« previous post | next post »

Lately, since Xi Jinping made himself President for Life of the People's Republic of China, wags and wits have taken to calling the country over which he rules "Xina".

It turns out that this is the Catalan word for "China".  Curious to know how Xina is pronounced in Catalan, I looked it up on Wiktionary:

  • Balearic, Central /ˈʃi.nə/
  • Valencian /ˈt͡ʃ

We may compare this with Shina 支那, the old Japanese name for China, which is pronounced /ɕiꜜna/.  The Japanese name is directly borrowed from medieval Sinitic, which in turn comes from Sanskrit.

The Sanskrit word Cina (चीन IPA: [tʃiːnə]), meaning China, was transcribed into various forms including 支那 (Zhīnà), 芝那 (Zhīnà), 脂那 (Zhīnà) and 至那 (Zhìnà). Thus, the term Shina was initially created as a transliteration of Cina, and this term was in turn brought to Japan with the spread of Chinese Buddhism. Traditional etymology holds that the Sanskrit name, like Middle Persian Čīn and Latin Sina, derives from the Qin state or dynasty (秦, Old Chinese: *dzin) which ruled China in 221–206 BC, and so Shina is a return of Qin to Chinese in a different form.

The Sanskrit term for China eventually spread into China, where its usage was closely related to Buddhism. A Tang Dynasty (618–907) poem titled Ti Fan Shu (literally "preface to a Sanskrit book") by Emperor Xuanzong of Tang (685-762) uses the Chinese term Zhīnà (支那) to refer to China:

“Tí fàn shū”
hè lì shé xíng shì wèi xiū,
wǔ tiān wénzì guǐshén chóu.
Zhīnà dìzǐ wú yányǔ,
chuān ěr hú sēng xiào diǎntóu


Preface to a Sanskrit Text

Crane standing, snake shaped, its configuration never rests,
The script of India makes even ghosts and gods worry;
It leaves the Chinese disciples speechless,
While the pierced-eared Indo-Iranian monk nods smilingly.


Just as I was about to conclude this post, I noticed that Xina is also the name for "China" in Tetum, an Austronesian language spoken on the island of Timor.


[h.t. Jichang Lulu]


  1. Victor Mair said,

    November 26, 2018 @ 9:34 am

    From Johan Elverskog:

    The Swedish "Kina" is pronounced pretty much like Xina.

  2. J.W. Brewer said,

    November 26, 2018 @ 3:17 pm

    I guess there are already various Mandarin words such as 吸納 that would be pinyinized as "xina" and pronounced not that dissimilarly to the Catalan – maybe with the Balearic first syllable and the unreduced Valencian second syllable? If we went back to Wade-Giles, we would romanize the President-for-Life as Mr. Hsi (or maybe Chairman Hsi, after the example of Mao?), and I personally think Hsina would be a more attractive-looking word, although de gustibus etc.

  3. David Marjanović said,

    November 26, 2018 @ 6:39 pm

    India is "five heavens"?

  4. Victor Mair said,

    November 26, 2018 @ 7:13 pm

    Wǔ Tiānzhú 五天竺

    Someone else may post more on this later tonight.

  5. Chris Button said,

    November 26, 2018 @ 9:01 pm

    Incidentally it's actually this 天竺 that is often used erroneously in my opinion as some of the evidence behind reconstructing the original 天 character in Old Chinese with a lateral onset rather than the coronal as attested by its original phonetic 丁 (天 originally being composed of 丁 on top of 大). The 天 in 天竺 is associated with 祆 which appears to reflect a borrowing of the original "minor" Shang character 天 for the supreme Zhou deity after the Zhou conquest of the Shang.

  6. Shuheng Zhang said,

    November 26, 2018 @ 9:14 pm

    – A transcription of this poem in "Middle Sinitic":
    ghak lip zhia ghieng shiet mùi heuu
    ngó then mun dzí kúi zhin jou
    ci na déi tsí mu ngen ngió
    chion nhí gho seong sièu tém dou

    – My own English translation:
    Crane standing, snake shaped, configuration ceaseless;
    The Hinduka script worries ghosts and gods.
    While Chinese disciples are left speechless,
    Pierced-ear Indo-Iranian monk smilingly nods.

    – Notes:
    1) "eu" stands for /ɨ/, and "eo" stands for /ə/. There is still room for improvement in transcribing these two vowels, but one goal among all for this transcriptional system is typability.
    2) 支那, in Middle Sinitic, is "ci na" just like in Sanskrit. Whoever loaned the word 支那 into Chinese may be seeking for precision in reflecting its Chang'an sound at that time.
    3) For the English translation, I'm trying to make it rhyme, for I sense a jesting tone to the poem. Does it work?
    4) "Script of India" is the precise translation for 五天(竺)文字; I'm putting down Hinduka here for this is the Old Iranian word that may correspond to 胡僧 (Indo-Iranian monk). Do you think this opinion is far-fetched, or it might work in any sense?

    I would like to thank Mr. John Carlyle from the University of Washington for his help on both Chinese historical phonology and the transcription. The transcription system is also greatly attributed to his ideas and hard work. However, if there is any mistake, it would be my own.

