Wiktionary:Etymology scriptorium/2018/January

Definition from Wiktionary, the free dictionary
Jump to: navigation, search
discussion rooms: Tea roomEtym. scr.Info deskBeer parlourGrease pit ← December 2017 · January 2018 · February 2018 → · (current)


Does anyone have Sinhalese references to check cornac and provide the original script? —Μετάknowledgediscuss/deeds 09:11, 1 January 2018 (UTC)

Hmm. If I put "kūravanāyaka" into {{pi-alt}}, the Sinhala script that comes out is "චොර්නච්". Putting that into {{l|si}} results in චොර්නච් (cornac). I'm not sure what to make of the fact that the automatic transliteration is exactly the same as the spelling of the French word. At any rate, I don't feel confident enough to add චොර්නච් as the spelling of the Sinhala word. —Mahāgaja (formerly Angr) · talk 09:53, 1 January 2018 (UTC)
My Sinhala resources aren't that great, but I'll be looking. {{pi-alt}} is doing something wrong since that is c + o + r + virama + n + c + virama, and I don't see how "kūravanāyaka" needs any virama (compare the Devanagari: कूरवनायक). Also, c is pronounced [t͡ʃ].
नायक (nāyaka) means "leader, guide" in Sanskrit. —AryamanA (मुझसे बात करेंयोगदान) 00:45, 2 January 2018 (UTC)
@Metaknowledge Found the first element finally. It's කූරුව-නායක (kūruva-nāyaka). —AryamanA (मुझसे बात करेंयोगदान) 00:57, 2 January 2018 (UTC)

Outcome of *#RHC- in PGmc and PBS[edit]

I'm trying to clean up *néh₂s (see User:Victar/*néh₂s) but I'm not sure if PIE *n̥h₂s-réh₂ > PGmc *nustrō is a predictable outcome, or if it would have instead yielded *nastrō (as with *nh₂s-réh₂) or *unstrō (as with *n̥s-réh₂). Any other examples of *#RHC- out there in either PGmc or PBS? --Victar (talk) 02:13, 2 January 2018 (UTC)

On a quick check I found the following:
However, we must also consider the following contradictions:
The second set of examples is interesting because we know that -uR- is the normal result in this environment; -Ra- cannot be the regular outcome of any of the root grades. Thus, we know that analogy must have occurred in these cases, and I think that this kind of analogy is regular for #RHC-, since otherwise the zero grade would have looked like *uRC-, which doesn't even begin with the root consonant anymore. We must also consider what happened to *brekaną: the zero grade would have been *burk-, but this was modified. This same treatment was very likely given to laryngeal roots, except that the vowel was -a-. —Rua (mew) 12:29, 2 January 2018 (UTC)
Reconstruction of laryngeal in *meHndʰ- seems to be erroneous, Orel and Beekes agree. Crom daba (talk) 14:05, 2 January 2018 (UTC)
Also *nasō, *nusō alternation is morphological (analogical), not phonological check Griepentrog Wolfgang. Die Wurzelnomina des Germanischen und ihre Vorgeschichte (available at shady Russian sites I hear). Crom daba (talk) 14:37, 2 January 2018 (UTC)
@Crom daba: HA! 29 pages on *néh₂s. If only I spoke German... @Rua, JohnC5 --Victar (talk) 21:00, 2 January 2018 (UTC)
@JohnC5, did you happen to read the paper? Any new insights? --Victar (talk) 07:31, 8 January 2018 (UTC)
@Victar: Sorry for the delay. Griepentrog seems to think that the zero-grade, regardless of whether it had a laryngeal (#n̥Hs- or #n̥s-), would result in *#uns- and that *nus- was rebuilt as *nus- on analogy to *nas-. Alternatively, *nus- was entirely secondary built from *nas-. I'm not sure that helps particularly. He never seems to propose anything like *#R̥HC- > *#RuC-, but that they are all secondary. I'd mention that he says that he finds the paradigm *Hnā́s- ~ *Hnás- for be "unproblematic" at the send of this discussion. —*i̯óh₁nC[5] 05:00, 12 January 2018 (UTC)


If there are any Sinhalese speakers out there, please help with the etymology of pingo. Thanks. — SGconlaw (talk) 18:14, 2 January 2018 (UTC)

We don't have any yet. All I can find for pingo in Sinhala is කද (kada). —AryamanA (मुझसे बात करेंयोगदान) 18:55, 3 January 2018 (UTC)
Any reason to say it's from Sinhalese in particular, other than the country? DTLHS (talk) 00:42, 4 January 2018 (UTC)
One 19th-century encyclopedia (cited in the entry) says it is from Sinhalese. @AryamanA, I also came across this, but I can't read it as the headwords are in Sinhalese. Is it of any help? — SGconlaw (talk) 02:54, 4 January 2018 (UTC)
@Sgconlaw: When I search for "pingo", I can't find any thing remotely close to the word "pingo" in there. I just see කද (kada) again... —AryamanA (मुझसे बात करेंयोगदान) 20:20, 5 January 2018 (UTC)
Hmmm. OK, thanks! Could it be a Tamil word, I wonder? — SGconlaw (talk) 05:09, 6 January 2018 (UTC)
@Sgconlaw: The shape of the word seems weird for Tamil IMO, but you can try this dictionary. —AryamanA (मुझसे बात करेंयोगदान) 01:01, 11 January 2018 (UTC)
Thanks. I found a few terms meaning a carrying pole, but none similar to pingo: காவடி (kāvaṭi), காவுதடி (kāvutaṭi, a pole for the shoulder with ropes attached for carrying burdens or gifts to a temple etc.), சீர் (cīr, shoulder staff for carrying burden), தண்டாயம் (taṇṭāyam, a long pole for carrying burdens), துட்டாப்பு (tuṭṭāppu, a pole with a load carried on the shoulders between two persons), மரம் (maram, a staff for carrying anything by suspension), வாரை (vārai, a pole for carrying burdens). — SGconlaw (talk) 02:43, 11 January 2018 (UTC)

etymology of the word sardonic[edit]

Moved from Requests for Verification:

The etymology for "sardonic" is stated as relating to the plant ranunculus sardous, however, there is evidence to suggest it may also be attributed to oenanthe crocata (https://en.m.wikipedia.org/wiki/Water_dropwort)

Kiwima (talk) 03:13, 3 January 2018 (UTC)


Based on nothing but instinct, I conjecture that "squizz" is an abbreviated form of "squint."

Yes, partially ! I've updated the etymology Leasnam (talk) 21:50, 11 January 2018 (UTC)


I suspect this comes from a rebracketing of the compound spondylolisthesis (spondyl- + olisthesis) as spondylo- + listhesis. The Greek word is indeed ὀλίσθησις (olísthēsis), from ὀλισθάνω (to slip); there's no such word as **λίσθησις.

@SemperBlotto: do listhesis and olisthesis seem to be alternative forms of one another? --Per utramque cavernam (talk) 17:59, 3 January 2018 (UTC)


This shouldn't be a doublet of riff, right? Ultimateria (talk) 18:56, 3 January 2018 (UTC)

Why not? They're both from *krépos. —Mahāgaja (formerly Angr) · talk 19:05, 3 January 2018 (UTC)
I was also surprised to find out this was a doublet. — Eru·tuon 22:52, 3 January 2018 (UTC)

PII language codes[edit]

Language code for Younger Avestan[edit]

Hey, can we finally get a language code for Younger Avestan? Younger Avestan was written centuries after Old/Gathic Avestan, with its own distinct morphological changes. @-sche, AryamanA --Victar (talk) 17:38, 4 January 2018 (UTC)

@Victar: This may cause issues, since we currently have some Young Avestan entries in CAT:Avestan lemmas. We also treat Classical and Vedic Sanskrit as the same language, and they also underwent distinct morphological and phonemic changes. I think we should at least have etym-only codes for Young and Old Avestan though. Also @JohnC5. —AryamanA (मुझसे बात करेंयोगदान) 17:50, 4 January 2018 (UTC)
@AryamanA: And as with Latin. I'd be fine with that: ae-gth or ae-old and ae-yng. --Victar (talk) 18:21, 4 January 2018 (UTC)
I support etym-only codes for Younger and Gathic, but not a full-fledged code for Younger. —Mahāgaja (formerly Angr) · talk 18:55, 4 January 2018 (UTC)
Etym-only codes seem appropriate to me, and accent codes and regional labels if they don't exist. —*i̯óh₁nC[5] 02:25, 5 January 2018 (UTC)
OK, I've added etymology codes for them. Feel free to change the code/name for Old Avestan to Gathic Avestan if that name is preferable (ngrams and my perusal of which name collocates with "Younger Avestan" more both suggest "Old Avestan" is more common). - -sche (discuss) 23:57, 5 January 2018 (UTC)
Thanks @-sche. What do you think @AryamanA, ae-gth or ae-old? --Victar (talk) 00:22, 6 January 2018 (UTC)
@Victar: ae-old, since "Gathic Avestan" seems to be dated. Great work on Iranian btw, it was sorely needed. —AryamanA (मुझसे बात करेंयोगदान) 00:24, 6 January 2018 (UTC)
@AryamanA: Cool, and thanks! --Victar (talk) 05:35, 6 January 2018 (UTC)
@Victar: Do you think we also need etym-only codes for Digor and Iron? —*i̯óh₁nC[5] 04:21, 6 January 2018 (UTC)
@JohnC5: Good idea. Done. --Victar (talk) 05:35, 6 January 2018 (UTC)
How intelligible are the varieties? Do they need full codes, even, as proposed on Wiktionary:Requests for moves, mergers and splits#Ossetian? - -sche (discuss) 23:19, 8 January 2018 (UTC)
@-sche: I believe they're pretty intelligible -- atleast enough to say the differences are highly predictable. I can't profess to be extremely knowledgeable, however. --Victar (talk) 18:54, 9 January 2018 (UTC)

Kermanic/Central Iranian dialects[edit]

@-sche, there are over a dozen dialects of Kermanic/Central Iranian (glottolog/ethnologue) that lack language codes. Even if I just had a language family code, I could use that, but I right now I have nothing. See *druwištah. --Victar (talk) 21:26, 4 January 2018 (UTC)

