20.2.22.21:42: WHITE RAT 1.29

? qulugh ai nai sair juri ish nyair

'white rat year, head month, twenty nine day'

1. My hypothesis of ergativity in pre-Mongolic got me thinking about ergativity in Mandarin, an idea that's existed longer than I had thought. Last night I found Frei (1957) who warns of the danger of viewing Chinese through an Indo-European prism on p. 45. His point remains valid over sixty years later for any non-Indo-Euroepan language, though one must also be wary of the opposite peril: the temptation to make the ordinary exotic.

2. Poupard (2019) on Tibetan literature in Mandarin makes me wonder what might be learned from a comparative study of colonized authors writing in colonial languages. Out of ignorance, I might have stumbled upon a cliché topic in comparative literature.

3. Dwight Decker, one of my favorite authors, coined the term déjà lu for the feeling of having read something before.

4. Today I learned from Wikipedia that

The Mughal designation for their own dynasty was Gurkani (Persian: گورکانیان‎, Gūrkāniyān, meaning "sons-in-law").

The word "Gurkani" derives from "Gurkan", a Persianized form of the Mongolian word "Kuragan" meaning "son-in-law".

The actual Written Mongolian word is ᠬᠦᠷᠭᠡᠨ kürgen. What I don't understand is why the Persian version has g and k instead of k and g.

5. I just heard Chávez pronounced as [ʹʃavɛz] with a French-style reading of Ch on 48 Hours on CBS.

20.2.21.23:51: WHITE RAT 1.28

? qulugh ai nai sair juri nyêm nyair

'white rat year, head month, twenty eight day'

1. Tonight I had 淺蜊蕎麥 asari soba for dinner.

There are at least three kanji spellings of asari 'Venerupis philippinarum':

• 淺蜊 (postwar 浅蜊) <SHALLOW BUG.ri>

• 淺 asa 'shallow' is a phonogram

• 蜊 ri, the second half of 蛤蜊 kōri, a rare Sino-Japanese word for 'clam', is a semantically relevant phonogram (there needs to be a shorthand term for that)

• the surname Asari, possibly from the clam name, can be spelled 淺利 without the 虫 <BUG> radical
  

• 蛤仔 <CLAM CHILD> looks like it should be read kōshi. - 仔 is a Chinese noun suffix. 蛤仔 (Mandarin gézāi) is the Chinese equivalent of asari and its spelling has been repurposed for the unrelated word asari
  

• 鯏 <FISH.ri> is a made-in-Japan semantophonetic kanji with the right side representing the final syllable of asari
  

asari has been derived from the verb asar- 'to look for', but I find it hard to believe that asar-i 'looking for' could become the name of a specific kind of clam.

I've written about the spelling of 蕎麥 soba here.

The complications of the Japanese script make even the name of a prosaic food a fascinating object of interest. I suppose native Japanese wouldn't even give 浅利蕎麦 (the spelling in the menu) a second thought.

2. Written Mongolian -(y)i is normally an definite accusative suffix. But Grønbech and Krueger (1976: 43) give examples where it is following a subject rather than an object:

cima-yi kür-üged saca tedeger bügüde bos-cu ire-müi.

2SG-yi come-COORD.GER as.soon.as they all rise-CVB come-DUR

'As soon as you have come, they will all rise and come [at you]'

nama-yi ire-küi cagh-tur

1SG-yi come-INF time-LOC

'at the time of my coming'

They explain that

[a]t the beginning of a sentence or clause, an accusative may be used to indicate that the word is not subject to the final verb but to the closest verb.

Today I wondered if such uses of -(y)i were remnants of its earlier function as an absolutive in an earlier ergative system. I would expect an absolutive to mark subjects of intransitive verbs (like kür- 'come' and ire- 'come') and objects of transitive verbs: e.g.,

odqan köbegün ber bars-i üje-bei.

youngest son NOM tiger-DEF.ACC see-PST.

'The youngest son saw the tiger.'

