I had a heavy book which introduced tons of games of all kinds when I was a kid. I was fascinated by go back then. Now, about 15 years later I am actually getting to know the game. I was doing some tsumegos from the Cho Chikun’s Encyclopedia of Life and Death when the 21st tsumego caught my attention. The objective is to kill a stone chain in the corner. I considered all options and I came into conclusion that the stones cannot be killed. Although of course they can be killed.
I discovered that playing into the corner is the correct solution, but I am still puzzled why it is. White responds by connecting his stones on the 2-5-point and then what? To me it looks like a seki situation, mutual life. The chain of three black stones and the white chain have mutual liberties on S1 and T4.
This situation will result into what we call “bent four in the corner” (a term that is often thrown around but not always in a correct way so be carefull :D) and white will actually (at least almost always) die.
You got the first part right, for white this “is” seki. Meaning if he/she tries to atari the black stones he/she will inevitably die. So far so good (or not really, not for white anyway)
Because for black, this is NOT really a seki…
Because black can start a ko for the life of the whole group! (white could technically threaten elsewhere and recapture 5…) But is it really a ko??? Well technically yes, but we have already established that only black can start it, white can just wait or die…
Thus black can wait for “the opportune moment” as they say. and start it when he is sure to win (there are no more ko-threats). Under Chinese rules this is easy to do, after everything else is played out, black can just fix every possible ko-threat with no loss (in chinese playing inside your own territory costs nothing) - usually both players just agree it is dead as to not waste time - when there are unremovable ko-threats it gets complicated… In Japanese rules the position is considered dead by default (even in the case of unremovable ko-threats).
Wow, the same question six days ago! I tried searching but couldn’t find that topic which is quite embarrassing really. I don’t even have the excuse that I would be unfamiliar with this software. I am actually an admin on a Discourse forum.
But yeah, at least I am not the only beginner who couldn’t figure this out.
I commend your swift uptake and eagerness to share your newfound knowledge. It is quite impressive to see 20k pretty much correctly identifying and applying this technique in a game. Keep up the good work! Study all the tsumego so you will know how to solve them in real games.
I am very sorry for being a nitpicking bastard, but upper right was not actually a bent four situation. It is the same ko that results from bent four, but not the bent four itself. (since white had the ability to start the ko, before black can fix all ko threats)
There are quite a few other critical weaknesses in my game which is why I suffer painful losses all the time. As @HowToPlay pointed out in the game chat, white J5 deep in the endgame on turn 62 would have been a disaster for me. But neither player saw it which is how my games flow. It’s blunders and counterblunders. Anyway, sharing the pattern with another beginner made me happy.
Edit. Never mind. I played that just fine.
Yes @AdamR, I see. I had that idea for the opponent in the game but as the game went on, I forgot. I fully agree.
Keeping a hierarchical washlist of moves ordered by urgency and constantly reassessing said order as the game progresses is one of the most difficult aspects of the game in the long term. Recognizing that a shape can be killed, remembering or reading on the spot how to initiate that killing sequence and ultimately, if belatedly, finishing the killing shape cleanly, … don’t sell yourself short.
Note that the “bent four in corner” rule is a peculiarity of Japanese (and some other) rulesets.
For me, the original attraction of Go was the simplicity of its rules. The Japanese rules, IMHO, spoil this simplicity by introducing ko as a special rule (instead of a rule which simply prohibits a move that repeats a previously-occurring board position, with the same player to move), and then the “bent 4 in corner” rule, which actually refers to a number of positions that can lead to the bent 4.
As pointed out on the Sensei’s net page (HowToPlay gave a link, above) some other rulesets, for example the New Zealand rules, don’t suffer from this disfigurement. It can make a difference to the game result because some ko threats cannot be removed except at large cost.
- when there are unremovable ko-threats it gets complicated… In Japanese rules the position is considered dead by default (even in the case of unremovable ko-threats).
don’t be sad. It just means that you have a life and I don’t
