Talk:Ruki sound law

There was an error in the article: but the change s > š happened after, not before r, u, K, i. I have fixed it and added some more information.

--Grzegorj 17:18, 6 August 2005 (UTC)

I've cleaned this article up a little, but there are some things I have questions and concerns about;

• How much do we really know about the precise value of the retracted sound?
• When we say the sibilant was "probably an apical sibilant", does that mean apico-alveolar? Or apico-postalveolar? (Some people think the Spanish apico-alveolar sounds like ʃ, but there is a difference.)
• Shouldn't we use either š or ʃ throughout, instead of using both? Which one should we use? Philological tradition I think would warrant the use of š; and if we don't know the precise value of š, shouldn't we use š instead of the IPA?
• What is meant by "It associates with a Russian word which means 'hands' or 'arms'"?
  • I'm guessing it's referring to the fact that since "ruki" has a meaning in Russian, that fact can be used as a mnemonic to remember after which sounds the change occurred. -- pne (talk) 14:41, 30 April 2008 (UTC)
• What is meant by "It was later proposed to be valid for all Satem languages, except for Indo-Iranian languages"? Obviously it is also valid for Indo-Iranian; after all, that's where the rule was first observed. Perhaps the writer meant "not just the Indo-Iranian languages"? I'm not changing it, since I'm not sure what was meant.
• The "Exceptions in Indo-Iranian languages" section does not seem to actually deal with exceptions. It should be renamed. There probably are some exceptions, which could be added.
• "[T]he two sounds must have been very close (r/l), so that both could have triggered the change in Indo-Iranian" -- I don't see how that's a "drawback" to the theory. Has it been cited as one in other sources?
  • And nobody is requiring [l] to have triggered the change. PII may simply have merged the liquids before ruki occured.
• "[A]nd what's more, there are no real examples of this change working in Slavic" -- what change is meant? The merger of l and r?
  • s > š after PIE l probably.
• Finally, I changed all the examples of phonemes/allophones to italics with no marking (e.g. //, []) around them. I'm not sure if they should be treated as phonemes or allophones or some other more vague category. Also, I couldn't figure out how in MediaWiki to put brackets around something that is double-bracketed :)

71.82.214.160 01:34, 4 June 2007 (UTC)

• And one new issue...

This velar fricative changed back into š before a front vowel or j.

I'd be interested in knoing if there's any proof that the sound ever was [x] in this context - couldn't š → x have been the conditional part? Or was it that the yers were elided in-between, thus giving some modern [x] from s + i...? --Tropylium 08:40, 4 October 2007 (UTC)

Since a number of vowels (I think) became proto-Slavic front vowels before the first regressive palatalization that changed velar consonants to postalveolar ones, there would be variance as to the conditions for the appearance of *š as opposed to *č and *ž. There might also be evidence from loanwords, but I'm not sure. I'm also not sure of the evidence that *š was an intermediate step for Proto-Slavic other than that that's the realization in other language families. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 14:20, 4 October 2007 (UTC)

Exceptionless

I edited 'unexceptionless for Indo-Iranian languages' to 'exceptionless for Indo-Iranian languages', on the assumption that what was meant was that there are no exceptions in Indo-Iranian languages. If I'm wrong--if there are significant exceptions in these languages--then I suggest replacing "unexceptionless" with something like "there are few/some/numerous/ exceptions". Having said that, there is a confusing discussion later on of the exceptions in Indo-Iranian languages, so maybe my guess was wrong. Mcswell (talk) 21:07, 28 September 2008 (UTC)

Examples

It would be nice to have some examples of words that changed. -- pne (talk) 14:41, 30 April 2008 (UTC)

Here's one example:

PIE *h₂ous

• Proto-Slavic *uxo

Russian: ухо

Polish: ucho

Bulgarian: ухо

And some Centum languages to compare to

Latin: auris

Proto-Germanic: *auson

I'm sure they're relatively easy to find at wiktionary. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:44, 30 April 2008 (UTC)

Vowels i and u?

The article states that RUKI applied after the semivowels *y and *w. But these also had vowel allophones *i and *u in PIE. Did the law apply after those too? For example, would a hypothetical *dus- become *duš-? What about the vowel allophones of *r? CodeCat (talk) 13:56, 3 May 2016 (UTC)

@Rua:, yes, it does happen after syllabic r, i, and u, e.g. Proto-Indo-Iranian *tŕ̥šnas < *tr̥snós or the Proto-Balto-Slavic locative plural endings *-išu and *-ušu in i- and u-stem nouns. —Mahāgaja · talk 08:19, 1 August 2020 (UTC)

re Albanian

TBH this does not really seem like RUKI to me — especially since there are additionally other, fairly neutral positions where *s > sh. These include *st > sht, *skr > shkr, *sk > shk before voiceless stops, and s → sh in loanwords from Greek and Latin. I'm not going to add this as my own WP:SYNTH, but I wonder if there's already a source out there saying the same about Orel's proposal — e.g. a review of his book? Do any experts in Albanian historical linguistics happen to be reading? --Trɔpʏliʊm • blah 19:10, 7 December 2019 (UTC)

Nuristani

I rcall seeing a critique of Hegedűs that pointed out the middle Indo-Aryan merger of ṣ and s (which essentially undoes RUKI, possibly aside from *kṣ and/or (*ḱs >) *tṣ > ch) and suggests that this may have happened in Nuristani as well. Does this ring a bell for anyone else? (It's possible I'm only remembering a discussion instead of a published work.) --Trɔpʏliʊm • blah 17:28, 10 February 2021 (UTC)