, 25 November 2025 (UTC)Reply
That's possible. However, relation to magpie (the bird) could be merely coincidental or indirect. The parallel example ofchewet really doesn't make a lot of sense, since the "meat dish" sense isprobably derived fromchew is unknown, and the "chough, jackdaw" sense is derived from a different word altogether (Frenchchouette). I'm debating whether to even keep it in the etymology. And the correspondence betweenhaggess "magpie" andhaggis "haggis" is not reliable either, since, as in the case withchewet the dish is older than the name of the bird. On the other hand,pie "magpie as a name" is attested earlier thanpie "pastry" if we consider surnames (1177 forRadulfus Pie; 1199 forHenricus Piehus). Maybe the pastry dish was named after a creator whose last name wasPie and it caught on (?). At the time the namePiehus was formed,pie was not used for "magpie" - the word at that time for a magpie wasaguster ("haggister").Leasnam (talk)16:22, 26 November 2025 (UTC)Reply
The speculative portion of the etymology consumes 2 inches of vertical space on my screen. It needs to be hidden beneath a show-hide box.DCDuring (talk)17:28, 26 November 2025 (UTC)Reply
Thanks for all the hard work. We now have ‘pie’ possibly connected to magpies rather than insects (insect pie doesn’t sound very tasty but magpies might be) and an entry for what is essentially the Old English word for a bee. Much betterǃOverlordnat1 (talk)20:24, 26 November 2025 (UTC)Reply
Just a guess: it might have something to do with the well-known magpie behavior of picking up small shiny/colorful objects and sticking them in their nests, with the crust representing the nest and the filling representing the assortment of objects.Chuck Entz (talk)23:15, 28 November 2025 (UTC)Reply
It sounds somewhat strained. On a side note, apparently ornithologists have noticed that the idea is incorrect. Magpies are very intelligent birds, and they would be wary of small, shiny objects instead of being attracted to them.Wakuran (talk)00:30, 29 November 2025 (UTC)Reply
Latest comment:17 days ago5 comments2 people in discussion
The OED must have been all ears when they wrote the etymology that + away. It is clearly anaptyctic for 'that way'. Any other such English formation? Also, this 'pronominal adverbs' is very ugly without a header.Saumache (talk)14:14, 26 November 2025 (UTC)Reply
Latest comment:17 days ago1 comment1 person in discussion
With due respect to our good friend Thadh, I have looked in a handful of places and cannot find any indication of a dialectal Old Slovak variant ofjež being*již. Inherited Proto-Slavic word-initial*e- and*ě- usually result inє-(je-) in Pannonian Rusyn, such asєшень(ješenʹ) andєдло(jedlo). I'm personally more convinced that this is a Carpathian Rusyn borrowing - Carpathian itself has several varieties, and it's not unthinkable that in addition toї́жо(jížo), there also exists the variantїж(již). In fact,this particular Slovak-Rusyn dictionary givesїж(již) when you look up "jež".
If anyone can find an Old Slovak record of*již, let me know. But for the time being, I'm changing the etymology to a Carpathian borrowing.Dijacz (talk)18:51, 27 November 2025 (UTC)Reply
Latest comment:17 days ago1 comment1 person in discussion
So as per sources, e.g. OED, seems to originate c. 1500 straightforwardly from the figurative sense of cloud, which makes sense to me. Ourcloud has the closest sense (11.) restricted toaspect of something positive though, is it worth creating subsenses and split the cross references? Does the restriction actually exist?
Latest comment:15 days ago3 comments2 people in discussion
Currently,Sanskritvā́ha is transcribed with accent on the first syllable and derived fromProto-Indo-European*wóǵʰ-o-s. However, as far as I can tell, the Sanskrit term actually should derive from PIE*woǵʰ-ó-s and should place the accent on the second syllable. CompareProto-Indo-European*bʰóros, another tomos-type noun, which producedSanskritभर(bhára). However,*bʰorós producedSanskritbhārá. The only sources on the Sanskrit term I could find agree with this interpretation: Laura Grestenberger (page 7) and the workLaws and Rules in Indo-European (page 246) both derive the Sanskrit word from*woǵh-ó-. It may also be worth noting that there is supposedly an Avestan cognate[script needed](vāza-), which would provide a nice parallel for the Sanskrit term.
