Open main menu

āter: length in related termsEdit

Is it correct that some of these related terms have a short first vowel? Even if atrōx shortened for some reason, shouldn't this then apply to ātrōcitās?

--Hiztegilari (talk) 12:24, 1 June 2016 (UTC)

I'd have to double-check to be 100% sure, but I suspect that the vowel is long in all these words, and that the words that don't show a long vowel were taken from dictionaries where vowel length isn't shown before consonant clusters. Some Latin lexicographers care only about vowel length with respect to poetry scansion, so they don't bother marking vowels before clusters since all vowels are "long by position" there. —Aɴɢʀ (talk) 12:34, 1 June 2016 (UTC)
An annoying habit for sure. —CodeCat 13:20, 1 June 2016 (UTC)
Lewis & Short mark atrox, atrocitas, and atrociter with ā̆, suggesting that it sometimes scans long and sometimes short. That's only expected for inherently short vowels before "muta cum liquida" clusters, suggesting that these are actually ătrox etc., which in turn suggests either that atrox and āter are unrelated or that atrox comes from a different ablaut grade (e.g. āter could be from full grade *h₂eh₁tro- and atrox from zero grade *h₂h₁tro-). —Aɴɢʀ (talk) 13:47, 1 June 2016 (UTC)
De Vaan says the vowel of atrōx is short, though he does try to connect it to āter by looking for an explanation for the short vowel. The hypothesis he gives (by Schrijver) he calls "somewhat cumbersome". —CodeCat 13:53, 1 June 2016 (UTC)


Is there anything more modern than Pokorny to confirm that all these terms do, in fact, come from this adverb? —CodeCat 13:55, 2 June 2016 (UTC)

Proto-Germanic *alīnō/alinōEdit

Was wondering about this when I was about to create a reconstructed entry for this, which would be the lemma - with a short or long i? Don't currently have access to my uni library so I cannot validate this with external sources. Given the diphthong in 𐌰𐌻𐌴𐌹𐌽𐌰 (aleina) I would think that it ought to be *alīnō, but other editors seem to have favoured alinō in the past, so I added both as possibilities on the Gothic entry when I created it. And I am still figuring out how all these sound changes work so I cannot be entirely sure of which it ought to be from my own knowledge. Which is the right one? (Or perhaps there were two forms in P.Gmc.?) What do you think? — Kleio (t · c) 16:36, 2 June 2016 (UTC)

*alinō would certainly be the parent for North-west Germanic. Only Gothic seems to have the long vowel (doesn't ei = ī in Gothic ?). As far as I can tell, The Gothic is the unexplained radical Leasnam (talk) 20:50, 2 June 2016 (UTC)
Yes; as far as I know, Gothic ei corresponds to iː and reflects P.Gmc. ī -- thence my confusion. — Kleio (t · c) 21:07, 2 June 2016 (UTC)
Since it's reconstructed as both *alinō and *alīnō, you could create 2 separate entries, OR (--what I would do...) you could create the former and list the latter as an alternative form Leasnam (talk) 22:52, 2 June 2016 (UTC)
That seems like the best course and I added *alīnō as an alternative form of *alinō. — Kleio (t · c) 20:08, 3 June 2016 (UTC)
  • Ringe's The Development of Old English has *alinō and says "the long vowel in Goth. aleina is puzzling".
  • Boutkan says the PGmc was "probably" *alīnō. Boutkan and Lehmann both reject the suggestion of some authors that aleina is a writing error for *alina, citing Feist.
  • George T. Rikov, Hittite haliya- 'to kneel, genuflect' [...] 'elbow'..., in Балканско езикознание, volume 45, page 335, says:
    OIcel. ǫln, aln, OE eln, OFris. elne and OHG elina point to PGmc. *alinō which can correspond to Lat. ulna, if it continues PIt. *olenā (cf. Krahe-Meid I: 65). Yet the alternative explanation as *h2olineh2 (cf. Pokorny 1959: 307-308, Lehmann 1986: 26) is also possible. However[,] the Gothic hapax aleina 'ell' (only acc. sg., M. 6.27) points to PGmc. *alīnō (cf. for instance Krahe-Meid III: 107) which might correspond to OIr. uilen and W., OCorn. elin, Bret. elin, ilin, if these Celtic forms reflect *h2olīneh2-.

- -sche (discuss) 00:40, 3 June 2016 (UTC)

Twice-borrowed term or term derived from an older stage of the same language?Edit

Modern Hebrew פסחא‎ is "[l]ikely a reborrowing from Aramaic פַּסְחָא ‎(pasḥā) or Ancient Greek πάσχα ‎(páskha, Passover), ultimately from Biblical Hebrew פֶּסַח ‎(pésaḥ, Passover)." Normally when a term in language X is a loanword from language Y, which in turn borrowed from language X, we call it a "twice-borrowed term". But in this case, the word was borrowed from language Y, which in turn borrowed it from an older stage—an etymology-only language stage—of language X. So do we want to put this in Category:Hebrew twice-borrowed terms, or in Category:Hebrew terms derived from Biblical Hebrew? I would be prefer the former, but in order to achieve that we have to write "from Biblical {{der|he|he|פֶּסַח||Passover|tr=pésaḥ}}", rather than what we usually do in such cases, "from {{der|he|hbo|פֶּסַח||Passover|tr=pésaḥ}}". When I wrote the former, Wikitiki89 changed it to the latter, which also changed the category to Category:Hebrew terms derived from Biblical Hebrew. So how do we want to proceed? Ideally, it would be great if the modules recognized terms borrowed from an etymology-only language into the corresponding primary language as twice-borrowed terms, so that we could write {{der|he|hbo}} and still get the twice-borrowed term category. Is it possible to edit them to do that? —Aɴɢʀ (talk) 14:07, 3 June 2016 (UTC)

