Go (board game)

From Citizendium
Jump to navigation Jump to search
This article is developing and not approved.
Main Article
Related Articles  [?]
Bibliography  [?]
External Links  [?]
Citable Version  [?]
This editable Main Article is under development and subject to a disclaimer.
(CC) Photo: S. Venga
A modern Go board

Go is a board game played by two players. It is called Weiqi (圍棋; 围棋) in Chinese, Baduk (바둑) in Korean, and Igo or Go (囲碁; 碁) in Japanese.

Go is the world's oldest game that is still played in its original form, with a documented history of over 2,500 years. It originated in China, is a traditional game throughout East Asia, is considered the national game of Japan [1], and is now played around the world. It is the second most played board game in the world, after Chinese chess.

Go reached the West via Japan, so in English the name of the game and many of the words used to describe positions or strategic concepts are borrowed from Japanese. We indicate such words by putting them in italics the first time they are used below.

There is a large Wiki devoted to discussing the game, Sensei's Library.

[edit intro]


Go is played on a flat board (a goban) with a grid of 19x19 intersections. Two sets of stones, white and black, are used. The game is played in turns and unlike Chess, black makes the first move in go. Each stone is placed on an intersection and the goal is to capture more territory than the opponent. In go, it often matters whether a given move is beautiful or produces good shape.

It is possible to play essentially the same game on a board of size different from the standard 19-line board. 13-line and 9-line boards also moderately common, mainly used to get a quicker game or to keep things relatively simple for teaching. Each of these sizes roughly doubles the number of intersections: 92=81, 132=169, 192=361. Historically, 15-line boards were once normal in China and 17-line in China, Tibet and Japan. There has been some experimentation with 21 and 23-line boards.


There are a few different Rule Sets for playing Go. They are very similar to each other and for most games the different rules give similar results.

The core rules are:

  1. Each turn a player places a stone on one empty intersection. Afterwards it's the other player's turn.
  2. A stone is a member of a "chain" of stones if one stone is connected to another stone of the group with the same color via a single line (i.e. they are on orthogonally adjacent points; there are no direct diagonal connections).
  3. To remain on the board, at least one stone in a chain must have a connection via a line to an empty intersection (called a liberty). Otherwise, that chain is dead and has to be removed from the board, and the stones become prisoners.
  4. You aren't allowed to make a move that directly reverses the move of your opponent — the opponent has just captured just one of your stones and you want to immediately replay that stone and capture only the stone he just played. This is called ko. In some variants you aren't allowed to return to any previous position, which could be done by means of three kos in turn. More usually, if both players insist on doing this, the game is annulled.
  5. You also aren't allowed to fill in the last liberty of a chain of your own, unless doing so also fills in the last liberty of an opposing chain, in which case you get priority and capture. (The complexity of the game can result in situations where you might want to "commit suicide", but this isn't allowed.) This rule is omitted in some variants.
  6. Either player may pass, skip their move. If both players pass in the same position, the game is over. All stones that could be captured through adding additional stones are dead and get removed from the board and become prisoners.

While the above rules are sufficient for most games, specific rules that exactly define the terms that are used in the rules can be a bit longer.[1]

Also see Go rules.

Scoring and handicaps

There are two systems of scoring, Japanese and Chinese. The Japanese method is generally used in Western countries (except New Zealand). Both systems count territory — the number of empty intersections that a player's stones surround — plus something else, but the "something else" is different. Japanese scoring counts territory plus the number of enemy stones captured. Chinese scoring counts territory plus the number of your stones on the board. Since capturing some number of stones reduces the stones the opponent has on the board by exactly that number, the two systems almost always give the same result. In either system, the winner is the player with the higher score.

However, a seki, an area where neither player can move because the opponent could then capture, is scored differently in the two systems. In Japanese scoring, a seki does not count for either side since it is not territory and no stones have been captured. In Chinese scoring, both sides count the stones they have on the board in the seki. If one player has more stones involved in the seki than his opponent, the scores will be different in the two systems. It is possible, though extremely rare, to have a game where either player can be scored as the winner depending which system is used.

Go has a widely used and fairly convenient handicap system. Players of more-or-less equal strength start with an empty board, black plays first, and white gets a small bonus in the scoring, called komi, to compensate for that first move advantage. Komi is normally 5.5 points if Japanese scoring is used. For unequal players, some black stones are placed on the board before play begins, giving black a substantial advantage. White then plays first and komi is half a point to make a drawn score impossible. On a 19-line board, handicaps from two to nine stones are normal and up to 17 is possible. On a 13-line board, five is the usual maximum; this gives roughly the same advantage as nine on a 19-line board. The board has markings for the hoshi points where the handicap stones are placed.


Go has a system of ranks similar to those used for martial arts. A number of different regional groups — at least Chinese, Japanese, Korean, American, and European — have ranking systems and the major sites for online play, Internet Go Server and Kiseido, each have their own. All these systems are roughly comparable, but no two are exactly the same. It is moderately common to see a player whose ranks in two systems are a few levels different.

A beginner is ranked as some large number of kyu, 25 or 35 in different systems, and the number decreases as strength increases. In martial arts, the kyu ranks wear colored belts. Above 1 kyu comes 1 dan, shodan, corresponding to a black belt in martial arts. There are amateur ranks up to 6 or 7 dan.

The appropriate handicap can be calculated from the ranks. For amateur players, ranks are about one stone apart in strength so a 2 dan would take 5 stones from a 7 dan and would give a 1 kyu 2 stones; both games would then be approximately even. Except for teaching, a 2 dan would not usually play against a 20 kyu since no possible handicap would make an even game of that.

