In a Go game, players take turns in placing their stones on empty intersection on the board.
However, it is legal (in most sets of rules) that a player doesn't place a stone on the board by making a pass (i.e. not playing a stone, doing nothing). Then it's the other player's turn.
The article on resigning has some good methods of resigning in unspoken ways, such as placing two stones on the board at the same time. These are effective in breaking the language barrier (or circumventing it, at any rate). Do such customs exist for passing?
Bill Spight: Under most sets of rules a pass is free. However, under AGA rules of territory scoring a pass costs 1 point. The passer hands his opponent a stone as a captive, called a pass stone. Two consecutive passes end play, but White makes the last pass. The practical effect of pass stones is to reconcile territory and area scoring. Several people have had the idea of pass stones, but the AGA incorporation of them can be traced back to an American Go Journal article I wrote in the 1970s on the Chinese rules in which I called them bookkeeping stones. Pass stones sounds better, doesn't it?
Andrew Grant: Can Bill (or anyone) explain why "pass stones" are given in the USA? The only reason I am aware of is that that makes area scoring agree with territory scoring - but that implies an unspoken assumption that territory scoring is "correct", and therefore area scoring needs to be modified to agree with it. The Chinese don't give 'pass stones', and seem happy with the idea that their scoring system is as good as the Japanese territory scoring.
Either stick with territory scoring or change to area scoring 'without' pass stones, but don't try to reconcile two incompatible scoring systems.
Bill: My idea was not to try to reconcile two different scoring systems, but to make it easy for players who were used to territory scoring to play under area scoring. The use of pass stones allows them to use their familiar method of counting territory when counting area.
Mef: It also has the convenient side effect of a universal way to communicate that you are passing, for players who may not share a native language.
Anon: Grant said "... implies an unspoken assumption that territory scoring is 'correct' ..." but in fact it is the other way round. Pass stones here modify the territory score to be equal to the area one. I also made this mistake on first reading, mainly because it is usually said that the territory score is superior to the area one, for reasons of being "sharper". Contrast this with button go, which modifies the area score to have the same sharpness as the territory score.
Ending the game with two consecutive passes can lead to positions similar to Zugzwang, in the sense that not having even a neutral move leads to a loss. Such positions involve moonshine kos.
Andre Engels: The number of passes to end the game need not be two. The rule I have learned, and which I think is the standard in Japanese rulesets, is that three passes are necessary. Depending on the ruleset in use, this might resolve some problems with kos (I may not take back the ko, so I pass, if then my opponent passes as well, I can retake, so he cannot leave an open ko on the board).
Bill: I believe that the first person to suggest a 3-pass rule was Yasunaga Hajime, before the codification of go rules. My rule for stopping play is that the same player passes twice in the same whole board position. That usually amounts to a 3-pass rule.
Jozsef: Since two passes may only stop the game, with resumption being possible, I don't think three passes are necessary to avoid such open kos. But for both cases it is important to allow taking back a ko after a pass.