Also, whilst doing some searching, I found that the etymology ofSanskritघन(ghana) is also inaccurately listed on the PIE pages. The actual page for the Sanskrit entry derives it from*gʷʰónos, but it is also listed as a derivative of*gʷʰonós.Graearms (talk)03:31, 28 November 2025 (UTC)Reply
You're right on both counts. SeeGrassman (search for vAha) andNWS. According tow:Jochem Schindler's old theory*tómh₁-o-s-type nouns action nouns ‘(act of) cutting’, sometimes particularised into ‘(individual) cut, slice’, while*tomh₁-ó-s-type nouns are "possessive" derivations of the former: ‘possessing cutting’ → ‘sharp’ → ‘cutter’ (active) or ‘possessing cutting’ → ‘(having been) cut’ → ‘hole’ (passive). That's the theory at least. —Caoimhin ceallach (talk)00:05, 29 November 2025 (UTC)Reply
@Caoimhin ceallach I've gone and removed the erroneous information from the PIE pages. I am, however, not willing to mess with the Sanskrit inflection or transliteration tables. So, I suppose another user will have to fix that.Graearms (talk)00:26, 30 November 2025 (UTC)Reply
Latest comment:14 days ago5 comments4 people in discussion
I was looking through the Indo-European Comparative Dictionary by Stuart Mann, and I found that it implies thatOld Armenianերաշտ(erašt) and perhapsPersianرشت might be derived from the word*tr̥stós (onpage 1447). Currently, the further etymology of both the Armenian and Persian terms is left somewhat unexplored. Would a connection with at least the root*ters-(“dry”) be possible?Graearms (talk)00:57, 30 November 2025 (UTC)Reply
You shouldn't look at Mann, an unreliable work. The regular reflex of PIE*tr̥s- is Armenianթառ-(tʻaṙ-).երաշտ(erašt) is borrowed from Middle Iranian *rašt, with prothetice- added in Armenian because we don't tolerate words starting withr-. The further origin of that Iranian word should be solved by Iranists.Vahag (talk)09:56, 30 November 2025 (UTC)Reply
They alsosuggest (based on prevailing views) that Dutch-ster comes from Latin-istria, which it likely doesn't. Their reasoning is that for a suffix to be borrowed from Latin, it would have to be borrowed in a large enough number of terms containing the suffix before it can be viewed as a morpheme attachable to new words, and we do not find this to be the case for either-ster or-egge. Their explanation that-egge waspossibly formed ultimately from a rebracketing of Vulgar Latin-trica,triga (Latin-trix) also doesn't fill me with confidence; seems like a stretch (btw,Old English-iċġe cannot simply come from*-iga). It's easier to postulate a Germanic origin for this suffix, as many PWGmc verbs also contain the ending-igōn and can easily produce such a suffix by adding-jā.Leasnam (talk)18:26, 1 December 2025 (UTC)Reply
I did a quick search of Old English words containing-icge (in another db) and I wasn't able to find any terms using the suffix that are attached to a Latin base. This tends to speak rather loudly against it originating from a Latin source.Leasnam (talk)18:31, 1 December 2025 (UTC)Reply
That makes sense. The original proposal is inBeiträge zur Geschichte der deutschen Sprache und Literatur 56 (1932), p.24, but I can't access it.Exarchus (talk)19:27, 1 December 2025 (UTC)Reply
Latest comment:13 days ago2 comments2 people in discussion
Currently, these terms are listed simultaneously as descendants of*h₁réwdʰeti and*h₁rudʰrós. So, I suppose one of these etymologies must be inaccurate. Moreover,օրոյդ(ōroyd) andորույդ(oruyd) are also listed as descendants of*h₁réwdʰeti. Fortunately, it appears a certain Hrachia Acharian already attempted to find the etymology of these terms (seehere on page 331 andhere on page 106). However, Acharian appears to treat them as reconstructed.Graearms (talk)17:07, 1 December 2025 (UTC)Reply
Acharyan conducted a mental experiment of what would a native Armenian word look like if it had survived. A stupid IP took his asterisked forms and spread them all over Wiktionary. Why are IPs like this? I will delete them. --Vahag (talk)10:17, 2 December 2025 (UTC)Reply
Latest comment:13 days ago2 comments2 people in discussion
Currently, the origin is left as unknown. I recently stumbled across the workIn-laws and outlaws in Indo-European societies by the linguist Birgit Anette Olsen, who (on pages 327-328) tentatively suggests a possible derivation from*h₂m̥bʰi +*sokʷ-eh₂ +*h₂onk-yo-. The second component of this compound is the source of*sokʷh₂ṓy and perhapsAncient Greekὀπάων(opáōn). The third compound is apparently paralleled by the element-ունչ(-unčʻ) inOld Armenianանտերունչ(anterunčʻ,“without master”), which Olsen explains as a derivation from*h₂onk-yo-.Graearms (talk)22:06, 1 December 2025 (UTC)Reply
This seems to imply that the non-simple metathesis/r-k-l/ </l-k-r/ occurred independently in various languages evolving from Vulgar Latin. How plausible is this? ‑‑Lambiam22:25, 1 December 2025 (UTC)Reply
Latest comment:13 days ago10 comments3 people in discussion
Hi! I just added a etymology source from Polish,kilo is just borrowed from Greek κιλό but the problem is, is it direct borrowing or learned borrowing? is it also rfv-etymologyフィリピン人 (talk)09:29, 2 December 2025 (UTC)Reply
I suspect the first language to clip its word forkilogram tokilo was French, then everyone else followed suit, but it's probably impossible to prove at this point. Even our entry for Greekκιλό(kiló) says it's borrowed from Frenchkilo. —Mahāgaja ·talk10:26, 2 December 2025 (UTC)Reply
@Mahagaja@Adamnewwikipedianaccount The EIEC suggests that the term may be a Semitic borrowing. Apparently, in one inscriptionHurrian[script needed](šittanna) is used to translate Mitannisatta-wartana(seven laps of a horse (around a track)). This word could be a compound of terms related toSanskritsaptán andvartana. Moreover, the EIEC notes a comparison withProto-Kartvelian*šwid-. Based on this evidence, the EIEC postulates that it was a culture word spread throughout Anatolia and Upper Mesopotamia during the 14th-century BCE.[1] Václav Blažek also seemingly suggests thatProto-Uralic*ćäjćemä (or at least the descendants of this term) may be borrowed from Balto-Slavic, that the terms currently derived fromProto-Ugric*säptɜ may actually be borrowed from Tocharian, and thatBasquezazpi was probably borrowed from a Semitic source close to Egyptian.[2] There is alsoEtruscansemφ, which has also been explained as an Indo-European borrowing. But I also foundthis source which (on page 215) considers the Etruscan to term actually mean "eight." Regardless, his argument is essentially that the term is wanderwort.