Here's how pre-Mongolic might have worked:

3. Today the Honolulu Star-Advertiser reprinted a New York Times story about "the secretive Shincheonji church". I instantly knew shin had to be 新 'new' and cheon had to be 天 'heaven', but it only gradually dawned on me that ji had to be 地 'earth'.

The full name of the church is

新天地예수敎 證據帳幕聖殿

Shinchhŏnjiyesugyo chŭnggŏjangmaksŏngjŏn

which breaks up into

• 新天地 Shinchhŏnji (= the New York Times' "Shincheonji") 'new heaven [and] earth' = 'new world'
  

• 예수 yesu 'Jesus'

• -敎 -gyo 'teaching' (a suffix for names of religions)

• 證據 chŭnggŏ 'testimony'
  

• 帳幕 jangmak 'tent' = 'tabernacle'
  

• 聖殿 sŏng-jŏn 'holy palace' = 'temple'
  

4. Another Honolulu Star-Advertiser reprint of a New York Times story introduced me to the word zoonotic. The word reminded me of hypnotic. Both are adjectives derived from -osis nouns, but the two have different components:

• zoo-nosis < νόσος 'disease'

• hypn-osis < -ωσις 'state, abnormal condition, action'

• by coincidence English has a 'long' o corresponding to Greek long ω

I don't understand why zoonosis doesn't end in -nosos or even a Latinized -nosus. Is -nosis by analogy with borrowed Greek -sis nouns like hypnosis and gnosis?

5. Today's Honolulu Star-Advertiser has a column on heka or hekka, a southern Japanese dialect word for 'sukiyaki' that survives in Hawaii. Here's a blog on "Heka: The Mystery Dish". I wonder what its etymology is. The blogger, a native Japanese speaker, at first thought hek(k)a was Chinese, Thai, or Malaysian - before learning from her grandmother that it is a Hiroshima dish.

The column goes on to mention the Big Island terms tomato beef and ice shave for what I call beef tomato and shave ice. I've never heard of tomato beef before. I just realized that the Hawaiian-like modified-modifier order of ice shave might reflect a Hawaiian substratum.

6. Tonight's Hawaii Five-0 involves an English-language novel manuscript from 1920 containing the word ʻukulele. The spelling with an ʻokina (ʻ; 'cutting off' = 'glottal stop') is an anachronism since the ʻokina wasn't used for Hawaiian words in English text a century ago. The use of the ʻokina when using Hawaiian words in English is a modern phenomenon and still not universal: Hawaii is still more common than Hawaiʻi in Hawaii.

20.2.20.23:55: WHITE RAT 1.27

? qulugh ai nai sair juri ? nyair

'white rat year, head month, twenty ? day'

1. Last night I was reading about 佛田洋 Butsuda Hiroshi.

洋 <OCEAN> is commonly read Hiroshi 'wide' as a name, presumably because oceans are wide.

佛田 <BUDDHA FIELD> Butsuda has a perfectly transparent reading, but ... does it really mean 'Buddha field', or is 佛 <BUDDHA> a phonogram for something else?

佛田 got me thinking about another Buddha-name, 大佛 Osaragi. My host mother in Japan in 1991 showed me a book by 大佛次郎 Osaragi Jirō, a novelist from her prefecture, and drew my attention to his surname 大佛 <GREAT BUDDHA> which looks as if it should be read Daibutsu ('Great Buddha', the normal reading).

Wikipedia has three explanations for why 大佛 is read Osaragi. I find none plausible:

A. < Osana-gi 'young-tree', the name of a mountain thought to resemble a big Buddha statue. The change of n to r is sui generis.

B. < Ō-saragi 'great-newcomer'. The shortening of ō 'great' is irregular (making it sound like o- 'small'¹!), and saragi 'newcomer' isn't in any dictionaries I've checked.

C. < Ō-sore-ki 'great-slash.and.burn.field.-place'. The shortening of ō 'great' is irregular, the change of o-e to a-a is sui generis, and sore 'slash and burn field' and ki 'place' aren't in any dictionaries I've checked.

