The Game of Go/Chapter III

by Arthur Smith

III

RULES OF PLAY

The players play alternately, and the weaker player has the black stones and plays first, unless a handicap has been given, in which case the player using the white stones has the first move. (In the olden times this was just reversed.) They place the stones on the vacant points of intersection on the board, or “Me,” and they may place them wherever they please, with the single exception of the case called “Ko,” which will be hereafter explained. When the stones are once played they are never moved again.

The object of the game of Go is to secure territory. Just as the object of the game of Chess is not to capture pieces, but to checkmate the adverse King, so in Go the ultimate object is not to capture the adversary’s stones, but to so arrange matters that at the end of the game a player’s stones will surround as much vacant space as possible. At the end of the game, however, before the amount of vacant space is calculated, the stones that have been taken are used to fill up the vacant spaces claimed by the adversary; that is to say, the captured black stones are used to fill up the spaces surrounded by the player having the white pieces, and vice versa, and the player who has the greatest amount of territory after the captured stones are used in this way, is the winner of the game. However, if the players, fearing each other, merely fence in parts of the board without regard to each other’s play, a most uninteresting game results, and the Japanese call this by the contemptuous epithet “Ji dori go,” or “ground taking Go.” I have noticed that beginners in this country sometimes start to play in this way, and it is one of the many ways by which the play of a mere novice may be recognized. The best games arise when the players in their efforts to secure territory attack each other’s stones or groups of stones, and we therefore must know how a stone can be taken.

A stone is taken when it is surrounded on four opposite sides as shown in Plate 2, Diagram i. When it is taken it is removed from the board. It is not necessary that a stone should also be surrounded diagonally, which would make eight stones necessary in order to take one; neither do four stones placed on the adjacent diagonal intersections cause a stone to be taken: they do not directly attack the stone in the center at all. Plate 2, Diagram iv, shows this situation.

A stone which is placed on the edge of the board may be surrounded and captured by three stones, as shown in Plate 2, Diagram ii, and if a stone is placed in the extreme corner of the board, it may be surrounded and taken by two stones, as shown in Plate 2, Diagram iii.

In actual practice it seldom or never happens that a stone or group of stones is surrounded by the minimum number requisite under the rule, for in that case the player whose stones were threatened could generally manage to break through his adversary’s line. It is almost always necessary to add helping stones to those that are strictly necessary in completing the capture. Plate 2, Diagram v, shows four stones which are surrounded with the minimum number of stones. Plate 2, Diagram vi, shows the same group with a couple of helping stones added, which would probably be found necessary in actual play.

It follows from this rule that stones which are on the same line parallel with the edges of the board are connected, and support each other, Plate 2, Diagram vii, while stones which are on the same diagonal line are not connected, and do not support each other, Plate 2, Diagram viii. In order to surround stones which are on the same line, and therefore connected, it is necessary to surround them all in order to take them, while stones which are arranged on a diagonal line, and therefore unconnected, may be taken one at a time. On Plate 2, Diagram iii, if there were a stone placed at S 18, it would not be connected with the stone in the corner, and would not help it in any way. On the other hand, as has been said, it is not necessary to place a white stone on that point in order to complete the capture of the stone in the corner.

In order to capture a group or chain of stones containing vacant space, it must be completely surrounded inside and out; for instance, the black group shown on Plate 2, Diagram ix, while it has no hope of life if it is White’s play is nevertheless not completely surrounded. In order to surround it, it is necessary to play on the three vacant intersections at M 11, N 11, and O 11. The same group of stones is shown in Diagram x completely surrounded. (It may be said in passing that White must play at N 11 first or the black stones can defend themselves; we shall understand this better in a moment.)

In practice it often happens that a stone or group of

Plate 2

stones is regarded as dead before it is completely surrounded, because when the situation is observed to be hopeless the losing player abandons it, and addresses his energies to some other part of the board. It is advantageous for the losing player to abandon such a group as soon as possible, for, if he continues to add to the group, he loses not only the territory but the added stones also. If the circumstances are such that his opponent has to reply to his moves in the hopeless territory, the loss is not so great, as the opponent is meanwhile filling up spaces which would otherwise be vacant, and against an inferior player there is a chance of the adversary making a slip and allowing the threatened stones to save themselves. If, however, the situation is so clearly hopeless that the adversary is not replying move for move, then every stone added to such a group means a loss of two points.

