Wiktionary:Etymology scriptorium/2013/December

The etymology and cognates don't quite add up. *sinþaną would not give sinnan in OE, it would become sīþan. The only sources for a form sinnan are *sinnaną and *sinjaną. The same applies to the Old High German cognate as well. The relationship to sendan is even more obscure. —CodeCat 19:58, 1 December 2013 (UTC)

Then what of the etymology of OE findan, from *finþaną? I suppose a relation to sendan would be from the causative *sandijaną (“to cause (one) to go”).

Look at *sinþaz if you have any doubts about the plausibilty of what CodeCat said. A possible explanation: if you look at the etymology you added to the entry at *finþaną, you'll notice it cites two Proto-Indo-European roots: *pent- and *penth- (the etymology at pons analyzes the source of that cognate- corresponding, I suppose, to *penth- in the *finþaną etymology- as *pónt-h₁s, while πόντος (póntos), पथिन् (pathin) and հուն (hun) have *pónteh₁s as the source for those- which looks like only a difference in ablaut grade) . That suggests to me that there must also have been a Proto-Germanic form with a *d to correspond to the *dʰ in *penth- / *pen*dʰ . Another part of the etymology seems to support this, saying "Related also to Old High German fendo, fendeo (“pedestrian, footsoldier”), Old Saxon fāþi (“walking”), Old English fēþe (“locomotion, walking, gait, pace”)". The Old Saxon and the Old English would be what one would expect from *finþaną as far as the consonants and the vowel length are concerned, given the action of the w:Ingvaeonic nasal spirant law, though I don't know enough about the sound changes affecting vowels to have any clue beyond that. I wonder if there's a similar dual-form situation, with *sinþaną having a voiced doppelganger (would w:Verner's Law apply here if the accent were shifted from the first to another syllable?). By the way, the etymology at path says that it's descended from the same *pent- root as above via a borrowing into Proto-Germanic from Scythian (an Indo-Iranian language). Chuck Entz (talk) 04:02, 3 December 2013 (UTC)

The sound changes still don't really add up though. I'm not sure what *penth- is supposed to signify. What is the th here? If we go with a root *pent- (which is straightforwardly the root of *finþaną) then there is no way that -nþ-, nor its voiced counterpart -nd-, can become -nn- in Old English and Old Saxon. -nþ- must undergo the nasal spirant law, and -nd- simply appears unchanged in both (compare hand). So there is no post-Germanic sound law that can account for -nn- which means that it must have been present in Proto-Germanic already, implying the existence of a Proto-Germanic *sinnaną as the direct ancestor of sinnan. *sinnaną itself could come from an earlier *sent-na-, in which the -t- would probably have been assimilated away (I think there are other examples of this, but I'm not sure). —CodeCat 03:03, 4 December 2013 (UTC)

I am also finding some sources having OE sinnan answering to PGmc sinþnan or senþnan. Is this plausible? Leasnam (talk) 05:58, 11 December 2013 (UTC)

Yes, but I doubt that Proto-Germanic still had -nþn-. —CodeCat 15:06, 11 December 2013 (UTC)

While I don't doubt the etymology of this word in any way whatsoever, I've always found myself confused when I find words that derive from an Old English word that has a medial cg, yet have a "g" or (even more confusingly) a "y".

I first chalked it up to a word becoming more literary and it being rewritten by partially-bilingual Old English|Anglo-Norman scribes when I saw a resulting word having a "g", but then I find words like "lair" and "lay"--which I would assume were relatively common terms--having an "i" or "y" where a "dg" would be expected.

Can someone explain these conundra to me? Was it a dialectual (I know it's dialectal; force of habit) thing I'm unaware of, some... common change that occurred depending on the placement of a vowel, or does no actual existing explanation exist?