¹The vowel length distinction between ō- 'great, big' and o- 'small' is unintentionally iconic. In Old Japanese the two sounded quite different: 'great, big' was əpə-, whereas 'small' was wo-. (The latter might be from an earlier *wə-, but in any case there was no primordial distinction between long /oː/ 'big' and short /o/ 'small'.)

2. I write a lot about unpredictable Japanese name readings here, but American English name readings of non-English names are also unpredictable, albeit not to the same degree: e.g., Heidi Przybyla's last name is [ˌpʰɹɛzˈbɛlə] ("Prezbella" according to her Twitter bio) which is far from what I think is the Polish pronunciation: [ˌpʂɨˈbɨla]. Nicolle Wallace calls her [ˌpʰɹɛzˈbɛlə] at 1:40 in this video. If I had not heard the video or seen "Prezbella" on Twitter, I would have guessed that her surname was pronounced [ˌpʰɹɪˈbɪlə] or [pʰɚˌzɪˈbɪlə] in American English.

3. I've read about Qing dynasty attempts to 'correct' earlier Chinese transcriptions of foreign names but never seen any examples until now. Shudder.

4. When looking up the Jurchen Empire era name 天輔 Tianfu 'Heavenly Assistance', I discovered that Wikipedia provides equivalents for each year in the calendars of the other (Greater) Sinospheric countries as well as the Western calendar: e.g.,Tianfu 3 equals

• 1119 AD, the 金史 History of the Jin Dynasty's date for the creation of the Jurchen script. Other sources give other dates. Last year may have been the nine hundredth anniversary of the script.

• 重和 Chonghe year 2 and 宣和 Xuanhe year 1 (the era name changed in the middle of the year) in China

• a less anachronistic reading of the names in quasi-Pinyin would be Chungho and Hüanho

• 天慶 Tianqing year 7 in the Khitan Empire

• *Tênking in Sino-Khitan

• <HEAVEN pu.?.l.ge.er> in the Khitan small script

• 元德 Yuande year 1 in the Tangut Empire

• 𗤩𗸣 3628 0644 *1ghwan4 1ty1 in Sino-Tangut?

• what was the Tangut era name?
  

• 會祥大慶 Hội Tường Đại Khánh year 10 in the Great Viet Kingdom (now Vietnam)

• the 12th century pronunciation was probably slightly different
  

• 元永 Gwen'weĩ (now Gen'ei) year 2 in Japan

¹<GREAT 089.ge.en> in the Khitan small script. What was its equivalent in the Khitan large script and Jurchen?

5. Wiktionary provides a phonological history of Sino-Japanese 祥 (the character in 會祥大慶 above):

/sɨau/ → /sʲɨɔː/ → /sjɔː/ → /sjoː/

Back in the 90s, I proposed that Sino-Japanese had /ɨ/ but never wrote a complete phonological history of Sino-Japanese. So it was neat to see someone else reconstruct /ɨ/ in earlier Sino-Japanese. Here's my take:

/sɨaŋ/ → /sjaũ/ → /sjɔː/ → /sjoː/ (or /ɕoː/)

Old Japanese saŋga 'omen' reflects the /sɨaŋ/ stage.

6. Looking for the Khitan equivalents of Tianfu and Tianqing in Andrew West's "A Chronology of Khitan Inscriptions" led me to realize that the first character of the Khitan large script spelling of pulugh 'intercalary'

might be read pu if its reading is ultimately from some para-Japonic cognate of Old Japanese puk- 'blow':

• 吹 represented Middle Chinese *tɕʰwie 'blow'

• 吹 was borrowed in Parhae as a semantogram to write the *pu- of an unrelated para-Japonic *puk- 'blow' (cf. how 吹 represents fu- of fuk- 'blow' in modern Japanese)
  

• 吹 was borrowed in Khitan as a phonogram to write the Khitan syllable pu

Here's another possibility with the same starting point:

• 吹 represented Middle Chinese *tɕʰwie 'blow'

• 吹 was borrowed in Parhae as a semantogram to write an unrelated Koreanic *pur- (or *pul-?) 'blow' (cf. Middle Korean 불- pur- 'blow')
  

• 吹 was borrowed in Khitan as a phonogram to write the Khitan syllable pul

If Manchu fulu < *pulu 'surplus, excess' is a borrowing from Khitan pulugh (centuries before 'Manchu' had that name: i.e., when Manchu was still Jurchen), it seems Jurchen dropped the final consonant -gh which its phonotactics would not permit.