I'll wait to see if any editors who are knowledgeable of Iranian languages want to weigh in, particularly Vahag (pinged below) and @ZxxZxxZ, but I'll look into the matter if they don't comment. Do you think these lects need full codes (so they can have their own entries with their own language headers, etc), or just "etymology-only" codes? - -sche (discuss) 23:30, 8 January 2018 (UTC)
I don't have anything useful to contribute. --Vahag (talk) 16:51, 9 January 2018 (UTC)
@-sche: Kermanic/Central Plateau/Central Iranian is very much a dialectical continuum. Some people try to break them up it cardinal directions, but I'm not even going to bother. The more widely spoken lects already have codes, ex. gzi, gbz, nyq, but again, I don't find them needed and etym-only codes will do fine. I just need that Kermanic code for the smaller ones. --Victar (talk) 18:54, 9 January 2018 (UTC)
To be clear, I'm looking to get a family code for Kermanic/Central Plateau/Central Iranian. --Victar (talk) 20:44, 9 January 2018 (UTC)
@Victar: my knowledge of Iranian languages is limited, and I'm having trouble finding evidence that "Kermanic" is a common name for a group of languages (Google Books doesn't show me any books using it; I only see some linguistic websites), and "Central Iranian" seems(?) like a name for a different (larger) group, so I'm not sure what to call the family if I add a code for it. I also note that Wikipedia seems to call it an "areal grouping". Where is the family code going to be used, anyway? It seems possible to have pages like *druwištah without such a code. Is it for etymologies, and if so, is it really a genetic grouping or an areal one? - -sche (discuss) 18:47, 11 January 2018 (UTC)
@-sche: The family is mostly referred to by the more generic Central Dialects (CLI, Cheung:2007, Novak:2013, Cathcart:2015) and often abbreviated as CD(s). As I said, it is a dialect continuum and they're all extremely similar to one another, and its not merely an areal grouping (despite the infinite wisdom of Wikipedia). If you actually look at the code on *druwištah, you'll see I'm temporarily using {{l|unk}} for the entries, and those are just a small sampling. If you can't give me a family code, I'll need codes for each missing dialect, which number in the dozens. Your call. --Victar (talk) 03:42, 12 January 2018 (UTC)
For a paper that uses Kermanic/Kermanian, see Recherches sur les dialectes kermaniens (Iran Central), which also explains why it is a better term to use than the very generic Central Dialects. --Victar (talk) 17:54, 12 January 2018 (UTC)
OK, thanks (sincerely, not sarcastically) for your persistently informative comments! I've added ira-ker as "Kermanic". I'm not sure it helps your use case, though, since you can't use a family code in {{l}}...? If you want etymology-only (or full) language codes for the dialect or for "Kermanic" as a conflation of them all into one language, let me know. (If the latter is what you wanted in the first place, I apologize for misunderstanding.) - -sche (discuss) 18:03, 12 January 2018 (UTC)
@-sche: Yeah, I need a silo code for all the 40-some unnamed dialects that I can use in {{l}} and even create entries for. --Victar (talk) 20:58, 12 January 2018 (UTC)
A family code and a language code can't have the same name (or code, but that seems easier to get around), so if we make "Kermanic" a language, there can't be a family named "Kermanic". So would it be better, in your view, to remove the family code (so ira-ker "Kermanic" can be a language code), or to try to give the family and language different names? What names? Naming one "Kermanian" could work although the name seems little attested; naming one "Kermanic Iranian" or "Kermanic Central Iranian" would seem somewhat more confusing. - -sche (discuss) 23:56, 12 January 2018 (UTC)
Actually, we do have Aramaic language and Aramaic languages. Module:data consistency check doesn't issue a warning for this. But it seems like a bad idea. — Eru·tuon 01:02, 13 January 2018 (UTC)
A full list of canonical names used by both languages and families here. — Eru·tuon 01:27, 13 January 2018 (UTC)
Hm! Apparently it's not as problematic as I thought. It introduces ambiguity when an etymology says "from X", and WT:Families has said since its first drafting eight years ago "the name chosen for a family should not conflict with the name of a language" (hence Japanese language, Japonic family). But if it's not actually breaking anything, I guess we can have both a Kermanic family and a Kermanic language. ira-ker for the family, ira-krm (not added yet) for the language? - -sche (discuss) 03:51, 13 January 2018 (UTC)
@-sche: Cool! Would you mind using ira-krm for the family and ira-ker for the language? --Victar (talk) 18:06, 13 January 2018 (UTC)
Yes check.svg Done. - -sche (discuss) 05:14, 14 January 2018 (UTC)
@-sche, can you have Kermanic point to Kermanic languages instead? Can you please also set the script to fa-Arab (it's specifically nyq-Aran)? Thanks. --Victar (talk) 19:02, 7 February 2018 (UTC)
It also dawns on me that these need to be moved under ira-ker: nyq, vaf, atn, kfm, ntz, soj, gzi, gbz. And if you can set fa-Arab as the script of those as well. Thank you. --Victar (talk) 21:28, 7 February 2018 (UTC)
OK, I've updated the Wikipedia article link for ira-ker, and the script code. By "moved under ira-ker", can you clarify if you mean that they should have ira-ker as an ancestor, and/or that they should have ira-krm as their family code? Or that those codes should be removed, "subsuming"/"merging" them into ira-ker? - -sche (discuss) 02:20, 8 February 2018 (UTC)
@-sche:, thanks, and good questions. I meant under the family code. I do wonder if those codes should be done away with and subsitued with ker-* etymology codes. I'm thinking about he future, when I have Kermanic entries where I treat each dialect as I would Ancient Greek variants. What do you think? --Victar (talk) 02:56, 8 February 2018 (UTC)
What do you think about this entry, بز? --Victar (talk) 05:10, 8 February 2018 (UTC)

PIr proto languages[edit]

@-sche, there are also several proto language codes we could use:

Speaking of Yaghnobi, we're spelling it Yagnobi, which is super rare. It should be renamed to Yaghnobi. --Victar (talk) 00:22, 6 January 2018 (UTC)

"Proto-Ossetic" might be best simply identified with Alanic (which also has a number of attested words). "Pamir" is probably a language area and should therefore not have an associated proto-language. The others are in principle doable, but are there any usable sources on any of these? At least Novák and Cheung do not give any usable intermediate reconstructions. --Tropylium (talk) 21:49, 7 January 2018 (UTC)
@Tropylium: I think you're correct to equate Proto-Ossetic with Alanic, but I think the code for Alanic xln should be deprecated, and Alanic should be an alternative name for Proto-Ossetic, which is what it is commonly referred to and reconnectructed as. We also have one text we're calling Old Ossetic, but I'm not where that fits on the development stage -- perhaps the parent of Digor and Iron. What are your thoughts?
See my sources below for Proto-Pamir. It's seem to me to be a widely accepted genetic language family. --Victar (talk) 22:06, 7 January 2018 (UTC)
The thing with "Pamir" is, it was originally defined simply as "everything pre-Tajik spoken in the Pamirs", which is obviously not a subgroup (that includes things like Wakhi, which is rather a Saka descendant). Payne's chapter in Compendium… outright states: Postulation of a common stage linking these three groups into a hypothetical proto-Pamir unity distinct from other East Iranian languages seems however to be impossible.
The narrower MY-SI-ShY, then, is more of a leftover working hypothesis than anything that has been shown to be its own subgroup. Novák covers in another paper reviewing earlier classification schemes that Munji-Yidgha has some competing affinities towards Pashto and particularly Bactrian, and that in general Eastern Iranian as a whole is more of a dialect continuum as well. There has been at least one work attempting to reconstruct "Proto-Pamir", but it doesn't come out sufficiently distinct from Old Eastern Iranian to be its own thing entirely (and I gather there are problems of chronology — no reason to think that a few shared "Pamir" features would be older than the competing isoglosses shared with other Iranian languages). --Tropylium (talk) 02:54, 8 January 2018 (UTC)
Eastern Iranian has pretty much been debunked. Have you read Cathcart (2015)? It's an excellent paper. The position of Munji-Yidgha is an bit unsure, but I believe Shughni-Yazghulami and Sanglechi-Ishkashimi are still thought to be related. Either way, I'm fine without that one for now. --Victar (talk) 04:24, 8 January 2018 (UTC)
@AryamanA, JohnC5, Vahagn Petrosyan, माधवपंडित, Calak, any opinions on or additions to the recommendations above? --Victar (talk) 00:18, 8 January 2018 (UTC)
@Victar: Looks like the whole gang's gonna be here :) I prefer Proto-Pashto since Pathan is an exonym. Proto-Saka seems to also be called Proto-Scythian. For everything else, I have no opinion (yet) or I think it's good. The codes are definitely necessary IMO. —AryamanA (मुझसे बात करेंयोगदान) 00:23, 8 January 2018 (UTC)
Thanks, @AryamanA. Haha, yeah, thought I'd just tag everyone. I think Novák's thought behind Proto-Pathan is that Proto-Pashto represents Pashto and its various dialects, and Proto-Pathan is the parent of Pashto and Waneci (previously thought to be a dialect, but is now considered a separate language). I'm on the fence though.
Scythian/Scythic gets thrown around a lot and is even used for Altaic, so it might be better to avoid it. Proto-Saka is actually sometimes called Proto-Scytho-Khotanese, and Proto-Scythian/Scythic is a hypothetical ancestor of it and Proto-Ossetic/Alanic, but I'm a bit wary of that grouping. --Victar (talk) 00:55, 8 January 2018 (UTC)
@Victar: Cool! I kinda like Proto-Pathan, sounds delightfully archaic and also like you said it represents the identity of the language better as the ancestor of not just Pashto but also Waneci. -- माधवपंडित (talk) 01:16, 8 January 2018 (UTC)
I'm always in favor of more fine-grained family branching... —*i̯óh₁nC[5] 02:54, 8 January 2018 (UTC)
I've added a family code for Sakan (ira-sak) and proto-language code for Proto-Saka(n) (ira-sak-pro). (My understanding is that whenever we have a proto-language, we recognize / have a code for the family of which it is the proto-language.) - -sche (discuss) 23:36, 12 January 2018 (UTC)
@-sche: Thanks and perfect! Looking forward to adding some entries! --Victar (talk) 02:05, 13 January 2018 (UTC)
I've added "Proto-Sogdian" (ira-sog-pro) as such, because that name seems noticeably more common, although I appreciate that "Proto-Sogdic" might make clearer that it is ancestral to more than just Sogdian. If there is an issue with this, let me know. I thought I recalled the spelling of Yag(h)nobi being discussed somewhere before, but I can't find where; in any case, the spelling with 'h' does indeed seem to be more common (since the 1960s), so I'll rename it in a while (because that will entail updating a few hundred entries and categories). - -sche (discuss) 00:36, 17 January 2018 (UTC)
Thanks, @-sche, but Proto-Sogdian and Proto-Sogdic and two different proto languages, one being the ancestor of all Sogdian lects, and the other being the ancestor of Proto-Sogdian and Proto-Sogdian and Proto-Yaghnobi. Please see this clipping. -Victar (talk) 05:03, 17 January 2018 (UTC)
OK, I updated the "sog" parts of the codes to "sgc" and made them codes for (Proto-)Sogdic instead. Proto-Sogdic seems to be mentioned considerably less often than Proto-Sogdian, at least in the sorts of places Google has indexed. - -sche (discuss) 23:19, 20 January 2018 (UTC)