If no actual reason exists, could someone at least present some sort of possible theory as to how this happened? Tharthan (talk) 02:19, 4 December 2013 (UTC)

Not all of the Old English verb forms have -cg-. Two forms of the present tense have ġ, and so does the entire past tense. So that may be the source of the discrepancy. The same also seems to have happened with lie and say. —CodeCat 02:55, 4 December 2013 (UTC)

"cg" was the geminate form of "g" (or ġ), and so they alternate in the same way as other geminates. Compare habban where the geminate "bb" alternates with the non-geminate "f". --WikiTiki89 03:04, 4 December 2013 (UTC)

CodeCat is right. Many irregular verbs in Old English were normalised when they passed into later Middle English (as opposed to earlier Middle English), and the second and third person singular forms, as well as the past and imperative forms were used to reconstruct the verb stem anew (cf also maken < maca/macast/macaþ). Leasnam (talk) 03:34, 4 December 2013 (UTC)

What about nouns, like dog? Tharthan (talk) 11:47, 4 December 2013 (UTC)

The spelling -cg- actually represents two different sounds. One is a palatal affricate, the other is just a geminate g. We should really make a habit of writing one as -cġ- (or -ċġ-?) and the other as plain -cg-. —CodeCat 15:23, 4 December 2013 (UTC)

I thought that the "geminate g" is "cg" and is always pronounced /ddʒ/. --WikiTiki89 15:45, 4 December 2013 (UTC)

No, not always. There are also a few rare cases, like docga, where it's /gg/. Palatalisation only occurs when it's derived from Germanic -gj-, which is the usual source of geminates in West Germanic. But whenever -cg- doesn't come from -gj-, there is no palatalisation. —CodeCat 15:51, 4 December 2013 (UTC)

Is there any evidence for that other than the modern pronunciation of dog? (I'm not disputing, just curious.) --WikiTiki89 15:58, 4 December 2013 (UTC)

Probably, but I can't think of any examples. Like I said it's rare. —CodeCat 16:25, 4 December 2013 (UTC)

So should we fix the pronunciation at docga then? --WikiTiki89 16:29, 4 December 2013 (UTC)

Yes, definitely. Here is more information if you want, as well as some more examples of geminate : . —CodeCat 16:32, 4 December 2013 (UTC)

Remember too that it is usually believed that the -ga in docga represents a suffix, so this may have aided in the solidification of the sound as hard g. (deprecated template usage) earwig, (deprecated template usage) stag, and perhaps (deprecated template usage) flag (if we derive it from flacg) are other examples of this. Leasnam (talk) 20:26, 4 December 2013 (UTC)

So are you saying that it was in fact pronounced /doddʒa/? --WikiTiki89 20:40, 4 December 2013 (UTC)

Nay, just the opposite: more likely /dok.ga/ Leasnam (talk) 20:43, 4 December 2013 (UTC)

I took "this may have aided in the solidification of the sound as hard g" to mean that it originally wasn't a hard g, but was reanalyzed as such. --WikiTiki89 20:49, 4 December 2013 (UTC)

Yes, my bad. What I originally wrote was confusing. What I meant was the suffixal hard -g would have continued to be seen as a suffix, but written the same way as -ċġ. I corrected what I wrote. Sorry. Leasnam (talk) 21:06, 4 December 2013 (UTC)

Can this really be from Scots? I'd like to see a citation for that even being a possibility; every source I can find says it comes directly from German or Yiddish. —Μετάknowledge^{discuss/deeds} 22:38, 7 December 2013 (UTC)

It's conceivable that the word was first borrowed in Scotland, from where it later spread. --WikiTiki89 23:00, 7 December 2013 (UTC)

I think it's a mistake. The word speel is old Scottish and northern English, but with a different meaning (probably from speeler, an acrobat). I suppose it's possible that the Old English word spel (news) was retained in Scotland and became speil, but I can't find any evidence. I think several different words have been confused over time, but the modern usage (schpiel) seems to have been a relativlely recent importation from Yiddish. I don't know any Scottish Yiddish speakers, but I suppose there might be a few. I'm happy to be proved wrong, but, in the absence of any evidence, I seriously doubt the Scottish origin. The fact that they have a similar word that has recently become confused with the modern meaning does not constitute an etymology. Dbfirs 22:59, 20 December 2013 (UTC)

The second meaning of the word - "bird of prey", later "gyrfalcon" - seems, according to its plural nominative listed in Cleasby-Vigfússon, to have an i-stem instead of an a-stem, however, this contradicts the etymologies given by the same dictionary, since both *walaz and *habukaz are a-stems. Can there be any solution (or an alternative etymology) provided to this word, or am I possibly misunderstanding something? - Myndfrea (talk) 15:46, 8 December 2013 (UTC)

