Dutch Treat 
by Hans Ree

21st Century Chess?

I make mistakes, therefore I exist, wrote my hero Tartakower. As
long as I make mistakes, I do not exist, Kasparov seems to think.
For years he has been pushing the form of chess he calls "advanced
chess", in which man and computer, like a modern centaur, will
form a unity. Chess for the 21st century, Kasparov calls it. As far
as I know, the first time he came up with it was during a
tournament in Amsterdam where he claimed to have lost two
games because he could not remember his old opening analysis.
This unfortunate and unjust accident would not happen in
advanced chess, where players would have at their disposal a chess
computer, database, and files of old analysis.

It took some time, but it was to be expected that Kasparov would
have his way. In the second week of June, in the Spanish town of
Leon, Kasparov and Topalov played a six-game match with the
help of computers. Originally I thought that it would also be a
match between different manufacturers of chess software,
comparable to an automobile race, but Ricardo Calvo corrected me
on The Chess Cafe's Bulletin Board. In fact, Kasparov and Topalov
had the same packet available: the playing program Fritz 5 and the
database Chessbase 7.0. The final decision what move to make was
still for man, not for Fritz. Both players got one hour for the entire
game, which did not give them much time for extensive
deliberations with Fritz.

I certainly do not hope that this will really be the chess of the
future, for it lacks everything that makes chess attractive to
millions of players. The rationale behind it is that chess will be
raised to a higher level by a harmonious division of tasks between
man and machine. Man will think of long-range strategies and will
check the tactical variations with the help of a computer, so that his
deep strategical thinking will not be spoiled by stupid blunders.
But as Freud wrote, one has to be able to live with some
incertitude. The exhilarating feeling that one walks in a minefield,
that one has to be constantly on the alert for a tactical surprise,
plotting at the same time to surprise one's opponent, these are the
very things that make chess exciting, and Kasparov wants to
eliminate them in an irrelevant search for certitude and perfection.

But as a one-time light-hearted exhibition, it probably was fun,
especially for the spectators, who for the first time in chess history
could have the idea that they were looking into the players' heads
and seeing what variations they were investigating. What they
actually could see were the screens of Fritz 5, the variations that
Kasparov and Topalov were feeding it and Fritz's evaluations.

According to reports of Leontxo Garcia in the Spanish newspaper
"El Pais", Topalov made little use of his computer, which is
understandable as one hour for a game is not much and sensible
discussions with chess computers can take a long time. Kasparov
leaned on Fritz more heavily; after all, it was his idea from the
beginning. That it helped him, I doubt. He was seen putting
variations to Fritz for a long time, then angrily shaking his head
and gesturing with his hands as if to say: "What idiot do I have to
put up with?" and then going back to the board, now thinking for
himself, but having lost quite a lot of time in fruitless discussions
with the beast.

The final result was 3-3, both players winning two games. Then
some blitz games were played - no time for computer consultations
then, I suppose - Kasparov winning the final sudden death game. A
few months ago Kasparov beat Topalov 4-0 in a short match of
rapid games without computer assistance. For the moment and at
the time controls in Leon, I suppose that Kasparov on his own is
quite a lot stronger than Kasparov with computer prosthesis.

Fortunately mistakes were still made and beauty could be seen, as
in the fourth game. 

White: Kasparov Black: Topalov, Fourth Game 1. Ng1-f3 Ng8-f6
2. d2-d4 g7-g6 3. c2-c4 Bf8-g7 4. g2-g3 c7-c5 5. Bf1-g2 Qd8-a5+
6. Nb1-c3 Nf6-e4 7. Qd1-d3 c5xd4 8. Nf3xd4 Ne4-c5 9. Qd3-d1
Nb8-c6 10. e2-e3 Nc5-e6 11. 0-0 The first pawn sacrifice, a nice
one, though seen before in a game where Kasparov's second
Azmaiparashivili was black. Black declines, but soon will be
forced to win a pawn after all.11...Qa5-c5 12. Nd4-b5 a7-a6 13.
Nc3-a4 Qc5xc4 This pawn will give black little pleasure.14.
Nb5-c3 Ra8-b8 15. Na4-b6 Qc4-c5 16. Nc3-d5 0-0 17. b2-b4
Qc5-d6 18. Ra1-b1 Ne6-c7 19. Nb6-c4 Qd6-e6 20. e3-e4 d7-d6 21.
Bc1-e3 Qe6-d7 22. Nc4-b6 Qd7-d8 23. f2-f4 Nc7xd5 24. e4xd5
Nc6-a7 25. Rb1-c1 Bc8-f5 26. g3-g4 Na7-c8 27. Nb6-c4 Bf5-d7
28. Qd1-e2 Bd7-b5 29. Qe2-f2 e7-e6 30. Rf1-d1 e6-e5 31. f4-f5
g6xf5 32. g4xf5 Qd8-f6 33. a2-a4 Had this pawn sacrifice been
accepted, Kasparov probably would have followed it up with an
exchange sacrifice: 33...Bxa4 34. Rf1 Bb5 35. Nd2 (heading for
e4) Bxf1 36. Rxf1 33...Bb5xc4 34. Rc1xc4 Nc8-e7 35. Rd1-f1
Kg8-h8 36. Qf2-g3 b7-b5 37. Rc4-h4 Ne7xf5 (See Diagram)

