Talk:Indo-European s-mobile

North Germanic bull
In the tauros/steer example, there has been some to-ing and fro-ing about mentioning Icelandic. The Icelandic word for bull, as Nema Fakei pointed out, is þjör, which "bucks the Germanic trend" (thus NF's edit summary). Angr then pointed out that Old Norse has stjórr. The trouble is, if all other Germanic languages have the s, including the main current of North Germanic, it is unlikely that Icelandic has preserved the PIE form without it. It is more likely that Icelandic has lost the s in some other way. I would say the jury is still out on this, so better not to mention it. Apart from that, though, we have to resist the temptation to overload an example with all possible details, whether they help the point which is being made or not. Even if Icelandic does preseve the form without an s, that doesn't change the validity of the statement that the form with the s is preserved in Germanic, which contrasts with Latin, Greek and Slavic, which is all that needs to be said here. A discussion of Icelandic at this point can only make it harder of the reader to glean the point that is really at issue. --Doric Loon 20:59, 4 December 2006 (UTC)
 * A better phrasing would be "unlike modern Scandinavian". Losing the s might be possible, but I don't know whether it's more probable than two parallell forms, since North Germanic loss of initial s is a very rare sound evolution, as far as I'm aware of... 惑乱 分からん 23:19, 4 December 2006 (UTC)
 * This word has been passed down as 'tjur' in Swedish, 'tyr' in Danish , and 'þjór' (rare) in Icelandic. The first link says Old Norse 'þjórr' and 'stjórr' (poet.) so it seems that the form with the s may have existed in Old Norse but was rare. Stefán 23:58, 4 December 2006 (UTC)
 * Yeah, I think that could be interpreted as two parallell forms. I thought about checking out Svenska Akademiens Ordbok for further etymology, but they haven't arrived at entries starting with tj, yet. 惑乱 分からん 00:19, 5 December 2006 (UTC)
 * Okay, the tj entries have finally arrived. Well, they do connect the Swedish word to both "steer" and "taurus", but they don't discuss it any further. 惑乱 分からん * \)/ (\ (&lt; \) (2 /) /)/ * 17:27, 13 April 2007 (UTC)
 * I think another indication for a PIE origin is the alternative Norse forms with stj- and þj-, which indicate a PIE origin. If the s- was dropped later than Grimm's law came in effect, I think a Norse/Icelandic tj-form of the word would be mnore likely (the t in modern Scandinavian is due to a much later sound shift.) Cf. English thane, Old Icelandic þjónn (servant) (PIE tek-, beget) En thief, OI þjófr (PIE teup- crouch(?)) En thigh, OI þjó (PIE teu- swell) OE þeod, OI þjóð (people, nation) (PIE teuto-, people). Cf. En tar, OI tjara (PIE deru-, tree) En tether -> ON tjoðr (PIE deu- fasten +tro- -er ) There are other examples of North Germanic root words lost or not attested in the other Germanic languages, such as Old Icelandic vár for spring season. (Old Icelandic forms instead of Old Norse, since I found a good free dictionary online, the differences are however minimal). þjórr inherited directly from PIE sounds plausible. 惑乱 分からん * \)/ (\ (&lt; \) (2 /) /)/ * 10:28, 14 April 2007 (UTC)

Claudere/schliessen
Could Latin "claudere" and German "schliessen" be cognates going back to some root with S-mobile? There has been etymological connections proposed before, but maybe they're considered deprecated, possibly *(s)k(l)u- (cling, clasp, close) Cf. Lat claudere, Old High German sliozzan, Greek kleio, Old Church Slavonic kljucu etc... 惑乱 分からん 14:26, 3 January 2007 (UTC)


 * Yes, they are connected - see Pokorny p. 604. Hey good example, well done! --Doric Loon 22:21, 3 January 2007 (UTC)


 * Yeah, I read some pages suggesting it. If it's an example that should be cited, (possibly, the selection is fine, already) we'd need a credible source connecting the root to S-mobile, I don't know if any languages other than Germanic has an S-form... 惑乱 分からん 07:25, 4 January 2007 (UTC)