Honestly, I think etymology-only languages should be handled as if they were their parent language, and only in addition to that should they be added to a "derived from etymology-only language" category. --WikiTiki89 14:26, 3 June 2016 (UTC)
Exactly. We treat Biblical Hebrew as a kind of Hebrew, so it's similar to having a term for "English terms derived from US English". They were already Hebrew/English before, so how can they be derived from anything? —CodeCat 14:31, 3 June 2016 (UTC)
So we agree that Category:Hebrew twice-borrowed terms is the appropriate category, rather than Category:Hebrew terms derived from Biblical Hebrew? How do we implement that? Currently, the only way is the cumbersome "from Biblical {{der|he|he}}". @CodeCat, can you edit the relevant modules so that "from {{der|he|hbo}}" will categorize in the desired way? —Aɴɢʀ (talk) 14:38, 3 June 2016 (UTC)
How is this? —CodeCat 14:46, 3 June 2016 (UTC)
Looks good, CodeCat, thanks! —Aɴɢʀ (talk) 17:45, 3 June 2016 (UTC)
(e/c) What I mean is, {{der|en|hbo}} should place the entry in Category:English terms derived from Hebrew and secondarily also in Category:English terms derived from Biblical Hebrew. Similarly {{der|he|hbo}} should place the entry in Category:Hebrew twice-borrowed terms, but should probably make an exception and not place it in Category:Hebrew terms derived from Biblical Hebrew. --WikiTiki89 14:49, 3 June 2016 (UTC)
Category:English terms derived from Biblical Hebrew is already a subcategory of Category:English terms derived from Hebrew. —CodeCat 14:54, 3 June 2016 (UTC)
So what? Category:English nouns is a subcategory of Category:English lemmas, but we still put entries in both. --WikiTiki89 14:56, 3 June 2016 (UTC)
But we don't duplicate entries in Category:English terms derived from French into Category:English terms derived from Romance languages, which is more pertinent a comparison. —CodeCat 16:25, 3 June 2016 (UTC)
Well maybe we should? --WikiTiki89 16:55, 3 June 2016 (UTC)
Here are the category breadcrumbs from Category:English terms derived from French:
  • All languages » English language » Terms by etymology » Terms derived from other languages » Indo-European languages » Italic languages » Romance languages »
If we're going to include some of these in the entry, how many? Chuck Entz (talk) 17:22, 3 June 2016 (UTC)
That's a good point, we don't need to include families. --WikiTiki89 17:24, 3 June 2016 (UTC)

Calques Taking Place In Other LanguagesEdit

Currently gogravius, Gaugraf, and добродушный are showing module errors because they're attempting to use a non-existent |nocat= parameter. These are etymologies where a term was calqued into another language before being borrowed or inherited into the current one. I wasn't sure about the best way to fix these, since I'm not sure how much information about the calque we want to show in the etymology and in the categorization. Currently, the template shows:

  1. Calque of <source language> <source term> or
  2. <target language term> + <target language term>, calque of <source language> <source term>

It also has parameters for translations, transliterations, glosses, parts of speech, etc. It categorizes into [[Category:<target language> calques]] and [[Category:<target language> terms borrowed from <source language> <]], and [[Category:<target language term> compounds]] for the second configuration

Do we want to show the usual display of the template and:

  1. suppress all categorization? That would mean implementing the missing |nocat= parameter
  2. add only the calques category?
  3. add [[Category:<target language> terms derived from <source language>? This would require a new parameter.
  4. as previous, but add the calques category?
  5. replace {{calque}} with some combination of {{etyl}} and {{m}} with various parameters?
  6. something else I haven't thought of?

Chuck Entz (talk) 19:14, 3 June 2016 (UTC)

The {{calque}} template doesn't have much purpose besides categorization, so I would just say not to use it. Just like we don't use {{bor}} when a term was borrowed into another language. The {{calque}} template needs to be reworked anyway, it's really bad. --WikiTiki89 19:22, 3 June 2016 (UTC)
I marked some of the parameters as deprecated in the documentation, and without them it's basically like {{bor}} plus the extra "calques" category (a calque is a borrowing after all). So we should presumably treat it similarly. {{bor}}, {{inh}} and {{der}} all lack a nocat= parameter because the category is the main purpose of the template (that and adding extra meaning into our wikicode). Without the categories, {{inh}} and {{der}} would basically become like {{cog}}. —CodeCat 19:28, 3 June 2016 (UTC)
I've tidied up the three entries mentioned above and removed the {{calque}} template, since these words are not calques within Latin, German, and Russian respectively. —Aɴɢʀ (talk) 19:54, 3 June 2016 (UTC)


I think this etymology is mistaken. We reconstruct a Proto-Germanic form, but the German and Dutch standard sources agree that their respective words are of Romance origin ([1][2]). The word is first attested in German in 1361 and in Dutch in 1477, that is significantly younger than the Old French (1270). Moreover, the widespread Upper German form Schrauf instead of Schraube (cf. Yiddish שרויף(shroyf)) also hints at its being a borrowing. The Romance etymology makes good sense. And, finally, the word means only "screw" in all Germanic languages although the Germanics did not have screws. So, a Proto-Germanic form is very speculative at best and shouldn't be treated like a fact. But actually there seems to be little justification at all for such a reconstruction. Kolmiel (talk)


From αναχωρώ(v.pre.)/αναχώρησα(αναχώρηση(n.) or from ανα- + χωρώ? Sobreira (talk) 10:45, 6 June 2016 (UTC)

French future/conditional stem of être, ser-Edit

If *essere led to être, I wonder what the sound changes to result in the future stem ser- were. Hillcrest98 (talk) 13:36, 6 June 2016 (UTC)

The difference is probably due to stress placement: *éssere > être but *esserā́t > sera. The original final stress is shown by the modern forms in Italian sarà, Spanish será, Portuguese será, etc.; these also suggest that loss of the initial e happened already in VL. —Aɴɢʀ (talk) 15:06, 6 June 2016 (UTC)
I was thinking that as well. Hillcrest98 (talk) 16:35, 6 June 2016 (UTC)

dims derived in Latin or PIE, paxillus, vexillum, from palus, velumEdit

Can paxillus and vexillum be diminutives of palus and velum? As they do not conserve -l- of the root but -x- from Reconstruction:Proto-Indo-European/weǵʰ-, should they be from a suffixed diminutive form in PIE? (the only PIE diminutive suffix/infix I found is the one supposedly used for mægden). Then palus doesn't actually come from GRC... Sobreira (talk) 12:17, 8 June 2016 (UTC)

