21.11.12.23:45: WHITE OX 10.8

𘬈𘭌𘲐𘰯𘭧𘬌𘲺

? uni ai par sair nyêm nyair

'white ox year, ten month eight day'

Tonight I found the New Pohnpeian-English Online Dictionary edited by Prof. Kenneth L. Rehg, who introduced me to the Micronesian world back in the 90s, his longtime colleague Damian Sohl, and Robert Andreas.

Stephen Trussel worked on the software, so it's not surprising that the dictionary resembles the Austronesian Comparative Dictionary that he coauthored with Prof. Robert Blust, who introduced me to the Austronesian world twenty-five years ago.

The first entry in the English finderlist is sorapang 'abacus' from Japanese 算盤 soroban 'id.' I wonder why the word isn't *soropang. (Pohnpeian has no b, and -ng is closer to Japanese -n [ɴ] than -n.) Clicking on sorapang took me to the s-entries. Sohseng 'Korea' from Japanese 朝鮮 Chōsen 'id.' caught my eye.

Pohnpeian has no affricates, so Ch- is approximated as s-. (That pattern of borrowing is parallel to how Old Japanese borrowed all Chinese voiceless affricates as s-. The word now pronounced Chōsen was Teusen in Old Japanese. Japanese ch- is from the affrication of t-.)

Pohnpeian <oh> is [oː]. I should have guessed that the use of <h> to indicate vowel length was a convention adopted from German.

I don't know whether the <e> of <Sohseng> is [e] or [ɛ]. Why is <e> ambiguous in Pohnpeian orthography if /e/ and /ɛ/ are distinct phonemes according to Wikipedia? Are there too few minimal pairs to justify a digraph *<ea> for [ɛ] parallel to the digraph <oa> for [ɔ]?

Ah, seems I misunderstood. eh₁ is defined as

name of the letter e, the second letter of the Pohnpeian alphabet, used to represent the phoneme /ɛ/, a lower-mid front vowel which occurs in both the Northern and Kitti dialects, as well as the phoneme /e/, a mid front vowel, found only in the Northern dialect.

So there is only one nonhigh front vowel phoneme whose realization varies by dialect. Hence Sohseng would be [soːsɛŋ] or [soːseŋ] depending on dialect.

I wish the online dictionary had its own pronunciation key so I didn't have to rely on Wikipedia.

Who is the online dictionary primarily for? Learners or native speakers? How common is online access on Pohnpei?

Now that so many have smartphones, do print dictionaries make sense any more for small languages when one can access online dictionaries for free with the latest words like sehlpwohn 'cell phone' and interned 'internet'? (<d> is [t].)

21.11.11.23:57: WHITE OX 10.7

𘬈𘭌𘲐𘰯𘭧𘲧𘲺

? uni ai par sair ? nyair

'white ox year, ten month seven day'

1. Is Korean 까치 kkachhi 'magpie' related to Japanese kasasagi 'id.'? Let me try to force a relationship.

As far as I know, kasasagi is not phonetically attested in Old Japanese. According to Martin (1987: 441), the word appears in the 鎮國守國神社 Chinkoku-shukoku jinja manuscript of 名義抄 Myōgishō as kasasaki. Let's assume that -g- (originally /Nk/) in kasasagi is an innovation, perhaps by analogy with sagi 'heron'.

Now let's suppose that kasasaki is borrowed from an earlier Koreanic compound like *kàsá-tsàkí. Each half of the compound has the canonical pitch pattern *low-high typical of Koreanic disyllabic nouns (Ramsey 1991: 219). I am not going to speculate what each half meant.

At some point between borrowing into Japanese and Late Middle Korean, the Koreanic form underwent reduction:

• *kàsá- > *kàzá- > *kàá > kǎː

• *-s- lenited to *-z- and then disappeared

• but normally *-z- wouldn't disappear before Late Middle Korean!
  

