Wiktionary:Etymology scriptorium/2016/October

discussion rooms: Tea room • Etym. scr. • Info desk • Beer parlour • Grease pit ← September 2016 · October 2016 · November 2016 → · (current)

---

искоусъ

In the etymology section of искусство (iskusstvo), this noun is said to mean "test, experiment"; but in its own entry, it is translated as "temptation". Is there some confusion between "attempt" and "temptation"? --Fsojic (talk) 13:31, 1 October 2016 (UTC)

Btw, does it have something to do with искать (iskatʹ, “to seek, to look for”)? --Fsojic (talk) 13:34, 1 October 2016 (UTC)

It's probably not a mistake, Serbo-Croatian doesn't have a perfectly analogous formation, but words built with kus-/kuš- range over these meanings: kušati (“taste, try”), iskušenje (“temptation”), pokus (“experiment”), iskustvo (“experience”)Crom daba (talk) 08:38, 2 October 2016 (UTC)

искоусъ (iskusŭ) has two meanings: "test" and "temptation". For the 2nd sense the Russian descendants are и́скус (ískus) / иску́с (iskús) and искуше́ние (iskušénije). --Anatoli T. ^{(обсудить}/^вклад) 12:10, 2 October 2016 (UTC)

There's nothing unusual about having those two meanings associated- it should be quite familiar to anyone who knows Biblical Greek, for instance. If you think about it, a temptation is something that tests one's character. Chuck Entz (talk) 16:11, 2 October 2016 (UTC)

Why must the Russian word be a borrowing from OCS, rather than an inheritance from PS? --WikiTiki89 18:08, 5 October 2016 (UTC)

paliativo

Is it correct to say that these words are "interfixed with -t-"? DTLHS (talk) 15:33, 2 October 2016 (UTC)

I'd say the -t- comes from the past participle. —CodeCat 15:37, 2 October 2016 (UTC)

catling

What does a knife have to do with a cat? Is there another etymology here? DTLHS (talk) 03:04, 5 October 2016 (UTC)

Catgut was formerly used in surgery; it may have something to do with that. —Aɴɢʀ (talk) 16:29, 5 October 2016 (UTC)

More likely would be an association with the sharp teeth and claws of cats. Another possibility: Catling and Catlin are both English surnames, so it might be named after someone. Chuck Entz (talk) 02:21, 6 October 2016 (UTC)

amo, Olav Hackstein claim needs source

(citation needed) Hillcrest98 (talk) 04:00, 5 October 2016 (UTC)

Celtic descendants of Reconstruction:Proto-Indo-European/bʰébʰrus

@Angr, Anglom, JohnC5 The Proto-Celtic descendant is also given as *bebrus, a u-stem. But is there actually evidence to point to a u-stem in any of the Celtic languages? —CodeCat 18:38, 8 October 2016 (UTC)

I don't think so. The word is barely attested in Celtic; mostly in names, many of which have derivational suffixes added to them. But nothing is really inconsistent with its being a u-stem either. —Aɴɢʀ (talk) 19:54, 8 October 2016 (UTC)

It seems to survive into modern Welsh, and there's also an Old Irish attestion as a name (going by what's listed on the PIE page). If the Brythonic descendants have a u-stem plural -ow, or if any u-stem endings survive for Old Irish, I think we have enough positive evidence. —CodeCat 20:27, 8 October 2016 (UTC)

I can't find the name in any entry in DIL, so I don't know what its genitive is. The plural of Welsh befer is beferod, and GPC says it's a loanword from Old English anyway. It's apparently not attested until the 14th century, and the attested spellings listed in GPC are befyr, befer, and befar. But Welsh doesn't usually spell its epenthetic vowels; if it were actually inherited from a PC *bebrus, we'd expect the spelling *befr to predominate. (On the other hand, it's not a very well attested word anyway; the usual Welsh word for beaver is afanc.) —Aɴɢʀ (talk) 22:01, 8 October 2016 (UTC)

I see. Are there even any attestations that unambiguously mean "beaver"? Otherwise, the Celtic descendant tree probably has no business being in the PIE entry. —CodeCat 22:36, 8 October 2016 (UTC)

I once thought about updating this PIE entry, but there are as many reconstructions as authors to discuss this word (*bʰébʰrus, *bʰébʰros, *bʰibʰrós, *bʰébʰr̥) with some claiming that the *u-stem represents a later adjective and others claiming thematicizations from an original *u-stem noun. BSl. and IIr. have *o- and *u- stems side by side in evidence, so, if there is no Celtic evidence for a *u-stem, I say drop it. —JohnC5 04:11, 9 October 2016 (UTC)

The closest thing to an attestation that unambiguously means "beaver" is the Gaulish place name Bebriacum, which Tacitus says means "locus castrorum". The next closest thing is Vulgar Latin *biber/beber, which is presumed to be a loanword from Gaulish (the Italic inherited word being fiber). Welsh befer and Old Breton beuer unambiguously mean "beaver", but they don't unambiguously come from Proto-Celtic as they could also be loanwords. And no, nothing in Celtic has to come from a u-stem. —Aɴɢʀ (talk) 09:44, 9 October 2016 (UTC)

Old Irish attá, Welsh taw etc

@Angr, Anglom, JohnC5 These are derived from the root *steh₂-, but I'm not sure what the Celtic form should be. In the entries themselves, *tāti, an athematic verb, is given as the ancestor, but on the page *steh₂- itself, the Celtic forms are given as a ye-present, *tāyeti. I don't know enough about how Irish or Brythonic developed to know if this is correct or not. In particular, what would happen to the -y-? —CodeCat 19:44, 8 October 2016 (UTC)