The Hurrian borrowings appear a bunch of other similar terms, such asHurrianaikawartanna, which might parallelSanskritekavartana.[3] This is all related to the Indo-Aryan superstrate in Hurrian. I personally am not convinced that these terms really reflect the borrowing of the numerals themselves into Hurrian. English terms likeoctogon showcase a component borrowed from the Latin numeralocto, butocto itself is not an English word. I think we maybe should deleteHurrianaika(“one”) unless there is specific attestation of the word functioning as a numeral and not as a component of a compound term.Graearms (talk)21:22, 2 December 2025 (UTC)Reply
The oldest evidence of septḿ̥ appears to be in Old Assyrian transmission, probably from a Luwian loanword in Hittite, Kanišite Hittite that is. Cuneiform numerals are mostly spelled logographically, leaving rather scant evidence of phonetic spelling. Numeralia from*sḗm instead of*h₁óynos and speculative derivation of*mey vel sim. (missing in Wiktionary) instead of*kʷetwóres do not bolster confidence in the Proto-Indo-European reconstruction of a decimal number system. Borrowing from another language which had developed writing already by 3000 BCE for the specific purpose of accounting is not unrealistic. Wiktionary does not distinguish late core-PIE from early Indo-Anatolian, although there is growing concensus. Anything you could say about it is possibly fringe, because there is no agreement on important details. The loanword hypothesis of *septḿ̥ is one of those problems and I would argue vehemently for it to be included in the entry. Nota Bene: Bjørn (2023) has published in Historische Sprachforschung vol. 135, being one of the leading journals of Indo-European studies, not a sort of self-published hobby horse.Friya Willie (talk)08:01, 6 December 2025 (UTC)Reply
I'm inclined to agree that the possibility of a Semitic origin should at least be mentioned, even if just because of how many prominent philologists have supported the theory.
After doing a bit of research I found the idea affirmed by:
^Blažek, Václav (1999),Numerals: comparative-etymological analyses of numeral systems and their implications (Opera Universitatis Masarykianae Brunensis, Facultas philosophica;322)[2], Brno: Masarykova Univerzita,page246
^Bjørn, Rasmus G. “The Lexicon of an Old European Afro-Asiatic Language. Evidence from Early Loanwords in Proto-Indo-European1.” Historische Sprachforschung / Historical Linguistics, vol. 135, 2022, pp. 3–42. JSTOR,https://www.jstor.org/stable/27257695. Accessed 6 Dec. 2025.
Latest comment:9 days ago2 comments2 people in discussion
I've noticed that the template for pronunciation of Sanskrit used here in etymology section produces automatically two lines with two variants of pronunciation (classical Sanskrit, plus reconstructed Vedic pronunciation). It would be good to find another solution how to list pronunciation of Sanskrit, that would list only classical Sanskrit pronunciation. The Vedic pronunciation is relevant only in rare situations: when comparing across ancient Indo-European languages (though even there we can do without, usually). Otherwise it is irrelevant. What's worse, it'll only be confusing for any reader other than historical linguist. Moreover, in case of words which we know did not exist in the period when pronunciation was such as the "IPA (Vedic)" line gives, it's a misinformation. This situation would be similar to listing Old English pronunciation to English lexemes of French origin.
