User talk:Pulimaiyi

Archives
2017 2018 2019 2020 2021 2022 2023

Out of process deletions of rfv-quote at अमृत

The discussion of the validity of the quotation at Sanskrit अमृत (amṛta) is still in progress, so kindly do not delete the notice until the discussion is finished. --RichardW57m (talk) 14:29, 30 May 2024 (UTC)

@RichardW57m: The RFV is not unresolved. It's just you refusing to concede even after being shown evidence contrary to your conviction. -- 𝘗𝘶𝘭𝘪𝘮𝘢𝘪𝘺𝘪^{(𝘵𝘢𝘭𝘬)} 14:42, 30 May 2024 (UTC)

Derogatory comments regarding other editors

Greetings,

I am writing to formally warn you of violations of the Wikimedia Foundation's Universal Code of Conduct (UCoC), as well as Wiktionary policy on wiktionary:Civility.

You wrote a comment speculating on another editor's mental health. It is extremely arrogant, judgemental and rude to speculate on the mental condition of an editor you don't know. Please refrain from commenting on editors and stick to the content. Thank you.

I have redacted your comment appropriately, and I hope and pray that I will not have occasion to do it again. Elizium23 (talk) 19:52, 8 July 2024 (UTC)

Old Awadhi

Would it be appropriate to upgrade Old Awadhi awa-old from an etymology-only language to a full-fledged chronolect as inc-oaw? It might be more accurate to make a distinction between

• modern Awadhi terms such as बल्लेबाज (ballebāj) versus
• older terms found in the Ramcharitmanas, Hanuman Chalisa, Padmavat, etc.
• Also, CAT:Awadhi semi-learned borrowings from Sanskrit would imply that a term was first borrowed into Old Awadhi rather than modern Awadhi.

This would be similar to how terms found in Kabir’s work are currently considered as Old Hindi and terms found in the Guru Granth Sahib are currently considered as Old Punjabi.

(see [[User_talk:Pulimaiyi/2022#Old_{{desc|LANG|TERM}}_in_descendants_trees]])

The case for distinguishing Old Awadhi is perhaps stronger than Old Marwari or Old Maithili because

• Ramcharitmanas, Hanuman Chalisa, Padmavat, etc. are all fairly well-known texts that have been analysed several times.

(see MOD:Quotations/awa/data)

• the Eastern Hindi languages do not have an Early New Indo-Aryan common ancestor.
• There already exists Old Gujarati for Gujaratic-Rajasthani languages and Old Bengali for Bengali-Assamese languages, and their relationship to Old Marwari or Old Maithili would need to be clarified.

The etymology-only code for Old Braj bra-old is a descendant of Old Hindi rather being contemporaneous with it. So, with the current arrangement it is possible for an Old Braj term from the सूर-सागर to be a descendant of an Old Hindi term from Kabir’s work.

Old Awadhi was tangentially discussed with Svartava with respect to Chhattisgarhi being an ancestor of it in two edit summaries at the revision history of भाखा

Kutchkutch: Is it appropriate to … make Chhattisgarhi a descendant of Old Awadhi?

Svartava: No idea about Chhattisgarhi being descendant of Old Awadhi

Making Chhattisgarhi a descendant of Old Awadhi is not possible without upgrading Old Awadhi from an etymology-only language to a full-fledged chronolect.

Apparently, Svartava has a WT:AutoWikiBrowser bot that can replace instances of one code with another. (also AryamanA)) Kutchkutch (talk) 09:35, 11 August 2024 (UTC)

@Kutchkutch: I gladly support this move. I want to ask, however, is Awadhi sufficiently different from Old Awadhi natively (ie, without the intervention of any foreign loanwords)? As such, would the Hanuman Chalisa in modern Awadhi be any different? -- 𝘗𝘶𝘭𝘪𝘮𝘢𝘪𝘺𝘪^{(𝘵𝘢𝘭𝘬)} 05:38, 13 August 2024 (UTC)

Thanks for the feedback. Since many of lemmas in CAT:Awadhi lemmas other than बल्लेबाज could possibly refer to both Old Awadhi and modern Awadhi, it is completely understandable to ask how different they are. Although a distinction between the two has been made in descendants trees, since they are merged as one language, I have avoided creating any Awadhi entries.