At the end of the game such abandoned groups of stones are removed from the board just as if they had been completely surrounded and killed, and it is not necessary for the player having the advantage actually to surround and kill such a group. It is enough if they obviously can be killed. The theory on which this rule proceeds is that if the players play alternately, no advantage would be gained by either side in the process of actually surrounding such a group, and its completion would only be a waste of time. But let us suppose that a black group at the end of the game is found to be hopeless and also completely surrounded with the exception of one point. The question arises, can the Black player demand that his adversary play on the vacant space in order to kill this group, for, if he could, it is obvious he would gain one “Me” by so doing. The answer is, he cannot so demand, and his adversary is not bound to play on this point, and the hopeless or abandoned stones are removed without further play. We might call such groups “dead.” They may be distinguished from stones that are “taken,” because these latter are removed at once, whereas “dead” stones are removed only at the end of the game.

As a corollary to the rule for surrounding and taking stones, it follows that a group of stones containing two disconnected vacant intersections or “Me” cannot be taken. This is not a separate rule. It follows necessarily from the method by which stones are taken. Nevertheless in practice it is the most important principle in the game.

In order to understand the rule or principle of the two “Me,” we must first look at the situation shown in Plate 3, Diagram i. There, if a black stone is played at F 15, although it is played on an intersection entirely surrounded by white stones, it nevertheless lives because the moment it is played it has the effect of killing the entire white group; that is to say, a stone may be played on an intersection where it is completely surrounded if as it is played it has the effect of completely surrounding the adversary’s stones already on the board. If, on the other hand, we have a situation as shown in Plate 3, Diagram ii, a black stone may indeed be played on one of the vacant intersections, but when it is so played the white group is not completely surrounded, because there still remains one space yet to be filled, and the black stone itself is dead as soon as it touches the board, and hence it would be impossible to surround this group of white stones unless two stones were played at once. The white stones, therefore, can never be surrounded, and form an impregnable position.

This is the principle of the two "Me," and when a player's group of stones is hard pressed, and his adversary is trying to surround them, if he can so place the stones that two disconnected complete "Me" are left, they are safe forever. It makes no difference whether the vacant "Me" are on the edges or in the corners of the board, or how far from each other they may be.

Plate 3, Diagram vi, shows a group of stones containing two vacant "Me" on the edge of the board. This group is perfectly safe against attack. A beginner might ask why the white group shown on Plate 3, Diagram v, is not safe. The difficulty with that group is, that when Black has played at S 9, there are no "Me" in it at all as the word is used in this connection, not even a "Kageme" as shown in Plate 3, Diagram iii, because a "Me," in order to be available for the purpose of defense, must be a vacant intersection that is surrounded on four sides, just as a captured stone must be surrounded and therefore on the sides of the board it can be made by three stones, and in the corner of the board by two stones, but it is absolutely necessary, in addition to the minimum number of surrounding stones, to have helping stones to guard the surrounding stones against attack. This brings us to what the Japanese call "Kageme."

In actual play there are many groups of stones that at first glance seem to have two vacant "Me" in them, but which on analysis, will be found vulnerable to attack. A "Me" that looks somewhat as if it were complete, but is, nevertheless, destructible is called "Kageme." "Kage"

Plate 3

means "chipped" or "incomplete." Plate 3, Diagram iii, is an illustration of this. A beginner might think that the white group was safe, but Black can kill the upper six white stones by playing at E 3, and then on the next move can kill the remainder by playing at G 2. Therefore, E 3 is not a perfect "Me," but is "Kageme." G 2 is a perfect "Me," but one is not enough to save the group. In this group if the stone at F 4 or D 2 were white, there would be two perfect "Me," and the group would be safe. In a close game beginners often find it difficult to distinguish between a perfect "Me" and "Kageme."

Groups of stones which contain vacant spaces, can be lost or saved according as two disconnected "Me" can or cannot be formed in those spaces, and the most interesting play in the game occurs along the sides and especially in the corners of the board in attempting to form or attempting to prevent the formation of these "Me." The attacking player often plays into the vacant space and sacrifices several stones with the ultimate object of reducing the space to one "Me"; and, on the other hand, the defending player by selecting a fortunate intersection may make it impossible for the stones to be killed. There is opportunity for marvelous ingenuity in the attack and defense of these positions. A simple example of defense is shown in Plate 3, Diagram iv, where, if it is White's turn, and he plays in the corner of the board at T 19, he can save his stones. If, on the other hand, he plays anywhere else, the two "Me" can never be formed. The beginner would do well to work out this situation for himself.