I'm not sure any Insular Celtic language would have different outcomes for *tāti and *tāyeti, and Matasović doesn't list any Continental descendants. He reconstructs it as *tā-yo- but doesn't say why it's that and not *tā-. —Aɴɢʀ (talk) 19:58, 8 October 2016 (UTC)

The only argument I could make is that the root verb was *stéh₂t in PIE, a perfective/aoristic verb. Could it be expected that Celtic created a present out of this just by replacing the endings? Germanic is known to have merged the imperfective/present and perfective/aorist classes, but Celtic seems to have merged the latter into the stative/perfect instead. So this would be an unusual exception to that process. —CodeCat 20:24, 8 October 2016 (UTC)

It could also have simply lost the reduplication of *stísteh₂ti since reduplication in Celtic marked the future and subjunctive preterite but not the present indicative. Our entry *steh₂- indicates that Balto-Slavic and Germanic both have verbs coming from *steh₂-ye-ti; maybe Matasovic reconstructs *tāyeti simply to bring it into line with those rather than for inner-Celtic phonological reasons. —Aɴɢʀ (talk) 22:26, 8 October 2016 (UTC)

Reduplication in the subjunctive? I thought the subjunctive was never reduplicated even if the future was? Anyway, what of my earlier question regarding the -y-? Would it not surface in any way? Are there any remnants of the athematic inflection that *tāti would have had presumably? In fact, would the lemma form itself not give Old Irish ·táith as the conjunct? —CodeCat 22:32, 8 October 2016 (UTC)

Sorry, I meant preterite, not subjunctive. And y disappeared between vowels, so I don't think it would surface in any way. The fact that the oldest form of the 1st person singular is ·táu, while táim appears only much later, could be taken as evidence of an original thematic conjugation, since *tāyū would have given ·táu, but *tāmi wouldn't. The conjunct forms come from PC forms with an early apocope of final -i, so *tāti > *tāt > ·tá (but also *tāyeti > *tāyet > ·tá, so it's no help). The absolute form *táith is attested only with suffixed object pronouns (táthum (“there is to me → I have”) etc.), but it could be equally well from *tāyeti as from *tāti. The 1st person singular is really the only thing I can think of that says "old thematic form". —Aɴɢʀ (talk) 10:00, 9 October 2016 (UTC)

From what I've seen, Proto-Celtic -t- appears as -th- after a stressed syllable, but often appears instead as -d- after an unstressed one. The 3rd person singular absolute ending -id is a good and relevant example of this. Since the -th- in táthum reflects this ending, it might be expected to be *tádum instead if the stem was originally bisyllabic. Does -th- appear in other hiatus verbs with suffixed object pronouns, too, or do they have -d-? —CodeCat 13:10, 11 October 2016 (UTC)

@CodeCat: Well, nóïd (“praise, extol”) is a hiatus verb; with suffixed pronouns in the 1st and 2nd person singular we get nóithium (“extols me”) and nóithiut (“extols you”). AFAIK the 3rd singular absolute ending -id only shows up as d before suffixed pronouns if it is synchronically preceded by an unstressed vowel. If the vowel before the /θ/ is syncopated, regardless of whether it's a hiatus verb or not, then the voicing to /ð/ doesn't happen: *beirith + i > beirthi (“(he) carries it”), but *sóerafaith + ut > sóerfudut (“(it) will free you”). —Aɴɢʀ (talk) 08:20, 15 October 2016 (UTC)

adannai

@Angr The etymology itself is sound, I think. But according to DIL, andaid doesn't seem to be attested very well, if at all, and I'm not sure if the attestation is Old Irish or Middle Irish. Does this indicate that the base verb was already pretty much no longer used by the time Old Irish was written? When did the compounding actually take place? —CodeCat 21:17, 9 October 2016 (UTC)

The sole citation for andaid in DIL is from An Leabhar Breac, which is Middle Irish. That could mean that andaid is a back-formation from ad·annai, or it could mean the unprefixed version was always extremely rare, and almost by accident the one time it did get attested it happened to be in a late text. —Aɴɢʀ (talk) 15:11, 10 October 2016 (UTC)

What do you suggest we put in the etymology for ad·annai? We can't link to andaid if it's not attested in Old Irish, a red link implies that an entry needs to be created. Do we need a reconstructed *andaid instead? To make a reconstructed entry would imply that we suppose it existed, but just isn't attested, which isn't really true either. I think all the evidence points to andaid never having existed as an independent verb in Old Irish, but that there is an ancestor somewhere along the line where it did still exist. I'm just not sure what that ancestor would be. Primitive Irish? Proto-Celtic?

An important point to consider is that when ad·annai was first created, both the base verb and derivative existed side by side (this is a logical necessity). If we reconstruct *adandāti for Proto-Celtic, there'd need to be some evidence that it was already formed in Proto-Celtic, I presume. But if it's not Proto-Celtic, then it would have to be Primitive Irish. How would we write a reconstructed Primitive Irish ancestor of ad·annai? Do we use Ogham or Latin?

I've also seen a bunch of entries such as ·icc and ·muinethar, which act as base entries for derivatives, but are not actually attested terms. We had a similar case with Russian verb entries, where the base verb didn't exist but we did have an entry for it anyway. I still have my reservations about this approach, since e.g. ·icc does not, in itself, meet WT:CFI. —CodeCat 12:51, 11 October 2016 (UTC)

Actually, maybe we can create andaid for Old Irish: finite forms aren't attested, but the participle andithe (Ml. 56d3) and the verbal noun andud (Ml. 131d14) are. As for the conjugation class, the final vowel of ad·annai strongly suggests it was *andīti. VKG (I, 457) says, "Das Verbum war wohl ein -ī-Stamm mit zum großen Teil verlorener Mouillierung" (Mouillierung being Pedersen's word for palatalization). As to the origin of *andīti itself, I have no idea. Matasović doesn't list it, and Pedersen's etymology is wild speculation. As for ·icc and ·muinethar, I see your point, but I wasn't sure where else to put their etymologies without redundancy. —Aɴɢʀ (talk) 09:43, 12 October 2016 (UTC)

Words for fist

Very very messy.

pugnus... says that it's from PIE *pewǵ-, *pewḱ-. Okay. But then it says "near cognate of πυγμή". "Near cognate" is very confusing, whether if it's not cognate or if it is cognate but formed differently.