• Old Awadhi terms are mentioned in the dictionary {{R:inc-ohi:Callewaert:2009}}. The term “Old Awadhi” has been used since Grierson in 1904 ({{R:inc:Masica:1993}} page 53). There are Perso-Arabic borrowings in Old Awadhi as well as modern Awadhi.
• Modern Awadhi terms are mentioned in dictionaries such as {{R:awa:Dikshit}} and the {{R:hi:Hindwi}} database. Although, a modern Awadhi dictionary could possibly refer to Old Awadhi texts, this would be similar to how a modern Marathi dictionary would refer to the Dnyaneshwari.

• Regarding how the grammar (rather than the lexicon) of Old Awadhi differs from modern Awadhi, {{R:awa:Saksena}} is an entire book on that topic. Although it is from 1937, this work would perhaps be more helpful in imagining what the Hanuman Chalisa would be like in modern Awadhi. There are several comparisons in {{R:awa:Saksena}} between Old Awadhi and modern Awadhi that clearly show how the language has changed.

• Since you said that you support keeping Old Awadhi separate from modern Awadhi (and just wanted some further clarification regarding how different they are), it seems that the implementation towards it can proceed.

@Svartava After the language code for inc-oaw is created (and if you have nothing further to say about the matter), could you run your bot to replace awa-old with inc-oaw?

Kutchkutch (talk) 12:50, 13 August 2024 (UTC)

@Svartava Thanks for helping out with the implementation by using the bot. The implementation seems to be done now. Kutchkutch (talk) 15:40, 13 August 2024 (UTC)

@Kutchkutch Your work is appreciated as well, we managed to finish this quite quickly with the teamwork. Congrats on completing 19000 edits, also. Svartava (talk) 03:29, 14 August 2024 (UTC)

@Svartava: You seem to have missed the Reconstruction namespace- I just fixed a module error at Reconstruction:Sanskrit/तिथिवार. It seems like this has happened before, since I vaguely remember on at least one other recent occasion finding Indic-related module errors only in that namespace. Chuck Entz (talk) 14:47, 14 August 2024 (UTC)

@Chuck Entz: I just want to clarify that I did check CAT:E more than once but it wasn't updated (technical delay) when I checked (I also put it in my watchlist for some time around these events, so that any new addition in it to be shown, but again that didn't happen because as far as I have seen such changes only appear when a recent edit on that page causes some error rather than when the functioning of some pre-used template/module is impacted). It happens because search results by default don't show namespaces other than mainspace (if possible, we might want to change this because reconstructions are content pages / dictionary entries, but whatever) and the last such instance recently was related to the merging Apabhramsa lects. Svartava (talk) 15:14, 14 August 2024 (UTC)

@Svartava: I wasn't expecting you to spend all your time patrolling CAT:E- that can take days for the errors to pop up. It's just that you obviously were very careful to prevent the errors in mainspace by changing the language codes via AWB, but you didn't do the same for the Reconstruction namespace. I had assumed it was merely that you did't think of it. If it's a matter of not knowing how to search the Reconstruction namespace, that can be fixed. It's actually not hard at all to add namespaces at Special:Search, but I'll make it easier for you: this is a search for "inc-oaw" in the namespaces that are most likely to be a problem. Just replace the "inc-oaw" with the language code you're interested in. Thanks! Chuck Entz (talk) 04:07, 15 August 2024 (UTC)

Pali āhacca

What is your objection to showing the structure of āhacca as being composed of the Pali root har (soon to appear) plus the suffix cca?. (The suffix is listed as 'tya' in Duroiselle's Practical Grammar of the Pali Language.) --RichardW57 (talk) 15:17, 2 January 2025 (UTC)

@RichardW57: The analysis har +‎ -cca is considered valid if 1) -cca is a productive suffix occuring in cases where the word is fully synchronic and non-inherited; and 2) har + -cca can convincingly give hacca in any other synchronic non-inherited formations. If you have such examples, please bring them. Svārtava (tɕ) 17:45, 2 January 2025 (UTC)

@RichardW57: Exactly what @Svartava: said. Deriving hacca from हृत्य (hṛtya)/हत्य (hatya) makes a lot more sense than claiming it was synchronically coined from har + "-cca". -- 𝘗𝘶𝘭𝘪𝘮𝘢𝘪𝘺𝘪^{(𝘵𝘢𝘭𝘬)} 02:30, 3 January 2025 (UTC)