 * I've put it in the table. Surely Pokorny IS a credible source?  He writes quite specifically "mit beweglichem s".  --Doric Loon 11:53, 4 January 2007 (UTC)


 * Hmm... schließen goes back to Proto-Germanic *sleutan, doesn't it? Does PIE *skl- regularly become *sl- in Germanic? (It might, as I'm pretty sure Germanic doesn't have any words with *skl-.) —Angr 13:08, 4 January 2007 (UTC)
 * Hmmm, good point. Norwegian has skli (slide), but it seems to be some variant form of PG *skrīðan. (Cf. German schreiten, Old English scriðan. 惑乱 分からん 14:49, 4 January 2007 (UTC)

s+sonorant in Latin
We have to be careful with sonorant-initial examples in Latin like nix and nurus, because *sl sm sn became *l m n in Latin by regular sound change anyway. If no language has examples unambiguously pointing to *neigwh- and *nuru- instead of *sneigwh- and *snuru-, we can't assume s-mobile here. —Angr 13:06, 4 January 2007 (UTC)


 * Good point, which probably should be noted in the article somewhere, perhaps underneath the table. But when I made up the original list of examples (to which claudere is the only one we have since added) I restricted myself to cases where Calvert Watkins notes the PIE root with the s in brackets.  I have tried to stick to one example of a derivative for each field in the table, and to focus on fairly well known languages (plus Gaelic because I am strictly POV there - love is blind), but to give a full picture, more examples would certainly be needed.  --Doric Loon 20:08, 4 January 2007 (UTC)
 * Well, Watkins gives "snow" as *sneigwh-, not **(s)neighwh-, and *snuru- isn't in his dictionary at all. —Angr 21:10, 4 January 2007 (UTC)
 * Doesn't Watkins list only contain roots with derivations found in modern English? 惑乱 分からん 00:05, 5 January 2007 (UTC)
 * Yes, and AFAIK *snuso- (I just remembered the -r- in both Germanic and Latin is secondary) has no descendants in Modern English. But he does mention it in the preface, and doesn't parenthesize the s. On the contrary, he suggests (but doesn't come out and say so directly) that sn- > n- is phonological in Greek νυός and Albanian nuse as well as Latin nurus. So Greek and Albanian are two more languages where the absence of an s before a sonorant (or at least before n) doesn't prove an s-mobile. —Angr 00:22, 5 January 2007 (UTC)
 * I moved the uncertain examples to a separate table and added two examples from LIV. --ἀνυπόδητος (talk) 18:23, 29 September 2009 (UTC)
 * Looks good. Further comments? 惑乱 Wakuran (talk) 19:32, 7 November 2009 (UTC)

Suggested move to S-mobile
Since no form of "s-mobile" exists outside of IE, we should keep it at as short of title as possible. Does anyone oppose this? The ikiroid (talk·desk·Advise me) 17:44, 18 March 2007 (UTC)


 * No objection if you are certain there is no form outside PIE. However, I have learned to be very wary of such assumptions.  Especially since people have a tendency to broaden terms: what about the negating proclitic s- in Italian (which is actually a remnant of Latin dis-)?  I would be loathe to see people muddle this article by adding info on vaguely parallel phenomena which as a linguist I think of as quite different.  Wikipedia has two kinds of articles on linguistic phenomena - broad comparative ones (like apophony) and specific ones (like Indo-European ablaut) and it is important to keep these different, so we can have both the big picture and depth.  And my experience is that the specific ones only stay specific if the title doesn't  allow them to be anything else.  --Doric Loon 21:44, 18 March 2007 (UTC)

Origin
Is it possible that s-mobile
 * (A) comes from the s which appears at the end of prepositions in Greek and Latin:
 * Greek: eks, eis <*ens, pros
 * Latin: abs, ex, trans   ?