@Victar, Tropylium: xln is apparently attested(?), so I don't think it makes sense to deprecate it, even if we want to rename it. If it is or can be treated as the same language as Proto-Ossetic / Proto-Ossetian, it seems like we should handle it like Latin and like Proto-Norse, where the directly-attested words have mainspace entries and reconstructions have Reconstruction: entries, all using the same code. Regarding the other things: we should avoid splitting very similar proto-languages (compare the flattening we undertook recently of how many Austronesian proto-languages we included), but if these proto-languages are distinct, i.e. we won't just have etymologies with long strings of homographs like "from Proto-Shughni-Roshani *foobar, from Proto-Shughni-Yazghulami *foobar, from Proto-Iranian *foobar", and if there are resources reconstructing words in them, I see no problem with adding them, except Proto-Ossetic (redundant to xln and/or oos?) and, based on the discussion above, Proto-Pamir (apparently areal). - -sche (discuss) 19:09, 11 January 2018 (UTC)
@-sche: Alanic xln is not attested. All we have are borrowing and some personal names. Ossetian os is it's only descendant, with its three dialects, which is why it only makes to call its proto form Proto-Ossetic, but if you want to keep the family name Alanic, that's fine with me. We do also have some fragments what has been called "Old Ossetic" but scholars are still undecided if this is actually attest Proto-Ossetic, or if this is a dialectical form. Either way, yes, it should probably be handled like Proto-Norse, with attested and reconstructed entries. --Victar (talk) 20:14, 11 January 2018 (UTC)
-sche, I'm not asking for these codes frivolously. I have painstakingly sourced each one, in which you can find actual and reconstructions, all distinct from one other, ex. Proto-Iranian *gawarĉas > Proto-Shughni-Yazghulami: *ɣæwærs > Proto-Shughni-Roshani: *ǰöwörs. --Victar (talk) 20:14, 11 January 2018 (UTC)
I see and appreciate your sourcing for each one; I'm sorry if my comment seemed to doubt the distinctness of the languages; it was intended mainly to hedge/caveat my own unfamiliarity with them and with whether or not they were overly fine-grained.
John Tzetzes is said to have preserved some words of Alanian, which (unless we prefer to consider the words to belong to some other language instead) would mean that the language is attested.
- -sche (discuss) 20:53, 11 January 2018 (UTC)
@-sche: Regardless, I would appreciate it if you create code for me to under the name Proto-Ossetic for entering reconstructions and attested forms alike. Using Alanic xln would be confusing. ira-oss-pro would work just fine. --Victar (talk) 21:27, 11 January 2018 (UTC)
FYI, Old Sogdian is limited to the text (12 words) of the Ashem Vohu, so yes, we still need that code, and theoretically can also be reconstructed. We can also reconstruct a stage between Old Sogdian and Proto-Sogdic called Proto-Sogdian. --Victar (talk) 05:20, 12 January 2018 (UTC)
A thing with Alanian/Alanic is, Scythian xsc is just as not directly attested (only mentions by Greek authors), and yet we keep it around as a regular language. Should we depreciate that, too (maybe make it etymology-only? we don't seem to have any lemmas for it around).
BTW if we switch to the usual proto-language framework, we should probably also clean up and delete the etymology-only "Old Iranian" and "Middle Iranian", right? --Tropylium (talk) 10:14, 12 January 2018 (UTC)
@Tropylium: Yep, Scythian is thought of as the ancestor of Alanic/Proto-Ossetic, Sauromatian/Sarmatian, and possibly all Northeastern Iranian languages, like Saka and Sogdian. So yeah, we're not going to see anything that isn't reconstructed from it. I support getting rid of Old and Middle Iranian as well. --Victar (talk) 15:44, 12 January 2018 (UTC)


  1. 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Novák, Ľubomír (2013) Problem of Archaism and Innovation in the Eastern Iranian Languages (PhD dissertation)[1], Prague: Univerzita Karlova v Praze, filozofická fakulta
  2. ^ Kim, Ronald I. (2007), “Two problems of Ossetic nominal morphology”, in Journal of Indo-European Studies and Historical Linguistics[2], volume 112, DOI:10.1515/9783110192858.1.47, ISSN 0019-7262
  3. ^ Cheung, Johnny (2007) Etymological Dictionary of the Iranian Verb (Leiden Indo-European Etymological Dictionary Series; 2), Leiden, Boston: Brill, ISBN 978-90-04-15496-4
  4. 4.0 4.1 4.2 Schmitt, Rüdiger (1989) Compendium Linguarum Iranicarum[3], Wiesbaden: Reichert Verlag, ISBN 978-3882264135
  5. ^ Windfuhr, Gernot (2010) The Iranian Languages (Routledge Language Family Series), Oxon, New York: Routledge, ISBN 978-0700711314
  6. 6.0 6.1 Edelman, D. I. (1980) History of the consonant systems of the North-Pamir Languages, in Indo-Iranian Journal
  7. ^ Parpola, Asko (2015) The Roots of Hinduism: The Early Aryans and the Indus Civilization
  8. ^ Tremblay, Xavier (2005). “Bildeten die iranischen Sprachen ursprünglich eine genetische Familie oder einen Sprachbund innerhalb des indo-iranischen Zweiges? Beiträge zur vergleichenden Grammatik der iranischen Sprachen V.”


@-sche Balochi has ISO codes for its dialects, but all of them aren't active: bgp, bgn, bcc, ktl. I'm cool with that, but I'm wondering if I should create some etym-only codes: bal-eas, bal-wes, bal-sou, bal-kor. --Victar (talk) 19:04, 13 January 2018 (UTC)

When varieties have been assigned codes by the ISO, I think the usual/preferred practice is to use those codes even if we make them etymology-only, like "tmr" and "frc". So, I've just used the ISO codes. If you prefer the other names, they could be added as aliases (m["bal-sou"] = m["bcc"], etc) so that you could use them instead. (For anyone interested, the discussion which led to bgp, bgn and bcc being merged is archived in this WT:LT thread. As always, if you or anyone else has evidence that this decision should be reversed, bring it up.) - -sche (discuss) 00:17, 14 January 2018 (UTC)
@-sche: Cool, thanks. I think I'll create some aliases because those codes are whack. --Victar (talk) 00:29, 14 January 2018 (UTC)
@-sche: I... added the alias, but {{desc|bal-eas}} isn't working. --Victar (talk) 00:55, 14 January 2018 (UTC)
Hmm, and {{desc|bcc|foobar}} and {{desc|ira-mid|foobar}} also don't work, but e.g. {{desc|de-AT|foobar}} and {{desc|frc|foobar}} do work. I'm not sure what's going wrong. To the Grease Pit? - -sche (discuss) 05:43, 14 January 2018 (UTC)
Figured it out; the parent has to be a language, in this case bal. Southern Balochi: foobar and Eastern Balochi: foobar work now. - -sche (discuss) 05:51, 14 January 2018 (UTC)
@-sche: Thanks!--Victar (talk) 14:27, 14 January 2018 (UTC)

Revert at Reconstruction:Proto-Indo-European/(s)poH(y)-[edit]

@Rua reverted me attempting to incorporate a later etymology from De Vaan and Lubotsky regarding this root (see page 115) cited in Derksen's Baltic Etymological Dictionary particularly because I changed the page's name (I'll stand corrected on that) while doing so. She then told me to discuss it here. mellohi! (僕の乖離) 19:21, 5 January 2018 (UTC)

My objection is how the descendants fit the form you moved it to. Neither Germanic nor Latin can come from it, and Balto-Slavic and Sanskrit look off too. Moreover, the entry is given as a root, whereas you moved it to a name that isn't a root. If we are to change the entry to a fully-formed noun then one of the Latin descendants has to go, as one word can't have two different descendants. —Rua (mew) 19:26, 5 January 2018 (UTC)
So can I add De Vaan and Lubotsky's later claim into the etymology section itself then? mellohi! (僕の乖離) 19:36, 5 January 2018 (UTC)
What is their claim, exactly? —Rua (mew) 20:05, 5 January 2018 (UTC)

@Rua Several claims.

  1. That the laryngeal was h₁.
  2. That this root was derived from a y-extended speh₁- "to be full to the brim".
  3. That the laryngeal in the root metathesized "which is typical for i-presents and their derivatives" or from "paradigmatic levelling".
  4. "Because of the M in Germanic and Latin, it is attractive... to reconstruct *(s)ph₁oi-men-." (Me misinterpreting this probably motivated the page move in the first place, sigh...)

mellohi! (僕の乖離) 20:32, 5 January 2018 (UTC)

Claim 4 seems a bit dubious to me. You would expect an n-stem in Germanic and probably also Latin, but there's no sign of that. Compare *uhsô, which survives as an n-stem in Germanic. As a general rule, Germanic tends to create n-stems out of other stems, not eliminate existing ones. In fact, none of the descendants preserve any kind of mn-suffix. Some have only m, some only n, but none have both. —Rua (mew) 22:39, 5 January 2018 (UTC)


The article claims there is a scholarly consensus, whereas dictionaries and online sources seem to show the opposite of consensus. Is this a good summary? --Espoo (talk) 18:32, 6 January 2018 (UTC)

Which article? —Rua (mew) 18:42, 6 January 2018 (UTC)
pontifex#Etymology_2 --Espoo (talk) 19:02, 6 January 2018 (UTC)
I don't see any links to an external article there. Do you mean the link in the references section? —Rua (mew) 19:04, 6 January 2018 (UTC)

New attempt: The etymology section of pontifex#Etymology_2 claims there is a scholarly consensus, whereas dictionaries and other sources seem to show the opposite of consensus. Is this a good summary? --Espoo (talk) 19:11, 6 January 2018 (UTC)

  • See “pontifex” in Douglas Harper, Online Etymology Dictionary, 2001–2018., a rather reliable source. DCDuring (talk) 19:22, 6 January 2018 (UTC)
Does that mean you disagree with the more recent explanations provided at the link above that reject those? --Espoo (talk) 19:32, 6 January 2018 (UTC)
No. It means that Online Etymological Dictionary has adjusted the entry from what it was when your source accessed it. It is hard to say that there is a scholarly consensus on anything other than the inconclusiveness of the various speculations. DCDuring (talk) 20:12, 6 January 2018 (UTC)

*kъnjiga again[edit]

So it seems that the supposed Proto-Turkic word is pretty much fictive. Chuvash кӗнеке (kĕneke) is from Russian, Uyghur [script needed] (küjn, свиток) is

  1. not found three online Uyghur dictionaries
  2. nor Schwarz nor Haji Qutluq Qadiri
  3. not found in Clauson nor Caferoğlu (assuming Old Uyghur)
  4. nor Kashghari, nor Dictionnaire Turc Oriental, nor Şeyh Süleyman, nor Budagov (assuming any Turkic language located in Tarim)

even if it existed, I don't think it would be enough to reconstruct Proto-Turkic. There's also Chuvash кунӗ (kunĕ)/коньӑ (konʹă, (rare) a special kind of weaving), but it doesn't look too promising.

