Reconstruction talk:Proto-Indo-European/ḱh₂d-

From Wiktionary, the free dictionary
Jump to navigation Jump to search

RFV discussion: April 2019–September 2021[edit]

This entry has survived Wiktionary's verification process (permalink).

Please do not re-nominate for verification without comprehensive reasons for doing so.

This "root" has no e in it, which makes it suspicious. IEW is 60 years old and thus not adequate as a source, and the Wiktionary page name doesn't match the form given in IEW anyway. —Rua (mew) 15:43, 11 April 2019 (UTC)[reply]

@Rua, this should just be deleted. --{{victar|talk}} 13:52, 27 May 2019 (UTC)[reply]
@Victar I added De Vaan as a more trustworthy source, but it's still somewhat dubious that there are no full grades anywhere. —Rua (mew) 13:57, 27 May 2019 (UTC)[reply]
Rua: I deleted the PII forms because those were all impossible. Now we're just left with the Latin and a couple dubious extra-Latin forms. --{{victar|talk}} 14:08, 27 May 2019 (UTC)[reply]
@Rua, Victar: I don't know about this particular root, but *bʰuh₂- is one root that doesn't seem to have had a full grade, so it wouldn't be without precedent. —Mahāgaja · talk 05:40, 22 June 2019 (UTC)[reply]
@Mahagaja: Like I said, I'm more concerned with the lack of indubious cognates outside of Latin, which is grounds enough for deletion. --{{victar|talk}} 23:44, 22 June 2019 (UTC)[reply]
What are the IE cognates of Skt. √śad ("fall, fall out, fall off; collapse; decay, wither, perish")? Hölderlin2019 (talk) 01:15, 24 June 2019 (UTC)[reply]
@Victar: Hölderlin2019 raises a good point. It's clear that Proto-Iranian *kat- can't be connected and should be deleted, but why is Sanskrit शशद (śaśada) = Latin cecidī impossible? It could even be evidence for a full grade *ḱh₂ed- (since zero grade *ḱh₂d- would have given לid- rather than śad-). —Mahāgaja · talk 08:56, 24 June 2019 (UTC)[reply]
Well for one, ×शशद (לaśada) does not exist -- it's actually शशाद (śaśāda), and that word reflects PIE *ḱe-ḱód-e, which itself points to a *ḱed- root. Secondly, I've only seen that word in the context of listing cognates for the Latin -- it's not in any of my Sanskrit dictionaries -- so it's very poorly attested which makes me question its meaning entirely and think that it's just a bunch of semantic massaging by Latinists. Not all Indo-European words need cognates, nor are all words actually from PIE. --{{victar|talk}} 13:09, 24 June 2019 (UTC)[reply]
It's listed in both Whitney and MW? Hölderlin2019 (talk) 14:55, 24 June 2019 (UTC)[reply]
I think you meant to write, "is it", and no, in Sanskrit the word शशाद (śaśāda) actually means "eating rabbits", as seen in Monier-Williams. --{{victar|talk}} 15:01, 24 June 2019 (UTC)[reply]
No, I meant that it's listed in both Whitney and MW. Hölderlin2019 (talk) 15:10, 24 June 2019 (UTC)[reply]
Ah, accidental question mark than. That's the danger of using older sources; often inflection tables are assumed and not based on actual sources. The only form that is found in Sanskrit, and again, very poorly attested, is (only found once in AV) शत्स्यति (śatsyati) and everything else is fabricated. So again, ×शशद (לaśada) did not exist. --{{victar|talk}} 15:20, 24 June 2019 (UTC)[reply]
It was a quizzical question mark. I don't know why you think the inflections are fabricated (by whom? Whitney/MW? the Indian grammarians?); both Whitney and MW assert that this particular one is attested in the Brahmanas. Hölderlin2019 (talk) 15:33, 24 June 2019 (UTC)[reply]
Inflection tables are made up all the time. We do so even on en.Wikt. Modern sources ({{R:ine:LIV}}) cite the word as शशाद (śaśāda) and not ×शशद (לaśada) as seen in MW. Also, MW mistakenly associates this word with the unrelated शीयते (śīyate, “to fall down”), which is from *ḱey-. --{{victar|talk}} 15:40, 24 June 2019 (UTC)[reply]
@Victar: But शशाद (śaśāda) does exist? If so, it can come from *ḱe-ḱod-e as you say, but surely it can also come from *ḱe-ḱh₂od-e, from a root *ḱh₂ed-, of which both the zero grade and the full grade could give Latin cadō. —Mahāgaja · talk 18:39, 24 June 2019 (UTC)[reply]