πυγμή (pugmḗ) postulates it to be from *puǵnos or similar. Here's where the real trouble begins: it says that it's cognate to fist.

But fist says it comes from *pn̥kʷ-sti-s < *penkʷ- , so it isn't cognate after all. Hillcrest98 (talk) 00:03, 10 October 2016 (UTC)

It looks like 2 possible and competing etymologies are being shown respectively. Perhaps we should show both (--if each is still valid to-date) at all entries Leasnam (talk) 15:24, 10 October 2016 (UTC)

The only thing the other words have in common with fist is the first letter of their PIE reconstruction, a nasal (that seems to be in different parts of the morphology) and a rather tenuous semantic association. As for *pewǵ-/*pewḱ- and *puǵnos, the latter looks like the zero grade of the former with a noun suffix, so pugnus and πυγμή (pugmḗ) would then be be the same root with different noun suffixes in their PIE reconstructions. Chuck Entz (talk) 15:53, 10 October 2016 (UTC)

falcinellus from falx

I just wanted to create falcinellus as mult or neolatin for Plegadis falcinellus and Limicola falcinellus, as a diminutive of falx but realised that falx is feminine. Isn't this contradictory? Derivates into an adjective? Sobreira ►〓 (parlez) 11:06, 11 October 2016 (UTC)

Seems like it's being used like an adjective, given that both Plegadis and Limicola are both masculine. Hillcrest98 (talk) 15:22, 11 October 2016 (UTC)

-ware

-ware#Old English says

From Proto-Germanic *warjaz (“dwellers of”). Cognate with Old High German -āri ("inhabitants of"), Old English wer "man". More at wer

but *warjaz says

Derived from *warjaną (“to hinder, defend”).

Lysdexia (talk) 11:34, 12 October 2016 (UTC)

The origin of agent enclitics

• -er < -ere < -arius < *-aricos < *-laricos < *-laris < laris < lar
• -tor < -ter < *-atta < atta

Do you agree? Therefore mother and father mean *meh₂d- (“moistener”) and *peh₂- (“feeder”), so their roots come from verbs like their suffixes imply rather than from baby calls. Lysdexia (talk) 12:20, 12 October 2016 (UTC)

Well there is a suffix involved in PIE words for mother and father (not to mention sister and daughter), but I don't think these are from those verbs. Rather, the suffix probably has to do with these being terms for family members. As for your Latin derivations, those are pretty much nonsense. For instance, atta goes back to PIE basically unchanged. It's not enough to have the pieces- you need to know the rules for putting them together. Chuck Entz (talk) 14:48, 12 October 2016 (UTC)

Is that nonagent suffix published somewhere? Brother should mean bearer; sister is *swésōr (“exceptional”); daughter should mean suckler. Wiktionary already has the suckler conjecture. Nonsense there should be doublets for old words? You put too much faith into regular sound shifts. -ather doesn't heed the Great Vowel Shift. Lysdexia (talk) 03:35, 13 October 2016 (UTC)

Wow... I just... wow. You should read the work of Edo Nyland. I think you'll find his conjectures similarly well-researched and well-constructed. —JohnC5 04:12, 13 October 2016 (UTC)

frol#Portuguese

@Ungoliant MMDCCLXIV, Daniel Carrero and anyone else interested in Portuguese historical linguistics: The etymology of frol says this is a metathesis of flor, from Latin flōrem, but since /l/ regularly becomes /r/ after labials in Portuguese and its closest relatives (e.g. praia from plagia and branco from *blancus), isn't it more likely that it's a dissimilation of fror, which is the expected form? The modern standard form flor would then either also be a dissimilation (in the opposite direction) or a learned re-borrowing of the Latin. —Aɴɢʀ (talk) 16:20, 12 October 2016 (UTC)

@Angr both derivations are mentioned in my sources. I’ll amend the entry. — Ungoliant ^(falai) 20:57, 15 October 2016 (UTC)

badb

@Angr I find the etymology a bit doubtful here. My experience is that old IE languages tend to derive the names of their gods from the things they are associated with, not the other way around. Compare *Þunraz, Aurora etc. —CodeCat 17:05, 12 October 2016 (UTC)

Well, the oldest form of the word appears to be bodb, which Matasović takes back to *bodwos, *bodwā. The only other Celtic cognate he mentions is Old Breton bodou (“heron”). He connects the Celtic word with Proto-Germanic *badwō (“battle, fight”), saying "the crow is the bird symbolizing the carnage in battle". If 'battle, fight' is also the oldest meaning in Celtic, then the goddess's name could come directly from that, and the meaning 'crow' could come either directly from 'battle, fight' or from the goddess's name. The latter seems a little more plausible to me. —Aɴɢʀ (talk) 17:16, 12 October 2016 (UTC)

karyorrhexis

RFV of the etymology.

A dispute over this etymology was brought to my talk page, which I'm bringing here because it raised some tricky theoretical considerations. The current etymology says it comes from English karyo- prefixed to English rhexis. This is looks wrong, because it doesn't explain the double r. It's also suspect because the term may not have been coined in English: the earliest Google Books hit is in a 1910 French book title.

