The rules debate as seen from Ancient China

John Fairbairn

THERE IS currently (June 2006) a debate raging about scoring in ancient Chinese go. There are many confident assertions but a paucity of evidence. Some evidence does exist. Unfortunately it is sometimes only suggestive. Worse, some evidence may be contaminated. This article is a survey of what is available.

It is an opportunity also to introduce a major new contribution to the subject by Chen Zuyuan in Weiqi Tiandi, Issue 1, 2006, page 71, "Correcting the Wangyou Qingle Ji."

The Wangyou Qingle Ji (Carefree and Innocent Pastime Anthology, C&IP), from the first quarter of the 12th century, is often described as the oldest extant go book. It is the oldest complete book, but the Qi Jing (Go Classic, QJ) from Dunhuang is substantially older and is almost complete. The claims for C&IP are strong, however, in that, being an anthology, it brings in data older and broader than in QJ. It has the first known examples of diagrams and game records. It is altogether richer fare. Indeed, it is the only source of games before around 1550.

The real reason C&IP is often touted as the oldest book is actually more prosaic. Oriental scholars did not realise until quite recently that the QJ even existed. It was brought from Dunhuang to the British Museum in London by Aurel Stein and lay catalogued but essentially hidden until the late 20th century.

It should be stressed, however, that there are plenty of references to go in China that predate both these books. Some allude to scoring. There are even some tantalising references in Japan. I have recently drawn attention to a reference in the Tale of Genji which hints strongly at a scoring system similar to the modern Japanese one - the same system, in other words, that is often assumed to have existed in early China. But, on the whole, there is no surviving evidence about early rules of go from Japan (or Korea).

Readers not familiar with the arcane debates on rules will need some orientation before we can proceed. In modern Japanese rules, the winner is decided by each side totalling up the empty points they have successfully surrounded plus the number of enemy stones (pieces) they have captured and removed from the board.

Two things are to be noted about this. One is that it is necessary for each side to keep track of the stones he has captured, usually by physically keeping them somewhere. The other is that, at the end of the game, the areas that a player has successfully captured ("territory") may include enemy stones that are effectively captured - they do not have "two eyes" and are regarded as "dead". Dead stones are simply removed at the end of the game and are counted as captured stones, even though they have not been captured in the normal way, and the points vacated by them become part of that territory.

Capture of an enemy stone normally involves filling in all the points adjacent to it (its "liberties") with your own stones. If you were actually to capture dead stones in modern Japanese rules, you would have to add extra stones to the board, which means taking away points of territory - a bad thing.

In modern Chinese rules, however, the approach is rather different. The winner is decided by who controls more of the board, control being defined as the players' own stones on the board plus the empty points he has successfully surrounded (one point of score for each stone or point of territory). In this system, captured stones and dead stones are disregarded in the count. This means that it is (1) not necessary to keep track of captured stones, and (2) dead stones can either just be removed or can actually be captured on the board without any points being subtracted from the total score.

Remarkably, the two systems nearly always give the same final score. But because filling in one's own territory has no penalty under modern Chinese rules, it is possible to play out some freak positions that arise occasionally and resolve them by actually capturing. Under Japanese rules, resolving these positions by capturing would be penalising oneself, but not playing out the freak positions can lead to obscure anomalies which can only be decided by arbitration.

The merits of each set of rules are a major part of the current debate, though it has extra layers of complexity. (It should be added that this is almost entirely theoretical - in everyday practice go rules are very simple and the freak positions are very rare.)

One approach to sorting out the theory is to try to track down the ur-rules: to try to get inside the minds of the people who first crafted the game. Regrettably for rules theorists, who tend to believe that some form of the modern Chinese rules produces the most logical way of counting the game, it seems that ancient Chinese rules were more or less the same as modern Japanese rules.

The strongest, but not the only, proof of that is C&IP.

This proof centres mainly on a small number of games, one of which is called the Four Immortals of Chengdu. We know quite a lot about this game as it happens, and not just from C&IP. Chengdu was then an important government town, being in the centre of a rich farming area and trading with the west. Even today it has special affinities with go, being a major go publishing area. In Song times, the era of C&IP, it was a cultural centre known overseas.

