A Proverb Revised

British Go Journal No. 64. March 1985. Page 11.

A Broader Discussion

Francis Roads

A presentation of several symmetry problems and answers are on previous pages.

At this stage I would like to broaden slightly the scope of the symmetry proverb. The centre of symmetry of some positions does not lie on a line or vertex. Your opponent would take a dim view if you played on the line of symmetry in either of the endgame positions shown in Dia 14 and 15. As you know, according to the rules of Go, the chief of all courtesies is to play on the vertices.

Diagram 14





Diagram 15




In Dia 14 Black gains about 12 points in gote by playing at T or U. Of course if White gets there first he plays at the equivalent points opposite. In Dia 15 Black can gain four points in sente by playing V; White's move would be W.

(If the last two sentences don't make complete sense, then refer to chapter 6 of "Basic Techniques of Go" by Haruyama and Nagahara, or better still, to "The Endgame" by Ogawa and Davies. The latter is a must if you are one of those players who get superb positions in the middle game but make the wrong decisions in the endgame.)

So my revised proverb is simply this: "In a symmetrical position the player with sente has the advantage".

Diagram 16




Diagram 17




I hear you asking: " Is that all the lengthy preamble has been leading to? Isn't that rather obvious?" Well maybe it is, but many people fail to appreciate its significance in some very simple situations, such as Dia 16.

Whoever plays first here gains a considerable advantage. You often see weaker players filling in a liberty of a white stone in a situation where it is clear that they imagine this represents some sort of atcak on the stone. Perhaps they have a sequence like that of Dia 17 in mind, in which White obediently ignores three white moves in succession.

Maybe this idea is generated by the way we teach Go to beginners. Often when a beginner plays his first game, sequences like Dia 17 are the only ones he has seen. I have the impression it can often take a long time to eradicate the idea that this is how stones get captured in actual play.

Diagram 18




Diagram 19




Diagram 20




In practice, of course, what happens is that White takes advantage of being first to play in a symmetrical position, with a hane like 2 in Dia 18. Black's stone is already reduced to half its birthright of four liberties.

In situations where a more defensive move is required, White can choose X (of course the equivalent points of Y and Z are available too). What White is unlikely to do is ignore the contact play altogether - thereby handing over to Black the advantage of playing first in a symmetrical position.

Now, if following Dia 18 both players persist in playing symmetrically, Black comes to grief first, as Dia 19 shows.

So Black will always have to break symmetry with a defensive move first. He must play 5 at 6 in Dia 19, and ideally he should defend earlier: black A or B in Dia 20 is often best, while C and D are possible alternatives, leading to the cross cut of Dia 19.

"I know the contact paly is always bad," a DFK said to me at Wanstead a few weeks ago. No, that is not the point; in fact it isn't always bad. The point to remember is that the contact play is not normally an attacking play.

The contact play forces your opponent to strengthen himself more than you yourself. It may of course be that he is so strong in the area that you can make him overconcentrated; more likely, you may have some local strength to nullify the additional strength he creates. In any event you won't go far wrong if you think of a contact play as a defensive manouvre, and remember that its one great advantage is that it usually forces a reply of some sort.

Diagram 21









Diagram 22









So, let's now apply our revised proverb to the first four moves of Dia 19, the notorious cross-cut (or "kiri-chigae" if you like Japanese terms) - See Dia 21.

Why does this formation have a reputation for complexity? The reason is, unless one of the stones is sacrificed, four groups, two of each colour, are going to have to find living space in the area.

The notorious "Tai-sha" ('great slant') joseki owes its mind-boggling complexity to a cross-cut for this very reason. In Dia 22, after white 8, Black must cut at 9, and now the stones 4, 5, 6 and 9 form a cross-cut.

Of course there are other stones present, but the continuation to 20 (one of the simplest lines) shows clearly how the four resultant groups are jostling for space.

So, going back to Dia 21, how does the player with sente seize the advantage?

Diagram 23




Diagram 24




Diagram 25




Stating the conclusion first, and assuming it is Black to play, then in the majority of cases it is better to play one of the simple extensions E, F, G or H, rather than one of the ataris I, J, K or L (Dia 23).In other words, white 4 in Dia 19 was not necessarily the best move.

After an extension as shown in Dia 24, Black may threaten a ladder at M, or at least the extension at N. If White pulls out his stone with O or P in Dia 25, Black immediately plays Q or R to "play hane at the head of two stones" in accordance with another proverb.

Diagram 26





Diagram 27





Diagram 28





If White pulls out his stone withthe diagonal move at S in Da 26, Black has a good contact play at T or U, which puts white into bad shape (I leave you to investigate, look out for empty triangles).

So after black 1 in Dia 24, White will probably strengthen his threatened stone. But how? V in Dia 27 preserves too much symmetry and invites W. If instead the iagonal move at X, Black extends to Y, and White needs to play again to avoid bad shape.

So the best move is often Z in Dia 27. But this invites black 3 in Dia 28, and White is again faced with a dilemma. A invites B (cf Dia 25); C is unsatisfactory as we know; and D preserves too much symmetry, preserving Black's advantage.

The conclusion is that after 1 in Dia 24 White has the unpalatable choice between inferior shape, or preserving symmetry and Black's advantage.

Diagram 29





Diagram 30





Diagram 31





What about the four atari moves I, J, K and L in Dia 23?

After 1 in Dia 29 White must normally play 2. This leaves Black with a cut at E. If Black protects it, eg with 3 in Dia 30, then White has the ladder at F or the extension to G - compare with Dia 25. But if he fails to protect the cut it will remain a thorn in his flesh for the future.

Atari moves like 1 in Dia 29 are similar to conatct plays; they provoke a local response, but are best kept in reserve.

Naturally any of the sequences shown above can be upset by local circumstances. A nearby stone or the edge of the board can turn good shape into bad, and vice versa.

Diagram 32





Here is just one example. The marked stone in Dia 31 turns the bad black atari 1 into a good move. After black 3 the white stone is cut off on a rather poor point. This shape crops up in the 6-3 point joseki shown in Dia 32. It is most often played when there is a white stone at or around H.

This article has turned out rather like a Bruckner symphony - rambling on, but with a theme running right through it. So like Bruckner I'll return to my first theme at the end.

Summary: "Play at the centre of a symmetrical formation" is useful advice, but not always correct. I can be generalised to: "In a symmetrical position the player with sente has the advantage." This principle applies especially to contact plays and to cross-cuts.

[Start] Plagal cadence.


This article is from the British Go Journal Issue 64
which is one of a series of back issues now available on the web.





Last updated Wed Jan 02 2013. If you have any comments, please email the webmaster on web-master AT britgo DOT org.