 * (B) is a marker of a verb of motion or movement?Tudhaliya 18:02, 10 May 2007 (UTC)

There is a connection between (A) and (B) in that the prepositions eks, eis, pros and trans all show movement either out of, into, toward, or through a place. Since eis is from *ens (en+s) and pros is from pro+s, it seems as though s might be an ending showing movement.Tudhaliya 15:46, 1 June 2007 (UTC)


 * From the article:

One theory of the origin of the s mobile is that it was influenced by a suffix to the preceding word. Since the nominative of Indo-European nouns often ended in *-s, it follows that verbs were frequently preceded by this phoneme. If the verb itself began with an s-, the result was a double s, which was eliminated in late PIE through a simplification of geminates which is also observable elsewhere in the language. Obviously this could not happen to related forms which were used in different syntactic positions. By this view, the forms with the *s- are original. This seems more likely than the opposite view, that roots which originally had no *s- gained one by assimilation from the preceding suffix.
 * I don't think there seems to be any evidence for any of the ideas. Especially proposal (B) looks strange to me. Also, I don't think prepositions in PIE necessarily ended with an s, just because they later did in Greek and Latin. 惑乱 分からん * \)/ (\ (&lt; \) (2 /) /)/ * 19:26, 18 May 2007 (UTC)


 * Without published sources to back it up, this is original research and doesn't belong in the article anyway. —Angr 20:23, 1 June 2007 (UTC)

S-movable
Angr, you questioned the sentence that this is sometimes called s-movable. I'd never heard it either, but read it today in Sihler's comparative grammar of Latin and Greek. That would be the reference if you feel it is needed. On the other hand, I don't actually like "s-movable", and since it is rare, i could be persuaded to delete it. But I think it would be worth having a redirect for it. --Doric Loon 18:58, 25 July 2007 (UTC)
 * I could live with a redirect. The biggest problem with Sihler's comparative grammar is that it doesn't cite its sources, so it's pretty much a dead end. —Angr 19:11, 25 July 2007 (UTC)

OK, I'll do that. --Doric Loon 22:01, 25 July 2007 (UTC)

S-mobile related to Klingon?
Is it possible that the s-mobile is like in that Klingon language? Just wonderin'. —Preceding unsigned comment added by 24.77.205.229 (talk) 01:51, 10 March 2008 (UTC)
 * Only if the guy who invented Klingon got the idea from Proto-Indo-European. —Angr If you've written a quality article... 06:05, 10 March 2008 (UTC)

Analogy to few cases of a "moveable n" in Modern English?
Is it worthwhile to compare this phenomenon to the few cases of a "moveable n" which occur in Modern English?

Namely:

"nuncle" from "an uncle"/"mine uncle" -> "a nuncle"/"my nuncle" "orange" from "a norange" -> "an orange"

Or, in spoken vernacular:

"nother" from "an other" -> "a nother"


 * Only if you can find a reliable, published source where such a comparison has already been made. —Angr 11:11, 4 May 2008 (UTC)


 * nickname comes from Middle English an ekename "supplied name" or the like; from The Cassell Concise English Dictionary, ISBN 0-304-31806-X. Nice analogy, that. --ἀνυπόδητος (talk) 18:59, 14 November 2008 (UTC)

Sparrow and Parrot ?
small brownish-gray bird, O.E. spearwa, from P.Gmc. *sparwan (cf. O.N. spörr, O.H.G. sparo, Ger. Sperling, Goth. sparwa), from PIE *sper- (cf. Cornish frau "crow;" O.Prus. spurglis "sparrow;" Gk. spergoulos "small field bird," psar "starling"). http://www.etymonline.com/index.php?term=sparrow

c.1525, perhaps from dial. M.Fr. perrot, from var. of Pierre "Peter;" or perhaps a dial. form of perroquet (see parakeet). Replaced earlier popinjay. The verb "repeat without understanding" is first attested 1596. http://www.etymonline.com/index.php?term=parrot

Evidently not an example.

