Wiktionary:Etymology scriptorium/2014/February

discussion rooms: Tea room • Etym. scr. • Info desk • Beer parlour • Grease pit ← January 2014 · February 2014 · March 2014 → · (current)

---

serikali and سرکار

I assume these are from the same source, probably Persian or Arabic, but I don't know what it is. Does anyone know? —Μετάknowledge^{discuss/deeds} 00:22, 1 February 2014 (UTC)

See سرکار#Persian. --Z 13:13, 1 February 2014 (UTC)

Thank you for adding the entry and its etymology! But why did you say the Swahili was derived via a South Asian language? I don't know how the vowels were realised, but glancing through the descendants, I reckon it could just as easily have come through Turkish, for example. —Μετάknowledge^{discuss/deeds} 19:45, 2 February 2014 (UTC)

Because the sense of "government" for سرکار was developed in India. --Z 19:53, 2 February 2014 (UTC)

hysteresis

Is there an etymological connection between hysteresis and hysteria? The Greek roots given, husteros and hustera, are very similar and suggest a common origin. SpinningSpark 15:21, 1 February 2014 (UTC)

Werewolf = war wolf?

I've read in the past that a werewolf is wer + wolf, i.e. a man-wolf, an idea explored in detail at w:werewolf. However, I noticed the Wikipedia entry w:Werwolf, about the failed attempt to recruit Nazi partisans, uses "Werwolf" and "Wehrwolf" interchangeably. Indeed, wehrwolf is listed as an archaic English werewolf. Now Wehr is familiar enough from Wehrmacht and such, tied to wehren, which means "to defend". I see no specific mention of it in the etymology of war, but I'd think there ought to be a connection... I'm aware, of course, that a lycanthrope is indeed a man-wolf by etymology, and the Wikipedia article gives some impressive medieval history, but I'm still suspicious there might have been more than one concept conflated here.

Anyway, I think it would be interesting if someone could nail down this etymology, because potentially "wehrwolf" seems to me to hint at a more mundane origin for the 20th-century legends, i.e. a wolf which simply is particularly clever and aggressive in attacking humans to defend its territory or family. Wnt (talk) 16:08, 1 February 2014 (UTC)

Are you sure the Wikipedia Werwolf article uses Wehrwolf interchangably? I'm not seeing it. SpinningSpark 16:51, 1 February 2014 (UTC)

ROFL, looking over the article more closely, it actually has a reference about this issue [1]. I haven't gone through all that as of the moment, but the article says "the name was chosen after the title of Hermann Löns' novel, Der Wehrwolf (1910).[4] Set in the Celle region, Lower Saxony, during the Thirty Years' War (1618–48), the novel concerns a peasant, Harm Wulf, who after his family is killed by marauding soldiers, organises his neighbours into a militia who pursue the soldiers mercilessly and execute any they capture, referring to themselves as Wehrwölfe." Wnt (talk) 17:55, 1 February 2014 (UTC)

You didn't need to go to the reference, Löns' book is described in the article. I saw that, but the WP article itself does not use wehrwolf as a synonym. Why ROFL? Do you think I am being particularly stupid or slow? Very possibly, but you will need to explain. SpinningSpark 10:18, 2 February 2014 (UTC)

Indus

The current etymology doesn't really give any hint about why the word lost its s- on the way from India to English. Does anyone else know? —CodeCat 23:49, 1 February 2014 (UTC)

The current etymology also doesn't explain why Sindus, first attested in Pliny the Elder's Historia Naturalis, would be the source rather than Indus, attested in the writings of Ovid and Cicero, and borrowed from Greek Ἰνδός (attested in Herodotus), which presumably was borrowed from Old Persian. My hunch is that Sindus was a later re-borrowing directly from India, and that it coexisted with the older Indus, rather than completely replacing it. Chuck Entz (talk) 03:58, 2 February 2014 (UTC)

Does Old Persian have a regular loss of s, or s > h? —CodeCat 04:01, 2 February 2014 (UTC)

The h seems to have been lost between Persian and Greek. --WikiTiki89 04:05, 2 February 2014 (UTC)

I don't know, but if we restore the commented-out section in the etymology, we get:

From Latin Sindus, a Latinization of Hindu, an Iranian variant (compare Old Persian 𐏃𐎡𐎯𐎢𐏁 (hiⁿdūš)) of Sanskrit सिन्धु (síndhu, “river, stream; Indus”).

