Geoff Kaniuk's referee exercises

One player thought that the game was over because he knew that bent four is dead. With the AGA rules he would have continued to play.

Edit: my post is irrelevant.

This is not a bent-4 is dead situation (it’s not like Topic Dia. 1 or Topic Dia. 2 in Bent four in the corner is dead at Sensei's Library). So that peculiarity is irrelevant here.

It’s just an unsettled situation. Both players are leaving a 20-point move hanging, while playing 1 point endgame moves elsewhere. When both players would be making the same evaluation mistake under AGA rules, it would depend on the number of dame points who plays here first. As soon as that happens, both players might realize their mistake.

Under Japanese rules, it would depend on the referee. If the referee is strong enough to realize that the situation is unsettled, they might let them resume the game, or score this as a seki or decide that both players lose (but I think the latter rarely happens in Europe). If the referee is not strong enough, they might make a mistake in scoring this position.

Additionally, bent four is dead under any ruleset. It’s not a rule, but a consequence of a rule.

1 Like

Could I ask you gentlemen to spin this discussion off into its own thread, please?

“Bent four under Japanese and AGA rules” or something?

Well there can be very rare situations under area scoring (unremovable ko threats), where it is not fully dead.

Yeah, I meant “usually dead”. But still, it’s not a rule that says “this is dead by definition”, but a consequence of actually playing it out in Chinese, or hypothetically playing it out in Japanese rules (and in the latter, unremovable ko threats are not considered, resulting in the always-deadness).

I think it’s very relevant to be able to correctly judge the given problem, though, to know the status of this group under the ruleset of the tournament. Without it, how could one answer 9. ?


I’m trying to make the point that this is irrelevant for question 9.

1 Like

But it’s relevant to know that it is irrelevant :wink:


With the Aga rules, players fill dame points. The position wouldn’t have been left unsettled. With Japanese rules, players don’t need to fill dame points, that’s why they mistakenly left the position unsettled.

Also under Japanese rules, players are recommended to fill all neutral points before counting, and players generally do so in tournament games IME.
But I don’t see how that would have made a difference in this situation. Both players refused to play here before passing. And it’s especially strange that black refused to play here to close their territory before passing, if they considered this black group to be alive. That is the actual problem here. Not closing off your territory before passing is a problem under any rule set.

1 Like

There are some interesting exceptions to this. In fact, it is possible for two bent fours to be alive in seki under Japanese rules. The status under area scoring rules depends on the existence and size of unremovable ko threats.

1 Like

If I were the referee for question 9, I might tell them to fill all neutral points before passing again. Given the apparent misunderstandings of both players, it might turn out like this:
And then black may or may not play this out correctly. Because this situation came up at all, I suppose that black is a DDK player and there is a real chance that black would fail to secure unconditional life.

If instead both players would still refuse to play in this area when filling all neutral points, I would score it as a seki. But I would not tell the players that beforehand, because I would not want to give any hints.

If the 3k agrees to resume (even though that’s not in their best interest, I assume), I would allow it and the 3k gets to play first. But if the 3k disagrees, the position after passing stands.

Edit: Apparently, my decision would not conform to the Official Japanese rules. (Oops :face_with_hand_over_mouth:)
Article 9.3. If a player requests resumption of a stopped game, his opponent must oblige and has the right to play first.

So I learned something new :slight_smile:

Strangely enough, it seems that Geoff would also consult the 3k player, although it’s not really clear what he would do if the 3k disagrees.

1 Like

Nowadays this is usually not allowed. You can record the game on a smartphone in plain sight of the opponent if they have no objections, but you should leave it on the table if you go for a walk. So I would be asking them to show me what they are doing. If they don’t seem to be cheating with AI, I would warn them that they risk disqualification with this behaviour. If they are cheating, I might disqualify them (perhaps after consulting other referees if there are any).

These questions are under Japanese rules, so I would explain to them that the game will be voided if they both continue to cycle the triple ko. So they should both consider if they rather continue the game by playing elsewhere, or finish the game right away with no result/jigo (whatever fits in the pairing system).

If instead some superko rule applies, it might be a challenge to reconstruct who violated the superko rule first, especially if these are weaker players and this cycling has been going on for a while already.


Extra case: what would you do as a referee?

1 Like

Similar to studying josekis during correspondence games.

This case with its spread of time of a correspondence game and review of a live game doesn’t translate well into a real-life tournament with a referee like Geoff’s exercises. However, here is a case I have wondered about as it’s almost happened to me, I think I even discussed it with Geoff but can’t remember his thoughts.

I am playing in the London open, a live tournament with slow time settings (90 mins main plus overtime). During my games I like to get up and wander around, look at how the other games are going in the room, stretch my legs, get a drink. In my game I am playing a complicated joseki like the taisha, and have forgotten how the variation I am half way down continues. On my wander I noticed another of the games has the same taisha variation played out in one of the corners, and is about 20 moves further along, plus the rest of the board is different of course. But seeing that end joseki position is enough to jog my memory of what the continuation should be, though I still need to think a bit about the correct move order. I then return to my game and play the correct joseki move, and the game continues with us both playing the joseki correctly. Did I cheat? How about if after I play the right move, my opponent messes up and I correctly punish him, using my own thinking. How about if opponent messes up, and then I mess up too?

How does the above change if the other game I see doesn’t simply jog my memory of something I know from a further developed position, but is at the exact same point in the variation and they play a move I didn’t know before?


Surely the answer must be no.

Otherwise nobody would be allowed to leave the board and spectate other games.


What if also my memory of the taisha corresponds to some weird memory aid.

I mean people remember whole decks of cards and digits of pi by making up stories right?

What if the move I’m forgetting in the taisha was associated with a blue jumper and while wander I see a person playing another game with a blue jumper and I remember the position and how to answer it by remembering the story :slight_smile:

1 Like