Mikkola (according to Skok) thinks it's Iranian in origin, citing Ossetian кинугӕ (kinugæ)/чиныг (ḱinyg) and Armenian կնիք (knikʿ), but Abaev claims that Ossetian is also from (early) Russian with vowel metathesis. @Vahagn Petrosyan, could the Armenian be from Iranic rather than Hurrian?

The only tangible words here are Hungarian könyv and Erzya конёв (konjov), @Tropylium, they look so similar, could they really have independently developed from something like proposed *künig? Did Mordvinic have contact with Hungarian?

Crom daba (talk) 19:28, 6 January 2018 (UTC)

@Crom daba, an Iranian mediation has been suggested for Armenian on phonetic grounds. See the source I added to կնիք (knikʿ). --Vahag (talk) 21:26, 6 January 2018 (UTC)
A correctly timed Turkic source looking like *künig would work in principle for both Erzya and Hungarian. However getting /o-ó/ (and not /ú-e/ or /ú-i/) in Erzya, with final stress, would seem to require loaning from Tatar specifically, not some kind of a Wanderwort from early Chuvash or directly from Proto-Turkic. (Stressed noninitial /o/ firmly rules out native origin, too.) To my knowledge there has not been any direct Mordvinic-Hungarian contact either, the furthest northwest that Hungarian has been is contacts with Permic. --Tropylium (talk) 21:44, 7 January 2018 (UTC)
Some further snooping: A magyar nyelv történeti-etimológiai szótára (1967, ed. Loránd), (which we don't seem to have a template for yet) attributes the supposed Uyghur word to "Gab." — any idea who or what this might be? --Tropylium (talk) 01:11, 8 January 2018 (UTC)
Thanks, it's Gabain. She writes "küin (lies kuin ?) < chin. küan Buchrolle || tomar kitap". Since it's not found in Clauson, I'm guessing this is from Middle Mongol, perhaps written in Mongol script. Might as well be a more recent Chinese loan, certainly not enough material to reconstruct Proto-Turkic *küiniŋ.
How about Iranic? Any front rounded vowels in Hungarian Iranisms? Lowered /u/ in Erzya? Crom daba (talk) 14:05, 8 January 2018 (UTC)
Given the Ossetic forms, Old Hungarian *kińVw and later labiality assimilation > *küńüw > könyű (> könyv) would also work. TESz actually has some old attestations as kenw, kenyew (but /ö/ may have been spelled as e; these are 200 years younger than the first attestation: qunwes for modern könyves, already in the 13th century).
Erzya seems like a bigger issue in any case. Final stressed /o/ is a fairly new feature, pointing to origin sometime in the 1000 to 1500 CE range. Within those constraints Old East Slavic would work better than anything Iranian — but in that case we'd still probably expect **/końija/ instead (and also the meaning 'book' and not 'paper').
I've updated the entry accordingly, thanks for your input guys. Crom daba (talk) 05:29, 9 January 2018 (UTC)
For completeness: it seems that the idea that the Uralic words stem from Turkic is from Räsänen, 1938, "Nochmals über ung. könyv 'buch' und mord. kon'ov 'papier'" (Finnisch-Ugrische Forschungen XXVI) — he acknowledges that the Uyghur is a Chinese loanword, but argues that Chuvash кунӗ (kunĕ) is regardless a regular cognate, hence they'd go back to *küinig. --Tropylium (talk) 15:57, 11 January 2018 (UTC)

πρεσβύτερος a semantic loan?[edit]

@Palaestrator verborum has posted sources that say that the ecclesiastical term πρεσβύτερος (presbúteros, priest, elder, presbyter) is a semantic loan from Classical Syriac ܩܫܝܫܐ (qaššīšā). How does this work? The term appears in the New Testament, and the New Testament was translated from Koine Greek to Syriac, not the other way around. Or is there a theory that the New Testament works that feature this term were originally written in Syriac or Aramaic, translated to Koine Greek, and then the original Aramaic or Syriac was lost? Or that these works were written after the office of Christian elder or priest or presbyter first existed? — Eru·tuon 22:19, 6 January 2018 (UTC)

It does not matter in which languages the New Testament was written: Christianity of course predates the New Testament canon, there were various scriptures about deeds of Jesus around, and the semantic loan does not need to be a learned borrowing. Jesus has supposedly spoken Aramaic, remember? Even if Jesus did not exist, the communities appeared in Aramaic-speaking Palestine and Christianity spread from there. It’s of course a modern believer’s view that Christianity is the New Testament but this has of course also a date and is not passed from Jesus himself – well, needless to say, not even Christians claim this. Palaestrator verborum sis loquier 🗣 22:28, 6 January 2018 (UTC)
Yes, it is possible that it's a semantic loan from the spoken Aramaic of the time into written Koine, but I don't know how to go about proving it. (While we're at it, I suppose senses 4–8 of elder#Etymology 1 are themselves a semantic loan from πρεσβύτερος (presbúteros). As for the NT being originally written in Aramaic, there is such a theory: see w:Aramaic_New_Testament#Minority_view_-_Aramaic_original_New_Testament_hypothesis. It isn't very widely held, though. —Mahāgaja (formerly Angr) · talk 22:36, 6 January 2018 (UTC)
Yes, we are using {{semantic loan}} way to little. I have sourced this semantic loan from Aramaic now well, and for the record I count from the Category:Semantic loans by language: This is the first semantic loan at all categorized in Wiktionary for Ancient Greek, while English has merely five words categorized, Latin seven. For Arabic I have by now collected 22 single-handedly.
Theoretically semantic loans are more often, it’s just not regularly distinguishable from independent formations, and people are unlikely to consider them, therefore the bad terminology outside of Wiktionary which had lead me to use {{calque}} for the here discussed entry instead of {{semantic loan}}, for which I must say sorry. Palaestrator verborum sis loquier 🗣 23:01, 6 January 2018 (UTC)
Some nitpicking: this sense of the word does not appear in the scriptures describing the life of Jesus, at least the canonical ones. According to Strong's Concordance, it appears in Acts and in 1 Timothy, Titus, 2 John, 3 John, 1 Peter, James. The gospels do have the related sense of an elder in the Sanhedrin (which, incidentally, could also have been a semantic loan from Aramaic).
After further thought I agree that the theory is likely, but like @Angr, I don't know if it can be proved, unless there are early Aramaic or Syriac works that use this term. My other concern is how to explain it so that it does not unnecessarily puzzle readers. The mention of Classical Syriac puzzled me. I don't know the exact time-frame or geography of Classical Syriac as it is defined on Wiktionary, but it's a literary language and we don't really know how different the spoken language used at the very beginning of Christianity would have been. It might be more accurate to call it a semantic loan from an unknown Aramaic term probably cognate to the Classical Syriac term. — Eru·tuon 23:19, 6 January 2018 (UTC)
@Erutuon: I don't get a ping if you use my old name. —Mahāgaja (formerly Angr) · talk 15:51, 7 January 2018 (UTC)
@Mahagaja: Whoops! It was reflexive, I guess. I'll try to remember next time. — Eru·tuon 21:47, 7 January 2018 (UTC)
Another obstacle: it's trivially easy to find זִקְנֵי יִשְׂרָאֵֽל translated as οἱ πρεσβύτεροι Ισραηλ in the Septuagint (e.g. in Numbers 11:30). Granted that Aramaic represents a myriad of languages, dialects and historical stages, and that the writers of the Septuagint no doubt had some knowledge of at least one of those, but the semantic overlap between זקן and πρεσβύτερος (presbúteros) is enough that Occam's Razor would suggest direct translation from Hebrew to Greek to be a better explanation. Chuck Entz (talk) 01:09, 7 January 2018 (UTC)


Does anyone know where this got its s from? It's not in any of its ancestors. —Rua (mew) 22:38, 7 January 2018 (UTC)

fr.wikt says it's an orthographic device used to note a long vowel. --Per utramque cavernam (talk) 22:49, 7 January 2018 (UTC)
If that can be referenced, let's add it (templatize it if the same thing is applicable to many words), like Togolese explains the "l". - -sche (discuss) 23:47, 8 January 2018 (UTC)


Does anyone (esp. @Mnemosientje) know where the Dutch term pino comes from? My first guess would be that the term originates from Pino ("Big Bird" from Sesame Street), that then got a cheeky slang reinterpretation, as that explains why this term would be common gender rather than neuter like object. But I'm far from sure about that. I think pino can also be used as an informal, irreverent substitute for any person ("tering, een of andere pino heb me brommerd gestolen") and that could be an intermediate stage. Lingo Bingo Dingo (talk) 13:15, 8 January 2018 (UTC)

Your guess is as good as mine; I can't really think of another evolution of its meaning that would make more sense than Sesamstraatvogel -> gozer -> neukobject backronym. I'm afraid there won't be any reliable sources to back up our guesses in any case. — Mnemosientje (t · c) 13:55, 8 January 2018 (UTC)

Norman sound changes[edit]

Since I'm not an expert on Norman, I wanted to see if anyone knows more about this. Are all Norman dialects supposed to be descended from Old Northern French, and hence initial Latin -c- remains hard as opposed to the change that the rest of Old French undergoes into -ch-? Or does it vary? So in the case of ch'na, could this be inherited, or was it taken from a different dialect of Old French rather than its normal parent one (the one that gave "regular" French chenal?) Where would the form canal fit in then; is that a normal variation that can occur in Norman dialects or was it more likely a learned form like the French counterpart? The -al endings indicate some alteration, at least. Also, where does our Old French entry canel fit in? The form canal was definitely a borrowing, but I'm not sure about this one, which seems like it could be a northern variant of chanel? Word dewd544 (talk) 01:06, 9 January 2018 (UTC)

Announcement: New templates {{rebracketing}} and {{univerbation}}[edit]

