Tangut Manuscripts with Tibetan Phonetic Glosses

British Library Or.12380/3495

IDP : British Library Or.12380/3495


This is a single sheet of a Tangut Buddhist manuscript held at the British Library in London. It has the pressmark Or.12380/3495 (originally K.K.II.0280.s), and was collected by Aurel Stein from Khara-Khoto during his expedition to Central Asia of 1913–1916. It has been studied by Arakawa Shintarō (Arakawa2008) and Tai Chung Pui (TCP2008), and is available online at the International Dunhuang Project.

The Tangut text comprises five columns of clear and neatly-written characters inside a double-lined frame, giving it the appearance of being a faithful manuscript copy of a printed text. The Tibetan phonetic glosses, which are written in a rather untidy and not very legible cursive headless script, are written in lighter coloured ink, and do not appear to be contemporary with the copying of the Tangut text. For some reason the glosses only cover the inner three columns of the Tangut text, in total 27 characters.

At the top left of the page is a line of Tibetan text in the same ink and calligraphy as the phonetic glosses, unfortunately missing its beginning as the top left corner of the sheet is torn off, reading ... ?se 'dzwar 'jo ste སེ་འཛྭང་འཇོ་སྟེ or ... ?se 'dzwang 'jo ste སེ་འཛྭར་འཇོ་སྟེ (Arakawa reads it as s[t?]e / 'dzwang / 'jo / te). The meaning of this is unclear to me. On the top right of the page, in the same ink as the Tibetan glosses, is a drawing of a flame-crowned "wish-fulfilling jewel" (ཡིད་བཞིན་ནོར་བུ), presumably in illustration of the "precious jewel burning bright and dignified" that is described in the fourth line of the Tangut text.


Position TCP2008 Tangut Tibetan Note
Character LFW2008 Reading Arakawa
TCP2008 BabelStone
01:01 180101 𗌮 L1543 mjor [1]
01:02 180102 𗆐 L2373 ljịj
01:03 180103 𘕾 L5851 njijr
01:04 180104 𗟭 L1274 wo
01:05 180105 𗣼 L2748 tśhja
01:06 180106 𗳦 L1737 ka
01:07 180107 𗫨 L3613 dwewr
01:08a 180108 𗩴 L2636 ne̱w
01:09a 180109 𗦻 L2639 mji̱j
01:08b 180110 𗪛 L2634 dźjwow
01:09b 180111 𗣼 L2748 tśhja
01:10 180112 𘉐 L4587 ·io̱w
01:11 180113 𗢳 L2852 tha
02:01 180201 𗉛 L1262 źjị g‑zhi g‑zhi གཞི gzhi གཞི
02:02 180202 𗷫 L1120 njɨ̱ nag g‑na གན gn{i} གནི [2]
02:03 180203 𘛥 L5130 rjur ru ru རུ ru རུ
02:04 180204 𗌻 L1346 ·jar ʼ‑gyar 'gyar འགྱར
02:05 180205 𘕕 L5865 sọ so so སོ so སོ
02:06 180206 𘈽 L2392 sjwɨj b‑se b‑se བསེ bse བསེ
02:07 180207 𗄊 L0010 źji g‑za g‑za གཟ gza གཟ
02:08 180208 𘂤 L5993 kha kha kha kha
02:09 180209 𗠁 L0206 bu̱ ʼ‑bu v‑bu འབུ 'bu འབུ
03:01 180301 𗒘 L5057 ɣiej ʼ‑ge v‑ge འགེ 'ge འགེ
03:02 180302 𗆤 L1913 dźjiar ʼ‑ji '[?]i ? འ​ི [3]
03:03 180303 𗄻 L2699 nwə nag gn{e} གནེ [4]
03:04 180304 𗥤 L3574 tsjij tse tse ཙེ tse ཙེ
03:05 180305 𘕕 L5865 sọ so so སོ so སོ
03:06 180306 𗐯 L4719 kiẹj ke ke ཀེ ke ཀེ
03:07 180307 𘛽 L1546 ljụ le lu ལུ
03:08 180308 𗦳 L3266 dzju ʼ‑dzu 'dzu འཛུ [5]
03:09 180309 𗇋 L3818 mjijr rme rme རྨེ
04:01 180401 𘊛 L2191 dzjọ g‑zo g‑zo གཟོ gzo གཟོ
04:02 180402 𘏨 L5655 ljɨ̣ glu ldi ? ལྡི [6]
04:03 180403 𗊏 L2583 nji b‑rnu brnu ? བརྣུ [7]
04:04 180404 𗍊 L0290 sju su su སུ su སུ
04:05 180405 𘉡 L2596 pjụ pu pu པུ pu པུ
04:06 180406 𗆬 L1902 wer we we ཝེ we ཝེ
04:07 180407 𘉍 L4573 bji ʼ‑bi v‑bi འབི 'bi འབི
04:08 180408 𘔉 L4628 dwər ʼ‑dwar v‑??r འ◌ར 'dwar ? འདྭར [8]
04:09 180409 𗯼 L5185 dźja̱ ʼ‑rbu '[?]u ? འ​ུ [9]
05:01 𗌮 L1543 ? mjor [10]
05:02 𗆐 L2373 ? ljịj
05:03 ?
05:04 180501 𘞃 L1329 dźjow
05:05 180502 𘟙 L3830 njij
05:06 180503 𗗙 L1139 ·jij
05:07 180504 𗡙 L4531 ·jow
05:08 180505 𗍺 L0752 tśja
05:09 180506 𘐔 L5771 tshwew


