queen's gambit accepted - smith, hall

173

Upload: car-dusan

Post on 04-Dec-2014

722 views

Category:

Documents


79 download

TRANSCRIPT

Ken Smit� Jo�n Hal

Chess Digest, Inc.

Queen's Gambit Accepted

Copyright© 1995

Chess Digest, Inc.

All rights reserved under Pan American and International Copyright conventions. Cover photograph is a painting copyright© 1995 by Keith Halonen (see page 7)

ISBN 0-87568-255-3

This is a revised and enlarged edition of Queen's Gambit Accepted- Chess Digest (Ken Smith) 1974.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form, or by any means: electronic, electrostatic, magnetic tapes, mechanical photo­copying, recording or otherwise, without prior and current permission from the publisher.

Authors: Ken Smith and John Hall Editors: Ken Smith and Roy DeVault Computer Typesetting: Roy DeVault Cover: Elaine Smith Final Proof: Sid Pickard Final Preparation and diagrams: Roy DeVault

Publisher: Chess Digest, Inc.®, 1601 Tantor (P.O. Box 59029) Dallas, Texas 75229

Send the publisher $2. 00 for the new Chess Guide that catalogs every chess book for general sale in the United States. You are given publishers, page counts, notation and critical reviews. Also included in the catalog is a free Chess Improvement course for beginners up through Master level players.

3 TABLE OF CONTENTS

Introduction 8

Familiarization 9

CHAPTER ONE 14 Classical Variation 6 0 -0 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 8xc4 c5 6 0-0

I A 6 ... cxd4 7 exd4 1 4 I A1 7 .. . Nc6 15 IA2 7 ... 8e7 1 7

I 8 6 .. . a6 18 I 81 7 Qe2 19

I 811 7 .. . b5 2 1

I 81 1 1 8 8b3 8b7 21 I 81 1 1 1 9 Rd1 Nbd7 22

1 81 1 1 1 1 10 e4 22 I 81 1 1 12 10 Nc3 25

I 8 1 1 1 121 10 .. . Qc7 25 I 81 1 1 1 22 10 ... 8d6 27 I 81 1 1 123 10 ... Qb6 28 I 81 1 1 1 24 10 .. . Qb8 30 I 81 1 1 1 25 10 ... b4 32

I 8 1 1 12 9 a4 33 I 81 12 8 8d3 35

I 81121 8 . . . Nc6 35 I 81122 8 ... cxd4 36

I 812 7 ... Nc6 39 I 8121 8 Rd1 39 I 8122 8 Nc3 41 I 8123 8 dxc5 43

I 8124 8 a3 44 I 82 7 8� �

I 83 7 Nc3 47

I� 7� N� �

I 841 8 Nc3 8e7 48

4 TABLE OF CONTENTS

I 841 1 9 Qe2 51

I 8412 9 dxc5 52 I 8413 9 Ne5 53

I 842 8 Qe2 54

I 85 7 e4 I 86 7 a3 I 87 7 b3 I 88 7 dxc5 I 89 7 Bd3

I 8421 8 ... cxd4 55 I 8422 8 ... Qc7 9 Nc3 58

I 84221 9 ... Be7 58 I 84222 9 ... Bd6 60

I 842221 10 b3 60 I 842222 10 dxc5 61 I 842223 10 Rd1 63 I 842224 10 Bd3 64 I 842225 10 Bd2 65

I 8423 8 ... Be7 66

I 84231 9 dxc5 I 84232 9 Rdl 0-0

66 68 69 70 71 72 74

CH APTER TWO 76 Furman's Variation 6 Qe2 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 Qe2

II A 6 ... cxd4 76 II B 6 ... a6 78

CHAPTER THREE 80 Modern Variation 4 ... Bg4 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 Bg4

Ill A 5 h3 80 Ill B 5 Bxc4 e6 82

Ill 81 6 Qb3 BxfJ 7 gxfJ Nbd7 8 Qxb7 c5 82 III 81 1 9 dxc5 83 Ill Bl2 9 Rg1 85 Ill 813 9 Nc3 85

TABLE OF CONTENTS 5

III B2 6 Nc3 87 III B3 6 aJ 88 III B4 6 0-0 89 III B5 6 Nbd2 90 III B6 6 h3 Bh5 7 Nc3 Nbd7 8 0-0 Bd6 9 e4 e5 91

III B61 10 g4 92 III 862 10 Bel 94

CH APTER FOUR 96 Smyslov's Variation 4 .. . g6 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 g6 96

CH APTER FIVE 98 Winawer's 4 ... Be6 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 eJ Be6 98

CHAPTER SIX 1 00 Black Plays Alekhine's 3 .. . a6 1 d4 d5 2 c4 dxc4 3 NfJ a6 1 00

VI A 4 a4 1 00

VI B 4 eJ Bg4 101 VI B1 5 h3 102

VI B2 5 Bxc4 e6 103

VI B21 6 Qb3 104 VI B22 6 Nc3 105 VI BD6 � 1 � VI B24 6 Nbdl 107

VI C 4 e4 b5 5 a4 Bb7 109

VI C1 6 Nc3 109

VI C2 6 axb5 1 1 1

VI C3 6 b3 1 12 VI D 4 Nc3 1 13

CHAPTER SEVEN 1 15 Mannheim Variation 4 Qa4ch 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 Qa4ch 1 15

VII A 4 .. . c6 5 Qxc4 Bf5 1 17

6 TABLE OF CONTENTS

VII A1 6 Nc3 1 18 VII A2 6 g3 1 19

VII B 4 ... Nc6 121

CHAPTER EIGHT 123

Elastic Variation 4 N c3 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 Nc3 1 23

VIII A 4 ... c5 1 24 VIII A1 5 d5 1 24 VIII A2 5 e4 1 26

VIII B 4 ... a6 5 e4 b5 6 e5 Nd5 1 28 VIII 81 7 Ng5 129 VIII 82 7 a4 130

VIII 821 7 ... e6 131 VIII 822 7 ..• Nxc3 133

CHAPTER NINE 1 36 Anti-Elastic Variation 3 ... c5 1 d4 d5 2 c4 dxc4 3 NO c5 4 d5 e6 5 Nc3 exd5 6 Q:x:d5 Q:x:d5 7 Nxd5 Bd6 8 Nd2 136

IX A 8 ... Ne7 137 IX B 8 . . . Nc6 139

CHAPTER TEN 141 White Plays 3 e4 1 d4 d5 2 c4 dxc4 3 e4 141

X A 3 ... e5 4 NO 141 X A1 4 .•. Nr6 142 X A2 4 ... Bb4ch 143

X A21 5 Nc3 144 X A22 5 Bd2 146

X A3 4 ... exd4 147 X B 3 . . . c5 150

X 81 4 NO 150 X 82 4 d5 151

X B21 4 ... Nf6 152 X 822 4 ... e6 153

TABLE OF CONTENTS

X C 3 .. . Nf6 4 e5 Nd5 5 Bxc4 Nb6 X C1 6 Bd3 X C2 6 Bb3

X 0 3 ... Nc6 X 01 4 Be3 X 02 4 NfJ Bg4

X 021 5 Be3 X 022 5 Bxc4 X 023 5 d5

CHAPTER ELEVEN Unusual White 3rd Move Alternatives 1 d4 d5 2 c4 dxc4

XI A 3 Nc3 XI B 3 e3

BIBLIOGRAPHY

Informants #I thru # 60

New in Chess # I thru #32 Das Damengambit- Pachman (1993) Play the Queen's Gambit- Marovic (1991) Trends in the Queen's Gambit Accepted#! -Martin (1994) Trends in the Queen's Gambit Accepted#2 -Baburin (1994) New Ideas in the Queen's Gambit Accepted- Flear (Batsford 1994)

COVER

7

156 156 158 160 160 162 162 163 165

167

167 168

Cover photo is of a painting by Keith Halonen. The original 18" x 46" oil

painting on reinforced untempered masonite paneling is entitled

"Concentration" and is for sale. Contact Keith Halontm; 2195 Dutton

Avenue, Santa Rosa, CA 95407-6977. Phone (707) 578-6125.

8 INTRODUCTION

The Queen's Gambit Accepted has long been the most direct answer to the sequence I d4 d5 2 c4. White offers a gambit with 2 c4, so Black accepts the challenge by simply taking 2 . . . dxc4. Originally, some theorists were horrified; isn't 2 . . . dxc4 a positional blunder, giving up the central pawn for nothing? The answer was soon clarified: No! Black is temporarily giving up the center. Black will restore the central balance by soon playing . . . c5 (or . . . e5 in some cases). Then White's only real satisfaction is the possession of the initial tempo. But even so, Black can fight for a dynamic equality by rapid piece development and - quite often - with a Queenside pawn advance after . . . a6 and . . . b5 (this normally taking place after a preliminary . . . c5).

Several main lines of the Q.G.A. evolve into positions in which White accepts an isolated d-pawn in return for a slightly freerer development. But Black's position, if properly played, is both solid and resilient. In many instances Black will have good winning chances (in dynamic equilibrium with White's) in the resulting positionally and tactically complex positions. This is why the Q.G.A. has been adopted by a host of top GM's over the years. Names such as Smyslov, Flohr, Keres, Petrosian, Fine and Reshevsky (of the "old guard") and modern GM's such as Ivanchuk, Seirawan, Dlugy, Miles, Hubner, Ehlvest, Kramnik and Anand give absolute testimony to its soundness.

The following analysis is written from both the White and Black side, giving the current best play for each. This will be "the reference book" on the Queen's Gambit Accepted for years to come.

FAMILIARIZATION

l d4 d5

2 c4 dxc4 JNO

9

For 3 e4 (the most current variation Grandmasters are playing) see Unusual White 3rd moves, Chapter XI.

By developing his KN before attempting to regain the gambit pawn, White prevents the freeing move . . . e5. Against 3 e3 or 3 Nc3 Black has the move 3 . . . e5 (see Chapter XI for these possibilities).

3 ... Nf6 This is designed to prevent 4 e4, which would give White a strong

center. If Black tries to keep his pawn, he loses time and weakens the

Queenside: 3 . . . b5? 4 a4 c6 5 e3 e6 6 axb5 cxb5 7 b3 a5 8 bxc4 b4 9 Ne5 Nf6 10 Bd3 Be7 1 1 0-0 0-0 1 2 Nd2 Bb7 13 f4 Nbd7 14 Qc2 Nb6 15 c5 Nbd5 16 Ndc4 Ra7 1 7 Bd2 Qc7 18 Ra4 (Hybi-Ericson, 5th World Correspondence Champ. 1965-68). Black has a weak pawn and no counterplay.

The passive 3 . . . e6 gives White superiority in the center: 4 e4 c5 5 Bxc4 cxd4 6 Nxd4 (6 Bb5ch is worth considering) 6 . . . Nf6 7 Nc3 Bf5 8 Be3, Capablanca-Bogoljubov, Moscow 1925.

The move 3 . . . c5 can be answered by 4 e3, transposing into "normal" lines. For the other responses to 3 . . . c5, see Chapter lX Anti-Elastic Variation.

3 . . . Nd7 can easily transpose into regular lines, but . . . Nb6 may hold the gambit pawn for some lime. For example, 3 . . . Nd7 4 e4 (not 4 Nbd2 b5 5

10 THE QUEEN'S GAMBIT ACCEPTED

b3 c3 6 Nb1 b4 7 a3 c5 8 dxc5 Nxc5 9 Qc2 Be6 favoring Black, Borisenko­Dorfman, USSR 1975) 4 . . . Nb6 5 Ne5 Nf6 6 Nc3 e6 7 Nxc4 with White holding a strong pawn center.

Premature (and bad) is 3 . . . Bg4 due to 4 Ne5 Bh5 5 Nc3 ! Nd7 (or 5 . . . e6 6 g4! Bg6 7 h4 f6 8 Qa4ch c6 9 Nxg6 hxg6 10 Qxc4 KP 1 1 e4 with a superior position for White, Alekhine-Grunfeld, Semmering I926) 6 Nxc4 (also good is Alekhine's 6 Qa4 c6 7 Qxc4 Nxe5 8 dxe5 e6 9 g4 Bg6 10 Bg2 Qc7 1 1 /4) 6 . . . Ngf6 7 f3 Nb6 8 Na5 Rb8 9 e4 e6 10 a3 Nfd7 I I Be3 Bd6 I 2 Qd2 c6 13 b4 0-0 I4 Be2 with clear advantage to White, Andersson­Kavalek, Bugojno 1982.

3 . . . c6 leads to a variation of the Slav Defense: 4 e3 Bg4 5 Bxc4 e6 6 Nc3 (6 Qb3 Qb6 7 Ne5 B./5 8 Nc3 Nf6 9 j3 Nfd7! with equality according to Alekhine) 6 . . . Nd7 7 h3 Bh5 8 0-0 Ngf6 9 e4 Be7 10 Bg5 h6 I I Be3 0-0 I 2 e5 Nd5 13 Nxd5 cxd5 I4 Bd3 ReS (or 14 . . ./5) with approximate equality, Antunac-Kovacevic, Yugoslavia I975.

Finally, 3 ... a6 is playable and is the basis for Alekhine's system, which is discussed in Chapter VI.

4 eJ For 4 Qa4, the Mannheim Variation, see Chapter VII, and 4 Nc3,

Elastic Variation, which is seen in current Grandmaster practice, see Chapter VIII.

4 ... e6 The key position of the Classical System. Black prepares to bring out

the KB and challenge White's pawn preponderance in the center by playing . . . c5. Otherwise White would eventually seize the center after Nc3 and e4. The transposition 4 . . . c5 and then 5 . . . e6 is also possible.

FAMILIARIZATION 1 1

Taking on independent significance from the Classical System are 4 . . . Bg4 (Modem Variation, Chapter III), 4 . . . g6 (Smyslov Variation, Chapter IV), and 4 . . . Be6 (Winawer Variation, Chapter V).

The defense of the gambit pawn by 4 . . . b5 is, of course, dubious: 5 a4 c6 6 axb5 cxb5 7 b3 Be6 8 bxc4 bxc4 9 Qa4ch Nbd7 10 Ne5 or 9 Ne5. White has a clear advantage.

·

Also good for White is 5 . . . b4 (after 4 . . . b5 5 a4) 6 Bxc4 e6 7 0-0 Bb7 8 Nbd2 Nbd7 9 e4, Taimanov-Benediktson, Reykjavik, 1968.

5 Bxc4 c5

Black may be courting problems if he delays too long in placing pressure on White's center by 5 . . . c5. Illustrating this danger is the game Gelfand-Gulko, Linares 1990: 5 . . . a6 6 Qe2 (6 0-0 c5 transposes into our main line, but not 6 . . . b5? l Bd3! Bbl 8 a4! b4 9 Qe2 Nbdl 10 e4 Bel 1 1 Nbd2 c5 1 2 e5 with a big plus for White) 6 . . . b5 7 Bd3 c5 8 dxc5 Bxc5 9 a3 Nbd7 10 b4 Be7 I I e4 Bb7 and now 12 Bb2 favors White somewhat.

In the position after 5 Bxc4 c5 White's plan is to castle and then work towards utilizing his central pawn majority. At the same time Black must find a suitable method to develop his QB, and the usual plan involved is . . . a6, . . . b5, and . . . Bbl. In order to cut across this idea, White frequently plays a4 - though this weakens the b4 square. This square can then become a convenient haven for a Black Knight.

After I d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 (the above diagrammed position), Black has excellent prospects when White plays 6 0-0, the main line Classical Variation, Chapter I, or 6 Qe2, Furman's Variation, Chapter II. Some current Grandmasters are short-circuiting this

12 THE QUEEN'S GAMBIT ACCEPTED

by playing 3 e4, Chapter X. Another way White has of dodging the Classical is 4 Nc3, the Elastic Variation, Chapter VIII. Black sometimes counters White's plan by 3 . . . c5, the Anti-Elastic Variation, Chapter IX. These mentioned variations are all the rage.

After I d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 (or 4 Qa4ch, the Mannheim Variation, Chapter VII):

There are other possibilities for change: Modem Variation 4 . . . Bg4, Chapter III, Smyslov's Variation 4 . . . g6, Chapter IV, and Winawer's Variation 4 . . . Be6, Chapter V.

After I d4 d5 2 c4 dxc4 3 NO

FAMILIARIZATION 13

Black can play 3 . . . a6, Alekhine's Variation, Chapter IV and do well. In the Queen's Gambit Accepted Black as well as White needs to know what he is about.

14 CHAPTER I

CLASSICAL VARIATION

(1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5)

6 0-0

Now on 6 . . . Be7 White can play 7 Qe2 . Then, the "automatic" 7 . . . 0-0 allows 8 dxc5 ! Bxc5 9 e4 followed by e5 with excellent attacking chances for White.

After 6 0-0 we analyze: I A6 . . . cxd4 and I B 6 . . . a6 . .J

lA (1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0)

6 ... cxd4

CHAPTER I 15

Black's plan is to isolate White's d-pawn, then immobilize it, and finally exploit it. This strategy was first used by Steinitz against Zuckertort ( 1 886), with great success. Black plays to slow or prevent the breakthrough d4-d5, while seeking simplifying exchanges to weather White's pressure.

The isolated pawn seems to be an excellent target - Steinitz felt that Black would gain the better chances. However, Tarrasch wisely countered that White's active piece play and possibilities of attack in the center and on the Kingside would lead to intense encounters.

Black can best exploit the isolated d-pawn in the ending, but it isn't easy to convince White to simplify so easily.

7 exd4 7 Nxd4 is innocuous; White needs the isolated d-pawn to stake out

central territory while freeing up the cl-h6 diagonal for the QB.

After 7 exd4 Black has: I AI 7 . . . Nc6 and I A2 7 . . . Be7.

I At ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 cxd4

7 exd4)

7 ••• Nc6 (See next diagram)

16 THE QUEEN'S GAMBIT ACCEPTED

Black attacks the d-pawn and thereby tries to hinder the plan Qe2 andRfd1.

8 Nc3 Also possible is 8 Qe2, gambiting the d-pawn. The game Matich­

Uhlmann, Zinnowitz 1966, continued 8 . . . Nxd4 9 Nxd4 Qxd4 10 Rdl Qb6 I I Bb5ch Bd7 12 Bxd7ch Nxd7 13 Nc3 Nf6. White has enough for the pawn, and held a slight advantage after 14 Be3 Qc6 15 Rae I a6 16 Nd5 Qxd5 1 7 Rxd5 .

8 ... Be7 Relatively best. With 8 . . . a6 Black does nothing for his development,

and White can play 9 Qe2. Then it's bad to take the pawn: 9 . . . Nxd4 10 Nxd4 Qxd4 I I Rd l Qg4 and White gets a strong attack, e.g. 1 2 Nd5 Qxe2 1 3 Nc7ch Ke7 14 Bxe2 Rb8 15 Bf4 Nd7 16 Bd6ch Kd8 17 Nxe6ch fxe6 18 Bxb8 wins the Exchange.

Instead of 9 . . . Nxd4, 9 . . . Be7 is playable. In this line, however, White secures the better chances with 10 Rd l , e.g. 10 . . . 0-0 I I d5 exd5 1 2 Nxd5 Nxd5 1 3 Bxd5.

9 Qe2 0-0 Too dangerous is 9 . . . Nxd4 10 Nxd4 Qxd4, e.g., I I Rd l (also strong

is 1 1 Nb5) I I . . . Qb6 (I 1 . . . Qg4 12 Bb5ch) 1 2 Be3 Qc7 13 Nb5 with a terrific position for White.

10 Rdl Probably best. In Spassky-Petrosian, World Championship Match

game 1966, (by transposition) White played 10 Be3 . Then I O . . . Na5! was good for Black: I I Bd3 b6 12 Bg5 (or 12 Ne5 Bb7 13f4 Nd5l) 12 . . . Bb7 1 3 Radl ReS 14 Rfe l (on 14 Ne5 Nd5 15 Bd2 Nb4 is good for Black) 14 . . . h6 ! 1 5 Bel Bb4! 16 Bd2 Bxc3 17 bxc3 Qd5.

CHAPTER I 1 7

1 0 ... Nb4 After IO . . . a6 then 1 1 d5 ! exd5 12 Nxd5 Nxd5 13 Bxd5 is good for

White. Or I O . . . Qc7 I I Bg5 followed by 12 Raci is strong for White.

Somewhat better is IO . . . Na5 I I Bd3 b6 I2 Bg5 Bb7 13 Rac i Rc8 ( 13 . . . Nd5?! 14 Qe4! g6 15 Qh4 f6 16 Bh6 Nxc3 1 7 bxc3 Bxj3 18 gxj3 j5 19 Qg3 Bd6 20 /4 RP 2 I ReI Qd7 22 Qe3 Bj8 [if 22 . . . ReS then 23 Qe2 and Bb5] 23 Bxj8 Kxj8 24 Qxe6 Qxe6 25 Rxe6 with a big advantage, Vukic­MaJjanovic, Nice I979). If now I4 Ba6 Bxa6 I 5 Qxa6 then I 5 . . . Nc4 is equal, but 14 Ne5 slightly favors White.

l l Bg5 In Botvinnik-Aiatortzev, Leningrad I932, White played I I Ne5.

After I I . . .N4d5 I2 Bg5 h6 (12 . . . Nxc3 13 bxc3 Nd5 14 Bd2 is good for White) 1 3 Bh4 Bd7 (or /3. .. Nxc3 14 bxc3 Nd5 15 Bxe7 and 15 . . . Nxe7 is forced, since 15 . . . Qxe7 is answered by 16 Ng6/) I4 Nxd5 Nxd5 (14 . . . exd5? 15 Bxf6 gxf6 1 6 Nxd7 Qxd7 1 7 Bd3) I 5 Bxe7 (not 15 Bxd5 exd5 16 Nxd7 ReB! 1 7 Bxe7 Qxd7 18 Rei Rac8! with Black on top) I5 . . . Qxe7 (15 . . . Nxe7 I 6 d5 I) I6 Ng6! fxg6 I7 Bxd5 and with Black's pawn structure shattered White is on top (Botvinnik's analysis).

However, after I I Ne5, better is I l . . .Bd7. For example, I 2 d5 exd5 I 3 Nxd5 N4xd5 I4 Bxd5 Nxd5 I5 Rxd5 Bg4! and Black's light-squared control gives good chances.

1 1 ... N4d5 12 Racl Nxc3

On 12 . . . b6 White plays 13 Nxd5 exd5 (13 . . . Nxd5? 14 Bxd5) I4 Bd3 with a strong initiative. Or if 12 ... Bd7 then 13 Nxd5 Nxd5 (13 . . . exd5 14 Bxd5) I4 Bxd5 Bxg5 1 5 Bxb7 Bxc l 16 Bxa8 Bxb2 17 Be4 White will win a pawn.

13 bxc3 Bd7

14 Ne5 Rc8 15 QO

White's position is preferable, Alekhine-Honlinger, Vienna 1936.

I A2 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 cxd4

7 exd4)

7 ..• Be7

18 THE QUEEN'S GAMBIT ACCEPTED

The development of the KB to e7 follows the same play as 7 . . . Nc6. Therefore, it often arises through transposition. This move has independent importance only when Black does not adopt Steinitz's plan of development. In the 9th Match game between Zuckertort and Steinitz ( I 886), Steinitz tried . . . Nbd7-b6-d5, . . . Qa5 and a Rook to d8 to work against the isolated d-pawn.

8 Qe2 0-0

9 Nc3 Nbd7

10 Rd1 The strongest reply to Black's plan.

10 .. . Nb6 On IO . . . a6 White gains the advantage with I I Bg5 or the more

energetic I I d5 .

1 1 Bb3 Nbd5 12 Bg5 Qa5 13 Racl Rd8 14 Ne5

Black has difficulty completing his development since (as Euwe points out) the natural I4 . . . Bd7 gives White a decisive advantage after I5 Nxd5 Nxd5 (15 . . . exd5 16 Bxf6 Bxf6 17 Rc5) 16 Bxe7 Nxe7 17 Nxd7 (/7 Q/3 is also strong) 17 . . . Rxd7 18 Bxe6! .

18

( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e4 e6 5 Bxc4 c5 6 0-0)

CHAPTER I

6 ... a6 j 19

This i s the popular way to play today, supplanting Steinitz's variation (lA), which - as we saw - leaves White with the better chances.

With 6 . . . a6 Black leaves his QN uncommitted and prepares Queenside expansion with . . . b5.

Now White has many choices: I 81 7 Qe2, I 82 7 Bb3, I 83 7 Nc3, I 84 7 a4, I 85 7 e4, I 86 7 a3, I 87 7 b3, I 88 7 dxc5 and I 89 7 Bd3 .

I 81 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 eJ e6 5 Bxc4 c5 6 0-0 a6)

7 Qe2

20 THE QUEEN'S GAMBIT ACCEPTED

White plans to protect the d-pawn with Rdl and prepare e4. Now Black can play: 11 �and I B l 2 7 . . . Nc6.

CHAPTER I

I Bl l ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2)

2 1

The most direct. Now White has two moves for the Bc4: I B 1 1 1 8 Bb3 and I 8 1 1 2 8 Bd3 .

I B l l l ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5)

8 Bb3

22 THE QUEEN'S GAMBIT ACCEPTED

On b3 the Bishop doesn't block the d-file, while exerting pressure on d5 and e6. Also, Black must be watchfu l for the breakthrough sacrifice at d5.