I'm no expert, but I believe the sources claiming hard 'g' in Old English are inaccurate. It's clear that there must have been palatisation of the 'g' (evidenced byencgel and in derivativesencgelcyn) that would have resulted from the i-mutation that also produced the mutated vowel in the first syllable ('æ' and 'e') from earlier*angil. I-mutation causes Germanic 'ng' to invariably become Old English /nd͡ʒ/. And of course, the French influence is evident in some of the later Middle English variants likeaungel (though the spellingau can also be found in words that are clearly not of French origin, likelaund (land),haund (hand) [cf.haundmaidis (handmaids)],wraung (wrong),faung (grasp, catch), etc.); yet we're talking about adeeply established term in English.Leasnam (talk)18:42, 4 December 2025 (UTC)Reply
I see that Middle Englishaungel lists a few alternative forms (engel,ængel,ongel,ennꟑell) that could not have been borrowed from Old Normanangle (as per the etymology) but which are assigned to Old English. I note too that in one of the MED passages [a1325 (c1250) Gen.& Ex.(Corp-C 444)1803-7] it reads:Iacob..ðor wrestelede anengel wið..Get held he wið ðisangel fast., soengel andangel are alternatives of the same word and are interchangeable. The etymology ataungel should either state thatengel~angel (from Old English) andangel~aungel (from Old French) are different terms; OR, preferably that the Middle English word is an assemblage of both (likely a French influence on a pre-existing English word).Leasnam (talk)19:39, 4 December 2025 (UTC)Reply
If you're using the spelling "cg" as proof of a pronunciation with /d͡ʒ/, that doesn't actually follow: we see "cg" occasionally after the letter "n" even in contexts where palatalization was not regular, such as in the ending "ung"/"uncg" in words likeþenuncg forþenung, or inancgmod forangmod. In that kind of context, "cg" probably represented the velar plosive (which it is speculated may have become devoiced for some speakers) as opposed to the voiced velar fricative that "g" usually represented. I agree that the i-mutated vowel indicates that palatalization should have applied: however, I think forms that contained "engl-" after syncope would be expected to have converted the preconsonantal palatal to a velar, and this could have been leveled into the nominative/accusative.--Urszag (talk)20:29, 4 December 2025 (UTC)Reply
Right, the combination 'cg' doesn't universally represent the affricate sound in all positions (docga,picgbrēad, etc.), but under the conditions inencgel, before a front vowel (< 'i'), it does. Whether the velar 'g' of the oblique cases replaces the affricate 'g' in the nom/acc singular is a matter of debate. Personally, I don't find it likely (comparewenċel~wencle~wencles) - though plural forms with 'c'/'k' hung around into Middle English, a hard 'k' was never refitted to the nom/acc singular.Leasnam (talk)00:09, 5 December 2025 (UTC)Reply
The formennꟑell is consistently found for "angel" in the mostly phonemic orthography of theOrmulum, where the grapheme ⟨ꟑ⟩ represents/ɡ/. This probably comes from the nominative/accusative singular being levelled out, confirming Urszag's hypothesis. However, this levelling didn't necessarily occur in all Middle English texts that have a direct descendant ofenġel, especially since the pronunciation with/d͡ʒ/ would've been supported by Medieval Latinangelus and Old Frenchangel,angele.Hazarasp (parlement ·werkis)02:29, 5 December 2025 (UTC)Reply
Update: a cursory Googling reveals thatSlovakDžuňa is also a thing. Not that that brings us any closer to the etymology, but at least we've got a timeframe, i.e. pre-Pannonian settlement (pre-~1800s).Dijacz (talk)21:45, 4 December 2025 (UTC)Reply
Latest comment:7 days ago4 comments2 people in discussion
Currently, no etymology is mentioned. I think this term might relate to the Serbo-Croatian termmrva that Pokorny mentions on page 736 of the IEW,[1] which is derived from the root now reconstructed as*merh₂-.Graearms (talk)04:10, 7 December 2025 (UTC)Reply
Latest comment:6 days ago11 comments8 people in discussion
I have noticed several times that the etymology of Italian words (e.g.adiuvare) often includes the phrase "borrowed from Latin," which is even funnier because if you click on the linked article, you will read: "A word that was adopted (borrowed) from another language,rather than formed within the language orinherited from a more ancient form of the same language" – and Latin is precisely the latter case. So there is a serious contradiction here that needs to be resolved somehow. It doesn't make much sense to say that a language "borrowed" words from itself, from an earlier period of itself – or from another variant of the same language.
Whatever we call this relationship between Italian words and their Latin origins, one thing is certain: this relationship cannot be described as borrowing.Bennó (talk)19:34, 7 December 2025 (UTC)Reply
"It doesn't make much sense to say that a language "borrowed" words from itself, from an earlier period of itself – or from another variant of the same language" is wrong - in fact, it makes perfect sense. Latin was used as a relatively fossilized written language for over a millennium, and the 'vernacular' forms of Latin that developed later, e.g. Italian, could absolutelyborrow words and terms from that Latin. —SURJECTION/ T/ C/ L/19:54, 7 December 2025 (UTC)Reply
For an example from Portuguese (another Romance) language, Latinarticulus ended up developing intothree Portuguese words:artelho (inherited),artículo (learned borrowing), andartigo (somewhere in between). These were definitely not all threeinherited from Latin, despite the fact that Portuguese as a language developed from Latin. —SURJECTION/ T/ C/ L/20:00, 7 December 2025 (UTC)Reply
I am a philologist specializing in ecclesiastical Latin, and I am familiar with the linguistic history and with "scholarly borrowing".There are complicated cases. However, as is evident from the example I provided, I am not talking about this, but about words belonging to the core vocabulary, such as the above-mentionedadiuvare, which are clearly not the result of subsequent or scholarly borrowing.It seems to me that someone regularly adds/has added the phrase "borrowed from Latin" to words of Latin origin, but in the vast majority of cases this label is misleading and incorrect.Bennó (talk)20:17, 7 December 2025 (UTC)Reply
Ecclesiastical Latin is only one route of borrowing: Latin was used for laws and other legal communications, and as the international language of philosophers and scholars. It's possible to find Latin writings on just about any subject fit for discussion in formal settings.