  1. There are no Tibetan glosses on the right hand side of the first column of Tangut text.
  2. The reading of the Tibetan gloss is uncertain. Arakawa reads it as nag ནག, but Tibetan transcriptions of Tangut do not normally indicate a final ‑g. On the other hand, Tai Chung Pui reads it as gna གན, but without an expicit final letter 'a (i.e. གནའ) this would normally be read as gan which makes this reading dubious. Neither of these readings (nag or gna) closely refelects the reconstructed reading of the corresponding Tangut character (njɨ̱), so I suspect that the gloss is a mistake for gni གནི, with a missing vowel sign.
  3. The Tibetan letter under the vowel sign is an illegible blob, but from the reconstructed reading of the Tangut character (dźjiar) it is probably a j , giving the reading 'ji འཇི.
  4. This is the same phonetic gloss as discussed in Note 2, but for a Tangut character with a quite different reading (nwə). Arakawa again reads it as nag, but Tai Chung Pui does not provide a reading. I suspect that the gloss is a mistake for gne གནེ, with a missing vowel sign.
  5. The Tangut character could plausibly be either L3240 "complete" or L3266 "host" (both reconstructed as reading dzju). Arakawa and Tai Chung Pui both prefer to read it as the latter character.
  6. The Tibetan gloss is unclear. Arakawa reads it as glu གླུ, but I tentatively read it as ldi ལྡི with a black smudge of ink above the vowel sign.
  7. The reading of brnu for the Tibetan gloss is uncertain, and the vowel does not match the reconstructed reading of the Tangut character (nji). Another possible reading is bdwu བདྭུ, but that would be an even poorer match for the Tangut reading.
  8. The reading of the Tibetan gloss is not certain as there is a black smudge at the bottom of the letter d. I follow Arakawa, who takes the smudge to be a subjoined letter w and reads the gloss as 'dwar.
  9. The reading of the Tibetan gloss is not at all certain. Arakawa reads it as 'rbu འརྦུ or possibly 'dwu འདྭུ, neither of which correspond to the reconstructed reading (dźja̱) or the Tibetan phonetic gloss given in other manuscripts ('ja' འཇའ).
  10. There are no Tibetan glosses on the right hand side of the last column of Tangut text. The top left corner of the page is torn off, so all but the right edge of the top three Tangut characters are missing. Arakawa plausibly reconstructs the first two characters as L1543 and L2373.


Published: 2011-12-07. Last modified: 2011-12-19.

Tangut Manuscripts with Tibetan Phonetic Glosses

BabelStone Home Page