That would seem to indicate that the s>h happened in Old Persian (I think there was a variant with an s, as well, but I have no source of my own for either). It would make sense to assume that the h-variant made its way to Greece from Persia, losing the h, then into Latin, while a doublet with the original s was borrowed into Latin in later times, but didn't displace the version with the plain i, and that the version with the plain i is what has been passed down to the present day. I have no references that say so, but the idea that the s somehow materialized in the transition to Latin, then re-disappeared again seems a bit far-fetched. Chuck Entz (talk) 04:33, 2 February 2014 (UTC)

s > h sound change (in word-initial pre-vocalic position) was regular in Iranian, probably nearer 1500 BC (here is more discussion on the date and this particular OIr. word for "Indus" which is one of the affected words), but certainly not in Old Persian, in which s is always preserved in this position. (which also means Chuck Entz idea about existence of a re-borrowed *Siⁿdūš in OP is possible, though I can't see why Sindus should be necessarily borrowed via Old Persian?) Regarding loss of h: I don't know if there was any regular loss of h, but: the cuneiform sign 𐏃 (h(a)) is omitted when the etymologically expected h is followed by u or i (example). However, I haven't seen any source talk about loss of [h] in OP, it is usually assumed that it's just a writing convention and [h] was actually pronounced. P.S. Interestingly, 𐏃𐎡𐎯𐎢𐏁 (hiⁿdūš) is the only word among the attested OP words in which h is followed by i or u and still it has been scribed by the sign for h. --Z 10:46, 2 February 2014 (UTC)

esti

AFAIK, Esperanto doesn't take words from Latin, right? I suspect it's from Spanish estar. Please correct me if I'm wrong. --kc_kennylau (talk) 14:44, 2 February 2014 (UTC)

Esperanto often takes words from Latin: see kvankam, sed, and okulo, for instance. That being said, it seems that esti was probably based on Spanish estar, French est, Ancient Greek ἐστί, and German ist, in addition to Latin est. André Cherpillod's Konciza Etimologia Vortaro lists all of these as well, although for some reason it lists Spanish "es" instead of "estar". I'll edit the entry accordingly. —Mr. Granger (talk • contribs) 15:56, 2 February 2014 (UTC)

Ebbe Vilborg in his Etimologia Vortaro De Esperanto says "Estas fakto, ke Z ne utiligis la hispanan" ("It's a fact that Zamenhof did not make use of Spanish.") He cites Latin, French and Italian as the direct ancestors, and English, German, Polish and Russian as influences, with Greek relegated to the same historical sentence as Sanskrit and Proto-Indo-European.--Prosfilaes (talk) 10:27, 20 April 2014 (UTC)

Transgressive in Proto-Slavic

What is with transgressive in Proto-Slavic? Shouldn't this also be included? And how would reconstruction look like?89.201.150.245 02:04, 5 February 2014 (UTC)

I'm not sure what you mean with "transgressive"? Do you mean what is the Proto-Slavic word that means "transgressive"? —CodeCat 03:17, 5 February 2014 (UTC)

Maybe they want the transgressive forms of Proto-Slavic verbs? —Mr. Granger (talk • contribs) 03:22, 5 February 2014 (UTC)

(after e/c) I think he might mean w:Transgressive (linguistics), which is found in Czech, Slovak, and Polish. --WikiTiki89 03:23, 5 February 2014 (UTC)

Judging from the Wikipedia article, it looks like the transgressive is just a remnant of what were still fully inflected participles in Proto-Slavic. Polish -ąc derives from the present active participle in -ę ~ -ęť-, while -wszy is from the past active participle -vъ ~ -vъšь. —CodeCat 03:55, 5 February 2014 (UTC)

