Mahjong(g), before and after Mahjong(g): Part 2

Michael Stanwick

In Part 1 (see Stanwick, 2006) I proposed that the Wilkinson/Laufer Pattern survived to at least circa 1923. In this concluding article I introduce reconstructions, from new archival documentation, describing a new tile set pattern with its variations which can be dated to before 19141. I will compare these sets with a late 19th century tile set, its description dating from 1934, uncovered by Andrew Lo (2004), and cited here in English from the original description in Chinese. From the limited material, I will make tentative propositions for further discussion and research. Finally, I will further propose that one modified variation of the basic 1915 Pattern persisted during the 1920’s, whilst another may have influenced a variation of the1909 Culin pattern to form a hybrid pattern that was exported to the West by J. P. Babcock.

  • [1]

    I am indebted to Thierry Depaulis for providing me with this important documentation and for some of its translation and comments.

4. The Mauger sets' patterns and their reconstructions

4.1 Document and Description.

Captain George E. Mauger was born in 1867 in St Helier on the British Island of Jersey. From 1912 to 1918 he resided in Paris and was a member of the Société d’Anthropologie de Paris. He died in Jersey in 1918. On the 18th November 1915 he gave a talk on Chinese games to the Paris Anthropological Society and this talk subsequently appeared as an article in the Society’s Bulletin in 1917 (see Mauger, 1915).

Toward the end of the article Mauger introduces a Chinese game called “Ma-Tchio-Pai”.

“As for the domino cards we also have for the latter [that is, money-suited cards]2 a form with bamboo backs with the Chung fa p’ai of which there are specimens in the British Museum and in Pennsylvania [University Museum] and from where derives the great Chinese current game of chance which is played everywhere in Peking, Shanghai, Suchong [Suzhou], Hankow [now Wuhan], in America and in Europe by all the educated Chinese classes who want entertainment of the great games of chance. I had great fortunes to be able to obtain explanations on this game which is called Ma-Tchio-Pai [ma qiao pai].”

Following this text are three, large Chinese characters for 麻 ma 雀 qiao 牌 pai3 , literally ‘hemp bird’ tiles, but in the bonded form meaning ‘sparrow’ tiles. This excerpt throws up some important observations. Firstly, we are informed of the 中‘chung’ [zhong] and 發 ‘fa’ tiles found in the 1889 Wilkinson sets, although Mauger seems unaware of the 1901 Laufer and 1909 Culin sets. Secondly, it appears that Mauger has referenced Wilkinson (1895), and Wilkinson in O’Donoghue (1901) but is unaware of Culin’s 1909 expedition to China4. Thirdly, he provides a list of geographical locations where the game is played and also who is playing the game in 19155. This is interesting as it shows that the game was played (albeit by the educated Chinese classes) in areas of North and Central China (no mention of South China) – and in the USA, five or six years before Babcock introduced it in 19207.

Mauger then states:

“I will use the terminology which is used in the native province of the friend who explained it to me.”

According to Thierry Depaulis, from the spelling of the Chinese words that Mauger uses it can be inferred that his ‘Chinese friend’ is using Mandarin Chinese. However, some details in the spelling (and therefore in the actual pronunciation) would lead to the Hankow (Hankou, now called Wuhan, in central China) style of Mandarin. Mauger’s Chinese friend(s) may therefore have come from Hankow [Wuhan] in Central China8. But the sets described may not necessarily be from there. Mauger continues (with some grammatical alteration):

“The game consists of one hundred and thirty six pieces with the different symbols painted in different colours for each series on the bone face which has a bone dovetail crimped into the back of the bamboo. There are three series of nine repeated four times: Thus Tungtze 銅字 [Tong zi] [‘copper’ = ‘Cash’ ~ ‘Circles’] or Pingtze 餅字 [Bing zi] or cakes9Tchantze 簽字 [qian? zi] [slip of bamboo?10] or Saultze 索字 [Suo zi] or cords [‘Strings of Cash’ or ‘Bamboos’]; Wanza (za should be zi) 万字 [Wan zi] myriads [‘Myriads of Cash’ or ‘Characters’] represented by numbers. The aces are represented by special drawings. A very decorative circle for the pingtze [bing zi] [‘cakes’ ~ ‘Cash’], a bird for the Tchantze [qian zi? ~ suo zi] [‘slip of bamboo?’ ~ ‘Strings of Cash’] and the quantifier for the myriads [this is explained below].”