Let me explain my understanding of how these scientific coinages work: the coiner either

1. takes Ancient Greek morphological elements and combines them using Ancient Greek morphophonological rules, then converts the result to Latinized Greek (Ancient Greek κάρυον (káruon) + Ancient Greek ῥῆξις (rhêxis) →Ancient Greek *καρυόρρηξις (*karuórrhēxis) → Latin karyorrhexis)
   or
2. takes Latinized Ancient Greek morphological elements and combines them into a Latinized Greek term using morphophonological rules borrowed from Ancient Greek (^{Ancient Greek κάρυον (káruon)→} Latin karyon + ^{Ancient Greek ῥῆξις (rhêxis)→}Latin rhexis→^{Ancient Greek *καρυόρρηξις (*karuórrhēxis)→}Latin karyorrhexis)
   or
3. takes Latin morphological elements derived from Ancient Greek and combines them into a Latin term using special morphophonological rules (Latin karyo- + Latin rhexis → Latin karyorrhexis)
   or
4. takes morphemes in the language of the coiner derived from Latinized Ancient Greek and combines them into a term in the language of the coiner using special morphophonological rules (English karyo- + English rhexis → English karyorrhexis)

The last 2 models are difficult to apply here because they would require some mechanism for knowing when to add the extra r, while the first 2 are simply using the well-known assimilation in Ancient Greek of ν (n) to a following ρ (r). Nonetheless, I suspect that the trend among coiners of terms as a whole is toward progressing down the list as Classical literacy declines within the scientific community.

There's also the matter of back-formation and borrowing causing contamination between the models: for instance, I believe there are cases where a term was coined in French, and someone else converted the French term into its hypothetical Ancient Greek form to use it in coining a new term using one of the first two models.

My hunch is that this etymology probably uses the second model, because Ancient Greek κάρυον (káruon) really means nut (the modern concept of cells with nuclei was unknown to the ancient Greeks).

Is there any way we can incorporate this into our etymologies? Chuck Entz (talk) 22:12, 14 October 2016 (UTC)

Did a pretty simple wording of your second model. Hillcrest98 (talk) 22:43, 14 October 2016 (UTC)

Other terms from entries in Category:en:Medicine ending in rhexis: capsulorrhexis, amniorrhexis, keratorhexis (keratorrhexis more common at Google books), enterorrhexis.

capsulorrhexis does not follow any of your models as Latin capsula "small box. casket" is the likely source of capsulo=. Keratorrhexis does not follow them very closely.

The possibility exists that the formation was more imitative than rule-following. LSJ contains entries for 8 words (6 attested in the Perseus corpus) ending in -ρρηξις, all of which are preceded by prepositional prefixes. All of the English medical terms are prefixed by biological nouns, suggesting that their formations are much more recent, but possibly in Medical Latin, except for those that would have required a microscope to be conceivable, eg, karyorrhexis. As it is a late-comer, imitation of the pattern of the preceding terms terms ending in rrhexis seems quite plausible.

Given the abundance of medical literature in several modern languages in the 19th century, the origin could have been in any of the leading languages. MW3 uses the label ISV ("International Scientific Vocabulary"), perhaps to avoid the sometimes contentious assignment of a term to origin within one of those languages. DCDuring TALK 00:47, 15 October 2016 (UTC)

Reconstruction:Proto-Germanic/skukkōną

Discussion moved from WT:Requests for verification#Reconstruction:Proto-Germanic/skukkōną.

Can somebody verify and/or cleanup the following related terms:

• shog
• shock
• choquer

All four(+proto tree) pages seem to claim a different origin of the same words. I can't correct these myself since I can't tell which ones are right or wrong, or if both are correct. --Eliot (talk) 21:40, 14 October 2016 (UTC)

Latin toponyms from Iberian

@Metaknowledge, Isomorphyc, I'm so meta even this acronym: Thanks to the excellent work of @Samubert96, the set of Latin toponyms from Iberia containing brīga has come to my attention. It is quite productive, with Augustobrīga, Centobrīga, Flāviobrīga, Iūliobrīga, Lacobrīga, Lambrīca, Latobrīgī, Medobrēga/Medubrīga/Mundobrīga, Nertobrīga, Nitiobrīgēs, Rubricātus/Rubricātum?, Sēgō̆brīga representing a portion of its compounds, but I was wondering about its origin.
This source would propose a Basque origin of bri, vri, uri (“peopled”) + -ga (negative particle?; locative particle?), but I have not the experience to evaluate this claim. The geographic distribution seems a bit wide to me to be Basque.
This Spanish wiki page goes into some detail about the etymology and examples, also pulling the more northerly examples of Brigantēs and Brigiānī. My Spanish is not great, so I'm having trouble fully evaluating the argument, but I think the proposed origin is Celtic from *brixs from *bʰerǵʰ-. This seems better than the Basque but does not explain the long ī that appears in most forms. So my questions would be whence the ī and should we create a Latin entry -brīga (Celtic toponym-forming suffix)? —JohnC5 19:13, 16 October 2016 (UTC)

The Basque thing is silly. I don't know how the i got long, though, given that *brixs is the only reasonable etymon. —Μετάknowledge^{discuss/deeds} 19:37, 16 October 2016 (UTC)

Post–edit-conflict:
@JohnC5: Geiriadur Prifysgol Cymru cites the Gaulish placenames Nemeto-briga and Nerto-briga (≟ Nertobrīga) as relations in its entry for bre¹ (“hill, hillock, mountain, hill-country, upland, peak”), which *brixs lists as a Welsh descendant, all of which is supportive. I have no idea whence the ī; what is the evidence for that quantity? Re creating an entry for -brīga, is there any evidence that -brīga was ever a Latin suffix, or was it just a Celtic one? Either way, it looks like a suffix entry (in at least one language) is warranted. — I.S.M.E.T.A. 19:41, 16 October 2016 (UTC)

