Friday, 28 August 2009

How Complex is Tangut ?

Last year my friend Nathan Hill kindly invited me to give a talk on Tangut at my Alma Mater. I accepted with some trepidation because I am still very much at the start of a long and steep learning curve with regards to Tangut, but I hoped that by the time the talk was due to be given in May this year I would have something interesting and exciting to talk about. Unfortunately I got tied up with other stuff (Tangut, ironically), so in the end my talk turned out to be more of a general introduction to the structure of the Tangut script and some of the issues that I have faced over the last year or so in preparing an encoding proposal for Tangut. But anyway, the talk didn't go too badly, and so I thought that I would convert my PowerPoint slides into a four-part series of blog posts.



Notes for an introductory talk on the Tangut script given at SOAS on 21st May 2009



1.1 The Age of New Scripts

During the 10th to 13th centuries a number of new scripts were devised by peoples who had come into contact with (and conflict with) China, and who wanted to assert their national identity and cultural superiority by means of their own, unique and distinct writing systems (colour-coded to show their current Unicode status):


[See Documents relating to the encoding of the Tangut, Jurchen and Khitan scripts for Unicode encoding proposals]


Three of these scripts, Large Khitan, Jurchen and Tangut, are structurally similar to Chinese, and I will look at their similarities and differences, both amongst themselves and in relation to Chinese, below.



1.2 Khitan Large Script

  • Closely modelled on Chinese
  • Many characters borrowed directly from Chinese
  • Some with the same meaning (e.g. 皇帝 in the text below)
  • Some as phonetic borrowings
  • Many other characters derived from Chinese characters by adding or removing strokes (e.g. 東 with two extra strokes on the 6th line from the right in the text below)
  • Few or no characters composed of multiple elements with large numbers of strokes (i.e. no characters like Chinese 雙)
  • Uses exactly the same stroke types as Chinese
  • Largely undeciphered

Transcription of a Khitan Memorial Stone

Source: Miínzú Yǔwén 民族语文 2005 no.4 page 54

Click here to highlight Khitan characters that are the same as Chinese characters



1.3 Jurchen

  • Very similar to Khitan Large Script
  • Many characters derived from Khitan and/or Chinese
  • Relatively few direct borrowings from Chinese compared with Khitan
  • No characters with large numbers of strokes or composed from multiple complex elements
  • Uses exactly the same stroke types as Chinese
  • Largely deciphered

Drawing of a "Medallion" with a Jurchen inscription

Source: S. W. Bushell, "Inscriptions in the Juchen and Allied Scripts" in Actes du Onzième Congrès International des Orientalistes (1897) 2nd section page 21
(originally from Fāngshì Mòpǔ 方氏墨譜 [Mr. Fang's Catalogue of Inkstones] (1588) vol. 1 folio 33)


Table of Chinese, Khitan and Jurchen Numerals

Source: Daniel Kane, The Sino-Jurchen Vocabulary of the Bureau of Interpreters (1989) page 21



1.4 Tangut

  • Only superficially similar to Chinese
  • Characters are not obviously derived directly from Chinese or Khitan characters, although they are clearly influenced by Chinese
  • Discrete elements arranged into a square character
  • Appears crowded compared with Chinese, with few non-complex characters
  • Most characters composed of two or three distinct components, and only a few characters are themselves elemental components
  • Mostly written using the same stroke types as used for writing Chinese, but some stroke types and stroke constructions are unique to Tangut
  • Higher proportion of diagonal and oblique strokes than in Chinese
  • No closed elements (i.e. no box elements like Chinese 口 and 囗)

Chrysographic Edition of the Lotus Sutra

Source 中国少数民族文字字符总集


Fragment of a Memorial Stone from the Western Xia Royal Tombs

Source: 大夏寻踪——西夏文物特展 (Vanished Exhibition on Western Xia artefacts at the National Museum of China)

[Can you spot the characters meaning "one" and "three" ?]



1.5 Stroke Complexity

Tangut is renowned as being very complex in terms of the structure of its individual characters, but I wanted to try to determine exactly how complex Tangut is, and how it compares with Chinese, Khitan and Jurchen, so I produced the following graphs to show the distribution of characters by stroke count in these various scripts.


Distribution of Tangut Characters by Stroke Count

Data derived from Proposal for a revised Tangut character set for encoding in the SMP of the UCS (SC2/WG2/N3577) Appendix A.


Distribution of Traditional CJK Characters by Stroke Count

Data derived from the kTotalStrokes field of the Unihan Database for those characters defined in Unicode 1.0 (i.e. U+4E00 through U+9FA5), excluding simplified characters (mostly those characters with a kTraditionalVariant field).


Distribution of Simplified CJK Characters by Stroke Count

Data derived from the kTotalStrokes field of the Unihan Database for those characters defined in Unicode 1.0 (i.e. U+4E00 through U+9FA5) that have the kXHC1983 field but do not have the kSimplifiedVariant field (i.e. most simplified characters in the 1983 edition of Xiàndài Hànyǔ Cídiǎn 现代汉语词典).


Distribution of Large Khitan Characters by Stroke Count

Data derived from the transcription of a Khitan memorial stone given in Miínzú Yǔwén 民族语文 2005 no.4 page 54 and page 55.