8 •.• Bb7 In the game Spassky-Karasev, USSR 1963, White played 9 Nc3 .

After 9 . . . Nbd7 Spassky essayed 10 e4?! . There followed IO . . . cxd4 1 1 Nxd4 Nc5 1 2 Rd l Nxb3 13 axb3 Nd7 14 NO! Qb6 15 Bg5 f6 16 Be3 Bc5 1 7 Nd4 0-0 1 8 b4 Bxd4 19 Rxd4 with advantage for White. But instead of I I . . .Nc5, correct was I I . . .Bd6, as after 12 Rd l Qb8! play transposes into I B l l l l 2 1 (JO . . . Bd6), note to White's 12th move.

After 8 . . . Bb7 White has: I 8 1 1 1 1 9 Rdl and I B l l l 2 9 a4.

I 81111 ( l d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7)

9 Rdl

Immediately utilizing the d-file with ideas of Nc3 followed by d5.

9 ... Nbd7 Now White has: I 8 1 1 1 1 1 10 e4 and I B l l l l2 10 Nc3 .

I 811111 ( l d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7 9 Rdl Nbd7)

CHAPTER I

10 e4

I -.I.

23

This pawn sacrifice, previously considered to pose problems for Black, has been defused by recent games.

10 .. . cxd4 Bad is l O . . . Nxe4 1 1 d5 ! Bxd5 1 2 Bxd5 exd5 1 3 Nc3 with a big

advantage for White. However, worth investigating is 1 0 . . . Bxe4 !?. For example,

l O . . . Bxe4 !? 1 1 Ng5 Bg6! (IJ . . .Bxb1 12 Nxj7 Qe7 13 Nxh8 c4 [on 1 3 . . . Bf5 14 dxc5 Nxc5 1 5 Be3 Nfd7 16 Racl White is on top; or if 1 5 . . . Nxb3 then 16 axb3 - threatens 17 Qxb5ch! as well as 17 Rxa6 and 1 8 Qxb5ch -16 . . . Qb7 1 7 Rac l g6 1 8 Bf4 Nd5 19 Be5 Bh6 20 Rc5 with White in control] 14 Bxc4 bxc4 15 Rxb 1 strongly favors White, Agzamov-Vera, 1984) 1 2 d5 (not 12 Bxe6? .fxe6 13 Nxe6 Qe7 14 Nc7ch Kj7 with Black on top) 1 2 . . . e5 1 3 Ne6 Qb6 1 4 Nxf8 Kxf8 and play is unclear, according to Gorelov.

1 1 e5 Nd5 Not 1 l . . .Ne4? 1 2 Bc2 d3 1 3 Bxd3 Nec5 14 Bc2 with clear advantage

to White, Magerramov-Tzchekov, 1982. In the game Timman-Seirawan, Indonesia 1 983, Black essayed

l l . . .Bxf3 !?. There followed 12 gxf3 Nh5 13 f4 g6 14 Rxd4 (on 14 j5 Black shouldn't play 14 . . . exj5? because of 15 e6!; but instead of 14 . . . exj5? he can play 14 . . . Ng7 and on 15 .fxe6 Nxe6 retaining a good position) 14 . . . Qb6 1 5 Rd 1 Rd8 1 6 Nc3 Be7? (a serious error; he had to play . . . Ng7) 17 f5 ! 0-0 1 8 Be3 Bc5 19 Rd6 ! Bxe3 20 Rxb6 Bxb6 2 1 Rd 1 ! with a powerful position for White.

However, in the game Salov-Kupreichik, USSR Championship 1 987, Black improved on 1 3 . . .g6 with 13 . . . Qh4 ! . There followed 14 Rxd4 Bc5 1 5 Rxd7! (on 14 Re4 Black has 15 . . . Bb6 with the follow-up . . . Nc5 and

24 THE QUEEN'S GAMBIT ACCEPTED

excellent play) 15 . . . Kxd7 16 Qf3 Rac8 17 Nc3 f5 18 Bxe6ch Kxe6 19 Qd5ch Ke7 20 Qb7ch Ke6 2 1 Qd5ch Ke7 22 Qb7ch and now Black, wanting more than a draw by perpetual, tried 22 . . . Kd8!? though the game was eventually drawn anyway.

1 2 Rxd4 Qc7 In the game Vera-Valdez, Cuba 1985, Black played 12 . . . Bc5. There

followed l 3 Rg4 Qc7 (13 . . . g6 14 Bh6 strongly favors White) 14 Bxd5!? (14 Bd2 comes into consideration) 14 . . . Bxd5 15 Nc3 Bxf3 16 Qxf3 Rc8 17 Rxg7 Nxe5 1 8 Qh3 Bd4! 19 Bf4 Qc4! 20 Ne4 Bxf2ch 21 Nxf2 Qxf4 22 Qh5 and now with 22 . . . Nd3 ! Black is winning. After this game, Vera gave 20 Qg3 as an improvement, claiming an edge for White. Nonetheless, in the game Perez-Valdez, Cuba 1987, when White tried out the move 20 Qg3 Black responded with 20 . . . Ng6 2 1 Ne4 Rc6! 22 Rxg6 hxg6 23 Nd6ch Rxd6 24 Bxd6 Rh5 and Black had the upper hand.

Another choice is 1 2 . . . Be7. Now the game Vaiser-Damjanovic, Vmjacka Banja 1984, continued 13 Nbd2 Qc7 14 Nfl 0-0 15 Rg4 Rfc8 16 Bh6 g6 1 7 h4 with a slightly better game for White. In the game Szymczak­Bernard, Posen 1984, White varied with l 3 Rg4 (instead of 13 Nbd2). There followed 13 . . . Qc7 14 Bd2 g6 15 Nc3 Nxc3 16 Bxc3 Nc5 17 Bc2 Bd5 18 Qe3 ReS 19 h4 Nd7 20 a3 Qb6 2 1 Qe2 Rc4 with mutual chances.

1986.

13 Bd2 N5b6

14 NcJ Bc5

15 Rg4 BxfJ 16 QxfJ Nxe5 17 QgJ Bd6 18 Ne4 Nec4

With at least even chances for Black, Michalchisin-Balashov, Minsk

CHAPTER I

I 81 1 1 12 ( 1 d4 d5 2 c4 d:xc4 3 NfJ Nf6 4 eJ e6 5 B:xc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7 9 Rd1 Nbd7)

10 Nc3

25

A natural developing mov'e, exerting pressure on d5 and e4. Now we examine: I 8 1 1 1 12 1 l O ... Qc7,"l\ 8 1 1 1 1 22 l O . . . Bd6, I B l l l l 23 I O . . . Qb6, I B l l l l 24 l O . . . Qb8 and I B l l 1 125 lO . . . b4.

I Bl l l l21 ( 1 d4 d5 2 c4 d:xc4 3 NfJ Nf6 4 eJ e6 5 B:xc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7 9 Rd1 Nbd7 10 Nc3)

10 ... Qc7 (See ne:xt diagram)

26 THE QUEEN'S GAMBIT ACCEPTED

A relatively old line, but it's still alive and kicking today!

1 1 e4 In the game Granda-Magem, Pamplona 199 I , White tried I I d5 .

There followed I l . . .c4! I2 dxe6 fxe6 13 Bc2 Bd6 14 Rd4 0-0 I5 Rh4 Nc5 16 e4 e5 17 Bg5 h6 18 Bxf6 Rxf6 19 Nd5 Bxd5 20 exd5 Raf8 and Black's strong pressure on the f-file gives him the better game.

1 1 ... cxd4 Not l l . . .b4 because of 12 e5! bxc3 13 exf6 and after either 13 . . . Nxf6

or 1 3 . . . gxf6 White has the strong thrust 14 d5. Also dangerous is the sacrificial 1 2 Nd5 ! , tried in Addison-Berliner, U.S. Championship 1962.

12 Nxd4 Nc5! Best. After I 2 . . . Be7 13 Bg5 (in Levitt-Ben Menachem, Holon I986,

White tried the speculative sac 13 Bxe6!?. There followed 13 . . . fxe6 14 Nxe6 and now instead of the actually played 14 . . . Qe5?!, correct was 14 . . . Qc6 and if 15 Nxglch Kj7 16 Nf5 Rhe8 play is approximately even; or if 15 Nd5!? then 15 . . . Kj7 leads to murky complications) l3 . . . b4 14 Na4 Qe5 15 Bxf6 Nxf6 I6 Nb6 Rd8 I7 Ba4ch K18 I8 Bc6 White held an edge, Smyslov­Keres, Budapest 1950.

13 Bel Bad is 1 3 Nd5? Nxd5 ! I4 exd5 Nxb3 15 Nxb3 Qc4 ! . Also, on 1 3 e5

Nfd7 I4 Bf4 Black must avoid I4 . . . Nxb3? since I5 axb3 Be7 I6 Raci Qb6 I7 Be3 Qa5 I8 f4 strongly favors White, Farago-Dobosz, Lodz I980.

Also inferior for Black is I4 . . . 8e7. The game Kharitonov-Gorbatov, Leningrad I99 I , continued I5 Bc2 b4 16 Ne4 Nxe4 17 Bxe4 Nc5 I8 Raci Qb6 I9 Bxb7 Qxb7 20 Qg4 g6 2 I Bg5 with tremendous pressure for White. Black's accurate response to this line is I4 . . . b4 ! ; e.g. the game Levitt­Chemin, Moscow 1988, continued 15 Na4 Nxb3 16 axb3 Bd5 I7 Raci Qb7

CHAPTER I 27

18 h4 Be7 19 Qg4 g6 20 Rd3 Rc8 and Black's Bishop pair gave him a perfectly satisfactory game.

In the game Christiansen-Anand, Las Palmas 1993, White tried 13 Bg5. There followed 1 3 . . .Bd6 (but not 13 . . . Nfxe4 because of /4 Nxe4 Nxe4 15 Nxb5 axb5 16 Qxb5ch Bc6 1 7 Ba4l Rxa4 18 Qxc6ch! Qxc6 19 Rd8 mate) 14 Rae 1 Bxh2ch 15 Kh 1 Be5 16 Bxf6 gxf6 17 Ncxb5 Qe7 18 Rxc5 Qxc5 19 Nxe6 fxe6 20 Qh5ch Ke7 2 1 Bxe6 Kxe6 22 Qg4ch Kn 23 Rd7ch Qe7 24 Qh5ch Kffi 25 Qh6ch Kn 26 Rxe7ch Kxe7 27 Qg7ch Ke6 28 Qg4ch Ke7 29 Qg7ch Ke6 30 Qxb7 axb5 3 1 Qd5ch Ke7 32 Qb7ch Ke6 33 Qd5ch and White forced a draw by perpetual.

13 .. . Be7 14 Racl 0-0 15 fJ

Somewhat better seems 15 Bc2 and 16 a3 - Pachman.

15 . . . Rac8

16 Bc2 Rfd8 17 a3 Qb8 18 Nb3 Nxb3 19 Bxb3 Bd6

20 g3 Be5 Black has equal chances, Salov-Chernin, Wijk aan Zee 199 1 .

I 81 1 1 122 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7 9 Rd1 Nbd7 10 Nc3)

10 ... Bd6

(See next diagram)

28 THE QUEEN'S GAMBIT ACCEPTED

Smyslov's innovation, which he introduced against Petrosian in the 1959 Candidates Tournament.

l l e4 On 1 1 d5 exd5 12 Nxd5 (12 e4? 0-0 favors Black) 12 . . . c4 1 3 Nxf6ch

Qxf6 14 Bc2 Ne5 15 Nxe5 Bxe5 16 Rbl 0-0 Black is better, Wexler­Olafsson, Buenos Aires 1960.

Another sequence is 1 1 dxc5 Bxc5 12 Ne5 Qc7 13 Nxd7 Nxd7 14 e4 0-0 1 5 Bg5. Now correct is 15 . . . Qe5 and on 16 Rxd7 Bc6! attacking both Rook and Bishop, or if 16 Bh4 then 16 . . . Nf6 17 Bg3 Qg5 is good.

1 1 ... cxd4 12 Rxd4!

Best. In Petrosian-Smyslov, Zagreb 1959, Petrosian played 12 Nxd4. After 1 2 . . . Qb8 13 NO b4 14 Nd5 exd5 15 e5 Nxe5 16 Nxe5 0-0! White had little compensation for the pawn.

Other 1 3th move choices (instead of 13 Nj3) give Black no problems: 1) 1 3 h3 0-0 14 a3 Rd8 1 5 Bc2 (15 Bg5 Be5) 15 . . . Bh2ch 16 Kh1 Bf4

17 Nb3 Bxc 1 1 8 Raxc 1 Ne5 19 Na5 Qc7 with a freer position for Black, Portisch-Reshevsky, Amsterdam 1964.

2) 13 g3 b4 14 Na4 Bxe4 (too dangerous is 14 . . . Nxe4 because of 1 5 j3 N4f6 1 6 Nxe6; however 14 . . . 0-0 is possible) 15 0 (to 15 Bg5 Black replies 15 . . . 0-0 and 16 Nxe6 is answered by 16 . . . Re8!) 15 . . . e5 16 Ne6 (on 16 .fxe4 exd4 1 7 Rxd4 0-0 Black is fine, Sokolov-Janosevic, Yugoslavia 1962) 16 . . . fxe6 17 fxe4 BeSch 18 Nxc5 (on 18 Kg2 Qb5, on 18 Be3 Black should avoid winning the Exchange with 18 . . . Bxe3ch 19 Qxe3 Ng4 20 Qf4 Qa7ch 21 Kg2 Ne3ch 22 Kh3 Nxd1 23 Rxd1 since White would get a strong attack. Instead of 18 . . . Bxe3ch, good is 18 . . . Qb5) 18 . . . Nxc5 19 Qc4 Qb5 and Black halts White's initiative, retaining a material advantage.

CHAPTER I 29

3) 13 e5?! (dubious) 13 . . . Bxe5 14 f4 Bxd4ch 1 5 Rxd4 0-0 16 Be3 Nb6 1 7 Radl Nbd5 18 Nxd5 Bxd5 with Black on top, Letelier-Ciarke, 1966 Olympiad.

12 ... DeS

13 Rd3 Ng4! Once thought to be inferior, this now has been given a vital boost by

Sadler.

14 Bg5 Qb6 The obvious 14 . . . Bxf2ch 15 Kfl Qb6 16 h3 N4e5 17 Rxd7! is very

strong for White.

15 Nd5 Bxd5 The only good move. After 15 . . . exd5 16 exd5ch Kf8 1 7 d6 Bxd6 (or

1 7 . . . Re7 18 Belch Kg8 19 Qe6!!) 18 Radl ReS 19 Be3 White has a tremendous position.

16 exd5 Bxflch! Sadler's improvement. The older move 16 . . . e5 (Veresov-Suetin,

USSR 196 1 ) is poor after 1 7 Bh4 0-0 18 Ng5 (not 18 h3 e4!) 1 8 . . . N4f6 19 Ne6 with strong pressure.

17 Kfl Nc5 18 dxe6

Now in the game Levitt-Sadler, London 1988, Black played 18 . . .f6!? and after 19 Bf4 a draw was agreed. GM Flear says that 18 . . . fxe6! may be even stronger.

I 8 1 1 1 123 ( 1 d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7 9 Rd1 Nbd7 10 Nc3)

10 ... Qb6 (See next diagram)

30 THE QUEEN'S GAMBIT ACCEPTED

This is currently very popular.

1 1 d5 In Haik-Piiester, 1984, White played I I e4. After l l . . .cxd4 1 2 Nxd4

Bc5 1 3 Be3 0-0 (also interesting is 13 . . . Ne5 14 a4 Neg4 15 e5 Nxe3 16.fxe3 Nd7 1 7 Qg4 h5 18 Qxg7 0-0-0 19 Qxfl Rhg8 20 Rd2 Bxd4 21 exd4 Rdj8 22 Qxh5 bxa4 23 Bd5 - if 23 Ba2 then 23 . . . Bxg2! 24 Rxg2 Qxb2! - 23 . . . exd5 24 Qh3 Rh8 25 Nxa4 Qg6 26 Qe3 Kb8?!- 26 . . . Qe6 is slightly better - 27 e6 Nf6 28 Qe5ch and Black won on time, Korchnoi-P. Nikolic, Reykjavik 1988) 14 n (or 14 a4 Rad8 15 axb5 axb5 16 Qxb5 Qxb5 1 7 Ncxb5 Bxe4 with even chances, Haik-Pieister, 1984) 14 . . . Ne5 1 5 Rd2 Rfd8 16 Rad1 Rd7, with equality, Bouchaud-Fayard, Torey 199 1 .

I n the game Savon-Ribli, Barcelona 1989, White tried I I Ne5. The game continued 1 l . . .Rd8 12 f4 Be7 13 a4 b4 14 a5 Qa7 15 Na4 0-0 16 Bc4 cxd4 17 exd4 Nd5 with even chances.

Finally, if White tries to probe Black's Queenside with 1 1 a4 then Black can conveniently respond with 1 l . . .b4 12 a5 Qc7 1 3 Na4 cxd4 14 exd4 Bd5 with a satisfactory position.

1 1 ... Nxd5 Best. After 1 l . . .e5 12 Bc2 (or 12 Ng5 Bd6 13 Bc2!? [ 1 3 Nce4 is

slightly better for White] 13 . . . 0-0 14 Q/3 Rad8 15 Bj5 h6 16 Nge4 Rfe8! 1 7 b3 Bj8 1 8 Bb2 g6 1 9 Bxd7 Nxd7 with equality, Nogueiras-Sisniega, Novi Sad 1990) 1 2 . . . c4 13 Bf5 !? Bd6 14 Bxd7ch Nxd7 1 5 e4 Bc5 16 Rb1 ! 0-0 17 b4 Bd4, as in Lobron-Nikolic, Novi Sad 1990. Now Lobron gives 1 8 Be3 ! Bxe3 19 Qxe3 Qxe3 20 fxe3 with an edge for White.

12 Bxd5 Bxd5

13 Nxd5 exd5

14 Rxd5 Be7

CHAPTER I 31

Also good is 14 . . . Qb7 15 e4 (in the game P. Nikolic-Seirawan, Manila Olympics 1992, White tried the aggressive Rook maneuver 15 Rg5, but after 15 . . . g6 16 e4 Bgl 1 7 e5 Nj8 18 a4 Rb8 19 axb5 axb5 20 e6 Nxe6 21 Rxc5 0-0 Black had equal chances) 15 . . . Be7 (15 . . . Nb6 16 Rh5 Bel 1 7 e5 0-0 18 e6l Bf6 19 Ng5 Bxg5 20 Rxg5l Rae8 21 Qg4 f6 22 Rxc5 Qdl 23 Re5l Qd6 24 Bh6 is strong for White, Cifuentes-Siipak, Mar del Plata 1990) 16 Bg5 f6 1 7 Rad l Nb6 18 Bf4 0-0! 1 9 R5d2 Rfe8 20 e5 and now Ivanchuk gives 20 . . . c4! with Black having at least equality. In the game Vaganian­Garcia Palermo, Reggio Emilia 1992, Black "tempted fate" by accepting White's sacrifice with 18 . . . Nxd5?1 (instead of the aforementioned 18 . . . 0-0l). There followed 19 exd5 0-0 20 d6 Bd8 2 1 d7 Bc7 22 Qe6ch Rf7 23 b4 ! c4 24 Be3 and Black was in real trouble and lost.

Also good is 1 5 . . . Qc6. 15 e4 Rd8

16 Be3! Qb7! On 16 . . . Qe6 17 a4! Qxe4 18 Rad l White is well on top. In the game

Damjanovic-Ehlvest, Manila Interzonal 1990, 16 . . . 0-0 was played. After 1 7 b4 Qc61 1 8 Radl ! Nf6 (or 18 . . . cxb4 19 Nd4 Qb l 2 0 N.f5 Nf6 21 Qb2l Rfe8 22 Bg5l with advantage to White) 19 Bxc5 Rde8 20 Bxe7 Rxe7 2 1 Rd6 Qxe4 22 Qxe4 Rxe4 23 a3 Ra8 24 Nd2! Re2 25 Kfl Rae8 26 h3 h6 27 Rxa6 White had a winning ending.

1 7 Rad1 0-0 18 Qd2

Piling up on the Knight, but Black has a clever way out.

18 ... Qa8! Now White can't play 19 Rxd7, as after 19 . . . Rxd7 20 Qxd7 Rd8 2 1

Qxd8ch Bxd8 22 Bxc5 Bc7 Black is on top.

19 Qc2 In the game Zviagintsev-Ivanov, Moscow 199 1 , White played 19

Bg5. After 19 . . . Nf6 20 Bxf6 Bxf6 2 1 b3 Qc8 22 Qc2 Qe6 Black has a sound position (though he managed to play inaccurately and lose).

19 ... Nf6 20 Rxd8 Rxd8

Play is dead even, in Salov-Ivanchuk, Linares 1990; the players agreed to a draw here.

32 THE QUEEN'S GAMBIT ACCEPTED

I 81 1 1 124 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7 9 Rd1 Nbd7 10 Nc3)

10 .. . Qb8

Note that both IO . . . Qb6 (I B I I I I 23) and IO . . . Qb8 quite often transpose into the same position after several moves. However, there are several earlier branching-off points of independent significance.

1 1 d5 On I I Ne5 Nxe5 I2 dxe5 Black mustn't play I 2 . . . Qxe5? 13 Nxb5 ! as

in Boban-Kindljic, 198 I , but instead 12 . . . Nd7 1 3 f4 Bc6! 14 a4 c4 I 5 Bc2 Qb7 I 6 axb5 axb5 I7 Rxa8 Qxa8 18 Qd2 Nc5 I9 f5 !? exf5 20 Nd5 Bxd5 2 I Qxd5 Qxd5 2 2 Rxd5 Be7 with even chances, Barczay-Brilla Banfalvi, postal game I 983 . In Puskov-Razuvaev, Rostov I993, Whiteplayed I I h3, but this is no threat to Black. After I I . . .Be7 I2 d5 exd5 13 Nxd5 Nxd5 I4 Bxd5 Bxd5 I5 Rxd5 Qb7 I6 Rd i 0-0 17 b3 Rfd8 Black had fully equal play.

1 1 . . . exd5 Another way is I l . . .c4 !?. The game Schwarts-lbragimov, Podolsk

I990, continued I2 dxe6 fxe6 1 3 Bc2 Bd6 14 e4 0-0 15 Bg5 (or 15 a4?! Ne5 16 Nxe5 Bxe5 1 7 axh5 axb5 18 Rxa8 Bxa8 and Black has a good game; from Tozer-Shaw, British Championship I992) I5 . . . Ne5 I6 Nxe5 Bxe5 I7 h3 Qc7 I 8 Rd2 Rf7 I9 Rad l Raf8 20 a3 Bc6 2 1 Qe3 Nd7 with good play for Black (who went on to win).

12 Nxd5 Sharper is I2 e4 but since some very strong players have been

accepting this position for Black, it seems that I2 e4 is not to be feared.

CHAPTER I 33

Black has an extra pawn and the onus is on White to show any compensation.

move.

12 .•• Nxd5

13 Bxd5 Bxd5

14 Rxd5 Be7

15 e4 Qb7 On 15 . . . Nb6 play can transpose into I B 1 1 1 123, note to Black's 1 4th

16 Bg5 Nb6 Note that 16 . . . f6 is fully playable - see I B1 l l l 23 , note to Black's

14th move.

1 7 Rad1 Or 17 Re1 f6 (not 1 7 ... Nxd5? 18 exd5 f6 19 d6 .fxg5 20 Qxe7 Qxe7

21 Rxe lch Kj8 22 Nxg5 and Black must resign - mate by Ne6ch and Rxg7 is too much to handle) 18 Bf4 0-0 19 Rh5 g6 20 Rh3 Rfe8 with a good position for Black, Brooks-Rachels, US Championship 199 1 .

After 17 Bxe7?! Nxd5 18 Bxc5 Nf4 19 Qe3 Ne6 2 0 Bd6 f6 2 1 Qb3 Kf7 Black is better, Donner-Portisch, Varna 1%2.

In the game Lin Ta-Rachels, Manila Interzonal 1990, White tried 1 7 Rd3, but after 1 7 . . .f6! 1 8 Bf4 c4 19 Rdd 1 0 -0 20 Nd4 Rfe8 2 1 Ne6 Bf8 22 Qg4 Qc8 2 3 Rd6 h5 24 Qxh5 Rxe6 25 Rxb6 Rxb6 Black won.

1 7 ... f6! But not 17 ... Nxd5 18 exd5 f6 19 d6 fxg5 20 Re1 Rd8 2 1 Qe6 Rd7 22

Nxg5 with a terrific position for White, Neronsky-Schuk, postal game 1 966.

18 Bf4 0-0 Interesting is 1 8 . . . Nxd5!? 19 exd5 0-0 20 d6 Bd8 2 1 d7 Bc7 22

Qe6ch R1i 23 Be3 Qb6 with unclear play, Donner-Szymczak, Prievidza 1980.

19 RSd2 Rfe8 In Yijola-Agzarnov, Sochi 1984, Black played 19 . . . Rad8 and after 20

e5 Rxd2 21 Rxd2 Nc4 the players agreed to a draw.