• after the medial consonant, the low-pitched first vowel and the high-pitched second vowel became adjacent and fused into a single long vowel with a rising pitch

• *tsàkí > *tsʌ̀kí > *tskí > tshí

• the first vowel was reduced to ʌ and then nothing

• *-k- was reduced to aspiration of the preceding *ts-

Putting kǎː and tshí together results in Late Middle Korean 가치 kǎːtshí, the earliest attested form of the word. The word should have become modern Korean *kachhi with the same hangul spelling, but instead became kkachhi. Wiktionary explains:

The spontaneous gemination of the initial consonant occurred in the late nineteenth century. Spontaneous gemination is a recurrent phenomenon in Modern Korean, motivated by sound-symbolic effects.

See Wiktionary for non-Koreanic etymologies of kasasagi 'magpie'.

2. In modern standard Mandarin, 鵲 què 'magpie' and 雀 què 'sparrow' are homophones with aspirated initials. But in the Middle Chinese lexicographical tradition, 鵲 'magpie' had initial aspirated *tsʰ- whereas 雀 'sparrow' had initial unaspirated *ts-. Modern Chinese forms have both aspirated and unaspirated initials. How can the variation be explained? I have derived Tangut aspiration in part from a preinitial *K-. Could the aspirated forms of 雀 'sparrow' reflect a preinitial *k-? My Old Chinese reconstruction requires a minor syllable *CI- with a high vowel to account for the high vowel in Middle Chinese:

*CItsekʷ > *CItsiekʷ > *tsɨakʷ > *tsɨakʷ

Could aspirated readings be from *kts- < *kIts-?

3. Today I learned from Wiktionary that the Chinese name of the Crested Myna (八哥 <EIGHT OLDER.BROTHER>: Mandarin bāgē) is a phonetic transcription of Arabic ببغاء <bbghā`> babghā` ~ babaghā` ~ babbaghā` 'parrot'. Perhaps the Arabic original was disyllabic babghā`.

4. Forty-five years and a week ago today, Gaiking fought the 白鯨　Hakugei 'White Whale'. Was that Sino-Japanese term made up to translate Moby Dick (白鯨記 'Record of the White Whale' in Chinese)? Scripta Sinica has only one instance of it in its database of premodern Chinese texts in a 1499 entry in  朝鮮王朝實錄 Veritable Records of the Yi Dynasty.

11.12.1:45: I struggled to make up a Latin compound for 'white whale' but failed, so I settled for Greek leukophallaena. I don't know of any words with *albo-, which is what I'd expect for a combining form of Latin albus 'white'.

Wiktionary derives Latin ballaena 'whale' from Greek φάλλαινα. Why was Greek [pʰ] borrowed as Latin b-? Greek [pʰ] is from Proto-Indo-European *bh-. Could the Latin form reflect a variety of Greek that had not devoiced *bh-? That seems unlikely, as I know of no other evidence for such a variety.

Latin ballaena cannot be inherited from Proto-Indo-European, as b- would go back to the rare consonant *b-; the true Latin reflex of *bh- is f-: e.g., frater 'brother' corresponding to Greek φράτηρ 'member of a community' and Sanskrit bhrātr̥ 'brother', all from Proto-Indo-European *bhréʕtēr 'brother'.

21.11.10.23:55: WHITE OX 10.6

𘬈𘭌𘲐𘰯𘭧𘭬𘲺

? uni ai par sair ? nyair

'white ox year, ten month six day'

1. In today's installment of finally learning the incredibly obvious, I never realized that the -bek- in Uzbekistan was the Turkic title beg. What precedes it is uncertain.

2. I always thought beg was a loanword from Middle Chinese 伯 *pæk. Chinese unaspirated p- could be perceived as b, and p- was not originally possible as an onset in Turkic (Erdal 2004: 100). Although Chinese loans in Turkic could begin with p-, such loans could postdate beg which may date from an earlier period before initial p- was possible. The final -g is harder to explain, as Turkic phonotactics allowed -k. Had *-k already lenited to *-ɣ in the Chinese source variety? Turkic phonotactics did not allow *peɣ mixing front *e with back *-ɣ, so *beg preserved the vowel at the expense of the coda.