Distribution of Jurchen Characters by Stroke Count

Data derived from Jin Qizong 金啓孮, Nüzhenwen Cidian 女真文辞典 [Dictionary of Jurchen Characters] (Beijing: Wenwu Chubanshe, 1984).


Stroke Count Data for Traditional CJK, Simplified CJK, Tangut, Jurchen and Khitan

StrokesCJK TraditionalCJK SimplifiedTangutJurchenKhitan
1102030
23722066
3806002528
4157143316552
52402153228760
63863516540141
766456816029334
895775931014718
91,1258515243710
101,369923773134
111,55590184702
121,63687088500
131,54676178200
141,44659464000
151,50253447300
161,25140933600
171,02031117300
1879317510600
197161686000
205191052900
21394791500
2230447600
2324040100
2414921100
2510722000
26546000
27521000
28261000
29131000
3080000
3150000
3231000
3341000
3400000
3510000
3611000
3700000
3800000
3910000
4000000
4100000
4200000
4300000
4400000
4500000
4600000
4700000
4810000
Total18,3738,9436,2211,377255
Mean13.4611.4912.096.015.43
Mode12101265

Comparison of CJK, Tangut, Jurchen and Khitan Stroke Counts


Jurchen and Large Khitan are the two scripts that appear to be most similar to Chinese, yet actually they are the most different when it comes to stroke count, both having only half the number of strokes as traditional CJK characters on average. This difference is probably due to the fact that Large Khitan and Jurchen characters do not have any high stroke count radicals such as 言 "speech" (7 strokes), 金 "gold" (8 strokes), 馬 "horse" (9 strokes) and 鳥 "bird" (9 strokes) that are very common in Chinese characters.

On the other hand, it was a surprise (to me at least) to see how closely the contour of Tangut matches that of traditional Chinese, as I had always assumed that Tangut characters must, on average, be much more complex than Chinese characters. But although Tangut does not have any characters with very few strokes (less than 4 strokes) or very many strokes (more than 24 strokes), which distinguishes it from Chinese, if you ignore the lower and upper ends of the graph the distribution of stroke counts for Tangut is very close to that of traditional Chinese. Why then does Tangut text look so much more complex and more crowded than Chinese? That could be answered with another graph which took into account each character's frequency of occurence. A large proportion of high frequency Chinese characters have very few strokes (e.g. 一二三人女山火水大小中), and conversely Chinese characters with very many strokes tend to occur less frequently, with the result that normal Chinese text always has a large proportion of characters with few strokes. In contrast to the situation with Chinese, there does not appear to be any relationship between frequency and stroke count for Tangut characters, so that normal Tangut text is uniformly composed of characters with 12±6 strokes, with the result that it appears denser and more crowded than Chinese.



1.6 Structure of Tangut Characters

  • Individual Tangut characters not obviously derived directly from Chinese or Khitan characters
  • Limited set of component elements
  • Elements are themselves built from simpler elements by the addition of 1 or 2 strokes
  • Most characters constructed from 2 or 3 component elements
  • Very few basic elements are also characters in their own right

Series of components are constructed from a basic element, on the one hand by the addition of strokes to the basic element to make other simple components (vertical progression in the diagrams below), and on the other hand by combining these simple components with other components to make complex components (horizontal progression in the diagrams below).


Series of Tangut Components (Example A)


Series of Tangut Components (Example B)


Due to this incremental process many character components are very similar to each other, and when two or three such similar components (coloured red in the diagram below) are combined together in different combinations to make different characters (coloured blue in the diagram below), the results are confusingly confusable.


Eleven Characters composed from different combinations of Five Components



1.7 Tangut Radicals

  • Not true radicals (determinatives)
  • But simply aids to character lookup
  • Chinese dictionaries select leftmost or topmost character element as the radical
  • Most Russian dictionaries base the radical on the character element at the bottom right corner of the character

In the example below, the same radical is used in both Li Fanwen's dictionary and Kychanov's dictionary, but in the former it is a lefthand radical, and in the latter it is a bottom right radical. This shows how most horizontally aligned components can occur equally on the left side or on the right side of a character, and it is largely an arbitrary decision of dictionary compilers as to whether it is treated as a lefthand side radical or a righthand side radical.

Li Fanwen 2008 Kychanov 2006

The proposed Unicode character ordering is based on 527 left-based radicals (including some top, bottom and enclosing radicals where there is no lefthand component). The advantage of this system of ordering is that it is consistent and allows for deterministic lookup of characters, but the disadvantage is that there are some high stroke-count radicals with very few members.

N3577 Appendix A



1.8 Structural Analysis

  • Because Tangut characters are composed of a limited set of component elements arranged in different configurations they are very amenable to structural analysis
  • Nishida’s 1966 dictionary gives structural analysis of each character

Table of Tangut Component Configurations identified by Nishida

Source: Nishida Tatsuo 西田龍雄, Seikago no kenkyū 西夏語の研究 (1964) page 246


Entry in Nishida's 1966 Tangut Dictionary

Source: Nishida Tatsuo 西田龍雄, Seikabun Shōjiten 西夏文小字典 (1966) no. 10-103


The Unicode proposal gives an Ideographic Description Sequence (IDS) for each proposed character. This borrows a character description syntax designed for CJK characters (but which will no longer be restricted to CJK characters from Unicode 6.0).