Old Japanese puk- 'blow' and Middle Korean pur- 'blow' are sound-symbolic soundalikes no more related to each other than either is to English blow which also has a labial stop, a labial vowel, and a liquid like Korean. The mismatch between -k- and -r- cannot be explained away as suffixes (with no functions - a red flag!) or as different reductions of a *cluster (Korean has -rk-stems, but 'blow' is not one of them).

7. Reading about calendars led me to learn about the short-lived state of 大爲 Taewi which had but one era name, 天開 Chŏn'gae (1135-1136, a contemporary of the Jurchen Empire).

Wikipedia explains the role the Jurchen indirectly played in the establishment of Taewi:

It was during this period an organized Jurchen state [the Jurchen Empire] was putting pressure on Goryeo [Koryŏ dynasty Korea]. The trouble with the Jurchens was partly due to Goryeo's underestimation of the newly established state and the ill treatment of its envoys (i.e. killing them and humiliating their corpse). Goryeo's dislike for the Jurchens stemmed from the fact that they were once a subservient tribe under Goryeo's predecessor state Goguryeo [Koguryŏ], and took Jurchen assertion of equality with Goryeo as an offense. Taking advantage of the situation, Myo Cheong purposed to attack the Jurchens and that moving the capital to Pyongyang would assure success.

The king actually did listen to him and was persuaded. However, the rest of the court and the bureaucracy did not support the move, and the king had to back out of his commitments to Myo Cheong.

Eventually, Myo Cheong led a rebellion against the government. He moved to Pyongyang, which at the time was called Seo-gyeong (西京, “Western Capital”), and declared the establishment his new state of Daewi [= Taewi].

The Wikipedia entry on 妙淸 Myochhŏng's enemy 金富軾 Kim Pu-shik gives more insight on Koreo-Jurchen relations of the period: e.g.,

In fact, [Pu-shik's brother] Kim Bu-cheol (voicing a position of Gim Busik [= Kim Pu-shik] who was at the time in China) submitted a memorandum proposing to accede to the demands of Emperor Taizu of Jin [= the Jurchen Empire] giving the following rational[e]: "Now even the great Song calls itself the younger brother of the Khitan and they have gotten along peacefully for generations. And although there is nothing under heaven that can measure up to the dignity of the Son of Heaven [of Goryeo], submitting to and obeying the barbarians like this is the proper policy, one that the sages called 'the temporarily putting aside of one’s principles as circumstances demand it' and 'the protection of the whole country.' "

8. Which city in Vietnam used to be known as 大羅 Đại La? (Answer here.)

9. These etymologies for the names of the 三韓 Samhan 'Three Han' from Wikipedia based on 盧國屏 Lu Guobing (2002: 243) are puzzling:

Ma means south, Byeon means shining and Jin means east.

Lu provides no evidence for his equations which would imply that there are Korean(ic) words for 'south', 'shining', and 'east' resembling the early 1st millennium AD pronunciations of the three Chinese characters now read as Ma, Byeon, and Jin in Sino-Korean:

• 馬 *mæʔ
  

• 弁 *bɨan̥
  

• 辰 *dʑɨən
  

But I know of no such words. I don't even know of any surviving native Korean terms for the directions.

10. I am also puzzled by Wilkinson's (2001: 928) statement that 滿洲 Mǎnzhōu 'Manchu') is a transcription of a "non-Han" (presumably Manchu) word for 'brave'. I can't find any Manchu word for 'brave' resembling Mǎnzhōu or Manchu Manju 'Manchu'.

Wilkinson's romanization of Manchu ᠴᡳᠩ 'Qing' as Csing instead of Cing is unusual. Cs looks like what I'd expect in a Hungarian-based romanization of Manchu (Mandzsu?).