Added these etymology templates today as the categorization structure was already present. Now we can collect these efficiently too. Formerly one added appendix links and categorized with {{catlangname}} manually. @Erutuon, JohnC5, Vorziblix, Mahagaja, Vahagn Petrosyan Palaestrator verborum sis loquier 🗣 13:12, 9 January 2018 (UTC)

etymology of "bampot"[edit]

bampot Posted by Grant Barrett on August 16, 2005 · 1 Comment bampot n. a crazy person; a fool or dolt. Etymological Note: Most likely a form of barmpot. According to OED, barm, “the froth that forms on top of fermenting malt liquors; the head of a beer,” is used attributively as a formative to indicate a crazy or feeble-minded person or idea. This is, obviously, related to barmy or balmy ‘crazy.’ Thanks to Michael Quinion for the tip on bampot‘s etymology. Probably not related to the Irish Gaelic bambairne ‘dolt, stupid person, lout.’ In “Some Modern Irish loanwords describing people” (Celtica vol. 18, p. 53, 1986, School of Celtic Studies, Dublin) Mícheál Ó Siadhail connects bambairne to the old Spanish slang bambarria, which, according to the Velasquez Spanish and English Dictionary (1985, New Win Publishing) means ‘a fool; an idiot.’ Bambarria is glossed as “blockhead” in Carnoy, Albert. “Apophony and Rhyme Words in Vulgar Latin Onomatopoeias. American Journ. of Philology. vol. 38, no. 3. (1917) 271. (source: Double-Tongued Dictionary)

if need be[edit]

I've added a tentative etymology (need is the noun, be the subjunctive), but the variant if needs be makes me wonder if that's correct. needs could be the plural, I guess? --Per utramque cavernam (talk) 22:18, 10 January 2018 (UTC)

FWIW, I always interpreted needs in if needs be as the plural. (Native English speaker, grew up near Washington, DC.) ‑‑ Eiríkr Útlendi │Tala við mig 22:31, 10 January 2018 (UTC)
The OED mentions if need be and similar phrases in need, n. 1, but doesn't have a quotation for if needs (be, require, were). So perhaps that phrasing is just a random alteration, because the grammar of the phrase has been forgotten. — Eru·tuon 22:45, 10 January 2018 (UTC)
This StackExchange thread agrees with your analysis. "If needs be" seems like it can be analysed identically: like "if need be" = "if need exists", so "if needs be" = "if needs exist". - -sche (discuss) 23:57, 10 January 2018 (UTC)


Why didn't this become *steffather? Confer Dutch, Low German, Swedish, Danish and so on. I see that in Middle English, there was the form stef- for step-. Why didn't this become the dominant form?

Was it because, as was the case with Icelandic, English was somewhat isolated from other Germanic languages? Tharthan (talk) 23:14, 10 January 2018 (UTC)

It seems like there just wasn't enough pressure to change away from the p. Maybe the other step- words and step provided enough reinforcement / pull in the p direction to counter whatever pull father was exerting in the f direction. - -sche (discuss) 03:29, 11 January 2018 (UTC)
In Middle English we do see forms with f (e.g. steffader, stifader), but they did not prevail over those with p. Analogy with other terms (stepbrother, stepsuster, et al.) may have played a role in the preservation of step- Leasnam (talk) 03:32, 11 January 2018 (UTC)


The etymology of this Nanzhao word for "tiger" is quite unclear. Every source I look at gives a different explanation. Here are my findings:

  • 张锡禄 (1990), 南诏国王蒙氏与白族古代姓名制度研究:
  • 陆家瑞 (1996), 南诏膘信与清平官赵叔达唱和诗试析:
  • 吴裕成 (2010?), 说“寅为虎”:屈原以生于“三寅”为自豪:
    《通雅》还记“南诏谓虎为波罗,蛮人呼虎为罗罗”,连同汉代《方言》所记虎称“李父”、“李耳”、 “伯都”等,有关虎的这些称谓中,或许保留着古代崇虎的风习。
  • 唐善纯 (?), 语言学视野里的大东亚文化圈:
  • Relevant entries in 白汉词典 (1996, which uses the old orthography):
    • 【jinl bol lo】 公虎。
    • 【jinl mox lo】 母虎。
    • 【jinl】 金。
    • 【jinl】 筋。
    • 【jinl】 关。
    • 【jinl】 粘,贴。J-zorx zix. 用胶粘鸟。
    • 【jinl】 斤。
    • 【jinl】 尖。
    • 【jinl】 熊。
    • 【bol】 表示阳性:Deirt bol. 公猪。用在人名后兼表尊敬、亲昵之意:At zirl ~阿志哥;Guanl~. 官儿;~ Xil gul yil ~. 新郎官;Gux~. 老头,老大爷。
    • 【lo】 老虎,寅虎。 (1993 orthography lod)
    • 【banrt】 豹子。 (1993 orthography banp)
  • Relevant entries in the Nuosu Yi-Chinese-English Glossary:
    • ꆿ₂ lat [la⁵⁵] n. 老虎 tiger
    • ꀞ₁ bat [pa⁵⁵] adj. 公;雄 male (of a given species)

What should we make of all this? — justin(r)leung (t...) | c=› } 08:51, 12 January 2018 (UTC)

@Wyang, Zcreator, any thoughts? — justin(r)leung (t...) | c=› } 14:39, 12 January 2018 (UTC)
@Justinrleung, Zcreator There are several issues affecting the methodology of the publications listed above (as well as much of the old-style Chinese linguistics literature). The most prominent is that the claim of etymology from other languages is not accompanied by a careful study of the lexicon of the cited language. Besides, there is also anachronism in the research ― none of these modern languages existed at the time of Nanzhao, which probably spoke various, more immediate descendants of Proto-Lolo-Burmese, as well as some kind of Proto-, or Old Bai. I think this may be the reason for disagreement amongst these sources, and for us something like "From the language of the ancient kingdom of Nanzhao, likely derived ultimately from PST *k-la ("tiger"). Compare ..." may be enough. Wyang (talk) 15:28, 12 January 2018 (UTC)
@Wyang: Thanks! I've updated the etymology. — justin(r)leung (t...) | c=› } 15:43, 12 January 2018 (UTC)
Concerning the translation of the examples, it seems rather metaphoric and tangential to another translation. I was digging into a claim that the "three stars", a name for the "belt of Orion" were also called the three generals, but came up empty which should be no surprise as I have no clue about Chinese. Anyhow I still associate star and warrior. Now, I read "star" in a place where it doesn't make much sense (except as a proper noun) and recognize that sign 星回 looking similar to 宿 (used to denote star constellation?) - not at all, I have to admit in hindsight, but the final form 青 of 星is a bit closer and the given form is looking human enough, while 宿 definitely includes the short form of the radicals 人 denoting a person. And it is transliterated similar to the minister in w:Xiangqi, which is xiang (versus xīnghuí)- again, to the untrained observer only. Now, I'm not blogging, if you think I was being annoying. I'm trying to suggest the quote uses metaphors and tiger here means war gear, so that the second quote in the lemma means war clothes of leather, not "tiger skin", as the second character of the lexem in question can also mean light clothes. The first one means wave, which I guess could be a metaphor for turmoil, ie. war. If that gear happened to be tiger leather, that would explain the origin of this reading as the war clothes.
This is surely still far from the truth. The idea is only that tiger and star are backfashioned metaphors. If this offends anyone in case I am completly wrong, my only defense is that Chinese is highly chaotic and any guess is as good as the next one. If anyone verifies this or takes me serious enough to falsify, I would be highly surprised, though. If I opened a new topic to inquire the etymology of the meaning celebrity for the star character (third meaning at ), that would be likely OK, I hope, so I post this here in the same spirit.Rhyminreason (talk) 01:44, 17 January 2018 (UTC)
@Rhyminreason: I'm afraid your connections are based on a series of unsupported claims, probably because you are unfamiliar with Chinese. It is interesting how you have managed to read into weird details to cobble up some sort of explanation in support of your "star-warrior connection". Yes, to your surprise, I will try to falsify your claims and present you with what the evidence points to. Chinese is not as chaotic as you think it is.
星回 refers to 星回節 (given in the title of the poem), which, according to 太平廣記, was a festival in Nanzhao on the 16th day of the 12th month. I have no idea why you're bringing 宿 and 青 into any consideration here; the fact that 星 or 青 looks like a human (which really doesn't mean much) has nothing to do with the 亻 radical appearing in 宿. Moreover, the similarity between xiàng (MC sɨɐŋH) and xīng (MC seŋ) is simply coincidence. All the ancient sources I've looked at (including 太平廣記 and 蠻書) annotate 波羅 as meaning tiger in the language of Nanzhao. 波羅 is no doubt a transcription of a foreign word, which is usually not analysable by the individual characters.
Now about the "celebrity" sense of 星, it's simply an extension of "star", just like in English, "star" can refer to a celebrity (e.g. pop star; movie star). — justin(r)leung (t...) | c=› } 02:48, 17 January 2018 (UTC)
What do you make of that poem, though? It is certainly not to be taken literall (with the tiger and wild horses in mind). The ancient sources would need a date. The ones named are second hand sources written, or at least compiled, by outsiders. So there is room for doubt. Renaming festivals and idols is a common theme in conquest and I suppose, indeed without credible proof, this might lead to new readings of existing symbols and cryptic subtext. A pure metaphor would actually rely on a preexisting meaning, I have to admit, and If you deem the context certain, that is all I was asking for. As the etymon is not certain, according to the article, I hope questioning it doesn't overstay my welcome, yet.
Regarding the star general I should open a new topic. Rhyminreason (talk) 16:30, 17 January 2018 (UTC)
@Rhyminreason: I'm no expert in literature, so I can't be sure of what the poem is actually referring to. About the tiger and the wild horse, 太平廣記 says "驃信昔年幸此,魯射野馬並虎", which seems to be referring to an incident where the king shot a wild horse and a tiger. FWIW, 陆家瑞's 南诏膘信与清平官赵叔达唱和诗试析 does analyse the poem in a somewhat unconventional way, and he does connect 波羅 to warriors wearing tiger skins (coincidence?), but I for the most part feel quite sceptical of the interpretations given in the article. — justin(r)leung (t...) | c=› } 16:42, 17 January 2018 (UTC)


@Eirikr The likelihood of a theory is not for some random anonymous person on the Internet to decide and dictate as worthy of inclusion. If it doesn't come from a reliable source, it constitutes original research, and therefore is worth questioning and if necessary, subjected to complete removal. I've done my share of adding theories on this site, and if people find them to be dubious eyesores, I won't object their removal. So I hope you won't either, because your strenuous insistence on maintaining your own theories without providing any sources and on rejecting any opposition is starting to be very irritating. That kind of arrogance is only acceptable if you could verify your actual credentials with verifiable documents, not just some half-hearted "proficiency" tags. ばかFumikotalk 20:44, 12 January 2018 (UTC)