Sunday, 16 August 2009

Antedating the Caron

The Holy Grail for historians of character encoding is to discover why the háček (ˇ)—a diacritical mark which looks like an inverted circumflex—is called a caron in character encoding standards such as Unicode. Until the recent popularisation of Unicode, the term caron seems to have been virtually unknown outside of character encoding standards, and has no obvious etymology or source, whereas the Czech name háček (diminutive form of hák "hook") was and still is a widely-used name for this mark (it is also less-commonly referred to as a wedge, inverted circumflex or inverted caret).

The question of the caron has time and time again vexed the Unicode and Unicore mailing lists, but thusfar no-one has been able to pin down when the term was coined or what the etymology of the term is. In 2001 Unicode Vice President Rick McGowan wrote a short tongue-in-cheek screenplay explaining how the term might have come into being :


How These Things Get Started, Chapter Two

Scene: The tiny cramped office of General Frump, commander of an obscure military base somewhere north of the 89th parallel... Everyone and their baby fur seal is suffering from walking pneumonia and dysentary...

General Frump: Thank you for your report on the enemy code, Corporal, is there anything else? I'm quite busy...

Corporal Dolt: Well, sir, there's the matter of the "hacek", which some thought might be confused with "hatchet", sir. But as you know, sir, there simply is no other term, and I thought, perhaps...

General Frump (eyes glazed): Carry -- ahem -- on, Corporal. Ahem. Carry -- ahem -- on.

Corporal Dolt (saluting): Very well, sir. "Caron" it is.

General Frump: Ahem... Dismissed, Corporal...


And seven years later Unicode guru Ken Whistler suggested that the following little dialogue at an ISO committee meeting may be closer to the truth:


Delegate from Slovakia: You can't spell "WITH HACEK" that way — it has a hacek on the C.

Convenor from Switzerland: Well, we can only use ASCII A-Z in character names.

Delegate from Slovakia: Well, it's spelled wrong, and that isn't acceptable to us.

Convenor from Switzerland (with a straight face): In Swiss French we call it a "caron", and there wouldn't be any trouble spelling that.

Delegate from Slovakia: Really?

Convenor from Switzerland: Yes, so let's just use that term instead. (aside to editor) Just change them all to "WITH CARON" and let's move on.


A couple of the more serious etymologies that have been proposed for the word "caron" are :

  1. that the word "caron" is a portmanteau for "caret" (a inverted v shaped editorial symbol used to indicate an ommission) and "macron" (a diacritic mark shaped like a horizontal line that is placed over a letter), the name suggesting the (inverted) shape of the caret and the above-letter positioning of the macron (suggested by James Naughton in October 2001);
  2. that the word "caron" derives from the Russian word короной (or the related word some other Slavic language) meaning "crown" or "corona", as the accents sits on top of a letter like a little crown (suggested by Alexander Savenkov in December 2003).

The history of the term "caron" is equally obscure, and so far the furthest back that this character name can be taken is to the mid 1980s, when, according to a Unicode FAQ, the "caron" is referred to in publications such as Frank Romano's The TypEncyclopedia (1984) and IBM's "Green Book" (National Language Support Reference Manual, 1986).

I thought that I once read that someone had once offered a $100 reward for anyone who could establish the origins of the term, but I can no longer find any evidence for this possible delusion of mine. Anyway, although I'm far from being in a position to claim this hypothetical reward, with a little digging it is still possible to take the history of the Caron back quite a bit earlier than the mid 1980s.

But before looking in detail at the history of the name, let's take a look at the 45 characters in Unicode 5.1 that have the term CARON in their name :