@Mahagaja: Is *ḰHeT- even a valid root in PIE? In PII, you might also expect the laygyeal to have some sort of aspirating power. --{{victar|talk}} 19:00, 24 June 2019 (UTC)[reply]
@Victar: We have entries for *kh₂em-, *kh₂eyd-, and *sh₂ey-, so the root shape seems to be rare but not impossible. As for aspiration, you'd expect it after a stop, but probably not after ś. —Mahāgaja · talk 19:17, 24 June 2019 (UTC)[reply]
@Mahagaja: I was asking *ḰHeT- (=*KʲHeT-), not *KHeT-. I don't think it is. Aspiration would have been pre-PII. --{{victar|talk}} 19:31, 24 June 2019 (UTC)[reply]
@Victar: In both cases I suspect there aren't enough examples to allow us to generalize. CHEC is a rather rare root shape to begin with, so the apparent lack of ḰHeC- could be coincidental. And even if *-ḱH- became *-śʰ- in PII, are there enough examples of PII *-śʰ- to be sure that it didn't simply become ś in Sanskrit? —Mahāgaja · talk 19:37, 24 June 2019 (UTC)[reply]
@Mahagaja: I found an example with *sḱeh₂i- ~ *(s)ḱh₂ey- ~ *(s)ḱeyh₂- (“cut open, sting”) which allegedly yielded σχάω (skháō) and ἔσχασα (éskhasa). Theoretically, I would think PIE *ḱe-ḱh₂ód-e would have yielded > śeśʰh₂óde > PII *ćaćʰHáda (=/tśatśʰHáda/) > PIA *śaśʰHáda > शहद (śaháda). Problem is, we lack clear examples of *ḱh₂. --{{victar|talk}} 20:28, 24 June 2019 (UTC)[reply]
My point exactly. We know that *źʰ became h in Sanskrit, but that doesn't mean *śʰ did. It may have become simply ś. —Mahāgaja · talk 20:40, 24 June 2019 (UTC)[reply]
@Victar: This discussion is surreal. Whitney does not "make up" inflection tables; Whitney does not deal in inflection tables, except in his introductory grammars. What he does do is catalogue attested Vedic + Skt. forms and assign them to roots. MW is not in error; he's simply following the synchronic analysis of the ancient grammarians, whose fiat assignment in this particular case Whitney explicitly discusses. The reduplicated perfect in question is multiply attested in the Brahmanas. I'm frankly astounded by how tenuous your grasp of how to understand the Sanskrit is, never mind the Sanskrit itself Hölderlin2019 (talk) 20:16, 24 June 2019 (UTC)[reply]
But Whitney doesn't list *शशद (śaśada); he lists शशाद (śaśāda), whose existence Victar isn't denying. — This unsigned comment was added by Mahagaja (talk • contribs) at 16:40, 24 June 2019.
For one, I've only been referring to MW, not Whitney. Secondly, I'm not infallible and if I misread some source, I'm happy for someone to point it out, but personally attacking me is unnecessary. And lastly, any author is also fallible, especially in older works when our understanding of PIE and Sanskrit was not as developed and defined as it is today. I'm not sure which part you are claiming MW isn't in error of, but Rix agrees that MW mistakenly lumped शीयते (śīyate, “to fall down”) into this root, which probably had bearing on the definition he gave for it and calls into question the semantic connection to the Latin. --{{victar|talk}} 20:51, 24 June 2019 (UTC)[reply]
For what it's worth, {{R:ine:LIV}} reconstructs the root as *ḱad- (*ḱád-e-ti > cadō) and {{R:ine:LIPP}} as *ḱed- (*ḱd-é-ti > *ḱₔd-é-ti > cadō). --{{victar|talk}} 19:00, 24 June 2019 (UTC)[reply]
That could work too for those of us who believe that PIE primary a was merely rare but not nonexistent. —Mahāgaja · talk 19:17, 24 June 2019 (UTC)[reply]
@Mahagaja: Yeah, I'm not a fan of PIE a entries either. I think, however, there is some argument to be made that it existed natively in onomatopoeic roots. My preference would be for *ḱed- though. --{{victar|talk}} 21:01, 24 June 2019 (UTC)[reply]

It seems fairly clear that this root existed, with descendants in three indo-european branches, the only question is its shape. I'm in favour of *ḱad-, because it's the simplest shape that explains all forms, with *ḱh₂ed- as alternative (problematic because of Greek). The existance of *a in PIE, though rare, is beyond doubt at this point.


Retrieved from "https://en.wiktionary.org/w/index.php?title=Reconstruction_talk:Proto-Indo-European/ḱh₂d-&oldid=64048781"