The series of diagrams commencing at Plate 3, Diagram v, show the theoretical method of reducing vacant spaces by the sacrifice of stones. This series is taken from Korschelt, and the position as it arose in actual play is shown on Plate 10, depicting a complete game. In Plate 3, Diagram v, the white group is shown externally surrounded, and the black stone has just been played at S 9, rendering the group hopeless. The same group is shown on the opposite side of the board at Plate 4, Diagram i, but Black has added three more stones and could kill the white group on the next move. Therefore, White plays at A 12, and the situation shown in Plate 4, Diagram ii, arises, where the same group is shown on the lower edge of the board. Now, if it were White's move, he could save his group by playing at J 2, and the situation which would then arise is shown on Plate 4, Diagram iii, where White has three perfect "Me," one more than enough. However, it is not White's move, and Black plays on the coveted intersection, and then adds two more stones until the situation shown in Plate 4, Diagram iv, arises. Then White must again play at S 8 in order to save his stones from immediate capture, and the situation shown at Plate 5, Diagram i, comes about. Black again plays at J 18, adds one more stone, and we have the situation shown in Plate 5, Diagram ii, where it is obvious that White must play at C 11 in order to save his group from immediate capture, thus leaving only two vacant spaces. It is unnecessary to continue the analysis further, but at the risk of explaining what is apparent, it might be pointed out that Black would play on one of these vacant spaces, and if White killed the stone (which it would not pay White to do) Black would play again on the space thus made vacant, and completely surround and kill the entire white group.

Plate 4

Plate 5

A group with five vacant “Me,” as shown in the preceding diagrams, is a situation well known to the Japanese, so much so that they have a special phrase or saying that applies to it, to wit, “Go moku naka de wa ju san te,” which means that it takes thirteen turns to reduce a group having five such “Me” in the center.

As we have previously seen, in actual play this white group would be regarded as “dead” as distinguished from “taken,” and this series of moves would not be played out. White obviously would not play in the space, and he could not demand that Black play therein in order to complete the actual surrounding of the stones, and the only purpose of giving this series of diagrams is to show theoretically how the white stones can be killed. However, the killing of these stones would be necessary if the surrounding black line were in turn attacked (“Semeai”), in which case it might might be a race to see whether the internal white stones could be completely surrounded and killed before the external white group could get in complete contact with the black line.

Stones which are sacrificed in order to kill a larger group are called “Sute ishi” by the Japanese, from “Suteru,” meaning “to cast or throw away,” and “Ishi,” a “stone.”

It may be noted that if a group contains four connected vacant intersections in a line it is safe, because if the adversary attempts to reduce it, two disconnected “Me” can be formed in the space by simply playing a stone adjacent to the adversary’s stone, as shown in Plate 5, Diagram iii, where if Black plays for instance at K 11, White replies at L 11, and secures two “Me.” Even if these four connected vacant intersections are not in a straight line, they are nevertheless sufficient for the purpose, provided the fourth "Me" is connected at the end of the three, and the Japanese express this by their saying "Magari shimoku wa me," or four "Me" turning a corner. Neither does it make any difference whether the four connected "Me" are in the center of the board or along the edge. On Plate 5, Diagrams iv and v, are examples of "Magari shimoku wa me," and they both are safe. It is interesting, however, to compare these situations with that shown at Plate 4, Diagram ii, where the fourth intersection is not connected at the end of the line, and which group Black can kill if it is his move, as we already have seen.

If, however, such a group contains only three connected vacant intersections, and it is the adversary's move, it can be killed, because the adversary by playing on the middle intersection can prevent the formation of two disconnected "Me." We saw a group of this kind on Plate 2, Diagram ix, which can be killed by playing at N 11. Obviously, if it is Black's move in this case, the group can be saved by playing at N 11; obviously, also, if White, being a mere novice, plays elsewhere than at N 11, Black saves the stones by playing there and killing the white stone. Plate 5, Diagram vi, shows another group containing only three vacant intersections. These can be killed if it is Black's move by playing at A 1. On the other hand, if it is White's move, he can save them by playing on the same point.

Of course, if a group of stones contains a large number of vacant intersections, it is perfectly safe unless the vacant space is so large that the adversary can have a chance of forming an entire new living group of stones therein.

