Jump to content

Talk:h₂e-conjugation theory

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

undue emphasis on Olivier Simon's work and possible conflict of interest

[edit]

The section "Some new developments" states definitively that Simon's study "unifies the Hittite ḫi-conjugation with the Proto-Indo-European mainstream model". I've read the study and I don't think it does that. It posits a model for unifying them but that doesn't mean the problem has been solved, it's just one person's idea for how it might be solved. From what I can tell this work is self published and has not been cited elsewhere by any of the experts in this area, like Jasanoff or Kortlandt, who might offer an opinion on whether it's correct. There's also the fact that this text was added to the article by Mundialecter, who I'm fairly certain is Simon himself (he goes by this name on Reddit and other places online). Self publishing a work with a theory on solving a long standing problem in a field, then editing an article to claim the work has solved the problem, seems like a conflict of interest to me. Cyllel (talk) 21:24, 11 February 2024 (UTC)[reply]

I've been bold and removed the section Simon added. The more I read of his work the more I realize he is an enthusiastic dilettante. It is not appropriate for him to insert his own work here. --Cyllel (talk) 23:36, 18 June 2024 (UTC)[reply]
Hi the dilettante does greet you :-). What I do regret is that you delete the reference to Kloekhorst's work. It's him who shows, in his PhD thesis, the identity between the Hittite H2e conjugation and a PIE unreduplicated perfect (stative) conjugation like *woidH2e (while Jasanoff is unable to explain this one and just treats it, in his 2003 book, as a latter development, though it's found in a lot of IE languages....). I only show (I list) tens of reduplicated (nearly exclusively) Vedic and Old Greek reduplicated perfects (with the famous -H2e endings) having a medio-passive meaning though their conjugation belongs to the regular set of active conjugations in their respective grammars. What matters is to understand how PIE worked and to analyse the available reconstructed material, not whether someone is famous or self-published... (by the way, Pr. Kortlandt himself self-publishes a lot) Mundialecter (talk) 11:18, 15 August 2024 (UTC)[reply]
Thanks for your reply. I'm sorry for my dismissive language before, thank you for taking it in good humor. I'm not going to get into the details of the issue, the only important point is that this is an encyclopedia, not a forum for discussing or promoting personal theories. What matters for Wikipedia is not, in fact, understanding how PIE worked, what matters is citing reliable third party sources and avoiding conflicts of interest. --Cyllel (talk) 12:13, 1 October 2024 (UTC)[reply]
Well, I really wonder what Wikipedia is for if it is not to explain, as much as possible, how the subject matter can function. A theory is not "personal" but a tentative non-personal explanation of how something functions/functioned. If it's just about compiling names of authors, Google Search can already do the job. In the case of PIE, the material we have are the classical daughter languages and the reconstructed forms we can reasonably deduce from them : whether we like reconstruction or not, let's just remember Wikipedia's sister project Wiktionary using reconstructed material. Who's promoting "personal theories" ? I think it's rather Jasanoff. Have a look at his book "Hittite and the Indo-European Verb" :
- Page 34: He advocates a "PIE pluperfect" while as far as I know nearly no or only very few scholars reconstruct a pluperfect for PIE; this shows that he doesn't understand a lot about verbal aspect, since aspectual systems (as opposed to systems functioning with tenses, as we have in English) don't need pluperfects.
- §84: He brings out of his hat some "protomiddle" (likewise, is this a category commonly reconstructed by most scholars of PIE ?)... And much farther, §97, links it to the perfect, though the perfect - whether we analyse it as a tense or as an aspect in PIE - has nothing to do with a voice like active, middle or passive. One does not needs to have a degree in linguistics to understand that "I have hunted" does not mean the same as "I hunt myself" or "I have hunted myself".
- After inventing a lot of new verbal categories (should we call them "personal theories" ?), Jasanoff ends up with putting into question the well-attested *woide verb : (p.232) : *u̯óid- / *u̯id-ɛ ‘know’, by far the best known and most frequently cited perfect in the PIE verbal system, thus emerges not as an archaism but asan inner-IE neologism —a back-formation from its own middle, which in origin was not a perfect at all but a root stative-intransitive present.
So, Jasanoff is not even consistent with the very mainstream and well-established (because attested by similar irregular conjugations in several classical languages) reconstruction, even found on Wiktionary ! : https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/w%C3%B3yde
(and notice that Jasanoff's labelling of *woide as "intransitive" is nowhere to be found except in his imagination. So, is this not a "personal theory" ?)
What do I do on my side ? I take back, (with due credentials) Alwin Kloekhorst's PhD thesis, a fully cross-reviewed scholarly work, and how it analyses the Hittite -hi conjugation as a non-reduplicated PIE perfect (or stative), and then compare this analysis with a few tens of irregular reduplicated PIE perfects-statives with an active form and a passive meaning (nearly all from the Lexikon der Indogermanischen Verben, i.e. the leading handbook for PIE verbal reconstruction widely used in Wiktionary...), and I just put things right by explaining that the "irregular" forms ought simply to be viewed as formerly regular (a common process in languages, the irregular verbs of English were regular in the past...). If *woide is viewed as the remnant of an older PIE paradigm, this simply explains the regular -hi conjugation of Hittite as the last remnant of this PIE unreduplicated stative (stative being an aspect while present is a tense, so that they don't exclude each other, just like English has a progressive present conjugation [to be + -ing] that roughly corresponds to a PIE imperfective aspect vs. the simple model of the general present conjugation which corresponds to PIE perfectives and sometimes even statives, depending on the meaning of the verb).
So I find that labelling "personal theory" an analysis more consistent with the general model (found even on Wiktionary !) than Jasanoff's invented verbal categories and final tentative but unsuccessful destruction of one of the best-attested PIE verbal forms is not fair. Wikipedia should be a place of scientifical thinking, based on factual evidence (in our case the attested or well-reconstructed verbal forms) and therefore I shall restore the former version of the article. Mundialecter (talk) 15:20, 18 November 2024 (UTC)[reply]
By the way, I have just stumbled upon this article by Kloekhorst (2018) which clearly dismisses Jasanoff's construction : https://www.academia.edu/37703436/The_origin_of_the_Hittite_%E1%B8%ABi_conjugation_2018_?email_work_card=view-paper
So, was I wrong when I took back Kloekhorst's analysis and extended the analysis of the active to the medio-passive, where I quote tens of reduplicated remnants in Sanskrit and Greek ? Mundialecter (talk) 18:01, 22 November 2024 (UTC)[reply]
If you have constructive contributions to the article that you'd like to make, you should do so, with the caveat that they must be supported by reliable sources. Talk pages (and Wikipedia in general) are not forums to discuss the subject at large; rather, they are designed for discussion constrained to the article and its contents. AviationFreak💬 05:21, 23 November 2024 (UTC)[reply]