We may fairly assume that the players there were of the first rank in China. At any rate, an observer called Xu Zongyan made a note of the game, telling us that it was played on the 10th day of the first month of 1094 at Peng City close to Yangzhou, another major city with strong associations with ancient go. Xu also tells us that the players were Liu Zhongfu from Jiyang (Jiangxi Prov.), Wang Jue from Kunling (now Changzhou Wujin County, Jiangsu Prov.), Yang Zhonghe from the Ye Commandery, Zhangde City (Henan Prov.) and Sun Shen from Yimen or the Barbarians Gate - the western gate of Bianjing, Kaifeng City (the capital of Henan).

The implication is that these are the best players from throughout the empire who, as ever, gather in the capital cities to find the best opponents. Presumably they were based in Chengdu but were in Yangzhou for a special occasion. That speculation seems justified by the fact that this was a special game - a relay game, intended no doubt to show the skill of the four best players at once.

C&IP does not give all these details. They come from a later and non-standard version of the Xuanxuan Qijing (Gateway to All Marvels). C&IP simply gives the title of the game, Chengdufu Si Xiuanzi Tu, followed by the courtesy names of the four players (no places of origin), each prefaced with the order in which they played, "Playing 1st Zhonghe, Playing 2nd Xian, Playing 3rd Jue, Playing 4th Zhongfu." Incidentally, it is not mere speculation that these are master players. We know quite a lot about Liu Zhongfu from elsewhere. He was the premier player and imperial go tutor in the time of the Northern Song Emperors Shen Zong and Zhe Zong (1068~1101). His Go Secerets is on the GoGoD CD. Co-incidentally, this game was played almost exactly as the same time as the games described by Murasaki Shikibu in Tale of Genji.

Then C&IP gives the game record, in one diagram, followed by just two lines of text:

White has killed 9 black stones and after filling in has 43 points.
Black has killed 7 black stones and after filling in has 42 points.

白殺黒九子填外有四十三路
黒殺白七子填外有四十二路

Several things to note here: (1) the winner is not specified; (2) there is no reference to number of moves; (3) there is no reference to group tax or other arcane practices.

The significance of this note is manifold. First, consider the games given in C&IP. The first three have a title incorporating the players' names (e.g. the first is Sun Ce summons Lü Fan to Play Go), a single diagram and a note saying "In all XX moves, no further moves given." In the case of the Sun Ce game, the specific reference is to 43 moves in all:

共四十三着
分局面停

The next two games have exactly the same formula, but have 83 moves and 77 moves respectively. In no case is a result given. But a distinctive feature of all these games is that White plays first, and all relate to then old and datable events. In other words, they are very old games. The Sun Ce game can be dated to 196. Game 2 features the Jin emperor Wu Di and so can be put between 265 and 290. Game 3 features the Tang emperor Xuan Zong and can be put at between 712 and 756. These last two emperors are described as major go nuts in several sources.

Then follow 10 games under a general title 諸國手野戰轉換十格圖 which may be rendered as Ten Fighting Master Games.

They each have an identical formula. There is a brief title, based on the venue, then a longer note on the venue with the players' names, then the formula "Black plays first, altogether XXX moves". There are no other notes with the one exception of one game which lists ko moves. It is the only game with ko captures, but the number of moves are limited to between 118 and 178. Again, no results are given, and in this case there is no reference to "game left unfinished".

For example, one game is headed:

瓊林圖
東京新鄭門瓊林苑相公 龎李百祥饒晉士明黑先 共一百六十三着

Qionglin Diagram
The Xianggong Pang had Li Baixiang cede Black and first move to Jin Shiming at Qionglin Garden, beyond the Xinzheng Gate, Kaifeng
Altogether 163 moves

Xianggong was a respectful word for a high minister. There was a person called Pang Ji (988~1063) who became the Grand Guardian of the Heir Apparent with the high rank of Princely Attendant in the Song court in the Renzong era. It is probably him, and this gives us an approximate date. The Xinzheng gate was the southernmost (and main) gate in the west wall of the city, and the Qionglin garden was on the south side of the main street which passed through the gate. All these 10 games are located in Kaifeng, and nearly all the locations can be identified on an old map of the city (the names have changed since). From these circumstances we can assume that the games were played by the best players of the time who enjoyed patronage of high officials.

The significant new piece of information here is that Black played first in every case, and for some reason this had to be specified. It is possible that who played first changed over time. It is possible that White played first in the three historical games because they involved royal personages.