The etymology (wet > unk) does not make sense to me. --WikiTiki89 22:26, 11 December 2013 (UTC)

*wet is the nominative form. —CodeCat 22:56, 11 December 2013 (UTC)

Well either we shouldn't have etymologies in inflected forms like that, or we should put in a less misleading etymology as I just did. --WikiTiki89 23:02, 11 December 2013 (UTC)

There's a bunch of unsourced Proto-Germanic and PIE in the etymology at heaven that the OED doesn't support and (at its shame entry) contradicts. It seems to have come from somewhere: could we get that part of the etymology sourced? and (still better) have someone knowledgeable compare the two sources for probable accuracy? — LlywelynII 10:28, 12 December 2013 (UTC)

I don't see the connection with "shame", but the Germanic reconstructions are correct as far as I can tell. Gothic directly corroborates one of them, and the other can't have been much different. —CodeCat 14:15, 12 December 2013 (UTC)

Thanks, but I'm not saying it's wrong. At least where it ends up looks fine (apart from the *kem- / *kam- issue). I'm saying we should add a source for it. (The OED Online confirms "shame", on the assumption that a derivation from roots related to "pre-Germanic" *kem-, "cover" is correct. It doesn't confirm any of the rest.)

It's hard to see how the Gothic could confirm anything: it's a redlink. Are you talking about one of the other ones? or do you have a Gothic source we should use to create the entry for that word? — LlywelynII 01:17, 13 December 2013 (UTC)

I'm not sure what you mean. Gothic "himins" straightforwardly matches Germanic "himinaz". —CodeCat 01:22, 13 December 2013 (UTC)

FWIW, Gerhard Köbler's dictionary of Old English says the following in the entry for heofon:

• heo-f-on, ae., st. M. (a): nhd. Himmel; Vw.: s. -bū-en-d, -cyn-ing, -dréa-m, -lic, -līc-e, -rīc-e (1), -stō-l, -wear-d; Hw.: s. heo-f-on-e; vgl. got. himins, an. himinn; E.: germ. *hemina-, *heminaz, st. M. (a), Decke, Himmel; idg. *k̑emen-, *k̑ōmen-, Sb., Stein, Himmel, Pokorny 22; s. idg. *k̑em- (3), V., bedecken, verhüllen, Pokorny 556; vgl. idg. *ak̑- (2), *ok̑-, Adj., Sb., scharf, spitz, kantig, Stein, Pokorny 18; L.: Hh 156, Hall/Meritt 177b, Lehnert 110a

- -sche (discuss) 19:08, 16 December 2013 (UTC)

Proto-Slavic word tanьc is a all Slavic borrowing from Proto-Germanic *dansōną, and not a German borrowing as commonly understood, thus we should include tanьc. Duh (talk) 16:26, 16 December 2013 (UTC)

Seems possible, but do you have a source? --WikiTiki89 16:29, 16 December 2013 (UTC)

I don't understand your reasoning at all. It can't be borrowed from Proto-Germanic because it has the High German consonant shift. Secondly, Proto-Slavic words can't end in a consonant. And thirdly, where does the ь come from? It's not in the German word. If the word was really borrowed in Proto-Slavic times, it would be *tǫcь or *tǫsъ instead. Something like Proto-Slavic *tanьc makes no sense on many levels, and I don't know why you think it does. —CodeCat 16:33, 16 December 2013 (UTC)

Firstly, is it impossible that it was a borrowing from Old High German into Common Slavic? That is, after the High German consonant shift and after the nasalization of Slavic vowels. Secondly, it presumably would have been tanьcь. And thirdly, the ь could have been inserted to maintain the CV syllable structure. It makes perfect sense to me and I'd believe it if a source were given. --WikiTiki89 16:42, 16 December 2013 (UTC)

There's not a single example that I know of that inserts an extra vowel in such situations. For example, compare *korljь, which was borrowed from High German no earlier than the 9th century. —CodeCat 17:10, 16 December 2013 (UTC)

