Appendix talk:Marathi Swadesh list

@Kutchkutch: हे भरून टाका!! 😃 -- mādhavpaṇḍit (talk) 04:29, 8 October 2017 (UTC)

Of course, also do it leisurely, no need to fill out the 207 words all at once. -- mādhavpaṇḍit (talk) 04:33, 8 October 2017 (UTC)

@माधवपंडित: Thanks for creating it! I've been looking at Category:Swadesh_lists_by_language for a while now and noticed Marathi didn't have one until now. It appears to be something every language should have, but since it is in the Appendix namespace, it appears to be harder to find compared to articles mainspace. I also found Appendix:Indo-Aryan_Swadesh_lists and Wiktionary:Requested_entries_(Hindi)/Multilingual_list_of_Indian_family_relation_names. Perhaps the second one is perhaps not an official appendix since it is a subpage. Kutchkutch (talk) 08:05, 8 October 2017 (UTC)

@Kutchkutch: True, as it is, the coverage of Indian languages on wiktionary was poor and it is even poorer for Marathi & Konkani. Thankfully more of us are showing up and I really wish to document Marathi and Konkani so well that wiktionary should be the most reliable resource for these languages. -- mādhavpaṇḍit (talk) 09:40, 8 October 2017 (UTC)

@माधवपंडित: Since Wikipedia is one of the top ten websites in the world, its coverage of India-related information appears to be better than Wiktionary's coverage of Indian languages. Wiktionary has a global rank closer to 500. I noticed that Indian-language pages on Wiktionary do not get very many page views compared to other languages. User:Aryamanarora predicts that Wiktionary's coverage of Indian languages will improve as Wikipedia users come to Wiktionary. Hopefully Aryamanarora's recent promotion to administrator and continued 'guidance' means that Wiktionary's coverage of Indian languages will grow and become 'the most reliable resource for these languages' as you mentioned. Wikis are usually criticised for having incorrect information because they are not made by users with proven expertise, but with this in mind they are still one of the the best ways to acquire general information.

When I first found about Swadesh lists a while ago I thought the word 'Swadesh' was the Sanskrit word स्वदेश (svadeś) and was surprised that the word 'Swadesh' in 'Swadesh list' is a surname.

For this Swadesh list I was wondering if declined forms should be included. For example, the feminine form of कुत्र (kutra), कुत्री (kutrī), has an entry. However, I noticed that on Appendix:Hindi_Swadesh_list कुत्ता (kuttā)'s feminine form कुत्ती (kuttī) is not mentioned so it seems declined forms are not included. Kutchkutch (talk) 02:16, 9 October 2017 (UTC)

@Kutchkutch: Haha! Even I thought that Swadesh referred to the Sanskrit word and wondered why they had "Swadesh" lists of European languages as well. The deal here is that even though Marathi certainly has an extensive literature, its documentation on the internet is restricted to google translate, globse & shabdkosh. The first two are outrageously inaccurate. And not many people have exposure to the DSAL project dictionaries we refer to. Old Marathi has absolutely no presence outside here and Tulpule's dictionary. So wiktionary has a decent shot at being the #1 go-to place for Old Marathi & Marathi. Aryamanarora is right, more Indian language editors have been showing up this year, we have Konkani, Marathi, Kannada, Gujarati, Bengali and Assamese taken care of by now. If i'm not mistaken Aryaman is the first Indian-language editor to have been made Admin. Hopefully this gets even more contributors for us. -- mādhavpaṇḍit (talk) 02:45, 9 October 2017 (UTC)

@Bhagadatta, Kutchkutch Could you check this table whenever you can? I noticed that someone who may not be a native Marathi speaker filled it out completely a bit ago, so not sure about the quality of translations; seems to be an excessive number of tatsamas. I modularized it just now, which is why I noticed. Its data is at Module:Swadesh/data/mr. (Also ha, two of us changed usernames since we last spoke here.) —Aryaman^{A (मुझसे बात करें • योगदान)} 03:29, 12 August 2020 (UTC)

Yes, @Kutchkutch: has now updated the list and replaced tatsamas with actual translations. -- Bhagadatta (talk) 04:50, 13 August 2020 (UTC)

@AryamanA: Yes, as Bhagadatta said, while updating the list I've only included tatsamas and Persian borrowings that I feel are necessary. This would be the cases in which there are no suitable ʻnativeʼ words or when the tatsamas and Persian borrowings are extremely common.