There are four more games in C&IP, however, and they offer some clues. They are a little bit of a hotchpotch. The first, Meeting an Immortal, has a title which describes it as Liu Zhongfu meeting an old lady at Blackhorse Mountain (a spa used as a summer retreat), but "also called the Coughing up Blood Diagram". Now although Liu Zhongfu was then a contemporary player, there is a strong allusion to the Tang dynasty tale of Wang Jixin meeting (and losing to) a woman in the mountains beyond Chengdu who played blindfold before disappearing into thin air. Also, the fact that the game had clearly been around long enough to acquire a totally different (and unexplained) name suggests that it was a very old game, perhaps hijacked by Liu as a party-piece. Significantly, White plays first, so if it indeed was old at the time, that would fit the pattern so far. Incidentally, this game is marked "altogether ( 共 ) 112 moves", and has no result.

The next game is the Four Immortals of Chengdu, which we will return to later, but remember that it has no total move number or result. It has Black playing first, but as we can date this game to 1094, that would put it in the same bracket as the Kaifeng games, which also have Black playing first.

The next game switches back to White playing first, but this game is called the Rotten Axe-Handle Diagram, and the headnote mentions the tale of Wang Jixin, so we know this game relates to the Tang dynasty. Once again, the colours fit the date. This game, however, differs in some respects. First, the header says: "White plays first, Black wins by 1 point". Then after the diagram it says: "145 moves each. Black has killed 22 white stones. White has killed 9 black stones. Black has 18 points. White has 17 points."

The differences in this formula from the Chengdu (and other) games is that it says "each" 各. And unlike the Chengdu game there is no reference to filling in. It was also felt necessary to specify White played first.

The next game, Gold-Petalled Bowls Diagram, is another historical reference, and so again White plays first. As the header tells us, "The Expectant Officials Yan Jingshi and Gu Shiyan battled for a pair of gold-petalled lidded bowls."

The Expectant Officials were the court go masters. Gu Shiyan is referred to in a story in the Yuhai (Sea of Jade) and elsewhere, where it says, "a Japanese prince came to the Tang court in the 4th month of the 7th year of the Taizhong era of the Tang emperor Xuan Zong (853). The emperor ordered Gu Shiyan to play the prince. Shiyan on move 33 played the Pressing the Divine Head manoeuvre and the prince did not know [what to do]. Finally he submitted."

Then we are again specifically told that White played first, and Black won by 1 point. After the diagram, with exactly the same formula as the Rotten Axe-Handle game, it says, "122 moves each. Black has killed 6 white stones. White has killed 6 black stones. Black has 40 points. White has 39 points."

The final game is another supposedly from the Tang dynasty, featuring Jia Xuan, an Expectant Official playing with Yang Xican. However, in this case Black played first, so the theory that White historically played first seems to take a beating. However, the Tang dynasty, though it lasted a very long time, ended in 907. There was a Posterior Tang dynasty from 923 to 936. The previous games would belong to the early or mid part of the main Tang era. Jia Xuan, though, flourished between about 976 and 995. Though he may have been born and become established under the Posterior Tang, he is in reality closer to the Song era, and so playing Black first would be plausible.

On this basis, we can put forward a tentative hypothesis that White orginally played first, but the practice changed in the late Tang, perhaps some time in the 10th century. We can strengthen the hypothesis, though not the dating, if we accept Japanese references to the superior player originally taking Black and the inferior player taking White and playing first, the order later being reversed. For this see Dr Ogawa Takuji's work on the "Origin and Development of Go in China" in Shinagaku Vol. VI No. III (1932) and Vol. VII No. I (1933). Ogawa (1870~1941) was a geologist but made many field trips to China and as a by-product became an eminent go historian.

Before leaving the topic of who moves first, it might be worth mentioning a totally different approach, relating to a very different time span, by Jiang Zhixin in Weiqi Tiandi 2006/5, page 64: "On the Origins of White Playing First."

Reverting to Jia Xuan for a moment, this game against Yang Xican too ends with a formula: "119 moves each. White has killed 21 black stones. Black has killed 9 white stones. White has 43 points. Black has 51 points."

But Black and White have been reversed in this formula, for some reason. And there are two other important factors. One is the presence of a seki (the first known!) It appears to have no effect on the score, however counted, but unfortunately it is the less interesting type of seki with no points inside it.

The more important factor is that the score - Yang lost by 8 points - does not fit if we count by Japanese rules. That way, Black wins by 6: Black has 55 points and White 49. However, if we make an assumption - and it is a big assumption as there is nothing specific in any of the old texts to justify it - that group tax applies, then we can make the score fit.