This situation is actually somewhat well attested.
  • āla < *akslā < *h₂eḱs-leh₂ vs. axilla < *aksi- (whence axis) + secondary Italic diminutive *-lā < *h₂eḱs-is
  • māla < *maxlā < *méh₂ǵ-s-leh₂ vs. maxilla < *makso- + secondary Italic diminutive *-lā.
I think a similar situation is occurring in these words. —JohnC5 15:07, 8 June 2016 (UTC)

PIE for flanEdit

This are the origins in English Etymology 1:

Borrowed from French flan (cheesecake, custard tart, flan),
from Old French flaon,
from Late Latin fladō (flat cake),
from Frankish *flado, *flatho (flat cake),
from Proto-Germanic *flaþô (flat cake),
from Proto-Indo-European *pl̥h₂t- (broad, flat),
from Proto-Indo-European *pelh₂- (to spread out, be broad, be flat)....

and below in French:

From Old French flaon,
from Late Latin fladonem,
accusative of flado (flat cake),
from Frankish *flado (flat cake),
from Proto-Germanic *flaþô (flat, broad),
from Proto-Indo-European *plat-, *pla- (flat, broad),
from Proto-Indo-European *pele- (to spread out, broad, flat)...

Apart from skipping the "fladonem, accusative of" and the Frankish "flatho", why are the PIE different? So different in fact. Exactly for this I don't care about the shape, but the same Proto-Germanic should come from exactly the same PIE.

And now the shape. I've been searching the etymologies of What links here of *pele-, *plat-, pla-, and others (even with and without hyphen) and most of them make sense (EN place, flatter, LA (a)placo ("please", "appease"), LA plaga ("beach"), GR pélagos, LA later ("brick")), but there is a messy lot of them and I don't know how to relate them (supposed derivatives like here? different grades? modified by H?).

Without hyphen:

I didn't check them all, but so far I found only "plat" as extension of "pelh2" in place and this plºh2t<pelh2 and plat<pele of flan. Any idea or suggestion of help? Sobreira (talk) 13:24, 10 June 2016 (UTC)

