Khoton/Khotong

I’ve seen mention of a rather mysterious language spoken in western Mongolia in some of the literature: Khoton. As far as I can tell, it is attested in only one document: Владимiрцовъ, Б. and А. Самойловичъ. 1916. “Турецкій народецъ хотоны.” Записки Восточнаго Отдѣленія Императорскаго Русскаго Археологическаго Общества 23, 265-290.

I was finally able to find a copy of this article online (thanks in no part to the shoddy citations many authors have given), and have started to add forms to the database. To say the least: it’s weird. A lot of the literature has used Khoton as a justification for inserting vowels at the end of certain reconstructed forms, because some of the forms given in the aforementioned publication have unexpected short vowels at the end of certain words. There’s some justification for this, as certain Mongolian correspondences with Turkic have unexpected final vowels as well. However, I’m skeptical that these vowels in Khoton are actually significant, as it would be very strange that only one poorly known variety of Turkic would exhibit these vowels, while others have lost them. These same short vowels appear in the forms for “6” and “7”, yet were not lost in any other Turkic language (even though the resulting forms would still be pronounceable).

As I work through this pre-reform Russian document, I hope to be able to figure out where Khoton fits in the overall family tree. It may be related to Chanto Uyghur, another poorly attested language of Mongolia that is currently spoken in the same aimag. I’ve suspected for some time that Karluk might be a wastebin taxon given that many of the non-standard varieties of Uzbek and Uyghur meet none of the criteria for being in that family, and the Khoton data might be a further nail in that coffin.

Map Updates

I’ve been working on a new version of my map of Turkic languages. I released the first one in 2021, and it’s due for an update.

To that end, I have been gathering data to create the best map possible, which means pulling Census data to create new maps. The latest is a county-level ethnicity map of China, which will vastly improve mapping for that country. Check it out here. There were a few issues mapping the census figures to the counties, but I’m working to remedy that. It appears that some county lines have shifted since the data of the census data I’m using, and that makes some of the counties appear in black. Regardless, I’m pleased with the outcome.

Site issues fixed!

At some point the version of SQL my database was using was updated, and that deprecated the syntax I was using for regular expressions. This messed up everything. I was able to find a relatively easy fix. This was all after I considered completely redoing the website. I still may do so, but for now I’ll focus on minor fixes.

I’ve gotten a complaint that I haven’t been adding much new data. Please keep in mind that I don’t get paid for this. Also, people have been sending me dubious posts on random websites as data sources. Not useful.

I’ve been otherwise very busy with life in general and new endeavors, so please be patient. One potential project is a re-do of the map of the Turkic world I first produced in 2021. There are issues (as there are with all linguistics maps), and I’d like to fix those.

Site Issues

My host recently moved servers, and, as a result, some of the code within the database has become wonky. The data is still all present in the background, but it’s currently inaccessible to the public. I hope to figure out the issue and correct it soon.

Shirts and such

I’ve had a few ideas for t-shirts, so I’ve created a Redbubble page. There’s more to come, but for now, check out my profile here:

ElegantLexicon.redbubble.com

So far I’ve got one for Adelaide Hasse, one from the Irk Bitig, and a random cracker design that I had sitting around.

More to come soon!

New Map, New Language?

In addition to working on point-based maps, I tried my hand at an SVG based map that can be altered by selecting various criteria. The result in this map of Kazakhstan. Like many of the other maps I’ve created, this map shows ethnic groups rather than languages spoken. It’s the best we can do as a proxy for language. You can display a number of different ethnic groups and choose to display either percentage of total speakers in a district or total number of speakers relative to the district with the highest number of speakers. The shading of the map will change, as will the key if you opt for number of speakers. It’s been fun and educational to create and I may try my hand with other countries. China has been particularly difficult to sort out, so that may be next. Incidentally, if anyone has data for Mongolia, I would love to see it.

I’ve been aware of Romanian Tatar for some time, but haven’t seen much information on it. It’s basically a variety of Crimean Tatar. Some sources call it Kipchak, but I’ll withhold judgement on that classification until I can inspect the data. Omniglot has a great bunch of resources on it. I’ll likely add it to the database once I’ve got some better data.

Maps & Stuff

I’ve been working on doing a massive overhaul of the maps page. There were some issues with Leaflet.js, the map creation app that I’ve been using. Also, I’ve wanted to create a more granular, more accurate set of maps than I’d previously had.