An RFV is fine. Removing content you personally disagree with, with insufficient research and no regard for usability, is not so good. As you describe yourself, you don't read Japanese. How am I to judge your abilities to understand Japanese source materials as anything but non-existent? Are you lying when you say you don't read Japanese? Or are you incompetent (i.e. "lacking competency") in your editing of Japanese entries? I really don't know how to interpret your actions. Your deliberate rudeness and abusiveness doesn't help. I'll leave the judgement of arrogance to others. ‑‑ Eiríkr Útlendi │Tala við mig 22:24, 13 January 2018 (UTC)
"As you describe yourself, you don't read Japanese. How am I to judge your abilities to understand Japanese source materials as anything but non-existent? Are you lying when you say you don't read Japanese? Or are you incompetent (i.e. "lacking competency") in your editing of Japanese entries?" That's the whole point. I don't know who you are either. Why should I believe anything you say while I have no idea who or how capable you are, apart from the blanket "competency" tags on your user page, again? It's not that I disagree with the theories you provide necessarily, but you keep pulling these gigantic claims seemingly out of nowhere, and you've continuously been reluctant on specifying what sources you used to deduce (original research, mind you, are not of any "usability" if it doesn't corroborate with verifiable sources). Who the heck are you? And who the heck are you to decide if something is of any "usability" when no one can verify any of your claims based on verifiable sources? Do you think you're God, and everyone has to take your claims as a matter of fact or something? Point to where you get your information. Give some links. It doesn't matter whether I understand the Japanese sources or not, because that doesn't freaking matter as far as the fact that you don't get to decide what is of usability all by yourself behind that username goes; for all anyone knows, you could be just a phony. As long as you don't do that, don't be annoyed if someone can't stand your utmost arrogance. ばかFumikotalk 11:31, 17 January 2018 (UTC)
I see an ad-hominem attack. That occludes the lofty goals of your inquiry.
The etymology as it were, wasn't very satisfying. I'd assume benefit of the doubt for the edit. In defense of the existing effort, I'd assume the links from the article to the different writing systems provide an obvious explanation. If somebody could vouch for this, that would probably help, but keep in mind prolific contributers do check new edits.Rhyminreason (talk) 18:34, 17 January 2018 (UTC)
Ad-hominem attack or whatever you think it is, when someone essentially anonymous keeps on writing whole paragraph-worth of text making big claims and refusing to give the sources to back those claims while deciding that they're valid by default because he's an admin, I feel like I'm entitled to express my frustration against power abuse. To be completely honest, I do have some feelings against this guy's character, but the only thing I care about here is whether he should be allowed to have free rein to do as he's pleased given his behavior described above. Every time I've read his etymologies, I always scratched my head and went "This is surely a lot of information! Where the heck does all this come from? Is it okay to just leave huge chunks of unverified claims around like this?" ばかFumikotalk 09:32, 18 January 2018 (UTC)
disclaimer: I'm not picking "sides".
If this is a recurrent feeling, I suggest you start keeping a list somewhere of all the etymologies you've found unconvincing and insufficiently sourced; at least you'll have a clear picture of what you're displeased with. Being embroiled in so many revert wars, and involved in conversations on so many talk pages is a very inefficient and irritating way of going about things. Reducing the number of "fronts" to one might help make the whole matter less frustrating to you.
Also, could you point to some conversations where Eirikr was obviously reluctant to share his sources? --Per utramque cavernam (talk) 11:11, 18 January 2018 (UTC)
Fumiko, I too would be interested in cases where you think I'm denying you sources. I am perfectly happy to admit that you and I have trouble communicating with each other. I have no recollection of refusing to provide you with sources. When you've asked, I've provided, even in cases where you seem to think the sources are inadequate. I myself only began adding in references just in the past two or three years, possibly at your insistence -- for which, thank you, as I do think that adding in references more consistently helps to improve the quality of our entries. However, I point out here that adding in references has not been common practice, and indeed most of our content has no references. Do you similarly object to the cheer entry? What about Duits, or 你好, or yáʼátʼééh, or голова (golova)? It is one thing to request references, and another to delete content just because it doesn't have a reference.
There are cases where I cannot find a source that clearly states the provenance of a term, such as 堕天使 (datenshi, fallen angel). In such cases, I will write out the etymology to show what can be backed up, and what is conjecture. For instance, from all I've read, 堕天使 is an alien concept in traditional Chinese and Japanese religious thinking, and all reference works that I've read that list 堕天使 specifically state that this is a Christian term (see three such sources listed at the Kotobank entry). Christianity was introduced to Japan by the Spanish and Portuguese, as described in numerous easily findable works and also at w:Christianity_in_Japan#History. It is reasonable to infer that this 堕天使 term may have arisen as a calque of either Portuguese anjo caído or Spanish ángel caído, as I've added to the etymology multiple times (with wording intended to clarify that this is inference and not established fact), and that you've removed without substantive discussion.
The core of your argument regarding etymologies, as best I can understand it by synthesizing the above with other comments and actions by you, appears to be that you only accept etymological content that has multiple references that you personally agree with, and will delete anything else. Is that your argument? ‑‑ Eiríkr Útlendi │Tala við mig 20:14, 18 January 2018 (UTC)

Etymology of Iroquois[edit]

Before I start an edit war with an anon, can someone provide any insight in this matter? If we look beyond the bad formatting, are the recent changes at least plausible? --Robbie SWE (talk) 19:04, 13 January 2018 (UTC)

@-sche might have knowledge about or interest in this matter. DCDuring (talk) 19:39, 13 January 2018 (UTC)
At a bare minimum, the spelling would strongly suggest a French presence somewhere in the chain. ‑‑ Eiríkr Útlendi │Tala við mig 22:10, 13 January 2018 (UTC)
Thanks for the ping. :)
The origin of this word has been a subject of speculation for a long time. General references (Merriam-Webster, Collins; Dictionary.com, OxfordDictionaries.com, EtymOnline, etc) give the agreed-upon details: it entered American English in the 1660s via French. (It entered French as early as 1603.) Those references also take the ultimate origin as probably (some more hedgingly say perhaps, others use no qualifier) Algonquian, although this is less certain.
I will update the entry later. - -sche (discuss) 22:40, 13 January 2018 (UTC)
Curious -- "Bakker argues that the French suffix -quois is added to many other tribal names by the French and hence may not represent anything". I note the surface similarity to English turquoise, deriving ultimately from Old French turquois (Turkish), where the suffix in question is actually -ois (-ish) added to a term ending in /k/. Any chance the Native American etymon was closer in form to /irok/? Or are we certain that the -quois or /-kwa/ ending was part of the etymon? ‑‑ Eiríkr Útlendi │Tala við mig 00:22, 14 January 2018 (UTC)
I don't know any way to be sure. Loewen asserts that "all ethnonyms ending in -quois can be traced to a Basque origin [...] the suffix -ko+a that attaches to place-names to mean 'those of'", but I am unconvinced that this is so broadly and breathlessly true. Loewen does compare Excomminquois to Les Escoumins, Quebec, and Canadaquois does seem to be exactly Canada (Laurentian kanata) + -quois. He and Bakker assert that the French Souriquois ("Mi'kmaqs") is from Basque zurikoa (those of the white) (zuri (white)), roughly translating the (or a cognate of the) autonym of the Abenaki with whom the M'kmaqs federated (a better translation of Wôbanaki is "people of the dawn land i.e. the east" but the root does also mean "white"). But that last tribe pokes a hole in the absolutism of his argument, since the French (and English) sometimes called them Abenaquois, and the /k/ is clearly part of the native name. (Other names with the element are "Armouchiquois" and "Charioquois".) - -sche (discuss) 01:56, 14 January 2018 (UTC)


@Vahagn Petrosyan HSB 2010 claims that this word comes from the same root as γλυκύς (glukús) and dulcis. Is this phonologically possible? mellohi! (僕の乖離) 22:54, 14 January 2018 (UTC)

@Hillcrest98, none of the proposed PIE roots are phonologically possible, so they have to assume a contamination. See the updates I made to the page. --Vahag (talk) 21:04, 15 January 2018 (UTC)

Not all is horse that hobbles: "Variants" of hepo[edit]

User @Liedes has been adding extensive coverage of Finnish 'horse' words recently. A lot of their etymological coverage is being bold to the point of what I would call wrongheaded, though. I would like to discuss a bit what the situation looks like from the viewpoint of established linguistic understanding.

  • hepo is the only truly native form (with exact equivalents in e.g. Veps hebo).
  • hevonen is the most widely used variant, though basically a loanword from southwestern Finnish in most dialects.
  • heppa is a well-established child-speak variant. It is however hard to tell if this should be considered a pseudo-derivative hep(o)+pa or perhaps he(vonen)+ppa. The latter seems better at least semantically, since it is based on the truly basic term for 'horse', not on a rarer poetic variant. (For similar derivation, compare e.g. peppu (butt) as pe(rse)+ppu, poppa (hot) as po(lttaa)+ppa, omppu (apple) as om(ena)+ppu.)
  • hopo seems to only appear in folk poetry. Suomen Murteiden Sanakirja (SMS) does not know of this word in the meaning 'horse'. Therefore it seems likely to be a purely poetic variant, created by alliteration, and so probably without any direct relation to Estonian hobu (but with the same phonetic motivation — see below). I find the specific glossing 'forest horse', 'stronger than hepo' to be unsubstantiated, but I'm willing to hear an argument at least.
    • hopoti in the sense 'forest horse' is an ad hoc further derivative that probably doesn't even fulfill our criteria for attestation. It's also phonetically impossible as a native word in most dialects: it seems to be derived by a suffix by which we would actually expect hopotti as the standard Finnish form. The interjection sense (also hopoti hoi) might be attested well enough, but I don't think it has anything to do with the marginally attested 'forest horse' word. Phonetic identity and vaguely horse-related semantics are not good enough, since the best option seems to be derivation from the interjection hop.
  • hoppa is another a rarer varient, but at least known in actual use from several dialects (and even in at least one dialect of Karelian [4]). This could be partly originally unrelated: an old but untenable loan etymology for hepo compares it with Danish hoppe from Old Norse hoppa, which however matches this Finnish variant perfectly. However, it is usually restricted to child-speak, which to me suggests that it is a variant of the widespread heppa, formed again by alliteration (and so independently of hopo).