Well or/er/ol/el seem to have been treated more or less like vowels in Proto-Slavic. Are there any other words we can compare? --WikiTiki89 17:14, 16 December 2013 (UTC)

Or if you prefer, it could have been tancь with the epenthetic vowels inserted later in each language. --WikiTiki89 17:19, 16 December 2013 (UTC)

Yes I aree with Wikitiki89 *tanьcь is correct form (my mistake "Tippfehler"). It is impossible that it would look as *tǫcь or *tǫsъ because a wouldn't change, and perhaps the root would be only *dan- thus d shifted to t as common in Slavic languages with suffix *-ьcь. Also it is impassible that word came from German, because of Old Dubrovnik literature which language had non German influence, and word tanac, or verb tačati is very common. Perhaps even *tanьcь isn't derived from Proto-Germanic at all, but from Proto-Indo-European *tens-. All of that has perfect sense. Duh (talk) 17:25, 16 December 2013 (UTC)

Which descendants does *tanьcь actually have in Slavic? So far we've talked about borrowing but not about the reconstruction. —CodeCat 17:37, 16 December 2013 (UTC)

These:

• East Slavic:
  • Belarusian: та́нец (tánjec)
  • Russian: та́нец (tánec)
  • Ukrainian: та́нець (tánecʹ)
• South Slavic:
  • Bulgarian: танц (tanc)
  • Macedonian: танц (tanc)
  • Serbo-Croatian: tanac/танац
  • Slovene: tanec
• West Slavic:
  • Czech: tanec
  • Polish: taniec
  • Slovak: tanec

Feel free to correct me if I made any mistakes. --WikiTiki89 17:48, 16 December 2013 (UTC)

Then I think we can include *tanьcь as a borrowing from Old High German. This is the only way to account for the -an- (rather than -ǫ- as in earlier borrowings) and the t-. —CodeCat 17:51, 16 December 2013 (UTC)

We can only conclude that it was not borrowed earlier than that. We cannot yet conclude that it was not borrowed later. --WikiTiki89 17:54, 16 December 2013 (UTC)

It was probably a late borrowing, but it wasn't so late that the word only spread to a few Slavic languages. What we do know is that the (word-final) cluster -nc was not permitted, and needed a vowel to break it up. So it was borrowed as *tanc-, but this became *tanьc in the nominative singular. We can't be sure if the other forms ever had a yer in them too, because it would have disappeared even if it had existed, and the forms you listed don't have any trace of it (as expected). We might be able to set an upper boundary if we can find later borrowings that leave final -nc without an epenthetic vowel. This word must have then been borrowed before that. It would also be helpful to know the earliest attestations of this word in the Slavic languages. —CodeCat 18:02, 16 December 2013 (UTC)

Then what happened to the ь in Bulgarian and Macedonian? Or could it have been a later re-borrowing? --WikiTiki89 18:06, 16 December 2013 (UTC)

Also, Ukrainian has танок (tanok), which is puzzling. It could be a back-formation, but I don't know of any other back-formations of this kind. --WikiTiki89 18:40, 16 December 2013 (UTC)

I don't know what happened in Bulgarian and Macedonian, this could be a later re-borrowing similar to tanc and tenc in some Slovenian dialects, or tanc in some Croatian dialects, which all have had a significant German influence. If so, they should not be included. Also Ukrainian танок (tanok) is a different form of tanec, it is made from suffix *-ъkъ similar to Czech tanek, Polish tanok, Croatian tanak, tančak, tančić, tančac or Slovenian tanek, tanček all with roughly the same meaning only made from different suffices, which indeed indicates of *tanьcь Proto-Slavic origin. Duh (talk) 19:29, 16 December 2013 (UTC)

Then what happened to the c when the ъkъ was added? The c should have come out as a č rather than disappearing. --WikiTiki89 19:34, 16 December 2013 (UTC)

Also indicates that original word is tan and the latter are suffices (there is also word tan in Old Croatian, with meaning "fun", "party"), but this indicates a native Proto-Slavic word, which mixed with Proto-Germanic one. Duh (talk)

• Duh (talk • contribs) is one of the sockpuppets of Slavić (talk • contribs) who has a history of deliberately fabricating etymologies. Don't waste time on this guy. --Ivan Štambuk (talk) 21:18, 16 December 2013 (UTC)

