The Chess Mind

And yet, it just won't stay dead, even in a theoretically ancient line like the one found in the famous game Karpov-Korchnoi, Moscow (Candidates match) 1974 (game 2). Karpov slaughtered Korchnoi in that game, which began

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 g6 6.Be3 Bg7 7.f3 Nc6 8.Qd2 O-O 9.Bc4 Bd7 10.h4 Rc8 11.Bb3 Ne5 12.O-O-O Nc4 13.Bxc4 Rxc4 14.h5 Nxh5 15.g4 Nf6 16.Nde2 Qa5 17.Bh6 Bxh6 18.Qxh6 Rfc8 19.Rd3



Here Korchnoi played 19...R4c5 and went down in flames: 20.g5! Rxg5 21.Rd5! Rxd5 22.Nxd5 Re8 23.Nef4 Bc6 24.e5! Bxd5 25.exf6 exf6 26.Qxh7+ Kf8 27.Qh8+ 1-0

Victor Reppert is a C. S. Lewis scholar (commendable), a philosopher (also praiseworthy), but, sadly, has been infected by the Dragon bug for many decades. In this post, Reppert argues that Korchnoi's 19th move was Black's 4th best move; 19...R8c5, 19...Qd8 and 19...Be6 are all better tries.

In the comments to that post, I knock out 19...R8c5 and 19...Qd8, and now it's time to start shaking 19...Be6 out of Reppert's tree of viable Dragon variations.

Here's his main line:

19... Be6 20.g5 Nh5 21.Ng3 Qe5 22.Nxh5 gxh5 23.Qxh5 Kf8 24.Qh2 Qxg5+ 25.f4 Qf6 26.f5 Rxc3 27.bxc3 Bxa2 28.Qxh7 Ke8 29.Qh8+ and now both 29...Kd7 and 29...Qxh8 are satisfactory for Black.

Now, there's no question about it: 19...Be6 is a tougher nut to crack. Finding promising leads was no trouble, but somehow Black always seemed to have survive. Practically speaking, Black's task may be hopeless, but insofar as we're just after the truth, those lines couldn't count as refutations.

I think, though, that I've managed to find a line in which White maintains at least a clear edge. After 19...Be6 20.g5 Nh5 21.Ng3 Qe5 22.Nxh5 gxh5 23.Qxh5 Kf8, not 24.Qh2 but

24.Qxh7 Qxg5+ 25.Kb1 Ke8 26.Qh2 f6



and now, after failing to prove a serious edge after 27.Rg1 and 27.f4, the quiet

27.a3

seems to work. Here's my main line:

27...Kd7 28.f4 Qa5 29.Qd2 Rg8 30.Qe1 Kc8 (30...Qc5 31.b3+/-) 31.f5 Bd7 32.Rg3 Re8 33.Qe2



White has a clear advantage, thanks to the open files for the rooks, the d5 square, and the potential weakness of e7.

Sometimes justice isn't swift, but it's sure. Of course, there's also grace - Black doesn't have to play the Dragon! (Okay, okay - he can play the Dragon. Just not this line.)

In the late 1990s, New in Chess Magazine had a couple of "SOS" (Secrets of Opening Surprises) articles on gambits for Black, both arising from openings with ...g6 and then ...d5.

The first, from the 1998/8 issue, featured a specialty of Peter Svidler's: 1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Bg5 and now not the usual 4...Ne4, but simply 4...Bg7 instead. White is invited to grab the d-pawn with 5.Bxf6 Bxf6 6.cxd5 (6.Nxd5 Bg7 7.e3 c5 8.Nf3 Nc6 and Black regains the pawn) 6...c6! 7.dxc6 Bxd4 8.cxb7 Bxb7.



White's up a pawn, but Black's magnificent bishop pair provide ample compensation.

The second idea comes from the 1999/4 issue and goes like this: 1.e4 g6 2.d4 Bg7 3.Nc3 d5. Cop-out moves like 4.e5 c5 or 4.Nf3 dxe4 promise nothing, while 4.Nxd5 c6 immediately regains the pawn. That leaves only 4.exd5, but after 4...Nf6 5.Bc4 Nbd7 6.Nf3 Nb6 7.Bb3 a5



we have a position reminiscent of a well-known (temporary) gambit in the Panov-Botvinnik variation of the Caro-Kann: 1.e4 c6 2.d4 d5 3.exd5 cxd5 4.c4 Nf6 5.Nc3 g6 6.cxd5 Bg7, when one possible continuation is 7.Bc4 Nbd7 8.Nf3 O-O 9.O-O Nb6 10.Bb3



In both cases, White has chances for a small edge, but nothing special.

And now, for gambit idea #3, brought to my attention by Brian Karen:

1.e4 g6 2.d4 Bg7 3.c4 and now, not 3...d6 or 3...c5, both quite common, but 3...d5!? followed by 4.cxd5 c6 5.dxc6 Bxd4.



Reminds one of the first gambit, doesn't it? Unfortunately, while it's a clever idea, it's not of the same quality of the preceding gambit lines.

For starters, White can vary with 5.Nc3, when 5...cxd5 6.Nxd5 e6 doesn't work out as charmingly as 4.Nxd5 c6 in the second gambit line. After 7.Nc3 Bxd4 8.Nge2, White's lead in development and Black's somewhat porous queenside and dark squares give White a clear advantage.

Further, even grabbing on c6 seems to be fine: 5.dxc6 Bxd4 6.cxb7 Bxb7 7.Bb5+ followed by 8.Ne2 is again clearly better for White. For instance, even 7...Bc6 8.Ne2 works well, e.g. 8...Bxf2+ 9.Kxf2 Qb6+ (9...Bxb5 10.Qxd8+ Kxd8 11.Nbc3 with a clear advantage thanks to the vulnerable Black king) 10.Be3 Qxb5 11.Nbc3 and White's impending dark-square domination and the Black king's lack of a safe haven ensure fun for White and trials for Black.

Verdict: Not recommended.