20 e5 fxe5 21 Qxe5 Bf8 22 Qg5 Qe7 23 Qxe7 Rxe7

The position is equal, lvanchuk-Seirawan, Tilburg 1990.

34 THE QUEEN'S GAMBIT ACCEPTED

I 81 1 1 125 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 eJ e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 BbJ Bb7 9 Rd1 Nbd7 1 0 NcJ)

10 .. . b4

Driving away the Nc3 and thus hindering e3-e4.

1 1 Na4 Qa5 1 2 Bd2!

An innovation by Mikenas. In Keres-Spassky, Amsterdam 1956, 1 2 e4 was played. Now 12 . . . Nxe4 is bad because of 13 d5, while 1 2 . . . Bxe4 is strongly met by 13 Ng5. Spassky played 12 . . . Qb5, then after 13 Qe 1 Bxe4 (13 . . . Nxe4 14 d5!) 14 Ne5 (now 14 d5 is met by 14 . . . c4!) 14 . . . c4 1 5 Nxc4. Now Spassky played 1 5 . . . Be7 but after 16 Bf4! 0-0 17 Nd6 White was on top. Better was 1 5 . . . Bd5 16 Ne5 Be7 with only a slight pull for White.

1 2 . . . Be7! Better than 12 . . .Qb5 13 Qxb5 axb5 14 Nxc5 Nxc5 1 5 dxc5 Bxc5 16

Rac1 Rac8 17 Ne5 with an edge for White - Black's pawns are weak. In Spassky-Mikenas, USSR 1963, Black tried 12 . . . cxd4 but after 13

Nxd4 Be7 14 a3 0-0 15 axb4 Qg5 16 n Qh5 17 e4 White was clearly superior.

After 1 2 . . . Be7! 1 3 a3 has little effect: 1 3 . . . Qb5 ! 14 Qxb5 axb5 1 5 Nxc5 Nxc5 16 dxc5 bxa3 and the ending is about equal.

CHAPTER I

I 81 1 12 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 Bb3 Bb7)

9 a4

Probing Black's extended Queenside pawns.

9 . .. Nbd7

10 axb5

35

After the sharp gambit move lO e4 Black mustn't play l O . . . Bxe4 because of 1 1 Ng5! or l O . . . Nxe4 1 1 d5 ! . Correct is l O . . . cxd4 1 1 e5 Ng4 1 2 axb5 Bc5 1 3 bxa6 0-0! 14 axb7 Rxa l 15 Qe4 and Black i s slightly better, Petrosian-Smyslov, USSR 1950.

After 10 Rd l Black has lO . . . Qb8! . Now 1 1 axb5 axb5 1 2 Rxa8 Bxa8 leads to equality. The game Loffier-Yakovich, Munich 1992, continued 1 1 e4 cxd4 1 2 Nxd4 and now 12 . . . Bd6! gives Black a very satisfactory position (1 2 . . . Bc5 was actually played and led to a draw in the game).

10 .. . axb5 1 1 Rxa8 Qxa8 12 Nc3 b4 13 NbS BxfJ!

Best. The older choices 13 . . . Qb8 and 1 3 . . . Qa5 are not as good.

14 gxfJ Qb8 15 Rd1 Be7

16 e4 cxd4 1 7 Nxd4 Bd6

Not 1 7 . . . 0-0? because of the winning shot 18 Nxe6! fxe6 19 Bxe6ch.

36 THE QUEEN'S GAMBIT ACCEPTED

18 eS Anand later gave 18 h3 0-0 19 Qb5 as better; but GM Flear states

that 19 . . . Rc8 is equal.

18 •.. BxeS

19 Nxe6 White tries to exploit Black's centralized King, but - as we shall see -

the Black defenses are sufficiently strong.

19 . . . fxe6

20 f4 Ke7 Not 20 . . . Bxf4? 2 1 Qxe6ch Kd8 22 Ba4 with an overwhelming attack

for White. Anand suggests 20 . . . Kf7!?.

21 Qc4 22 fxeS 23 Bd2

NdS QxeS N7b6

Kf7 Qxd4 Nxb4 NdS

here.

24 Bxb4ch 2S Qd4 26 Rxd4 27 Rxb4 28 Rb7ch

And the game Yusupov-Anand, Las Palmas 1993, was agreed drawn

CHAPTER I

I Bll2 ( 1 d4 d5 2 c4 dxc4 J NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5)

8 Bd3

37

By staying on the fl -a6 diagonal, White hopes to pressure the Black Queenside pawns with a following a4. Now the "usual" 8 . . . Nbd7 is not so good: 9 a4 b4 (or 9 . . . c4 /0 Bc2 Bb7 J l e4 Bel /2 Bg5 ReB /3 Rdl with a big plus for White, Lilienthai-Landau, Matchgarne 1934) 10 Nbd2 Be7 I I e4 cxd4 1 2 e5 Nd5 1 3 Nb3 Bb7 14 Bd2 0-0 1 5 Rfcl Qb6 16 aS Qa7 1 7 Rc4 ! g6 18 Rxd4 with a powerful position for White, Ribli-Xu Jun, Subotica Interzonal 1987.

After 8 Bd3 we examine: I B l l 2 1 8.Jc6 and I B l l 22 8 . . . cxd4.

I Bll21 ( 1 d4 d5 2 c4 dxc4 J NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 BdJ)

8 ... Nc6 (See next diagram)

38 THE QUEEN'S GAMBIT ACCEPTED

Placing pressure on d4. Now if 9 dxc5 then 9 . . . Bxc5 10 a3 Qc7! 1 1 b4 Bd6 1 2 Bb2 Ng4 gives Black equality according to GM Henley.

9 Nc3 After 9 a4 the game Polugaevsky-Toprover, 1959, continued 9 . . . bxa4

10 Nc3 (10 Bc2!? Nb4 1 1 Ba4ch Bd7 is equal) IO . . . Nb4 1 1 Bc4 Bb7 1 2 Rdl Qc7 1 3 dxc5 Bxc5 14 Nxa4 Be7 1 5 b3 0-0 16 Bb2 with equal play.

9 ••• ud4 Another good line is 9 . . . Bb7 10 dxc5 Bxc5 1 1 e4 Nd4 12 Nxd4 Bxd4

with equality according to GM Hubner.

10 exd4 Nb4 1 1 Bb1 Be7

Hubner gives equality here.

I Bl 122 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 b5 8 Bd3)

8 ... cxd4 (See next diagram)

CHAPTER I 39

As in many variations of the Queen's Gambit Accepted, Black gives White an isolated pawn. White (as is usual) must accept the isolani since 9 Nxd4 loses pawn control of the center.

9 e:xd4 White obtains nothing with 9 Nxd4. For example, the game Granda­

Magem, Madrid I992, continued 9 . . . Bb7 IO Nd2 Nbd7 I I N2b3 e5 ! 1 2 Nf5 g6 13 Ng3 e4 14 Bc2 h5 with Black on top.

9 ... Nc6 Insufficient is 9 . . . Bb7; Spanjaard-Devos, Deveter 1959, continued IO

a4 bxa4 1 1 Bc2! Be7 1 2 Bxa4ch Nbd7 (better was 12 . . . Bc6 though after 13 Bxc6ch Nxc6 14 Rxa6 Nxd4 15 Nxd4 Rxa6 16 Qxa6 Qxd4 1 7 Be 3 White retains some advantage) and Black was in serious difficulties.

In the game Lputian-Meister, USSR Championship 199 I , Black played 9 . . . Be7. There followed 10 a4 bxa4 I I Nc3 ! (an improvement on the older 1 1 Rxa4 0-0 12 Nc3 Bb 7 13 Rd1 a5 14 Ne5 Nc6 15 Bg5 Nb4 with even play, as seen in Nogueiras-Ehlvest, Interzonal 1987) I l . . .0-0 12 Nxa4 Nc6 1 3 Rd l Nb4? (this allows White's QR to shift to the Kingside with powerful effect) 14 Bb l Bb7 15 Ra3 ! Rc8 16 Ne5 Qd5 17 Rg3 with much the better game for White.

10 a4 bxa4

1 1 Rxa4 Nb4 12 Bb5ch

Or 1 2 Bc4 Be7 13 Bg5 a5! 14 Bb5ch Bd7 15 Ne5 !? 0-0 16 Rxb4 (16 Bxf6 Bxb5 is equal) 16 . . . axb4 17 Bxd7 Nxd7 18 Nc6 Bxg5 19 Nxd8 Rfxd8 with complex play and mutual chances, Portisch-Seirawan, Olympiad 1 986.

12 •.. Bd7

13 Bxd7ch Qxd7

40 THE QUEEN'S GAMBIT ACCEPTED

14 Nc3 Be7

15 BgS Qb7! Better than 15 . . . 0-0? 16 Bxf6 gxf6 17 Ra5! Kh8 18 Rh5 with a

tremendous position for White, Kozul-Psakhis, Zagreb Zonal 1993.

16 Bxf6 Bxf6

1 7 Ne4 Be7

18 Nc5 Bxc5

19 dxc5 0-0 As in Pergericht-Garcia Palermo, Brussels 1985. Play is about equal.

CHAPTER I

I 812 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 8xc4 c5 6 0-0 a6

7 Qe2)

7 ... Nc6

41

An active system; Black intends to exchange the dangerous light­squared Bishop by . . . b5, . . . c4 and . . . Nb4.

White has several replies: I B l 2 1 8 Rdl , I B l 22 8 Nc3, I B l 23 8 dxc5 and I B l 24 8 a3.

I 8121 ( 1 d4 d5 2 c4 d:x:c4 3 NfJ Nf6 4 e3 e6 5 B:x:c4 c5 6 0-0 a6

7 Qe2 Nc6)

8 Rd1 (See next diagram)

42 THE QUEEN'S GAMBIT ACCEPTED

Allowing the exchange of the light-squared Bishop in return for a pawn center.

8 ... b5 9 BbJ

With 9 dxc5, White plans to attack the opposing Queenside with a4 (the Bc4 will drop back to d3}, e.g. 9 dxc5 Qc7 10 Bd3 Nb4 1 1 a4 bxa4 (not J J . . . Nxd3 12 Qxd3 b4 13 c6!) 12 Rxa4 Rb8 13 Nc3 Bxc5 with equal chances according to Taimanov.

After 9 Bd3 Black plays 9 . . . cxd4! 10 Nxd4 (10 exd4 Nb4 exchanges off the Bishop} 10 . . . Nxd4 1 1 exd4 Bb7 with fully satisfactory play.

Be6.

line.

Also, 9 d5 is met by 9 . . . exd5 10 Bxd5 Nxd5 1 1 e4 Qe7! 12 Rxd5

9 ... c4! 10 Bel Nb4 1 1 NcJ

Or 1 1 e4 Nxc2 12 Qxc2 Bb7 13 d5, with transposition into our main

1 1 .. . Nxcl 12 Qxcl Bb7 13 d5

In Szabo-Euwe, Groningen 1946, White tried 1 3 e4?! , but after 1 3 . . . b4 14 e5 bxc3 15 exf6 gxf6 16 Qa4ch Qd7 17 Qxc4 Rc8 Black was on top. In the game M. Tseitlin-Kudshievich, USSR 1987, White tried to improve on the Szabo-Euwe game by playing 16 d5 !? (instead of 16 Qa4ch). There followed 16 . . . Bxd5 17 bxc3 Qa5 18 Rb1 Bg7 19 Be3 0-0 20 Bb6 Qa3 21 Rb4 a5 22 Rb5 Rfc8 23 Nd4 Bffi 24 Rdb l Ra6 25 Rxd5 ! exd5 26 Nf5 with a very dangerous initiative for White. But Black can improve

CHAPTER I 43

with 17 . . . Bg7 !? and if 18 Ba3 then l 8 . . . Bffi; also, instead of 18 . . . Bg7 Black should have played 18 . . . Rg8! .

13 ... Qc7! The idea of this move, introduced by Flohr in a game against

Reshevsky (Nottingham 1936), is to meet 14 e4 with the blocking move 14 . . . e5.

Too risky is 1 3 ... exd5 14 e4 Be7 15 e5 Nd7 16 Nxd5 0-0 17 Qf5 Nc5 18 Nf6ch Bxf6 19 Rxd8 Bxd8 20 Ng5 Bxg5 2 1 Bxg5 with clear advantage for White, EJ,Jwe-Grunfeld, Zandvoort 1936.

14 e4 White gets nothing with 14 dxe6 fxe6 15 Nd4 Kf7, Knip-Kmoch,

Amsterdam 194 1 , or 1 5 Ng5 Qc6 16 f3 Be7.

14 ... e5 15 Bg5

On 1 5 Be3 Ng4 is good.

15 ... Nd7

16 Racl In Reshevsky-Fiohr 16 Be3 was played; then after 16 . . . Bc5 17 Bxc5

Qxc5 18 b3 0-0 19 bxc4 Qxc4 Black was a hair better. Alekhine suggested that 16 . . . Bd6 was more accurate.

16 ... Bd6

17 Ne2 0-0

18 Ng3 Rac8 Best. After 1 8 . . . f6 19 Be3 g6 20 h4 Rac8 2 1 h5 White has a strong

attack, Stahlberg-Aiexander, London 195 1 .

19 NfS Rfe8 The position is equal according to Neistadt.

I 8122 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 Nc6)

8 Nc3 (See next diagram)

44 THE QUEEN'S GAMBIT ACCEPTED

White leaves the square b I for the light-squared Bishop.

8 ... b5 After 8 . . . Qc7 9 Bd3 ! Be7 lO dxc5 Bxc5 I I Ne4 Be7 1 2 b3 Nxe4 1 3

Bxe4 Bd7 (13 . . . 8/6 1 4 Bb2 Bxb2 1 5 Qxb2 is very strong for White -Timman) 14 Bb2 0-0 15 Racl Bd6 16 Rfdl f5 17 Qd2 ! fxe4 18 Qxd6 Qxd6 1 9 Rxd6 exf3 20 Rd7 White had a big advantage, Timman-Miles, Tilburg 1986.

9 BbJ Bb7 Or 9 . . . Be7 l O dxc5 Bxc5 I I e4 b4 12 e5 bxc3 1 3 exf6 gxf6 1 4 Qc4

Qb6 (as in Euwe-Aiekhine, Matchgame 1937) now Taimanov gives 1 5 Ba4 ! Bd7 16 bxc3 with a clear advantage to White.

10 Rd1 Na5! Clearly best. After IO . . . Qb6 I I d5 exd5 12 e4 ! d4 13 e5 0-0-0 14 exf6

dxc3 1 5 Bf4 c4 16 Bc2 g6 17 bxc3 White is in control, Furman-Byshev, 1952.

On lO . . . Qc7 I I d5 exd5 12 e4 Black should try 12 . . . 0-0-0! (but not 12 . . . dxe4? 13 Nxe4 Nxe4 14 Qxe4ch Qe7 15 Qf4 Nd8 16 Rei Ne6 1 7 Qg4 c4 18 Bg5! with a big advantage for White, Kotov-O'Kelly, Groningen 1946). This needs some testing, though.

1 1 Bel Qb6 12 e4

Or 12 dxc5 Bxc5 13 a3 Be7 14 e4 ReS 15 Ne5 draw, Tatai-Velikov, Rome 1983.

12 ... cxd4 13 Nxd4 Bc5 14 Bel! 0-0

Not 14 . . . Nc4? 15 Na4! .

CHAPTER I 45

15 eS Nd7

16 Ne4 Nxe5

1 7 Nxc5 Qxc5

18 Racl ! Taimanov assesses the position as unclear.

I 8123 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 Nc6)

8 dxc5

White intends to follow this up with the thrust e4 and e5 to gain central space.

8 ... 9 e4 tOeS 1 1 Bf4 12 Nbd2! 13 NxeS 14 Bg3

Bxc5 Qc7 Ng4 (6 Ngxe5 fxeS

Better was 14 Be3 ! Nd4 1 5 Qh5ch g6 16 Qh6 with a small edge to White.

14 ... 0-0 15 Racl Qe7

46 THE QUEEN'S GAMBIT ACCEPTED

16 Bd3 Bd7! 17 NO

On 17 Qh5 Black has 17 . . . Rf5! .

1 7 ._ Bd4

18 Nxe5 Nxe5 19 BxeS BxeS 20 QxeS Bc6

Play is equal, Nogueiras-Seirawan, Montpellier 1985.

I B124 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 Qe2 Nc6)

8 a3

Planning dxc5 followed by b4. White waits with dxc5 to gain a tempo if Black moves his KB in the interim.

8 ... bS 9 Ba2 Bb7

10 dxc5 Bxc5

1 1 b4 Be7

12 Bb2 0-0 13 Nbd2 Qb6 14 Nb3 Rfd8 15 Racl

CHAPTER I 47

As in Eliskases-Muffang, Warsaw 1935. Now 15 . . . a5 ! gives Black satisfactory play.

48 THE QUEEN'S GAMBIT ACCEPTED

I 82 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6)

7 Bb3

A finesseful idea first seen in several games from the Varna I962 Olympiad and the Botvinnik-Petrosian Match ( 1963). Now 7 . . . b5 is met by 8 a4 ! , e.g, 8 . . . Bb7 9 axb5 axb5 10 Rxa8 Bxa8 I I Na3 with clear advantage for White. Also, if 8 . . . b4 then White's pieces gain a fine post at c4.

Finally, on 8 . . . c4 9 Bc2 White follows up with 10 b3 with great pressure on Black's Queenside pawns.

7 .. . Qc7 8 Qe2 Nc6 9 Nc3 Bd6 10 Bd2 0-0 1 1 Racl b6

12 dxc5 Bxc5 13 Bc2 Bb7

14 Ne4 Be7 As in Langeweg-Radulov, Amsterdam I973 . Black has a satisfactory

position.

CHAPTER I 49

I BJ ( 1 d4 d5 2 c4 du4 3 Nf3 Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6)

7 Nc3

This move can transpose into any of several earlier variations. We give a typical example of such a transposition.

7 ••. b5 8 BbJ

8 Be2 presents no problems; the game Andersson-Tarjan, Indonesia 1 983, continued 8 . . . Bb7 9 b3 Nbd7 10 Bb2 Bd6 1 1 dxc5 Bxc5 1 2 Rc1 0-0 with even chances.

8 ... Bb7 9 Qe2

After 9 a4?! the game Korchnoi-Gurgenidze, 34th USSR Championship, continued 9 . . . c4 10 Bc2 b4 1 1 Ne2 Nbd7 with excellent chances for Black.

9 ... Nbd7

10 Rd1 Qb8 1 1 d5 exd5!?

And after 1 l . . .Nxd5 1 2 Nxd5 Bxd5 13 Bxd5 exd5 14 Rxd5 we have transposed into I B 1 1 1 123.

50 THE QUEEN'S GAMBIT ACCEPTED

I 84 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6)

7 a4

First employed by Rubinstein, and in modem times by Botvinnik. White prevents . . . b5 but at the cost of weakening b4 and b3.

7 . . . Nc6 The natural place for the Knight, pressing toward the weakened b4

square. Now White has two lines: I 84 1 8 Nc3 and I 842 8 Qe2.

I 841 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6)

8 Nc3 (See next diagram)

Ne5.

CHAPTER I

This move gives Black no real problems.

8 . . . Be7

B t

51

Now White can play: I 84 1 1 9 Qe2, I 84 12 9 dxc5 and I 84 1 3 9

I 841 1 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Nc3 Be7)

9 Qe2

Preparing to shift the KR to the d-file. The immediate central push 9 d5 is well met by 9 . . . exd5 10 Bxd5 Bg4! with equality.

9 . . . cxd4

52 THE QUEEN'S GAMBIT ACCEPTED

10 Rd1 e5 1 1 exd4 exd4 12 Nxd4 Nxd4 13 Qe5 Qd6

Best. Not as reliable is 1 3 . . . 0-0!?. For example 14 Rxd4 Qb6 15 Be3 Bc5 (not 15 . . . Qxb2 16 Qxe7l Qxa1ch 1 7 Rd1 Qxc3 18 Qxpchl) 16 Ne4! Bxd4! 1 7 Bxd4 Qd8 18 Rd1 Be6? (a serious error. Correct was 18 . . . Ne8l 19 Bd5l Qc7 20 Qh5l with unclear play) 19 Rd3 ! ReS 20 Nxf6ch Qxf6 21 Qxf6 gxf6 22 Rg3ch � 23 BeSch with a powerful game for White, Gauglitz­Gelfand, Halle 1987.

14 Qxd6 In Georgiev-Semkov, 1985, White tried 14 Qxd4 Qxd4 1 5 Rxd4 Bc5

(In Vyzmanavin-Greenfeld, Pardubice 1993, Black varied with 15 . . . 0-0 and after 16 Bg5 Bj5 1 7 Re1 Rfe8 18 Re5 Bg6 19 h3 h6 20 Bh4 Rac8 21 Bb3 Kj8 22 Re2 a draw was agreed) 16 Rf4 and now 16 . . . 0-0 17 Nd5 Nxd5 18 Bxd5 Rb8 19 Rc4 Bd6 gives equality.

14 ... 15 Rxd4 16 Rh4 17 Bf4 18 Bxe5 19 Rd1

Bxd6 Be5 0-0 ReS Rxe5

The chances are even, Tukmakov-Gulko, USSR 1985.

I 8412 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 S Bxc4 cS 6 0-0 a6

7 a4 Nc6 8 Nc3 Be7)

9 dxcS (See next diagram)

CHAPTER I

This is rather innocuous.

9 • • .

10 Rxd1 1 1 Bd2

Qxdl Bxc5 b6

53

More energetic is l l . . .Na5 ! . For example, Smejkal-Chernin, Moscow 1 989, continued 1 2 Ba2 b6 1 3 Ne2 Nc6 14 aS Bb7 1 5 Rdc1 Nd7 16 Ned4 (or 16 axb6 Bxb6 1 7 Ned4 Nxd4 18 Nxd4 Nf6 19 Bb4 Nd5/ with an edge for Black) 16 . . . Nxd4 17 Nxd4 b5! 18 Nc2 ! Ke7 with even chances.

1 2 Racl Bb7 13 Na2 0-0 14 Bel a5 15 Nc3 Rfd8 16 Kfl Kf8 1 7 Bb5 Rxd1 18 Rxdl Ke7 19 h3 h6 20 Nd2 Rd8 21 Rcl Nb4

With equality, Karpov-Hjartarson, Candidates Matchgame, Seattle 1 989.

I B413 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Nc3 Be7)

9 Ne5

54 THE QUEEN'S GAMBIT ACCEPTED

This immediate occupation of e5 is not dangerous to Black.

9 ... cxd4 10 Nxc6 bxc6 I I exd4 a5 12 QfJ Nd5

Black's Nd5 "stands like a house" while blocking diagonal pressure on c6.

13 Rdl Ba6 14 b3 0-0 15 Ne4 Nb6 16 Bxa6 Rxa6 17 Ba3 B:x:a3 18 Rxa3 Nd5 19 Raal Rb6

Black has a fine game and went on to win, Gohii-Baburin, Liechtenstein 1993.

I 842 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6)

8 Qe2 (See next diagram)

CHAPTER I 55

Preparing the "usual" Rfd 1. Now Black has three replies: I 842 1 8 . . . cxd4, I B422 8 . . . Qc7 and I 8423 8 . . . Be7.

I 8421 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Qe2)

8 . . . cxd4

Playing to isolate White's d-pawn.

9 Rd1 Be7 Not good is 9 . . . d3 10 Bxd3 (also good is 10 Rxd3) IO . . . Qc7 I I Nc3

Be7 1 2 h3 0-0 13 e4 Nd7 14 Be3 Nb4 15 Bc4 Nc2 16 Qxc2 Qxc4 17 Rac 1 with White well on lop, Zagorovsky-Romanov, postal game 1966.

56 THE QUEEN'S GAMBIT ACCEPTED

10 exd4 Less ambitious is 10 Nxd4 Nxd4 I I exd4 0-0 (or JJ . . . Bd7 I 2 Nc3

Bc6 equal - Ribli) 12 Nc3 Qd6 13 Bg5 Nd5 14 Bxe7 Nxe7 1 5 d5 exd5 16 Nxd5 Nxd5 17 Bxd5 draw, Nikolic-Ljubojevic, Belgrade 1987.

10 .. . 0-0 1 1 Nc3

Interesting is I I Bg5 Nd5 12 Bxe7 Ncxe7 13 Ne5 b6! 14 a5 (14 Nd2!? - Botvinnik) 14 . . . Bb7 15 axb6 Qxb6 16 Nc3 Nxc3 17 bxc3 Ng6 18 Bd3 Nxe5 19 Qxe5 Rfd8 with equality, Petrosian-Kotov, 1972.

1 1 . . • Nb4 Also played is l l . . .Nd5. For example, 1 2 Qe4 Ncb4 1 3 Ne5 Qd6!