  It's an idea worth looking into, regardless of who proposed it. As I said, though, we still need more evidence. --WikiTiki89 21:24, 16 December 2013 (UTC)

These words are not derived from CSl. Polish borrowed MHD tanz (→ taniec, with pseudoetymological -iec). From Polish it entered Russian (and Ukrainian and Belarusian) and Czech (and Slovak). In BCS and Slovene it's a Bohemianism; i.e. borrowed during the Illyrian movement. In Bulgarian it's a Russianism: Bulgarian has танц which is a respelling based on NHG Tanz, while the older variant танец is now obsolete. Macedonian has both танец and танц, reflecting the different transmitter languages. It's absent in Church Slavonic which uses ликовати, играти and плѧсати (and others) when translating 'to dance'. --58.7.112.243 11:02, 5 February 2014 (UTC)

You clearly know a lot about this word, but do you have citations to back that up? And by the way, no one mentioned Church Slavonic. Church Slavonic is not the ancestor of most Slavic languages. --WikiTiki89 18:20, 5 February 2014 (UTC)

I wasn't suggesting that Church Slavonic was the ancestor of any modern Slavic language but, rather, that—if this word was indeed derived from Common Slavic—we might reasonably expect *таньць (and a denominal verb) to occur in at least some Bible translations given that the words 'n. dance' and 'v. to dance' occur in a total of about forty passages. Max Vasmer's etymological dictionary (for Russian, and widely available online) states that it entered Russian through Polish (oldest recorded use is by Grigory Kotoshikhin), and from Middle High German to Polish. Skok's dictionary (for BCS) lists it as a "an-Slavic borrowing from MHG Tanz". --124.149.93.161 03:33, 6 February 2014 (UTC)

Unless I'm mistaken, "pan-Slavic borrowing" is exactly what we said it was. It could have been a colloquial term, which would explain it's late attestation in Russian and why it wasn't used in Church Slavonic. --WikiTiki89 03:38, 6 February 2014 (UTC)

Skok uses "svesl" to broadly mean 'pertaining to all modern Slavic languages'; i.e. not necessarily a (Proto-)/Common Slavic origin. --124.149.93.161 04:20, 6 February 2014 (UTC)

Well then how could all of the Slavic languages have borrowed the term exactly the same way so as to fit Proto-Slavic *tanьcь? --WikiTiki89 04:28, 6 February 2014 (UTC)

Much in the same way that BCS kuhinja, Belarusian кухня, Bulgarian кухня, Czech kuchyně, Macedonian кујна, Polish kuchnia, Russian кухня, Slovak kuchyňa, Slovene kuhinja and Ukrainian кухня are not from CSl. --124.149.93.161 05:30, 6 February 2014 (UTC)

And how exactly is that? By the way, it would help if you didn't use ambiguous abbreviations. I thought that "CSl" meant "Church Slavonic"; now I realize it means "Common Slavic". --WikiTiki89 05:38, 6 February 2014 (UTC)

But that is more believable anyway, because the words you just listed are all a bit different, some implying *kuxynja and others *kuxъnja, and one of them even *kujna. Therefore, it is more likely that they came from somewhere else. --WikiTiki89 05:44, 6 February 2014 (UTC)

Maybe this isn't the best place for this, but I've noticed a user keeps reverting the Romanian descendant for this Latin word to something that is incorrect. The actual etymological descendant for it in Romanian is 'duroare', an archaic word that is no longer used, but still identified in older speech and documents and found in the etymological dictionary, and thus the one that belongs there in the list. It follows the expected transformation and sound shifts in Ro. Instead, the users (who do not have a username or account but just IP addresses 213.233.104.28 and 78.96.154.234), place the terms 'durere' and 'dor' under it, which are not descendants of the term. Durere, also meaning pain, is simply a construction made from the verb 'durea' with the "long infinitive" suffix '-re', changing it to a noun based on or resulting from the action of the verb, as happens with most verbs in Romanian, such as 'putea' > 'putere', 'mânca' > 'mâncare', etc. And 'dor' is most likely descended from Latin 'dolus' instead, which is related but not the same term as 'dolor'. I just didn't want to have to keep monitoring the page all the time for changes so I thought I'd mention it. Word dewd544 (talk) 23:25, 16 December 2013 (UTC)