Code PointCharacterCharacter NameUnicode 1.0 Name
02C7ˇCARONMODIFIER LETTER HACEK
030C ̌COMBINING CARONNON-SPACING HACEK
032C ̬COMBINING CARON BELOWNON-SPACING HACEK BELOW
010CČLATIN CAPITAL LETTER C WITH CARONLATIN CAPITAL LETTER C HACEK
010DčLATIN SMALL LETTER C WITH CARONLATIN SMALL LETTER C HACEK
010EĎLATIN CAPITAL LETTER D WITH CARONLATIN CAPITAL LETTER D HACEK
010FďLATIN SMALL LETTER D WITH CARONLATIN SMALL LETTER D HACEK
011AĚLATIN CAPITAL LETTER E WITH CARONLATIN CAPITAL LETTER E HACEK
011BěLATIN SMALL LETTER E WITH CARONLATIN SMALL LETTER E HACEK
013DĽLATIN CAPITAL LETTER L WITH CARONLATIN CAPITAL LETTER L HACEK
013EľLATIN SMALL LETTER L WITH CARONLATIN SMALL LETTER L HACEK
0147ŇLATIN CAPITAL LETTER N WITH CARONLATIN CAPITAL LETTER N HACEK
0148ňLATIN SMALL LETTER N WITH CARONLATIN SMALL LETTER N HACEK
0158ŘLATIN CAPITAL LETTER R WITH CARONLATIN CAPITAL LETTER R HACEK
0159řLATIN SMALL LETTER R WITH CARONLATIN SMALL LETTER R HACEK
0160ŠLATIN CAPITAL LETTER S WITH CARONLATIN CAPITAL LETTER S HACEK
0161šLATIN SMALL LETTER S WITH CARONLATIN SMALL LETTER S HACEK
0164ŤLATIN CAPITAL LETTER T WITH CARONLATIN CAPITAL LETTER T HACEK
0165ťLATIN SMALL LETTER T WITH CARONLATIN SMALL LETTER T HACEK
017DŽLATIN CAPITAL LETTER Z WITH CARONLATIN CAPITAL LETTER Z HACEK
017EžLATIN SMALL LETTER Z WITH CARONLATIN SMALL LETTER Z HACEK
01C4DŽLATIN CAPITAL LETTER DZ WITH CARONLATIN CAPITAL LETTER D Z HACEK
01C5DžLATIN CAPITAL LETTER D WITH SMALL LETTER Z WITH CARONLATIN LETTER CAPITAL D SMALL Z HACEK
01C6džLATIN SMALL LETTER DZ WITH CARONLATIN SMALL LETTER D Z HACEK
01CDǍLATIN CAPITAL LETTER A WITH CARONLATIN CAPITAL LETTER A HACEK
01CEǎLATIN SMALL LETTER A WITH CARONLATIN SMALL LETTER A HACEK
01CFǏLATIN CAPITAL LETTER I WITH CARONLATIN CAPITAL LETTER I HACEK
01D0ǐLATIN SMALL LETTER I WITH CARONLATIN SMALL LETTER I HACEK
01D1ǑLATIN CAPITAL LETTER O WITH CARONLATIN CAPITAL LETTER O HACEK
01D2ǒLATIN SMALL LETTER O WITH CARONLATIN SMALL LETTER O HACEK
01D3ǓLATIN CAPITAL LETTER U WITH CARONLATIN CAPITAL LETTER U HACEK
01D4ǔLATIN SMALL LETTER U WITH CARONLATIN SMALL LETTER U HACEK
01D9ǙLATIN CAPITAL LETTER U WITH DIAERESIS AND CARONLATIN CAPITAL LETTER U DIAERESIS HACEK
01DAǚLATIN SMALL LETTER U WITH DIAERESIS AND CARONLATIN SMALL LETTER U DIAERESIS HACEK
01E6ǦLATIN CAPITAL LETTER G WITH CARONLATIN CAPITAL LETTER G HACEK
01E7ǧLATIN SMALL LETTER G WITH CARONLATIN SMALL LETTER G HACEK
01E8ǨLATIN CAPITAL LETTER K WITH CARONLATIN CAPITAL LETTER K HACEK
01E9ǩLATIN SMALL LETTER K WITH CARONLATIN SMALL LETTER K HACEK
01EEǮLATIN CAPITAL LETTER EZH WITH CARONLATIN CAPITAL LETTER YOGH HACEK
01EFǯLATIN SMALL LETTER EZH WITH CARONLATIN SMALL LETTER YOGH HACEK
01F0ǰLATIN SMALL LETTER J WITH CARONLATIN SMALL LETTER J HACEK
021EȞLATIN CAPITAL LETTER H WITH CARON
021FȟLATIN SMALL LETTER H WITH CARON
1E66LATIN CAPITAL LETTER S WITH CARON AND DOT ABOVE
1E67LATIN SMALL LETTER S WITH CARON AND DOT ABOVE

As can be seen from the above table, in the first version of the Unicode Standard, published in October 1991, these characters were actually named using HACEK, but when Unicode and ISO/IEC 10646 merged the character names were changed to use CARON, in line with the character names given in the draft version of ISO/IEC 10646, and so from Unicode version 1.1 (published June 1993) onwards the characters have been named using CARON. So where did the term CARON come from ?

Well, the immediate inspiration for using the name CARON in ISO/IEC 10646 must have been earlier ISO character encoding standards such as ISO/IEC 6937 (first published in 1983) and ISO/IEC 8859-2 (first published in 1985 as ECMA-94 "Latin Alphabet No.2"), which both use the term CARON. It was quite natural for ISO/IEC 10646 to borrow the term CARON from these earlier standards as the first editor of 10646, Hugh McGregor Ross, was also involved in the development of character encoding standards such as ISO/IEC 6937. It was equally natural for the first edition of Unicode to use HACEK, as the Xerox Character Code Standard (May 1986) which was developed by Unicode's founding father, Joe Becker, has "317 Hachek accent = caron" (see the comment by MMcM to this Language Hat post).

The earliest coded character sets to use the term CARON that I have managed to find are the bibliographic-usage character sets DIN 31624 (July 1979) and ISO 5426 (1980), which both have "CARON (HÁČEK)" at position 4/15, which is the same position as the plainly named "CARON" in ISO 6937. It would thus seem that the original sources for the ISO use of the name CARON were character sets intended for bibliographic use. Such character sets needed the "háček mark so that, for example, the Library of Congress could correctly catalogue books such as Jaroslav Hašek's The Good Soldier Švejk, but why did they call it a "caron", and only give "háček" as a parenthetical comment? Looking back to earlier bibliographic standards turns out to be disappointing. During the late 1960s and early 1970s computer systems designed specifically for libraries began to support the háček diacritic, but it seems to have been almost universally referred to as either hacek or háček in these systems, with no mention of a "caron" :


It seems that this line of enquiry is a dead end. The authors of DIN 31624 must have got the name from some other non-bibliographic source, but where? A year earlier than the publication of the German bibliographic character standard in 1979, Stanley Rice's Book design—systematic aspects has a "caret" symbol and a "caron" symbol next to each other, and separate from a "circumflex" accent :


Stanley Rice, Book design—systematic aspects (1978) page 103


This suggests that Rice thought of the caron not as an inverted circumflex used as a diacritic mark, but as an inverted caret used as an editorial mark, which seems to be taking us down a different path. But unfortunately, at this point the trail seems to go cold, with no other sightings of the "caron" in the earlier years of the 1970s. Then suddenly in 1967 the caron turns up in a most authoritative source, the 1967 edition of the United States Government Printing Office Style Manual :


United States Government Printing Office Style Manual (1967 edition) page 180

[Click on image to see the whole page (and your starter for 10, what character or characters on this page and the next are not yet encoded in Unicode?)]