Italian came from Latin, but underwent irreversible sound changes that easily distinguish it from Latin. If it hadn't, it would still be Latin and not Italian.
There are a number of levels of borrowing: an Italian phrase might be "dressed up" as Latin by changing the form of the words to imitate Latin, a Latin phrase might be have its words individually replaced by the equivalent Italian words, or whole words or phrases might be borrowed unaltered. Then there are loans from other languages: Englishbaseball traces both parts back through other languages to Latin. Really, I'm just scratching the surface. To be inherited, it would have to have been in use by speakers as their language changed from Latin to Italian- which would have involved all the sound changes. There are exceptions where sacred texts were memorized word for word and taught to future generations to be memorized unaltered in turn, but we're not talking about those.Chuck Entz (talk)21:44, 7 December 2025 (UTC)Reply
"I am a philologist specializing in ecclesiastical Latin" - I really hope you're either lying or using that word in a more general sense, because it would be terrifying if a professional linguist specialising in Ecclesiastical Latin could say such nonsense as you did above. All Romance languages borrow from Latin and nobody denies that fact.~2025-38199-92 (talk)19:54, 8 December 2025 (UTC)Reply
If we are consequential, there is no such thing as inheritance at all, but rather people borrow words they heard from other people, whether in sound or in writing. When using language you repeat a mental representation of a word you can have anywhere from. Yet we have differentterminology, the technical term of borrowing and that of inheritance, depending on whether the same thing happened across language barriers (at least the categorization of a borrowing is not used when a term has been borrowed from an older stage of what we treat as the same language, say 20th-century English borrowing from 17th-century English → batlet, so the premise of the terminology being based on language barriers alone is evidently not strictly necessary, but it is just descriptive of a certain remoteness of the source of a word). We avoid it completely again increoles andmixed languages, where we speak oflexicalizers, or perhaps there is inheritance from two languages that are not related by means of descendance (like someSouth Cushitic languages got replaced by Bantu but some became Frankensteins). There is also the issue ofproper nouns having different dynamics.Fay Freak (talk)20:54, 8 December 2025 (UTC)Reply
Latest comment:6 days ago3 comments2 people in discussion
Can anyone attest something likeالربيع الحكة (which literally means "itchy grass") being used in Moroccan Arabic (or any other variety) with the meaning ofnettle. If not, it could also be an independent development.Lankdadank (talk)00:24, 8 December 2025 (UTC)Reply
**الربيع الحكة is grammatically incorrect, it would be*ربيع الحكة(“rbīʕ el-ḥukka, rbīʕ el-ḥakka”). I've never heard of this term and I can't find any ghits. The common term in Morocco isحريقة(ḥurrīqa). It is likely an internal coinage. —Fenakhay(حيطي ·مساهماتي)21:02, 8 December 2025 (UTC)Reply
Phonologically speaking I can see how you can derivedeifyaw from either PIE*dh₂u-yé-ti or*déh₂u-ye-ti or*doh₂w-éye-ti. All formations are also plausible for PC. It's difficult to decide. Middle Irishdóid doesn't prove a long*o. I think LIV prefers*doh₂w-éye-ti because Middle Welshkynnut(“firewood”) seems to be derived from it. Zair seems to agree.[4] Greekδαίω(daíō) has the same problem and anyway LIV thinks it's a new formation. The only thing which could plausibly point either way is if Germanic*tuskjan- is somehow morphologically related to PIE*dh₂u-yé-ti. Don't know. @Mellohi!? —Caoimhin ceallach (talk)15:45, 10 December 2025 (UTC)Reply
Latest comment:6 days ago1 comment1 person in discussion
Can someone take a look at this etymology? Particularly the partFirst witnessed as a verb in a variant reading 欲太知 (よだつ, yodatu) at Manyōshū 3480 (万葉･三四八〇) quote "欲太知来ぬかも" seems incorrect. The man'yōgana for this portion of poem 3480 is欲太知伎努可母(yo1dati ki1no1 kamo), usually glossed as夜立ち来のかも(yodachi kino kamo), which would make欲太知(yo1dati) a simple compound of夜(yo,“night”) +立ち(tachi,“departure”), and unrelated to any(y)e ~ yo word for "duty"/"labor".Horse Battery (talk)01:16, 9 December 2025 (UTC)Reply
på (Danish, Swedish, Norwegian): rebracketing or elision?