11. Why are Hungarian dz /dz/ and dzs /dʒ/ so often realized as geminates (according to Wikipedia)?

12. Also according to Wikipedia: the Hungarian spelling gy /ɟ/ "is a remnant of (probably) Italian scribes who tried to render the Hungarian sound": cf. Italian gi [dʒ]. Has gy ever been used as a digraph in any Romance language?

The spelling gy is 'asymmetrical' because the hypothetical 'symmetrical' voiced counterpart of voiceless ty /c/ would be dy.

13. I'm listening to "Bye Bye Szása" (1990) on the offisöl csenöl 'official channel' of PA-DÖ-DŐ. I'm guessing those are loanwords from English with ö corresponding to schwa. When I first came up with a phonetic alphabet for Hawaiian Creole English in the mid-80s, I used ö from German for schwa.

PA-DÖ-DŐ is French pas de deux in Hungarianized disguise.

László Király explains the significance of the song:

The lyrics is full of sarcasm:

• "Bye, bye Sasha, you treated me so bad, you were gone and I'm still looking for my watch." (It refers that Soviet solders in the 2nd world war forced people to give them their watches.)

• "I was so enthusiastic while learning your language. But how is it [i.e., how do you say] in Russian that My lover is gone?" (It refers that learning Russian in schools was obligatory but only a few people could speak it efficiently.)

14. The word featured in this week's Star-Advertiser Japan section is エゴサ egosa < エゴサーチ egosāchi 'ego-search': looking for oneself online. Turns out there's an English term from 1995 that I've never seen before: egosurfing.

20.2.19.21:03: WHITE RAT 1.26

? qulugh ai nai sair juri ? nyair

'white rat year, head month, twenty ? day'

1. The future of translation is now:

"We've Just Seen the First Use of Deepfakes in an Indian Election Campaign" (via reddit)

Out-of-sync dubs are a thing of the past.

2. I had never heard of Haryanvi before reading that story. The sample phrases on Wikipedia are in an ad hoc transcription. I assume k, kae, and kay represent the same interrogative word ke (cf. Hindi kyā 'what').

3. The Transformer Sixshot is known in Taiwan as ... 西夏 Xīxià 'Western Xia' = 'Tangut'! Are there any robots marketed as 契丹 Qìdān 'Khitan' or 女真 Nǚzhēn 'Jurchen'?

4. Gankeshi looks like 'gun-erasing' but refers to Gundam

5. These statements on Wikipedia puzzle me:

5a. In the entry for Kawamori Shōji:

"Patlabor: The Movie - Mechanical Design (Credited as Masaharu Kawamori)"

"Patlabor 2: The Movie - Mechanical Design (Credited as Masaharu Kawamori)"

5b. In the entry for Genesis Climber MOSPEADA (1983):

The incidental music was composed by Joe Hisaishi, who would later gain renown for incidental music for the movies of Hayao Miyazaki, though it is accidentally credited, because of a misreading of the name characters, to a "Yuzuru Hisaishi."

Normally Japanese names are written without furigana in credits, so there would be no way to distinguish between homographic names such as

• 河森正治 Kawamori Shōji¹ and 河森正治 Kawamori Masaharu

• 久石譲 Hisaishi Jō² (i.e., Joe) and 久石譲 Hisaishi Yuzuru.

I am guessing that "Masaharu" and "Yuzuru" are (understandable) errors in the translated credits of foreign (English-language?) releases of Patlabor and MOSPEADA.

¹I've misread 河森正治 as Kawamori Seiji. 正治 can be read as Seiji - it is the 104th spelling that Windows 10's Japanese IME suggests when I type <seiji>. But Seiji is not how Kawamori Shōji reads his own name. In many cases, there is no way to be absolutely sure about how to read a Japanese name without knowing how that person reads their own name.

²Windows 10's Japanese IME automatically suggests the entire name 久石譲 if I type <hisaisi>!

6. When I first saw the Japanese word itasha today I thought it was from 板 ita 'board' plus　車 sha 'car'. It turns out to be from　痛 ita 'pain' plus sha 'car' - which was a repurposing of an earlier イタ Ita (<  イタリア Italia 'Italy') plus sha 'car'.