Is this use similar to OCS where present active participle -y/-ę was for masculine and neuter, and -ǫšti/-ęšti for feminine? So it would be that *-ę is for masculine and neuter and *-ęť is for feminine? But wouldn't then reconstruction look like *-ęťi for femenine?89.201.150.245 08:00, 5 February 2014 (UTC)

Not quite like that. The stem of the word was -ǫť- or -ęť-, from Proto-Indo-European -ont-(ye)- (or -ent-(ye)- for athematic verbs). It was an old consonant stem adjective, the only one of its kind to survive in Slavic (all other adjectives became o-stems), so its masculine nominative singular originally had the ending -ont-s, and the neuter had no ending at all (-ont) while the feminine had the ending -ont-ī (from PIE -ont-ih₂) in the nominative, and the stem -ont-ja- (PIE -ont-yeh₂-). But Proto-Slavic words couldn't end in a consonant, so the final consonants were removed in the masculine and neuter nominative singular, which gave just -on (which became -y, or -ę after a soft consonant). All the other forms did have an ending though, so then the final consonants were not removed. This is the same as consonant stem nouns, like *kamy ~ *kamen-, *slovo ~ *sloves-, *otročę ~ *otročęt- and so on. —CodeCat 14:21, 5 February 2014 (UTC)

Molly

I was sitting in my Japanese class last semester, wondering how Molly came to about as nickname for Mary. I was thinking of the Irish name Máire, and being in Japanese at the time, I thought about how English speakers sometimes hear the Japanese /r/ as /l/. Since the Irish slender /rʲ/ is also a tap (and one notoriously difficult for English speakers to get), I wondered if Molly was a result of English speaking Irish trying to pronounce the Irish name. I looked through the categories "English terms derived from Scottish Gaelic" and "English terms derived from Irish", and the only example that I found that could have been formed similarly was capercaillie (which is capal-coille in Gaelic). Does anybody know how "Molly" came to exist?--Gormadoc (talk) 04:35, 6 February 2014 (UTC)

Nicknames tend to play games with the sounds of the original names, so there doesn't seem to be any need to look outside of English for an origin (there are parallel examples like Hal from Harold, and Sally from Sarah). Chuck Entz (talk) 21:47, 7 February 2014 (UTC)

squick, squicky

Our entry for (deprecated template usage) squick claims "Believed to have originated as an onomatopoeia…". Our entry for (deprecated template usage) squicky claims "Blend of squeamish and icky". I assume not both of those are correct. Anyone know which is?​—msh210℠ (talk) 19:53, 7 February 2014 (UTC)

Eh, neither is very wrong, I think. I'd call squick an instance of w:sound symbolism (of which onomatopoeia is the best-known type, but not the type in question here, so the misnaming is understandable). The w:phonaesthemes which this sort of sound symbolism is based on arise from clustering patterns in extant vocabulary (much of it sound-symbolic itself, but some of it not): so in squick I suppose we have a squ- phonaestheme that draws from squeam and an ick phonaestheme that draws from ick(y). 4pq1injbok (talk) 11:30, 20 February 2014 (UTC)

*korenь

See also: Etymology scriptorium/2013/December

Reconstruction *korenь should be moved to *korę. *korenь is originally accusative of *korę.

Božanski boj, Radoslav Katičić, page 65 for reference. 95.178.211.114 20:42, 7 February 2014 (UTC)

Appendix:Etymology/cowabunga

RFV of the etymology, specifically this part:

• While initially believing Cowabunga held its origins in the nonsense word "kawabonga", modern linguists now believe it originated from the ancient Native American exclamation Kwa Bungu

It references " Harold Orbach, US Urban Indians as a Diaspora in Process of Creation, Kansas State University, US", which seems to be a presentation at a symposium in 1999 by a non-linguist.

There are several red flags here: it says "modern linguists now believe", which seems like an appeal to authority, but is so vague as to be meaningless (a quick Google search for "Kwa Bungu" turns up unrelated hits for African languages and discussions that are either quoting or cribbing from us). Given that most American Indian languages are relatively recently attested, it's usually meaningless to speak of a term in one of them as "ancient". Finally, it doesn't mention which language (there are hundreds in North America, alone). The fact that it was added by an IP who tweaked the text a few times, but otherwise made no other edits, is also suspicious.