Germanic *flaþô can only come from something like *plót-, a form like *pl̥h₂t- would become *fuld- in Germanic. Unlike languages like Greek or Latin, Germanic shows no distinction in the outcome of syllabic sonorants whether a laryngeal follows or not. —CodeCat 13:51, 10 June 2016 (UTC)
Lots of editors seem to be in a bad habit of uncritically listing a zillion "competing" or morphologically related PIE forms in etymology sections, which often simply amount to transcription variants (perhaps outdated) and may not be warranted by the word being treated. For example, piac refer to "*plat-, *pla-", when a closer look at πλατύς (platús) (listed as an intermediate) shows that the modern reconstruction for the ancestor of this form in particular is *plh₂tus. Even without taking any stance on the PIE reconstruction issue per se, it is clear that a Hungarian word deriving from a Greek word should not state a different proto-form from the Greek word itself.
Or, flatter in the first etymology referred to "*plewd-, *plew-, *plōw-" although the Germanic word is a descendant of the first stem specifically. Listing the other two in the entry does not seem to provide anything else but confusion (I've removed them). Similarly a variant "*plet-" seemed to make an appearence here for no purpose whatsoever.
This might be too obvious to mention, but in terms of establishing a "lemma" reconstruction, we should always refer to (modern) sources on PIE over tertiary sources like etymological dictionaries of individual (and possibly non-IE!) modern languages. --Tropylium (talk) 21:51, 10 June 2016 (UTC)
Thanks for the answer, I understand it a bit better. Oh, a lot of topics arise boiling into my mind. First of all, my ignorance still about derivation within PIE, how and why they relate to each other, and yet from late PIE to proto-families. And about the validity/correspondence of variants. Second, the sources: I guess Pokorny is authoritative but outdated on laryngals and the variants can be outdated too. Also the recentness of the reconstruction, considering new data available. And the difficulty of finding references, at least for me, with scarce biblio (the web is not very plenty of PIE however, I found only ). My library has the IEED for Celtic, Baltic, Iranian and Frisian, but not Latin, which is outrageous absurd. And also, the absence of references stated by the editor. Third, what I don't know is whether the PIE studies are advanced enough for having consensus, as I read about different theories over and over again and on different topics (Urheimat, phon laws, exceptions, s-mobile....). Forth, some of the conflating proto-forms given when the intermediate derivative doesn't match the one given for the original (this would very simply be fixed with transclusions, like taxonavs in wikispecies!). Sobreira (talk) 22:50, 10 June 2016 (UTC)
To keep it simple and start somewhere: can that Á of Special:WhatLinksHere/Reconstruction:Proto-Indo-European/plÁt- be even right? Second, do we have a list of special characters for PIE? Sobreira (talk) 22:57, 10 June 2016 (UTC)
Well, it's completely incompatible with our notation, so, in that sense, it's wrong. You would have to know what notation is being used in order to figure out what it actually represents, after which it can be restated in our notation. Chuck Entz (talk) 23:04, 10 June 2016 (UTC)
As to your second question, see WT:AINE. Chuck Entz (talk) 23:06, 10 June 2016 (UTC)
After looking through your list above, I would say most are consistent with *pleh₂- . First of all, w:Indo-European ablaut means that the same root can exist with *e, *o or without a vowel in a given slot, so *ple-, *plo- and *pl̥- are all what we would call *ple-. Secondly, almost all instances of alleged PIE *a can be easily explained as the result of a laryngeal or other syllabic consonant/semivowel, whether directly, or as the result of a vowel being colored by the neighboring sound (generally h₂). Third, when sounds are lost by phonological processes, neighboring sounds are often lengthened, and laryngeals were lost just about everywhere they didn't become a vowel. That means that a PIE long vowel is often just another way of indicating a PIE vowel + laryngeal combination. Finally, roots can have various consonants tacked onto them, so *plat- and *plak- are probably best analyzed as *pleh₂-t- and *pleh₂-k-. I'm not sure where that puts *pele-, which perhaps is treating the laryngeal as just another consonant to be tacked onto the root instead of part of the root, or perhaps is just pretending it isn't there. I wonder where etymonline got it. Chuck Entz (talk) 01:16, 13 June 2016 (UTC)
(edit conflict) The problem starts from the fact that most of the information available on the web is from pre-1922 (read: ancient) sources and from general works like dictionaries. The latter are especially a problem because they use their own transliterations, and they use different forms than the dictionaries in those languages (e.g.infinitive rather than first-person singular for Ancient Greek).
The people using these sources don't understand them, so they just regurgitate what they think applies into the etymologies next to what others have regurgitated from other sources. As far as most editors are concerned, one source is as good as any, so they all should have equal space.
The real problem, though, is that not enough of those who edit etymologies have both the knowledge and the time to reconcile the conflicting systems represented and distill everything into a concise and internally-consistent statement that gets to the point and avoids padding. Data is easy to find, but understanding is rare. Chuck Entz (talk) 22:59, 10 June 2016 (UTC)
Danke böth! Sobreira (talk) 11:02, 11 June 2016 (UTC)
I have been thinking all day about this topics (9-10 forms and variants for one concept!) and I have ended up realising that due to 1) my lack of knowledge on PIE evolutions 2) the lack of consensus on PIE as I understand (I have just read about the Leiden vs. Beekes conjugation, not to mention the glottalic theory) and 3) the lack of material at my disposal (not to mention Pokorny is not legitimate any more); then I have to give up on doing anything systematic about Appendix:List of Proto-Indo-European roots, and even less about nominals/verbs. Unless you all feel I could be entitled to protract all the conflicting variants and intermediate etyms to the ultimate root, I have to desist. I am quite unsure and picky on this topic, and meticulous, systematic and think very theorically in general, so if I don't do it that way then I don't feel comfortable. I will indeed try to state the etymologies for the GL words I meet following the simple pattern GL<LA<PIE (starting from PIE, not from GL), so I will probably go through all of them, or at least try a few of them. I will rely only on the few modern available materials I found or could find that I think worth it (if you know any other, please let me know):
  • GENERIC for LA>PIE on the net:, AmHeritage, utexas updating Pokorny, etymonline, wikipedia articles on PIE, roots at ielex and -maybe- the etimologiasdechile aforementioned when izquierda which I don't like and wouldn't trust so much;
  • GENERIC for LA>PIE on book: unless (until) I buy or get the Roberts-Pastor specific for ES or I make my library buy the IEED Leiden-Brill for Latin by de Vaan;
  • SPECIFIC for GL>LA on book: 4 Galician (Ir Indo/Estraviz/SéculoXXI/GDXeraisL) and 1 Portuguese Houaiss general dictionaries stating etymology) and try to source my edits.
Of course I will try to leave track of my search on User:Sobreira/*pele- and so on (including all of them in upcoming User:Sobreira/PIE), to save time and work in case anyone with knowledge enough engage in the task of uniformizing the chaos. In that case, I would appreciate the courtesy of being warned. Thanks all for the recommendations, knowledge shared, respectful and patient treatment and time.
  • And because all of this, I think we should consider seriously the proposal of transclusion I said above, or some other way of solving this problem. Sobreira (talk) 17:49, 11 June 2016 (UTC)
I've digitized for my own research use a list of all roots in {{R:ine:LIV}} some time ago; I'd still need to verify it against Kümmel's online repository of Addenda und Corrigenda, but once that's done, it should make a fairly new and fairly consistent working list. (Individual entries, when we get around to writing them, may still need more detailed consideration though.) --Tropylium (talk) 20:32, 11 June 2016 (UTC)
Oh, thanks for the link. If I can help anyhow, just tell me. Do you contrast them online? If I have time and it doesn't take me too long, maybe tomorrow I send you something that could ease your job.
By the way, about sourcing the etyms, I think we should specify what part is sourced (with ref superindex), as simply saying LSJ of AmH is too generic for some elaborated etyms given and there is no way of knowing which part is refererenced. Sobreira (talk) 12:22, 12 June 2016 (UTC)
For the current discussion (PIE transcription/reconstruction standardization), just digging thru the Addenda und Corrigenda for any new roots / disproven older roots would suffice; and I could similarly already dump the root list by itself. What will take more work is updating for the corrections about reflexes in the IE languages, if we wanted to do something similar to our current List of PIE roots.
(Also, what's "LSJ of AmH"?) --Tropylium (talk) 17:24, 12 June 2016 (UTC)
My guess is that of is a typo for or, LSJ refers to A Greek–English Lexicon, and "AmH" refers to either the American Heritage Dictionary as a whole or specifically to its Appendix of Indo-European Roots. —Aɴɢʀ (talk) 17:35, 12 June 2016 (UTC)
Good three guesses, Angr. Sorry, I am sometimes too cryptic by shortening and lack of time. Tropylium, do you mean a list of reflexes/derivs from one post-PIE to the next languages? I would have never thought of that if you didn't mention! Good idea. Ambitious and huge, but it would be wonderful. I'll see what I can do while working on GL, mainly in the ones heritage from Latin and French/Provençal, maybe also Old Greek. Sobreira (talk) 19:54, 12 June 2016 (UTC)


Per WT:REF, the etymology for "ZOMG" requires sources. The listed etymology, that ZOMG is "presumably" a typo of OMG caused by a missed shift key, is unsourced. I've provided this sourced edit, whose sources verifiably date to 2003 and 1999 respectively. The sources support the etymology that the word is either an excited typo or a deliberate parody of chatspeak, or plausibly both, and the popular use of the word can be traced directly back to communities which Curbo, Milan and myself were members of in 2003. 22:26, 12 June 2016 (UTC)