(13 . . . Ra7 14 Qg4! Kh8 15 Qh3 b6 16 Ne4 Qe8 1 7 Be2 slightly favors White, Vegh-Kallai, Budapest 1984) 14 Qg4 (weaker is 14 Bd3 .f5 15 Qe2 Bf6 I 6 Bb I b6 I 7 Ra3 Bb 7 with an edge for Black, Marin-Garcia Palermo, Zaragoza 199 1 , or 14 Bd2 b6 I 5 Nxd5 Nxd5 I 6 Bd3 .f5 I 7 Qe 2 Bf6 I 8 ReI ReB 19 Rae/ Bb 7 with equality, Vera-Garcia Palermo, Bayamo 1983) 14 . . . f5 15 Qe2 b6 16 Bxd5! Nxd5 (16 . . . exd5 17 Bf4!) 17 Nc4 Nxc3 18 bxc3 Qc7 19 Ba3 with an edge for White, Marin-Garcia Palermo, Andorra 199 1 .

Another move for White (after I I . . . Nd5) is 1 2 Bd3. For example, 12 . . . Ncb4 13 Bbl (13 Be4 b6 14 Ne5 Bb 7 15 Bd2 .f5! 16 Bbl Bf6 is equal, Tuomala-Kivisto, 1984) 13 . . . b6 14 Qe4 (or 14 Ne5 Bb7 15 Ra3 [In Malaniuk-Yakovich, Moscow 1992, White tried 15 Ne4. There followed 1 5 . . . f5 16 Nc3 Rc8 17 Ra3 Nc2! ! 1 8 Ra2 Nxd4! 1 9 Qe1 Nb3 20 Nf3 Nxc l 2 1 Rxc l Nf4 22 Qe3 Qc7 23 Ra 1 Bc5 and White resigned] 15 . . . Rc8 16 a5 b5 1 7 Ne4 j5 18 Nc5 Bxc5 19 dxc5 Rxc5 20 Rg3 Rc7 21 Bg5 is unclear, Gligoric-Portisch, Pula 197 1 ) 14 . . . g6 15 Bh6 Re8 16 Ne5 Bb7 17 Qf3 f5 1 8 Qg3 Bh4 19 Qh3 Rc8 with equal chances, Nikolic-Petrosian, Vrsac 1983.

In Tataev-Y akovich, Moscow 1992, White tried 12 Ba2. There followed 12 . . . Ncb4 13 Bbl b6 14 Nxd5 Nxd5 15 Qe4 g6 16 Bh6 Re8 17 Ne5 Bts 18 Bxf8 Rxf8 19 Qf3 Bb7 20 a5 bxa5 2 1 Be4 Rb8 22 Qg3 Ne7 23 Bd3 Bd5 24 Rd2 Rb4 and Black eventually won.

12 Ne5 After 12 Bg5 Nfd5! 1 3 Nxd5 Nxd5 14 Bxd5 Bxg5 15 Be4 Bf6 Black

has equalized, Tukmakov-Balashov, USSR 1985. Interesting is 12 Ng5 Bd7, as in Bareev-Ivanchuk, Linares 1994,

which continued 13 Ne5. Now, instead of the actually played 1 3 . . . Rc8 14 Rei Be8 15 Radl Nfd5? allowing White to win a pawn with 16 Nxd5 ! Nxd5 1 7 Bxd5 Bxg5 18 Bxb7, Black should have played 1 3 . . . Bc6, since after 14 Nxc6 bxc6! 15 a5 Ra7 play is dynamically equal.

12 . . . Bd7

CHAPTER I 57

Other possibilities: I) 1 2 . . . b6 13 QfJ (13 Ra3!? or 13 g4!? Bb7 14 g5 Nfd5 15 Ne4 is

unclear, Dragic-Paunovic, 1985) 13 . . . Nfd5 (interesting is Pachman's suggestion 13 . . . Ra7!? and on 14 Qg3 Bb7!; not so good is 14 . . . Rc7?! 15 Bh6! Ne8 16 d5 Bd6! 17 dxe6 Bxe6 18 Bxe6 fte6 19 Ne4 Bxe5 20 Qxe5 Qe7 with a small advantage) 14 Nxd5 (on 14 Qg3 Kh8! 15 Qh3 Qe8 16 Be2 White is only a little better) 14 . . . exd5 1 5 Bb3 Be6 16 Bd2 f6 17 Ng4 Nc6 18 Bc3 Qd7 1 9 h3 ! Rad8 20 Qe2 a5 2 1 Re 1 Bn 22 QfJ with a small plus for White, Pinter-Korchnoi, Beersheva 1988.

2) 12 . . . Nfd5 13 Qg4 (13 Ne4 b6 14 a5 b5 15 Bb3 Bb 7 /6 Bd2 ReS 1 7 Racl Nc6! equalizes, Flear-Kupreichik, Hastings 1984) 13 . . .Kh8 14 Qf3 Kg8 1 5 a5, as in Kouatly-Marjanovic, Marsei 1986.

13 Bf4 Be8 14 Bg5 Rc8 15 Bb3 Bc6 16 Nxc6 Rxc6 17 Bxf6 Bxf6 18 d5 exd5 19 Nxd5 Nxd5 20 Bxd5 Rd6 21 Bxb7 aS

With unclear play and mutual chances, Ivkov-Gheorghiu, Hamburg 1965.

I B422 ( I d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Qe2)

8 .. . Qc7 (See next diagram)

58 THE QUEEN'S GAMBIT ACCEPTED

Maintaining the central tension while controlling the important b8-h2 diagonal.

9 Nc3 Less active is 9 Bd2. For example, the game Speelman-Hubner,

Barcelona 1989, continued 9 . . . Bd6 10 dxc5 Bxc5 I I Bc3 0-0 12 Nbd2 e5 1 3 Rac l Qe7 14 Ba2 Bf5 1 5 e4 Bg4 1 6 h3 Bd7 17 Nc4 Be6 18 Ncxe5 Bxa2 19 Nxc6 Qxe4 20 Qxe4 Nxe4 2 1 Bb4 Rfc8 draw.

Now play divides: I 8422 1 9 . . . Be7, and I 84222 9 . . . Bd6.

I 842221 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Qe2 Qc7 9 Nc3)

9 . . . Be7

CHAPTER I 59

This may be Black's best choice here, though the alternative 9 . . . Bd6 is very playable, too. If White now plays 10 dxc5 then 10 . . . Bxc5 transposes into I B42222.

10 Rd1 0-0 1 1 h3

Insipid is I I d5 exd5 12 Nxd5 Nxd5 l 3 Bxd5 Bf6 ! . Now 14 h3 gives equality; in the game Plachetka-Mikhalchishin, Tmava 1988.

After I I b3 play may proceed l l . . .b6 12 Bb2 Bb7 13 Racl Rfd8 14 h3 Na5 15 Ne5 Nc6. At this point White may repeat the position with 16 Nf3 Na5; in the game Pinter-Ehlvest, Zagreb INterzonal 1987, White tried for more with 16 Nb l . There followed l6 . . . Nxe5 17 dxe5 Qc6 18 f3 Nd5 19 Nd2 Bg5 20 Nfl Bh4 2 1 Rd2 Nc7 22 Bd3 Nd5 with about equal chances.

1 1 ... Bd7 Also possible is l l . . .Rd8. Then Botvinnik-Keres, USSR 194 1 ,

continued 1 2 b3 Bd7 l 3 Bb2 Be8 1 4 d5 exd5 15 Bxd5 Nd4 with equality. In the game Bonin-Zaltsman, New York 1989, White varied with 14 dxc5, but after l4 . . . Bxc5 15 Ng5 Ne5 16 Nce4 Nxe4 17 Nxe4 Be7 18 Rdc l Qb8 19 f4 Ng6 with a quite decent game for Black.

Finally, in Fahnenschmidt-Ribli, Germany 199 1 , White played 12 dxc5, and after l 2 . . . Rxd lch 13 Qxd l Bxc5 14 e4 Bd7 1 5 Bg5 Be8 16 Qe2 Nd4 17 Nxd4 Bxd4 18 Bd2 Rd8 19 Bel Nh5! with the upper hand for Black.

1 2 dxc5 Bxc5

13 e4 Ne5! In Drasko-Vaganian, Sarajevo 1987, Black played l 3 . . . Nh5, but after

14 Be3 NbS 1 5 Ba2 Nf4 16 Qd2 Bxe3 17 Qxe3 Ng6 1 8 Racl White was on top.

14 Bf4 Nxf3ch 15 Qxf3 Qa5!

Not l 5 . . . Bxf2ch 16 Qxf2 Qxc4 17 Be5 with clear advantage for White, Lukacs-J. Adamski, Naleczow 1985.

199 1 .

16 Qe2 On 16 e5 Black has 16 . . . Bc6.

16 ... 17 Bd2 18 e5 19 f4

Bc6 Bb4 Nd7 Rfd8

With a very good game for Black, Hegler-A. Bykovsky, Gausdal

60 THE QUEEN'S GAMBIT ACCEPTED

I B4222 ( 1 d4 dS 2 c4 dxc4 3 NO Nf6 4 e3 e6 S Bxc4 cS 6 0-0 a6

7 a4 Nc6 8 Qe2 Qc7 9 Nc3)

9 ...

xa a t il � . '

t -.a

:Jt -

!ffj -� � � �

� ' a , �

Bd6

*- B - � t � -7-' -

-- -

� �{)-. � � "«1� -� :! � ·� z z

-1!� � . �

This can transpose into 9 . . . 8e7 (I 8422 1) after 10 dxc5 . After 9 . . . 8d6 we examine: I 84222 1 10 b3, I 842222 10 dxc5, I

842223 10 Rdl , I 842224 10 8d3 and I 842225 10 8d2.

I B42221 ( 1 d4 dS 2 c4 dxc4 3 NO Nf6 4 e3 e6 S Bxc4 cS 6 0-0 a6 7 a4 Nc6 8

Qe2 Qc7 9 Nc3 Bd6)

10 b3 (See next diagram)

CHAPTER I

Fianchettoing the QB; but this is not dangerous for Black.

10 .. . 0-0 1 1 Bb2 cxd4 1 2 exd4 Na5 13 d5

61

After 13 Bd3 !? Nxb3 14 Rad1 Bd7 1 5 Ne4 Hubner gives 1 5 . . . Nd5 ! , but not 1 5 . . . Be7? 16 d5 ! Nxd5 17 Neg5 or 17 Bxg7!?.

13 ... Nxc4

14 bxc4 Bd7! 15 Ne4 Nxe4 16 Qxe4 Rae8

Black has a fine position, Van der Sterren-Hubner, Wijk aan Zee 1988.

I B42222 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6 7 a4 Nc6 8

Qe2 Qc7 9 Nc3 Bd6)

10 dxc5 (See next diagram)

62 THE QUEEN'S GAMBIT ACCEPTED

A good plan. White intends to advance in the center with e4; so first he exchanges the d-pawn to make that possible.

10 . . . Bxc5

1 1 e4 Ng4

12 gJ 0-0 13 Bf4 e5 14 Bg5

Or 14 Nd5 Qd6 15 Bd2 Nf6! 16 b4 Nxb4 17 Bxb4 Bxb4 18 Nxb4 Qxb4 1 9 Nxe5 as in Dizdar-Nikolic, Sarajevo 1987. Now Black can play . . . Qe7 and . . . Be6 with even chances.

14 .. . h6 15 Nd5 Qd6 16 Bd2 Nf6 1 7 b4 Bd4! 18 Rab1 Bg4 19 QdJ Rfd8!

Better than 19 . . . Nxd5? 20 Bxd5 Rfd8 2 1 b5 axb5 22 axb5 Ne7 23 Bb4 Qf6 24 Bxe7 Qxe7 25 Nxd4 exd4 26 Qxd4 Be6 as in Nikolic-Hubner, Wijk aan Zee 1988. Now 27 Qe5 is very strong for White.

20 b5! axb5 21 axb5 Ne7!

An improvement on 2 l . . .Bc5 22 Bc3 ! (not 22 bxc6 bxc6 with an edge for Black) 22 . . . Bxf3 23 Qxf3 Nd4 24 Bxd4 Bxd4 25 Nxf6ch Qxf6 26 Qxf6 gxf6, as in Li Zunian-Ye Rongguang, China 1990. Now 27 Kg2! allows White a small plus.

22 Bb4 Bc5

wins.

CHAPTER I

23 Bxc5 24 Nxe5 25 exd5 26 b6

Qxc5 Nfxd5! Be6

63

A better try was 26 Rfd 1 . After 26 dxe6 Rxd3 27 exf7ch Kf8 Black

26 ... Nxd5

27 QfJ RaJ 28 Qd1 Qe7

As in Samo-Lin Ta, Novi Sad Olympiad 1990. Black has a very satisfactory game.

I 842223 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Qe2 Qc7 9 NcJ Bd6)

Bolstering d4 and d5.

10 Rd1

10 . . . 0-0 1 1 h3 b6

In Lputjan-Hubner. Rotterdam 1988, Black played I I . . . Re8. After 1 2 dxc5 Bxc5 13 e4 Nd7 14 Ba2 b6 15 e5 ! Ndxe5 16 Bf4 f6 17 Racl Bb7 18 Nxe5 fxe5 19 Bg3 White held a slight advantage. However, a promising new method is l l . . .Bd7 !?. This essentially transposes to I B422 l 9 . . . Be7.

12 d5 Ne5!?

64 THE QUEEN'S GAMBIT ACCEPTED

13 NxeS On 13 dxe6 Bxe6 is equal.

13 ... BxeS 14 dxe6 Bxe6

IS Bxe6 fxe6 16 Qc4

White is slightly better, Karpov-Timman, Linares 1989.

I B42224 ( I d4 dS 2 c4 dxc4 3 NO Nf6 4 e3 e6 S Bxc4 cS 6 0-0 a6

7 a4 Nc6 8 Qe2 Qc7 9 Nc3 Bd6)

10 Bd3

Contesting e4 while probing the bl-h 7 diagonal.

Not l 3 . . . Nb4 14 Bb2 ! .

10 ... 0-0 II dxcS BxcS 12 Ne4 Be7 13 b3 NdS

14 Bb2 Bd7 IS Rfd Qb6 16 Nd4 NeS

Even better is 16 . . . Rfc8! .

17 aS Qa7 18 Bel

CHAPTER I

18 Nb5 Qb8 19 Nc7 Nxd3 20 Qxd3 Nxc7 2 1 Qxd7 Nd5 is equal.

18 . . . Ng6 19 Qh5 Rac8

Play is even, Lin Weiguo-Lin Ta, China 1989.

I 842225 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Qe2 Qc7 9 Nc3 Bd6)

10 Bd2

65

This looks harmless, but Black must play accurately to hold his own.

10 . . . 0-0 1 1 d5

In Cvitan-Mikhalchishin, Zenica 1989, White played 1 1 Rac l . After 1 l . . .b6 1 2 Bd3 Bb7 1 3 h3 Rfd8 14 dxc5 bxc5 15 Ne4 Nxe4 16 Bxe4 Qe7 17 Bc3 Nb4 the game was drawn.

1 1 . . . Ne5!? After l l . . . exd5 12 Nxd5 Nxd5 13 Bxd5 Ne5 14 Nxe5 Bxe5 1 5 f4 !

Bxb2 16 Rab1 Bf6 17 a5 White has more than enough for the pawn, Miles­Ye Rongguang, Beijing 199 1 .

Also of interest is l l . . .exd5 12 Nxd5 Nxd5 1 3 Bxd5 and now, instead of 1 3 . . . Ne5, Black can try 13 . . . Bg4, and after 14 h3 Bh5 1 5 Bc3 Rae8 16 Rfd1 Re7 we reach a position from the game Belov-Mesropov, Podolsk 199 1 ; now Belov gives 17 Rae 1 with a small plus to White.

12 Nxe5 Bxe5

13 f4!? Bxc3

66 THE QUEEN'S GAMBIT ACCEPTED

14 Bxc3 NxdS 15 DeS

White has compensation but the position is unclear.

I 8423 ( 1 d4 dS 2 c4 dxc4 3 NO Nf6 4 e3 e6 S Bxc4 cS 6 0-0 a6

7 a4 Nc6 8 Qe2)

8 . . . Be7

The most natural and solid developing move. Now White has: I 8423 1 9 dxc5 and I 84232 9 Rdl .

I 84231 ( 1 d4 dS 2 c4 dxc4 3 NO Nf6 4 e3 e6 S Bxc4 cS 6 0-0 a6

7 a4 Nc6 8 Qe2 Be7)

9 dxcS (See next diagram)

CHAPTER I

In order to follow up with e4.

9 . .. Bxc5

67

On 9 . . . Ne4 10 Nd4 ! Nxc5 I I Nxc6 bxc6 1 2 Nc3 gives White a small positional plus.

10 e4 Ng4! On IO . . . Qc7, I I e5 is strong. For example, Donner-Kinnmark, Halle

1 963, continued I I e5 Nd7 12 Bf4 b6 1 3 Nbd2 Bb7 14 Rac l Qb8 15 Rfdl with advantage for White. Also, bad is IO . . . e5 I I Bxf7ch! .

11 e5! A dangerous pawn sac, but Black can hold his own with precise play.

In the 8th Botvinnik-Petrosian Matchgame ( 1963), I I Bf4 was played; there followed l l . . .Qf6 12 Bg3 Nge5 1 3 Nxe5 Nxe5 14 Nbd2 (14 Qh5 Bd4) 14 . . . 0-0 15 Radl b6 16 Qh5. Now l 6 . . . Bd4! 1 7 b3 Bb7 18 Khl Qg6 is equal.

11 .. . Nd4

12 Nxd4 Qxd4

13 Na3! Botvinnik's innovation.

13 .. . Bxa3

14 Rxa3 Nxe5 15 b3

Keres recommended 15 Rdl , e.g, l 5 . . . Qg4 16 Qd2 ! 0-0 17 Be2 with unclear play; while if l 5 . . . Qxc4 16 Qxe5 0-0 17 Rg3 f6 18 Qd6 e5 19 Bh6 Bg4 20 h3 ! is strong. Also, Matanovic gives 1 5 Rc3 !?.

15 ... Qc5 16 Ral

68 THE QUEEN'S GAMBIT ACCEPTED

After Tal's 16 Bb2, Boleslavsky gives 16 . . . Nxc4 17 bxc4 Bd7! 18 Rg3 0-0-0 with equality.

With equality.

16 .. . Nxc4

1 7 bxc4 Bd7

18 Ba3 Qf5!

I 84232 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6

7 a4 Nc6 8 Qe2 Be7)

Pressuring the d-file.

9 Rdl

9 . . . 0-0 A clever move-order finesse. After 9 . . . Qc7 10 Nc3 0-0 play

transposes into I 84222 1 .

1 0 dxc5 Qc7 1 1 b3

In I. Ivanov-Seirawan, US Championship 199 1 , play continued 1 1 Nbd2 Nd7 1 2 Nb3 Nxc5 13 Nxc5 Bxc5 14 Bd2 b6 15 Bc3 Qe7 16 Ne5 Nxe5 1 7 Bxe5 with a comfortable position for Black.

1 1 .. . e5! Note that this central reinforcement is made possible because 8 . . . Be7

blocked the e-file.

12 Nc3 e4

CHAPTER I 69

13 Nd2 Bg4 14 tJ extJ 15 gxtJ Bh5 16 Bb2 Rad8 1 7 Nce4 Ne5 18 Rfi Nxe4 19 Nxe4 Bg6 20 Racl Nxc4 21 Rxc4 fS 22 Nf2 Bf7 23 Reel Rfe8 24 Nd3 Bxb3

As in P. Nikolic-Seirawan, Skelleftea 1989. A complex position with mutual chances (the game ended in a draw).

I 85 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6)

7 e4!?

This pawn sacrifice was introduced by Geller in 1958. If Black takes the pawn White gets the open e-file and a lead in development.

7 ... b5 Possible is the bold acceptance by 7 . . . Nxe4, e.g, 8 d5 Be7! (but not

8 . . . exd5? 9 Bxd5 Nd6 10 Re 1ch Bel 1 1 Bg5 f6 12 Bf4 with White on top) 9

70 THE QUEEN'S GAMBIT ACCEPTED

dxe6 Bxe6 10 Qxd8ch Rxd8 I I Bxe6 fxe6 12 Rei Nf6 1 3 Ng5 0-0 14 Nxe6 Re8 15 Nc3 Nbd7 with about even chances, Neikirch-Ciarke, Leipzig 1960.

In Petrosian-Sherwin, Portoroz 1958, Black played 7 . . . cxd4 but after 8 e5 Nfd7 9 Qxd4 Nc6 10 Qe4 Qc7 I I Bf5 b5 12 Bb3 Nc5 1 3 Qe2 Nxb3 14 axb3 Bb7 1 5 Nc3 Be7 16 Ne4 0-0 17 Racl White held the edge.

8 Bd3 Not 8 Bb3 Nxe4 9 d5 c4 with advantage to Black.

8 . . . Bb7 9 eS

Gulko gives 9 Bg5 !? cxd4 10 a4 ! with unclear play.

9 ... NdS 10 a4 b4 1 1 Nbd2 n:d4

1 2 NbJ Nd7

13 Rel NcS 14 NxcS BxcS IS NgS Ne3!

16 Nxe6! fxe6 1 7 fxeJ QgS 18 e4 QxeS

As in Geller-Keres, Bled 196 1 . After 19 Rdl play is unclear.

I 86 ( 1 d4 dS 2 c4 dxc4 3 NO Nf6 4 e3 e6 S Bxc4 cS 6 0-0 a6)

7 aJ (See next diagram)

CHAPTER I

This allows the Bc4 to pull back to a2 after . . . b5.

7 ... b5 8 Ba2

Also possible is 8 Be2 cxd4 9 Nxd4.

8 ... Bb7 9 Nc3

71

Or 9 Qe2 Nbd7 10 Rdi cxd4 I I exd4 Be7 12 Nc3 Nb6 I3 Bg5 0-0 14 Rac l Nbd5 with equality, Miles-Ribli, Riga 1979.

9 ... Nbd7

10 Qe2 Be7

1 1 Rd1 Qb8 12 e4! ud4

I 2 . . . b4 I 3 axb4 cxb4 14 e5! bxc3 15 exf6 slightly favors White.

13 Nxd4 Bd6 Not 1 3 . . . 0-0 because of I4 Nxe6!

14 g3 0-0 15 fJ! Rd8 16 Be3

White retains a slight pull, Kozui-Dorfman, I99 I .

I 87 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6)

7 b3

72 THE QUEEN'S GAMBIT ACCEPTED

Intending to fianchetto the QB.

7 ... cxd4! The most accurate way to equalize.

8 N:x:d4 Bd6 9 Bb2 0-0

10 Nd2 Bc7

l l Rcl Qd6 12 f4 b5

13 Bd3 Bb7

14 Qc2 Bb6 15 Ne4 B:x:e4

16 B:x:e4 Ra7

1 7 Bf3 Rc7

18 Qe2 Rfc8 Play is completely equal, Miles-Ivanchu.k, Moscow 1990.

I 88 ( 1 d4 d5 2 c4 d:x:c4 3 NO Nf6 4 e3 e6 5 B:x:c4 c5 6 0-0 a6)

7 d:x:c5 (See next diagram)

CHAPTER I 73

White steers into an early ending, but Black already has nothing to fear with accurate play.

7 ... Qxd1 Also good is 7 . . . Bxc5 8 Qxd8ch Kxd8 9 b3 (or 9 Nbd2 Nbd7 10 Be2

b6 1 1 Ng5 Ke8 12 Nge4 Bel 13 Nxf6ch gxf6l? 14 b3 Bb7 15 Bb2 Ne5 with approximately equal play, Wojtkiewicz-Ibragimov, Bern 1993) 9 . . . Nbd7 10 Bb2 Ke7 1 1 Be2 b6 12 Nbd2 Bb7 13 Ne l b5 14 a4 Bb4 15 Ndf3 bxa4 16 Rxa4 Bd6 17 Nd3 Rhc8 18 Rfa 1 with even play - the game was drawn shortly, Nikolic-Ivanchuk, Manila Olympiade 1992. Less good is 9 . . . b5. For example, 9 . . . b5 10 Be2 Bb7 1 1 Bb2 Ke7 1 2 a4 b4 13 a5 ! Nbd7 14 Nbd2 Rhd8 15 Rfc l and White retains a slight advantage, Spassky-Hubner, 19S9.

8 Rxd1 Bxc5 9 b3

Or 9 a3 Ke7 10 b4 Bd6 1 1 Bb2 b5 12 Be2 Bb7 1 3 Nbd2 Nbd7 14 Nb3 Rac8 1 5 Na5 BaS 16 Nd4 Nb6 17 Rac l Rxc l I S Rxc l ReS 19 RxcS NxcS 20 f3 Na7 2 1 e4 Nd7 22 Ndb3 f6 with equal chances, Rashkovsky­Kramnik, Moscow 1992.

9 .•. b5 10 Be2 Bb7 1 1 Bb2 Nbd7 12 a4

In Spassky-Fischer, 14th match game Belgrade 1992, White continued 12 Nbd2 and after 12 . . . 0-0 13 Rae I RfcS 14 h3 � 15 Kfl Ke7 16 Ne1 Bd6 17 a4 Bc6 1S axb5 axb5 19 Rc2 Rc7 20 Rdcl RacS play was quite even - the game was drawn in 32 moves.

12 ... bu4 13 Rxa4 0-0

74 THE QUEEN'S GAMBIT ACCEPTED

Chances are even, Spassky-Nikolic, Barcelona 1989.