Group tax is a concept known from more modern but still old Chinese go. The Chinese term is "cutting penalty." What it means is that every group has to have two eyes to live. By some unknown logic it was felt that these two points should not count as part of the score. The effect therefore is to subtract two points from the score for every separate live group you have. In this game Black had just two groups so his score becomes 55 - (2x2) = 51. White has three groups so his score is 49 - (3x2) = 43. Voilà!

Although group tax cannot be applied so convincingly in the Chengdu game, for reasons we shall look at later, it also makes the scores fit in the Rotten-Axe Handle game and the Gold-Petalled Bowls game, so the assumption seems justified after all.

There has never been any evidence that group tax was used in Japan, though, so to that extent we have to be cautious about saying that ancient Chinese and Japanese rules were the same. There is only tiny and disputed evidence that Japan used the four starting stones that are characteristic of Chinese go, so it may be that the Japanese inherited or devised a variant of old go rules.

Nevertheless, the look and feel of the two sets of rules is the same. As testimony of that, let us briefly look at some low-level pointers and anecdotes.

The pointers first:

As to the anecdotes, the most famous, though somewhat ludicrous, is the tale of the Japanese envoy Kibi no Makibi. He is credited with taking knowledge of go back to Japan in 735, but it was there much earlier. He features in a story in the encyclopaedia Gushi Leiyuan (Garden of Ancient Things), in the Games section. When he played go with a Chinese player and the game was about to end in a tie, Kibi stole one of the Chinese player's black stones and swallowed it. When they counted up, the Chinese side had lost by one point. Suspicious, they consulted a diviner and learned the stone had been stolen and swallowed. They therefore gave Kibi a laxative, but the Japanese side had him swallow an anti-diarrhoea medicine, and in the end they won.

A more sensible version on the same theme, which also mentions an otherwise unknown rule for handicap play, features Jia Xuan, the late-Tang player already mentioned. The Gujin Tushu Jicheng (Collection of Old and Modern Books) relates that the Tang emperor Tai Zong was playing Jia Xuan, who kept losing by one point. Tai Zong knew that Jia was losing deliberately, and so he said to him, "If you lose again I shall punish you!" The game eventually ended in a tie. Tai Zong said, "This is not right. Let us play again. If you win I will give you a scarlet robe. If you lose. you will be punished and thrown in the mud." The game again ended in a tie. The Emperor said, "I gave you a handicap. If it is a tie, you lose." So he ordered his servants to throw Jia in the mud. But Jia then exclaimed, "I still have one prisoner in my hand!" Tai Zong laughed and gave him the robe.

There is one other niggling point, however, and that bring us back eventually to Chen Zuyuan. It is to do with the use of "altogether" and "each". On the face of it, this seems like nothing more than elegant variation. After all, the other parts of the counting formulas also appear in slightly variant forms.

Still, there is just the possibility that the more modern (in C&IP's terms) "each" may represent some sort of rule to do with equal number of plays, and there is the anomaly that the Gold-Petalled Bowls diagram says 122 moves each, but shows only 243 moves.

That is just a niggle. However, it becomes a festering sore when modern go publishers get their hands on C&IP. That is the theme of Chen's article. He looks at five editions published between 1985 and 2004. He is rightly annoyed that editors have made not just incorrect changes but also unnecessary ones in what purport to be facsimile versions. So where the original says "145 moves each" an editor may change that (without comment) to 290 moves total. It may seem of no consequence, but, as Chen points out, if some evidence is later unearthed to do with, say, playing equal number of moves, the terminology used can matter greatly.

Chen also shows how different versions of the Four Immortals of Chengdu appear in the various editions. The source of this contamination is that the position shown in the original C&IP leaves a point still to play for, and the original now has a reference that Black won by 1. Some editors have creatively re-crafted the final few endgame moves (as ever, without comment) to make the position fit the score. But is this wise? Chen's argument is that the correct procedure is to reproduce the original, warts and all. In this case, it seems safe to assume that the Song original was in error. The details are left to Chen's article, but an essential point is that we luckily have the corroborative evidence of Xu Zongyan, who tells us White won by 1 point.

Although Chen is right to insist that it is necessary to go back to the original, that is not without its problems. Access is one problem. There is only one copy and it is in the Rare Books section of Beijing National Library. But there is a more subtle problem. The original has been altered.

One alteration is the addition of a move 244 on the Gold-petalled Bowls diagram. Also, the "Black wins by 1" has been brushed on to the Four Immortals from Chengdu diagram (instead of White winning by 1).

Now although Chen has spotted this, he does not make it clear whether he is also aware of a photographic copy of C&IP made in 1916 by Xu Nanchang of Nanling. In that version, there is no "244" and no "Black wins by 1", confirming his suspicions.