@Svartava The solution to that is to use |se=1 with {{pi-root}}, which is much easier for uses of suffixes that were falling out of use but were interpretable. I've put that in the page, after the assertion of inheritance from presumably Vedic Sanskrit, so the reader sees the connection with Sanskrit and then can see how one goes about interpreting it as other than a suppletive memory item. --RichardW57 (talk) 09:04, 3 January 2025 (UTC)

@RichardW57 As above, the question is about the validity of the analysis har + -cca giving hacca, which needs to be verified by other examples or innovations within Pali, otherwise there is no convincing reason to believe that har + -cca will become hacca and not anything else. (Pali sources mentioning har + -cca = hacca don't count as they just mean to provide an explanation for the term rather than accurate historic and phonological analysis and Wiktionary is not forced to follow other sources when they are wrong, similar to how Sanskrit पीत (pīta) is not analysed as पा (pā) + -त (-ta).) Svārtava (tɕ) 11:38, 3 January 2025 (UTC)

@Svartava: But surface analysis is meant to show how the term is/was understood, not the history of the term! Moreover, in Pali terms -hacca is har + t + ya, but the presence of -t- is now lexically determined. One could also offer a more abstract analysis as har + ya, but that has several possible outcomes. Note that hacca seems to be a different formation. On a phonological level, there is ablaut going on, which the template doesn't handle, and the ablaut causes the final consonant to drop (and sometimes the vowel to change), which is why we get -acca- and not -añca- from the root han. --RichardW57 (talk) 17:00, 3 January 2025 (UTC)

I'm not sure that rejecting that analysis of पीत (pīta) is correct; one problem is that there is more behind the root than the citation form. --RichardW57 (talk) 17:00, 3 January 2025 (UTC)

@RichardW57: how the term is/was understood is generally not the correct approach if we intend to provide historically and phonologically accurate etymologies unlike Pali-only sources (like PTS, etc.) may be giving. Adding surface analysis of har + -cca for hacca is equivalent to writing kar + -ta for kaṭa and kata < कृत (kṛta), which is agreeably a very bad idea and also finishes the importance of surface analysis: here also one can argue that kar + -ta is how the term kata is/was understood - and that it is, because of it's meaning as the past participle. Svārtava (tɕ) 20:05, 3 January 2025 (UTC)

But Template:surf is not intended to provide historically accurate etymologies. Now {{pi-root|har|cca|sa=हृ|se=1}} displays By surface analysis, har (root) +‎ -cca and {{surf|pi|har|-cca|pos1=root}} also displays as By surface analysis, har (root) +‎ -cca. The former also categorises. As the latter is valid, I see no reason not to succinctly use a single call to {{pi-root}} rather than invoking both. Or do you repent of the parameter |se=?

@Svartava: For kar + ta, it's a question of how we choose to handle allomorphs and sporadic sound changes. In this formation, retroflexion is a sporadic sound change, and the reduction of the root to zero grade is a suffix-dependent rule, with a rather vague rule for the resultant vowel. The form -cca warns the hearer (or at least, the reader) that a final resonant may have been elided, just as final -tta in a past participle announces an unspecified voiceless stop in the root. --RichardW57 (talk) 11:49, 4 January 2025 (UTC)

@RichardW57: The template or display text is not the issue, the issue is the validity of the surface analysis. Retroflexion is not a sporadic sound change, it is due to the ṛ in कृत (kṛta) (similar to how r in प्रति (prati) gives paṭi-) and a result of non-surface-analysable inheritance.

The well defined system of roots in various grades works well in most cases in Sanskrit and even if some aspects of it are inherited into Pali, it may not work in all cases. For example, कृ (kṛ) explains कृत (kṛta) (zero grade using suffix -त (-ta)), कर्तव्य (kartavya) (first grade using suffix -तव्य (-tavya)) but I don't think it can be said that Pali root kar can explain either kata/kaṭa or kattabba, due to language degradation over time, though it continues to explain karaṇa where the surface analysis can be legitimate. Where the surface analysis is not able to adequately explain the word, it should not be given. You can still argue saying that zero grade of kar is ka and -tabba is a suffix that "sporadically" geminates the first t when the original root drops it consonants or so on define rules to fit all known formations, but this is just unuseful fictious stuff. Svārtava (tɕ) 13:42, 4 January 2025 (UTC)

The retroflexion of dentals to retroflex under the influence of /r/ or /r̩/ in Pali, where not shared with Sanskrit, is 'sporadic', even if common. There seems to be no phonetic rule for it to definitely happen, though there may be some morphological conditions. You may recall that, contrary to the Young Grammarians' belief, a sound change tends to spread through a language word by word. If it stops spreading before affecting most of the vocabulary, it is 'sporadic' by definition.