We now come to the one exception to the rule that the players may place their stones at will on any vacant intersection on the board. This rule is called the rule of "Ko," and is shown on Plate 6, Diagram i. Assuming that it is White's turn to play, he can play at D 17 and take the black stone at C 17 which is already surrounded on three sides, and the position shown in Plate 6, Diagram ii, would then arise. It is now White's turn to play, and if he plays at C 13, the white stone which has just been put down will be likewise surrounded and could be at once taken from the board. Black, however, is not permitted to do this immediately, but must first play somewhere else, and this gives White the choice of filling up this space (C 13) and defending his stone, or of following his adversary to some other portion of the board. The reason for this rule in regard to "Ko" is very clear. If the players were permitted to take and retake the stones as shown in the diagram, the series of moves would be endless, and the game could never be finished. It is something like perpetual check in Chess, but the Japanese, in place of calling the game a draw, compel the second player to move elsewhere and thus allow the game to continue. In an actual game when a player is prevented from retaking a stone by the rule of "Ko," he always tries to play in some other portion of the board where he threatens a larger group of stones than is involved in the situation where "Ko" occurs, and thus often he can compel his adversary to follow him to this other part of the field, and then return to retake in "Ko." His adversary then will play in some part of the field, if possible, where another group can be threatened, and so on. Sometimes in a hotly contested game the battle will rage around a place where "Ko" occurs and the space will be taken and retaken several times.

Plate 6

Korschelt states that the ideograph for "Ko" means "talent" or "skilfulness," in which he is very likely wrong, as it is more accurately translated by our word "threat"; but be this as it may, it is certainly true that the rule in regard to "Ko" gives opportunity for a great display of skill, and as the better players take advantage of this rule with much greater ingenuity, it is a good idea for the weaker player as far as possible to avoid situations where its application arises.

There is a situation which sometimes arises and which might be mistaken for "Ko." It is where a player takes more than one stone and the attacking stone is threatened on three sides, or where only one stone is taken, but the adversary in replying can take not only the last stone played, but others also. In these cases the opponent can retake immediately, because it will at once be seen that an endless exchange of moves (which makes necessary the rule of "Ko") would not occur. A situation of this kind is shown on Plate 6, Diagrams iii, iv, and v, where White playng at C 8 (Diagram iii) takes the tree black stones, producing the situation shown in Diagram iv, and Black is permitted immediately to retake the white stone, producing the state of affairs in Diagram v. The Japanese call such a situation "Ute keashi," which means "returning a blow." It forms no exception to the ordinary rules of the game, and only needs to be pointed out because a beginner might think that the rule of "Ko" applied to it.

We will now take up the situation called "Seki." "Seki" means a "barrier" or "impasse"—it is a different word from the "Seki" in the phrase "Jo seki." "Seki" also is somewhat analagous to perpetual check. It arises when a vacant space is surrounded partly by white and partly by black stones in such away that, if either player places a stone therein, his adversary can thereupon capture the entire group. Under these circumstances, of course, neither player desires to place a stone on that portion of the board, and the rules of the game do not compel him to do so. That portion of the board is regarded as neutral territory, and at the end of the game the vacant "Me" are not counted in favor of either player. Plate 6, Diagram vi, gives an illustration of "Seki," where it will be seen that if Black plays at either S 16 or T 16 White can kill the black stones in the corner by playing on the other point, and if White plays on either point Black can kill the white stones by filling the remaining vacancy. Directly below, on Diagram vii, is shown the same group, but the corner black stone has been taken out. The position is now no longer "Seki," but is called by the Japanese "Me ari me nashi," or literally "having 'Me,' not having 'Me.'" Here the white stones are dead, because if Black plays, for instance, at T 4 White cannot kill the black stones by playing at S 4, for the reason that the vacant "Me" at T 1 still remains. The beginner might confuse "Seki" with "Me ari me nashi," and while a good player has no trouble in recognizing the difference when the situation arises, it takes considerable foresight sometimes so to play as to produce one situation or the other.

Plate 6, Diagram viii, shows another group which might be mistaken for "Seki," but here, if White plays at J 19, the black stones can be killed, further proceedings being somewhat similar to those we saw in the illustration of "Go moku naka de wa ju san te." Plate 7 shows a large group of stones from which inevitably "Seki" will result. It would be well for the student to work this out for himself. "Seki" very seldom or never occurs in games between good players, and it rarely occurs in any game.