As to how and when the book was altered, we can only guess from its history. That tale highlights another egregious error spotted by Chen.

The book, being by Imperial request, presumably only existed in a few copies anyway. It appears to have become virtually unknown as early as the Ming dynasty (which began in 1368). An official named Gu Shan, courtesy name Tingyu, built a library in Huayong-qiao, Wuxian County (in Jiangsu Province, near Suzhou City). Gu stored the book there in the Qing Jiaqing period (1796~1821), but it was later acquired by the famous bibliophile Huang Pilie who regarded it as such a treasure that he specially made a postscript for the book. By the end of the Qing period, the book had already passed through many hands before it reached the hands of Qu Liangshi of Changchu. Qu (1873~1940) was the fourth owner of his own library (now a famous state-owned library in Changchu City).

During the Communist insurgency, Qu took part of his library to Shanghai. Part of it was bought by the famous Buddhist scholar and bibliophile Ding Fubao (1874~1952). This part included C&IP. After the Liberation, Ding Fubao presented the book to the state.

One of the modern editions claims that the book also went through the hands of a famous Qing scholar called He Shaoqi, on the basis that the book bears a stamp saying "Private property of Shaoqi." However, as Chen has discovered, Qu Liangshi's paternal great-grandfather, first owner of their library, was called Qu Shaoqi, and so he is obviously the owner of the stamp.

There are other serious errors in the modern Chinese editions spotted by Chen. He seems unaware, but surely would not be surprised, that the Japanese editions are just as spotty. The Kodansha edition of 2004, for example, with comments by Go Seigen, reproduces move "244" and so on, though there is some excuse in ths case. When the Japan-China goodwill exchanges started, the Chinese gave the Nihon Ki-in a facsimile of the original book. Unfortunately, the later interpolations showed up, too.

Naturally, none of the evidence provided by C&IP offers any clues as to when modern Chinese rules came into force. The most authoritative view on this is still the article by Yang Liansheng ("Some Thoughts on Changes in the Chinese Method of Final Counting in Go") which appeared as a translation from his Chinese, written in 1957, by Go Seigen in Kido, Vol. 36, October 1960, page 54.

Yang was a professor at Princeton, and friend of Go's family. He was a strong amateur at go. He said, "The period when the method of counting started to change in China was, in my view, more or less in the early Ming period, or perhaps a little earlier. By the late years of the Ming Wanli era (1573~1620), the present Chinese method had already become the norm."

Yang mentions the well-known case of the gambling game between Bai Laichuang and Chang Shijie in Chapter 54 of the novel Jin-Ping-Mei (Golden Lotus) of 1617 in which both sides fill in the neutral points then count. Bai's stones were cut at five places and Cheng's at two. Bai's cutting penalty (group tax) meant he had to take off three "stones", and he lost by the amount of the tax. This seems to be the earliest direct reference to group tax, but also the references to filling in dame and to counting in stones (zi) rather than points (1 zi = 2 points) strongly suggest the modern Chinese method of scoring.

Yang also mentions a book, Riben Xiao (Japanese ideas), which describes the method of counting then used in Japanese go. According to this, the rules of play were generally the same, but the method of counting differed. The captured stones were all collected in the hand and used to fill in the opponent's territory. If both sides had the same total number of points, it was a tie, otherwise the player with the fewer points lost.

This book (also known as Riben Difeng, Japanese customs), was published in 1592~1593. It seems that the Chinese author had no memory of Chinese go being played the Japanese way. That pushes the date of the completed switchover, according to Yang, to around the middle of Ming (say, 1500).

As to why there was such a drastic change, Yang is silent. But Go Seigen has ventured an opinion: "In my view there are two reasons why the method of counting in China changed from empty points to stones. One is that captured prisoners are not necessary and there is no risk that they will be stolen. The other is that it is possible to resolve rationally special problems such as bent four in the corner. The official rules (promulgated in 1949) of the Nihon Kiin in which bent four in the corner is simply adjudged unconditionally dead cannot be called rational."

It seems highly unlikely that Chinese players in the Ming were truly bothered by bent four in the corner. but the reference to the risk of stones being stolen is surely pertinent. As we see from the Golden Lotus, go was a gambling game. A characteristic of the Ming period was a broadening of popular culture. Go was being played by people below the top classes of officials (the same thing was happening at the same time in Japan). It is easy to imagine that, to players convinced they had lost money as well as the game because stones had mysteriously disappeared, the appearance of a foolproof method of counting would seem far better even than a new mousetrap. And with money involved, it is also easy to accept that the new method would take over in a flash.