Mauger provides diagrams11 for the 5 of ‘Cash’ or ‘Circles’, the 3 of ‘Strings of Cash’ or ‘Bamboos’ and a bird (a darting Sparrow) for the 1 of ‘Strings of Cash’ or ‘Bamboos’. A diagram is also provided of the 6 of ‘Myriads of Cash’ or ‘Characters’. However, the character 万 wan (found in wan zi) meaning ‘Myriads’, has been replaced by the character 品 pin meaning ‘rank’ or ‘grade’. There is no mention of ‘Flowers’ or ‘Seasons’ tiles or spare blanks so a tile set has 136 pieces. Further, the darting ‘Sparrow’ on the # 1 ‘String of Cash’ is very similar to those depicted in the circa 1909 Culin and Nagawa tile sets, which may suggest this symbol was becoming a dominant feature around this time. The # 1 ‘Cash’ is a “very decorative circle” which is also consistent with those found in both the Culin and Nagawa tile sets. Mauger continues:

“Moreover there are a series of special premiums and several general premiums. There are four special premiums. Sometimes they are represented by symbols representing the four directions 東 east, 南 south, 北 north, 西 west; other times by the titles: 公 duke, 侯 marquis, 將 general or intendant, [and] a 相 minister12. There are four specimens of each one…” [The general premiums are]: Tsung 中 [zhong] or center; 龍 Long or Dragon; 發 Fa [or] prosperity; Fong 鳳 [fèng] [or] Phoenix. White pae 白 [bai pai].”

This describes the appearance of alternative characters, the four ranks of nobility for the four Directions, and the characters 龍 long (Dragon) and 鳳 feng (Phoenix) for two of the ‘Honours’ characters zhong and fa (see Stanwick Part 1, 2004)13. Diagrams are also provided to illustrate the Chinese characters long (Dragon), jiang (General) and fèng (Phoenix).

In his report on a contemporary game called 叉 cha 麻 ma 雀 que (gambling at ma que or sparrow)14Xu Ke (1917) lists the three ‘Honours’ as long (‘Dragon’), fèng (‘Phoenix’), and bai (‘White’) (no mention of the character pin however), but also the alternative characters zhong (‘Centre’ or ‘Middle’), fa (‘Prosperity’) and bai (‘White’). Further, Xu specifically mentions that a tile set was composed of 136 tablets or 牌 pai. This description, as far as it goes, is consistent with Mauger’s. Again, this partial description may not necessarily reflect a set from Central China (Shanghai).

Whether this Mauger tile set and its variations can be classed as precursors to Mahjong(g) is not clear in my view15. The absence of the two quartets, commonly called the ‘Flowers’ and ‘Seasons’, and the four spare blanks would indicate that these should be classed as precursors. However, the presence of the bird on the # 1 ‘String of Cash’ and the ornate # 1 ‘Cash’ is consistent with the 1909 Culin set and in keeping with that set, the ‘Flowers’ and ‘Seasons’ may have been discarded, along with any spare blank tiles, as irrelevant to the local Chinese game or they were simply not bought with the set.

Joseph Park Babcock (1923a) claimed to have spent a great part of his time travelling in the interior of China between 1913 and 1923, witnessing different game-plays and encountering many tile sets16. If this is an accurate account then it is conceivable that he saw at least two forms of tile set patterns – as exemplified by the Culin (1909), Mauger (pre 1915) and Xu Ke (circa 1917) tile sets. Based on this limited data one may hypothesise that, prior to 1920, a form of the 1909 Culin pattern (see Stanwick, 2004b) may have undergone some slight modification via Babcock, through the inclusion of the alternative green ‘Phoenix’ and red ‘Dragon’ characters (used in the ‘Honours’ group in the Mauger pattern) instead of the green ‘Prosperity’ and the red ‘Middle’ or ‘Centre’ characters. Alternatively, one may hypothesise that this pattern may have already been in existence during Babcock’s tenure in China and was merely appropriated by him for his export sets. An example of the basic pattern appears in the 1922 ‘Mah-Jongg Sales Company of America’17 catalogue as ‘Style 3’, packed in a black, hardwood box.


