I am unable to report the final result of ...

By Rob Van Zeijst

I am unable to report the final result of the 27th World Amateur Go Championship, held May 28-31 in Sasebo, Nagasaki Prefecture, for deadline reasons. However, at the time of writing, South Korea and China were leading the pack with six wins out of six games after South Korea defeated North Korea in a direct confrontation. You can find out how the championship turned out by visiting www.nihonkiin.or.jp/amakisen/worldama/27/index-e.htm.

To get an idea about the level of the players, let us review what happened last year. South Korea's representative Seo Jung Hwi had just passed his country's exam to become a pro and this was his last tournament as an amateur. To turn pro in South Korea, a player must have attained a pro level of at least 3-4 dan. Therefore, Seo was generally considered to be a top candidate for the title. However, he dropped two out of eight games, finishing fourth. In other words, the top players of this tournament probably have a level equivalent to at least a 5-dan pro.

Ko

Now we know the ko rule, let us explore what its impact is on the game. As we learned last week, ko threats are important when playing a ko. These threats often are answered, but not always. If the ko is decisive--this is called tenka-ko or tenka-kikazu--the ko threat often will be ignored and the ko resolved. Let us take look at an example.

Diagram 1: The marked white stone is a pivotal stone. It splits Black's center groups. Fortunately for Black it is his turn. Let us see what happens when he captures the ko.

Diagram 2 (black 3 connects where the marked white stone was captured): When Black captures the ko with 1, White only has one ko threat at 2. Black ignores it and connects with 3, while White finishes his ko threat at 4. So, Black has connected his center stones while White bottles up three black stones with 4. When Black plays at 5, he makes his entire group alive.

Conclusion: Not being permitted to immediately capture black 1, White was forced to make a ko threat. But as the value of the ko is much larger than the ko threat, Black ignored the threat and made the connection.

Diagram 3: Here is an example of a huge ko in a game between top pros. Go Seigen, one of the most famous go players of the 20th century, is Black, while White is Shukaku Takagawa, who won nine Honinbo titles in a row and is known for his simple style. After a relatively quiet start with a familiar joseki (standard corner exchange) in the top left, the situation suddenly erupts into a ko when Black cuts at 33. Why does Black start a ko?

Diagram 4 (white 36 is played where the marked black stone was captured): Black started the ko because he has an excellent ko threat at 35. Since the value of the ko is incredibly large, White must ignore Black's ko threat and connect with 36 where he took the marked black stone. Black then captures the two marked white stones with 37. White pushes with 38 and 40, but the value of the ko is not so much what White can win, but what he can lose (see Reference 1). Not quite recovered from the shock of this sudden ko, White plays at 42, a move that is locally large, but has little global impact. Sure enough, instead of cutting at 44, Black attacks the entire group with 43. Having to spend four moves--42, 44, 46 and 48--to make life with a small group while Black plays four moves that have a much more global impact, is no good. In the end, White had to resign.

Reference 1: When Black plays a ko threat with 1, if White answers with 2, Black will take back the ko with 3. White's only ko threat is at 4. However, Black will ignore any threat and capture one more stone with 5. Next, White can capture the three marked black stones with 6, but black 7 bottles up the three white stones marked with X's to take the entire lower left corner.

Conclusion: In the case of a game-decisive tenka-ko, a person who stands to lose the most should ignore ko threats even if the value of these threats is large. However, don't start a tenka-ko unless you can win it or if you have a ko threat that is huge. We will look at some examples next week.