Does someone knows the etymology of Slovene words hrepeneti, jokati, še, kot and word ve, all of them are Serbo-Croatian Kajkavian as well, if so please share. 5.43.160.128 15:03, 18 December 2013 (UTC)

"Dug" meaning "udder" seems to derive from the Indo-European root *dheugh "to milk" (cf. Sanskrit dogdhi "milks"), yet I can find nothing in the usual places (e.g. American Heritage). If someone could look into this (for example look up the root in Pokorny) that would be great. If this etymology turns out to be correct, it would be an example of the well-known fact that Grassman's Law (which applies in Greek and Indo-Iranian) is not Indo-European. 78.29.206.28 22:19, 21 December 2013 (UTC)

Well, Etymonline suspects a borrowing from Scandinavian, citing Old Swedish dæggia "to suckle", but that verb is from a different root, namely *dʰeh₁(i̯)- "to suck (mother's milk)". However, the ablaut involved would be quite unexpected, although analogy can certainly lead to strange ablauts.

Also, I suspect that your suggestion doesn't work. First, the semantics of *dʰeu̯gʰ- are more complicated, and the LIV reconstructs "to hit" as the primary meaning of the root and its aorist. I can't go into all the detail right here, and you could check the book yourself if you can read German. Second, and more immediately damning, I'm pretty sure the sound laws don't work out. Something like Proto-Germanic *dugu-, *dugi- or *duga- would never have become dug, but something like *dow or *die I think. I also suspect that's why Etymonline thinks it's from Scandinavian – many words of this shape, ending in a short vowel plus /g/, are. It would have to go back to a form with gg or so, certainly not a simple /g/ because Old English didn't have one. Even different suffixes wouldn't change that. And with a suffix starting with n, Kluge's law would apply and you'd get Proto-Germanic *dukk-. I just see no way to make the connection work. It seems to be an udder impossibility (pardon the gratuitous pun). --Florian Blaschke (talk) 03:37, 18 February 2014 (UTC)

Where is the source? I found *kory in "Etymological Dictionary of the Slavic Inherited Lexicon". Which is correct, or are they both correct, if so where is a source or any kind of reference? 5.43.190.207 03:28, 27 December 2013 (UTC)

*-y might work if the word were an n-stem like *kamy. But n-stems have -en- and not -ěn- like this word has, so that seems to disqualify it. There's also nothing listed among the descendants that would corroborate *-y, so I don't know what that dictionary is basing its reconstruction on, I'm skeptical of it. Does this ending appear in OCS? —CodeCat 03:44, 27 December 2013 (UTC)

It is корень in OCS, this would also suggest a dual reconstruction as *korenь, *korěnь. Should we put that? Gorska vila (talk) 16:23, 27 December 2013 (UTC)

Also where is the evidence for ě in *korěnь. Maybe correct reconstruction would be only *korenь. Gorska vila (talk) 16:41, 27 December 2013 (UTC)

Then what about the Serbo-Croatian and Ukrainian descendants, which explicitly reflect ě? —CodeCat 16:56, 27 December 2013 (UTC)

Ukrainian has a natural change like in камінь (gen. каменя) so this is not a ě reflection. Serbo-Croatian case is little more complex; this could be a quasi-ijekavian invention like podrijeklo, but there is also ě reflex in old literature so case isn't so clear. Anyhow we should put *korenь in. Gorska vila (talk) 19:12, 27 December 2013 (UTC)

But why? You said yourself that ě is clearly attested in SC. You're contradicting yourself now. —CodeCat 19:17, 27 December 2013 (UTC)

It isn't attested, not clearly at all, 19th century linguistic works like "Ilirska Slovnica" from Věkoslav Babukić a leading linguist at that time puts ě everywhere like lěpo, pěvati, něšto... but always koren. Another example is modern Croatian linguist Bulcsú László who holds extreme views on jekavization and writes prjetočiti, nješto, prjetila, prjedstava... all extreme unnatural jekavization of language, but he always writes koren. The ě reflection could be found in Marino Darsa where he writes korien and Bartholomeo Cassio who writes korin. So again it isn't clear at all. Another explanation can be that korěn is a later, and koren is a primer form. Gorska vila (talk) 19:36, 27 December 2013 (UTC)