Figure 4.

4.2 Tile Set Reconstruction

Based on Mauger’s description given by his Chinese friends and the diagrams provided in his article, Figures 5 represents the basic pattern18 of the tile set – firstly, the requisite three suits, but with new characters to represent the ‘Myriads of Cash’ suit. Thus, on the # 1 of ‘Myriads of Cash’ the character for ‘ten thousand’ is a swastika 卐 wan, in the clockwise formation19 and for the rest of the suit the character 品 pin is used, in this context meaning ‘rank’ or ‘grade’20. Secondly, there is the combination of the ‘Directions’ group – containing the four ‘Directions’ characters, plus the ‘Honours’ group – containing the ‘White’, ‘Centre’ and ‘Prosperity’ characters. Two other possible combinations of these latter two groups, that also include the alternative characters described above, are shown in Figures 6 and 7. The fourth combination is shown in Figure 4 above. Further, based on the illustration of the 5 of ‘Cash’, the circles of the coins in the ‘Cash’ suit are extrapolated to have the distribution of [22c – 92c]21. On a similar basis, the ‘Strings of Cash’ suit is composed of a darting Sparrow for the # 1 ‘String of Cash’ and rods or bars for the # 2 – # 9 ‘String of Cash’.


Figure 5.


Figure 6.


Figure 7.

But prior to 1915, what is the earliest documented appearance of the pin character in possible precursor tile sets? To answer this question we can turn to another piece of documentation.

  • [2]

    Additional comments appear in square parentheses.

  • [3]

    Throughout this article I have used pinyin romanization of Chinese. Where other romanizations are used, the pinyin form is in parentheses.

  • [4]

    This view was arrived at by Thierry Depaulis after taking into account the relevant references in the entire article.

  • [5]

    Culin (1924) provides support for this observation. He was told by his Chinese informant Dzau Sing Chung of Shanghai that the game used to be played by rich people, but was at that time extending to all classes. Perhaps Mauger’s Chinese friend(s) were not aware that it was becoming more popular amongst other classes of Chinese Society.

  • [6]

    Culin (1924) noted that in 1909 he could find no tile sets for sale in Canton [Guangzhou in Guandong province] in South China, even though his Chinese informant said the game was beginning to be played there.

  • [7]

    However, there is a reference to a Ge Kunhua who left China in 1879, after working for the British Consulate in Ningbo, to take up a professorial teaching post at Harvard University in the United States. He is said to have introduced ma que to the American scholars there and the game became popular throughout the teaching staff at the University (Yu Haoxu, 2002).

  • [8]

    This is supported by the fact that there was a French concession in Hankow [now Wuhan] from 1896 to 1943. Hankow was a major treaty port on the Yangtze river, opening in 1862 (Wikipedia).

  • [9]

    Wilkinson (1895) states that the suit of 餅 ‘ping’ [bing], ‘cakes,’ was originally and properly that of 錢 ‘ch’ien’ [qian] meaning ‘money’ or ‘sapeks’ or ‘cash’. A ‘sapek’ is a round (as a rice-cake) copper or brass coin with a square hole in the middle, through which many coins could be strung together, hence ‘Strings of Cash’.

  • [10]

    If ‘Tchantze’ is indeed, in pinyin, ‘qian zi’ meaning a slip of bamboo or slips of bamboo used for gambling, then it is the first mention of the use of the term ‘bamboo’ in connection with the ‘Strings of Cash’. This is not altogether incongruous as bamboo discs were perhaps sometimes used for money (Eberhard, 2003).

  • [11]

    The drawings are initialled by Edouard Cuyer, past president of the Société d’Anthropologie de Paris.