The US GPO Style Manual has been revised and reprinted many times since its first edition in 1894 (linking to the Internet Archive copy as it seems to me that 90% of pre-20th century books that used to be downloadable in the UK from Google Book Search are no longer downloadable outside of the US ... grrr), but none of the editions prior to 1967 include a character named "caron". For example, this is the same table of characters in the previous edition to the 1967 edition, published in 1959 :


United States Government Printing Office Style Manual (1959 edition) page 178


The layout of characters in the 1967 edition of the US Government Printing Office Style Manual matches that in Stanley Rice's 1978 book on book design, and must surely have been the immediate source for it. It seems highly probable that the GPO Style Manual was also the source for the name "caron" in the 1979 German bibliographic character standard (DIN 31624). I imagine that someone on the committee that developed DIN 31624 must have had a copy of the 1967 edition of the GPO Style Manual, and when it came to naming the háček character the owner of this prestigious volume must have objected to the native name of the character as it evidently conflicted with international usage, and so as to ensure compatibility with the ISO version of the standard (ISO 5426) that the DIN committee must have realised it would evolve into, the accent was named with its apparent English name, CARON, and its real name, HÁČEK, was merely given as an annotation. The rest is history.

But where did the authors of the 1967 edition of the GPO Style Manual get the idea of including an inverted caret shaped mark called a "caron" from? That remains a mystery, as I have been unable to find any suggestion of the use of the name "caron" for such a mark in any earlier book on typography. Frank Romano has stated that "caron" was the name for the Slovak diacritic given in the "giant books in the center of the order department" of the Mergenthaler Linotype Company in Broooklyn, but none of the Linotype specimen books and manuals that I have seen use the term caron (see below). Nevertheless, I think it is probable that the earliest use of the name "caron" predates and is unrelated to the development of multilingual or bibliographic-usage character encoding standards (in 1967, ASCII was only just revised to include lower case characters, and character encoding standards with "exotic" letters, symbols and accents were still a decade away).

Not only do we not know where the authors of the 1967 edition of the GPO Style Manual got the name "caron" from, we also cannot be sure what exactly the mark labelled "caron" was intended to represent. In the manual the "caron" is paired with the caret mark, which is distinct from the similar-shaped circumflex accent, so it may well be that the "caron" was not intended to represent the háček diacritical mark at all, but was rather intended to represent an inverted caret used for proofreading purposes. If this was the case, as I believe, then the introduction of the name "caron" into DIN 31624, and thence, spreading like a virus, to sundry other ISO character encoding standards, and ultimately infiltrating its way into the Unicode standard, was a mistake.

The inverted caret mark used in proofreading is shown in the table of proofreading marks in the 1967 edition of the GPO Style Manual :


United States Government Printing Office Style Manual (1967 edition) page 4

[Click on image to see the whole page]


And books on typography, dating back into the 19th century explicitly name this mark an "inverted caret" :


Theodore Low de Vinne, Correct Composition: A Treatise of Spelling, Abbreviations, the Compounding and Division of Words, the Proper Use of Figures and Numerals, Italic and Capital Letters, Notes, etc., with Observations on Punctuation and Proof-Reading (New York, 1901) page 322


Even some Linotype specimen books have an "inverted caret", which is paired with the "caret", and does not appear to be the same thing as the háček diacritical mark :


Specimen Book Linotype Faces (Mergenthaler Linotype Company, 1939) page 892


But the question remains, when did the "inverted caret" become the "caron" ? Did the editors of the 1967 GPO Style Manual devise the name "caron" themselves on a whim, or did they borrow it from some other source ? (Answers on a postcard, please.)


Saturday, 8 August 2009

The Lost Game of Liubo Part 3 : Pictures of Immortals playing Liubo



One intriguing motif that is found mainly on engraved stone coffins from the area of modern Sichuan province is that of two elven "immortals" (仙人) or "feathered men" (羽人) engaged in a game of Liubo.


Source : Sichuan Handai Huaxiang Xuanji 四川漢代畫象選集 (Shanghai, 1955) fig.35


These immortals are very different to later depictions of Daoist immortals : both figures appear to be naked, with thin, sinuous bodies, wispy wings sprouting from their shoulders, and two distinctive protrusions on their heads. These protrusions have been variously described as horns or topknots, but I believe that they are in fact enormous ears, as can be seen from this painting of a winged immortal riding a dragon from a late Western Han tomb mural that was discovered in the ancient capital city of Chang'an in 2004 :


Source : Wen Wu 文物 2006.5 page 26

Is this evidence of alien visitation to ancient China ?

Or do the long nose, moustache and generally non-Chinese features indicate that the Chinese believed that the Indo-European peoples to the west of China were a race of supernatural beings ?


