.
Daily Yomiuri On-Line
The Magic of Go
Pros have a full arsenal of techniques to clinch a win or to turn the tide if they are behind. One of these techniques is a good understanding of sente, i.e. keeping or taking the initiative, and gote, which means the opposite. It may seem obvious that having sente would then always be good, while getting gote is a bad thing. That is actually far from the truth. Amateurs often attach too much value to the sente/gote problem, and forget to defend properly.

Defending properly

Often playing a sente exchange leads to the fixation of possibilities, while not making this exchange would leave those possibilities open.

Diagram 1: For old hands, this is a familiar game between Yoshio Ishida and Shuko Fujisawa in the 1971 Honinbo League. Let's look at this game in terms of sente and gote. The first problem occurs when White plays 10. The joseki (standard exchange) calls for exchanging white A for black B (or sometimes C), which is sente. However making this exchange forfeits a possible invasion around C (where White plays almost 60 moves later). White exchanges 10 for 11, then takes sente to make a shimari. A more difficult problem arises after black 13. White could have chosen Reference 1. The next problem is white 28, a move that sums up Shuko's thick style. Again, White takes gote, but it is the vital point of thickness for both sides. Imagine a black stone here. White 28 enables the invasion at C later and isolates 13. The last problem is the 30/31 exchange. Does Black need to respond? See Reference 2 for what happens if he doesn't. After black 31, white 32 and 34 are sente moves to limit the scale of Black's expansion and forcing him to defend his territory.

Reference 1: After black 1 and 3, White could have played 4 and 6, probably leading to the variation through 9. White then gets sente to switch to 10 (or A). In the game he loses sente, but he gets more territory and more thickness.


Reference 2: If Black doesn't defend with 31, White will invade at 1. If Black tries to kill him with 2 through 6, white 7 through 11 are nasty--if Black keeps insisting with 12, White captures six stones after 19. Of course, there are different variations, but the bottom line is Black can't kill the white invader.

Problem 1: After black 3, how should White respond?

Problem 2: White plays first and wins the game. Think about connecting, life and death, sente and gote.

Solutions to last week's problems


Solution 1: Do unto your opponent as he does unto you, should be your motto here. After black 3, do not answer immediately, but counter-invade yourself. The moves through 8 seem a reasonable continuation. The fighting skills of the individual players determine the outcome.

Reference 3: So what happens if White responds with 4? After 11, White needs to defend somehow because the lower right white group is too large to sacrifice now, and Black grabs the initiative with 13 and 15. The sente exchange 13/14 makes the invasion at A less of a threat. After 15, White will have his hands full defending his stones on the lower left, lower right and upper left at the same time.

Solution 2: Black captures 19 with 30, White recaptures with 33. Black passes with 34 and White connects at the marked stone with 33. White can first force 1 through 8, capturing the marked stone to prepare for 9 and 11. Black can barely live with both groups with 12 through 14, then has to defend with 16 and 18. Note Black should not play 25 as White resists with 21 to make ko--White has ko threats, Black has none. In the end, White gets 14 points in territory + 2 prisoners and Black gets 12 points + 2 prisoners. White wins by two points.

Solution 3: After black 1, White should respond with 2, etc. The sequence through 16 may look improbable, but if White defends even once, he loses. Of course, Black can defend, but this would also give White that same chance.
backhome
Copyright 1999-2004 The Yomiuri Shimbun