@I'm so meta even this acronym: I do not know the evidence for the length. Perhaps the length may be educed from the development of the Romance descendants? Formations like Augustobrīga, Caesarobrīga definitely seem like Roman formations. —JohnC5 20:00, 16 October 2016 (UTC)

@JohnC5: L&S offer the following perplexing set of quantities: Jūlĭōbrĭga, Lacobrĭga, Sēgŏbrīga (Σηγόβριγα), Centobrīga, Nertobriga; I haven't checked the others, but the short vowel in Lacobriga differs from Gaffiot's long reading. There are certain confusions in the Spanish Wikipaedia article: it attempts to derive Germanic -burg from a different IE etymon than -briga, from *bʰergʰ- (“protect”), as in bury, instead of from *bʰerǵʰ- (“high place”). I don't believe this is very justified. It also offers a dubious Greek folk-etymology for Πρεττανία (Prettanía, “Britannia”) while making the obvious point that Britannia indeed is not related to -briga. These issues aside, the catalogue of city etymologies seems reasonable. I feel the Basque theory would need to be more concrete to justify citing. Since the oldest attested Celtic language is from the eighth century AD, given the diversity of the Iberian peninsula in classical times, the long /ī/ could have come into Latin from anywhere. I have trouble understanding the modern dissent, because I am not sure how Latin quantities can be distinguished, even in some of these cases by Romance descendents. If it helps at all, I have made a catalogue of dictionary entries for words in -briga: here. Regretting this is not very conclusive, I broadly agree with your reading of the sources. It might be possible to question the Latinity of the suffix, but I would favour being catholic about it.

I will add here, in case I forget, once you add this suffix, it will also be possible to update the etymology of Coimbra. It feels so useful when we have something interesting to add to a modern etymology. Isomorphyc (talk) 23:30, 16 October 2016 (UTC)

We can tell the length if we know where the accent falls in Romance, since it is contrastive in this case (Conimbrī́ga vs. Conímbrĭga). If I were better with Spanish etymology, I could maybe tell better. —JohnC5 00:37, 17 October 2016 (UTC)

@Metaknowledge: Seeing as you speak Spanish, can you help at all with this? @JohnC5: Are any of these -briga places located outside areas that contemporarily spoke Celtic languages? @Isomorphyc: N.b. Coimbra's Spanish name, Coímbra. — I.S.M.E.T.A. 10:25, 17 October 2016 (UTC)

I was going to mention the Spanish name; also note the quasi-Latin name of the archaeological site, Conímbriga. What puzzled me is whether the names with dictionary quantities (which do not apparently include Conimbriga) are the same as the names with Romance descendents. Isomorphyc (talk) 12:08, 17 October 2016 (UTC)

I'm a little confused on what I could do to help, unless you need me to translate something from Spanish. I bet Coímbra is borrowed from Pt. Coimbra (or from OPt.), and I don't know whether Conímbriga is correctly formed, but if it is, that means the i is short. But maybe this vowel quality just varied from place to place. —Μετάknowledge^{discuss/deeds} 19:48, 17 October 2016 (UTC)

@JohnC5, Isomorphyc, Metaknowledge: I used the numerous foreign-language editions of en:w:Coimbra to add all the translations of Coimbra that I could to the entry. I note that not even one has retained the -ig- of the Latin Conimbriga. I read somewhere that French words descended from Latin drop any syllable after the one that's stressed (hence the view that all French words are stressed on the ult); accordingly, Coïmbre is consistent with Conímbrĭga, whereas Conimbrī́ga would suggest something like *Coïmbrigue (AFAIK). Does the same rule of derivation exist in other Romance languages? The issue, of course, is that most of those translations are probably just borrowings or transliterations of the Portuguese Coimbra, and not terms inherited from the Latin. I assume it's too much to hope that these -briga toponyms turn up in poetry, where we could simply scan the quantity… — I.S.M.E.T.A. 20:07, 17 October 2016 (UTC)

w:Coimbra#Early history says that the Latin Conimbriga became Colimbria over time and, when the city was captured by Muslim forces in 714, it became قُلُمْرِيَة (qulumriya, “Qulumriyah”). Does that help in any way? — I.S.M.E.T.A. 20:14, 17 October 2016 (UTC)

(e/c) Note that there is an intermediate form Colimbria (I think in Medieval Latin, but I'm not sure). Compare Arabic قُلُمْرِيَّة (qulumriyya). --WikiTiki89 20:24, 17 October 2016 (UTC)

I'll also point out w:Segorbe as the descendant of Segobriga, which again looks more like Sēgóbriga > Linguistics stuff > Segobre > Segorbe. —JohnC5 03:46, 18 October 2016 (UTC)

@JohnC5: I created an entry for Segorbe and added all the translations, like I did with Coimbra; notable are the Aragonese Segorb and the Catalan Sogorb. The Valencian name is also Sogorb, but, for some reason, Valencian doesn't have an ISO 639 code (maybe it's too similar to Catalan). — I.S.M.E.T.A. 13:17, 18 October 2016 (UTC)

@I'm so meta even this acronym: Thanks for all this work. According to WT:LL, we consider Valencian a dialect of Catalan. —JohnC5 14:12, 18 October 2016 (UTC)

@JohnC5: So do most people, it seems. And you're most welcome. — I.S.M.E.T.A. 14:36, 18 October 2016 (UTC)