  • [12]

    The Japanese Mahjong Museum own at least two ‘Duke, Marquis, General, Minister’ tile sets that also have the character 品 pin instead of 万 wan in the ‘Myriads of Cash’ suit. (See Noguchi 2005).

  • [13]

    When portrayed together, the ‘Dragon’ and the ‘Phoenix’ represent the Emperor and Empress, respectively (Eberhard, 2003). The four ranks of nobility represent positions in the Chinese Court (Ebashi 2005). An extension of this hierarchy may be reflected in the ‘Myriads of Cash’ suit with the character 品 pin, meaning ‘rank’ or ‘grade’.

  • [14]

    This name could be synonymous with ‘cho ma-cho’ reported by Culin (1924) as the name for the game in 1909. See Stanwick 2004, No. 5, footnote 6.

  • [15]

    For what I have stipulated to be a precursor tile set, see Stanwick 2004a.

  • [16]

    Joseph Park Babcock was hired as a civil engineer by the Standard Oil Company in 1912 and sent to Soochow [Suzhou], China (Sloper, 2006). During a ten year period, he spent a great part of his time travelling in the interior of China learning about the game of ma que (The Saturday Evening Post, 1923), and started exporting tile sets sometime in 1920. Included with these sets was a copy of his manual, ‘Rules for Mah-Jongg’, dated 1920.

  • [17]

    The tile sets were imported into the US by W. A. Hammond. Hammond subsequently created the Mah-Jongg Sales Company of America in the early 1920’s. Hammond, Babcock and Albert R. Hager were partners in the Company.

  • [18]

    The colours are based on those of the 1909 Culin set as is the decorative # 1 ‘Cash’. The form of # 8 ‘String of Cash’ was not described or drawn so its representation here is a best guess. It could equally have been identical to its namesake in the Culin set.

  • [19]

    According to Eberhard (2003), the swastika is one of the oldest symbols in China (and India). It was a very old form of the character 方 fang, meaning ‘square’ or the ‘four regions’ of the world. From AD 700 it has been used to mean ‘ten thousand’, symbolising ‘infinity’, hence its use here.

  • [20]

    [20] See footnote 13. This character has been found in tile sets produced in China during the early 1920’s, for example, in a tile set produced for the famous Chinese female impersonator of Beijing (North China) drama, Mei Lan Fang (see Noguchi, 2005).

  • [21]

    For example, if each coin on the 2 [2] of ‘Cash’ is composed of 2 circles [2c], then the distribution is [22c].

5. The tile of set of Sheng Xuanhuai

5.1 Text Description and Discussion.

A description of a hitherto unknown earlier tile set of 140 pieces, owned by the Chinese official Sheng Xuanhuai (1844 – 1916), was uncovered and introduced by Andrew Lo (2004). This description originally appeared in Dai Yu’an, Gu shui jiu wen [Old Hearings from the Gu River], with a preface dated 1934. This account has some warrant since the author’s late, elder fourth brother, was Sheng’s English interpreter. According to the description22:

“When Sheng Xuanhuai came from Baiyue (an area near Canton, Guandong province in South China) to Tianjin, to take his post of customs commissioner, he brought with him a 麻雀牌副 ma que pai fu [ma que tile set], so introducing it [ma que pai] to Tianjin. The set is different from the current ones.”

Note that this excerpt claims Sheng came to Tianjin [Hebei province in North China] from Baiyue (Guandong province in South China) – but when? Based on Feuerwerker (1970), there were two instances when he had held the post of Tientsin Customs Commissioner. The first began in the summer of 1884, after he had left Shanghai in April of that year. The second began in 1892, after a six year posting in Cheefoo [now called Yantai, in Shandong province23, North China.].