Estonian has indeed the form hobu with an o, but this does not give support for reconstructing original *hopoi instead of *hepoi. Sound changes *e > *õ > o are well known in Estonian (other examples of this full chain include kollane, kord, otsima, onu versus Fi. keltainen, kerta, etsiä, eno), while the inverse development *e > **o in Northern Finnic (Finnish, Karelian-Livvi, Ingrian, Ludian, Veps) is not known. Same goes for the Votic variant opo: in Votic *e-o gives quite regularly *o-o, also e.g. toro (acorn) versus Fi. terho. Tiit-Rein Viitso explicitly proposes this as the explanation for hobu in his article Läänemeresoome esimese silbi õ ajalugu (though he suggests *e-o > *o-o directly, which to me seems incorrect). Compare also the almost total absense of **hoponen, **hovonen. The former is attested, per SMS, from one parish — it is probably best considered a secondary variant of heponen again by the same mechanisms as hepo > hopo.

hoputtaa (to rush) is fairly clearly not a part of the hepo cluster, even though it can be often used with a horse: it is derived from hoppu (rush), which has no especial connection with horses. I would suggest that the verb has become associated with the various hVp(p)V 'horse' words secondarily by phono-semantic matching. (Hoppu is, however, likely derived from the Scandinavian word family for 'to jump' etc, which is also where at least the Scandinavian hoppa group is derived from.)

Hessu Hopo (Goofy) has obviously nothing whatsoever to do with horses (contrary to the imaginative but purely speculative essay as of circa this edit). By the KISS principle, his last name is probably simply hopo (fool), a dialectal variant of more common hoopo. Compare also the "ö-grade" variant höpö (with some cognates even outside Finnish: Karelian höpö, Ingrian höpöi). On this cluster however, SMS notes that hopottaa [5] is used both for 'to blabber, nag' and 'to rush, urge' with no especially clear dividing line. This suggests that the cluster of "onomatopoetic" terms based on a "root" √hOp- (besides hopo and hopottaa also höpöttää, höpistä, höperö, höpsö and an abundance of dialectal terms such as hopeltaa, hopertaa, hopista, hoplo, hopakka) actually also ultimately originates through hoppu, explaining the phonetic similarity.

--Tropylium (talk) 22:09, 16 January 2018 (UTC)

TLFi abbreviations[edit]

I wasn't able to find an explanation of the abbreviations used by the TLFi e.g. http://www.cnrtl.fr/definition/moraine --Espoo (talk) 22:46, 16 January 2018 (UTC)

  • @Espoo If you mean for the biblio sources, the aide page takes you to here, but I didn't find e.g. BRUNHES.
    I didn't find Base Historique du Vocabulaire Français here either.
    Sobreira ►〓 (parlez) 11:14, 17 January 2018 (UTC)
I meant the abbreviations of everything else except the biblio. I wrote to contact@cnrtl.fr at the end of december but haven't received an answer. --Espoo (talk) 07:29, 18 January 2018 (UTC)
@Espoo You mean GÉOPHYS.; Prononc. et Orth.; Étymol. et Hist.; Fréq. abs. littér.; DÉR.; suff./rad. prérom./esp./cat./a. fr./b. lat./anc. prov/lat. pop./dér./id./cf./v./subs. fém./masc./adj./fig./gén./étymol. ....?? Sobreira ►〓 (parlez) 13:21, 30 January 2018 (UTC)

Curiosity on French[edit]

Anyone knows why FR cristal is not written with <y> as in etym and English? Sobreira ►〓 (parlez) 10:51, 17 January 2018 (UTC)

It was written with <y> until the 19th century. --Per utramque cavernam (talk) 11:45, 17 January 2018 (UTC)
No, the spelling with y was always only a variant that was listed as that in the dictionary of the French Academy. According to the etymology, both the French and English words should be spelled cristal as in Middle English, Old English, and Old French. The introduction of the y based on the Latin and Greek spelling was one of many examples of nonsense introduced in the 15th to 17th centuries (when linguistics didn't exist yet) by people who wanted to flaunt their knowledge of these languages and even tried to change English grammar according to similarly misunderstood differences between English and Latin, e.g. the split infinitive. --Espoo (talk) 10:28, 18 January 2018 (UTC)


Oh ffs. ばかFumikotalk 11:16, 17 January 2018 (UTC)

Needs clipping of all guesswork and history lesson. Korn [kʰũːɘ̃n] (talk) 23:13, 27 January 2018 (UTC)
Well, it's all besides the point of the perfectly fine, referenced etymology. I've just removed the rambling paragraph. —Μετάknowledgediscuss/deeds 23:44, 27 January 2018 (UTC)
Yes, that was in need of cleanup. I'm currently in the process of researching the term to find a date. The variance in reading (source EN term /koʊkoʊ/ would be expected to produce JA /koːkoː/ rather than /kokoa/) also deserves mention. ‑‑ Eiríkr Útlendi │Tala við mig 18:48, 29 January 2018 (UTC)


RFV of the etymology. Dokurrat (talk) 18:51, 17 January 2018 (UTC)

If it's indeed from Malay, I think it would be a phono-semantic matching. — justin(r)leung (t...) | c=› } 04:23, 18 January 2018 (UTC)
It can be found in:
Changed it to psm. Wyang (talk) 07:54, 18 January 2018 (UTC)
Roger... Dokurrat (talk) 12:52, 18 January 2018 (UTC)

etymology of moraine[edit]

The etymology given at moraine is very different from that on http://www.cnrtl.fr/definition/moraine And is there such a thing as Savoyard Italian? It's not even mentioned on https://en.wikipedia.org/wiki/Savoyard_dialect --Espoo (talk) 08:58, 18 January 2018 (UTC)


Is there a typo in the etymology of Latin fraus (deceit, fraud)? It is supposedly derived from Proto-Indo-European *dʰwer-, but that means "door". On the other hand, Proto-Indo-European *dʰrewgʰ- (to deceive, mislead) is said to be derived from *dʰrew- (to mislead). — SGconlaw (talk) 16:13, 18 January 2018 (UTC)

The entry has been expanded to show intermediate steps. The semantic shift is still unexplained, though. Considering fraud truly descendant from door, semantically, "behind closed doors" seems the only interpretation, but not particularly convincing (at least not if bent to make a case for the opposite). Maybe a trapdoor as could be used in hunting would be another. The semantic connection from door (*dʰwer-) to board, wood (PIE?) and by extension tree (*dóru, alternatively *drewo-) is obvious.
I have been told of before because I am not well informed about the phonological difference between d and dʰ. Ignoring that for a moment, I would assume door and fraud both related more to dead trees (wood) and the perhaps related adjectives *deru-, *drew- (hard, firm, strong, solid) than to each other. E.g. a wood-puppet would be a figure for deception and fraud generally implies hardship, while from the opposite point of view, deception seems firmly cunning.
The Indo Iranian branches do have d in the descendants for PIE door. Those may have well proven reasons. Likewise, the Slavic languages having forms of doru for wood might be later and not particularly related.
Still, may I ask why a d>dʰ shift would be completely out of the question?
But wait, there is more. We still say pull a trick. The etymon of trick is uncertain, the first variant given there is from latin trīcārī. Whereas the second is from *dreg- (“to drag, scrape”). I can't say much to the first one. But the second one is notable because drag gives *dʰreǵʰ- (to draw, drag) instead - note the different dʰ to the other form. And notably, latin for wood is "that which is collected" (lignum), semantically close to draw-in., while I stipulate German Holz (wood) attests to a similar relation via holen (get, collect) - Holz holen is very idiomatic - probably via the sense "hew", whence heavy (cf. hard). In the light of this, cari in trīcārī can be deemed related to [[carrus], whence carry. I won't even mention how I think the number three figures into this.
This seems to suggest a relation of tricking to wood. But I cannot close the semantic gap, far from it. Before I say more, what are your opinions? Rhyminreason (talk) 13:53, 19 January 2018 (UTC)
Homonymy does not always mean a connection. e.g. lie, drake, -most. mellohi! (僕の乖離) 16:54, 19 January 2018 (UTC)
Semantic pairing and homonymy together do more often indicate significance, though.
"True" is from *drewh₂- (steady, firm). Truth is obviously opposite of "fraud." So is d versus dh a question of good / bad? Let's suppose there was a prefix s-, to denote negation, (the phoneme doesn't matter for sake of the argument). Then "not-true" (not-strong, whatever) would become a standalone word and the prefix and beginning phoneme would merge. Softening a hard consonant would directly verbalize the contrast.
A similar semantic shift, red oak > robust, is also given to compare to the development from *dóru (tree) to true.Rhyminreason (talk) 18:56, 20 January 2018 (UTC)

toxify, detoxify[edit]

"detoxify" appears earlier (1907) than "toxify" (1917). Is "toxify" a backformation? What is the etymology of "detoxify" if not de- + toxify? DTLHS (talk) 01:50, 19 January 2018 (UTC)

With things like this, I sometimes wonder if we can be sure toxify was not used before 1917 (as opposed to just, it was not that common, and uses didn't survive), but it certainly seems that way: I too can't find any uses from before 1917, while, I can find detoxify as early as 1905, in George Gould's Dictionary of New Medical Terms, which does not include toxify. Stedman's Medical Dictionary from 1918 is interesting here, because it provides both an etymology ("detoxify (de-toksi-fi) [L. de, from, + toxicum, poison, + facere, to make.]") and a gloss, "Detoxicate.", which is attested at least one year earlier, in a Clinical Urinary Analysis by J. C. B. Statham in the 1904 Journal of the Royal Army Medical Corps. I would surmise "detoxify" was formed similarly, as de-+tox(i)-+-ify, possibly even influenced by detoxicate. I can find toxicate since at least 1899, but only in the sense "intoxicate" (a 1901 book explicitly states that some people say "toxicate" where other people say "intoxicate"); I'm not sure if the sense "toxify" is also attested that early (but I find it in a law dictionary by 1910). - -sche (discuss) 02:32, 19 January 2018 (UTC)

Scottish Gaelic slaod[edit]

Is this word a loan from English? —This unsigned comment was added by CecilWard (talkcontribs).