@I'm so meta even this acronym: Thanks very much for your extensive work. The article at w:la:Conimbriga offers two further Latin variants: Conembriga and Conimbrica with the interesting history that the modern city was named after the episcopal see in the 11th century, which in turn was named after the archaeological city, ten miles away, which had been destroyed in the fifth century. Sorry I don't have anything linguistic at add at this point. Isomorphyc (talk) 22:56, 18 October 2016 (UTC)

@Isomorphyc: You, too, are welcome; I often find myself in the mood to do that kind of grunt work. :-) Thanks for noting Conembriga and Conimbrica; they weren't mentioned in the English edition of the Wikipedia article. — I.S.M.E.T.A. 13:06, 19 October 2016 (UTC)

@CodeCat, KarikaSlayer, do y'all have an opinion or know Romance editors who could help out us poor Latinists? —JohnC5 18:56, 18 October 2016 (UTC)

The derivation from *brixs is sound, I would say. I don't have any immediate answers for the vowel length question, but it's possible that the long vowel arose as a consequence of the stress. In Celtic, the stress would have been word-initial, but with secondary stress on every odd-numbered syllable, so in Celtic the -rig- syllable would have been somewhat more stressed as well. Stress on the stem vowel, i.e. -óbrig-, was unheard of in Celtic. Therefore, it's possible that the Romans, hearing the Celtic name, borrowed the stress with it, so that the -rig- syllable remained stressed. However, to account for Latin stress patterns, this necessitated making the vowel long. —CodeCat 19:07, 18 October 2016 (UTC)

How early are some of these words attested? It would make a significant difference if we're dealing with Classical Latin vowels vs. a Vulgar Latin speaker attempting to fit a loanword with irregular stress into the CL stress system, which is what I'd suspect. KarikaSlayer (talk) 20:06, 18 October 2016 (UTC)

@KarikaSlayer: The earliest half dozen are in Ptolemy (in Greek) and Pliny; you can see the L&S citations in my list: User:Isomorphyc/Sandbox/briga. Giulius Caesar also has Magetobriga in DBG-1, though for some reason it is not in any of the dictionaries, I think. There are a few more citations in the Antonine Itinerary a few centuries later, but the vast bulk are mediaeval I believe. Thanks to everyone for the extensive work here. Isomorphyc (talk) 22:37, 18 October 2016 (UTC)

Merfyn

@Anglom, Angr, CodeCat, JohnC5, Victar, I'm adding some pages for Merfyn, and its Proto-Brythonic and Proto-Celtic forms. While I found "mer" (marrow) comes from PC "smeru", the second element supposedly means famous, but I can't find any Proto-Celtic words resembling it. Working backwards from Welsh I guessed Smeruminos or Smerumindos, which gave Proto-Brythonic Mɨrβ̃ɨn or Mɨrβ̃ɨnn, so the affection was a problem there. The best I could come up with was Smerumenjos, which gives Merβ̃ɨn and then Welsh Merfyn. Though even with that, which seems correct, I can't find any corresponding *menjos. In fact, I saw *menjos mentioned as meaning "mountain".

Secondly, what about the Welsh name Cado? The easiest explanation would be "Katu" (battle) and "φo" (from), giving Katuφo, Kado and Cado. But sometimes in Welsh his name is given as Cadwy, which complicates things. UtherPendrogn (talk) 17:10, 17 October 2016 (UTC)

I don't know anything more about the etymology of Merfyn than what it already says at Mervyn#Etymology. Besides "marrow", the first element could also mean "lake". I can't find any Welsh word myn that means "eminent, famous", just words that mean "kid, young goat", "will, wish, desire", "crown, wreath of flowers". But what makes you think this even has Proto-Brythonic and Proto-Celtic forms? Maybe it was coined in Welsh. As for Cado/Cadwy, it's pretty much inconceivable that a name would be a compound of which the second element is a preposition. It's probably a clipped form of some name or name beginning with cad (“battle”), like Cadogan/Cadwgan or Cadwaladr. If it is a full name on its own, all I can think of is cad + gwy (“water, liquid”) or ŵy (“egg”), but neither "battle water" nor "battle egg" is really the sort of thing ancient Celts would be likely to name their sons. —Aɴɢʀ (talk) 21:33, 17 October 2016 (UTC)

Battle Water is definitely possible, but not Battle Egg.

And the first Merfyn is mentioned as being alive in the 570's, which would make his name back then Proto-Brythonic and not Old-Welsh. UtherPendrogn (talk) 05:38, 18 October 2016 (UTC)

According to Old Welsh, the 570s is after Welsh separated from Proto-Brythonic; it belongs to the period known as Archaic Welsh or Primitive Welsh. —Aɴɢʀ (talk) 13:58, 18 October 2016 (UTC)

auger

Is the metanalysis explanation that I just added right? --Fsojic (talk) 21:11, 17 October 2016 (UTC)

It's right. But our category is called Category:English rebracketings, so maybe we should use that term in the etymology too, for consistency. —CodeCat 21:16, 17 October 2016 (UTC)

Thanks. I hadn't even noticed that the category was already at the bottom of the page. --Fsojic (talk) 21:23, 17 October 2016 (UTC)

Talk:پنداشتن

DTLHS (talk) 23:48, 18 October 2016 (UTC)

The IP is correct. I changed the etymology with references. --Vahag (talk) 08:02, 19 October 2016 (UTC)

aktuelle begivenheder

Someone was annoyed when I RFVed some terms under the same header, so now these get a section each.