But Wikipedia mentions an alternative etymology from an Iranic reflex of *baga-. The problem with the Iranic etymology is the vowel: why isn't the Turkic form *bagh with back vocalism and back gh instead of front g? Was a front vowel needed to preserve -g?

3. Why will Kyrgyzstan soon "be the only independent Turkic-speaking country in a few years that exclusively uses the Cyrillic script"? In other words, what makes it different from the other ex-Soviet Turkic countries?

21.11.9.1:15: WHITE OX 10.5

𘬈𘭌𘲐𘰯𘭧𘬦𘲺

? uni ai par sair tau nyair

'white ox year, ten month five day'

Yesterday I found this page confirming my memory of brainwashing as a calque of 洗腦 xǐnǎo 'wash brain'). The page has a couple of other etymologies of interest:

• yen 'craving' < Cantonese 癮 [jɐn] 'addiction' (CantoDict entry)

I always assumed that had something to do with Japanese yen (also of Chinese origin but a different morpheme 圓 'circle').

Did Cantonese /ɐ/ ever have a fronted allophone before /j/? I can't think of any other Anglicization of Cantonese /ɐ/ with fronting.

• chop chop 'right away' < Cantonese 速速 [tsʰok tsʰok] 'hurry, urgent'

The expression makes me think of "a quickly moving knife" as the page puts it. Could it be English with Chinese-style reduplication?

速速 isn't in either CantoDict or Bauer's huge Cantonese-English dictionary. Is it obsolete?

I suppose English -p reflects an overextension of the labiality of [o] to the final stop or a folk etymology involving chopping. It would be interesting if the expression were first attested as something like choke choke with [k].

21.11.8.23:56: WHITE OX 10.4

𘬈𘭌𘲐𘰯𘭧𘲂𘲺

? uni ai par sair ? nyair

'white ox year, ten month four day'

1. Forty-five years ago today, the Battlehawk team fought a monster named 大魔公望 Daima Kōbō. 大魔 Daima is 'great devil'. What is 公望 Kōbō? Is it from 黄公望 Kō Kōbō, the Japanese name for Huang Gongwang?

Naver's Japanese dictionary gives the odd hybrid native/Sino-Japanese reading Kō Kinmochi for 黄公望, citing Wikipedia - which as of today doesn't mention that reading.

2. The Chinese dictionary I use (重編國語辭典修訂本) has a new front page but no entry for 公望. Only now did I learn that its author is 李鍌 Li Xian. I had to look up 鍌 Xiǎn, defined as 人名用字 'a character for personal names'. I guess <GOLD> on the bottom is  supposed to symbolize a positive quality beneath the phonetic 洗 Xiǎn (a surname or plant name usually read xǐ 'wash': brainwashing is a calque of 洗腦 xǐnǎo 'wash brain').

21.11.7.23:42: WHITE OX 10.3

𘬈𘭌𘲐𘰯𘭧𘯙𘲺

? uni ai par sair ? nyair

'white ox year, ten month three day'

1. Today is the sixty-fifth anniversary of the Japanese movie 空飛ぶ円盤恐怖の襲撃 Flying Saucers: Attack of Terror. The label on this film container has some interesting variants and simplifications:

• variants of 年, 画, 沢, 円

• 𬮦 with PRC-style simplification of 門 as 门

• 监: PRC-style simplification of 監

• 擊: simplified as 𨊥 but without the horizontal stroke over 凵
  

2. Two days ago I was surprised to see Nanzhao referenced in a GI Joe wiki entry on the Oktober Guard.

3. Yesterday I learned about the new TV series 境界戦機 Amaim Warrior at the Borderline. I would never have guessed that Amaim is read as アメイン Amein in Japanese.

4. Yesterday I discovered that 聖闘士星矢 Saint Seiya is built into the Windows 10 Japanese IME. Somebody at Microsoft is a manga/anime fan. Probably quite a few somebodies.