Latest comment:3 days ago11 comments6 people in discussion
The claim that på in the Scandinavian languages is a rebracketing of uppå, and not just simply an elision of the first syllable, seems suspect. Is there any evidence that a reanalysis of uppå to upp + på really took place?~2025-33636-01 (talk)10:24, 9 December 2025 (UTC)Reply
Here is a translated version of the Swedish equivalent to OED,SAOB:
ETYMOLOGY: Old Swedishpa; cf. Danishpå (formerlypa), Norwegianpå; arisen fromuppå (fromuppe å andupp å') through incorrect segmentation[…], partly possibly also through loss of the initial vowelu; cf.punne
Yes, uppá -> upá -> paa -> på is attested at least in Danish sources if you see the Danish dictionary. For Norwegian, see similar stuff happening totu,tå andpoinni in many dialects. Concerning Swedish, see SAOB, part of which was quoted above, mentions older spellings likeppa, probably remnants of the double "p" inupp. Or do you mean it rather looks like a shortening?Tollef Salemann (talk)18:28, 9 December 2025 (UTC)Reply
I think its the sense change, from upon to simply on they're referring to. it's not the u that stands alone, but the a "on". it's re-bracketing like apron from napron. The n doesn't stand alone, but is re-analysed as part of a different word (an apron rather than a napron). Or French Sant Chamond rebracketed from Sanct Aumond. In this case Upp a is re-analysed as up pa via the compound form uppa, and pa takes on the sense of earlier a "on".Griffon77 (talk) 09:19, 10 December 2025 '(UTC)
Ehm...han går oppå skulen,han kom oppå middagen,kva kallast det oppå norsk - sentences like these are not possible nor in Norwegian, nor in Danish, nor in Swedish, so there is quite a sense change.Tollef Salemann (talk)19:50, 10 December 2025 (UTC)Reply
Exactly, see your source, "Ordet användes iä. sv. i stort sett på motsv. sätt som PÅ", what meansThe word is used inOld Swedish almost same way as PÅ — altså, Old Swedish is not same as modern Swedish. But the sense change for the figurative examples - is it not rebracketing?Tollef Salemann (talk)20:15, 11 December 2025 (UTC)Reply
@Griffon77: With napron → apron the rebracketing is from "a napron" to "an apron". Because both "a" and "an" are words in English,/əneɪpɹən/ can be interpreted as either "a napron" or "an apron". But if there's no wordu in the Scandinavian languages, then/ɵpoː/ can't be interpreted in two different ways, which means reducing it to just/poː/ is simply apheresis, not rebracketing. —Mahāgaja ·talk12:26, 10 December 2025 (UTC)Reply
I wouldn't say geminated consonants really play any major part in Scandinavia, and /ɵ/ in itself is just an interjection, similar to English 'uh'.Wakuran (talk)12:41, 10 December 2025 (UTC)Reply
Unless the rebracketing they're talking about isupp + å →upp + på, so from ‘uppå is the emphatic form ofå’ to ‘uppå is the emphatic form ofpå’. But was it really the case thatuppå was more emphatic than the other two? I can't find anything in SAOB which points to that. —Caoimhin ceallach (talk)19:12, 10 December 2025 (UTC)Reply
Do we need to reconstruct a PIIr. and PIE form when there are no direct cognates of the noun? Giving "ultimately from PIE *deh₂w-" is enough in my opinion.Exarchus (talk)10:25, 10 December 2025 (UTC)Reply
Latest comment:5 days ago3 comments2 people in discussion
Currently, we have a lemma entry for a PIE adjective*méǵh₂s, but no entry for a PIE root*méǵ-, and many entries reconstruct derived PIE words as being built on a stem *méǵh₂-: e.g. the PIE adjective page contains forms such as *méǵh₂-yōs, *m̥ǵh₂-nós. In contrast, Scrijver 1991 page 480 says "magnus most likely reflects *mǵ-nó-"; De Vaan likewise reconstructs *mǵ-nó-, *mǵ-i(V)s-. Phonetically, inter-consonantal h₂ is expected to become vocalized to Proto-Italic *a (although based on the single example ofsocius, deletion of *H might have been regular in *-CHi̯V-). While we could suppose original *magano- with subsequent syncope, that seems less parsimonious than just supposing the laryngeal wasn't there to begin with. Sihler, apparently citing Proto-Italic forms (not explicitly identified as such), gives reconstructions such as *magyos, mag-yōr-. Does any source explicitly discuss variation between *meǵh₂- and *meǵ- as forms of this root? @Mellohi!,Agamemenon,Kwékwlos,ExarchusUrszag (talk)22:42, 9 December 2025 (UTC)Reply
Similarly,gli/ʎi/ becomesgliela/ʎela/; so it is not really a special case, excluding the fact that/ʎe/, in Italian orthography, is spelled <glie> keeping <i>, and thatgli+accusative is spelled univerbated unlike other the dative+accusative clusters.
I think the+a + part should be removed from these pages if it cannot be verified.
I found a book that supports your argument that this is a case of gli showing the same behavior as other pronouns, and cited it on the pageglie. I edited the other pages to mark them as univerbations ofglie and the respective accusative pronouns.--Urszag (talk)02:12, 10 December 2025 (UTC)Reply
It wasn't really an argument: I was just mentioning in case someone didn't know, and to show that there is not much special.
I don't know what the origin of/i/→/e/ is, but whatever it is, it is probably the same as all datives, not something special aboutgli.