I 89 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 0-0 a6)

7 Bd3

Returning to the b1-h7 diagonal and controlling e4, but this shouldn't cause Black any problems.

7 ... Nbd7 This seems to be best. In Polugaevsky-Ivanchuk, Monaco 1992,

Black played 7 . . . Nc6 but after 8 Nc3 cxd4 (in the game Kasparov-Kamsky, Intel Grand Prix 1 994, Black played 8 . . . Qc7. After 9 a3 b6 10 dxc5 bxc5 1 1 Ne4 Bel 1 2 Bd2 Bb7 1 3 Bc3 Nxe4 1 4 Bxe4 0-0 1 5 Rcl h6 1 6 Qa4 j5 1 7 Bb 1 e5 18 Nh4! White had a powerful position) 9 exd4 Be7 1 0 Bg5 0-0 I I Rei h6 1 2 Bh4 b5 13 Bxf6 Bxf6 14 Ne4 Bd7 15 Bbl Be7 16 Qd3 f5 I 7 Nc5 Bxc5 I 8 Rxc5 White held the advantage.

8 a4 In M. Gurevich-Magem, Barcelona 1992, White played 8 Qe2. There

followed 8 . . . b6 9 Rd l Bb7 10 Nc3 Qc7 I I Bd2 Bd6 1 2 Racl 0-0 1 3 e4 cxd4 14 Nb5 Qb8 1 5 Nbxd4 Nc5 16 e5 Nxd3 17 Qxd3 Bxe5 18 Nc6 Bxc6 19 Rxc6 Rd8 20 Qe2 Bd6 2 1 Rdcl Qb7 22 Be3 Nd5 23 Bg5 Nb4! and Black consolidated and eventually won.

8 . .. b6 9 Qe2 Bb7 10 Rd1 Qc7

CHAPTER I 75

1 1 hJ Be7 12 NcJ 0-0 13 Bd2 Bd6 14 Racl e5 15 Bb1 Rfe8

Play is equal, Eingorn-Ehlvest, USSR 1988.

76 CHAPTER II

Furman's Variation

( I d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5)

6 Qe2

This variation comes from the great Soviet analyst Semyon Furman. White quickly moves the Queen off the d-file with the option of playing dxc5 and e4. Another idea is the gambit of the d-pawn. For example, 6 . . . cxd4 7 exd4 Nc6 8 0-0 Nxd4? 9 Nxd4 Qxd4 10 Rd l Qb6 I I Bb5ch! Bd7 1 2 Bxd7ch Nxd7 l 3 Nc3 Nf6 14 Be3 Qc6 15 Rac l h6 16 a4 ! with the threat of 17 Nb5 ! .

After 6 Qe2 we analyze: I I A 6 . . . cxd4 and I I B 6 . . . a6.

II A ( l d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 e6 5 Bxc4 c5 6 Qe2)

6 . . . cxd4 (See next diagram)

CHAPTER II

Isolating White's d-pawn.

7 exd4 8 0-0 9 NcJ

Be7 0-0 Nbd7

77

The QN will swing over to b6 from where it controls the important blockading square d5.

10 Bg5 Nb6

1 1 BbJ Bd7

12 Ne5 Bc6 Black willingly gives up the Bishop pair and allows an isolated c­

pawn in the process. In return Black obtains positional compensation: the formidable Ne5 has been lured off the board and the possibility of opening lines (while getting rid of the isolani) by d4-d5 is ruled out.

13 Nxc6 bxc6 14 Rfd1 Nfd5

15 Bxe7 Nxe7

16 Racl ReS 1 7 Bc2 Rc7

Black has a sarisfactory position, lzeta-Magem, Dos Hermanas 1993.

78 THE QUEEN'S GAMBIT ACCEPTED

li B ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 e6 5 Bxc4 c5 6 Qe2)

6 . . .

This seems to be Black's best line.

7 dxc5

a6

A typical exchange in the Funnan Variation. White prepares e4, or, in some cases, a3 and b4 (then the QB's diagonal will not be obstructed by the d-pawn).

7 .. . Bxc5 8 0-0

At this point e4 is not very effective: 8 e4 Qc7 9 e5 (or 9 0-0 Ng4 1 0 Nbd2 Nc6 1 1 Nb3 Bd6 1 2 h3 Nge5 with even play, Taimanov-Filip, Interzonal 1970) 9 . . . Ng4 10 0-0 Nxf2 1 1 Rxf2 Bxf2ch 12 Kxf2 b5 1 3 b3 0-0 ! (not 13 . . . bxc4 because of 14 Ba3! with attacking chances) 14 Ba3 bxc4 15 Bxffl Kxf8 with equal chances according to Neistadt.

8 . . . Qc7 Also good is 8 . . . Nc6. For example, the game Fedorowicz-Chandler,

London 1987, proceeded 9 e4 (in Lautier-Dorfman, Barcelona 1992, White varied with 9 Nbd2. After 9 . . . 0-0 10 a3 Nd5 1 1 Ne4 Be 7 12 b4 b5 13 Bd3 j5 14 Ng3 B/6 15 Bb2 Bxb2 16 Qxb2 Nxe3! 1 7 Bxb5 axb5 18fxe3 Q/6 19 Qb3 Kh8 20 Rad1 Ne5 Black had an excellent game) 9 . . . Ng4 (9 . . . b5 10 Bb3 Bb 7 1 1 Nc3 Qc7 12 e5 Nd7 13 Bf4 slightly favors White, Huzman-Lin Ta, Belgrade 1988) 10 Bf4 e5 1 1 Bd2 0-0 12 h3 Nf6 1 3 Nc3 Be6 14 Bg5 h6 1 5 Rad1 Nd4 16 Nxd4 Bxd4 with even play.

9 Nbd2 0-0

CHAPTER II

10 a3 b5 1 1 BdJ Bd6

12 b4 Bb7

13 Bb2 Nbd7 14 Racl Qb8 15 hJ Ne5 16 Bb1 NxOch 1 7 NxfJ BxO 18 Qxl3 Be5

19 Qe2 Bxb2 20 Qxb2 ReS

With equality, Christiansen-Diugy, U.S. Championship 1985.

79

80 CHAPTER Ill

Modern Variation

( I d4 d5 2 c4 dxc4 3 NO Nf6 4 e3)

4 ... Bg4

A dynamic idea, getting out the QB early on, before . . . e6 blocks it in. However, b 7 is now unprotected, which White can try to exploit. If White goes after the b-pawn (with an early Qb3) Black is planning on sacrificing it to obtain active counterplay.

Now we examine: III A 5 h3 and III 8 5 Bxc4.

III A ( I d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 e3 Bg4)

5 h3 (See next diagram)

CHAPTER III 81

White pursues the Bg4 with the intention of gaining the two Bishops.

5 . . . Bh5 6 g4 Bg6 7 Ne5

Also reasonable is 7 Nc3 e6 8 Ne5 Nbd7 9 Nxg6 hxg6 10 Bxc4 Nb6 1 1 Bfl ! with a small plus for White, van Seters-Smyslov, Hastings 1962.

7 .. . Nbd7 Or 7 . . . Be4 8 f3 Bxb l 9 Rxb1 Nfd7! 10 Nxc4 (unclear is 10 Bxc4.1?

Nxe5 I I dxe5 Qxdlch 12 Kxdl) IO . . . Nb6 1 1 Qc2 e6 12 Bg2 N8d7 1 3 b4 Bd6 with an edge for White, Vladimirov-Korelov, Leningrad 1964.

8 Nxg6 hxg6 9 Bg2 c6 10 Nd2 Qa5 1 1 0-0 Nb6

12 bJ cxbJ

13 NxbJ Qa4 14 Qe2 e6 15 Rb1 Nbd5 16 Qb2 b6 17 e4

White is a little better, Bany-Marjanovic, Istanbul 1988.

82 THE QUEEN'S GAMBIT ACCEPTED

III B ( 1 d4 dS 2 c4 dxc4 3 NfJ Nf6 4 e3 Bg4)

S Bxc4

The natural recapturing/developing move.

S ... e6 Now we have: III 8 1 6 Qb3, III 82 6 Nc3, III 83 6 a3, III 84 6 0-0,

III 85 6 Nbd2 and III 86 6 h3.

III 81 ( 1 d4 dS 2 c4 dxc4 3 NfJ Nf6 4 e3 Bg4 S Bxc4 e6)

6 Qb3 (See next diagram)

CHAPTER III 83

Trying to exploit the QB's absence from the Queenside. Black now must be willing to sacrifice his b-pawn for active counterplay.

6 ... Bxf3 Weakening White's pawns and creating possibilities of attack against

White's King if he castles Kingside.

7 gxf3 Nbd7 Defending by 7 . . . b6 is too weakening: 7 . . . b6 8 Nc3 Be7 9 d5 ! exd5

IO Nxd5 0-0 I I Nxe7ch Qxe7 12 Bd2 a6 13 Rgi b5 I4 Bd5 Ra7 I 5 Bc3 with a big advantage for White, A. Zaitsev-Spassky, I96 1 .

Nc3 .

8 Qxb7 c5 Now play divides into: III B I I 9 dxc5, III B I2 9 Rgi and III B I 3 9

I I I B t l ( 1 d 4 d5 2 c4 dxc4 3 N f3 Nf6 4 e3 Bg4 5 Bxc4 e6 6 Qb3 Bxf3

7 gxf3 Nbd7 8 Qxb7 c5)

9 dxc5 (See next diagram)

84 THE QUEEN'S GAMBIT ACCEPTED

Concerned over 9 . . . cxd4 after which White's pawns are terribly weakened.

9 .. . Bxc5

10 f4!? In Uusi-Luik, 1965, White tried 10 Qb3 . There followed 10 . . . 0-0 1 1

Nc3 Qe7!? 1 2 Be2 Rab8 1 3 Qc2 Rfc8 with murky play and mutual chances. On 1 0 Nc3 an analysis by Euwe runs I0 . . . 0-0 I I Qb3 Qc7! I 2 Be2

Rab8 l 3 Qc2 Bd6! I4 Bd2 Nd5 with an edge to Black.

10 ... 0-0

1 1 Nc3 Or I I Qg2 Rc8 I2 b3 Nb6 with even chances, Spassky-Gurgenidze,

USSR I963.

11 ... Nb6

12 Be2 Nfd5

13 0-0?! A mistake. Better was I3 BD ! with an unclear position.

13 .. . Qh4 14 BfJ Rab8 15 Qa6

On I 5 Qxa7? there follows I5 . . . f5 I6 Ne2 Ra8 I7 Qb7 Rf6! with the upper hand for Black.

15 ... NxcJ

16 bxcJ f5 1 7 Bg2 Rf6

Black is clearly superior, Quinteros-Miles, Amsterdam I977.

CHAPTER III

Ill 812 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 8g4 5 8xc4 e6 6 Qb3 8x0

7 gxO Nbd7 8 Qxb7 c5)

9 Rg1

This "logical" move only leads to trouble for White.

9 . . . g6

10 Nc3 8e7!

1 1 dxc5 Nxc5 12 Qc6ch Kf8!

13 f4 Nfe4! 14 8a6

85

14 Nxe4? ReS! 1 5 Qb5 a6 wins. We have been following Hodos-Tal, 1962; now Taimanov gives 14 . . . Nxc3 15 bxc3 Rb8 with a clear superiority for Black.

I l l 813 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 8g4 5 8xc4 e6 6 Qb3 8x0

7 gxO Nbd7 8 Qxb7 c5)

9 Nc3 (See next diagram)

86 THE QUEEN'S GAMBIT ACCEPTED

Ignoring Black's intentions of further disruption of White's pawn structure, White intends to play actively.

9 .. . cxd4 10 exd4 Bd6

Not 10 . . . Be7? 1 1 Bf4 0-0 12 Bc7 Qc8 l 3 Ba6! Qe8 14 0-0 as in Gligoric-Kozomara, Sarajevo 1963.

Playable is 10 ... Rb8, e.g. Maslov-Spassky, USSR 1963, continued 1 1 Qxa7 Bd6 1 2 Nb5 Bb4ch l 3 Kf1 Rc8 14 Be2 Nd5 1 5 Qa4 Qh4 with complicated play.

1 1 Ne4 l l Nb5 Bb4ch 12 Kf1 0-0 l 3 Rg l , Bisguier-van Scheltinga,

Beverwijk 1962, and now l 3 . . . Rc8 ! is good for Black. In Arzukevich-Korelov, Leningrad 1963, 1 1 Rg1 was played. There

followed 1 l . . .Rb8 ! 12 Qxa7 0-0 l 3 Bh6 Nh5 14 Ne4 Nb6 15 Qa5 f5 with excellent chances for Black.

After 1 1 Be3 0-0 12 0-0 Nb6 play is about even.

1 1 ... Nxe4

12 fxe4 Or 1 2 Qxe4 Rc8 l 3 Bb3 Nf6 14 Qd3 Qa5ch 15 Ke2 0-0 with an edge

for Black, Chernikov-Blagidze, Baku 1964.

12 .. . 0-0

13 e5 Nb6

14 BdJ Bb4ch 15 Ke2

Blagidze-Klavin, Tiflis 1962 . The position is approximately equal.

CHAPTER III

III 82 ( I d4 d5 2 c4 dxc4 J NfJ Nf6 4 eJ Bg4 5 Bxc4 e6)

6 NcJ

This simple developing move shouldn't trouble Black much.

6 ... Nbd7

87

Probably best. In Larsen-Speelman, Hastings 1990/9 1 Black tried 6 . . . a6. There followed 7 h3 Bh5 8 g4 Bg6 9 Ne5 Nbd7 10 Nxg6 hxg6 1 1 Qf3 Rb8!? 1 2 Bd2 c5 13 0-0-0 b5 14 Be2 cxd4 1 5 exd4 Nb6 16 Kb1 Nfd5 17 Bd3 with an edge for White.

7 Be2 Or 7 0-0 as in Szabo-Ghitescu, Beverwijk 196 7, which continued

7 . . . Bd6 8 e4 e5 9 dxe5 Nxe5 10 Be2 Bxf3 1 1 Bxf3 Qe7 1 2 Be2 0-0-0 ! 1 3 Qa4 Bc5 1 4 Bg5 c6 with even chances.

In Eingorn-Conquest, Bern 1993, White played 7 h3 and after 7 . . . Bh5 8 0-0 Black played a new move - 8 . . . Bb4. Play proceeded 9 Qb3 Qe7 10 Be2 0-0 1 1 Re1 Rab8 12 Bd2 c5 and Black had equalized. Instead of 10 Be2, White should have played 10 a3, as after 10 . . . Bxc3 1 1 bxc3 Bxf3 12 gxf3 White would have a small plus (the two Bishops and a center majority).

7 .. . Bd6 8 e4 Bb4 9 Bg5

After 9 e5 Nd5 10 Qb3 c5 Black achieves equal play.

9 .. . h6 10 Bxf6 BxcJch

88 THE QUEEN'S GAMBIT ACCEPTED

11 bxc3 Nxf6 l2 NeS Bxe2 l3 Qxe2 0-0 14 0-0 cS

As in Gligoric-Miles, Bugojno 1978. Black has the edge here. White should have tried 7 0-0 (or 9 e5 ) with even play.

Ill 83 ( 1 d4 dS 2 c4 dxc4 3 NO Nf6 4 e3 Bg4 S Bxc4 e6)

6 a3

Intending to expand on the Queenside. Now Black's best reply is 6 . . . Nbd7! with equal play.

6 ... a6 Best is 6 . . . Nbd7! as mentioned above.

7 b4 Bd6 8 Bb2 Nc6 9 Nbd2 0-0 10 Qc2 Qe7 1 1 NeS

White holds a positional plus, Belyavsky-Petrosian, USSR 1977.

pieces.

CHAPTER III 89

III B4 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 e3 Bg4 5 Bxc4 e6)

6 0-0

Completing Kingside development before committing any Queenside

Black strives for . . . e5.

6 . . . Nbd7

7 Be2 Bd6

8 Nbd2 0-0 9 Nc4

Discouraging . . . e5, but Black has other resources.

9 . . . Be7

10 b3 c5 This challenge to White's d-pawn effectively replaces Black's original

. . . e5 plan.

1 1 Bb2 Rc8 12 Ret a6 13 dxc5

And play is quite equal, Kavalek-Miles, Tilburg 1977.

90 THE QUEEN'S GAMBIT ACCEPTED

III B5 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3 Bg4 5 Bxc4 e6)

6 Nbdl

This modest looking move doesn't achieve any advantage against correct counterplay.

6 ... 7 Qb3

Nbd7

In Botvinnik-Smyslov, Monte Carlo 1968, 7 b3 was played. There followed 7 . . . c5 8 dxc5 Bxc5 9 Bb2 0-0 10 0-0 Qe7 1 1 Be2 Rfd8 1 2 Nd� Bxe2 1 3 Qxe2 Ba3 with even chances.

7 .. . 8 Ne5 9 0-0 10 a4 l l a5 1 2 Ndxc4 13 Rel

Black has judiciously simplified.

Nb6 Bh5 Bd6 0-0 Nxc4 Bel Bxc4

14 Nxc4 Rb8! The right idea. Bad is 14 . . . Bxh2ch 15 Kxh2 Ng4ch 16 Kg3 Qg5 1 7

e4 Qg6 1 8 Kfl and White refutes Black's sacrifice (Vasjukov's analysis).

15 h3 Ne4 Black has no difficulties, Ivanov-Lerner, 1979.

CHAPTER III 91

III 86 ( 1 d4 d5 2 c4 dxc4 J NfJ Nf6 4 eJ Bg4 5 Bxc4 e6)

6 hJ

This has obvious similarities to III A. The main difference is that the KB is on c4 whereas in III A it "stayed home" with the option of Bg2 (after g2-g4).

6 ••• Bh5 7 NcJ

In Donner-Matulovic, Wijk aan Zee 1974, White played 7 0-0, but after 7 . . . Nbd7 8 Be2 Bd6 9 Nbd2 0-0 10 b3 e5 1 1 dxe5 Nxe5 12 Bb2 ReS 1 3 Nxe5 Bxe2 14 Qxe2 Bxe5 15 Bxe5 Rxe5 16 Nc4 Re7 Black had equal chances.

In Damjanovic-Barczay, Pees 1964, White tried 7 Qb3 . There followed 7 . . . Bxf3 8 gxf3 Nbd7 9 Qxb7 c5 10 dxc5 Bxc5 1 1 f4 0-0 12 0-0 Nd5 1 3 Qb3 Qh4 14 Bxd5 exd5 1 5 Qxd5 Qxh3 16 Qg2 Qh5 with complications and chances for both sides.

The immediate thrust 7 g4 was tried in Gerusei-Dr. Hubner. The game continued 7 . . . Bg6 8 Ne5 Nbd7 9 Nxg6 hxg6 10 Qf3 (to be considered is 1 0 Bfl !? intending to redeploy with Bg2) 1 0 . . . Rb8 1 1 Nc3 c5 12 d5 exd5 1 3 Nxd5 Bd6 14 g5 Ne5 15 Nxf6ch KfS 16 Qd5 gxf6 17 Bd2 Be7 1 8 Qe4 Nc6 with unclear play.

7 ... Nbd7 8 0-0

Again the advance g4 is possible, but it should not be feared. For example, 8 g4 Bg6 9 Nh4 Nb6 10 Nxg6 hxg6 I I Be2 (better I I Bfl l with

92 THE QUEEN'S GAMBIT ACCEPTED

the idea of Bg2) l l . . .c5 12 g5 Nfd5 13 Nxd5 exd5 and now 14 f4 gives equality. Poor is 14 a4? because of 14 . . . Qxg5 ! 15 e4 Qf6 with an edge to Black, Moskalenko-Matulovic, Kastel Stari 1988.

The central advance 8 e4 is best met by 8 . . . Bb4 ! . Now 9 e5 Nd5 10 Bxd5 exd5 I I Qb3 Bxc3ch 12 bxc3 Nb6 gives Black even play. I n Sokolov­Matulovic, Yugoslavia 1988, White tried 9 Bd3 . After 9 . . . c5! 10 a3 Ba5 I I b4 ! cxd4! (1 l . . . cxb4 12 Nb5 bxa3ch 13 Kfl gives White the edge) 1 2 Nb5 Bc7 1 3 Qc2 ! Bb8 14 Nbxd4 0-0 15 Bb2 a5 16 Nd2 axb4 17 axb4 Rxa l 1 8 Bxa I Bd6 19 Qb3 Qe7 20 Bc3 Bg6 chances were about equal.

Finally, 8 Be2 is met by 8 . . . Bd6 9 e4 Bb4 10 e5 Nd5 I I Qb3 c5 1 2 0-0 Bxc3 13 bxc3 Qb6! with equality, Gligoric-Matulovic, Novi Sad 1 976.

8 . . . Bd6 9 e4

On 9 Be2 0-0 10 b3 c5 ! I I Bb2 cxd4 12 Nxd4 Bxe2 13 Qxe2 a6 14 Rfd l Qe7 1 5 Racl Rac8 the players agreed to a draw in Tukmakov­Smyslov, Hastings 1 972.

9 ... e5

Now White has two choices: III B6 1 10 g4 and III B62 10 Be2.

I I I 861 ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 eJ Bg4 5 Bxc4 e6 6 hJ Bh5

7 NcJ Nbd7 8 0-0 Bd6 9 e4 e5)

10 g4 (See next diagram)

CHAPTER III 93

The sharpest continuation here. White tries to use his Kingside pawn majority (after 1 1 dxe5) to seize the initiative.

10 ... Bg6 Not good is IO . . . exd4 I I Nxd4 Nxg4? in view of I2 Nf5 ! Bh2ch

(12 . . . Ne3 13 Nxg7ch Kj8 14 .fxe3 wins for White, as does 1 3 . . . Ke7 14 Qxh5 Nxc4 15 Bg5chf6 16 Nd5ch) I 3 Khi Nxflch I4 Rxf2 Bxdi I 5 Nxg7ch Ke7 I6 Rxf7ch Kd6 I 7 Be3 c5 I 8 Rxdich and White is on top, Skembris­Stramatopoiulos, Olympiad I980.

l l dxe5 Nxe5 Not I l . . .Bxe5? I 2 Nxe5 Nxe5 1 3 Be2 Ned7 I4 f4 ! Bxe4 I 5 g5 Bc6

I6 gxf6 with a very strong game for White, Lukacs-Sapi, Hungary I980.

12 Nxe5 Bxe5 l3 f4 Qd4ch! 14 Qxd4 Bxd4ch 15 Kh2 Bxc3

16 bxc3 Bxe4

1 7 g5 Bd5! 18 Rfe1ch Kf8!

Much better than I8 . . . Kd7? I9 Bd3 ! Ne8 20 c4 with excellent chances for White, Hulak-Matulovic I98 1 .

19 Balch Interesting and unclear is I9 gxf6 Bxc4 20 f5 ! c5 ! .

19 .. . Kg8 20 gxf6

To be considered is 20 Bfl !?.

20 .. . Bxc4

94 THE QUEEN'S GAMBIT ACCEPTED

Now the game Ftacnik-Matulovic, Vrsac 1981 continued 2 1 Re7 h5 ! 22 Rxc7 b5 23 Rgl Rh6! and due to the opposite colored Bishops, White cannot expect to win.

III 862 ( 1 d4 d5 2 c4 dxc4 J NO Nf6 4 eJ Bg4 5 Bxc4 e6 6 hJ Bh5

7 NcJ Nbd7 8 0-0 Bd6 9 e4 e5)

10 Be2

By unpinning the Nj3, White prepares to exchange on e5. Transposition to our main line by 10 dxe5 Nxe5 I I Be2 is also possible.

10 .. . 0-0 1 1 dxe5

Unzicker-Miles, South Africa 1979, continued I I Be3 Re8 1 2 d5 Bg6 1 3 Nd2 Bc5 14 Bxc5 Nxc5 15 Bb5 Re7 with equal chances.

1 1 . . . Nxe5 12 Nd4

Simple development by 12 Be3 doesn't wony Black, e.g. 1 2 Be3 Re8 1 3 Nd4 Bxe2 14 Qxe2 Ng6 1 5 Qb5 Qb8 16 f3 c6 17 Qb3 Bf4 with equality, Gligoric-Matulovic, Yugoslavia 1975.

The exchanges after 12 Nxe5 Bxe2 13 Qxe2 Bxe5 14 Bg5 Qe8 give even chances, Balashov-Miles, Tilburg 1977.

12 ... Bc5 Also good for equality is 12 . . . Bxe2 13 Qxe2 Ng6 14 Rd l Qc8 1 5 Bg5

Be5 16 Qe3 Nh5 17 Qf3 Nhf4 18 Nf5 Qe6, lvkov-Miles, Bled 1979.

CHAPTER III 95

Not good is 12 . . . Bg6? allowing 1 3 0! (shutting in the Bg6). For example, Tatai-Matulovic, Stip 1977, continued 13 0 ! Bc5 14 Be3 Nc6 15 Nc2 Qe7 16 Qcl Rfd8 17 Bxc5 Qxc5ch 18 Qe3 Qxe3ch 19 Nxe3 Rd2 20 Nc4 with strong pressure for White.

13 Nb3 Or 1 3 Bxh5 Bxd4 14 Nd5 c5 with equal play, Hubner-Miles, Wijk

aan Zee 1979.

