# What is this alternate cut after 3-3 against double keima round hoshi?

In 38 Basic Joseki, Kosugi & Davies remark in this position:

``````\$\$c
\$\$ --------------------
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . 1 . . O . . . .
\$\$ -. . . X . a . . . X
\$\$ -. . . 3 2 6 . . . .
\$\$ -. . O 4 5 . . . . .
\$\$ -. . . . 7 . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . X . . . . . ,
``````

that if White blocks at 2, Black will push and cut with 3 and 5 or 3 and a.

I cannot see how one is meant to cut with 3 and a in this situation. I have not found this in Ishida, or on Josekipedia (though someone has asked a question about a) or in Sensei’s Library; though with this severely pruned search pattern (only shaded points sought):

there were some matches (to which the “a” refers), but nothing useful that I could see.

## Questions

• Is this a misprint, or how is one meant to cut with 3 and a?
• When is this preferable?
• What continuations are there, and are they joseki?

## Follow-up

I am not sure what I can have thought (though I expected a to be an answer to a different White 4, rather than achieving a different objective), but now I think I see, as Daniel T says in his answer that the marked white stones are separated:

``````\$\$c
\$\$ --------------------
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . X . c W . . . .
\$\$ -. . . X b B . . . X
\$\$ -. . . X W d . . . .
\$\$ -. . O O a . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . X . . . . . ,
``````

He shows W@a–B@b, but W b and d also seem to fail, with a shaky ladder and too many cuts respectively. I still do not see when this is good; maybe the issue is which 6-3 stone B wants to separate 5-5 from, but he could just push 3 in the other direction.

It seems to me that playing 3 and 'a' cuts the E15 stone from the F17 stone.

``````\$\$Bcm1
\$\$ +----------
\$\$ |..........
\$\$ |..........
\$\$ |..1..O....
\$\$ |...X75...X
\$\$ |...32.....
\$\$ |..O46.....
\$\$ |..........
\$\$ |..........
\$\$ |..........
\$\$ |...X.....,
``````
• You are evidently right, though White 6 at E16 or F15 are rather more complicated to refute. I have (perhaps a little unfairly) extended my question to also ask when this cutting pattern is preferable. Apr 25, 2018 at 21:22
• Spot on that white cannot easily connect. As a side note if this actually happens in a game, I don't imagine w playing 6 in the diagram, as it will remove all options w could have with E16 at some point (now or in the future), probably tenuki is better. Black's 7 seems better at G16 or G17 maybe, to avoid white's good shape move in that area.
– mafu
Apr 27, 2018 at 9:47

The point is, that in this situation

``````\$\$c
\$\$ --------------------
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . X . c W . . . .
\$\$ -. . . X b B . . . X
\$\$ -. . . X W d . . . .
\$\$ -. . O O a . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . X . . . . . ,
``````

the moves at `b` and `c` are miai: Either white allows black to connect solidly at `b`, or white moves at `b` themselves. In that case, black can still take `c`, yielding this situation:

``````\$\$c
\$\$ --------------------
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . X . X O . . . .
\$\$ -. . . X W B e . . X
\$\$ -. . . X W . . . . .
\$\$ -. . O O . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . . . . . . . .
\$\$ -. . . X . . . . . ,
``````

Now white cannot capture the black cutting stone in a ladder by playing at `e`, because the two white stones lack a liberty, breaking the ladder in favor of black. The cutting stone at `E17` is safe as well, so there is no way that white can connect up.

Also, a black follow-up at `E14` is dangerous for white, so white will need to invest another move to either protect that cut, or make it irrelevant. I'm not sure which move would be the best continuation for white, but that consideration leaves the scope of this question anyway.