Yasser Seirawan's "Winning Chess Combinations," recently issued by Everyman Chess, is a new fascinating book in his Winning Chess series. How to define a combination turned into a big debate in the last century. Some experts believed that a combination is a sequence of forced moves leading to an advantage. Others, including the former world champion Mikhail Botvinnik, maintained that a sacrifice must be always present in a combination. The young Seirawan joked that he always wanted to sacrifice his opponent's pieces and keep his own, but hiding behind these words was a player with excellent tactical skills.
As in his previous books, Seirawan tackles his subject by first explaining the basic ideas and patterns and subsequently discussing some of the most complex and brilliant performances of the world's leading players. Garry Kasparov's "immortal" game against Veselin Topalov, played in the Dutch town of Wijk aan Zee in 1999, is one of the masterpieces analyzed in great detail. Kasparov turned a double rook sacrifice into a vicious king's hunt, but at one point missed a shorter and more brilliant finish. After I published the improvement in this column, Seirawan was impressed and sent me a box of smoked salmon posthaste. It was the first and last time my chess move was rewarded with a fish.
To explain the classic bishop sacrifice, Seirawan chose another Kasparov game for his new book, and I immediately smelled another salmon. It was the controversial fifth game from his 2003 match against Deep Junior that angered the New York fans after it ended in 19 moves with a boring move repetition. Kasparov allowed the bishop sacrifice and later lacked the courage to play for a win. What he really feared and saw during the game is still a mystery.
1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.e3 0-0 5.Bd3 d5 6.cxd5 exd5 7.Nge2 Re8 8.0-0 Bd6 9.a3 c6 10.Qc2? Bxh2+! 11.Kxh2 (The Hungarian grandmaster Joseph Horvath reached this position via a different move order in 1989 and thought that black can force a draw.) 11...Ng4+ 12.Kg3 Qg5 13.f4 Qh5 14.Bd2 (Suggested by Carsten Hansen in his 2002 Gambit Publications book "The Nimzo-Indian 4.e3," as an improvement to Horvath's 14.Ng1?! Qh2+ 15.Kf3 Qh5 18.Kg3 repeating the moves.) 14...Qh2+ 15.Kf3 Qh4 (Threatening 16...Nh2 mate. The game concluded with a repetition of moves: 16.Bxh7+ Kh8 17.Ng3 Nh2+ 18.Kf2 Ng4+ 19.Kf3 Nh2+ draw. Instead of 16.Bxh7+ Kasparov could have made the game fun to watch with 16.g3. The following analyses were published in this column shortly after the game. )
16.g3!? Qh2! (Tightening the mating net.) 17.f5 (A more exciting draw than in the actual game could be achieved after 17.Rae1 g6 [threatening 18...Nxe3! 19.Bxe3 Bg4+! 20.Kxg4 Qh5 mate] 18.f5 Nd7! 19.e4 [The only move, otherwise white gets mated either after 19.Kxg4 Qh5+ 20.Kf4 g5 mate; or after 19.fxg6 Nde5+! 20.dxe5 Nxe5+ 21.Kf4 Qh6 mate.] 19...dxe4+ 20.Nxe4 gxf5 21.N4c3 Re3+!? 22.Bxe3 Nde5+ 23.dxe5 Nxe5+ 24.Kf4 Qh6+ 25.Kxe5 Qg7+ 26.Kd6 Qf8+ draw by a perpetual check. This variation should be corrected: Instead of 21...Re3+, black can play 21...c5!, for example 22.Bf4 b5!, threatening 23...Bb7+ with good winning chances.) 17...Nd7! (The key move to black's triple-piece sacrifice. Threatening 18...Nde5+! 19.dxe5 Nxe5+ 20.Kf4 Qh6 mate, it forces white to accept the first gift. Seirawan claims that Kasparov saw this move and its consequences during the game, but at the postmortem Kasparov only talked about 17...h5.) 18.Kxg4 Qg2! (Not allowing the white king to drop back.) 19.Kf4 (After 19.Rh1 Nf6+ 20.Kh4 h6! 21.Rag1 Bxf5! 22.Rxg2 g5 mates.) 19...Nf6 20.e4! dxe4 21.Bxe4 (Prompting black to sacrifice a rook and a bishop.)
21...Rxe4+! (This spectacular sacrifice, suggested by Deep Junior, began to circulate on the Internet only a few days after the game. On 21...Nxe4 22.Nxe4 Qxe2 23.Rae1! wins.) 22.Nxe4 Nd5+ 23.Ke5 ( The white king is forced to walk into a lion's den. After 23.Kg4? Qxe2+ 24.Kh3 Ne3 black wins.) 23...Bxf5!! (Black quickly brings in his rook, speeding up the attack. Seirawan writes that 24.Rxf5 is now a forced capture. "Is the position after 24.Rxf5 good or bad for White?" he asks the reader. The answer was published in this column shortly after the game. Here goes...) 24.Rxf5? (The best defense, 24.Nf4!, is so complicated that it could fill another column. Seirawan does not even mention it. On the other hand, 24.Kxf5 loses to 24...Qh3+ 25.g4 [or 25.Kg5 h6 mate] 25...Re8 and white is mated soon, e.g. 26.Nf4 g6+ 27.Nxg6 fxg6+ 28.Kg5 h6+ 29.Kxg6 Qxg4+ 30.Kxh6 Qh4+ 31.Kg6 Ne7 mate.) 24...Re8+ 25.Kd6 Rd8+! (Seirawan shows 25...Qxe2 26.Re5?? Rd8+ 27.Kc5 Qb5 mate.) 26.Ke5 (On 26.Kc5 Qxe2 27.Nd6 b6+ 28.Kxc6 Rxd6+! wins.) 26...Qxe2 27.Re1 (Black also wins after 27.Raf1 Nb4!; or after 27.Rxf7 Ne3!) 27...Ne7!! 28.Rf6 (On 28.Rxe2 Ng6 mates.) 28...Rd5+ 29.Kf4 Rf5+! 30.Rxf5 Ng6+ 31. Kg5 h6 mate .
Do I get another salmon, Yasser?
Solution to today's study by S. Isenegger (White: Kb6,P:f6,h3; Black: Kb8,P:a6,c5,f7): 1.h4! c4 2.h5 c3 3.h6 c2 4.h7 c1Q 5.h8Q+ Qc8 6.Qh2+ Ka8 7.Qg2+ Kb8 8.Qg3+ Ka8 9.Qf3+ Kb8 10.Qf4+ Ka8 11.Qe4+ Kb8 12.Qe7! Qe6+ 13.Qxe6 fxe6 14.f7 wins.