I've added an etymology at slaod Leasnam (talk) 01:31, 22 January 2018 (UTC)

Latin Hispania[edit]

Is the presented Punic etymon attested in this form (I mean the spelling, not in the square script), or is it even attested at all? It was added in 2007 by an IP and all hits on Google derive from Wiktionary, so I don't exactly have high hopes. The etymology at Spain has a related but different term that may also be unattested. ←₰-→ Lingo Bingo Dingo (talk) 11:04, 19 January 2018 (UTC)

The theory that Hispania is from a Phoenician/Punic word for "land of rabbits / hyraxes" is old, though the spelling of the supposed etymon varies from source to source, and you are right that our entry's may not be attested. Simon says (well, M. A. Simón et al. say, in Ten years conserving the Iberian lynx, ISBN 8492807806, page 1950) that "Hispania, the name that the Romans gave to the peninsular, derives from the Phoenician i-spn-ya, where the prefix i would translate as “coast”, “island” or “land”, ya as “region” and spn[,] in Hebrew saphan, as “rabbits” (in reality, hyraxes). The Romans, therefore, gave Hispania the meaning of“land abundant in rabbits”, a use adopted by Cicero, Cesar, Pliny the Elder and, in particular, Catulo, who referred to Hispania as the cuniculus peninsula." And the theory does seem to have had some currency in Roman times, pun intended, as some coins depict Hispania and rabbits. But Wikipedia's article is right to say that the origin is uncertain and debated. Milton Azevedo (in Portuguese: A Linguistic Introduction, 2005, ISBN 0521805155, page 6) calls the rabbit theory "a charming legend [...of] a Phoenician name, i-shepham-im or 'land of rabbits'". Michael Dietler and ‎Carolina López-Ruiz (Colonial Encounters in Ancient Iberia, 2009, ISBN 0226148483), say: "Cunchillos 2000:224 [...] offers a new interpretation of the etymology of the Roman name for the Iberian Peninsula, i.e., Hispania, as derived from the Northwest Semitic word meaning “island/coast” ('i) and “north” (spn), therefore “northern island, island to the north,” or else “island of the metals (root spy/h, "beat metals", etc.). Both senses would fit well with geographic perceptions that the Iberian Peninsula might have triggered for the Phoenicians." Wikipedia lists several other theories. I will try to update Hispania and Spain later if no-one with more conclusive references beats me to it. :) - -sche (discuss) 18:04, 19 January 2018 (UTC)
Thanks. My intent isn't really to contest the theory, but specifically whether the term should be presented as attested or reconstructed. Maybe Wikitiki89 knows whether the name is attested? ←₰-→ Lingo Bingo Dingo (talk) 12:30, 20 January 2018 (UTC)
I've (centralized the content from Spain to Hispania and) simply commented out the Phoenician spellings, pending confirmation, since they don't resemble the ones I've saw most often mentioned in literature. I've updated the etymology to mention the uncertainty. - -sche (discuss) 23:31, 20 January 2018 (UTC)

Related but borrowed terms[edit]

If an item has been borrowed into a language A from the genetically unrelated language B, and the same word has also been borrowed from language B into languages C,D,E etc., which are all genetically related to the language A, what is the relationship between the word in the languages A and C-E called? What template should be used? See küçə for exemplification. Allahverdi Verdizade (talk) 13:46, 21 January 2018 (UTC)

I use the wording "whence also" and the template {{cog}}, but others insist that {{cog}} is only for inherited cognates. Why do you have to show the other Turkic borrowings on the Azerbaijani page at all? They are listed in the Descendants section of the Persian ancestor. --Vahag (talk) 22:12, 21 January 2018 (UTC)
Yes, and I am inclined to agree with them, as the notion of cognate rules out borrowing. Regarding your question. I believe that knowing into which other related and unrelated neighboring languages an item has been borrowed, and into what meanings it has come to evolve, is simply interesting and useful. One could surely go back to the etymon's entry and then click forward to the individual borrowed terms' entries, but this is simply tedious. Also, it may be more interesting for someone who looks up the Azerbaijani term ünvan whether this Arabic borrowing also exists in Persian, Ottoman Turkish, etc, and what it means there, than in Swahili or whatever other languages around the world where it also may be found. On the other hand, whoever is interested in the full range of descendants can go to the etymon's entry and check out the whole picture.Allahverdi Verdizade (talk) 00:17, 22 January 2018 (UTC)
Another interesting example is when a word is borrowed from an ancestor of another modern language, and it's descendant word does exist in the language's modern stage. Like Azerbaijani hündür (high, tall) is related to Mongolian өндөр (öndör, high, tall), but clearly we cannot claim cognateship because it is borrowed into Azerbaijani, not inherited, and neither can we say that it is borrowed from Mongolian. Since, technically, it is borrowed from some earlier stage of Mongolian (Middle Mongol language?) and not from it's modern form. Allahverdi Verdizade (talk) 13:17, 27 January 2018 (UTC)
I really see no reason not to say that hündür and өндөр (öndör) aren't cognates in a case like that. Who says that the word "cognate" necessarily requires a purely inherited pedigree of both words concerned? —Mahāgaja (formerly Angr) · talk 15:14, 27 January 2018 (UTC)
I will refer to this discussion that deals with the delimitation between the concepts of cognate and loanword, and also to this article written by Asya Pereltsvayg. Shortly, two words are only considered cognates if they are inherited from the same ancestor language. I think, to answer my own question, that the best way of doing what I wanted to do is to say "Borrowed from Middle Mongol, whence also {{l|mg|өндөр}}". Of course, only given that scholarly work supports the assumption that it is in the Middle Mongol period the borrowing was made, and not later or earlier. Allahverdi Verdizade (talk) 21:25, 27 January 2018 (UTC)
Pereltsvayg states her position strongly as if it were inconvertible fact, but other sources disagree. See for example the definition of "cognate" in Trask's Student's Dictionary of Language and Linguistics: he begins with the narrow definition that cognates must be inherited from a common ancestor (English foot and German Fuß) but then acknowledges the other meaning as well, that borrowed terms can also be considered cognates (English jail, Old French jaiole, Spanish jaula, Basque txabola). —Mahāgaja (formerly Angr) · talk 21:43, 27 January 2018 (UTC)
Yes, although Durkin gives a similar definition of cognate in Oxford Guide to Etymology on the page 290: Developed from a common ancestor. Among the cognates of Old English sæd are Old Dutch sat, Old Saxon sad, Old High German sat, Old Icelandic saðr, Gothic saþs; these words are all cognate.:; it is not a coincidence that all the examples he gives are indeed cognates in the narrow sense. Furthermore, throughout the entire book he uses the terms as opposed to each other, along the lines with "terms XY are cognate or rather borrowings". Anyway, I guess the opinion that loanwords should be included under the term cognate is legitimate, just as the opposite stance. I personally find it useful to distinguish between cognates (Foot and Fuß) and related words such as hündür and өндөр.


RFV of "characters selected as literally 'to overthrow Communist Party (of China)'" in the etymology. — justin(r)leung (t...) | c=› } 13:51, 22 January 2018 (UTC)

Shouldn’t you list this at RFV? — SGconlaw (talk) 20:48, 27 January 2018 (UTC)
No, it's RFV-etymology, not of any sense. — justin(r)leung (t...) | c=› } 20:55, 27 January 2018 (UTC)
Oh I see. — SGconlaw (talk) 21:02, 27 January 2018 (UTC)

Outcome of PIE *-kn- in PCelt[edit]

What's the normal outcome of PIE *-kn- in PCelt, simply *-kn- or *-gn-? --Victar (talk) 03:11, 27 January 2018 (UTC)

{{R:cel:Matasovic 2009}} reconstructs them separately, and the name Licnos is attested in Gaulish, so apparently they were still distinct in PCelt. But they did merge in Insular Celtic, so you can't tell whether a Goidelic or Brythonic word comes from -kn- or -gn-. —Mahāgaja (formerly Angr) · talk 09:57, 27 January 2018 (UTC)
Makes sense. Thanks. --Victar (talk) 21:34, 27 January 2018 (UTC)

Classical Latin pronunciations in some Vulgar Latin reconstructions[edit]

Is having what the Classical Latin version of an unattested Vulgar Latin verb's pronunciation would have sounded like needed for all entries? I get that it's useful for many if not most of the cases, but... For example some of the Germanic ones like *wadanio would not have really existed in Classical Latin times would they? Since the pronunciation templates there are automated, it gives the "Classical" pronunciation as /waˈda.ni.oː/ (although the letter 'w' wouldn't have existed and it would have been spelled with a 'u') and the Vulgar one as /βaˈda.ni.oː/, [baˈða.nʲo]. However, wouldn't the Vulgar Latin pronunciation (that led to the Romance descendants) actually be more like what we have listed as the classical, like /waˈda.ni.oː/... and therefore the shift to the initial -gua- in most Romance languages? I don't see the jump from -βa- or -ba- working as well, except for maybe Venetian and Friulian? Could be wrong here; I'm just asking. Word dewd544 (talk) 21:35, 27 January 2018 (UTC)

Gothic 𐍃𐌹𐍀𐍉𐌽𐌴𐌹𐍃 (sipōneis) from Celtic (*sepānios)?[edit]

Found this in Dennis Green, Language and History in the Early Germanic World (Cambridge 1998) pp. 156-157; he claims that 𐍃𐌹𐍀𐍉𐌽𐌴𐌹𐍃 (sipōneis, follower) is a borrowing from a continental Celtic language (a plausible connection; Celtic borrowings are not unheard of in the Gothic corpus), the specific form - unattested - being *sepānios, ultimately from Proto-Indo-European *sekʷ-. My question to anyone here with some knowledge of Indo-European and Celtic linguistics is if this is a plausible development morphologically? Can *sekʷ- yield *sep- and is there a precedent for the *-ānios part? What sound laws are at play here? — Mnemosientje (t · c) 21:52, 27 January 2018 (UTC)

The change of to p is found in Brythonic and in most varieties of Gaulish; indeed, that sound change is the origin of the term P-Celtic. I'm not aware of a Celtic suffix -ānios, but that doesn't mean there wasn't one. It seems plausible enough at any rate. —Mahāgaja (formerly Angr) · talk 22:26, 27 January 2018 (UTC)
Thank you, that's very interesting. I've tentatively added Green's proposed etymology to the entry. — Mnemosientje (t · c) 01:16, 28 January 2018 (UTC)

endotenon, epitenon[edit]

Why "tenon" and not "tendon"? DTLHS (talk) 04:02, 29 January 2018 (UTC)

Both words first appear in 1916. DTLHS (talk) 04:15, 29 January 2018 (UTC)
Is seems that they, and the tendon-y prefix teno- in any case, are from a Greek word [script needed] (tenōn, tendon), related to τείνω (teínō). - -sche (discuss) 19:25, 29 January 2018 (UTC)
[script needed] = τένων (ténōn) (gen. τένοντος)? - 01:17, 10 February 2018 (UTC)
Yep; τένων (ténōn) (gen. τένοντος (ténontos) is exactly right. —Mahāgaja (formerly Angr) · talk 07:29, 10 February 2018 (UTC)