This Stack Exchange thread does some useful research into the word, although it comes to no agreement: What is the origin of ZOMG? 22:32, 12 June 2016 (UTC)

The only thing that the references prove is that Curbo and Millan claimed in 2003 to have originated it. We quite often get posts here by people who say that they and/or acquaintances invented a term. Invariably someone comes up with examples of usage predating the alleged invention, which blows their claims out of the water. You're different only in introducing the prior quotes yourself so you can say they don't matter. Chuck Entz (talk) 23:12, 12 June 2016 (UTC)
The edits and citations reflect information other than Curbo/Milan's relatively unimportant credit:
  • This word is a strong candidate to require citations, as per Wiktionary guidelines regarding words of uncertain etymology
  • An alternative etymology is that ZOMG is an intentional parody of late 1990s/early 2000s chatspeak; compare OMFG and OMGWTF for the addition of letters signifying swear words as intensifiers, OMGWTFBBQ for the addition of nonsensical letters as a joke (attested 2001, popularised at least as far back as 2003 as the username of Something Awful contributor Ryan Adams), and LOLZ for the use of Z in leetspeak or chatspeak (attested 1999). The highly mutable nature of leetspeak and hackerspeak of this period makes this at least an equally plausible etymology.
  • Milan's FAQ, as a contemporary source, documents that the word was at least used in 2003, and that he describes it arising from a conversation involving the use of leetspeak. The initial use appears to have been lowercase, "zomg", which is consistent with someone missing the shift key. The uppercase version "ZOMG" and later construction "zOMG" would require someone to hit both shift and Z, which is harder to do without noticing before you hit enter.
  • The word "ZOMG" has humour and is easily transmitted, explaining its viral propensity and popularity in large internet communities. The user expects each letter in an acronym to stand for a word, but is unexpectedly given the uncommon letter Z which is highly unlikely to stand for anything. This occasionally flummoxes users who ask what the Z stands for, as a result there are several folk etymologies which were derived independently by guesswork or what made sense (e.g. Zerg OMG, Zombies OMG, missed shift). However, the answer that the Z means nothing is unsatisfying and therefore unlikely to be put forward or supported; people want answers that make sense. Therefore, over time, the "intentional joke" etymology would have become less popular, and the missed shift etymology more popular, regardless of which is true.
  • The later spelling "zOMG", suggestive of a missed shift etymology, does not appear in the early sources, which supports the hypothesis that the missed shift etymology became more popular over time without necessarily being accurate. Later users (2006 onward) were no longer exposed to the 1990s hacker culture and would be increasingly unaware of the fad of inventing ridiculous acronyms (e.g. NIFOC) and would be less likely to accept it as an explanation.
Hence, while the missed-shift etymology is widely believed, it is not necessarily definitive. The idea that it was invented deliberately as a joke is plausible and equally worthy of inclusion. 03:12, 13 June 2016 (UTC)

List of PIE verb rootsEdit

As mentioned above, you can now find all headwords of {{R:ine:LIV}} (for now with German glosses only) listed at User:Tropylium/Proto-Indo-European/Verbs. New and adjusted and reconstructions are next in line to be also added. --Tropylium (talk) 19:33, 13 June 2016 (UTC)


Is it just me or could *kirikǭ not possibly have existed in Proto-Germanic? As far as I know Proto-Germanic was not a thing anymore by the time the idea of a 'church' could have made its way to Germanic-speaking regions. Nor does Kroonen have it. Not sure if I'm missing something here. — Kleio (t · c) 19:58, 14 June 2016 (UTC)

I don't know, Germanic speakers surely encountered churches and may have needed a word for them even before they themselves were christianized. —Aɴɢʀ (talk) 20:21, 14 June 2016 (UTC)
But surely Proto- Germanic speakers would not? The earliest I can realistically imagine Germanic speakers needing a word for a church would be after the Goths started raiding the Roman province of Dacia and across the Danube in the Balkans from roughly the mid-3th century AD. The entry itself even notes that it would've spread via the Goths, but again, that means the time is way off for it to be Proto-Germanic, no? Seeing as Gothic is a descendant of Proto-Germanic. — Kleio (t · c) 20:31, 14 June 2016 (UTC)
D.H. Green in Language and History in the Early Germanic World (1998, p. 297) claims kyriakon was first used in Greek during the third century, and was in use through the fourth century, dying out in favour of ekklesia in the fifth. So it is likely to have been borrowed via the Goths sometime between 250-400. Considering Christianity only became the official religion of the Empire in the early 4th century, I think we would have to place the borrowing sometime in the 4th century. (Curiously, Wulfila, translating in the second half of the 4th century, has aikklesjo (from ekklesia), and no equivalent of kyriakon even once) — Kleio (t · c) 20:55, 14 June 2016 (UTC)
I too agree that the timing is off--this word is too late to be considered even latest Proto-Germanic :( Shame too, it's a very informative entry. Can we not move it to West-Germanic or something ? Leasnam (talk) 21:01, 14 June 2016 (UTC)
Idk, afaik there isn't a precedent for west-Germanic entries, but this could then be a first? Not sure honestly. Do agree it's a nice entry besides the chronological issues of calling it Pgmc. — Kleio (t · c) 21:06, 14 June 2016 (UTC)
We don't have a code for Proto-West-Germanic, and many linguists believe there never was such a thing. It is labeled "West Germanic" as a kind of dialect of Proto-Germanic. At any rate, it seems unlikely to have been borrowed more than once, since there's such a small window in which κυριακόν (kuriakón) rather than ἐκκλησία (ekklēsía) (the only term used in the New Testament) was the usual word. BTW, why do we assume the word was borrowed via the Goths when the word isn't attested in Gothic? —Aɴɢʀ (talk) 21:09, 14 June 2016 (UTC)
For lack of better candidates, I guess; the Goths after all were at the relevant border of the empire, where Greek was spoken. The source I cited above does not mention the Goths. Someone he appears to have debated does.Kleio (t · c) 21:19, 14 June 2016 (UTC)


Are nem- and νόμος the origin? See æ#Noun. Lysdexia (talk) 00:39, 20 June 2016 (UTC)