I don't have access to the referenced work, but a reference to an "ancient Indian phrase, 'Kwa Bungu'" sounds like the kind of joke someone might make during an oral presentation, knowing that everyone there would get the w:Howdy Doody reference.

At any rate, there's nothing in the etymology to suggest that the Howdy Doody usage wasn't an independent coinage, so it's probably irrelevant. If it hadn't been there for over a year and quoted outside Wiktionary, I might have just removed it. Chuck Entz (talk) 21:33, 7 February 2014 (UTC)

Stone translations

Discussion moved from Wiktionary:Tea room/2014/February.

When I saw this diff, my BS meter pegged. Strangely, the Palauan seems to check out (the Palauan is here, way at the end of the WMP list, just before CMP). The Parthian was added in this diff. The etymology at سنگ disagrees, but the Avestan translation that was added along with the Parthian one makes it more plausible, and it may be that the etymology at سنگ has Parthian and Middle Persian confused. Can someone with access to the proper references check this out? If real, this has to be one of the great strange coincidences of all time. Chuck Entz (talk) 14:06, 6 February 2014 (UTC)

I have created ʾsng, expanded سنگ with references and replaced Mahjoob's unreliable translations. --Vahag (talk) 21:21, 8 February 2014 (UTC)

I knew something was wrong, but I didn't have the resources to pin down exactly where. Thanks! Chuck Entz (talk) 03:33, 9 February 2014 (UTC)

jumbo

I know that this word comes from P. T. Barnum's outrageous imagination, but is this perhaps related to mumbo jumbo, or to Swahili jambo (“hello”) (which was often used with whites, more so than with other Africans)? Etymonline and Wikipedia have quite a lot of theories, some of which seem quite implausible to say the least. —Μετάknowledge^{discuss/deeds} 18:52, 9 February 2014 (UTC)

German Bildung

This term comes from the verb bilden; but the verb is, according to our entry, formed on Bild (“image, picture”). I don't see the semantical link. Isn't it rather the same root as English build; the metaphor would then be the same as in French édification (building of a person, education)? --Fsojic (talk) 22:45, 9 February 2014 (UTC)

To start with, build is from an umlauted back-rounded vowel (OE bylden)- in modern English there are a couple of fairly obscure nouns from the same root: bold and bottle. Chuck Entz (talk) 23:02, 9 February 2014 (UTC)

Bild (image) -> bilden (to create an image), cf. abbilden, einbilden -> bilden (to form, to construct). Wyang (talk) 23:21, 9 February 2014 (UTC)

rose

Is it possible a Proto-Germanic origin of rose as in word angel (engel)? So not necessary from directly from French. For German Rose isn't inherited from French, so maybe is possible a Proto-Germanic word, which would eventually came from Latin. — This unsigned comment was added by 89.201.199.224 (talk).

The Online Etymology Dictionary says that English rose comes from Old English rose, in turn from Latin rosa. --WikiTiki89 21:53, 15 February 2014 (UTC)

I traced our seemingly incorrect etymology to these two edits by 86.211.238.254 (talk). So I think that is enough reason to change it back. --WikiTiki89 22:00, 15 February 2014 (UTC)

Boswith/Toller shows rose as Old English, and states it's from Latin. Given that the IP geolocates to France and those were their only edits, I think a one-off POV hit looks like a real possibility. Chuck Entz (talk) 22:25, 15 February 2014 (UTC)

I don't know; the fact that the word is rose /ɹoʊz/ and not roose /ɹuːz/, as would be expected due to the Great Vowel Shift, makes me suspect the Modern English word is a loanword from French and not an inheritance from Old English rōse. On the other hand, if Old English actually has rŏse (and the o is apparently short in Latin, so why would it be lengthened in OE?), then the modern word could be inherited. —Aɴɢʀ (talk) 15:45, 16 February 2014 (UTC)