13 ... Q:xd1 14 B:xd1 Bb6! 15 a4 B:xd1 16 Rxd1 a5!

Black has fully equal chances, Andersson-Miles, Bled 1979.

96 CHAPTER IV

Smyslov's Variation ( 1 d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 e3)

4 ... g6

This idea - fianchettoing the KB followed by the maneuver . . . Nf6-d7-b6 - originated with Smyslov. Black allows his opponent a broad pamt center, hoping for active counterplay with his pieces. The position closely resembles the Grunfeld Defense variations.

5 Bxc4 After 5 Qa4ch simply 5 . . . c6 should be satisfactory for Black. In Donner-Bouwmeester, Beverwijk 1958, White tried 5 Na3, but

after 5 . . . Bg7 6 Nxc4 0-0 7 Be2 c5 ! 8 0-0 Nc6 9 dxc5 Ne4 Black has excellent chances.

5 ... Bg7 6 0-0

If 6 Nc3 Black must play 6 . . . 0-0, but not 6 . . . c5 because of 7 Qa4ch Nbd7 8 Bxf7ch! Kxf7 9 Ng5ch Ke8 10 Qc4 with a powerful attack, Gulko­Suchanov, Moscow 1963.

The ambitious 6 Ne5 can be met by 6 . . . 0-0 7 Qb3 e6 (or 7 . . . Qe8) 8 0-0 Nfd7 9 f4 c5 10 Rd l cxd4 I I exd4 Nc6! 12 Be3 Na5 with advantage to Black, Padevsky-Makarov, Bulgaria 1954.

6 ... 0-0 Now if 7 Qe2 then the game Sherzer-Kholmov, Moscow 196 1 ,

continued 7 . . . Nc6 8 Rd l Bg4 9 h3 Bxf3 1 0 QxO e5 I I Nc3 exd4 1 2 exd4 Nxd4 1 3 Qxb7 c5 14 Be3 Nd7 with equal chances. Also, if 7 h3 then 7 . . . c5 8 Nc3 cxd4 9 Nxd4 Bd7 and IO . . . Nc6 gives equality.

CHAPTER IV 97

7 b3 Karpov's move. Another course is 7 Nc3. For example 7 Nc3 Nfd7 8

e4 Nb6 9 Be2 Bg4 10 Be3 Nc6 1 1 d5 Bxf3 (not 1 J . . .Ne5 12 Nxe5 Bxe2 13 Nxj7!) 12 Bxf3 Ne5 13 Be2 N5c4 (on 13 . . . c6 14 Qb3 White is better, Terpugov-Smyslov, USSR 195 1 ) 14 Bf4! (an improvement on 14 BcJ ? l c6 with excellent play for Black, Evans-Smyslov, Olympiad 1952) and now if 1 4 . . . Nxb2 then 1 5 Qb3 ! Bxc3 16 Qxc3 N2a4 17 Qa5 is excellent for White. Also, if 1 4 . . . c6 then 1 5 dxc6! bxc6 16 Qc2 gives White the edge since 16 . . . Nxb2? is answered by 17 Ba6! - from Portisch-Gheorghiu, 1 966 01ympiade.

7 ... c6 8 Bb2 Bg4

9 Nbd2 Nbd7 10 hJ Bf5 1 1 Ret Nb6 12 Bfl

White holds a persistent slight edge, Karpov-Korchnoi, Matchgame 1974.

98 CHAPTER V

Winawer's 4 . . . Be6 (1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 e3)

4 .. . Be6

This particular move comes from Winawer, who first played it at Nuremburg 1 896 ( !). Although it can catch a poorly prepared opponent off guard, current theory gives White the nod.

5 NcJ The obvious S NgS is well met by S . . . BdS, e.g. Irmai-Vyslouzil 1 9S9,

continued 6 Nc3 h6 7 e4 hxgS 8 exdS NxdS 9 BxgS Nxc3 10 bxc3 QdS I I Qg4 e6 1 2 Be2 Nd7 13 BD QaS 14 0-0 cS with equal chances.

Or if S Nbd2 then Black plays S . . . cS, meeting 6 NgS with 6 . . . BdS, e.g. 7 e4 h6 8 exdS hxgS 9 dxcS ! (not 9 Nxc4 Qxd5 10 dxc5 Qxc5 1 1 Be3 Qc7 with Black slightly better, Szilagyi-Flohr, Marienbad 19S6) with an edge for White.

Finally, if S Na3 then play can continue S . . . Nc6 6 Nxc4 BdS 7 Be2 e6 8 0-0 Be7 9 a3 0-0 10 b4 Ne4 I I Bb2 Bd6 as in Luik-Schianovsky, USSR 1970. Now 1 2 NfeS is slightly better for White.

5 • • • c6 Others are not better. For example, S . . . a6 is met by 6 a4 Nc6 7 aS bS

8 axb6 cxb6 9 NgS NaS 10 Nxe6 fxe6 I I Qa4ch Qd7 12 Bxc4 with a clear advantage for White. Or if S . . . BdS then 6 NxdS QxdS 7 Qc2 bS (7 . . . e6 8 Bxc4 Bb4ch 9 Ke2 is very strong for White) 8 a4 a6 9 b3 cxb3 10 Qxc7 e6 1 1 Qc8ch Qd8 1 2 Qxd8ch Kxd8 1 3 axb3 aS 14 Bc4 is excellent for White, Chodos-tschemikov, USSR 1962.

6 Ng5

CHAPTER V 99

Also good is 6 a4, meeting 6 . . . g6 with 7 e4 Na6 8 Ng5 ! Nc7 9 e5 Nfd5 10 Nxe6 Nxe6 I I Bxc4 Nec7 12 h4. White is on top here.

6 ... Bd5 7 e4 h6 8 exd5 9 dxc6! 10 d5 1 1 Qd4

hxg5 Nxc6 NeS

Another good move is I I Bxg5. For example I I Bxg5 Qb6 12 Be3 ! Qxb2 1 3 Bd4 Nd3ch 14 Bxd3 cxd3 1 5 0-0 and White is definitely better, Timman-Ljubojevic, Matchgame 1987.

1 1 .. . Nfd7 f6

On 1 2 . . . g6 1 3 f4 ! . 12 Bxg5

13 Be3 g5! Playing to control e5, by taking on/4, aft.erfl-/4.

Not 1 7 Rcl Nc5! .

14 Be2 Bg7 15 f4 gxf4 16 Qxf4 Qc7! 1 7 Rd1 !

1 7 ... 0-0-0 18 Qd4 a6 19 d6!

A strong "shot". Now the game Akopian-Kirov, Palma de Mallorca 1 989, continued 19 . . . exd6 20 Ne4 Nc5 2 1 Nxd6ch! Rxd6 22 Qxd6 Ncd3ch 23 Bxd3 Qxd6 24 Bf5ch with a very strong game for White.

100

CHAPTER VI Black Plays Alekhine's 3 . . . a6 (1 d4 d5 2 c4 dxc4 3 NfJ)

3 . . . a6

An original idea of Alekhine's. Black intends to respond to Qb3 with . . . b5 or . . . Ra7. Although Black's Queenside pawns are somewhat weakened, the soundness of this system has been confirmed by its adoption by such Masters as Smyslov, Flohr, Bagirov, Timman and Miles.

After 3 . . . a6 we analyze: VI A 4 a4, VI B 4 e3, VI C 4 e4 and VI D 4 Nc3.

VI A ( 1 d4 d5 2 c4 dxc4 3 NO a6)

4 a4 (See next diagram)

CHAPTER VI 101

Restraining . . . b5, but weakening b3 and b4 and contributing nothing to piece development.

4 ... Nf6 5 e3 Nc6

Also reasonable is 5 . . . Bg4. The game Garcia Gonzales-Garcia Palermo, Bayamo 1983, continued 6 Bxc4 e6 7 Nc3 Nc6 8 h3 Bh5 9 g4 Bg6 IO Nh4 Bb4 I I Nxg6 hxg6 I2 Qf3 Na5 13 Ba2 c5 with equal play.

6 Bxc4 Bg4 7 0-0 e6

8 h3 Bh5

9 Be2 Bd6 10 b3 Qe7 1 1 Nbd2 e5

Chances are equal, Michalchisin-Donachev, Lemburg I983.

VI B ( 1 d4 d5 2 c4 dxc4 3 NfJ a6)

4 e3 (See next diagram)

102 THE QUEEN'S GAMBIT ACCEPTED

Now if 4 . . . b5 then White can respond with 5 a4 Bb7 6 b3 ! . Now the game Weiner-Haberditz, Vienna 1948, continued 6 . . . e6 7 bxc4 bxc4 8 Bxc4 (Euwe recommended 8 Ba3l?) 8 . . . Nd7 9 0-0 Ngf6 10 Nbd2 c5 1 1 Ba3 Be7 12 Qc2 cxd4 1 3 Nxd4 with a small edge for White.

4 .. . Bg4 Now White has two moves: VI 8 1 5 h3 and VI 82 5 Bxc4.

VI Bl ( l d4 d5 2 c4 dxc4 3 NO a6 4 e3 Bg4)

5 h3

"Putting the question" to the Bishop - as Nimzovitch used to say.

5 ... Bh5

CHAPTER VI 103

After 5 . . . Bxf3 6 Qxf3 Nc6 7 Bxc4 e6 8 Nc3 Nf6 9 Bd2 Bd6 10 0-0-0 0-0 I I g4 ! White had some initiative, Pachmann-Lundin, Olympiade 1960.

6 g4 Bg6 7 Ne5 Be4

Not recommended is 7 . . . e6. For example, the game Pachmann-Sajtar, Bucharest 1949, continued 8 Bg2 c6 9 Nd2 Nd7 10 Nxd7! Qxd7 I I Nxc4 Qc7 1 2 Bd2 a5 13 Qe2 with e4 coming and a clear advantage for White.

8 fJ Bxb1 9 Rxbl e6

Not 9 . . . b5 since then 10 Bg2 followed by I I f4 strongly favors White.

10 Bxc4 Nd7

1 1 Qb3 Nxe5

12 dxe5 b5

13 Bd3 Ne7 As in Nei-Soloviev, USSR 1 964. Now 14 a4 leaves White slightly

better.

VI B2 ( 1 d4 d5 2 c4 dxc4 3 NfJ a6 4 e3 Bg4)

5 Bxc4

The most straightforward.

5 ... e6 Now in the game Rukavina-Hulak, Budva 198 1 , White played 6 0-0.

After 6 . . . Nf6 7 b3 Bd6 8 Bb2 0-0 9 Nbd2 Nc6 10 h3 Bxf3 I I Nxf3 Qe7 12 Qe2 e5 13 dxe5 Nxe5 14 Nxe5 Bxc5 Black had a solid game.

104 THE QUEEN'S GAMBIT ACCEPTED

In Speelman-Vorotnikov, Leningrad 1984, White played 6 Be2. There followed 6 . . . Nf6 7 0-0 c5 8 b3 Nc6 9 Bb2 ReS 10 Nbd2 Be7 1 1 dxc5 Bxc5 1 2 Rc 1 Be7 l 3 Nc4 0-0 14 Qxd8. Now Speelman recommended I4 . . . Rcxd8! , giving the further moves 15 Nb6 Nd7 I6 Nxd7 Rxd7 I 7 Bxa6 bxa6 I S Rxc6 Bxf3 I9 gxf3 Rd2 with equality.

The thrust 6 d5 gets nowhere after 6 . . . exd5 7 Bxd5 Qe7! 8 Nc3 c6 9 Bc4 Nd7 10 Qc2 Ngf6 I I 0-0 Bxf3 12 gxf3 Qc5 with even chances, Boleslavsky-Fiohr, Pamu I94 7.

Now (after 5 . . . e6) we analyze four other main choices for White: VI B2 1 6 Qb3, VI B22 6 Nc3, VI B23 6 h3 and VI B24 6 Nbd2.

VI B21 ( 1 d4 d5 2 c4 dxc4 3 NfJ a6 4 e3 Bg4 5 Bxc4 e6)

6 Qb3

Trying to exploit the absence of Black's QB from the Queenside.

6 ... BxfJ Black takes the opportunity to weaken White's Kingside pawns. Not

very good is 6 . . . Nc6, which is strongly answered by 7 Bd2 ! (but not 7 Qxb7 Na5 8 Qe4 Bf5).

7 gxfJ Now if 7 . . . Qc8 then the game Uhlmann-Duckstein, 1957, continued

8 Nc3 c5 9 d5 b5 10 Nxb5 axb5 I I Bxb5ch Kd8 (on J J . . . Ke7 12 e4!) 1 2 dxe6 fxe6 1 3 e4 Nf6! I4 Be3 Ke7 15 Rei with a clear advantage for White.

Alekhine himself recommended the peculiar Rook move 7 . . . Ra7. However, it doesn't give Black satisfactory chances. For example, 7 . . . Ra7 8 Bd2 Nf6 9 Nc3 Nbd7 10 Be2 c5 (on 10 . . . b5 White plays 1 1 a4) I I d5 exd5

CHAPTER VI 105

(in Botvinnik-Flohr, 14th USSR Championship, Black tried I J . . . e5 but after I 2 a4! 0-0 I 3 0-0 White held a definite advantage) 12 Nxd5 Nxd5 13 Qxd5 Be7 14 BcJ 0-0 1 5 0-0-0 b5 16 BdJ with a powerful attacking game for White, Meyer-Unger, postal game 1957.

7 ... b5 8 Be2 Nd7

This seems best. The "energetic" 8 . . . c5 is less good. For example, 8 . . . c5 9 dxc5 Bxc5 (9 . . . Nd7 is well met by I 0 c6 Nc5 I I Qc2 ReB I 2 Nc3l Rxc6 [ 1 2 . . . b4 13 Na4 Qa5 14 Nxc5 Bxc5 15 b3 clearly favors White] 13 Nxb5 axb5 14 b4l QbB 15 Rbl with a big plus for White, Bratkovsky­Markovich, postal game 1952/53) 10 a4 b4 1 1 0-0 Nf6 12 Nd2 Nbd7 13 Nc4 0-0 I4 Qc2 ReS 15 bJ Nd5 I6 Bb2 with advantage for White, Denker­Lundin, Groningen 1946.

9 a4 Probing Black's extended Queenside pawns.

9 ... b4 10 Nd2

Or IO f4 Ngf6 I I Bf3 c5! I2 Bxa8 Qxa8 13 Rgi (on 13 0-0 Black plays 13 . . . Q/3l) l 3 . . . cxd4 I4 exd4 Bd6 15 BeJ 0-0, as in Ilivitsky-Gretskin, USSR 1954. Black has counterplay in a murky position.

10 ... Ngf6 1 1 Ne4

On I I 0-0 c5 12 dxc5 Bxc5 Black is fine.

1 1 . . . c5 So far as in a Botvinnik-Smyslov Matchgame ( I 954 ). Botvinnik

played I 2 Nxf6ch, but in Taimanov's opinion the right course is I2 Nxc5, e.g. I 2 . . . Nxc5 1 3 dxc5 Bxc5 14 Bd2 0-0 15 0-0-0 Qb6 16 Kb1 with White retaining a small plus.

VI 822 ( 1 d4 d5 2 c4 dxc4 J NO a6 4 eJ Bg4 5 Bxc4 e6)

6 NcJ (See next diagram)

106 THE QUEEN'S GAMBIT ACCEPTED

A normal (and good) developing move.

6 ... Nf6 7 0-0 Nc6

After 7 . . . c5 8 Qe2 Nc6 9 Rdl White has a clear advantage.

8 Be2 Bd6 9 b3 0-0 10 Bb2 Qe7 1 1 Nd2 Bxe2 12 Nxe2 Ba3 13 Bxa3 Qxa3

As in Bertok-Minic, 1970; now 14 Rae I slightly favors White.

VI B23 ( 1 d4 d5 2 c4 dxc4 3 NO a6 4 e3 Bg4 5 Bxc4 e6)

6 h3 (See next diagram)

CHAPTER VI 107

"Putting the question" to the Bishop.

6 •.. Bh5 7 NcJ Nf6 8 0-0 Nc6 9 Bel Bd6 10 bJ 0-0

In the game Rivas-Pastor-Vorotnikov, Leningrad 1984, Black varied with IO . . . Bg6!?. After I I Bb2 Nd5 12 Na4 b6 13 Re i NbS 14 Ne5 Nd7 15 Bf3 White was slightly better.

1 1 Bbl 12 Rcl

Qe7 Rfd8

As in Yusupov-Timman, Sarajevo 1984. At this point, Taimanov gives 1 3 Ne5 ! Bxe2 14 Nxc6 Bxd l 15 Nxe7ch Bxe7 16 Rfxdl with a slight plus to White.

VI B24 ( 1 d4 d5 2 c4 dxc4 3 NO a6 4 e3 Bg4 5 Bxc4 e6)

6 Nbd2 (See next diagram)

108 THE QUEEN'S GAMBIT ACCEPTED

a� '� 1t ' .

a :a: Playable, but certainly giving Black no difficulties.

6 ... Nd7 7 0-0 Ngf6 8 h3

Or 8 b3 Be7 9 Bb2 0-0 lO Be2 c5 l l Nc4 ReS 12 dxc5 Nxc5 with equality, Neistadt-Beresin, Leningrad 1964.

8 ... 9 b3 10 Be2 1 1 Nxd4 12 Qxe2 13 Bb2

Now on 14 Nc4 Black plays l4 . . . b5 .

14 Racl 15 Rfd1

As in Smyslov-Botvinnik, USSR approximately even.

Bh5 c5 cxd4 Bxe2 Bc5 0-0!

ReS Qe7

Championship 1948. Play is

CHAPTER VI 109

VI C ( l d4 d5 2 c4 dxc4 3 NO a6)

4 e4

A line investigated by Borisenko and Furman. White allows Black to retain his c4 pawn, seeking compensation in a strong pawn center.

4 ... b5 5 a4

Probing Black's pawns. Now 5 . . . c6 is weak: 6 b3 cxb3 7 axb5 or 6 . . . bxa4 7 bxc4.

5 ... Bb7 Counterattacking White's e-pawn. Now we analyze: VI C l 6 Nc3, VI

C2 6 axb5 and VI C3 6 b3 .

VI Cl ( l d4 d5 2 c4 dxc4 3 NO a6 4 e4 b5 5 a4 Bb7)

6 Nc3 (See next diagram)

1 10 THE QUEEN'S GAMBIT ACCEPTED

This can lead to dangerous tactical situations which currently seem to favor White.

6 ... b4 GM Flear suggests 6 . . . c6 - tests are needed.

Not 7 . . . bxc3 8 Qb3 . 7 Bxc4 e6

8 Qb3 Nc6 9 a5 bxc3 10 Qxb7 Nb4

1 1 0-0! A new move. Previously 1 1 Bxa6 Rb8 12 Bb5ch Ke7 1 3 a6 ! was tried

here, with unclear play according to Taimanov. However, even in this line 1 2 0-0 ! (instead of 12 Bb5ch) looks vigorous.

1 1 ... Rb8 12 Bxa6!

This Queen sac gives Black problems.

12 .. . Rxb7

13 Bxb7 Cllb2 14 Bxb2 Qb8 15 a6 Nxa6 16 Bc6ch Ke7

As in the game Izeta-Vladimirov, Ceuta 1992, which continued 17 Rfb 1 f6 1 8 Rxa6 K17 19 Ra8 and Black resigned. Improvement for Black is needed!

CHAPTER VI

VI C2 ( 1 d4 dS 2 c4 dxc4 3 NfJ a6 4 e4 bS S a4 Bb7)

6 axbS

This line allows Black to equalize comfortably.

6 ... a.xbS 7 Rxa8 Bxa8 8 Nc3 e6 9 NxbS Bxe4 10 Bxc4 Bb4ch 1 1 Ke2!

1 1 1

After I I Nc3 Black plays l l . . . Ne7! and if 1 2 Qa4ch then 1 2 . . . Nbc6 1 3 0-0 BxO 14 gxf3 0-0 followed by . . . Qa8. Or 12 0-0 BxO 1 3 QxO 0-0 (not 13 . . . Qxd4 14 Qb7 Qb6 15 Bb5ch Kj8 16 Rd1) 14 Rdl c6 with approximate equality (Krasenkov).

1 1 ... Ne7 12 Bf4 BaS 13 Qa4 Nbc6 14 Nd2 BdS 15 Nb3 0-0!

16 NxaS Nxa5 1 7 Bd3 c6 18 Nc7 Nb3

As in Korchnoi-Seirawan, Biel 1993, which continued 19 Rdl c5 20 dxc5 and the players agreed to a draw.

1 1 2 THE QUEEN'S GAMBIT ACCEPTED

VI C3 ( 1 d4 d5 2 c4 dxc4 3 Nf3 a6 4 e4 b5 5 a4 Bb7)

6 b3

A standard method of undennining Black's Queenside pawn chain. Current theory gives Black satisfactory play.

6 ... Bxe4

7 Nc3 Bb7 8 axb5 axb5 9 Rxa8 Bxa8 10 bxc4 e6! 1 1 Nxb5 Bxf3 12 gxf3 Bb4ch 13 Bd2 Bxd2ch 14 Qxd2 Ne7

As in Vaiser-Chekov, USSR 1983 . Chances are balanced.

CHAPTER VI 1 13

VI D ( 1 d4 d5 2 c4 dxc4 3 NO a6)

4 Nc3

This developing move leads to complex play with mutual chances.

4 ... b5 5 a4 b4 6 Na2

In Karpov-Portisch, Tilburg 1983, White tried 6 Ne4 !?. There followed 6 . . . Nd7 7 Ned2?! (best was 7 Qc2! with unclear play) 7 . . . c3 8 bxc3 bxc3 9 Ne4 Ngf6 10 Nxc3 e6 I I e3 Bb4 12 Bd2 c5 with a small edge for Black.

edge.

6 ••• Nc6 7 e3

Best. Not 7 d5? Na5 8 Nxb4 Bb7! and 9 . . . e6 with Black having an

7 .. . b3 8 Nc3 Nb4 9 Bxc4 Nc2ch 10 Kfl Nxa1 1 1 Ne5 e6 12 QO (6 13 Qc6ch! Ke7 14 Qxa8 fxe5 15 dxe5 Kn

1 14 THE QUEEN'S GAMBIT ACCEPTED

16 Ke2 Qh4 As in Mageramov-Vorotnikov, 1986. Play is unclear.

CHAPTER VII Mannheim Variation 4 Qa4ch

(1 d4 d5 2 c4 dxc4 3 NfJ Nf6)

4 Qa4ch

1 15

The main idea behind this Queen sortie is to transpose into a variation of the Catalan: 1 d4 Nf6 2 c4 e6 3 g3 d5 4 Bg2 dxc4 5 Qa4ch. Bogoljubov introduced the conception against Alekhine at Mannheim 1934.

A close relative of the Mannheim was seen in the game Agdestein­Michael Adams, Oslo I994. It began with the unusual move order I Nf3 d5 2 c4 dxc4 3 Qa4ch Nd7 4 g3 Ngf6 5 Bg2 c6 6 Qxc4 e5 7 0-0 Bd6 S Qc2 0-0 9 d4 (finally White transposes into a Queen's pawn structure - otherwise it would be an English or Reti set-up) 9 . . . exd4 10 Nxd4 Bc5 1 1 Rd i Qb6 1 2 e3 Ne5 1 3 f4 Bg4 I4 fxe5 Bxdi I5 Qxd l RadS I 6 exf6 Bxd4 I 7 Qb3 Bxf6 I S Qxb6 axb6 I9 Bf3 Rd3 20 Kf2 ReS 2 I a4 RaS 22 Be4 Rd6 23 Ra2 b5 24 a5 g6 25 b4 ReS and Black went on to win on the 40th move.

Before analyzing our main line replies for Black (4 . . . c6 and 4 . . . Nc6), we note some other Black 4th move alternatives:

In McCambridge-Tarjan, San Jose 19S3, Black played 4 . . . Bd7 and after 5 Qxc4 Bc6 6 Nc3 Nbd7 7 Bg5 e6 S e4 White had a small pull.

In Alekhine-Fine, Kemeri I 937, 4 . . . Qd7 was tried. There followed 5 Qxc4 Qc6 6 Na3 ! Qxc4 7 Nxc4 e6 S a3 c5 9 Bf4 Nc6 10 dxc5 Bxc5 I I b4 Be7 1 2 b5 NbS 13 Nd6ch Bxd6 14 Bxd6 Ne4 I5 Bc7 with White holding a positional advantage.

1 16 THE QUEEN'S GAMBIT ACCEPTED

Inferior is 4 . . . Nfd7. For example, Pachmann-Janetchek, Barcelona 1975, continued 5 e4 e6 6 Bxc4 c5 7 d5 exd5 8 Bxd5 Be7 9 Qb3 with a big advantage for White.

Now we return to our main line choices: VII A 4 . . . c6 and VII B 4 . . . Nc6.

CHAPTER VII

VII A ( 1 d4 dS 2 c4 dxc4 3 NfJ Nf6 4 Qa4ch)

A solid reaction.