That would require explaining the υ, which doesn't come from regular PIE vowels. If you're suggesting a compound with *h₂ékʷeh₂ (the source of the second etymology of ǣ), that might be able to explain the φ, but then you would have to explain what happened to the vowels both before and after the φ (η/ᾱ wouldn't cause the labiovelar to end up as φ, and one would expect a vowel from *h₂e regardless of the ablaut grade). Chuck Entz (talk) 13:57, 20 June 2016 (UTC)


Where did the "group of crows" sense come from? Should probably be noted in the etymology. —CodeCat 17:40, 20 June 2016 (UTC)

It's attested since at least the 1470s; William Caxton's 1477 edition of The horse the ghoos & the sheep has "an Herde of hertes a Murther of crowes" among its list of collective nouns, and a 1475 version supposedly has "a morther of crowys". But Grammarphobia quotes the OED as saying it is "one of many alleged group names found in late Middle English glossarial sources", i.e. it may have been made up by people writing lists of terms, rather than being in common use like flock; and the usage "apparently died out after the 1400s, [but] was revived in the 20th century". Most sources connect it to the traditional association of crows with death. - -sche (discuss) 17:58, 20 June 2016 (UTC)

EN derivates from PIE *h₂egʷnosEdit

The verbs to (y)ean, do they come from PGmc *aunōną or *gaaunōną? Because it's given in both. Sobreira (talk) 09:39, 21 June 2016 (UTC)

ean comes from *aunōną; yean comes from *gaaunōną, which itself comes from *aunōną Leasnam (talk) 19:29, 21 June 2016 (UTC)
Why is *gaaunōną reconstructed for PGmc, if it only has descendants in Old English? Incomplete list of reflexes? --Tropylium (talk) 20:56, 21 June 2016 (UTC)
I can't answer that question ; it probably doesn't need to exist, since it's apparent descendants (if this did exist in pgmc) are already at *aunōną Leasnam (talk) 21:32, 21 June 2016 (UTC)
Don't we create reconstructs when there are descendants in only one language? Do we write it "yean (< *gaaunōną)" in descendants of *aunōną then ? And delete *gaaunōną?? Sobreira (talk) 07:30, 23 June 2016 (UTC)
We can create pages for words with only descendants in only one language, although entries with a multitude of descendants in multiple languages are often seen as more solid. PGmc for only Gothic or Old Norse, and PItal for only Latin are examples. Leasnam (talk) 17:21, 23 June 2016 (UTC)
Considering the y- variant isn't even attested in Middle English or Old English, let alone outside the English branch of Germanic, it seems overly optimistic to reconstruct *gaaunōną for Proto-Germanic. "Yean" probably doesn't have the ge- prefix it all; it's probably just a sporadic variant of "ean", perhaps by hypercorrection in a dialect that says "east" for "yeast". —Aɴɢʀ (talk) 18:18, 23 June 2016 (UTC)
I believe the ga- form likely existed. Wycliffe has ȝene, ȝeene, yeene beside ene, eene (sheep with lambs). There is also the Old English ġeēane (yeaning), possibly an inflected form for *ġeēan, but in any event suggesting a prefixed form. The Saterland Frisian word bejääne also hints at one (be- + je- + ääne?), though I am not 100% sure. Is all this enough to warrant a page in my opinion ? No. Since most of the descendants are prefixless, I would use that, putting the other forms on the same page. Leasnam (talk) 21:47, 23 June 2016 (UTC)

from lamb to placentaEdit

From the PIE aforementioned gives GRC ἀμνός, which gives diminutive ἀμνίον. But how can it change its meaning into "bowl in which the blood of victims was caught" (in amnion/amnios and also mentioned as derivate by template:R:ine:RobertsPastor)? Was it made of lamb leather/skull or the lamb taken as a stereotypical sacrifice animal? Any source? Sorry for the disgusting topic. Sobreira (talk) 11:15, 21 June 2016 (UTC)

Probably from the shape and bloodiness of the placenta. —Aɴɢʀ (talk) 14:28, 21 June 2016 (UTC)

Order of sections in PIE: place for synsEdit

Where do we want the synonyms section?:

  • At the beginning, before Derivs/Descend:
between Root+Usage notes and Derived terms: krewh₂-
between Root and Derived terms: leǵ-, les-, pelth₂-, perd-, pesd-, pleh₂-, h₁ésh₂r̥
between Inflection and Coordinate+Descendants: ph₂tḗr
between Root+Inflection/Declension and Descendants: átta, bʰardʰeh₂, dáḱru, ǵómbʰos, h₁éḱwos, h₁er- (why a noun with "-"?), h₁n̥gʷnis,h₂ébōl, h₂eḱru, h₂éngʷʰis, márkos, médʰu, mélit, péh₂wr̥
between Inflection and Usage notes+Descendants: h₁ógʷʰis   Done (FIXED)
  • Later or end, after Derivs:
between Derived terms and Descendants: dʰéǵʰōm   Done (MOVED)
between Derived terms and References: kʷer-   Done (MOVED)

Nothing at Wiktionary:About_Proto-Indo-European. Sobreira (talk) 13:48, 21 June 2016 (UTC)

Per Wiktionary:Entry layout#Headings after the definitions, synonyms come after Inflection, before Coordinate terms, Derived terms, and Descendants. —Aɴɢʀ (talk) 14:29, 21 June 2016 (UTC)


Could someone conversant with Hindi and Sanskrit update hoon (etymology 4)? Thanks. — SMUconlaw (talk) 22:12, 21 June 2016 (UTC)