But was gambling at go extensive enough to warrant that? Even without statistics, that seems easy to answer affirmatively.

A little book has survived from the Wanli period of Ming times. It is the earliest known book of rules.

An official called Wang Siren from Shaanxi Province was a great jester. His book, Yi Lü (Go Rules), was a jokey-serious collection of regulations which specified punishments for infractions of go manners of the time. It was obviously meant for his local "club" but the manners he describes must have been widespread.

Taking as an example of a major punishment:

Telling lies by saying that a group will die, feigning joy or sadness, feigning surprise or entreaty, or disturbing the other person's mood is punished by 50 light lashes. This type of trickery is about aiming to cause the opponent to make a mistake by thinking that there is no other move here so that he plays elsewhere. Lying by saying that "that group will die" is referring to a group as dead when it is not clearly so and is a trick to make him make the last move elsewhere. Feigning joy refers to saying "I have won, I have won" when the game is not yet over, and is a trick to cause the opponent to lose his attention on the game. Feigning sadness is showing a loss of fighting spirit by saying "that is hopeless" in a game one has not yet lost, and is a trick to reduce the opponent's attention on the game. Feigning surprise is suddenly noticing something and exclaiming "Ah, so!" and is a trick to make the opponent treat one lightly. Feigning entreaty is saying thinks like "I don't want to play there but..." and is a trick to create unreasonably in the mind of the opponent a feeling of compassion, and is a trick to make the opponent play a slack move. To use all these tricks requires effort, but to use them is unsophisticated and vulgar. Nevertheless there is a big difference between this sort of deception and stealing the opponent's pieces by force, so the punishment is only 50 light lashes.

The crime of "moving the pieces" or taking advantage of the dark to move pieces on the board is punished by 60 heavy lashes, with the same penalty for "moving" by hiding. Removing pieces, keeping them and ad-libbing about it gets a punishment of 80 lashes and exile [not being allowed to play] for 2 years. If you move pieces hither and thither means you are shifting them from the original point. In the case of hiding and moving it means keeping them in a sleeve or hand, but to that extent it is still a light offence. However, removing pieces and keeping them and ad-libbing about it means placing pieces where there was no expectation to be any or placing a piece that one expected to be in the east in the west. This is a grave offence and is no different from being a robber.

As we can see, even assuming modern Chinese rules were in force then, stones could still disappear. But it seems the reference is to stones on the board, and presumably it was easier to keep track of those than captured stones taken off the board.

In night games 500 years ago under unreliable lighting, fingers no doubt became more dextrous and itchy. But sneaky behaviour did not seem anything like as reprehensible to Wang as physical force.

In a different example, he says removing a piece without the opponent noticing is a lesser crime than removing it in front of the opponent's eyes, even if we are talking about the same piece. In other words, sneak thievery is less reprehensible than using force. Among the worst three sins, are:

Seizing the other person's pieces in broad daylight = 90 heavy lashes, corvée for 2.5 years. However, a person who strongly repents later: 70 lashes. For a person who repents later and is overwhelmed by remorse the sentence is reduced to 50 lashes. In broad daylight also includes playing under candle light, Seizing pieces means suddenly snatching a piece from the grasp of the opponent when he is unsure and has not yet removed his finger even though he has placed the piece down, and placing it elsewhere, or when he has not placed it down and is still holding it.

Assuming this behaviour was genuine, the typical merciless Ming club makes the go servers sound like Sunday school.

Obviously these punishments are not for real. They can all be commuted to payment of a fine. In other words, the Ming equivalent of a swear box. Still, it seems pretty clear that go manners of the time did create an environment where a new method of scoring was urgently needed.

Assuming that is what really happened, it is possible to believe that a new set of international rules could be created, and adopted just as swiftly, even today - if the money is right. Tournament directors may also eye Wang Siren's rules enviously.



Unicode fonts are required to read the Chinese in this article. Many other details of ancient Chinese go and of the personalities mentioned here are on the GoGoD CD, as are the games in C&IP.

For those you would like to study the impact of sekis containing points on group-tax counting, the database games 1660~a, 1700JQXG232, 1730CLP2-25 and 1740hcem001 will provide a starting point.

This is a page from GoGoD's New In Go. If you have come to this page from an outside link and the index panel is not shown, click here to view.