Proto-mahjong

Assessment

Confidence

lost no rules text survives

Evidence

corroborated two or more independent sources agree on the facts that define this variant

Cross-comparison

What each source was read to say, set side by side. The numbers are the sources above.

• [1] Sloperama's family tree page gives the Matiao / Khanhoo / dominoes root structure the chart's root node stands for
• [2] Stanwick and Xu's tile-set research on early ma que, underlying the Hua Maque and Wang Maque material
• [2] dates the physical evidence: the ma que pai sets of Carl Himly and George Glover were collected in 1868-1876, 1872-1873 and 1875, and the wind tiles are found in most known ma que pai sets after the 1870s. The game is 麻雀 ma que, literally 'hemp bird' and in bonded form 'sparrow', also known in the Late Qing period as peng hu, 'encountering harmony'
• [3] names the ancestry directly: khanhoo is a non-partnership draw-and-discard Chinese card game, first recorded in the late Ming as a multi-trick-taking game, which 'belongs to the same family as Mahjong'. By the late Qing the rules Wilkinson and Culin recorded had changed from trick-taking to draw-and-discard, a shift attributed to the 18th and 19th centuries when trick-taking gave way to shedding games — mòhú (默和) and pènghú (碰和), which it calls the ancestors of mahjong. Wilkinson had Charles Goodall issue a khanhoo pack with a rules booklet in 1891
• [4] the history FAQ's own introduction sets the frame: mahjong 'as we know it today was created in the mid-to-late 1800s', based on money-suited card games; 'the earliest written record describing the gameplay of mah-jongg is no older than the 1890s, and the earliest known mah-jongg sets go back no farther than the 1870s'. Part VI (11f) names the node: 'Millington calls these unknown original rules "proto-mah-jongg"', and says 'No written rules have ever been discovered that describe how the game was played with this different set of tiles'
• [5] gives the node's physical definition, a composite of the two Glover sets of the early 1870s: the three suits; the four winds and four 'Center' (中) tiles but no 發; eight blanks; four seasons and no flowers; four rulers of Heaven, Earth, Man and Peace; and four direction rulers. Culin reported a fifth 'Center Ruler', which Sloper and Stanwick both doubt because it would give an odd tile count. 'It is not yet known how the "ruler" tiles were to be used in play'
• [6] PRIMARY: Culin's 1893 report lists the Long Island Historical Society set, 'the gift of the Hon. George Glover, formerly U.S. consul at Fuhchau', piece by piece: 4 suits of 9 in circles, rectangles and myriads; 4 each of north, south, east, west and chung 'middle'; one each of the northern, southern, eastern, western and middle rulers, the heavenly, earthly, human and harmony rulers, and the four seasons; and 8 blanks. It notes a similar Glover collection in the American Museum of Natural History. So the contested 'middle ruler' is in Culin's own text, not an error of later transcription
• [6] PRIMARY: records that Wilkinson exhibited 'a set nearly identical' from Ningbo, called chung fa, which lacked the five direction rulers, the heaven, earth, man and harmony rulers and the seasons, but had four pieces bearing fat (發). This is the first published side-by-side of the 1870s form and the c. 1890 form, and it shows the difference is the extra tiles, not the suits
• [7] traces the Glover sets to their museums: AMNH catalogue 1/2377, now 143 pieces, and Brooklyn Museum 28.779, now 143 pieces, which Stanwick and the museum's staff identify with the set Culin described. Hart's letters place Glover at Fuzhou from 22 February 1872 to 12 September 1873. Glover's own cover note, dated Shanghai 18 August 1875, calls the game 拷家雀 k'ao chia ch'iao, 'snatching the house sparrow', 'composed of 148 pieces' — the earliest association of 雀 'sparrow' with a tile set
• [7] the Himly set: 148 bamboo tiles described by Carl Himly (1889, 1901) as 'Ning bo zhu pai', bamboo cards of Ningbo, acquired during his Shanghai posting of 1868-1876. It has the three suits, one 化 hua joker per suit (tong hua, suo hua, wan hua), the winds, four direction kings plus a 摠王 'almighty king', the heaven, earth, man and harmony kings, the seasons and eight blanks — but no 中 tiles. Stanwick's reading: the directions were present before the dragons
• [7] sets out what the node lacks: 'From the documentation presented no date of an origin can be proffered, only the earliest date so far determined'. The earliest recorded game-play is Wilkinson's 1890 manuscript on chungfa, quoted in his unpublished 1925 memorandum: four players named for the compass points, 15 cards each as in khanhoo, sequences that 'score nothing', and a tariff 'not fixed, but ... determined by the players before commencing'. Its meld 'any 3 aces, or any 3 nines (not all of the same suit)' is one mahjong does not have
• [8] concludes that the tile set 'remained fairly stable in overall group and tile numbers between circa 1872/73 and circa 1920', though the composition of some groups varied, and that the set may have standardised around Ningbo and Shanghai between 1890 and 1901. Wilkinson's 1889 Ningbo set in the British Museum (1896,0501.904) has 140 pieces and a box stamped 中發, and in 1889 Wilkinson found chungfa confined to Zhejiang and Jiangsu