@Angr According to the OED, it is likely that Old English had both rōse and rŏse as variants (as well as the unattested masculine variant rōs). And it is unknown whether it was reborrowed, influenced, or entirely uninfluenced by the French rose. --WikiTiki89 22:33, 16 February 2014 (UTC)

bumbershoot

A severely damaged product of German origin or simply respelt thus for extra humour? —Μετάknowledge^{discuss/deeds} 23:01, 17 February 2014 (UTC)

What would be the supposed German origin? --WikiTiki89 23:27, 17 February 2014 (UTC)

*Bumber + Schutz. I can't tell what the first half is gunning for. —Μετάknowledge^{discuss/deeds} 00:48, 18 February 2014 (UTC)

I thought you might be thinking of Bummel + Schutz, but even that's a major stretch. One analysis I've seen is umbr- > um-ber- with the initial consonant of the second syllable copied to the beginning of the first. The last syllable seems to be random, or possibly from parachute. Chuck Entz (talk) 01:03, 18 February 2014 (UTC)

Yeah, I find parachute to be more likely than Schutz. --WikiTiki89 01:10, 18 February 2014 (UTC)

OCS orthography

Does anyone know a basic rules of OCS orthography? Or maybe has a some kind of online source? Please. 89.201.143.210 13:00, 22 February 2014 (UTC)

You mean for normalization? Or a purely descriptive analysis of the orthography actually found in the manuscripts? —Aɴɢʀ (talk) 13:40, 22 February 2014 (UTC)

Both I guess, anything would be great. 89.201.153.46 18:58, 22 February 2014 (UTC)

Well one place to start is w:Early Cyrillic alphabet and w:Glagolitic alphabet. --WikiTiki89 20:24, 22 February 2014 (UTC)

But this is only about script (alphabet or azbuka), not about orthographic rules (assimilation, morphology...). 89.201.153.46 21:57, 22 February 2014 (UTC)

Morphology is not part of orthography. Assimilation can be said to be part of orthography only to the extent that some orthographies do not represent or over-represent the assimilation in the spoken language. I think what you really want here is not an orthography of OCS, but a grammar. w:Old Church Slavonic talks about it a little. You can also look at our OCS entries and the conjugation/declension tables on them. If that is not enough, there are plenty of books on Amazon on OCS grammar. --WikiTiki89 22:05, 22 February 2014 (UTC)

I can definitely recommend the Old Church Slavonic grammar by Horace Lunt. It's very complete. —CodeCat 22:21, 22 February 2014 (UTC)

macaque

RFV of the etymology, which has been copied to other entries:

From French macaque, from Portuguese macaco, from Bantu makaku (“some monkeys”) ‘some monkeys’, from ma- numeric prefix + kaku ‘monkey’.

Anyone who knows anything about Bantu languages will tell you this is nonsense. First of all, Bantu is a family, not a single language, so it can't have words (the work-around was to give makaku a language code of "und"). Secondly, there's no such thing as a "numeric prefix" in Bantu languages- there are noun-class prefixes, which can indicate that the term belongs to a specific plural noun class as a plural.

That said, there are obviously traces of a real etymology in there, copied from a real dictionary, but it would be nice to make a correct and complete etymology out of it. Apparently Kongo has a root kaku which refers to monkeys, and takes ma- as a plural noun-class prefix, so it's definitely a prime candidate, but are there others? Are there reliable etymological sources that explicitly say which language(s) the Portuguese term is borrowed from?

Once we have the correct etymology,Special:WhatLinksHere/makaku seems to indicate that the etymology has only been copied in this form to makak and macaco, so it shouldn't be hard to fix all of the collateral damage. Chuck Entz (talk) 06:43, 23 February 2014 (UTC)

 Done. — Ungoliant ^(falai) 04:04, 28 February 2014 (UTC)

Etymoline says the opposite. --kc_kennylau (talk) 09:28, 28 February 2014 (UTC)

To say that it is of unknown origin seems entirely wrong. Some kind of African origin would be consistent with a pattern of word transmission via Portuguese shared by many words for plants and animals.

That we don't have very exact information that fits our language naming scheme for etymologies says much more about the limitations of our language-naming scheme than about the previous etymology. There is similar imprecision for many words for New World plants and animals. We obviously need some way of reflecting inexact knowledge. A system of geographic specifiers might help. DCDuring TALK 12:34, 28 February 2014 (UTC)

Kc kennylau, surely Portuguese etymological works are better sources for Portuguese etymology than English ones.