4 ••• c6

5 Qxc4 Bf5

1 1 7

Interesting is 5 . . . Bg4. Now if 6 Nbd2, the game Andrianov­Kharitonov, Junnala 1983, continued 6 . . . Nbd7 7 g3 e6 8 Bg2 Bd6 9 Qb3 Nb6 10 0-0 0-0 1 1 Nc4 Nxc4 1 2 Qxc4 Qa5 13 Bd2 Qh5 with even chances. After the more active 6 Nc3 Black can play 6 . . . Bxf3 !? (a new and very interesting concept evolves from this) 7 gxfl g6! 8 e4 Bg7 9 Bg5 (poor. Better was 9 Be3) 9 . . . 0-0 10 0-0-0 Nbd7 I I h4 Nb6 12 Qd3?! (1 2 Qb3 Nh5l - Krasenkov) l 2 . . . Nh5 with the upper hand for Black, Krasenkov-Garcia Palermo, Andorra 199 1 .

The tempting 5 . . . Be6 favors White: 6 Qc2 g6 7 e4 Bg4 8 Be3 Bxfl 9 gxO Bg7 10 Nc3 0-0 1 1 Be2 Nbd7 12 0-0-0 b5 1 3 Kb1 Qb8 14 f4 a5? ! 1 5 e5 Nd5 1 6 Nxd5 cxd5 1 7 f5 with a distinct advantage to White, Golz-Zinn, Germany 1963.

(See next diagram)

1 18 THE QUEEN'S GAMBIT ACCEPTED

After 5 . . . Bf5 ! 6 e3 e6 7 Bd3 Bxd3 8 Qxd3 c5 ! gives equal chances. We now explore two other White choices here: VII AI 6 Nc3 and VII A2 6 g3.

VII Al ( 1 d4 d5 2 c4 dxc4 3 NfJ Nf6 4 Qa4ch c6 5 Qxc4 Bf5)

6 Nc3

Developing while contesting e4.

6 . . . e6 7 g3

CHAPTER VII 1 19

In Smyslov-Hubner, Candidates Matchgame 1983, White played 7 Qb3 but after 7 . . . Qb6! 8 Qxb6 axb6 9 Nh4 b5 10 Nxf5 exf5 1 1 e3 Nbd7 12 Bd3 g6 Black has the edge.

7 ... Nbd7 Best. After 7 ... Ne4 the game Trifunovic-Radulescu, Budapest 1948,

continued 8 Bg2 Nxc3 9 bxc3 Be4 10 0-0 Be7 1 1 Nd2 Bxg2 12 Kxg2 Qd5ch 1 3 e4 with an edge for White.

8 Bg2 Be7 9 0-0 0-0 10 Bg5

Or 1 0 e3 Ne4 l l Nd2 Nxd2 12 Bxd2 e5 1 3 e4 Nb6 (13 . . . exd4 14 Qxd4 slightly favors White) 14 Qb3 exd4 15 exf5 dxc3 16 Bxc3 with even chances, Andersson-Hubner, Wijk aan Zee 1983.

10 .. . h6 1 1 Bxf6 Bxf6 12 Rfd1 aS 13 e3 a4 14 Qe2 Qa5 15 Nd2 Nb6

Black's two Bishops offset White's spatial plus, giving about equal chances, Seirawan-Portisch, Tilburg 1983.

VII A2 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 Qa4ch c6 5 Qxc4 Bf5)

6 g3

1 20 THE QUEEN'S GAMBIT ACCEPTED

Immediately transposing into a sort of Catalan.

6 ... Nbd7 7 Bg2 e6 8 0-0 Be7 9 Bg5

In Rashkovsky-Bagirov, Kusadasi 1990, White tried 9 a4. There followed 9 . . . 0-0 10 Nc3 Ne4 I I a5 Nxc3 1 2 bxc3 Be4 1 3 Nd2 Bxg2 1 4 Kxg2 c5 1 5 Nf3 Qc7 1 6 e4 and now Bagirov gives 1 6 . . . Nf6! 1 7 Rei Rfd8 18 Bf4 Qc8! with equality. In Agdestein-Illescas, Biel Interzonal 1993, White played 9 Nbd2. There followed 9 .. . 0-0 10 Qb3 Qb6 I I Nc4 Qa6 1 2 Bf4 Nb6 1 3 Bd6 Bxd6 1 4 Nxd6 Bg6 1 5 a4 Rad8 16 Nxb7 Qxb7 17 a5 Be4 with equal chances. Another possibility is 9 Nc3 0-0. The game Akopian­Shirov, Wijk aan Zee 1992, continued 10 e3 Ne4 I I Qe2 Nxc3 12 bxc3 Be4 13 c4 c5 14 Rd l Qc7 15 Bb2 Nb6 16 Rac l Bf6! with equality.

9 ... 0-0! After 9 . . . h6 10 Bxf6 Bxf6 I I NcJ White held a small plus in

Gheorghiu-Wittwer, Lugano 1983.

10 Nbd2 a5 1 1 Bxf6 Nxf6 12 Ne5 Qb6 13 e4 Bg6 14 Nxg6 hxg6

As in Kurajica-Seirawan, Indonesia 1983. Black has a solid position.

CHAPTER VII 121

VII B ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 Qa4ch)

4 ••. Nc6

This emphasizes piece play against White's center - an idea which would have pleased Tchigorin, who specialized in such stratagems.

After 4 . . . Nc6 the normal-looking 5 g3 is poor: 5 . . . Be6 6 Bg2 Qd7 7 Nc3 as in Tukmakov-Kozlov, 1984; and now 7 . . . Nd5 gives Black the edge.

Also, if 5 e3 then 5 . . . Nd7 6 Qxc4 Nb6 7 Qc2 g6 8 Bb5 Bd7 9 Bd2 Bg7 10 0-0 0-0 gives equality; from Andersson-Fernandez Garcia, Olympiad 1984.

Interesting is 5 Qxc4. For example, 5 Qxc4 Be6 6 Qa4 Bd5 7 e3 e6 8 Nc3 Bb4 9 Bb5 0-0 10 0-0 with a small edge for White.

5 Nc3 Nd5

6 e4 After 6 Qxc4 Black should play 6 . . . Ndb4. For example, the game

Andersson-Korchnoi, Johannesburg 198 1 , continued 7 Qb3 Nxd4! 8 Nxd4 Qxd4 9 Be3 Be6 10 Qa4ch Bd7 I I Qb3 Be6 12 Qa4ch and the players agreed to a draw.

In Odendahl-Dlugy, Philadelphia 1985, White played 6 Ne5. There followed 6 . . . Nb6 7 Nxc6 Qd7 8 Ne5 Nxa4 9 Nxd7 Bxd7 10 Nxa4 Bxa4 I I e3 Bc6 1 2 f3 b5 13 a4 bxa4 14 Bxc4 Rb8 with a clear superiority for Black.

6 ... Nb6 7 Qd1 Bg4 8 d5

122 THE QUEEN'S GAMBIT ACCEPTED

After 8 Be3 BxD 9 gxD e6 10 Be2 Qh4 I I 0-0 0-0-0 Black held an edge, Cruz-Sanguinetti, Buenos Aires 1963.

8 ... Ne5 9 Qd4!?

Or 9 Bf4 Ng6 10 Bg3 (better than 10 Be3 e6 with Black slightly better, Botvinnik-Petrosian, Matchgame 1963) 10 . . . e5 ! I I dxe6 Bxe6 1 2 Qxd8ch Rxd8 1 3 Bxc7 Rd7 14 Bxb6 axb6 15 Ng5. So far as i n the game Rajkovic-Barle, Yugoslavia 1983 . Now instead of the actually played 15 . . . Rd4?, correct was 15 . . . Ne5 ! with equal chances, e.g. 16 Nxe6 fxe6 17 Rd 1 g5 ! or 17 f4 Nd3ch 18 Bxd3 cxd3 ! .

9 . .. NxfJch 10 gxfJ BxfJ 1 1 Rg1 Qd6!

An improvement on l l . . . e6 12 Qe3 ! Bh5 13 Qh3 g6 14 dxe6 fxe6 1 5 Qxe6ch with the upper hand for White, Crouch-Sadler, Hastings 1992.

12 a4 a6 13 e5 Qd7

14 e6 fxe6

15 Bh3 Nxd5

16 Rg3 Nxc3

17 Qxd7ch Kxd7

18 RxfJ Ne4 As in Crouch-Sadler, Cappelle Ia Grande 1993. White had no

sufficient compensation for the missing pawns.

CHAPTER VOl Elastic Variation 4 N c3

(1 d4 dS 2 c4 dxc4 3 NO Nf6)

4 Nc3

1 23

This variation is currently quite popular. Now, If Black likes, he can transpose into the Slav Defense with 4 . . . c6. But for our purposes, we shall ignore this possible transposition.

An important move order question arises after 1 d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 Nc3 c5 5 d5 e6 6 e4 exd5, since now 7 e5 seems to create problems for Black (according to current theory - but "stay tuned" since improvements are hardly out of the question). To avoid this line Black can adopt the " Anti-Elastic" Variation which goes 1 d4 d5 2 c4 dxc4 3 Nf3 c5 !? since now after 4 d5 e6 5 e4 the Black KN isn't onf6, so White doesn't have the aggressive e5 available. See Chapter IX for this line.

Since 1 d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 Nc3 c5 leads to the "problem line" with 5 d5 e6 6 e4 exd5 7 e5, Black can opt for our other 4th move response of 4 . . . a6 !?.

So, after 4 Nc3 we analyze: VIII A 4 . . . c5 and VIII B 4 . . . a6.

1 24 THE QUEEN'S GAMBIT ACCEPTED

VIl l A ( 1 d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 Nc3)

4 .. . c5

As mentioned in our introduction, this line can lead to difficulties for Black according to current theory. Hence we recommend VIII B 4 . . . a6 as the best choice according to current analysis.

Now, after 4 . . . c5, White has: VIII AI 5 d5 and VIII A2 5 e4.

VIII At ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 Nc3 c5)

5 d5

CHAPTER VIII

Currently considered to be White's most promising line.

5 . . . e6

1 25

After 5 . . . Bf5 the game Checkov-Adamsky, Tiflis I 986, continued 6 e3 e6 7 Bxc4 exd5 8 Nxd5 Nc6 9 0-0 Bd6 10 Ng5 ! Bg6 (on 1 0 . . . Nxd5 White has 1 1 Nxj7l) I I f4 h6 I 2 Nh3 Be4 1 3 Nxf6ch Qxf6 I4 Nf2 Qe7 I 5 Nxe4 Qxe4 I 6 Bd5 Qe7 I7 e4 with a big advantage for White.

6 e4 exd5 After 6 . . . Nxe4!? 7 Nxe4 exd5 8 Nc3 Be6 9 Ng5 Nc6 IO Nxe6 fxe6 I I

g3 ! Qf6 I 2 Bh3 Be7 I 3 0-0 0-0 I4 f4 Rad8 I5 f5 ! White had a very strong game, H. Olafsson-Fiear, Wijk aan Zee I987.

7 e5! This move gives Black some problems. Poor is 7 exd5 Bd6 8 Bxc4

0-0 9 0-0 Bg4 10 Bg5 Nbd7 1 1 Ne4 Be5 with fine chances for Black, Scheeren-Pleister, I985.

7 . • . Nfd7 Bad is 7 . . . Ng8?. The game Vladimirov-Hodos, USSR 1958,

continued 8 Qxd5 Nc6 9 Bxc4 Be6 IO Qe4. Also no good is 7 . . . d4? 8 Bxc4! Nc6 9 exf6 dxc3 10 Qe2ch Kd7 1 1

Bf4 Qa5 1 2 Rdich Nd4 1 3 Bb5ch as in Uhlmann-Wade, Hastings 1958/59. Finally, 7 . . . Ne4 is answered by 8 Qxd5 Nxc3 9 Qxd8ch Kxd8 10

bxc3 B e6 I I Ng5 Nd7 I 2 Nxe6ch! fxe6 13 f4 Nb6 I4 a4 ! g5 ! I 5 a5 Nd5 I6 f5 ! with White in control, Gligoric-Nikolic, Niksic I 983 .

8 Bg5 Be7 8 . . . f6 has been receiving some rough treatment lately. For example,

8 . . . f6 9 exf6 gxf6 (weaker is 9 . . . Nxf6 10 Bxf6 gxf6 1 1 Nxd5 b5 12 Qe2ch KP 13 0-0-0 with a tremendous game for White, Korelov-Miesis, I 964) IO Qe2ch (after 10 Qxd5 .fxg5 1 1 0-0-0 Q/6 12 Bxc4 Nc6 13 Rhe 1 Be 7 14 Bb5 Kj8 15 Qd2 Nb6 Black managed to consolidate and win, Chlong­Skomorokin, Silvaplana 1993) IO . . . Kf7 I I Nxd5 Bg7 I 2 Be3 b5 I 3 0-0-0 Qa5 I4 Bd2 Qd8 I5 Ng5ch fxg5 I6 Qh5ch Ke6 I7 Bxg5 Qa5 I8 Nf4ch and Black resigned, from Ivanchuk-Z. Polgar, Melody Blitz I 992.

9 Bxe7 Qxe7 10 Nxd5 Qd8 1 1 Qcl Nb6

After I l . . .Nxe5 the game Levin-Bouton, Dortmund I992, continued I 2 Qe4 Be6 I 3 Qxe5 Qxd5 I4 Qxg7 M I 5 Rdi Qe4ch 16 Be2 Nd7 I 7 Ng5 Qg6 I 8 Nxe6 Qxe6 I9 0-0! ( a key improvement over /9 Qxh7 which allowed Black to equalize in van der Sterren-Z. Polgar, Wijk aan Zee I 990, with /9 . . . 0-0-0 20 Qc2 Rfe8 21 h4 Ne5/) 19 . . . 0-0-0 20 Bg4 f5 2 I Bh3 Nf6 22 Rde I Qd5 23 Re7 and Black could not find a good defense.

1 26 THE QUEEN'S GAMBIT ACCEPTED

12 0-0-0 N:x:dS! Best. After 12 . . . N8d7? 1 3 Nxb6 Qxb6 14 Bxc4 Qh6ch 15 Kbl 0-0 16

Rd6! Black is in real trouble, Lukacs-Z. Polgar, Budapest 1986.

13 Bxc4 0-0! Pachmann's recommendation. He now gives 14 Bxd5 Qe7 with

White holding a small advantage.

VIII A2 ( 1 d4 dS 2 c4 d:x:c4 3 NfJ Nf6 4 Nc3 cS)

S e4

Seizing e4 and preparing rapid development after Bxc4.

S . . . cxd4 6 Qxd4

After 6 Nxd4 Black plays 6 . . . e5 7 Ndb5 a6! 8 Qxd8ch Kxd8 9 Na3 b5 10 Nc2 Bb7 1 1 Bg5 with a big plus for Black; from Hebert-Spraggett, Quebec 1986.

6 ... Qxd4 7 Nxd4 eS

Poor is 7 . . . Bd7? because of 8 e5 Ng4 9 e6! fxe6 10 Bxc4 Kf7 1 1 Ne4 Nc6 1 2 Nxe6 Bxe6 1 3 Ng5ch Kf6 14 Nxe6 Nge5 1 5 Bb3 Na5 16 Bg5ch and Black was in trouble in Georgiev-Ehlvest, Reggio Emilia 1988/89.

8 NdbS Kd8! 9 Be3

CHAPTER VIII 1 27

After 9 Bxc4 Pliester gives 9 . . . a6 10 Na3 b5 1 1 Bd5 Ra7 12 Nc2 b4 l 3 Ne2 Nxd5 14 exd5 Rd7 1 5 Ne3 Bb7 or l 3 Na4 Bd7 14 Nb6 Bc5 1 5 Nc4 Bb5 with a substantial advantage to Black.

9 . . . Be6 Not 9 . . . a6? because of 10 Bb6ch and 1 1 Nc7.

10 0-0-0ch Best. Poor is 10 Nd5 Bxd5 I I exd5 Bb4ch 1 2 Nc3? (relatively best

was 12 Ke2 with only an edge for Black) 1 2 . . . Nbd7 l 3 Bxc4 ReS as in Thompson-Motwani, Troon 19S6.

In Gurevich-Z. Polgar, Las Vegas 19S6, White played 10 Bxa7. There followed 10 . . . Nbd7 1 1 Be3 Bb4! (not 1 J . . .Ng4? 12 Nd5.f) 12 f3 Ra5 ! 1 3 Kf2 Ke7 14 Be2 ReS 15 Rhd1 NeS 16 Nd5ch Bxd5 17 Rxd5 Rc6! with equal chances.

In Winants-Karoli, Brussels 19S6, 10 Nxa7 was tried. After l O . . . Nbd7 1 1 f3 Bb4 12 Nb5 Ra5 13 Kf2 Ke7 14 Be2 ReS play was even.

10 . . . Nbd7

1 1 Bxa7 12 Nxa7 13 Nd5!

Rxa7!? Ng4!

Now the right way for Black is l 3 . . . Nxf2! since after 14 Bxc4 Bc5 15 Nb5 Nxh1 16 Rxh1 White i s only a little better (Pachrnann).

128 THE QUEEN'S GAMBIT ACCEPTED

VIII B ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 NcJ)

4 . . . a6

Showing intentions of reinforcing the c4 pawn with . . . b5. Now if 5 a4 then 5 . . . Nc6! is O.K. for Black. For example, 5 . . . Nc6 ! 6 e4 (6 e3 Na5l) 6 . . . Bg4 7 d5 (poor is 7 e5 Bxf3 8 gxj3 Nd5 9 Bxc4 Nb6 with the upper hand for Black, Ardianshah-Hort, Indonesia 1982) 7 . . . Ne5 8 Be2 Bxf3 9 gxf3 e6 10 f4 Ng6 I I Bxc4 exd5 1 2 Nxd5 Nxd5 13 Qxd5 Bb4ch 14 Ke2 Qe7 1 5 Qxb7 0-0 with unclear play according to Neistadt.

In Aloni-van den Berg, Tel Aviv 1958, White tried 7 Be3 . Then Black (instead of the actually played 7 . . . Na5.1?, which led to murky play) should have played 7 . . . e5! 8 dxe5 Nd7 9 Bxc4 Ndxe5 10 Nxe5 Nxe5 I I Be2 Qxdl ch 1 2 Rxdl Be6 with equality (Pachmann).

5 e4 Clearly the most aggressive procedure. White has a proud center, but

Black holds an extra pawn.

5 ... b5 6 e5 Nd5

(See next diagram)

CHAPTER VIII 129

Now White has two main choices: VIII 81 7 Ng5 and VIII B2 7 a4.

VIII 81 ( 1 d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 Nc3 a6 5 e4 b5 6 e5 Nd5)

7 Ng5

Threatening 8 Qf3, attacking both d5 and .f7. Now, Black shouldn't play 7 . . . Bf5? since then 8 Nxf7! is powerful, e.g, 8 . . . Kxf7 9 Qf3 e6 10 g4 Bb4 1 1 gxf5 R18 12 Rg l exf5 1 3 Qh5ch Kg8 14 Bh6 Rf7 1 5 e6 Re7 16 Bxg7! Rxe6ch 1 7 BeSch! and White wins, Hottes-Bialas, Bad Pyrmont 196 1 . The move 7 . . . e6 is met by 8 Qh5 !? (not 8 Q/3? Qd7 9 Nxd5 exd5 10 a3 Nc6 1 1 Be3 Nd8 12 Be2 Qfl 13 Qg3 h6 14 Nh3 c6 15/4 Qc2! with the upper hand for Black, Alekhine-Bogoljubov, Matchgame 1934 ); for

130 THE QUEEN'S GAMBIT ACCEPTED

example, 8 Qh5 !? Qd7 9 Be2 Bb7 lO Bf3 g6 I I Qg4 b4 12 Nce4 Nc6 1 3 Be3 h6 1 4 Nh3 0-0-0 with complications and chances for both sides, Petursson-Zaltsman, Reykjavik 1984.

However, below in our main line we give Black's most promising move, 7 . . . Nxc3 .

7 . . . NxcJ

8 bxcJ This seems best. After 8 Qf3 Qd5 ! 9 bxc3 Qxf3 lO Nxf3 Black is

somewhat better.

8 ... (6

9 Ne4 On 9 Qf3 Pachmann recommends 9 . . . c6!.

9 ... Qd5

10 Nc5 Nd7

1 1 Nxd7 Bxd7 As in Buslaev-Gurgenidze, USSR 1964. Black has a clear advantage.

VIII B2 ( 1 d4 d5 2 c4 dxc4 J NO Nf6 4 NcJ a6 5 e4 b5 6 e5 Nd5)

7 a4

Pressuring Black's pawn chain. Now weak is 7 . . . Bf5 in view of 8 Nh4! Nxc3 9 bxc3 Be4 lO e6 ! Qd5 1 1 exf7ch Kd8 1 2 Be2; from Miles­Heinbuch, Ostende 1985.

CHAPTER VIII 131

The logical 7 . . . Bb7 is open to evaluation. For example, the game Chiburdenidze-Sturua, 1982, continued 8 e6 f6 9 Ne4 Nb4 10 Nc5 BxO 1 1 gxfl N8c6 1 2 Be3 Nxd4 1 3 Bxd4 Qxd4 14 Qxd4 Nc2ch 1 5 Kd2 Nxd4 16 Kc3 Rd8 and now Chiburdenidze gives 1 7 Nxa6! Nxe6 I8 axb5 and I9 Bxc4 with a clear advantage to White. However, in the recent game Nemet­Adianto, New York I994, White played 9 Be2 (does this indicate a lack of faith in the move 9 Ne4 in the above game? - probably so). There followed 9 . . . g6 10 O-O· Bg7 I I Ne4 Nb6 I2 Nc5 Bd5 l 3 Rei 0-0 I4 Bf4 Nc6 I 5 a5 Nc8 I6 Nb7 Qe8 I 7 Nc5 Nd8 I8 Nd2 Qc6 I9 BO Bd6 20 b3 c3 2 I Nfl BxO 22 gxf.3 b4 with a draw is 44 moves. Certainly there is "food for thought" here.

Now (after our main line move, 7 a4) we examine two lines for Black: VIII B2 I 7 . . . e6 and VIII B22 7 . . . Nxc3 .

VIII B21 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 Nc3 a6 5 e4 b5 6 e5 Nd5

7 a4)

7 . • • e6

Black prepares to continue with the development of his KB; in addition, Black is quite prepared to return the gambit pawn in order to gain vigorous piece activity on the Queenside and in the center.

8 axb5 Bb4

9 Qc2

132 THE QUEEN'S GAMBIT ACCEPTED

Also possible is 9 Bd2 Nb6 I 0 bxa6 R..xa6 I I R..xa6 Bxa6 12 Be2 Nc6 13 Bg5 ! Qd7 14 0-0 with only a small edge for White, V. Scherbakov­Fominic, USSR 1988. One has the impression that lvanchuk (our main line involves him on the Black side) would have an improvement on this line.

9 ... Nb6 10 Be2

Interesting is 10 Qe4 Qd5 I I Qg4 axb5! 12 Rxa8 Qxa8 1 3 Qxg7 Qa I ! ! (a stunning tactical thrust. After /3 . . . Rj8 /4 Be2 Bxc3ch 15 bxc3 Qal /6 0-0 White holds a clear advantage according to Fominic and Cesarski) 14 Qxh8ch Kd7 15 Kd l Bxc3 16 bxc3 Nd5 17 Qxh7 Nxc3ch 18 Kd2 Nc6 19 Qc2 Na4 20 Bxc4 bxc4 2 1 Qxc4 Nb6 22 Qc3 Qa2ch 23 Kdl Ba6 24 ReI with a very messy position.

Another way is I 0 bxa6 Rxa6 I I Rxa6 Bxa6 12 Be2 and according to Fominic and Cesarski, White holds a small edge.

10 .. . axb5 1 1 Rxa8 Nxa8

It isn't often that one finds a Knight situated on the Rook's square after I I moves ! Notice that all of Black's minor pieces are clustered on the Queenside, ready to stir up counterplay.

12 0-0 Qd7 Nb6 13 Qe4

14 Qg4 KfB! Not 14 . . . 0-0 since 15 Ne4 (threat: Nf6ch) would commence a

tremendous attack.

15 Ne4 This looks menacing, but lvanchuk proves Black has resources.

15 .. . Nc6 16 Ng3 Ne7 1 7 Nh5 Ng6 18 h4 Bb7 19 Bdl Be7 20 Bc2 b4 21 Ng5 Nd5 22 f4 Bxg5 23 fxg5 c3 24 b3 Nge7 25 QfJ Nf5 26 Qfl g6 27 g4 Nxd4!

CHAPTER VIII 133

As in the game Kramnik-Ivanchuk, Monaco 1994 (a blindfold game!) . After 28 Qxd4 gxh5 29 gxh5 Ke8 30 Qf2 Ba6 Black soon won.