• [9] Depaulis's translation of Himly's 1901 tile-by-tile colour description is what makes the lost Himly set reconstructable: 'Himly's early maque (Mahjong) set is now lost', but his text records each tile's engraving and colour
• [10] records Himly's own inference about the set: because two of its three suit names, 索 and 万, match the suits of a class of three-suited money cards, and the suits in both are quadruplicated to 36 pieces, Himly treated his tiles as thick cards of the 'draw and discard' family
• [11] the Sheng Xuanhuai set, known only from Dai Yu'an's Gu shui jiu wen (preface 1934), whose author's brother was Sheng's interpreter: 140 pieces, with four virtue tiles in place of the directions, a 品 'rank' suit in place of myriads, and four king tiles, three named for the suits plus a 摠王. It is classed as a precursor for its kings, and dated to Sheng's arrival at Tianjin in 1884 or 1892
• [2] proposes a function for the extra tiles, by analogy with two 1941 descriptions of living games: in Beijing/Tianjin hua ma que the three suit kings and the 摠王 are 'inner flowers', concealed jokers, and the heaven/earth/man/harmony kings, direction kings and seasons are 'outer flowers', exposed on draw. Applied to the Himly and Glover sets, this makes the 1870s extras jokers and flowers. The authors label it a hypothesis, and note that in Dai's account of the Sheng set the suit kings were played as outer flowers instead
• [12] carries the card ancestry back past 1850. Jin Xueshi's Mu zhu xianhua (1783) describes mò hú, a 60-card draw-and-discard game, and pèng hú, played with two decks (120 cards, four of each) or two and a half (150), in which three of a kind is a peng. Li Dou's 1796 Yangzhou huafang lu ranks ma diao first among card games, then peng hu, and notes that dominoes 'are now bamboo pai instead'
• [12] places the game in the 1870s-80s Treaty Ports: a Shen Bao report says that after gambling was banned in the Shanghai concession (the report covers 1865-1887) gamblers 'invented' cha mo ziang and peng hu, and police banned both from the brothels on 12 November 1887. 麻雀 in connection with the game first appears in Shen Bao on 18 February 1884; the name mo ziang is argued from dialect to be Ningbo's. Stanwick and Xu suggest the similar Glover and Himly sets spread 'via officials travelling by sea between the Treaty Ports'
• [13] CHINESE PRIMARY SOURCES: Zhang Deyi's diary 航海述奇·五述奇, entry for Guangxu 15, 1st month, 29th day (1889), describes fellow travellers playing a game 'named 麻雀, also 馬將': 1-9 of 索, 餅 and 萬, 中發 or 龍鳳, 白版 and the four winds, four of each 'as the usual rule'; and, if flowers are added, one each of the east/south/west/north kings, 總王, 索化, 萬化, 同化, the seasons and several further quartets. So the Himly-style kings and suit-jokers were still an optional layer in 1889
• [13] on place and date: Fan Zengxiang's 樊山續集 (printed 1902) says 'the game of 撮麻雀 arose in Ningbo twenty years ago and is now everywhere'; Xu Ke's 清稗類鈔 says it 'began at Ningbo in Zhejiang' and elsewhere that Taiping soldiers added 筒化, 索化, 萬化, 天化, 王化 and the direction 化 tiles, all Taiping titles, before it reached Ningbo; Du Yaquan's 博史 (1933) says it spread first among Fujian and Guangdong seafarers and reached Ningbo early in the Guangxu reign. The author concludes the game was fixed in form at Ningbo, born in the 1870s-80s
• [13] on the inventor: the earliest source naming Chen Yumen (陳魚門, 1817-1878) is Chen Ailu's 《麻雀牌考補》 in the magazine 紅玫瑰, 1925, which has Chen, a fish-market man, rework paper cards into 136 bamboo pieces plus four seasons. The author admits 'no late-Qing source directly records Chen Yumen inventing mahjong', that a single witness does not stand (孤證不立), and that 紅玫瑰 is not a scholarly journal, yet calls Chen the most probable candidate
• [14] traces the Chen Yumen story to J. B. Powell's 'Mah Chang: The Game and Its History', China Weekly Review, 30 June 1923, where a fisherman Sze invents the cards, General Chen Yumen adds 東南西北中發白 to make 136, and Chang Shiu-Mo adds eight flowers. It states that apart from Chen Yumen being a real person, none of the details has support, and that no late-Qing or early-Republican writer before Powell connects Chen with the game
• [14] notes, citing 中國賭博史, that the Taiping Heavenly Kingdom's repeated gambling bans mention dominoes and dice but not mahjong, so Xu Ke's Taiping origin 'remains doubtful' — while observing that Xu Ke's list of 化 tiles is the only independent written account matching the Glover-type sets' features
• [15] FIRST-HAND REPORT, not a fetched document: Sloper writes that a 2008 CCTV-9 programme claimed British consul F.E.B. Harvey's 1861 diary records Chen Yumen teaching him the game, and placed the diary in the Chiba Mahjong Museum. Sloper, Stanwick, Depaulis and Takunori Kajimoto wrote to the museum's founder, who 'replied and said that the diary is not in his museum'; in 2011 Yukio Mori added that he had never heard of it from Tianyi Pavilion's curators either
• [16] the Harvey story as Wikipedia relays it: Chen Yumen 'enabled him to teach visiting British Consul General F.E.B. Harvey the game in 1861'. The essay's own footnote for that paragraph is the 2008 CCTV 'New Frontier' documentary on YouTube. It does not say Harvey recorded any rules, and adds that 'information connecting individuals to the invention of specific rules is nonexistent'