Since he already had the tile set when he arrived in Tianjin, he may have acquired it in 1884 – perhaps after inspecting the construction of the South China coastal telegraph line in Guandong province24 – and anywhere from South China to North China, during his journey back to Tianjin. Alternatively, he may have acquired it in North China (contrary to that implied in the excerpt above), during his six year posting in Cheefoo [Yantai] between 1886 and 1892. The description continues:

“Apart from the 1-9 of “Strings” [Bamboos] and 1-9 “Cakes” [Circles], there are no Myriads 万 ‘wan’’, but 1-9 Ranks 品 ‘pin25. There are no East, South, West and North “Winds”, but instead four “Winds” [=Virtues]26 – benevolent, just, righteous, and virtuous.”

It is clear this circa 1888 set shares certain features with the pre-1914 Mauger tile set. Both replace the ‘Directions’ characters with alternative characters – either with the characters for ‘benevolent’, ‘just’, ‘righteous’ and ‘virtuous’ as in the Sheng Xuanhuai set, or with the characters for ‘Duke’, ‘Marquis’, ‘General’ and ‘Minister’ as in the Mauger set. Further, in the ‘Myriads of Cash’ suit, both sets have replaced the wan character with the character 品 pin27. As far as I am aware the Sheng Xuanhuai set is the earliest documented appearance of the pin character. Both sets are also missing the four ‘Seasons’ tiles, present in earlier tile sets (see Stanwick Part 1, 2004).

The four ‘Virtues’ I propose, are the four Confucian cardinal virtues of integrity, propriety, righteousness and modesty, respectively (see Eberhard 2003). If Sheng had this set made specifically for his own use, then the inclusion of these four characters would be consistent with, and be a reflection of, his status as a Confucian official (see Feuerwerker, 1970).

“There is a white tablet; there are neither “Red Success” [zhong ie, Red Middle] nor “Fortune” [fa], but scarlet 龍 Long [Dragon] and azure 鳳 feng [Phoenix].

This is also the earliest known documented appearance of the ‘Dragon’ and ‘Phoenix’ characters, and the presence of both characters is in accord with the 1915 Mauger set28 and the observation of Xu Ke (1917). However, the other contemporaneous set, collected in Ningbo in central China, by W. H. Wilkinson in 1889, does contain the two ‘Honours’ zhong (middle) and fa (occur/prosperity). Further:

“Apart from these there is a 万王 Wan Wang [King of ‘Myriads of Cash’. It is assumed to use the character for wan, rather than the character for pin] belonging to the “Rank” [pin] suit, a 索王 Suo Wang [King of ‘Strings of Cash’] for the “Strings” and a Bing Wang [King of ‘Cakes’] for the “Cakes” [‘Circles’ ~ ‘Cash’].”

It may appear strange that the King of the ‘Myriads of Cash’ suit is called wan wang and not pin wang, a King of the ‘Ranks’ suit, but this name is probably the original name of the suit being carried over, rather than a reflection of the name of the characters appearing on the tiles. Following this extract there follows a brief mention of the function of these tiles, in which a further wang tile is named – a “Supreme King” 摠王 [zong wang] “which is worth more than the other three”.

Notably, three of the wang tiles share a similarity with the three 化 hua tiles from the Himly set, collected between 1868 and 187629, that is, each is named for its respective suit. But in contrast, the fourth tile, zong wang, “Supreme King”, is identical to the zong wang “Almighty King” in the Himly set30. However, the Sheng Xuanhuai tile set is missing the ‘Seasons’ quartet present in both the Himly and Glover precursor sets. It is classed as a precursor since it contains the four extra ‘King’ tiles and is missing the four ‘Seasons’ and four ‘Flowers’ tiles31.

If the Sheng Xuanhuai set derives from North China, then the presence of the wang tiles in it may point to an influence from its earlier, Central China counterparts, such as the circa 1872 Himly tile set for example.

Further, if one assumes the 1915 Mauger and 1917 Xu Ke descriptions reflect tiles sets from Central China, one may also hypothesise that since the four documented tile sets from Central China (the circa 1872 Himly and circa 1875 Glover sets and, importantly, the 1889 Wilkinson set) did not use the ‘Phoenix’, ‘Dragon’ and pin characters, nor alternative characters for the ‘Directions’ group, then the later appearance of these characters in the sets described by Mauger (1915) and Xu Ke (1917) for example, may be a consequence of their diffusion from North to Central China, possibly around the beginning of the 20th century32. If these differential diffusions were the case, then one possible mode of diffusion may have been merchants, diplomats or Chinese Officials travelling by steamer between the major treaty ports.