It is a rule of the game to give warning when a stone or group of stones is about to be completely surrounded. For this purpose the Japanese use the word "Atari" (from "ataru," to touch lightly), which corresponds quite closely to the expression "gardez" in Chess. If this warning were omitted, the player whose stones were about to be taken should have the right to take his last move over and save the imperiled position if he could. This rule is not so strictly observed as formerly; it belongs more to the etiquette of the old Japan.

The game comes to an end when the frontiers of the opposing groups are in contact. This does not mean that the board is entirely covered, for the obvious reason that the space inside the groups or chains of stones is purposely left vacant, for that is the only part of the board which counts; but so long as there is any vacant space lying between the opposing groups that must be disposed of in some way, and when it is so disposed of it will be found that the white and black groups are in complete contact.

Just at the end of the game there will be found isolated vacant intersections or "Me" on the frontier lines, and it does not make any difference which player fills these up. They are called by the Japanese "Dame," which means "useless." (The word "Dame" is likely to be confusing when it is first heard, because the beginner jumps to the conclusion that it is some new kind of a "Me." This arises from a coincidence only. Anything that is useless or

Plate 7

profitless is called "Dame" in Japanese, but etymologically the word really means "horse's eye," as the Japanese, not being admirers of the vacant stare of that noble animal, have used this word as a synonym for all that is useless. Therefore the syllable "Me" does mean an eye, and is the same word that is used to designate the intersections, but its recurrence in this connection is merely an accident.)

It is difficult for the beginner at first to understand why the filling of these "Dame" results in no advantage to either player, and beginners often fill up such spaces even before the end of the game, feeling that they are gaining ground slowly but surely; and the Japanese have a saying, "Heta go ni dame nashi," which means that there are no "Dame" in beginners' Go, as beginners do not recognize their uselessness. On the other hand, a necessary move will sometimes look like "Dame." The moves that are likely to be so confused are the final connecting moves or "Tsugu," where a potential connection has been made early in the game, but which need to be filled up to complete the chain. In the Illustrative Game, Number i, the "Dame" are all given, but a little practice is necessary before they can always be recognized.

When the "Dame" have been filled, and the dead stones have been removed from the board, there is no reason why the players should not at once proceed to counting up which of them has the greatest amount of vacant space, less, of course, the number of stones they have lost, and thus determine who is the victor. As a matter of practice, however, the Japanese do not do this immediately, but, purely for the purpose of facilitating the count, the player having the white pieces would fill up his adversary's territory with the black stones he had captured as far as they would go, and the player having the black stones would fill up his adversary's territory with the white stones that he had captured; and thereupon the entire board is reconstructed, so that the vacant spaces come into rows of fives and tens, so that they are easier to count. This has really nothing to do with the game, and it is merely a device to make the counting of the spaces easier, but it seems like a mysterious process to a novice, and adds not a little to the general mystery with which the end of the game seems to be surrounded when an Occidental sees it played for the first time. This process of arrangement is called "Me wo tsukuru." It may be added that if any part of the board contains the situation called "Seki," that portion is left alone, and is not reconstructed like the rest of the board.

Plate 8 shows a completed game in which the "Dame" have all been filled, but the dead stones have not yet been removed from the board. Let us first see which of the stones are dead. It is easy to see that the white stone at N 11 is hopeless, as it is cut off in every direction. The same is true of the white stone at B 18. It is not so easy to see that the black stones at L and M 18, N, O, P, Q and R 17, N 16, and M and N 15 are dead, but against a good player they would have no hope of forming the necessary two "Me," and they are therefore conceded to be dead; but a good player could probably manage to defend them against a novice. It is still more difficult to see why the irregular white group of eighteen stones on the left-hand side of the board has been abandoned, but there also White has no chance of making the necessary two "Me." At the risk of repetition I will again point out that these groups of dead stones can be taken from the board without further play.

Plate 9 shows the same game after the dead stones have been removed and used to fill up the respective territories, and after the board has been reconstructed in accordance with the Japanese method, and it will be seen that in this case Black has won by one stone. This result can be arrived at equally well by counting up the spaces on Plate 8, but they are easier to count on Plate 9, after the "Me wo tsukuru" has been done.