Much of the Pali verb systems seems to have undergone rapid changes, and a very common verb like karoti seems to have preserved a lot of the stages. The grammarians list many forms not attested in the texts. I'm strongly attracted to the view that Pali is demagadhised Magadhi, which process invites a lot of variants. In the case of kattabba v. kātabba, we're looking at rather an abandonment of zero grading in the first case and I think a change of root in the second. Sanskrit करोति (karoti) is already a bit of a mess. As you point out, the root system was breaking down in Pali, and the present stem increasingly replaces the root as the basis of verb forms outside the present tense system, though for the later material we are no longer looking at native speakers' productions in any sense.

@Svartava: I think {{T:pi-root}} could be carefully enhanced with a note of the 'grade' where different from that of the root, for which we should find a good account of the Pali root rather than invent our own. It doesn't help that only Pali phonemes are used in the Western names of Pali roots, which is why some grammars and dictionaries give up and resort to the equivalent Sanskrit roots. (Pali names are further constrained to follow Pali phonology.) --RichardW57 (talk) 23:17, 4 January 2025 (UTC)

@RichardW57:

I am unsure of how to make this clearer to you without practically echoing what @Svartava: has said, but let me try.

Surface analysis may be acceptable, to a degree. For instance, for Pali haraṇa, I would write the etymology as the term being inherited from Sanskrit (Old Indo-Aryan), but then would throw in the surface analysis of har + -ana.

Next, in a term like gata, although by rights, an inheritance from OIA is the only legitimate explanation, I could *still* probably be convinced to include a synchronic analysis as gam- + -ta, as roots with nasals losing them in the zero grade is well understood in Sanskrit, and therefore, I assume, in Pali.

But here, ha is not the zero-grade of har. It exists in that shape because its earlier form was हृत्य. Adding that surface analysis in such situations is misleading. -- 𝘗𝘶𝘭𝘪𝘮𝘢𝘪𝘺𝘪^{(𝘵𝘢𝘭𝘬)} 07:21, 5 January 2025 (UTC)

@Pulimaiyi: Why not? The zero grade in Sanskrit is hṛ, and the regular Pali reflex of that is ha. Now, the correspondence of Sanskrit ṛ to Middle Indic may be disturbed by neighbouring sounds, but this is not such a case. Looking at the table of principal parts in Warder, which has pruned rare alternatives, for roots in -ar, I see this as the general pattern for past participles formed by adjoining -ta to the root. There are some exceptions with labial initials, where the reflex of ṛ is u, and a probably trumping complex process whereby initial 'va(r)' instead yields 'vu', e.g. vuta from var and vuttha from vas. --RichardW57 (talk) 14:02, 5 January 2025 (UTC)

I see this as the general pattern for past participles formed by adjoining -ta to the root.

There is of course a pattern because of inheritance from neatly patterned formations in Sanskrit. That doesn't merit surface analysis by itself.

Finally, trying to surface-analyse Pali terms when they are clearly not analysable like that is very unfruitful. Sanskrit and Pali are very close once a reader understands the patterns of sound changes (that are generally consistent) that take place. I would say it is nearly impossible to be able to study Pali grammar independently in an effective way without studying/knowing Sanskrit grammar system, at least in the aspect of roots and their derivatives.

Claiming that ha can be said as the zero-grade form of har is a fringe statement. There is observable pattern, and that is because of the Sanskrit connection between zero grade and first grade, not because Pali had a well-developed grammatical understanding of this, as you yourself pointed out, the past participle of var is vuta, a perfectly understandable term when you put it beside Sanskrit वृत (vṛta), but analysed as "exception" within Pali. Svārtava (tɕ) 15:54, 5 January 2025 (UTC)