In many cases the two immortals playing Liubo are on the edge of a larger scene centred on the Queen Mother of the West (Xiwangmu 西王母) and her retinue, including the three-legged sun crow (三足烏), the moon toad (蟾蜍), the hare who pounds the elixir of immortality (and who in later mythology displaces the toad as the main denizen of the moon), and the nine-tailed fox (九尾狐狸).


Source : Mysteries of Ancient China (London, 1996) figure 101.1

[Queen Mother of the West on her Dragon and Tiger throne, together with a three-legged crow, a dancing toad, a hare holding sacred fungus, and a winged nine-tailed fox]


The cult of the Queen Mother of the West was widespread during the Han dynasty, and she is the most commonly found divinity on Han picture stones, picture bricks and engraved stone coffins. I believe that the images of two immortals playing Liubo in scenes associated with the Queen Mother of the West are not just random pictures of any old immortals who happen to be playing a game of Liubo, but are two specific immortals, and the game of Liubo that they are playing reflects a specific story from the mythology surrounding the Queen Mother of the West. The images of these two fairy-like Liubo players are largely restricted to the region of modern Sichuan province, and so it is likely that they reflect the local mythology of the ancient Ba 巴 and Shu 蜀 cultures. Although the the story behind this game of Liubo is now lost, I think that we can at least identify who the two players are.

The key to unlocking the mystery of who these two Liubo players are is an unusual design of bronze pictorial mirror (畫像鏡) made exclusively by a certain Master Bo 柏師 of the Wu 吳 region of eastern China (see 5.1 and 5.2). This type of mirror depicts various mythological scenes and fabulous creatures, including two feathered immortals playing Liubo :


Source : 中秋到了,发一面;富贵寿宜神兽镜


The Liubo players are not identified directly, but on either side of the players stands a horse with its head in a trough facing the player. The horse next to one player is labelled 王喬馬 wáng qiáo mǎ "Wang Qiao's horse" and the horse next to the other player is labelled 赤誦馬 chì sòng mǎ "Chi Song's horse", which strongly suggests that the two players must be Wang Qiao and Chi Song, who were (or should I say "are") two famous immortals.

Wang Qiao 王喬 (also called Wang Ziqiao 王子喬) is variously recorded as originally being the son of King Ling of Zhou 周靈王 (reigned 571–545); the county magistrate of Ye County 葉縣 in Hedong 河東 (in modern Shanxi province) during the reign of Emperor Ming of Han (reigned 58–75); a man from the ancient State of Yue in the south of China; or a man from Wuyang 武陽 in the State of Shu 蜀 (modern Peng County 彭縣 in Sichuan). According to various sources each of these four attained immortality, and later became revered as a Daoist immortal. It is likely that in fact they all refer to the same immmortal, and are merely later attempts to provide a mortal history for an existing immortal.

Chi Song 赤誦 is an alternative form of the name of Chi Song 赤松 or Chisongzi 赤松子 meaning "Red Pine", who according to legend was the Master of Rain (雨師) during the reign of the legendary emperor Shennong 神農. He was able to set fire to himself without harm, and frequently visited the Queen Mother of the West in her stone dwelling on Mount Kunlun 崑崙山.

Wang Qiao and Chi Song are very often mentioned together, as in this quote from the 2nd century BCE Huainanzi, which is perhaps the earliest surviving account of either of these two immortals :


王喬、赤松,去塵埃之間,離群慝之紛,吸陰陽之和,食天地之精,呼而出故,吸而入新,虛輕舉,乘雲遊霧,可謂養性矣,而未可謂孝子也。《淮南子・泰族訓》

Wang Qiao and Chi Song turned their back on the dusty world and left behind the entanglements of multitudinous sins. They breathed in the harmony of the yin and the yang, and consumed the essence of heaven and earth. When they breathed out they expelled the old, and when they breathed in they absorbed the new. Empty and light they lift up, and ride on the clouds and travel with the mist. This could indeed be called nourishing one's nature, but it could hardly be called the acts of a filial son. Huainanzi chapter 20 (Grand Reunion)


Not only are these two immortals often paired together, but they are both mostly associated with southern China and Chu culture. The poem Yuǎnyóu 遠遊 "Far-off Journey" in the Chu Ci 楚辭 anthology features Wang Qiao and Chi Song prominently :


漠虛靜以恬愉兮,澹無為而自得。
聞赤松之清塵兮,願承風乎遺則。
...
春秋忽其不淹兮,奚久留此故居。
軒轅不可攀援兮,吾將從王喬而娛戲。
...
順凱風以從游兮,至南巢而一息。
見王子而宿之兮,審一氣之和德。
曰:「道可受兮,不可傳;
其小無內兮,其大夫垠;
毋滑而魂兮,彼將自然;
一氣孔神兮,於中夜存;
虛以待之存,無以為先;
庶類以成兮,此德之門。」
聞至貴而遂徂兮,忽乎吾將行。
仍羽人於丹丘,留不死之舊鄉。

《楚辭・遠遊》


In emptiness and silence I found serenity;
In tranquil inaction I gained true satisfaction.
I heard how once Red Pine had washed the world's dust off:
I would model myself on the pattern he had left me.
...
Spring and autumn hurried by, never delaying:
I could not go on staying in my old home for ever.
Xuan Yuan was too remote for me to aspire to;
But I could follow Wang Qiao for my delight.
...
Drifting in the wake of the gentle south wind,
I traveled to Nan-chao in a single journey.
There I saw Master Wang and made him salutation,
And asked about the balance made by unifying essence.
He said:
‘The Way can only be received, it cannot be given.
Small, it has no content; great, it has no bounds.
Keep you soul from confusion, and it will come naturally.
By unifying essence, strengthen the spirit; preserve it inside you in the midnight hour.
Await it in emptiness, before even Inaction.
All other things proceed from this: this is the Door of Power.’