• [17] CONTRADICTS THE NODE, weakly: the 'Mahjong in the West' section says Harvey met Chen Yumen, who taught him the game, and that 'In the same writings he details the rules he was taught by Chen'. The article's own footnote on that sentence reads 'Harvey's "papers" and "detailed" description of the game have never been found. This story is probably forged.' The History section also dates the earliest surviving tile sets to around 1870, from Fuzhou, Shanghai and Ningbo, and relays Depaulis's suggestion that the king tiles were dropped because they shared titles with the Taiping leaders
• [18] dates the honours: the earliest surviving sets 'date to the 1870s when the game was largely confined to Zhejiang, Shanghai, and Jiangsu'; the dragons 'were invented after the wind tiles beginning with the whites which were developed from replacement tiles, followed by the reds circa 1870, and finally the greens which entered the set by 1890'. One of the earliest sets lacked 中 and the earliest lacked 發
• [19] Culin's later view: 'The Chinese game we call ma-jong is a recent Chinese modification' of a card game played with money-derived cards, whose 120-card pack (four identical sets of three nine-card suits plus three extras) he calls 'The Immediate Predecessors of the Game of Majong'. He repeats Wilkinson: in 1891 the game 'like the present ma-jong is from Ningpo and called chung fat', confined to Zhejiang and Jiangsu
• [20] PRIMARY: three-suited money cards 'are known in central China by the name of kun p'ai, staff or baton cards, or ma chioh, "hempen birds"'. A set is thirty pieces: ace to nine of ping, tiao and wan, with the white flower, red flower and 'a thousand myriads'. So in 1895 the sparrow name belonged to cards, not to tiles
• [21] the biographical stub gives Chen Zhengyao, courtesy name Yumen, as 1817-1878, sourced to Xu Shidong's birthday address of 1866 and Dong Pei's tomb inscription. It states that he carved the cards onto bamboo, assembled 136 pieces and 'invented mahjong', and that 胡 comes from Ningbo dialect, with no citation for any of it. The man is documented; the invention is not
• [22] Sloper's own position on Chen Yumen: he tends to think Chen was the creator, but the attribution 'still needs more research', since 'it is not known precisely what changes Chen Yumen made to the precursor games - and the exact rules of the game that Chen Yumen devised are likewise not known'. He lists the rival attributions (Taiping officers near Ningbo 1850-55, Hung Hsiu-chuan, two Ningbo brothers 1870-80, palace concubines) as 'too vague or too colorful to be relied upon'
• [23] the precursor chain as Sloper gives it: matiao, a four-suited trick-taking game, then 'a game known as Peng He Pai used four sets of three-suited cards'. John Low's reading of the Ningbo Display Hall's panels describes 'Zhi Pai' or 'Peng He Pai' as a rummy-like game first of two, then four, decks of 30 money cards (120), won with three groups and a pair — mahjong needs four. Depaulis names kan hu, cheki, si se, pian jin and kuwaho as the nearest relatives, but says none 'seems to have given rise to Mahjong'
• [24] the earliest writings, dated: Culin 1893 ('earliest known written mention of the game anywhere, in any language', per the Mahjong Museum), Wilkinson 1895, Natsume Soseki 1909, and the earliest Chinese book, Shen Yifan's 繪圖麻雀牌譜 of 1914. Depaulis found Wilkinson's 1925 memorandum 'with the earliest description of the rules of Zhongfa in 1890' through WorldCat
• [25] page 5 names the transition this node stands for: tradition holds that the flower tiles had multiplied into confusion, and that the people of Ningbo boldly pruned them. The set before the Ningbo reform is called 花麻雀 'flower sparrow', and the set after it 清麻雀 'plain sparrow'. Ebashi reads Wilkinson's flowerless 1889 set as the surviving evidence that the 'plain' 中發 set was on sale in Ningbo by then
• [25] page 11, summarising: the Glover tiles of the 1870s have the 'qingfu' insect on the one of strings, no dragon triad yet, and one each of eight 'king' tiles (東王 南王 西王 北王 天王 地王 人王 和王) beside the winds, and 'how these tiles were used in the game is unknown'. In the 1870s, in Shanghai and Ningbo, the game turned into the modern form, the 中發 or Ningbo set, whose defining changes were adding 發 and dropping the framed borders on the character tiles
• [25] page 4, on the Himly set: with neither 中 nor 發, it is 'a sub-type that branched off before the dragon triad appeared', which Ebashi guesses happened in North China and the north-east (Beijing, Tianjin, Dalian). He reads 摠王 as the 'all directions' king, possibly the ancestor of 中, and notes that the Glover and Himly sets revive Xu Ke's Taiping-origin account
• [25] page 9 adds a set the English literature does not have: the Ebashi set, bought from a Shanghai antique dealer without provenance. It is 22 × 18 × 9 mm, bone on bamboo, concave-faced with square corners, and has 中發白 and no borders on the character tiles. By tile size it dates just before Wilkinson's, which makes it the earliest 中發 set he knows. Tile size grows steadily from Himly (14.4 mm) through Glover (18-19), Wilkinson (23), Laufer (24) and Culin (26)
• [26] the tiles article summarises the same sequence: the Glover sets circulated mainly in 1870s Zhejiang, Shanghai and Jiangsu; the dragon triad appears only with the 1890 中發 set; outer flowers spread only after 1920; and the special inner flowers of the early sets disappeared from most of China but survive in Vietnam and Thailand