Eirikr has already added the Sanskrit etymon; I assume the Hindi is written the same way. —Μετάknowledgediscuss/deeds 00:24, 22 June 2016 (UTC)
Thanks. Isn't Hindi script slightly different from Sanskrit? — SMUconlaw (talk) 07:09, 22 June 2016 (UTC)
The script is exactly the same; individual words may be spelled differently in Hindi and Sanskrit, though. —Aɴɢʀ (talk) 10:45, 22 June 2016 (UTC)
Strictly speaking, there is no Sanskrit script, just Devanagari as used for Sanskrit. There are some extra consonants in Hindi made by adding a diacritic or two, but they don't seem to be used as much in the inherited core vocabulary, and Sanskrit borrowings tend to be spelled as Sanskrit. More often than not the spelling is the same, but Hindi deletes some of the unwritten vowels in the pronunciation. Chuck Entz (talk) 13:43, 22 June 2016 (UTC)
Thanks for clarifying. — SMUconlaw (talk) 21:11, 22 June 2016 (UTC)

բուն Armenian "tree trunk"Edit

RFV of the etymology: book gives Armenian as cognate from PIE *bʰeh₂ǵos (tree). The meaning matches much better than PIE *bʰudʰmḗn (bottom). Sobreira (talk) 10:29, 23 June 2016 (UTC)

I added the standard references. The etymology from *bʰudʰmḗn (bottom) through Iranian is universally accepted. And you are wrong that *bʰeh₂ǵos matches better. It means “beech”, not “tree”. The sense development “beech” → “tree trunk” is unimaginable, whereas “bottom, foundation” → “tree trunk” is easy, seen also in the Greek cognate πυθμήν (puthmḗn, bottom; tree trunk). --Vahag (talk) 19:29, 23 June 2016 (UTC)
Well, I guess it's a quite sujective matter. I wouldn't assert so sure what matches more. For me, relating the part to the total is easier than the part to the generic part. Didn't PIE for "oak" (*dóru or *perkʷu-?) give tree in a bunch of PIE? I'm only certain that we have to follow the opinion of the primary sources (and data: that's why I RFV), but they are opinions too and can be wrong too. Sobreira (talk) 14:33, 24 June 2016 (UTC)

PIE for oakEdit

Is *eiḱ a noun form from the root *eiǵ-? Or should oak and aesculus point to Special:WhatLinksHere/Reconstruction:Proto-Indo-European/eiǵ-? Sobreira (talk) 10:39, 23 June 2016 (UTC)

I feel that both *eiḱ and *eiǵ- should be changed to *h₂eyǵ-. That is incidentally the form it's given at Appendix:List of Proto-Indo-European roots/h₂. —Aɴɢʀ (talk) 14:07, 23 June 2016 (UTC)
Which I guess it's sourced from Appendix:List of Proto-Indo-European roots#References, innit? Sobreira (talk) 14:57, 23 June 2016 (UTC)
Probably. Buck and Pokorny won't have written *h₂, but the others may have. —Aɴɢʀ (talk) 15:22, 23 June 2016 (UTC)
Pokorny, page 13. Russians have *h₂eiǵ-. Didn't find in AmH or EtymOnLine, UTexas follows Pokorny. The <y> for being diphthong? Sobreira (talk) 15:56, 23 June 2016 (UTC)
Yes, "ei" and "ey" are exactly the same thing, just two different transcription conventions. Wiktionary follows the "ey" convention. —Aɴɢʀ (talk) 16:16, 23 June 2016 (UTC)
The problem is that it seems to be another root h2ei´g with the meaning of "goat" (ožys, αἴξ, एड). Subs? I'll check abouit later. Sobreira (talk) 14:17, 24 June 2016 (UTC)
Kroonen and de Vaan's etym. dictionaries seem to be hesitant to reconstruct this further back than the "Common IE" stage *aiǵ- (without *h₂); they suggest that this may be a substrate word. --Tropylium (talk) 18:40, 24 June 2016 (UTC)
I've been bold and unified the situation of eiḱ- and eiǵ- for PIE "oak" into *h₂eyǵ- (still to create the entry and the categories [ [osx/ang/... term from PIE root xxx] ]). I left provisionally PIE "goat" as *h₂eiǵ-. I also changed some of this *eiǵ- (eckle and ickle) into *h₁eyH- just because it made some better sense for me (h1 ~ e; transcript (ei) = transcript (ey)) and it was the only created PIE root that I found with that meaning (notice as well that uncreated *yeg- (ice) is mentioned in Appendix:List of Proto-Indo-European roots/y and Appendix:List of Proto-Indo-European nouns#Atmosphere). Correct me and please forgive me if I'm wrong. Sobreira (talk) 11:14, 4 July 2016 (UTC)

mugger and mugEdit

Currently, the verb meanings of mug (to strike, to rob) are listed under the same etymology as the noun mug (cup). Mugger has its etymology listed as mug + -er, which looks reasonable enough until one notices the lack of explanation for the verb. I could certainly believe that the verb came from the "face" sense of mug, as Etymonline and the OED suggest, though the current entry mentions nothing of this.

w:Makara (Hindu mythology) suggests that mugger comes from मगर (magar, crocodile), which implies that the verb mug is backformed from mugger. I find this etymology and sense development just as plausible as the other, but the Wikipedia article cites a blog post where the "mugger" connection is mentioned only as a side note with no listed sources, it doesn't look reliable.

Does anyone have any resources on this word? If not, can we at least split mug (verb) into its own etymology, with the note that it could be from mug (noun), but is uncertain? Listing the two under the same etymology with such distinct meanings doesn't seem right. Eishiya (talk) 14:36, 23 June 2016 (UTC)

Yep: the source for the Wikipedia article appears to be only talking about the crocodilian sense of mugger, and the claim that "This animal's name is descriptive of it's behavior" seems to be original research, as well as implausible anyway (it would seem to require that "mugger" can be parsed as "one who mugs" also in Hindi, not just in English). I'm removing it. --Tropylium (talk) 18:56, 24 June 2016 (UTC)


Is this a cursive caoshu, vulgar varient, etc? Johnny Shiz (talk) 18:37, 23 June 2016 (UTC)

Latvian from Middle Dutch.Edit

Middle Dutch is often mentioned as a source for Germanic loans, e.g. sīpols. How likely is that, given the Low German settlements in the region? Korn [kʰũːɘ̃n] (talk) 07:20, 24 June 2016 (UTC)