Having heard this precious teaching, I departed,
And swiftly prepared to start off on my journey.
I met the Feathered Men on the Hill of Cinnabar;
I tarried in the ancient land of Immortality.

David Hawkes, The Songs of the South (Penguin Books, 1985) pages 194–196.

So it is my belief that Wang Qiao and Chi Song (Red Pine) are the two immortals shown playing Liubo on Master Bo's mirrors (but why the horses? I have found no source that refers to them as horse-riders, and indeed they are especially known as riders of clouds and mist); and that they are also the same two immortals depicted on all the Han dynasty stone coffins and picture bricks/stones from the Sichuan region. But why they are depicted playing Liubo, and why this depiction of immortals playing Liubo is only found in the Sichuan region, remains somewhat of a mystery. My suspicion is that the later motif of two immortals playing the game of Weiqi (Go) in the mountains was derived from some now lost story of Wang Qiao and Red Pine playing Liubo on Mount Kunlun : when Liubo lost its popularity after the end of the Han dynasty, the two immortals simply changed their game from Liubo to Weiqi.



4. Picture Stones and Picture Bricks

4B. Scenes showing Fairy Liubo Players

4B.1 Engraving on an Eastern Han stone coffin from Sichuan

四川郫縣新勝鄉出土東漢畫像石棺

Source : Zhongguo Meishu Quanji 中國美術全集・繪畫編 (Shanghai, 1988) vol.18 plate 91


Provenance : Xinsheng, Pi County, Sichuan. {30.895°N 103.765°E}

Current Location : Sichuan Provincial Museum (四川省博物館藏)

Date : Eastern Han (25 CE – 220 CE)

Size : 89.0 × 225.0 cm.

Notes : Engraving on the side of a stone coffin. The Queen Mother of the West, on her dragon and tiger throne, is in the centre of the picture, with the familiar toad, hare, nine-tailed fox and three-legged crow on her right side, and two fairy Liubo players on a mountain (presumably Kunlun mountain 崑崙山) to her left. The throwing mat is between them, and the liubo board closer to the foreground, but rather indistinct.



4B.2 Engraving on an Eastern Han stone coffin from Sichuan

四川新津宝子山崖墓出土東漢畫像石

Source : Sichuan Handai Diaosu Yishu 四川漢代彫塑藝術 (Beijing, 1959) fig.47

[also Sichuan Handai Huaxiang Xuanji 四川漢代畫象選集 (Shanghai, 1955) fig.35]

[also Zhongguo Gudaishi Cankao Tulu: Qian-Han Shiqi 中國古代史參考圖錄・秦漢時期 (Shanghai, 1990) page 233]

[also Zhongguo Diaosushi Tulu 中国雕塑史图录 (Shanghai, 1983) vol.1 page 273]


Provenance : Baozi Shan, Xinjin, Sichuan. {30.402°N 103.818°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : 57.0 × 99.0 cm.

Notes : End of a stone coffin. The two fairy players appear to be naked, and have wispy wings and enormous ears. The throwing mat is placed between them, with a low Liubo table closer to the foreground, and a large wine bowl with a ladle on the other side of the throwing mat. Behind one of the fairy players grows the sacred Lingzhi fungus, and a bird flies overhead. The player on the right seems to have just thrown the six throwing sticks, which lie in a row on the mat, whilst the player on the left is raising up his hands as if questioning the result of the throw.



4B.3 Engraving on an Eastern Han stone coffin from Sichuan

四川簡陽縣深洞村鬼頭山東漢崖墓3號石棺左壁畫像

Source : Wen Wu 文物 1991.3


Provenance : Guitou Mountain, Shendong Village, Jianyang, Sichuan. {30.232°N 104.310°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : Left side of an engraved stone coffin. The two figures on the right appear to be playing Liubo, although it is not very clear. Unlike the fairy players seen on other coffins, these two figures are wearing long feather head-dresses, and the one on the left appears to have fur or spines down his back.



4B.4 Engraving on an Eastern Han stone coffin from Sichuan

四川新津縣堡字山出土畫像石棺

Source : Sichuan Handai Huaxiang Xuanji 四川漢代畫象選集 (Shanghai, 1955) fig.26

[also Zhongguo Gudaishi Cankao Tulu: Qian-Han Shiqi 中國古代史參考圖錄・秦漢時期 (Shanghai, 1990) page 233]


Provenance : Baozi Shan, Xinjin, Sichuan. {30.402°N 103.818°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : A large bird on the left side of the players recollects the bird flying over the players in Picture 4B.2. The players are shaded by what appears to be peacock (or phoenix) feathers, but maybe it is meant to be sacred fungus.



4B.5 Engraving on an Eastern Han stone coffin from Sichuan

四川宜賓縣公子山東漢崖墓出土畫像石棺

Source : Wen Wu 文物 1982.7


Provenance : Gongzishan, Yibin, Sichuan. {28.698°N 104.527°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.