What’s up now is a map showing census data from Russia, Ukraine, and Romania. The data gathering is slightly different in each country, so it shouldn’t be seen as comparing apples to apples. Russia is complete, Romania is complete, and Ukraine is nearly complete. The biggest surprise so far is the vast range covered by Gagauz speakers in Ukraine. Crimean Tatar is pretty much only in Crimea, while Karaim is very scattered. These are the only three Turkic languages covered in the 2001 Ukrainian census (at the village level), so that’s all I can show.

Once Ukraine is done, I have a lot of work to do:

  • First, I need to edit the code so users can select only a single country or certain languages.
  • Next, I plan to continue to search national census/statistics websites to see if this kind of granular information is available elsewhere.
  • I’d like to include other types of maps where this data is not available. The Georgian census, for example, had province-level data about Azerbaijani speakers that I’d like to include is some other format.
  • Finally, I’d like to restore some of my former maps: Baraba Tatar dialects, Khalaj and Urum villages, etc. Each map needs to have properly cited sources to ensure that anyone looking at this site isn’t comparing very different types of information.

General updates

Some quick updates:

I’ve added some Salar and Ili Salar forms, bringing the total number of entries up to 25,400. However, I’ve been really busy lately and haven’t had much time to put too much work into this website.

One of the things I’ve been working on is Wikipedia. I used to edit it quite a bit, but stopped until this past spring. I worked on a Wikipedia Edit-a-thon in May, and since then I’ve been really active. You can see my profile here.

I’ve also been busy writing a paper in library science, which will be published sometime in the next few months in the journal Library Resources & Technical Services. Exciting stuff!

Other things that have kept me busy are proofreading titles published by Language Science Press and, as of yesterday, I’ve been asked to review a manuscript for Archiv Orientální.

I really need to update my CV. You can find that on the main website.

24000 entries!

While going through all of my sources again, I realized that I hadn’t entered much for Dolgan. This has been pretty quick going, thanks to Stachowski’s Dolganischer Wortschatz. Today I reached 24000 entries. Entry 24000 is the Dolgan word for “glass” – hǟrkälä. This word is pretty interesting – it’s ultimately a borrowing from Russian зеркало, which means mirror. It means mirror, too, in Dolgan, but also means glass. I haven’t been able to find any other forms with that meaning for Dolgan. Russian зеркало was borrowed into Sakha as сиэркилэ, where it also means mirror. I’m not convinced the Dolgan form is descended from the Sakha form, as the vowels are pretty different, but there is likely some relationship. It is likely that early Russian traders traded manufactured goods like mirrors with the locals, who borrowed the word. Apply some vowel harmony and Dolgan’s strong dislike for /s/-sounds, and you get hǟrkälä. As glass was the only unknown component of these traded mirrors, the terms became conflated.

As a side note, I chose the term “glass” because I wanted to see if the early Turks had access to this technology. Most Turkic languages either use terms for manufacture products (such as bottles or mirrors) to mean “glass”, or borrow from other languages. This indicates that glass was unknown to them in ancient times. Also, glassmaking was developed only about 4000 years ago in Mesopotamia and only in the 5th Century CE in China. So any glass objects that the oldest Turkic civilizations would have had would have to come from the Middle East or Europe, and would not have been made locally. This may tell us something about their metallurgical practices, as it is believed that glass was discovered as a byproduct of metallurgy, when hot metal came into contact with sand.

Chalk

Naturally, after adding 50 new glosses to the database I’ve run across a new one that I’d like to add: chalk. Chuvash has пурӑ, пур, Kazakh and Kymyk have бор

Wiktionary suggests that the Kazakh for comes from Russian бор “boron”, but this is clearly conflating the boron meaning with the chalk meaning. Fedotov says this is a native Turkic term and ties it to Sakha буор “earth, clay”. (Tuvan has пор and Tòfa has бор for clay as well). In Bashkir, the form is either бур or аҡбур, suggesting that the original term may have referred to crumbly stone or soil, with color terms used to distinguish between chalk, clay, etc.

I’ve entered forms for the latest 50 glosses for Turkish, Tuvan, Dzhungar Tuvan, Sakha, and Chuvash, and I’m working on Azerbaijani. Once I’ve made my rounds, I may add chalk to the database, plus whatever else I find.

As you can see above, there are a lot of cases where it could be useful to suggest related terms. Knowing that chalk is related to earth and clay could be beneficial. I may work on this in the near future as well.