I don't know how likely Dutch would be as a source. I suppose not too unlikely since they were densely populated, rich, and had a lot of trade. Some of their merchands probably sailed to Latvia. — But the point is there's little reason to say that a word well attested in MLG should be exclusively from Dutch. So just add MLG as a source whenever that (i.e. good attestation) is the case. I think. Kolmiel (talk) 20:41, 28 June 2016 (UTC)
I have often noticed a lot of Russian words that originated from Middle Dutch, but they didn't enter Russian directly, but were rather borrowed first into German, then into Polish, and then into Russian. Perhaps something similar is going on with Latvian, although I would not know the exact path of the chain of borrowings. --WikiTiki89 20:48, 28 June 2016 (UTC)
@Wikitiki89: Yes, but Russian was a bit farther off maybe, receiving the words second-hand (or else by seafarers). But Korn is right that there were Low German-speaking areas very nearby Latvia (or even in it, I don't know), and Low German (later replaced by High German) was a major culture language of the area. Kolmiel (talk) 16:01, 6 July 2016 (UTC)


RFV of the etymology. ばかFumikotalk 09:53, 24 June 2016 (UTC)


RFV of the etymology. Needs specific reliable source that says it was "Coined in Japan from Sinitic elements". ばかFumikotalk 09:54, 24 June 2016 (UTC)

Since this is a concept specific to Christianity, and since we know that Christianity came to Japan mainly via the Spanish and Portuguese, we can safely say that this did not come directly from Chinese, but instead from contact with the west.
If it's the specific format you object to, the template {{calque}} would probably be a better fit, strictly speaking.
You're welcome to do this research yourself, and edit the entry accordingly. It's quite easy to do when these online resources have entries for the term you're researching. ‑‑ Eiríkr Útlendi │Tala við mig 23:00, 24 June 2016 (UTC)
It's just "safely say", not "definitely say", so I have reasons to be dubious right? ばかFumikotalk 03:14, 30 June 2016 (UTC)

work and wyrcanEdit

The Old English verb had palatalisation, so why do we not have "worch" today? Is it analogy with the noun? —CodeCat 15:59, 24 June 2016 (UTC)

ME seemed to use spellings implying both pronunciations interchangeably, even in the same edition of Chaucer. If it was after that, perhaps the OED will tell you how late the -ch pronunciation lasted. Edited: I deleted a conjecture which made no sense whatsoever, suggesting the de-palatalisation was to disambiguate a confusion which it in fact caused itself. Isomorphyc (talk) 18:38, 24 June 2016 (UTC)
From the OED: The normal representative of OE wyrcan would be *worch (for the vocalism cf. worm, worse, wort); the substitution of k for ch, producing the modern standard form... is shown in North Midland areas c 1200 and is due mainly to Work (sb.), though Scandinavian influence (see various forms above) is possible. Isomorphyc (talk) 18:45, 24 June 2016 (UTC)
Also a lot of verbs had alternations between /tʃ/ before a vowel and /k/ before a consonant which were then leveled out differently, e.g. sēċan (to seek) ~ sēcþ ((he) seeks) generalized the /k/ in Modern English, but besēċan (to beseech) ~ besēcþ ((he) beseeches) generalized the /tʃ/ in Modern English. So between the /k/ of the noun, Scandinavian influence, and preconsonantal position, there were probably lots of paths for wyrċan to generalize /k/. —Aɴɢʀ (talk) 10:00, 25 June 2016 (UTC)


RFV of the etymology. Originally, this character was written by mistake. is the original etymology. Source it please. Johnny Shiz (talk) 17:14, 25 June 2016 (UTC)


From Arabic. See Deborah

saur (filth)Edit

Is the etymology as being from Irish through contraction reliable? Further down the page we find Icelandic saur < Old Norse saurr, so loaning from Norse would look much more straightforward. --Tropylium (talk) 22:15, 26 June 2016 (UTC)


Köbler points in his etymology to High German tocken, from tocke, from *dukkō- (doll, bundle), which to me implies a split etymology: One for 'to pull' from *tukkōną (to pull) via Old Saxon and one for 'to bait' from German. Can anyone shed some light? @Leasnam Korn [kʰũːɘ̃n] (talk) 09:17, 29 June 2016 (UTC)


‎Wähnen has no etymology given. One of the meanings listed is, fortuitously, ween, and one of those is referred to *wēnijaną. However, *gawahwaną has also laid claim to it, or rather to er‎wähnen, 'with a different prefix'. I don't know why this should be thought to apply to er‎wähnen but not ‎wähnen. Er‎wähnen itself of course just refers back to ‎wähnen. Do we need a sword to divide the baby, or have we got DNA to assign it to one or the other pleading parent? --Hiztegilari (talk) 11:28, 30 June 2016 (UTC)

Despite the similarity, I don't believe erwähnen is a derivative of wähnen. I think they are wholly separate words, and that erwähnen should be removed from that page. Leasnam (talk) 23:27, 30 June 2016 (UTC)
  • Na, ja -- Duden's entry gives wähnen as deriving from MHG wænen, OHG wān(n)en, related to modern Wahn (illusion; delusion).
Meanwhile, their entry for erwähnen gives a derivation from MHG gewähenen, OHG giwahan(en) with a meaning of “to say, to tell”.
I suppose the OHG terms might be cognate? ‑‑ Eiríkr Útlendi │Tala við mig 04:14, 1 July 2016 (UTC)
It doesn't look like they are. In light of the above, I have removed erwähnen, replacing Derived Terms with Related term: Wahn Leasnam (talk) 08:26, 1 July 2016 (UTC)
They aren't cognates. Correct. -- And generally: The etymologies at are curt and not without flaw. Preferably use Pfeifer at (along with Kluge if you have access to that). In this case Pfeifer says the same thing though. Kolmiel (talk) 14:07, 6 July 2016 (UTC)


An earlier etymology, sourced from some crank's personal essay, claimed an improbable derivation from Uralic *pićlä (rowanberry). Is the Albanian comparison reliable, however? --Tropylium (talk) 14:31, 30 June 2016 (UTC)