I've always wondered why the 音読み of the kanji 洗 is セン when, in all Chinese topolects I'm aware of, the reading ends in a vowel.

I always chalked this up to Japanese having borrowed a Middle Chinese reading at a time when 洗 still ended in a nasal (cf. the Japanese vs Chinese reading of 鳥), but I was surprised to find that this isn't the case, and that it's reconstructed as *sejX in Middle Chinese.

Why, then, is there a nasal in the Japanese reading (and according to Wiktionary, the Korean reading too)?


I assume you're referring to the Baxter-Sagart Middle Chinese transcriptions of the Qieyun rime dictionary. When I search these transcriptions for 洗, I find two readings listed:

  1. *sejX, corresponding to modern Mandarin xǐ, Japanese セイ・サイ, and Korean 세 se
  2. *senX, corresponding to modern Mandarin xiǎn, Japanese セン, and Korean 선 seon

So I think the final nasal already existed in Middle Chinese and was borrowed into Japanese. This may not have been obvious because xǐ is the usual reading in modern Mandarin while セン is more common today in Japanese.

As for xiǎn, I don't know when this reading is used, but it is at least listed in my dictionary. My Chinese is too rudimentary for me to say any more on the subject. The Japanese and Mandarin readings listed above are from 新漢語林, and the Korean readings are taken from Wiktionary.

(For those who are curious, the final X in Baxter's notation indicates a rising tone.)

  • 1
    @3to5businessdays What's so strange about that? /xjangH/ has an /x/ initial, which in Japanese appears regularly as /k/, and /ang/ appears regularly as /oː/ (via /au/). – Zhen Lin Sep 8 '14 at 7:58
  • 3
    @3to5businessdays - those are just the regular results of Japanese's borrowing strategies. Japanese didn't have /x/ or /ŋ/, and so it borrowed those as /k/ and /u/ (which were the closest it had) - so *xjangH became /kau/. /au/ later became /oo/, giving /koo/. – Sjiveru Sep 8 '14 at 16:22
  • 3
    (And don't be misled by /h/ in modern Japanese – at the time Chinese words were being borrowed, it was a labial sound like [p] or [f].) – Zhen Lin Sep 8 '14 at 18:28
  • 1
    @3to5businessdays, Middle Chinese /x/ evolved into an /h/ in some dialects, suggesting that a hard /h/ like the [ch] in Scottish loch was a possible realization of this sound. Hard /h/ to /k/ is a closer phonetic shift than starting from the slightly-affricated apical sibilant /x/ as realized in modern Mandarin. – Eiríkr Útlendi Nov 5 '14 at 9:12
  • 1
    @Eiríkr Middle Chinese /x/ here is an IPA /x/, not a Pinyin /x/. In other words, the /x/ in the Middle-Chinese reconstruction /xjang/ is the ‘hard h’ sound found in Scottish loch /lɒx/ which has remained in some dialects (like Mandarin) and has become /h/ in others (like Cantonese). In Middle Chinese, it was still /x/ throughout, though, and it was quite regularly represented in Japanese by /k/. The sound denoted by ⟨x⟩ in Pinyin is /ɕ/, which is the result of a much later palatalisation of /x/ (and /s/ and /ʂ/) before front vowels, which is why /xjaŋ/ is now /ɕaŋ/, but Pinyin ⟨xiàng⟩. – Janus Bahs Jacquet Feb 29 '16 at 3:00

According to Pulleyblank's Lexicon of Reconstructed Pronunciation (http://books.google.com/books?id=qWGIxP1R4P4C, p336), there's a reconstructed EMC pronunciation of 洗 as *sɛn' (no idea what the apostrophe means, though - glottal stop?). This apparently corresponds to a modern Mandarin pronunciation of xiǎn, which also has a nasal final.

There may also be a reconstructed pronunciation of 洗 corresponding to modern Mandarin xǐ (the other option a lookup gives, and the pronunciation I assume you're referring to), but Google Books won't give me the page it should be on. If anyone has that book (or Google Books does give you that page), look it up somewhere on pages 330-332 and let me know if it's there.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.