This is about verifying whethera/a/ somehow has something to do as the etymology sections added in 2007 claim. I doubt it, but what do I know! o/Emanuele6 (talk)02:18, 10 December 2025 (UTC)Reply
Yes, I just overlooked gliene by accident at first. As for the origin of the vowel alternation, it can be explained as follows in the case of me te. For these, /e/ is the original vowel sound. Unstressed pretonic /e/ was commonly raised to /i/ in Italian, as indicembre and the prepositiondi. Evidently unstressed pretonic pronouns were treated the same way, so Latin accusative/ablativemē andtē evolved to Italianmi andti in most contexts, but retained the vowel /e/ when stressed, as when after prepositions. The first and second-person singular dative pronouns either underwent the same phonetic evolution, or borrowed the form of the Latin accusative/ablative pronouns, so we also find /e/ in pronoun sequences starting with a dative such as me la, te ne, etc., where the first pronoun came to regularly receive stress. The source I added, Desouvrey 2005, argues that "the appearance of vowele instead ofi ingli is likely to be due to the phonological harmonization of the third person cluster with clusters including first and second person pronouns."--Urszag (talk)02:57, 10 December 2025 (UTC)Reply
As for your question about writing the etymology: it is often a difficult question whether we have an affix and which. The hyphen segmentation is an abstraction added by the intellect of the language describer. There may be even an anti-anti-Communism that is anti-anti- + Communism and one that is anti- + anti-Communism, if the first quote onanti-anti- is correct to assume that we have the first case there. There is no “such as”, each case must be judged individually, by the meaning you understand.Fay Freak (talk)20:15, 10 December 2025 (UTC)Reply
Yep, that's why we don't have it. But that was only for the argument how one might see that which is not used – the hyphens in etymologies, or separate affixes. We also report the rules (of grammar) which worked in users' heads, so generalized “preferences” here are weak points.Fay Freak (talk)04:38, 14 December 2025 (UTC)Reply
I agree with Mahagaja, and would add that if we view terms like "anti-anti(-)feminist" as using "anti-" + "anti-" + "feminist" rather than "anti-anti-" + "feminist", it raises the question of whether we should haveanti-anti- at all, or it should be RFDed as SOP... (For great-great-grandmother et al, we just havegreat-, notgreat-great-.)- -sche(discuss)18:38, 13 December 2025 (UTC)Reply
Latest comment:3 days ago2 comments1 person in discussion
Albanian, RFV of the etymology. two sources are cited without full titles, and I cannot locate the relevant parts in Matzinger (2017)ragweed theatertalk,user12:42, 11 December 2025 (UTC)Reply
additionally: do the postposed article and the noun-forming suffix come from the same source, as the etymology listed currently suggests?ragweed theatertalk,user10:44, 12 December 2025 (UTC)Reply
Latest comment:4 days ago1 comment1 person in discussion
Chinese, RFV of the etymology. I cannot source this, and the second part seems a little shaky on phonetic grounds: the MC-aewng rhyme requires OC *-oŋ, while 羌 is unproblematically an OC *-aŋ. formatting also needs some cleanupragweed theatertalk,user12:57, 11 December 2025 (UTC)Reply
Latest comment:1 day ago4 comments3 people in discussion
Meaning "marble (the ball)". Looking at the ending-ci, I assume they might be derived from Dutch diminutive suffix-tje as can be seen in other Sundanese words borrowed from Dutch (kelenci <konijntje,panci <pannentje,laci <latje, etc.) But what could the original, non-diminutive origin word of that be? Anyone?Udaradingin (talk)15:13, 11 December 2025 (UTC)Reply
In German and the Scandinavian languages, words likeKugel tend to be used for bullets, and balls of hard material (or fire), while word likeBall tend to be used for balls of soft material, but I'm not entirely sure about the semantics in Dutch.Wakuran (talk)13:11, 12 December 2025 (UTC)Reply
Latest comment:1 day ago2 comments2 people in discussion
isReghin really borrowed from German? The English Wikipedia article says it already had the nameRegun in the 13th century. This suggests if German was involved at all that the name wouldve been borrowed into German and then back into Romanian later.—Soap—02:16, 12 December 2025 (UTC)Reply
Latest comment:2 days ago2 comments2 people in discussion
The claimed etymology of the word "hit" in Wiktionary is the following: (Proto-Germanic) *hittijaną --> (Old Norse) hitta --> (Old English) hittan. This doesn't really make sense since the explanation also suggests that the meaning "to strike" was gained in Old Norse and then subsequently lost in Old English and then re-gained in Middle English. I also do not see any citations in the article that back up the claim of an Old Norse origin.~2025-32800-24 (talk)05:41, 13 December 2025 (UTC)Reply
According to Cleasby/Vigfussonp.264/p.265) Old Norse had both senses, but apparently the word was borrowed very late andBosworth-Toller gives only one quote (the "meet" sense), TheMED's etymology says Late Old English "hyttan" came from Old Norse. It looks to me like both senses were borrowed, but only the "meet" sense made it into writing during the Old English period.Chuck Entz (talk)06:50, 13 December 2025 (UTC)Reply
Latest comment:3 hours ago9 comments5 people in discussion
@J3133Doesn't share any semblance of semantic link with Latinpulso (the Latin is first and foremost transitive, whereas ours is intransitive, then read the definitions), I proposepulse +-ate or a morphollogical borrowing from the Latin with senses taken from Englishpulse, which is basically positing the former.