7. The derived term 痛単車 <ita tan sha> itansha (not itatansha!) for itasha-like motorcycles has an interesting spelling: the syllable ta is represented twice, once by 痛 ita and again by 単 tan.

8. The term for itasha-like bicycles, 痛チャリ itachari, has a clipping of the mysterious word charinko 'bicycle'. One proposed origin is from Cheju 自輪車 charyungŏ 'bicycle', a Sino-Korean word not in standard Korean.

There is an older, unrelated word charinko meaning 'child pickpocket', a synonym of 掏児 <PICK.POCKET CHILD> suri (whose parts 掏 tō and 児 ji are not read su or ri in isolation). There is another spelling of suri, 掏摸 <PICK.POCKET FISH.FOR>, that is not necessarily for child pickpockets. (And its spelling also has components not read su or ri in isolation: 掏 tō and 摸 bo or mo. 掏摸 can also be read tōbo, a Sino-Japanese synonym of native suri.)

20.2.18.23:59: WHITE RAT 1.25

? qulugh ai nai sair juri tau nyair

'white rat year, head month, twenty five day'

Andrew West drew my attention to Tangut characters

𘜦𘜦

2298 and 2299.

Do you see any difference between them? There isn't any because Unicode has a single codepoint for both of them (U+18726).

Here is how they are treated in various sources starting with Sofronov (1968 II: 346), the first that assigns two numbers to the same shape:

The 1.92 reading means something like 'fast'.

The 2.85 reading has no known meaning.

There are two schools of thought on the initial of 'fast'. One believes 'fast' has a Class VIII initial (Li 1986's ɣ- and Li 2008's ·-); the other believes that 'fast' has a class IX initial lh-. Which school is correct?

The entry for 'fast' is in the rhyme 1.92 section of the Tangraphic Sea (92A62) and contains the fanqie

𗿉𗪌

3673 1ghu4 + 3674 1kwyr'4 = 1ghwyr'4.

'Fast' is also in the section of Homophones edition A 45B74 and edition B2 46A64 for characters with Class VIII initials and without any homophones.

So I just changed 1lhwyr'4 in my database for 2298 to 1ghwyr'4.

It seems that the index of Sofronov (1968 II: 346) has mixed elements of two entries. What appears as

3095 𘜦 ·i̭ẹ 2.85 VIII-76 (i.e., Class VIII initial, 76th character without a homophone)

3096 𘜦 lhi̭ẹ 1.92 IX

perhaps should appear as

3095 𘜦 lhi̭ẹ 2.85 IX

3096 𘜦 ·wi̭ẹ 1.92 VIII-76

the Tangraphic Sea fanqie is ·i̭u 1.3 + kwi̭ẹ 1.92 in Sofronov's 1968 reconstruction

Sofronov's 3095 - equivalent to 2299 in Li Fanwen's numbering which I use - would be transcribed as 2lhyr'4 in my system. But I can't figure out how he came up with a second reading lhi̭ẹ 2.85 (= my 2lhyr'4) for

𘜦

with lh- and rhyme 2.85. There is no second entry for that graph in Homophones or the Precious Rhymes of the Tangraphic Sea. (The second volume of the Tangraphic Sea with rhyme 2.85 characters is missing.) There is no other syllable read as 2lhyr'4 in my database, and no graph sharing elements with 2299 sounds like it. Here are all the graphs with 𘪘, the left side of 2299:

• 𗙜 0777 1shwe2

• 𗉮 0824 2chhy3

• 𘜡 2303 1shar3

• 𘜤 2521 1ha3

• 𘜨 3723 1chhwe2

• 𘜧 3726 2kwuq1

• 𗇜 5014 1wya4

• 𘜢 6028 1kwar4

• 𘜥 6029 1kwar4

• 𘜣 6034 1dzyr4

• 𘜠 6042 2shar1

There are forty-eight graphs with 𘥦, the right side of 2299 (other than 2298 and 2299) - too many to list here. None have readings with rhyme 2.85.