Since apart from Serbo-Croatian, there is absolutely no evidence for ě, I would reconstruct it as *korenь. Ukrainian clearly does not support the ě because the "і" alternates with "е" in the declension, which would not be the case if the original vowel were ě. Serbo-Croation could be considered an anomaly in this case, as can be Russian as well, since Russian seemingly supports *korьnь. --WikiTiki89 20:49, 27 December 2013 (UTC)

You can't just brush away the anomalies though. They show that the reconstruction is not as straightforward as just *korenь either. When different descendants disagree in cases like this, it's often evidence of a post-Proto-Slavic origin (newly coined or borrowed), or a productive change/alternation of some kind. —CodeCat 21:03, 27 December 2013 (UTC)

Or it could be an innovation. The Russian case of confusion between etymological e/o and ь/ъ is actually quite common, although I can't remember any examples off the top of my head. In Serbo-Croatian, you would have to draw a timeline of the attested evidence of ě, but without further evidence, I think an innovation is the most likely scenario. --WikiTiki89 21:19, 27 December 2013 (UTC)

I remembered one example of Russian treating etymological "e" as if it were etymological "ь": лёд (ljod) from *ledъ. --WikiTiki89 03:34, 28 December 2013 (UTC)

I know for sure that in Serbo-Croatian korien and korin is first seen in 16th century, whilst before only form koren exists. Also Croatian language portal states *korenь as etymology . Gorska vila (talk) 22:17, 27 December 2013 (UTC)

I think that is enough evidence for *korěnь to be moved to *korenь. The dictionary that Gorska vila linked to also mentions Lithuanian keras as a cognate. --WikiTiki89 23:46, 27 December 2013 (UTC)

I moved the page (after accidentally moving it to Appendix:Proto-Slavic/korеnь, with a Cyrillic е). If we find more evidence of *kory, then we can consider moving it there. --WikiTiki89 00:06, 28 December 2013 (UTC)

I found an interesting passage in a Russian newsletter from 1888 (Русскій филологическій вѣстникъ):

Во второй половинѣ 41 стр. примѣры смѣшенія ѣ и е не убѣдительны: они приводятся изъ сѣверныхъ памятниковъ, гдѣ ѣ и е мѣнялись безъ разбору.
Примѣчаніе къ 41 стр., касательно вліянія сербскаго языка въ словѣ корѣнь, крайне наивно. Современное сербское коријень — сербская особенность, неизвѣстная ни въ одномъ изъ другихъ славянскихъ языковъ, ни въ древнерусскихъ памятникахъ, за ислюченіемъ галицко-волынскихъ (при конечномъ ь), гдѣ оно имѣетъ ѣ въ именит. и винит. корѣнь, а въ другихъ, гдѣ нѣтъ въ концѣ ь, напр. е—корене, въ сербскомъ вездѣ ѣ = ије.

Vo vtoroj polovině 41 str. priměry směšenija jě i je ne uběditelʹny: oni privodjatsja iz sěvernyx pamjatnikov, gdě jě i je měnjalisʹ bez razboru.
Priměčanije k 41 str., kasatelʹno vlijanija serbskago jazyka v slově korěnʹ, krajne naivno. Sovremennoje serbskoje koriјenʹ — serbskaja osobennostʹ, neizvěstnaja ni v odnom iz drugix slavjanskix jazykov, ni v drevnerusskix pamjatnikax, za isljučenijem galicko-volynskix (pri konečnom ʹ), gdě ono imějet jě v imenit. i vinit. korěnʹ, a v drugix, gdě nět v koncě ʹ, napr. je—korene, v serbskom vezdě jě = ije.

In the second half of the page 41, examples of the mixing of ѣ and е are not convincing: they originate from northern monuments, where ѣ and е interchange without distinction.
Note on page 41, concerning the influence of the Serbian lanuage in the word корѣнь, very naively. The modern Serbian коријень is a Serbian peculiarity, unknown in any other of the Slavic languages, nor in Old Russian monuments, excluding Galicio-Volyn ones (with final ь), where it has ѣ in the nominative and accusative корѣнь, but in other cases, without ь at the end, for example е, it is корене. In Serbian always ѣ = ије.