DCDuring, see macaco#Portuguese. Africa is not even the only proposed continent of origin.

— Ungoliant ^(falai) 17:18, 28 February 2014 (UTC)

I think that of unknown origin fails to provide the clue that the pt entry has a lot more. Perhaps "See macaco." would help. DCDuring TALK 17:54, 28 February 2014 (UTC)

Good idea; done. - -sche (discuss) 18:24, 28 February 2014 (UTC)

I think the evidence is very strong for a Bantu origin, since there's at least one language that has makaku as the plural form of the word for monkeys.

As far as the description of ma-: it's really a declensional morpheme analogous to the nominative plural ending for a given declension in Latin, but as a prefix instead of a suffix. You wouldn't describe "-as" as a suffix that "indicates plurality in Latin and several Romance languages". Noun classes are correlated with a variety of semantic information in a similar way to how declensions are correlated with gender in Latin, with a given singular noun class paired with a specific plural class.

As for the limitations of our language-naming scheme: a correct way to present the previous etymology would be to say: "ultimately from a Bantu language, some of which have words for monkeys with something like *makaku in the plural". There's no problem giving the code for Bantu, as long as we make clear it's a family, not a language, but the {{term}} template shouldn't be used for the asterisked forms, since we don't know what language they would link to. As it turns out, Kongo has the appropriate form, so a better version would be: "ultimately from a Bantu language, compare Kongo kaku (plural makaku)". It would be a good idea to get input from someone who's actually worked with Bantu languages, so I'll ping @Metaknowledge and @CodeCat. Chuck Entz (talk) 19:59, 28 February 2014 (UTC)

I'm not really sure what more I can contribute here; I think Chuck's suggestion is good (perhaps with slightly improved wording). —Μετάknowledge^{discuss/deeds} 02:59, 1 March 2014 (UTC)

Appendix:Proto-Finnic/apu

I'm a bit confused by the descendants. They are clearly related, but one half of them ends in -u while the other half ends in -i. Does anyone know how that could have happened? —CodeCat 00:42, 26 February 2014 (UTC)

I don't know whether this is a regular process in the languages in question, but I know from many languages that /u/ > /i/ is not uncommon (this can be seen for example in dialects of Yiddish, French, Arabic, etc.). --WikiTiki89 06:23, 26 February 2014 (UTC)

I don't know of any regular change from -u to -i in the Finnic languages. I do know that in Proto-Uralic rounded vowels couldn't occur in non-initial syllables, so that would point to -i being original. But that doesn't really explain why there are two different forms. Finnish has both auttaa and avittaa as synonyms, too. So maybe this should just be reconstructed as a doublet? —CodeCat 14:04, 26 February 2014 (UTC)

These are two parallel derivatives: North Finnic *apə-w > *apu, versus Central Finnic *apə-j > *api. I do not know if the former stem type underlying Livonian a'b is known (though the latter would predict *api > **ä'b). --Tropylium (talk) 03:18, 21 March 2014 (UTC)

*rǫža

Should we include Proto-Slavic reconstruction *rǫža ('rose')? The word is from Latin rosa, but this is an early (Proto-Slavic) borrowing. We have Polish róża, Czech růže, Slovenian roža, Croatian ruža..., so by my humble opinion we should include this reconstruction, shouldn't we? 95.178.193.111 01:12, 1 March 2014 (UTC)

The phonetics are strange though. Why does it have a nasal vowel and a ž, when the Slavs could have borrowed it as "roza" instead? Also, ǫ doesn't become ů in Czech as far as I know, so that doesn't add up either. —CodeCat 03:10, 1 March 2014 (UTC)

Nor does it become ó in Polish. In fact the ó/ů is the most confusing thing for me as I don't see why the vowel would have been lengthened no matter what it was originally. I suppose the ż is parallel to the sz in words like Mateusz, which I don't know the explanation for but it could have the same explanation. --WikiTiki89 03:19, 1 March 2014 (UTC)