So I conclude that 2299 did not get a reading from a similar-looking character.

20.2.17.22:47: WHITE RAT 1.24

? qulugh ai nai sair juri ? nyair

'white rat year, head month, twenty four day'

1. Today I learned that John Cena's surname is Italian in spite of the fact that it is pronounced [ˈsiːnə] in American English instead of a more Italian-like [ˈtʃɛnə] or [ˈtʃejnə]. (The un-Italian final [ə] is inevitable due to the reduction of English unstressed vowels.)

2. Did blue falcon originate as a reference to this cartoon character?

3. I last visited Washington DC a year too early for the opening of Planet Word. More at the official site and this video.

4. I first heard RISD [ˈrɪzdiː] (Rhode Island School of Design) spoken last week on Gilmore Girls. RISD is the only acronym I can think of that is pronounced as a syllable combining the first few letters ([rɪz]) followed by a letter name ([diː]).

5. Tonight it occurred to me that the bizarre Sino-Jurchen vocabulary entry

337. <BAD be PERSON> ehe-be niyalma 'bad-ACC person' = 'offender'

involves a scribal error of <be> for <i>. The two phonograms look similar and represent case markers:

<be> 'ACC' vs. <i> 'GEN'

Kiyose (1977: 116) may have been the first to suggest that the term for 'offender' was actually ehe-i niyalma 'bad-GEN person', but he did not make a graphic argument for his emendation.

I myself have miswritten <i> as <be> - or was it the other way around?

20.2.16.23:41: WHITE RAT 1.23

? qulugh ai nai sair juri ? nyair

'white rat year, head month, twenty three day'

Most of this entry contains items from last night after my midnight deadline for the previous entry.

1. I listen to Ukrainian pop music on 101.ru. Some acts have Ukrainian names in romanization: e.g.,  Polina Krupchak - Мій Козак 'My Cossack' is sung by (not Поліна Крупчак). Such romanized Ukrainian names are a midpoint on a scale between a Ukrainian name in Cyrillic and a foreign(-sounding) name in the Roman alphabet: e.g., Arturro Mass, which I assume is a stage name. Mass' Twitter is in Russian even though he records in Ukrainian.

Romanized Ukrainian isn't just for names: e.g., the sleeve for MamaRika's "Проліски" (the title displayed on 101.ru) has "PROLISKY".

What does the choice of Roman letters - or Russian - say about the culture of Ukrainian pop music?

2. Having mentioned Ukrainian козак:

• Why does it have -o- if the Turkic original was qazaq?

• The Russian form was originally spelled козакъ with о as in Ukrainian. Why was it respelled as казак?

• How did qazaq also end up as Russian казах 'Kazakh' with a final fricative?

3. Surprise: Shor қазақ and Khakas хазах mean ... 'Russian'! What do Shor and Khakas speakers call Kazakhs?

4. Is German Metzger 'butcher' Latin in disguise?

5. I just learned about Cho and Whitman's Korean: A Linguistic Introduction (2020) via Alexander Zapryagaev.

LOL at the first sentence of the description:

The 'Korean wave' in music and film and Korea's rise as an economic power have boosted the popularity of Korean language study worldwide.

Hallyu itself is no laughing matter - it is a pillar elevating South Korea to global prominence - but I envisioned some Hallyu fan infatuated with Korean opening the book and being intimidated despite the cartoon cover (unusual for a linguistics book).

6. Also via Alexander Zapryagaev: Cho and Whitman's The Cambridge Handbook of Korean Linguistics which isn't listed on Amazon yet.

Not to be confused with Brown and Yeon, eds., The Handbook of Korean Linguistics (2015). John Whitman's chapter on Old Korean from the 2015 volume is online for free.

John credits me with "the idea of reconstructing deriving at least some cases of LMK [Late Middle Korean] /ye/ from a primary vowel /e/".

I was inspired in 1999 by Leon Serafim who reconstructed *e as a Proto-Koreo-Japonic source of Late Middle Korean /ye/ [jə]. Although I don't believe in Koreo-Japonic, I did think reconstructing *e at the Old Korean level made sense.