Virginia-American (talk) 11:32, 31 May 2008 (UTC)
 * Parrot isn't mentioned in the article. Sparrow is compared to Latin parra, which, if your information is correct, is not the etymon of parrot. —Angr 13:43, 31 May 2008 (UTC)

more examples from the Greek language
"google translate" translates κόπτω as "skive" and κόβω as "pluck" |en|%CE%BA%CF%8C%CE%B2%CF%89%0A babelfish κόβω as "i cut". lol i think "skive" is also PIE from the same root with a kept "s"!
 * (s)kep- cut, scrape -	English scab -	Late Latin capulare ('cut'),--- Greek κόβω (kohvo) which comes from the (ancient) Greek κόπτω (kohpto), means "to cut".

|en|%CE%BA%CF%85%CF%81%CF%84%CF%8C%CF%82 and babelfish.com as "convex".
 * (s)ker- bend - English shrink -Latin curvus ('curved'),--- Greek κυρτό (kerto) means bent.

|en|%CE%BA%CE%BB%CE%B5%CE%AF%CF%89 google translate gives both κλείνω and κλείω as "close".
 * (s)kleu- close -German schließen -Latin claudere, --- Greek κλείνω (kleno), ancient Greek κλείω (kleo)


 * (s)teh2- 	stand 	--- |en|%CF%83%CF%84%CE%AD%CE%BA%CF%89 Greek στέκω "stekho"


 * (s)wagh- 	resound 	English sough --- |en|%CE%B7%CF%87%CF%8E%0A Greek ηχώ, echo.

and many more..

sorry if my whole post was unnecessary, but i couldn't help it, i'm a Greek-speaker and they were too obvious. i always do that when i see PIE stuff..150.140.229.175 (talk) 04:13, 8 October 2008 (UTC)
 * Hmmm, yeah, I guess some of these could be added when there aren't any sufficient examples already, such as ηχώ "echo", if it's valid. 惑乱 Wakuran (talk) 02:40, 6 January 2009 (UTC)

SVO?? Huh??
One theory of the origin of the s-mobile is that it was influenced by a suffix to the preceding word. Since the nominative of Indo-European nouns often ended in *-s and it seems to have been an essentially SVO language, it follows that verbs were frequently preceded by this phoneme. The s-mobile can therefore be seen as an interference between the words, a kind of sandhi development. So for example, while an alternation between *pekyont and *spekyont (both meaning 'they saw') might be difficult to imagine, an alternation between *wlkwoms pekyont and *wlkwoms spekyont ('they saw the wolves' [1] -here incidentally in OV order) is plausible.

I'm pretty sure if there is any concensus as to word order in PIE, it is that it is SOV, not SVO. Though I would assure you, I'm not suggesting that there is a concensus... but surely not SVO. At the very least please say that there is no concensus on SVO. 192.250.175.26 (talk) 15:20, 28 December 2009 (UTC)


 * Actually, that makes sense, since in the example we give, the object comes before the verb. But does anyone have literature on it? --Doric Loon (talk) 17:21, 28 December 2009 (UTC)


 * According to the article Subject Object Verb, most of the archaic IE languages, such as Classical Latin, Ancient Greek, Sanskrit and Hittite seemed to be SOV. 惑乱 Wakuran (talk) 12:18, 8 July 2011 (UTC)


 * Fortson (IE Language and Culture): "It is almost universally asserted that most of the ancient IE languages were verb-final, and that PIE was as well; more specifically, they were SOV..." Fortson goes on to explain that the only strictly verb-final ancient IE language actually was Hittite, and that all others could place the verb in practically any position, regardless of register and other factors. I'd say this is basically the same view as 26 expressed above ("but surely not SVO"). --ἀνυπόδητος (talk) 17:29, 13 July 2011 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Indo-European s-mobile. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20030609163731/http://indoeuro.bizland.com/project/phonetics/ie6.html to http://indoeuro.bizland.com/project/phonetics/ie6.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 12:17, 13 November 2017 (UTC)

Siebs's law
I wonder why there is almost no explanation on how could the Sieb's law be reconсiled with this article, especially with the statement that *s- forms are more likely older than forms without *s-, since the law postulates occurence of any plosive with s-mobile, which couldn't be restored properly unless there were clearly spelled "*zd-" and "*zdh-" cases. Is it considered outdated or dubious? Anyway Wiktionary never relates protoforms like *spX/*bX. 213.208.170.194 (talk) 13:55, 21 June 2019 (UTC)

is l a nasal as it's currently claimed in the text?
referring to "l" as a nasal might need some additional clarification. It's commonly referred to as a lateral approximant. 2A01:C22:CD3E:4D00:8C54:93EC:DCCC:3188 (talk) 15:43, 2 December 2023 (UTC)