According to Dai (1934), his late older brother was considered as “the first one among Tianjin amateurs [to learn] of “bamboo woods distractions” [a metaphor for ma que]. Sheng presented him with a small tile set whose tiles, much smaller than today’s tiles, had bone fronts and bamboo backs, and were finely crafted. The myriads were replaced by ranks…”

Dai reports the set remained in his family for 40 years and then passed to him33.

The name of the game “bamboo woods distractions” may also be a reflection of earlier tile sets that were used for the game, that is, the sets were made of bamboo only – as with the circa 1875 Himly set from Ningbo for example.

  • [22]

    Very special thanks to Laurent Long, Thierry Depaulis, Alan Kwan and John (Zi Rong) Low for the translation.

  • [23]

    Cheefoo [Yantai] was a major treaty port, opening in 1862 (Wikipedia).

  • [24]

    According to Feuerwerker (1970), by 1884 Sheng Xuanhuai was in charge of both the Imperial Telegraph Administration and also the construction of the Shanghai-Canton (Guangzhou in Guandong province) coastal telegraph line, which was completed in the summer of 1884. Sheng was in Shanghai in January 1884 and by April had returned to Tianjin. In the summer of that year he briefly held the post of Tianjin customs Commissioner as acting incumbent. In 1885 he was appointed Director General of the China Merchant’s Company and concurrently with this responsibility, continued to hold posts in the regular provincial hierarchy of Shandong in North China. In July 1886, he was appointed military administrative Commissioner and Superintendent of customs at Cheefoo [Yantai], “where he remained for six years … carrying out efficiently the traditional responsibilities of his office” (Feuerwerker 1970). In June of 1892 he was appointed to the post of Tianjin customs Commissioner. By June of 1894 he was on leave due to illness and then returned to Tientsin to serve as “Director-General of Army Transport”.

  • [25]

    In Imperial China there were 9 ranks of Chinese Officials. See footnote 13.

  • [26]

    That is, the four “Virtues” replace the four “Winds”.

  • [27]

    But for the Sheng Xuanhuai set, the symbol used for the # 1 wan is unknown.

  • [28]

    See footnote 13.

  • [29]

    See Stanwick 2004a and 2006.

  • [30]


  • [31]

    Unfortunately there is no description of the # 1 ‘Cash’ and #1 String of Cash’.

  • [32]

    Alternatively, while acknowledging the initial assumption, it is possible that tile sets from Central China that were using these alternative characters, may have co-existed – much earlier than the limited evidence suggests – with tile sets that were not using them.

  • [33]

    This would suggest the gifted set (not Sheng’s own set) was given sometime around 1893. However, it is not certain whether this gift was contemporaneous with the appearance of Sheng’s own set in Tianjin. It is entirely possible he bought his set to Tianjin in 1884 and gifted another set to Dai’s brother, perhaps in 1893/1894.

6. Conclusion

The tile set descriptions given by Dai (1934), Mauger (1915) and Xu Ke (1917), list alternative characters for the ‘Honours’ tiles. Both Dai and Mauger also list alternative characters for the four ‘Directions’ characters and cite the use of the pin character instead of the wan character for the ‘Myriads of Cash’ suit. The 品 pin character has also been found in tile sets produced in China circa 1920, along with alternative characters for the ‘Directions’ group. Further, it may have been the influence of a Mauger pattern variation – that used the alternative ‘Dragon’ and ‘Phoenix’ characters – together with the type of 1909 Culin pattern, that was reflected in Babcock’s tile sets as his so-called genuine tile set pattern34. Alternatively, a similarly modified Culin pattern may have already been in existence during Babcock’s tenure in China. Whatever the case, it was this form of modified pattern, exported to the West by Babcock35 at the very beginning of the 1920s, that was directly associated with his term Mah-Jong(g), to designate his version of the Chinese game of ma que36.