4B.6 Eastern Han picture brick from Sichuan

四川新都縣出土漢代畫像磚

Source : Sichuan Handai Huaxiangzhuan 四川漢代畫像磚 (Shanghai, 1987) fig.35


Provenance : Xindu, Sichuan. {30.825°N 104.155°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : Are those tails? And what is the tiny figure above the player on the right?



4B.7 Eastern Han picture brick from Sichuan

四川廣漢縣出土漢代畫像磚

Source : Sichuan Handai Huaxiangzhuan 四川漢代畫像磚 (Shanghai, 1987) fig.115


Provenance : Guanghan, Sichuan. {31.00°N 104.30°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : One of very few pictures of immortals playing Liubo that show an observer, in this case with an outstretched arm as if he is joining in the game.



4B.8 Eastern Han picture brick from Sichuan

四川德陽縣出土漢代畫像磚

Source : Sichuan Handai Huaxiangzhuan 四川漢代畫像磚 (Shanghai, 1987) fig.116


Provenance : Deyang, Sichuan. {31.13°N 104.40°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : Sacred fungus in front of the player on the right, and above the player on the left.



4B.9 Eastern Han picture brick from Sichuan

四川出土漢代畫像磚

Source : 中国画像石网


Provenance : Unknown.

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.



4B.10 Engraving on an Eastern Han stone watch tower from Sichuan

四川渠縣土溪鄉趙家村貳無銘東漢闕

Source : Sichuan Handai Shique 四川漢代石闕 (Beijing, 1992) fig.219


Provenance : Zhaojia Village, Tuxi, Qu County, Sichuan. {31.023°N 107.040°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : Detail from the back of the main building showing two figures playing Liubo. It is not clear whether they are mortal or immortal players, but if they are immortals their clothing and head-dresses are unusual.



4B.11 Engraving on an Eastern Han stone coffin from Yunnan

雲南昭通省耕塘西面梁子墓出土東漢畫像石

Source : Wen Wu 文物 1960.6 page 49


Provenance : Shenggengtang, Zhaotong, Yunnan. {27.355°N 103.725°E}

Current Location : Zhaotong County Hall of Cultural Relics (昭通縣文物館)

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : Decoration on a stone coffin, also depicting the Queen Mother of the West, a nine-tailed fox and the three-legged sun crow.



4B.12 Eastern Han picture brick from Henan

河南漢代畫像磚

Source : Henan Handai Huaxiangzhuan 河南漢代畫像磚 (Shanghai, 1985) fig.240


Provenance : Henan.

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : This is one of only a very few examples of a depiction of fairy Liubo players from outside of Sichuan.



5. Pictorial Mirrors

5.1 Bronze picture mirror from Zhejiang

浙江紹興灕渚出土東漢神獸帶鏡

Source : Zhejiang Chutu Tongjing 浙江出土銅鏡 (Beijing, 1987) plate 24


Provenance : Lizhu, Shaoxing, Zhejiang. {29.955°N 120.475°E}

Current Location : Unknown.

Date : Eastern Han (25 CE – 220 CE)

Size : Unknown.

Notes : This is one of several examples of a pictorial mirror with the same design made by a certain Master Bo 柏師, showing a pair of fairy Liubo players. The mirror has three concentric rings of decoration. The inner ring has a typical mirror inscription reading "wealth, honour and long life, bringing benefit to your children and grandchildren—very auspicious" 富貴長壽宜子孫大吉. The middle ring is separated into seven panels :

  • at 12 o'clock: a horse facing the left player, labelled 王喬馬 wáng qiáo mǎ "Wang Qiao's horse";
  • at 10 o'clock: the two fairy Liubo players;
  • at 8 o'clock: a horse facing the right player, labelled 赤誦馬 chì sòng mǎ "Chi Song's horse";
  • at 7 o'clock: a chained tiger-like creature, with the chain attached to a large hoop;
  • at 5 o'clock: a chained humanoid creature, with the chain attached to a post labelled 銅柱 tóng zhù "bronze pillar";
  • at 3 o'clock: a monstrous creature with a long neck and tail, labelled 辟耶(=邪) bì xié "ward off evil", the name of a mythical creature that wards off evil, later known as a Pixiu (貔貅 pí xiū)
  • at 1 o'clock: another monstrous creature with a long neck and tail, labelled 柏𨸲(師)作 bó shī zuò "made by Master Bo"

The outer ring is decorated with images of various animals and fabulous creatures, including a phoenix and the three-legged sun crow adjacent to the Liubo-players, and a feathered immortal holding on to the tail of a dragon.

Apparently other examples of this design are given in Zhongguo Tongjing Tudian 中國銅鏡圖典 (Beijing, 1992) and Zhongguo Gutongjing Jianshang Tulu 中國古銅鏡鑒賞圖錄 (Beijing, 2002).



5.2 Bronze picture mirror

東漢神獸帶鏡

Source : 中秋到了,发一面;富贵寿宜神兽镜


Provenance : Unknown.

Current Location : In a private collection.

Date : Eastern Han (25 CE – 220 CE)

Size : 22.2 cm. in diameter.

Notes : A very similar, but not identical design to 5.1. The central inscription is one character shorter (富貴壽宜子孫大吉), and runs in the opposite direction. The middle ring has the same seven panels and exactly the same labels, but the images are mirrored compared with 5.1. The outer ring is blank.