Professional players in Japan, China and Korea each have their own separate ranking system from 1 dan to 9 dan. However these ranks don't accurately their playing strength as it takes many years for a young professional player to reach the 9 dan even when he has the same playing strength as other 9 dans. The difference in playing strength between the average professional 1 dan and the average 9 dan is also a lot less than one handicap stone.

The overall range is enormous. A top pro wins easily against a 1 dan amateur at 9 stones; the 1 dan wins easily against a 10 kyu at 9 stones, and the 10 kyu wins easily against a beginner at 13.


There is no exact date for the invention of Go. One legend dates the invention to the Emperor Yuo who taught the game to his eldest son Dan Zhu. Most modern writers think, that this legend (and a few similar legends), were written down in the Han period, to make the game appear older than it really is. They date the invention to the period of 1000-400 BC.[2]

Comparison to chess

Go is conceptually simpler than chess (especially when go rules are properly formulated):

  • A go player has only one kind of pieces, called stones. A chess player has six kinds of pieces (king, queen, rook, bishop, knight, pawn, and one of her/his bishops runs on white squares, while the other one on the black--thus they are actually different too).
  • A go player makes only one kind of moves, namely setting a stone on an intersection point (the effect may be different each time, causing sometimes a group of opponent stones to be removed). A chess pawn has four kind of moves: 1.going one step forward, 2.going two steps forward from its initial position, 3.capturing an opponent's piece one step askew from it and landing where the opponent's piece was, 4. capturing opponent's pawn en passant. On the top of it, the pawn, which reaches the last row gets promoted.
  • Go has essentially only one (very natural) restriction on moves: a move which would lead to a repetition of position is illegal (not allowed). It has also another rule, which disallows a suicide, but it's only a cultural rule, not essential to the game. In chess too we have a cultural rule which disallows to put your own king in check. In addition, we have also several essential rules which contribute to the total chess rules complexity: castling, en passant, promotion and the rules about draw by repetition or by making 50 moves by both sides without any capturing and without any pawn move, or by stalemating your opponent.

Chess as a whole does not admit any natural, regular generalizations onto larger boards (but many chess endings do). In the case of go, one may play the game on the square boards of arbitrary n by n size. More than that, one may play go on rectangular boards or even on arbitrary finite simple graphs. Thus go is so mathematical that it provides a graph invariant: with arbitrary graph we may associate the result of the game played optimally by both players (mathematical theory of games states that such optimal strategies exist; it's a corollary to the respective Zermelo's theorem, 1913). There has been some work on computer programs that exhaustively analyse go on small boards; the 5 by 5 board has been solved [2].

The handicap system is another important difference. It is much easier for two players of significantly different strength to have an interesting game — one in which both have to work at it and the outcome is often in doubt — in go than in chess. Of course, there are limits to this; nothing will make a game between a master and a beginner interesting for the master, but even a master may have trouble against a strong amateur with a large handicap.

There is a saying that "Chess is a battle; Go is a war." The goal is not just to take out the enemy king, but to control more of the board than the opponent. Expert Go players take the overall board position into account in choosing almost every move.

Go is one of the most complex games in the world, far outweighing games such as chess in the number of possible game positions. Indeed, computer chess programs, by year 2000, became as strong or even stronger that the strongest (human) chess players. It took longer to achieve this for go.

A trivial difference is that in Western games such as chess and checkers, the pieces go on the squares of the board, while in many Far Eastern games such as Go, Gomoku and Chinese chess, they go on the intersections of the lines.

Cultural Dimensions

Go strategy is also studied as an metaphor for Asian strategy compared to western strategy. [3][4]

It is interesting to note that the US engaged the Soviets on multiple "fronts" during the Cold War with buffer states (eg. South Korea as buffer state and Japan as main ally), whereas today "China is in a crescent-shaped ring of encirclement" by US allies.[5][6]

Neuroscience Research

Neuroimaging research suggests players use the right hemisphere of their brain while playing Go, responsible for position and orientation, slightly more than the left side. The game requires a more global strategy than chess, and thus differs from the more left hemisphere activated activity of playing chess. A study of six subjects suggests successful Go playing requires practice and experience rather than sheer intellectual ability.[7] Another study suggests the ability of professional Go players to assess visual information and nonmotor learning memory processes is mediated in part by the cerebellum and precuneus.[8]


  1. Rules sets compiled and analyzed by Robert Jasiek
  2. The Game of Go Speculations on its Origins and Symbolism in Ancient China By Peter Shotwell
  3. Scott Boorman (1969). The Protracted Game: A Wei-Ch'i Interpretation of Maoist Revolutionary Strategy. Oxford University Press. 
  4. Learning from the Stones: A Go Approach to Mastering China's Strategic Concept, Shi, by Dr. David Lai
  5. Hara, 2001, pp. 371
  6. "NTI: Global Security Newswire - U.S. Wants to Surround China With Missile Defenses, Chinese Experts Claim." NTI - Global Security Newswire. 23 Feb. 2010. Web. 26 Mar. 2011. <http://www.globalsecuritynewswire.org/gsn/nw_20100223_8151.php>.
  7. Chen X, Zhang D, Zhang X, et al (2003). "A functional MRI study of high-level cognition. II. The game of GO". Brain Res Cogn Brain Res 16 (1): 32–7. PMID 12589886[e]
  8. Ouchi Y, Kanno T, Yoshikawa E, et al (2005). "Neural substrates in judgment process while playing go: a comparison of amateurs with professionals". Brain Res Cogn Brain Res 23 (2-3): 164–70. DOI:10.1016/j.cogbrainres.2004.10.004. PMID 15820625. Research Blogging.