38. Rh4xh7+ And a nice rook sacrifice to crown the
achievement.38...Kh8xh7 39. Bg2-e4 Kh7-g8 In all the reports I
saw, Fritz's evaluation that black could have forced a draw at this
point with 39...Kh8 40. Bxf5 Bh6 41. Qh3 Kg7 42. Qg3+ Kh8 was
repeated as Gospel. "Fritz found this in 5 seconds" wrote its
manufacturer Frederic Friedel proudly. Fast it sure is, but was it
right? Can black survive the attack after 39...Kh8 40. Dh3+ Qh4
41. Qxf5? Anyway it would be much preferable to the game
continuation. 40. Be4xf5 e5-e4 41. Qg3-h3 Black resigned.

All in all I prefer to see games played without brain prosthesis. As
an extra, here is the final game of the Kramnik-Shirov match. A
brilliant win by Shirov, but it has to be said that after the opening
he was in a bad state. Will he be able to escape so spectacularly
from his chains when playing Kasparov for the world
championship? Most chess fans hope so, because they like a
change of power, but it seems hard to believe.

White: Kramnik Black: Shirov, Ninth Game 1. d2-d4 Ng8-f6 2. c2-c4
g7-g6 3. f2-f3 An oldie, used several times by Alekhine.
Apparently Kramnik has not been able to find anything convincing
against Shirov's Gruenfeld Defence. 3...d7-d5 4. c4xd5 Nf6xd5 5.
e2-e4 Nd5-b6 6. Nb1-c3 Bf8-g7 7. Bc1-e3 0-0 8. Qd1-d2 e7-e5 9.
d4-d5 c7-c6 10. h2-h4 h7-h5 11. Bf1-e2 c6xd5 12. e4xd5 Nb8-d7
13. d5-d6 Nd7-f6 14. Be3-g5 Rf8-e8 15. Ra1-d1 Bc8-e6 16.
Ng1-h3 Nb6-c4 17. Be2xc4 Be6xc4 18. b2-b3 Bc4-a6 He wants to
keep the white king in the middle, but this is over-optimistic and
imprudent. Now white could have played 19. d7 Re6 20. Nf2, after
which black can hardly move a piece and in the long run will
succumb to the pin of Nf6. 19. Nc3-d5 The same idea, but wrongly
executed, because now black has a chance to turn loose. 19...e5-e4
20. Nd5xf6+ He could have played 20. d7 immediately, which is
very complicated. Also in this case black will lose material, but he
has many promising attacking possibilities. On the FIDE Internet
pages (www.chessweb.com) Luc Winants gives the following
beautiful variation: 20. d7 exf3+ 21. dxe8Q+ Qxe8+ 22. Qe3 Nxd5
23. Rxd5 Qc6 24. Rd8+ Rxd8 25. Bxd8 Qc2 26. Qe8+ Bf8
(threatening 27...Bb5 28. Qxb5 Bb4+) 27. Rg1 Qc3+ 28. Kf2 Bb5
29. Qxb5 Bc5+ 30. Kg3 f2+ 31. Kh2 Bd6+ 32. Kh1 Qg3 and white
will be mated. 20...Bg7xf6 21. d6-d7 Now if black had to move his
rook white would be fine. (See Diagram)

21... Qd8-b6! But there is better. A rook is sacrificed for a decisive
attack. 22. d7xe8Q+ Ra8xe8 23. Qd2-e3 The only way not to lose
quickly. After 23. Be3 black wins with 23...Bxh4+ followed by
24...exf3. 23...Bf6xg5 24. Qe3xb6 Bg5xh4+ 25. Ke1-d2 25. Nf2
exf3+ 26. Kd2 Bg5+ doesn't help white either. 25...a7xb6 26. f3xe4
Re8xe4 A formal count would grant white an exchange for two
pawns, but Tarrasch already pointed out that two bishops plus rook
are a formidable power, hardly inferior to two rooks plus knight.
Black is winning. 27. Kd2-c2 Re4-g4 28. Rd1-d2 Bh4-e7 29.
Rh1-g1 Kg8-g7 30. Nh3-f2 Rg4-f4 31. Nf2-d3 Rf4-e4 32. Rg1-d1
Ba6-b5 33. a2-a4 Bb5-c6 34. Rd1-e1 Re4xe1 35. Nd3xe1 Be7-b4
36. Rd2-e2 Bb4xe1 37. Re2xe1 Bc6xg2 38. Kc2-d2 h5-h4 39.
Kd2-e3 Bg2-d5 40. b3-b4 h4-h3 41. Re1-e2 f7-f5 42. Re2-d2
Bd5-e4 43. Ke3-f4 Be4-g2 44. Rd2-d7+ Kg7-f6 45. Rd7-h7
g6-g5+ 46. Kf4-g3 f5-f4+ 47. Kg3-g4 Kf6-e5 48. b4-b5 and white
resigned.

This column first appeared in the Dutch newspaper
"NRC-Handelsblad" June 13, 1998. 
Copyright 1998 Hans Ree. All Rights Reserved.