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

Is there a reason these are all specifically Danish calques from English? —JohnC5 00:58, 20 October 2016 (UTC)

Only that I encountered several such calques that seemed doubtful, so I went through Category:Danish terms borrowed from English and took the ones I doubted over here. Is there a problem with my request?__Gamren (talk) 14:30, 20 October 2016 (UTC)

There's no problem. I was just curious! —JohnC5 15:16, 20 October 2016 (UTC)

åbent forhold

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

begivenhedshorisont

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

curiepunkt

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

curietemperatur

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

dødvande

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

kulturchok

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

kulturrevolution

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

I have German Kulturrevolution attested in 1845 and 1892 (both spelt "Culturrevolution"). There are dozens of attestations from the 1950ies. So I suppose that English had the word before 1966, too. If not, its having spread the word becomes very unlikely. Kolmiel (talk) 00:19, 3 November 2016 (UTC)

And Ngram Viewer has "cultural revolution" (case-insensitive) from before 1966, too. With words like these, how does one go about determining whence (and whether) it has been calqued?__Gamren (talk) 15:53, 3 November 2016 (UTC)

It would seem very likely that they are calques. Such words can be formed independently, of course, but those should be isolated attestations (as the 19th-century ones may be). What one would do is probably to check when and in which language there's the earliest cluster of attestations, implying that it had become a "term". Maybe it was used in some classic, or by a party... This may be very difficult and time-consuming. Candidates would be German, Russian, also English, French, and maybe other languages. Kolmiel (talk) 17:51, 4 November 2016 (UTC)

væggene har ører

I ask for sourcing of this etymology.__Gamren (talk) 15:50, 19 October 2016 (UTC)

rapper

Does the "sword" sense also derive from rap or is it a variant of rapier? --Florian Blaschke (talk) 03:12, 20 October 2016 (UTC)

ὀξύς

Is the etymology given at ὀξύς (oxús) onto anything at all? The inherited cognate seems to be ἄκρος (ákros), so the former would have to be either pretty heavily reformed, or transmitted through something else. --Tropylium (talk) 03:31, 23 October 2016 (UTC)

The contributor who added the etymology is well known for not letting a lack of background in historical linguistics or ancient languages keep them from adding huge volumes of etymologies. They consult lots of references, but every once in a while they trip up on obvious details that the references don't bother to explain (for instance, that the Macedonian in Ancient Greek etymologies isn't the modern Slavic language). I think this is one of those cases.

It looks to me like this goes back to the same root, but that's about it. Just a hunch, but judging from the etymology at Latin acuō, Latin acus seems to have started out as a very similar adjective that must have lost out to Latin ācer, though someone who knows more than I do would have to explain the s in the Greek. Chuck Entz (talk) 05:04, 23 October 2016 (UTC)

Beekes does not think there are sufficient IE comparanda to support an *o-grade *u-stem which would be required here if *h₂eḱ- were the root. Apparently Narten also reconstructed *h₃eks- based on अक्ष्णुते (akṣṇute, “mark the ear of cattle”), but Beekes thinks this too speculative. —JohnC5 00:36, 24 October 2016 (UTC)

Yiddish ביק (bik, "bull")

Can we safely derive this from Proto-Slavic *bykъ or is there an umlauted version of the *bukkaz (germanic buck word) with divergent semantics it could come from? Crom daba (talk) 15:16, 24 October 2016 (UTC)

Considering the plural of ביק (bik) is ביקעס (bikes), it is most likely a loanword, and likely a relatively recent one. Although that oddly leaves Yiddish without a traditional word for "bull". --WikiTiki89 15:25, 24 October 2016 (UTC)

Maybe there's some descendant of *steuraz in early Yiddish that got replaced by ביק (bik). It's not really that odd for that sort of thing to happen. —Aɴɢʀ (talk) 15:59, 24 October 2016 (UTC)

I only knew of ביק (bik), which is definitively Slavic, but I see there's also בוהײַ (buhay), which I don't think I've heard before. —Μετάknowledge^{discuss/deeds} 17:43, 24 October 2016 (UTC)

What a peculiar word. It sounds so un-Yiddish. It's reminiscent of buey; I wonder if it's from Romance. Do you or Wikitiki89 have access to a Yiddish etymological dictionary? —Aɴɢʀ (talk) 18:13, 24 October 2016 (UTC)

It's from Ukrainian or Belarusian бугай (we currently only have a Russian entry), itself from Turkic. --WikiTiki89 18:15, 24 October 2016 (UTC)

Polish also has buhaj, currently undefined. --Anatoli T. ^{(обсудить}/^вклад) 19:38, 24 October 2016 (UTC)

pl:buhaj gives the odd etymology of Old Polish > Ukrainian > Polish. Why would the Ukrainian word have come from Old Polish? --WikiTiki89 20:27, 24 October 2016 (UTC)

I don't know. Apparently, Ukrainians change g to h but the original Polish word "bugaj" had g. Russians changed h back to g :) (I haven't checked the etymology, not sure if it's Slavic or Turkic) Polish hulajnoga (“scooter”) must of Ukrainian origin, even if Polish has gulać. --Anatoli T. ^{(обсудить}/^вклад) 21:07, 24 October 2016 (UTC)

The word is safely Turkic I'd say, although the -y ending isn't present in any of the numerous borrowings into various languages other than East Slavic nor is it attested in Turkic reflexes. But ESTJa explains it as parallel to formations like Crimean Tatar babay (“dad”), anay from baba (“father”), ana (“mother”). Doerfer has a nice list of borrowings of Proto-Turkic *bukā and their trajectories at page 299 of "Türkische und Mongolische elemente im Neupersischen" tome 2. He also leaves open the possibility of Proto-Slavic *bykъ being a Bulgharic loan or the Turkic word being Indo-European so בוהײַ (buhay) and ביק (bik) might turn out to be doublets (even though it's too speculative for a mainspace entry). Crom daba (talk) 21:59, 24 October 2016 (UTC)