The only thing I failed to understand is which text it is referring to (when it says "page 41"). --WikiTiki89 02:54, 28 December 2013 (UTC)

This actually reminds me of a class of PIE words with a nominative in -ēn (contracted from earlier -en-s), direct cases with -en-, oblique cases with -n-. For example *uksḗn. The alternation between ě and e in this Slavic word somewhat resembles this, and I wonder if they could be related. —CodeCat 03:14, 28 December 2013 (UTC)

I think the alternation there is the same as that in Ukrainian, as Galicia and Volyn are in modern-day Ukraine. --WikiTiki89 03:18, 28 December 2013 (UTC)

But where did it come from? —CodeCat 04:35, 28 December 2013 (UTC)

The ordinary Ukrainian alternation of о/е with і, but with the і spelled ѣ, since it is pronounced the same and the modern Ukrainian orthographic distinction between і and и did not yet exist. This is all pure speculation on my part, as I do not know where to look for the text of these "monuments". --WikiTiki89 04:40, 28 December 2013 (UTC)

My speculation is confirmed by the second paragraph of w:Yat#Ukrainian. --WikiTiki89 21:49, 29 December 2013 (UTC)

I believe that korijen is a poetic invention, similar to other ě inventions in Dubrovnik literature. Like cvijel common Croatian, Slovenian and Serbian word cvil from *kvil, also cviljeti and cvjeljeti when common word in Croatian, Slovenian and Serbian is cviliti form *kviliti, also word ćutjeti when common non ě word is ćutiti from *tjutiti and other similar words that came to being by necessitas metrica. Gorska vila (talk) 18:27, 28 December 2013 (UTC)

Yes it's like in *uksḗn: *korę < PIE *korēn, accusative *korenь < PIE *korenm̥. The form *korěnь (from SC and OESl.) is secondary from *korenь. --Ivan Štambuk (talk) 14:55, 1 January 2014 (UTC)

So are you saying it should be moved to *korę? --WikiTiki89 18:52, 1 January 2014 (UTC)

RFV of the etymology:

• "From Proto-Indo-European *bʰeikos. Cognates include Latin ficus and Old Armenian թուզ (tʻuz)"

I'm having a hard time figuring out how to derive σ from *bʰ (or թ from *bʰ, for that matter): even with palatization, I would expect *πτ, not σ. About the only thing that it has going for it is a possibility of a Semitic source or borrowing, as exemplified by Hebrew פג (“early fig”).

The etymology at Latin ficus raises the possibility of an aspirated dental via a "Mediterranean substrate form such as *thuiko-", which looks like it might very well work for all three. I suppose there's also, perhaps, the off chance of borrowing from something along the lines of Hebrew שִׁקְמָה (shikmá, “sycomore fig”).

I was tempted to just get rid of the PIE and use the etymology from ficus, but I thought I would run it by you folks first, just in case. Chuck Entz (talk) 07:27, 29 December 2013 (UTC)

Armeno-Greek borrowings from substrates is Hrach Martirosyan's favourite topic. And he has some recent publications about it. I'll add his views shortly. --Vahag (talk) 08:35, 29 December 2013 (UTC)

It's generally considered a non-IE Mediterranean substratum word (possibly Minoan or Tyrsenian or the like – something like tʰwiːk- or θyːk- would make phonetic sense), although I don't know if there is a good intra-Semitic etymology. Are you sure that the Hebrew word is not itself a borrowing from (Medieval?) Greek? --Florian Blaschke (talk) 03:48, 18 February 2014 (UTC)

שִׁקְמָה (shikmá, “sycomore fig”) and פג (“early fig”) go back at least to Biblical Hebrew, though the first is more solidly attested in the older parts. פג (“early fig”) is only found in the Song of Songs, which most scholars consider fairly late- but it still has to be at least a couple centuries BCE. That's not to say it couldn't be a common borrowing from some unknown third source- but the time frame is considerably earlier. Chuck Entz (talk) 05:07, 18 February 2014 (UTC)