Plate 10 shows another completed game. This plate is from Korschelt, and is interesting because it contains an instructive error. The game is supposed to be completed, and the black stone at C 18 is said to be dead. This is not true, because Black by playing at C 17 could not only save his stone, but kill the four white stones at the left-hand side. Therefore, before this game is completed, White must play at C 17 to defend himself. This is called "Tsugu." On the left-hand side of the board is shown a white group which is dead, and the method of reduction of which we have already studied in detail. On the right side of the board are a few scattering black stones which are dead, because they have no chance of forming a group with the necessary two "Me." The question may be asked whether it is necessary for White to play at C 1 or E 1 in order to complete the connection of the group in the corner, but he is not obliged so to do unless Black chooses to play at B 1 or F 1, which, of course, Black would not do.

On Plate 11, this game also is shown as reconstructed for counting, and it will be seen that White has won by two stones. Really this is an error of one stone, as White

Plate 8

Plate 9

Plate 10

should have played at C 17, as we have previously pointed out.

Sometimes at the end of the game players of moderate skill may differ as to whether there is anything left to be done, and when one thinks there is no longer any advantage to be gained by either side, he says, "Mo arimasen, aru naraba o yuki nasai," that is to say, "I think there is nothing more to be done; if you think you can gain anything, you may play," and sometimes he will allow his adversary to play two or three times in succession, reserving the right to step in if he thinks there is a chance of his adversary reviving a group that is apparently dead.

No part of the rules of the game has been more difficult for me to understand than the methods employed at the end, and especially the rule in regard to the removal of dead stones without actually surrounding them, but I trust in the foregoing examples I have made this rule sufficiently clear. Moreover, it is not always easy to tell whether stones are dead or alive. There is a little poem or "Hokku" in Japanese, which runs as follows:

"Iki shini wo
Shiranu nonki no
Go uchi kana,"

which might be translated as "Oh! what kind of a Go player is he who does not know whether his stones are alive or dead!" But while the Japanese author of this "Hokku" may have regarded it as a simple thing, the Occidental student of the game would not be likely to share his views. An instance of this is shown by the possibilities of the supposedly dead black stone on Plate 10, and I think it would be fairer to state that the skill of a good Go player is most

Plate 11

clearly shown by his ability to recognize immediately whether a group is dead or can be saved; the study of our chapter on Problems will give further illustrations of the difficulty and nicety of such decisions.

We now come to the question of handicaps. Handicaps are given by the stronger player allowing the weaker player to place a certain number of stones on the board before the game begins, and we have seen in the chapter on the Description of the Board that these stones are placed on the nine dotted intersections. If one stone is given, it is usual to place it in the upper right-hand corner. If a second stone is given, it is placed in the lower left-hand corner. If a third stone is given, it is placed in the lower right-hand corner. The fourth is placed in the upper left-hand corner. The fifth is placed at the center or "Ten gen." When six are given, the center one is removed, and the fifth and sixth are placed at the left and right-hand edges of the board on line 10. If seven are given, these stones remain, and the seventh stone is placed in the center. If eight are given, the center stone is again removed, and the seventh and eighth stones are placed on the "Seimoku" on line K. If the ninth is given, it is again placed in the center of the board.

Between players of reasonable skill more than nine stones are never given, but when the disparity between the players is too great, four other stones are sometimes given. They are placed just outside the corner "Seimoku," as shown on the diagram (Plate 12), and these extra stones are called "Furin" handicaps. "Furin" means "a small bell," as these stones suggest to the Japanese the bells which hang from the eaves at the corners of a Japanese temple. When the disparity between the players is very great indeed,

Plate 12

sometimes four more stones are given, and when given they are placed on the diagonal halfway between the corner "Seimoku" and the center. These four stones are called "Naka yotsu," or "the four middle stones," but such a handicap could only be given to the merest novice.

We have now completed a survey of all the actual rules of the game, and it may be well to summarize them in order that their real simplicity may be clearly seen; briefly, they are as follows:

1. The object of the game is to obtain vacant territory.
2. The stones are placed on the intersections and on any vacant intersection the player chooses (except in the case of "Ko"). After they are played they are not moved again.
3. 1. One or move stone which are compactly surrounded by the stones of the other side are to be taken are are at once removed from the board.
   2. Stones which, while not actually surrounded can inevitably surounded, are dead, and can be taken from the board at the end of the game without further play.
   3. Taken or dead stone are used to fill up the adversary's territory.
4. The game is at an end when the opposing groups of stone are in absolute contact (the case of "Seki" being the single exception).

It is not possible to imagine a game with simpler rules, or the elements of which are easier to acquire.

We will now turn our attention to few considerations as the the best methods of play, and of certain moves and formations which occur in every game, and also to the manes which in Japanese are used to designate these things.