Are you also planning to modularise the MIA list?

@Bhagadatta: I was wondering if you have any suggestions for improvement. Also, when you're able to continue with the Konkani list, it would be very interesting to compare the two. Kutchkutch (talk) 10:06, 13 August 2020 (UTC)

@Kutchkutch: The Marathi list looks great. To be honest, owing to my long absences from wiktionary in the last 2 years, I had forgotten all about Swadesh lists and had forgotten that a Konkani Swadesh list even existed. I plan to resume filling out the Konkani very soon. -- Bhagadatta (talk) 10:23, 13 August 2020 (UTC)

Also, Konkani encounters a problem not familiar to Marathi or Hindi: while Marathi has an extensive literature, Konkani is still some kind of a "colloquial" language, with official government notices, sign-boards and schools in even Goa favouring Marathi as the medium of communication. That isn't to say Konkani doesn't have literature. It does, except a disproportionately large percentage of it is in the Romi (Catholic) dialect, the speakers of which in real life account for maybe 40% of all Konkani speakers. This literature uses the Latin script and thus it can get quite challenging sometimes to ascertain the true Devanagari spelling of a Konkani word even if it is quite common. -- Bhagadatta (talk) 10:37, 13 August 2020 (UTC)

@Bhagadatta: Thanks for your feedback. I had also forgotten about this list. Despite the criticisms of lexicostatistics, these Swadesh lists are still effective at showing a few of the basic words of each language. For increasing the coverage of basic Marathi words, Aryaman suggested a long time ago that the uncreated entries here and User:AryamanA/Marathi be a good start.

I've tried looking at Konkani resources to try to make sense of the language such as the Konkani and English Wikipedias, Konkani Vishwakosh, A Higher Konkani Grammar by Janardhan and videos by Prudent Media. However, the problem that you've described makes it challenging. Perhaps a similar situation exists for the other Southern Indo-Aryan languages such as Khandeshi/Ahirani, Varhadi, Thanjavur Marathi and Judeo-Marathi. Kutchkutch (talk) 11:09, 13 August 2020 (UTC)

@Kutchkutch, Bhagadatta: I have modularised the MIA list as well, thanks for the reminder. I actually remembered the Swadesh list stuff after reading about the IE-CoR project, which is compiling a database of IE wordlists for cognacy judgement. So I thought I would try to clean up the South Asian lists and see if we can use those for something like that.

I'll probably be around a bit more from now, since I'm starting my undergrad in linguistics and computer science so can justify to myself spending some time here :) —Aryaman^{A (मुझसे बात करें • योगदान)} 21:53, 13 August 2020 (UTC)

@AryamanA: Good luck! -- Bhagadatta (talk) 01:06, 14 August 2020 (UTC)

@AryamanA: Thanks for the reply and good luck with your studies.

There was an empty Elu Swadesh list before the modularisation the page, but it was removed during the modularisation process. This is understandable since there was nothing there to begin with and it seems that this is the MIA language that we understand the least.

Based on the entries for Old Hindi, Old Gujarati, Old Marathi and Early Assamese, it seems that Early NIA Swadesh lists may be feasible. However, until there is more clarity about the Aprabramshas, Aprabramsha Swadesh lists don't seem feasible yet.

Some of the Swadesh lists have the |nativename= parameter and/or alternative scripts. It would be nice if Module:Swadesh supports these features eventually. Perhaps nativename to show the name of the language in the Swadesh list of that language (i.e. हिंदी, ગુજરાતી, ਪੰਜਾਬੀ / پنجابی, etc) is the easier of the two to implement.

Also, it would look better if Reconstruction:Ashokan Prakrit/𑀓𑀼𑀢𑁆𑀢 could be shortened to just *𑀓𑀼𑀢𑁆𑀢 when adding the translation for ‘dog’ in Ashokan Prakrit. Kutchkutch (talk) 07:23, 14 August 2020 (UTC)

@Kutchkutch, Bhagadatta: Thank you! (1) I added support for a 'nativename' parameter on the module data pages; (2) multiple scripts and dialects now work, but require two separate data pages, see AP:Punjabi Swadesh list; (3) fixed the problems with reconstructed links in transliterated form.

I agree that maybe early NIA lists are possible. Not sure about the Apabhramshas yet. —Aryaman^{A (मुझसे बात करें • योगदान)} 21:55, 15 August 2020 (UTC)