Reading it

Well dated from surviving objects, and now traced to where each origin story starts. The chart's c. 1870s rests on four tile sets and their collectors' own records. The two Glover sets are in the American Museum of Natural History and the Brooklyn Museum; Glover was at Fuzhou in 1872-73, and his 1875 cover note names the game after the sparrow and counts its 148 pieces. Himly's set from Ningbo is lost, but he described it tile by tile in 1889 and 1901. Culin's 1893 report, read here in scan, lists a Glover set piece by piece. They agree on what the proto set was: the three money suits that carried over, the winds, eight blanks and four seasons, plus a layer of 'king' tiles that did not survive. The Himly set adds suit jokers. There are no flowers, and one or both of the later dragons are missing.

'Rules lost' also holds up, and the one source that seems to contradict it falls apart when checked. Wikipedia says the consul F.E.B. Harvey wrote down the rules Chen Yumen taught him in 1861. Wikipedia's own footnote calls the story probably forged. The essay it cites takes the story from a 2008 CCTV documentary and does not mention rules. The museum the documentary said holds the diary told Sloper and three other researchers that it does not. The closest thing to proto-mahjong rules is Wilkinson's 1890 note on chungfa, and that describes the later, c. 1890 form, not the 1870s one.

Two things stay open. The first is who made the game. Chen Yumen (1817-1878) is a real, documented official of Ningbo, but the earliest texts that credit him with the game are Powell's 1923 magazine piece and Chen Ailu's 1925 one, and no late-Qing source names him. Late-Qing and early-Republican writers name Ningbo as the place, and Du Yaquan's 1933 account puts it among Fujian and Guangdong seafarers first. The second is what the kings did. Xu Ke's Taiping account matches the Himly set's tiles, and Zhang Deyi's 1889 diary still lists them as optional flowers. Stanwick and Xu's inner-and-outer-flower reading is a reasoned hypothesis, not a recovered rule.

Ebashi's Japanese study, read on 2026-09-24, gives the period a frame the English sources lack. The game before Ningbo's reform was 花麻雀 'flower sparrow', and the pruned 136-tile game after it was 清麻雀 'plain sparrow'. On that reading this node is the flower-sparrow stage, and Chinese archaic is the plain-sparrow stage that followed. Two branches that kept the proto extras and survived into the 20th century are now in the tree as this compendium's own additions: North China's hua ma que and Changsha's wang ma que.