  7. Chris Button said,

    November 26, 2018 @ 9:16 pm

    If The Sanskrit "Cina" चीन does indeed ultimately derive from "Qin" 秦 then what would be the reason for the final vowel? Surely Sanskrit "Cin" चिन् would have been more appropriate?

  8. Victor Mair said,

    November 27, 2018 @ 12:45 am

    From Fred Smith:

    The easiest answer is that Sanskrit usually placed a short a after consonants, unless there was a reason not to. I can’t think of any Sanskrit words offhand that are constituted of a single syllable cvc with the v or vowel as a long ī, and that does not have the short a after the second consonant. I know that there’s at least one Sanskrit text with the term cīnācāra- in the subject, the behavior of the Chinese. For example we have the Cīnācāraprayogavidhi, instructions on the practices used in Chinese behavior. I haven’t looked at it, but it should be interesting. Maybe I‘ll try and find it.


    Two more. I find reference to the Mahā-Cīnācāra-Sāra-Tantra. It might appear in the Penn library. Also, I’ve found a couple of references to cīnācāra- as Tibetan practice or Tibetan way. I‘ve been sitting at O’Hare airport for about 7 hours. So I have nothing better to do.

  9. Quim said,

    November 27, 2018 @ 6:20 am

    Some nice bilingual posters: (catalan-chinese)×1024.jpg×1024.jpg

  10. Victor Mair said,

    November 27, 2018 @ 8:37 am

    From Dan Boucher:

    Cīna doesn’t really mean China but the Chinese (occurs, for example, in a list of foreign peoples in Manu, ch. 10, which is fairly early, turn of the CE). So it does make more sense for it to occur in this form; Cin just looks entirely unnatural as a Skt. word to me, and it would be hard to know even how to decline it. Moreover, transcriptions between languages, esp. very unrelated languages, are never precise. This doesn’t seem something worth fretting a lot about in my opinion.

  11. Victor Mair said,

    November 27, 2018 @ 8:39 am

    What Dan Boucher says about Cīna matches perfectly with what Fred Smith says about it.

  12. Victor Mair said,

    November 27, 2018 @ 8:41 am

    From David White:

    The default declension in Sanskrit is the short -a (deva). Final consonantal -n endings are very rare except for adjectival/possessive forms in -an, vān or -in. But since there is no paradigm for a declension with a long ī -īn ending, the Sanskrit transliteration defaulted to the deva paradigm.

  13. Victor Mair said,

    November 27, 2018 @ 11:42 am

    From Andrew Ollett:

    i don't have references to hand, but lots of words ending in consonants were borrowed into indic and iranian languages as "thematic" stems, ending in -a. kuṣāṇa (貴霜) is another example. it is probably less strange that it would seem, since the final syllable was very weakly pronounced in contemporary indic and iranian languages anyway. final vowels are "just for show" in sogdian and khotanese, i think, and in the greek phonetic transliteration of middle indic words on 1st c. coins, final vowels are sometimes left out (ΙΑΗΑΡΑΤΑϹ for kṣaharātasa on the coins of nahapāhan).

  14. Chris Button said,

    November 27, 2018 @ 11:52 am

    @ Fred Smith, Dan Boucher, David White (all via Victor Mair)

    Thank you so much for clarifying that. So while theoretically possible, it's overridden by basic phonotactics.

  15. Chris Button said,

    November 27, 2018 @ 11:55 am

    @ Andrew Ollet

    it is probably less strange that it would seem, since the final syllable was very weakly pronounced in contemporary indic and iranian languages anyway.

    Thanks – yes, as the default "syllable" so to speak.

  16. Victor Mair said,

    November 27, 2018 @ 12:16 pm

    From Michael Witzel:

    the Sanskrit -a stems are the most common ones (like the related Greek/Latin -o ones) and they function as a sort of default.

    Note for example these northwestern peoples in Skt:
    Hūṇa (Huns) often mentioned together with Cīna in the Mahabharata, Kuṣāṇa for the Tukhara/Tuṣara (in Atharvaveda Pariśiṣṭa: Xinjiang people including the Kuṣāṇa), Śaka (Scythians/C. Asian Saka), Parthava (Parthians/Pahlava), Parada for Persians, Tājika (medieval, modern Tajiks), Kamboja in Afghanistan, etc. etc.

    An -n stem as in Cīn would have created many problems in declination and Sandhi.

    Hindi Cīn is just the modern North Indian pronunciation that cuts (like in Nepali) most of the final -a; they are kept in South Indian languages.

  17. Victor Mair said,

    November 27, 2018 @ 4:31 pm

    From Deven Patel:

    It's quite common in Sanskrit to assimilate foreign Proper names into the appropriate phonological and, in this case, morphological norms of Sanskrit. So we have the vowel gradation in cIna, that extends the 'i' to a long vowel as well as the addition of the masculine nominal affix 'a' to what one assumes is the foreign word 'cin' (qin). The same process is seen with the word 'hun' — 'hUna' is how it is rendered in Sanskrit.

RSS feed for comments on this post