“Your September 19 journal article ‘AGA Rules Win’ brings up something about the passing
stones rule that has always vexed me, but hopefully unnecessarily,” writes David Hast from Grand Rapids, MI. “Perhaps you can explain.
Here is a theoretical situation: It is the endgame. Black fills the last dame on the board. Now it is White's turn. White's only options for placing a stone on the board would result in the loss of a point (either playing in his own territory, or playing a move in Black's territory that Black doesn't need to answer). The same situation exists for Black when it will be his turn. So now, following AGA rules, White passes a stone, then Black passes a stone, and then White must pass one more stone, since the rules require White to pass last. This passing exchange has resulted in a net gain of one point for Black.
Could this result in a 0.5 point loss for White after komi is calculated? In other words, does the rule of passing stones, with White required to go last, make it possible for White to lose the game?”
Terry Benson responds:
Below is an article from 2001 which appeared in the Journal. I have updated the text only slightly and added commentary at the end.
Only a Passing Matter: A Note on the New AGA Rules (2001)
Terry Benson
In an even game with 7 and 1/2 komi if white must make the third pass at the end of the game, that stone *does* change the *score* (from the traditional Japanese count) but does not change the *result* - unless there is a seki situation with an odd number of shared liberties AND it’s a one point game.
The reason is a matter of parity.
Assume that the players alternately fill in any dame. (This doesn't affect the traditional score in any case.) Only if Black plays the last stone on the board is White - under AGA rules – required to hand over a third “pass stone.”
If Black plays last, then the number of stones played by both players must be odd. Since the board has an odd number of points (361), the total territory (empty space after filling in the prisoners but WITHOUT pass stones) will usually be even. (The exception is a game with an odd number of shared liberties in seki.)
When the territory is even, the difference in the traditional (Japanese) scores of the two players must also be even: 2, 4, 6, etc. (e.g. 33 - 27 or 32 - 26). (An odd difference would make the total territory odd, e.g. 33 - 28 or 32 - 27.)
Looking at a difference of 6 and 8, if White is behind by 6 points (traditional count) and gets 7 1/2 komi, White wins by 1 1/2. The additional pass stone reduces the victory to 1/2 point, but White still wins.
If White is behind by 8 points (traditional count) and gets 7 1/2 komi, White loses by 1/2 point. The additional pass stone makes the score 1 1/2; White's loss is just one point bigger.
The third pass stone also does not affect the result of handicap games of 3, 5, 7, and 9 stones; but, because of the parity of the stones played and the points of komi, it *can* affect the result in other handicap games: 0 stones (1/2 point komi to white), reverse komi of 7 1/2 to black (reverse komi should perhaps be 6 1/2 for this reason), handicaps of any even number of stones (2, 4, 6, etc.), and even games with non-standard even point komi.
New Commentary on the Rules of Go (9/24/24)
by Terry Benson
Not including bidding or betting, a strategy game where you can “pass” - even if you can legally play - is unusual. You can’t pass in chess or checkers or Parchesi or Backgammon or bridge (play).
In Chinese rules, AGA rules, and the ancient game, every intersection is valuable. Passing loses points. AGA rules incentivize playing on the board (and completing walls, captures, so called "dame") if you can. Fundamentally, you can’t pass. You either play a stone or pass a stone. There are no “neutral” points. The object of the game is to “control more of the board” not “to surround territory.” Surrounding is how you close off, capture, and show control.
The ancient game was: who can put the most stones on the board. Rather than a tedious filling up of the board, it’s was easier to count by territory once all the contestable points were occupied.
AGA rules are a hybrid. The logic of area (Chinese) scoring and the convenience and comfort of territory (Japanese/Korean) counting. By requiring equal stones to be played and then put back on the board, you can ignore the stones and just count the empty spaces. Viola!
(Note that AGA rules also allow Chinese and Ing counting.)
Note that “Equal stones” is *not* a modern concept. There are game records from China in the 8th century which talk about “equal stones” at the end of the game. (It appears that Black was prohibited from playing the last stone.) That made counting simpler – just the territory - to get the same result. But when the game traveled to Japan territory counting was retained without the supporting concept of equal stones or the underlying understanding that every living stone was as valuable as the intersections they surround. The distortions which resulted and the confusion about the end of the game for less skilled players are what AGA rules fix.