Your First Book of Go

Go is an ancient game. It was invented by the Chinese more than 4000 years ago. It had already evolved into its modern form 2500 years ago, and it spread first into Korea and then into Japan. The Japanese are the modern authorities on the game, they claim the best players, and their language supplies many of the words and concepts used in the normal play of Go.

The elements of Go can be divided into four parts: The board, the pieces, the rules of play, and the object of play. These simple parts, however, evolve into probably the most complex and subtle game ever invented by human beings. If you like any two person game at all, based on mental skill, you may find Go to be more than a little intriguing.

A standard Go board consists of 19 vertical lines intersected by 19 horizontal lines. Several non-standard sizes of Go boards exist, including the more popular 13 x 13 board, and the less popular 9 x 9, 11 x 11, 15 x 15, and 17 x 17 sizes. The points of play on a Go board are not the squares formed by the intersecting lines, they are the intersections themselves. Thus, a 19 x 19 line board has 361 intersections. Sometimes an intersection is called a point. There are three different “kinds” of points. The four corner points each have only two points adjacent to them. The points along the sides of the board are connected to only three adjacent points. However, all the remaining points on the board are connected to four adjacent points. These differences, and the fact that diagonal points are not considered to be adjacent, are very important to play.

Play is conducted with a set of 181 black stones owned by one player, and a set of 180 white stones owned by the opposing player. The player playing the black stones always plays first. With two exceptions, play may be onto any unoccupied point on the board. Play alternates between the black and white stones, with one stone being placed on the board at each play. A player may pass instead of playing a stone. When both players have passed in succession, the game is over.

The object of play is to surround points of territory and capture stones of the opposing color. Each intersection securely surrounded, and each stone captured, are worth one point. The winner is the player with the most points at the end of the game. The final score is the point difference between the players.

Two concepts are necessary to understand the two exceptions mentioned above, and how to accomplish the object of play. The first concept is that of a “group” of stones. A group is one or more stones that occupy adjacent points on the board. A group lives or dies as a unit. When a stone is placed on the board, it may be a group of one, detached from all the other stones of its color, or it may be attached to a group that already exists, extending that group. An unoccupied point adjacent to a group is called a “liberty.” Every group must have one or more liberties, or the opponent (having just made a play that deprives a group of its last liberty) gets to remove the entire group from the board. Each stone of the group becomes a prisoner of that player.

Now for the first exception: It is illegal to make a play that leaves a group of your stones without any liberties. However, since prisoners are removed before a play is complete, any play that takes a prisoner is bound to open up a liberty for the stone just played, and for any group to which it is attached.

The second exception involves a play that gives your opponent the same position twice in a row. There is only one way this can occur in Go, and it is called “Ko.” The position arises when one player captures a single stone, and the stone captured provides the new stone with its only liberty. If you were to play back into the intersection just vacated by your captured stone, capturing the stone just played by your opponent, it would leave your opponent with the original position, and play could alternate like this forever. Instead, you must play elsewhere. You may only retake the Ko if it remains after your opponent’s next move. If you are able to retake it, your opponent must then find another move before retaking. If three Ko sites happen to develop, white could play at the first, black at the second, white at the third, then black could retake at the first, white at the second, and black at the third. This could repeat forever and the game would be a stalemate. Fortunately, three Ko sites, together with a mutual desire for a stalemate, are a very rare occurrence in Go.

Consider a ring of stones surrounding a single point of territory. If enemy stones took all the liberties around the outside of this group, the only point remaining would be that single point inside. Now, the enemy could take that final liberty and capture the entire group. But, consider a group of stones that forms a figure eight, surrounding two separate points of territory. Now, the enemy can still surround the group and take all of the outside liberties, but it can’t perform the final “kill.” A play into either of the final points is illegal, since it does not kill the group and open a liberty for the stone being played.

This brings us to the concept of an “eye.” An eye is one or more points completely surrounded by stones of a single color. An eye of two or more points is called a “big eye.” A group with two separated eyes cannot be killed. A group that, despite opposition, can form two separate eyes from a big eye cannot be killed. A group that can be connected (again, despite opposition) to either of these types of live groups cannot be killed. Thus, the first objective of a player is to have a plan for each stone played as to how that stone will eventually become part of a live group (or how its sacrifice helps a larger group of stones to live). The second objective is to include as many points as you are able into the eyes of the groups that you form. If the groups live, these points constitute your territory at the end of the game.

Two additional concepts arise in a “real” game of Go. Japanese words are used to name them. When a player makes a play, it is considered “Sente” (sen-tay) if it requires (or forces) a response from the opponent. A play that does not require a response is called a “Gote” (go-tay) play. It is important to know the difference, and it is generally a stronger line of play to “retain Sente.” A position that invites a very strong Sente play is sometimes called a “Ko threat.” Let’s say your opponent takes a stone in Ko (often threatening more stones on the next play). You would like to take back immediately, but you can’t. If you have a Ko threat worth a sufficient number of points, however, you can play there and force an immediate response. Now, you can retake the original Ko and force your opponent to find a Ko threat. Ultimately, the player with the most threats can win the “Ko fight.”

The game of Go usually begins with single stones being played on the third or fourth lines in from the corners first, and then in from the sides. At some point, one of the players will play very close to an opponent’s stone, or “attach” to it, playing into one of its liberties. During the middle of the game, play alternates between heavy play (stones placed close together), and light play (stones played several intersections away from the nearest stone). This phase involves life and death struggles of large groups. The end game is reached when each Sente play threatens only a point or two, and each Gote play defends only a point or two.

At some point further play gains nothing, and finally both players will pass. Any groups of stones, encircled by enemy stones, that do not have two eyes or the potential to make them from a big eye, are considered dead and are removed as prisoners. Sometimes opposing groups of stones are interlocked in such a way that neither can dominate the other or be killed by them. Typically, they will share one or a very few liberties with each other, but neither will have the two eyes necessary for life. The entire collection should be considered as neutral territory. Open territory touched by stones of both colors is also neutral.

When both players have passed and the dead groups have been removed as prisoners, both players put their prisoners back into their opponent’s territory. Prisoners may be left over, or territory may remain. Usually the borders of the remaining territory are rearranged so that rectangles are formed. This makes counting the territory easier. The outcome of the game is the sum of your territory and the prisoners you have captured, minus the sum of your opponent’s territory and prisoners. The process of putting your prisoners into your opponent’s territory and rearranging territory into rectangles doesn’t affect the result, it simply makes it easier to calculate.

Further Numination on this subject will require you to obtain a Go set and an opponent. If you cross this threshold, you will find a whole new space in which you can Numinate. And, now you are ready for your first game of Go – and your “Second Book” of Go.