  • [34]

    This pattern appears in the ‘Mah-Jongg Sales Company of America’ catalogue No. 2. Circa 1923.

  • [35]

    See footnote 17.

  • [36]

    See Babcock, 1923a, 1923b.


Special thanks to Thierry Depaulis for procuring the Mauger text and some of its translation and the translation of some of the references’ titles. Special thanks also to Thierry, Laurent Long, Alan Kwan and John Low for help with the Dai Yu’an translation. Thanks to Tom Sloper for help with the Mauger diagrams.


This series of articles is dedicated to my late friend Troy Ozama.


Babcock, Joseph P.
– 1923a. ‘Mah-Jongg. Its Authentic Source’. In: The Saturday Evening Post, December 15th.
– 1923b. Babcock’s Rules for Mah-Jongg. The Red Book of Rules. 2nd Edition. Seventh Printing. (1st Edition 1920).

Culin, Stewart. ‘The Game of Ma-Jong, its Origin and Significance’. In: Brooklyn Museum Quarterly, vol. XI, 1924, pp. 153-168.

Dai, Yu’an. Gu shui jiu wen [Old Hearings from the Gu River]. Preface dated 1934. Tianjin: Tianjin guji chubanshe, 1986.

Ebashi, Takashi. ‘Proto Mahjong. Mahjong tiles in the 19th Century’. In: Mahjong Museum Report, Vol. 5, Nos. 2-4. Issues 9-11. 2005.

Eberhard, Wolfram. A Dictionary of Chinese Symbols. Hidden Symbols in Chinese Life and Thought. Routledge, 2003. (1st ed. 1973).

Feuerwerker, Albert. China’s Early Industrialisation. Sheng Hsuan-huai (1844-1916) and Mandarin Enterprise. Atheneum, New York, 1970. (Originally published by Harvard University Press, 1958).

Lo, Andrew. ‘China’s Passion for Pai: Playing Cards, Dominoes, and Mahjong’. In: Asian Games: The Art of Contest, Colin Mackenzie and Irving Finkel, eds. Asia Society. 2004. pp. 217-231.

Mauger, George Edward. ‘Quelques considerations sur les jeux en Chine et leur développement synchronique avec celui de l’empire chinois’ (Some considerations on games in China and their synchronic development with that of the Chinese empire), in: Bulletins et Mémoires de la Société d’Anthropologie de Paris, 6e série, t. VI 1915, p.228-281 (Offprint Paris, 1917).

Noguchi, Kyoichiro. (Compiler) Edited by the Mah-Jongg Museum. History and Culture of Mahjong. A Pictorial Record of the Mahjong Museum. Publisher: Takeshobo, 2005. (In Japanese. Some English text).

Sloper, Tom. (2006).

Stanwick, Michael.
– 2004a. ‘Mahjong(g) Before Mahjong(g): Part 1’. In: The Playing-card, Vol. 32, No. 4, 2004, pp. 153-162.
– 2004b. ‘Mahjong(g) Before Mahjong(g): Part 2’. In: The Playing-card, Vol. 32, No. 5, 2004, pp. 206-215.

Stanwick, Michael. ‘Mahjong(g), Before and After Mahjong(g): Part 1’. In: The Playing-card, Vol. 34, No. 4, 2006, pp. 259-268.


Wilkinson, William H. ‘Chinese Origin of Playing Cards’. In: The American Anthropologist, vol. VIII, 1895, p. 61-78.

Wilkinson, William H. (1890): Published in 1901 as pp 184-194 of Catalogue of the Collection of Playing Cards Bequeathed to the Trustees of the British Museum, F. M. O’Donoghue.

Xu, Ke. ed., Qing bai leichao (“Qing Unofficial Categorized Extracts” or “Classified Anecdotes from the Qing Dynasty”). Shanghai: Commercial Press, 1917, 7 vols. (reprint Taipei, 1966, 12 vols.; Peking, 1984).

Yu, Haoxu. True History and Culture of Mahjong. Publisher: Display Hall of the Birthplace of Mahjong. Ningbo, (2002).