Last night I realized my contributions to Korean and Japanese historical linguistics are vocalic: Old Korean *e and the Old Japanese seven-vowel system.

7. Alexander Zapryagaev wrote about

those kanji that have a reading (pure, without okurigana) that can be confused as a 入声 [entering tone; i.e., *stop-final] Sino-Japanese reading.

I would have mistaken those native Japanese (i.e., kun) readings as Sino-Japanese in the mid-80s before I knew any Korean or Vietnamese and knew very, very little Mandarin. Mixed native-Chinese compounds like 夕刊 yūkan 'evening edition' (instead of a theoretical entirely Sino-Japanese sekkan; cf. Sino-Korean 석간 sŏkkan 'id.') reinforce the false impression of such kun readings being Sino-Japanese.

8. Today when I looked up a map of Israel on my phone, I got a map with city names in English and Arabic (but not Hebrew!). I saw that Cairo was القاهرة‎ Al-Qāhirah. The absence of -h- and the final -o in English Cairo made me wonder if the name was borrowed from Italian ... and tonight Wiktionary tells me it was.

Wiktionary says American places called Cairo are pronounced /kɛɹoʊ/ or /keɪɹoʊ/. Are those obsolete pronunciations of the English name of the Egyptian city that have survived as pronunciations of that city's namesakes?

9. Tonight I was listening to 三枝成彰 Saegusa Shigeaki's 「猜疑の勝利」 "Victory of Suspicion" which made me recall a question I forgot long ago: why is 猜 pronounced 시 shi in Korean? Theoretically it should be sae corresponding to Mandarin cāi, Sino-Japanese sai, and Sino-Vietnamese sai ~ xai (whose initials are irregular; the expected reading is thai).

The structure of 猜 is also odd. Wiktionary says it's a semantophonetic compound, but 猜 has no obvious connection to dogs (see the definitions for the morphemes it represents) and doesn't sound like its supposed phonetic 青 (except in Sino-Japanese).

(2.17.19:52: Table and caveat about phonetic similiarity in Sino-Japanese added.)

10. Last night I dug out one of my favorite movies, Megazone 23 Part I (1985), which was Matrix before Matrix.

Tonight it finally occurred to me after over three decades that the main character's name Yahagi has a spelling I would not have guessed if I weren't familiar with the movie: 矢作. 矢 is normally ya 'arrow', but I've never seen hag-i 'making (an arrow' spelled 作 <MAKE> anywhere else. hag-u 'make (an arrow)' is normally spelled 矧ぐ. Windows 10's IME gives seven choices for Yahagi:

• 矢作 <ARROW MAKE>
  

• 矢萩 <ARROW JPN.BUSH.CLOVER>
  

• 矢矧 <ARROW MAKE.ARROW>
  

• 矢矯 <ARROW RECTIFY>
  

• 八萩 <EIGHT JPN.CLOVER>

• 八作 <EIGHT MAKE>
  

• 谷萩 <VALLEY JPN.CLOVER>
  

ya 'eight' and ya 'valley' sound like ya 'arrow', and hagi 'Japanese clover' sounds like hag-i 'making (an arrow)'. I have no idea why 矯 <RECTIFY> (normally read kyō or ta[me-]) represents hag-i.

11. Via Bitxəšï-史: Shiotani and He's A Grammar of Monguor Language (2019). Monguor is an umbrella term for two languages, Mongghul and Mangghuer. The book is strictly speaking a grammar of Mangghuer.

I have no experience with any modern Mongolic languages other than Khalkha, so it's quite jarring to see Mandarin loanwords (in bold) mixed with native words. From page 14:

bi xuesheng bi

'I student be' = 'I am a student.'

damei-du qianbi bi

'I-DAT pencil be' = 'I have a pencil.'

The suppletive first person pronoun paradigms of Mongolic languages (e.g., Mangghuer bi 'I' and damei-, one oblique stem for 'I') are so unlike anything in Chinese.

20.2.9-20.2.15