Obviously it makes sense that the Modern Polish word came from Ukrainian, I was only questioning that the Ukrainian word would have come from Old Polish. It's probably just a mistake. As far as the more distant etymology, I think it's likely that *bykъ has some connection to *bukā, but I guess it's hard to tell exactly what that connection is. --WikiTiki89 19:35, 26 October 2016 (UTC)

• See also Hungarian bika (“bull”), which is apparently traced to Turkic roots. ‑‑ Eiríkr Útlendi │^{Tala við mig} 20:53, 26 October 2016 (UTC)

-is#Irish

@Embryomystic: are you sure about the etymology of -is? I'm not aware of an Old Irish -as that forms language names, though to be honest I'm not sure how language names are formed in Old Irish at all. I always sort of assumed that, just like English -ese, Irish -is was a borrowing from Latin -ēnsis. —Aɴɢʀ (talk) 09:45, 25 October 2016 (UTC)

I'm not sure off the top of my head where I got that from, but I do remember being surprised at the time. It may be a Middle Irish thing rather than Old Irish; I'm not always sure where one ends and the other begins, which isn't helped by the fact that even some of the Old Irish texts we have date to a period when people were more likely speaking what one might call Middle Irish. Let me see what I can find, anyway. I wish I had been more diligent about referencing some of this stuff when I made the original edits. embryomystic (talk) 15:52, 25 October 2016 (UTC)

Okay, I've had a look, and I think I may have been overextrapolating from Old Irish (or, as I say, possibly Middle Irish) Bretnas, modern Breatnais. There are a few other examples of languages using a form of this suffix in the DIL, but they're all using -is rather than -as. As for how language names are formed in Old Irish, it's a bit more chaotic, it seems; either they're used with some unsuffixed form of an ethnonym, or in some cases, they seem to be formed with -da. embryomystic (talk) 16:50, 25 October 2016 (UTC)

Reconstruction:Proto-Balto-Slavic/glāˀdus

I find the claim that the root derives from *gʰelh₂- rather dubious, sourcing or not. No explanation is given for why -dʰ- was affixed to this root, what kind of morphological element this is and what it means. The random swapping of the vowel and l is not explained either.

In general, I am very skeptical of claims that derive roots from other roots. The method of derivation is often ad hoc, left completely unexplained, with no parallels, and sometimes the existing root is even "rebuilt" by dropping consonants. That is not the case here, but it's still just like "Yeah, here's the derivation, with some random change to the existing root. There's no parallels, and no explanation, I just pulled it out of my hat, and you'll just have to believe me." That it's sourced doesn't make it any less dubious, because the source does no better in explaining it. I'm led to believe that linguists (and Wiktionarians) are all too eager to make unqualified claims about relationships just so they can make things all connected. —CodeCat 23:46, 29 October 2016 (UTC)

This kind of root derivations might be best left for PIE entries themselves, unless the derivational relationship has remained opaque. There is no obligation to trace things to the morphologically simplest component (and it might be preferrable to give full PIE terms, where reconstructible).

Derksen doesn't seem to give the root etymology, BTW, so it's not even sourced in this case. --Tropylium (talk) 00:19, 30 October 2016 (UTC)

I may be misunderstanding you, but are you saying that when the derivation is opaque, it should be included in other entries, but when it's not opaque, it should only be in the PIE root entry? —CodeCat 00:25, 30 October 2016 (UTC)

Sorry, mixing up my words: transparent and not opaque. E.g. *-trom has relatively regular descendants, and we indeed note at e.g. ἄροτρον (árotron) that it remains analyzable. By contrast, in cases like the current *glāˀdus, pointing to *gʰleh₂dʰ- should be enough. --Tropylium (talk) 01:42, 30 October 2016 (UTC)

Maybe we should categorize PIE roots by extensions and see if some pattern emerges. I think I recall a paper by Lehmann in which he claimed that the -dʰ- root extension had passive meaning in Germanic, although it doesn't seem like it could be applied in this case. Crom daba (talk) 03:14, 30 October 2016 (UTC)

中東

RFV of the etymology. ばかFumiko￥^talk 08:55, 30 October 2016 (UTC)

Reconstruction:Proto-Germanic/glandaz

I don't think a single Saterland Frisian form is enough evidence to reconstruct a Proto-Germanic form. Victar had formerly listed Old High German and PIE forms too, but the OHG form didn't fit the High German consonant shift, and the PIE form didn't fit Grimm's law, so they're highly dubious at best and I've therefore removed them. —CodeCat 17:26, 30 October 2016 (UTC)

Old Frisian gliand, gland (“glowing”), a participle form. This would just hearken back to PGmc *glōandz wouldn't it ? Leasnam (talk) 20:25, 31 October 2016 (UTC)

Curiously though there is also PGmc *glantaz (“glowing" "shining”), whence OHG glanz with the same meaning. Could Ofs gland then be some sort of conflation of the twain ? Leasnam (talk) 20:27, 31 October 2016 (UTC)

• Any chance Proto-Germanic *glantaz meant something closer to shining than glowing? The phonology calls to mind English glint. ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:15, 1 November 2016 (UTC)

Ooops, yes, "shining". it means shining, not "glowing" Leasnam (talk) 04:40, 6 November 2016 (UTC)

clumsy

Could someone add the etymology? None of the words mentioned on etymonline has an entry here. --Fsojic (talk) 17:36, 30 October 2016 (UTC)

I did one on the fly, and created the related entry. I'll tweak it later when I have a spot more time Leasnam (talk) 20:51, 31 October 2016 (UTC)