There might actually be a medieval/New Latin medicalpulsō as I foundvena pulsatilis, "pulsing vein". I'll look for the corresponding verb.Saumache (talk)12:20, 13 December 2025 (UTC)Reply
Pulse is felt. Hence there is a measuring instrument or body part applied which is the object of the transitive action. As to mention it appeared repetitive, the object was omitted, and one was not conscientious to keep apart transitive and intransitive verbs when one was not a philologist but a physician in the first place.Fay Freak (talk)14:10, 13 December 2025 (UTC)Reply
Nice verbiage, but, you didn't read the definitions, 'pulse' is not in the Latin whatsoever, not classically anyway. Remainder does not compute.
I did not believe it to be classical – unlike the nounpulsus, a stub in Wiktionary, as we even find the sense already in thesilver Latin ofTacitusAnn. 6, 50 about the personal physician of Tiberiuserat medicus arte insignis, nomine Charicles, non quidem regere valetudines principis solitus, consilii tamen copiam praebere. is velut propria ad negotia digrediens et per speciem officii manum complexus pulsum venarum attigit, andPliny the Elder 2,218et in corpore extrema pulsum venarum, id est spiritus, magis sentiunt. I did not take position about which language the intransitive verb started in at all; the reasoning about usage development often begins neutral towards individual languages.Fay Freak (talk)20:39, 13 December 2025 (UTC)Reply
It seems plausible that back-formation from the nounpulsation played a role. The OED's earliest citation for the noun is in the 1400s whereas the earliest citation for the verb is from 1674. Also, the English verb "beat" is both transitive and intransitive, which could have contributed to the Latin-derived verb "pulsate" being given an equivalent intransitive use.--Urszag (talk)21:57, 13 December 2025 (UTC)Reply
the English verb "beat" is both transitive and intransitive, which could have contributed to the Latin-derived verb "pulsate" being given an equivalent intransitive use. I like this way of thinking. It works equally with Germanschlagen. It would mean, something the dictionaries also missed, that the Latin verb usage as quoted above is calqued from Germanic, if that's correct terminology for syntactic borrowing, which passes the easiest in language contact.Fay Freak (talk)04:33, 14 December 2025 (UTC)Reply
I have updated the page and you can add from what has been said by you or Urszag, why should so many input-able thoughts be left stranded in forum talks.Saumache (talk)14:13, 15 December 2025 (UTC)Reply
Le Trésor appears to give the etymology ofintolérant as:in- +tolérant, modeled after Latinintolerans (“Dér. detolérant* (préf.in-1*) d'apr. le lat.intolerans”). I consider it entirely plausible that the English analysisin- +tolerant is not superficial but a main pathway. Similarly forintolerance. The term is found in English as a noun in 1751.[10] ‑‑Lambiam04:21, 15 December 2025 (UTC)Reply
@Lambiam Let's then stive for completeness and add the third option of in- + tolerant as a main pathway. There is often no right answer in these matters and if anything, these words can be the products of one or the other or one and the others. I'm against the simplistic approach (to sources and material) of some. And he reverted back my edit again.Saumache (talk)09:48, 15 December 2025 (UTC)Reply
@Saumache: Your claim that that it is “nothing but what theOED says” is not true: as I wrote in my edit summary, it is stated in our other reference already in the entry (theOnline Etymology Dictionary),Collins, and others. After I specifically mentioned other dictionaries to you, this seems to be in bad faith. I also linked toWiktionary:Etymology: that etymologies should be referenced; there is not such an exception for your statement that we are a secondary source.J3133 (talk)09:37, 15 December 2025 (UTC)Reply
Between reliable sources such as the OED and other dictionaries, and what is essentially an unreferenced surmise, I would follow the reliable sources until further evidence is presented. —Sgconlaw (talk)13:44, 15 December 2025 (UTC)Reply
@Sgconlaw There is no definite evidence for any of the etymologies given, and it would be naive to think it as straightforward as this, find me the primary source which says "I took it from Latin". Aren't we more honest if we give a number of credible ways the word might have derived through? Dictionaries often take from other dictionaries, including etymologies, what would it be if we never added informations (or surmises) of our own, and in this case, not in the least controversial (or so I think). There is no evidence, either on my part or theirs. So now the page is cluttered with references that have no potency whatsoever, on me or this discussion.
The quantity is somewhat irrelevant. Five of the six are probably merely copying what Murray or his assistants came up with some time before 1915. This is hardly an issue to draw the scrutiny of scholarly philologists. Consider this:
Bothtolerant andtolerable are thought to have come to England the French way, not directly from Latin.
What would an English author, knowing all these words, have done when they needed an antonym fortolerant? Would they have thought, “hmm... let me see if I can find and adapt a Latin antonym of the Latin etymon of Frenchtolérant? ‑‑Lambiam14:24, 15 December 2025 (UTC)Reply
Japanese numerals on Wikipedia.Wikipedia).Chuck Entz (talk)04:55, 7 December 2025 (UTC)