VIII 822 ( 1 d4 d5 2 c4 dxc4 3 NO Nf6 4 Nc3 a6 5 e4 b5 6 e5 Nd5

7 a4)

7 . . . Nxc3

8 bxc3 Qd5 After 8 . . . Bb7 9 e6! play becomes intense. For example, 8 . . . Bb7 9 e6!

f6 (or 9 .. .fxe6 10 Be2 Qd5 1 1 Ng5 Qxg2 12 Rf1 Bd5 13 axb5 axb5 14 Rxa8 Bxa8 15 Bf4 g6 16 Nxe6 Na6 1 7 Qa1 Bb7 18 Qa5 Qc6 19 d5! with a clear advantage to White, Bronstein-Liavdansky, USSR 1965) 10 Be2 Bd5 ! I I 0-0 c6 ! 1 2 Nh4 Bxe6 13 axb5 cxb5 14 Bf3 Bd5! 15 Nf5 and now, instead of l 5 . . . g6? (as in Piket-Schlosser, Munich 1989) Black should play l 5 . . . e6 and after 1 6 Rei Kf7 17 Bxd5 Qxd5 ! 18 Ne3 Qd7 19 d5 play is unclear.

In Sosonko-Diugy, New York 1984, White played 10 Be3. Then play continued IO . . . Bd5 ! I I Nh4 g6 12 Qb l Qd6 13 axb5 Qxe6 14 Be2 (after 14 bxa6 Nxa6 15 Qb5ch Kj7 Black is on top - Sosonko) l4 . . . Nd7 15 0-0 with an unclear position.

However, a more recent game, Griego-Hertan, U.S. 1987, may hold a promising lOth move alternative for Black. Instead of IO . . . Bd5 Black played I O . . . Qd5. There followed I I Qbl Nc6! 1 2 axb5 axb5 1 3 Rxa8ch Bxa8 14 Be2 Nd8! 1 5 Ng5 !? fxg5 16 Bf3 Qxe6 17 Bxa8 c6 18 d5 ! Qxd5 19 0-0 e5 ! with complex play and roughly even chances.

1 34 THE QUEEN'S GAMBIT ACCEPTED

9 g3 Bb7 10 Bg2 Qd7 1 1 Ba3

In Kavalek-Miles, Wijk aan Zee 1978, White essayed 1 1 Nh4. There followed 1 l . . .c6 (not l l . . . Bxg2 12 Nxg2 e6 13 Q/3 Ra7 14 axb5 Qxb5 15 N/4 with advantage to White) 12 f4 e6 l 3 f5 !? exf5 14 0-0 g6 1 5 Bg5 with a dangerous initiative for White. However, Black can play 1 l . . .Nc6! (instead of Miles' I I . . . c6) and after 1 2 f4 e6 1 3 0-0 g6 play transposes into the game Morgulev-Zhumilin, USSR 199 1 , which we analyze in the paragraph below.

In Morgulev-Zhumilin, USSR 199 1 , 1 1 0-0 was played. Then play proceeded 1 l . . .e6 1 2 Nh4 Nc6 l 3 f4 g6 14 f5 ! gxf5 1 5 Bg5 Be7! 16 Bxe7 Qxe7 1 7 Qb1 b4! 18 cxb4 Rb8 19 d5 Nxb4 20 d6 Qd8 2 1 Nxf5 exf5 22 Qxf5 0-0 23 dxc7 and now, instead of the actually played 23 . . . Qxc7 (which led to a draw), Black should have played 23 . . . Qd4ch! as after 24 Qf2 Qxf2ch 25 Rxf2 Rbc8 26 Bxb7 Rc7 Black's position would be slightly preferable.

1 1 . . . g6 This move has become more popular than the older (but still

playable) I l . . .Bd5.

12 0-0 Sharper is 12 h4!?. The game Sakaev-lbragimov, Kherson 1 99 1 ,

continued 1 2 . . . Bg7 13 h5 0-0 14 Kf1 ! (the dangerous-looking tactic 14 hxg6 fxg6 15 Rxh7?! is refuted by 15 . . . Bxj3 16 Bxj3 Q/5! 1 7 Rxg7ch Kxg7 18 Bxe7 Qxj3 19 Bxj8ch Qxj8 20 axb5 a5 with an edge for Black - Sakaev) 14 . . . Bd5 15 hxg6 fxg6 16 Ng5 Bxg2ch 17 Kxg2 Qd5ch 18 Kg1 h6 19 Nh3 and White went on to win. Nonetheless, this 199 1 game is superceded by our main line game which occurred later in 199 1 - obviously lbragimov has an improvement for Black over the above game.

12 .. . Bg7

l3 Ret 0-0 14 e6

In the game Zakharstov-lbragimov, Smolensk 199 1 , White tried 14 Bc5, but after 14 . . . Bd5 15 Ng5 Nc6 16 e6 fxe6 17 Qg4 Bxg2 18 Kxg2 Qd5ch 19 f3 e5 20 Ne6 exd4 2 1 cxd4 b4 22 Nxc7 h5 23 Qe6ch Qxe6 24 Rxe6 Nxd4 25 Rxg6 Nb3 26 Bxb4 Nxa1 27 Nxa8 (27 Bc3 Rf6/) 27 . . . Kf7 28 Rxa6 c3 and Black was on top.

14 .. . fxe6

15 Ne5 Qc8 16 Bh3

CHAPTER VIII 135

Unclear is 16 Bxb7 Qxb7 17 Qg4 Qd5 18 Bxe7 ReS 19 Bc5 -Ibragimov.

16 .. . BdS 1 7 Bxe7?

Better 1 7 Ng4 and Ne3.

17 .. . ReS 18 Bg5 cS 19 dxc5 QxcS 20 Be3 Qc7 21 Bf4 Qb7

Black is definitely better, A. Schneider-Ibragimov, USSR 199 1 .

1 36

CHAPTER IX

The "Anti-Elastic" Variation

( l d4 d5 2 c4 dxc4 3 NfJ)

3 .. . cS

4 dS An aggressive space-gaining move. For 4 e4 see Chapter X B (this

coming about by a transposition of moves). As you may remember in our introduction to Chapter VIII, this

"Anti-Elastic" Variation can be used to avoid a line which is currently favorable to White, analyzed in VIII A.

4 ... e6 S Nc3 exdS 6 QxdS QxdS 7 NxdS Bd6

Also playable is 7 . . . Na6. For example, 7 . . . Na6 8 e4 Ne7 9 Bxc4 Nxd5 10 exd5 Bd6 1 1 Nd2 0-0 12 Bxa6 bxa6 13 Nc4 Rd8 14 0-0! Bf5! 1 5 Rd1 Bc2! 16 Rd2 Bg6 with only a small plus for White, Khenkin-Yuferov, USSR 1990.

8 Nd2 (See next diagram)

CHAPTER IX 137

Now we have two lines: IX A 8 . . . Ne7 and IX B 8 . . . Nc6.

IX A ( 1 d4 d5 2 c4 dxc4 3 NfJ c5 4 d5 e6 5 Nc3 exd5 6 Qxd5 Qxd5

7 Nxd5 Bd6 8 Nd2)

plus.

8 ... Ne7

This line is quite playable, though current theory gives White a small

9 Nxc4 Nxd5 10 Nxd6ch Ke7 1 1 Nxc8ch Rxc8

138 THE QUEEN'S GAMBIT ACCEPTED

12 gJ Nb4 13 BhJ Rd8 14 0-0 N8c6 15 Bg5ch

In a game Khuzman-Greenfeld, Beersheva 1993, White eschewed the "finesseful" 1 5 Bg5ch, instead preferring 15 Be3 . There followed 15 . . . b6 16 Rfcl a5 17 Bg2 Rac8! 18 a3 Nd5 19 Bxc5ch bxc5 20 R.xc5 Ke6 2 1 Racl Nde7 22 Bxc6 Kd6! 23 R.xa5 Nxc6 24 Ra6 Kd7 25 Rc3 Ne7 26 Ra7ch Ke8 27 Re3 Rd7 28 Ra5 f6 ! 29 b4 Kf7 30 a4 Reich 3 1 Kg2 Nc6 32 Rb5 Rc4 33 Rb3 Ra7 34 Rb6 R.xa4 and Black won quickly.

15 .. . (6 16 BeJ Nd4

Seirawan gave this move an ! mark. However, this evaluation is open to question, as we shall see.

In the game Khalifman-Seirawan, Reykjavik 199 1 , after 1 5 Bg5ch f6 16 Be3 (as in the game Lputian-Seirawan - which is our main line) Black played 16 . . . b6. There followed 17 Rfc l a5 18 Bg2. Now the correct defense is 1 8 . . . Ke6, though White would still be slightly better. In the actual game Seirawan erred with 1 8 . . . a4? and after 19 a3 Nd5 20 Bxc5ch! bxc5 2 1 Rxc5 Nd4 22 Rxd5 R.xd5 23 Bxd5 Rb8 24 e3 Ne2ch 25 Kfl Rxb2 26 Be4 White held a winning advantage.

In a later game Salov-Seirawan, Wijk aan Zee 1992, Black tried to improve by playing 17 . . . Nd4 (instead of 1 7 . . . a5 as in the Khalifman­Seirawan game). However, the game proceeded 18 Bxd4 R.xd4 19 a3! Na6 (or 19 . . . Nd5 20 e3 Rd2 21 e4 Nc7 22 b4 Kd6 23 Rd1 Rd4 24 Rxd4ch cxd4 25 Rd1 Nb5 26 a4 winning for White; also, after 19 . . . Nc6 20 b4! Kd6 21 e3! Rd2 22 Rd1 Rxd1ch 23 Rxd1ch Ke7 24 Rd7ch KJB 25 bxc5 bxc5 26 Rc7 White will win - Salov) 20 b4! Rd6 (not 20 . . . cxb4! 21 axb4 Nxb4 22 Rc7ch Kj8 23 Be6 winning - Salov) and now, instead of the actually played 2 1 Bf5, Salov gives 2 1 Bg2! as best, e.g. 2 l . . .Rb8 (or 2J. . .Rad8 22 Bb7) 22 Be4! cxb4 (22 . . . g6 23 Bd3! b5 24 bxc5 Rc6 25 Rab1 wins) 23 axb4 Nxb4 24 R.xa7ch Rd7 25 Rcc7 R.xc7 26 Rxc7ch KfS 27 Bxh7 wins for White. Even after 2 1 Bf5 (as in the actual game) Salov went on to win in 36 moves.

1 7 Bg2 Nbc2 18 Bxd4 Nxd4

19 eJ Ne6 Not 19 . . . Nxe2ch, as 20 Kh l Rd2 2 1 Bf3 Rxb2 22 Rfdl is winning for

White according to Lputian.

20 Bxb7 Rab8 21 Be4

CHAPTER IX 139

Now 2 l . . .Rxb2 is met by 22 Bxh7 as Black cannot trap the Bishop by . . . g6. Note that this shows up a drawback of 1 5 . . . f6, and hence a concrete reason for the finesseful 15 Bg5ch. Nonetheless, Lputian felt that 2 l . . .Rxb2 was Black's best try to hold the position.

21 ... 22 Bg2 23 h4 24 Rfb1 25 Bfl

Ng5 Rxb2 Ne6 Rb6

And White ground out a win from this slightly better position, Lputian-Seirawan, Manila Olympiade 1992.

IX B

( 1 d4 d5 2 c4 dxc4 3 NfJ c5 4 d5 e6 5 Nc3 exd5 6 Qxd5 Qxd5 7 Nxd5 Bd6 8 Nd2)

8 ... Nc6

This may well be Black's best choice.

9 Nxc4 Bb8 10 g3

In Aseev-Westerinen, Helsinki 1992, White tried 10 e4. The game continued IO . . . Nf6 I I Nxf6ch gxf6 12 Bh6 Be5 1 3 0-0-0 Bd4 14 Nd6ch Ke7 1 5 Nxc8ch Raxc8 16 f3 a6 17 g3 b5 1 8 Bh3 Rc7 19 Rhe l Rg8 20 Kbl Ne5 with a fine game for Black.

140 THE QUEEN'S GAMBIT ACCEPTED

In Razuvaev-Fedorov, Belgrade 1988, White played 10 Bf4 !? but after 10 . . . Bxf4 I I Nxf4 Nf6 play was even.

10 . . . Nge7 After IO . . . Nd4 the game Vladimirov-Westerinen, Cordoba 1990,

continued I I N4e3 Bd7 12 b4 ! Bc6 13 Bg2 Ne7 14 Bb2 with a small advantage to White.

I I Nxe7 Kxe7 12 Bg2 Nd4 13 Ne3 Bd6 14 f4 ReS IS Kf2 Kf8 16 Rd1 a5 17 Nd2 Ra7 18 Bfl bS

As in P. Cramling-Westerinen, Ostersund 1992. The players soon agreed to a draw.

CHAPTER X

White Plays 3 e4 (1 d4 d5 2 c4 dxc4)

3 e4

141

The most natural and classical response; White immediately creates a classical pawn center and anticipates rapid development. This line has been immensely popular in recent years, counting among its advocates both Karpov and Kasparov. Nonetheless, we show that current theory allows Black to secure an equal game with precise play.

We now examine: X A 3 . . . e5, X B 3 . . . c5, X C 3 . . . Nf6 and X D 3 . . . Nc6.

X A ( 1 d4 d5 2 c4 dxc4 3 e4)

3 . . . e5 (See next diagram)

142 THE QUEEN'S GAMBIT ACCEPTED

Striking back on the central dark squares. Now the capture 4 dxe5 leads to complete equality after 4 . . . Qxdlch 5 Kxdl Nc6 6 f4 f6 7 Bxc4 fxe5. In the game Labourdonnais-MacDonnel, Matchgarne 1834/35(!), 4 d5 led to an edge for Black after 4 . . .f5 5 Nc3 Nf6 6 Bxc4 Bc5 7 NO Qe7. Finally, the capture 4 Bxc4 is inferior; the game Kuzrninich-Sornov, Leningrad 1946, continued 4 ... Qxd4 5 Qb3 Qxe4ch 6 Be3 Qg6 7 NO Nd7 8 Nc3 c6 9 0-0?! (better 9 0-0-0 though Black would hold a slight advantage with 9 . . . Nh6l) 9 . . . Be7 1 0 Radl Ngf6! with the upper hand for Black.

4 NfJ Now Black has three choices: X A l 4 . . . Nf6, X A1 4 . . . Bb4ch and X

A3 4 . . . exd4.

X Al ( 1 d4 d5 2 c4 dxc4 3 e4 e5 4 NfJ)

4 . . . Nf6 (See next diagram)

1992.

CHAPTER X

An old move which allows White a big plus.

5 Bxc4 Nxe4

6 0-0 Nd6 7 Bb3 e4 8 Ne5 Be7 9 Nc3 Nd7 10 Nxe4 Nxe5 1 1 dxe5 Nxe4 12 QfJ Be6

13 Qxe4 Bxb3 14 axb3 e6 15 BeJ

1 43

White has a substantial advantage, Gutman-J. Adamski, Giessen

X A2 ( 1 d4 d5 2 c4 dxc4 3 e4 e5 4 NfJ)

4 ... Bb4ch (See next diagram)

144 THE QUEEN'S GAMBIT ACCEPTED

This line has been under pressure lately. Now White has two moves: X A2 1 5 Nc3 and X A22 5 Bd2.

X A21 ( 1 d4 d5 2 c4 dxc4 3 e4 e5 4 NO Bb4ch)

5 Nc3

This move has been giving Black some problems in recent games.

5 ... exd4 6 Nxd4

This seems to be the most promising course. Another way is 6 Qxd4 Qxd4 7 Nxd4 Nf6 8 f3 a6 9 Bxc4 b5 10 Be2 c5 I I Nb3 Be6 I2 Be3 Nbd7 I3 Kf2 ReS I4 Rhd I Nb6 15 Rae I Ke7 I6 g4 h6 I7 g5 Nc4 with about even play, Karpov-Z. Polgar, Roguebrune I993.

CHAPTER X 1 45

6 ... Ne7 The alternative is 6 . . . Qe7. However White has achieved excellent

chances against this move. For example, 6 . . . Qe7 7 Bxc4 Nf6 8 0-0 Bxc3 9 bxc3 0-0 IO Rei c5 I I e5! cxd4 I 2 exf6 Qc5 13 Qb3 Rd8 I4 Bxf7ch Kh8 1 5 cxd4 Qb6 16 Bg5 Qxb3 I7 Bxb3 Rm 1 8 Re7 and Black resigned in the game Gomez-Fernandez, Lisbon 1993. Another game Shirov-Hubner, Dortmund 1992, varied with the audacious pawn snatch 7 . . . Qxe4ch. There followed 8 Kfl Bxc3 9 bxc3 Be6 10 Qa4ch Nc6 1 1 Bg5 Bxc4 1 2 Qxc4 Nce7 1 3 Rei Qg6 14 h4 Kf8 1 5 Qxc7 f6 16 Rh3 Qd3ch and Black gave up.

7 Bxc4 Nbc6 8 8e3

Or 8 Ndb5 Qxdlch 9 Kxdl Bg4ch 10 f3 0-0-0ch 1 1 Ke2 Be6 1 2 Bxe6ch fxe6 1 3 Be3 a6 14 Na3 Nd4ch 15 Bxd4 Rxd4 16 Nc2 Bxc3 17 bxc3 Rc4 1 8 Kd3 Rc5 with equality, Alburt-Rachels, U.S. Championship 1 989.

8 ... 0-0 9 a3

Probably even better is 9 Nb5. The game Vyzhmanavin-Bagirov, Podolsk 1 992, continued 9 . . . Bd7 10 a3 Ba5 1 1 b4 Bb6 12 Bxb6 axb6 1 3 f4 Na7 14 Nxa7 Rxa7 15 0-0 with slight, but persistent pressure.

9 ... Bxc3ch 10 bxc3 Na5 1 1 Be2 Ng6

In Kamsky-Short, Linares 1994 (Match Game 1 ), Black tried 1 l . . .b6? which is far too slow and doesn't address the problem of the a5-Knight. Seirawan says "For fair chances to equalize it is necessary to play 1 1 . . . f5 ! 1 2 exf5 Nxf5 ! ".

1 2 0-0 Qe7 13 Qc2 Ne5 14 Rfel b6 15 NfS BxfS 16 exfS Rfe8 1 7 a4 Rad8 18 h3 Nec4 19 Bf4 Nd6 20 g3 Nab7 21 h4

With White on top, from Bareev-Ivanchuk, Dortmund 1992.

146 THE QUEEN'S GAMBIT ACCEPTED

X A22 ( 1 d4 dS 2 c4 dxc4 3 e4 eS 4 NfJ Bb4ch)

S Bdl

This causes Black fewer problems than 5 Nc3 (X A2 1 ).

S ... Bxdlch 6 Nbxdl

In Karpov-Ivanchuk, Monaco 1992, White recaptured with 6 Qxdl. There followed 6 . . . exd4 7 Qxd4 Nf6 8 Nc3 Be6 9 Qxd8ch Kxd8 10 Nd4 (or 10 Ne5 Nc6! l l Nxc4 Ke7 12 Ne3 Rhd8 13 /4 Kj8! 14 Be2 Nd4 with even chances, Salov-P. Nikolic, Reykjavik 1991 ) IO . . . Ke7 1 1 Nxe6 fxe6 12 Bxc4 Rd8 1 3 f3 Nc6 14 Rd l Rxd1ch 15 Kxdl Rd8ch 16 Kcl g5! 17 Rd1 M 1 8 Kd2 h5 and the game was eventually drawn.

6 ... 7 Bxc4 8 0-0

exd4 Nc6 Nf6!

An important innovation by Anand. The older 8 . . . Qf6 runs into difficulties. For example, the game M. Gurevich-Romanishin, Barcelona 1992, continued 8 . . . Qf6 9 b4! Nge7 (in Shirov-Z. Polgar, Brunn 1 99 1 , Black essayed 9 . . . a6. There followed 10 e5 Qg6 1 1 Nb3 Bg4 12 Bd3! Qh6 [on 1 2 . . . Qh5 13 Be4! Nge7 14 h3 Bxf3 15 Bxf3 Qe5 16 Bxc6 White is on top; equally so after 13 . . . Ne5 14 Qel ! Ne7 15 Nxe5 Qxe5 16 Bxb7 Qxe1 1 7 Rfxe l Rb8 1 8 Nc5 - variations by Shirov] 1 3 h 3 Bxj3 1 4 Qxj3 Nge7 1 5 Qe4 Rd8 16 a3! with a clear superiority for White) 10 b5 Nd8 (or 1 0 . . . Be5 1 1 Nxe5 Qxe5 12 /4 Qd6 13 Nb3 with an edge) 1 1 e5 Qg6 1 2 Nd4 Ne6 1 3

CHAPTER X 1 47

Nxe6 Bxe6 14 Rc1 Rd8 1 5 Bxe6 Qxe6 16 Qc2 0-0 17 NO c6 1 8 Ng5 Qh6 19 f4 and White held the advantage and went on to win.

9 e5 Nd5 In view of Karpov's improvement (given in the next note), Black

should opt for 9 . . . Ng4. For example, 9 . . . Ng4 10 h3 Nh6 1 1 Nb3 Nf5 1 2 Qd3 (after 12 Bb5 Qd5 13 Bxc6ch [or 1 3 Nbxd4 Nxd4 14 Nxd4 Bd7! 1 5 Bxc6 Bxc6 16 Nxc6 Qxc6 with even chances - Flear] 13 . . . Qxc6 14 Rcl Qb6 15 Nbxd4 0-0 16 Nxf5 Bxf5 17 Qd4 as in C. Flear-Slipak, Palma de Mallorca 199 1 ) 1 2 . . . 0-0 1 3 Rfe1 Nh4! 14 Rad 1 Bf5 15 Qe2 Nxflch 16 QxO Bc2 17 Rd2 Bxb3 1 8 Qxb3 with even play, G . Flear-Slipak, Palma de Mallorca 1 99 1 .

10 Nb3 A recent improvement was seen in the game Karpov-Hubner,

Germany 1 994: 10 Qb3 Na5 1 1 Qa4ch Nc6 12 Qb3 Na5 1 3 Qb5ch! c6 14 Qc5 b6 1 5 Qxd4 with advantage to White.

game.

10 ... 0-0

1 1 Nbxd4 Nxd4

1 2 Qxd4 Nb6 13 Bb3 Be6

14 Qc5 Re8 15 Rad1 Qe7 16 Qe3 Bxb3

1 7 Qxb3 Rad8 As in Karpov-Anand, Linares 1992. Black has a fully satisfactory

X A3 ( 1 d4 d5 2 c4 dxc4 3 e4 e5 4 NO)

4 ... exd4 (See next diagram)

148 THE QUEEN'S GAMBIT ACCEPTED

Currently this move is doing well for Black.

S Bxc4 After 5 Qxd4 Qxd4 6 Nxd4 Bc5 7 Nb5 Na6 8 Bf4 Be6! Black stands

better, Nilsson-Ekstrom, Stockholm 1949.

S ... Bb4ch 6 Nbd2 Nc6 7 0-0 Nf6 8 eS NdS

9 Nb3 Nb6 10 BgS

Or 1 0 Bb5 0-0 1 1 Bxc6 bxc6 1 2 Nbxd4 Qd5 1 3 Qc2 c5 14 Nb5 Qc6 15 a3 Qxb5 16 axb4 cxb4 17 Ng5 g6 18 Qxc7 Nd5 1 9 Qa5 Qxa5 20 Rxa5 Be6 2 1 Ne4 and here the game Speelman-Hubner, Munich 1992, was agreed drawn.

10 ... Be7 1 1 Bxe7 Nxe7

After l l . . .Qxe7 1 2 Bb5 Bd7 1 3 Bxc6 Bxc6 14 Nfxd4 Bd5 1 5 Qg4 0-0 16 f4 g6 1 7 Rfel f5 18 Qg3 c5 19 Nb5 Nc4 20 Radl Be6 2 1 Qc3 b6 22 Nd2 Nxd2 23 Rxd2 Rad8 24 Nd6 White was on top, Michaelsen-Putzbach, Hamburg 199 1 .

12 Bd3 Or 1 2 Be2 Ng6 1 3 Rei 0-0 14 Qxd4 Qe7 15 Qe4 c6 16 Bfl Be6 1 7

Nbd4 Bd5 1 8 Qg4 Qb4 1 9 Qh5 Bxf3 20 Nxf3 Nd5 with equality, Bareev­Rublevsky, Oviedo 1992.

1 2 ... Bg4 13 Nbxd4 NedS!

CHAPTER X

Hubner's 1 3 . . . Ng6 14 h3 Bxf3 15 Nxf3 0-0 is favorable to White.

14 h3 Bh5 15 Qb3 0-0 16 Be4

149

As in the game Timoschenko-Matulovic, Vmjacka Banja 1990. The game was agreed drawn here.

150 THE QUEEN'S GAMBIT ACCEPTED

X B ( 1 d4 d5 2 c4 dxc4 3 e4)

3 ••• c5

Again striking back on the central dark squares (as did 3 . . . e5 in X A). See Chapter IX for similar play (after 1 d4 d5 2 c4 dxc4 3 Nj3 c5).

After 3. . .c5 we analyze: X B l 4 NO and X B2 4 d5.

X B1 ( 1 d4 d5 2 c4 dxc4 3 e4 c5)

4 NfJ

A sound developing move while protecting d4.

CHAPTER X

4 ... cxd4 5 Qxd4

151

In Kuzmin-Lputian, Palma de Mallorca 19S9, White played 5 Bxc4. There followed 5 . . . Nc6 6 0-0 e6 7 e5 (7 Nbd2 Nf6 is equal) 7 . . . Nge7 S Bg5 h6 9 Bxe7 Bxe7 10 Nbd2 0-0 I I Qe2 b6 12 Qe4 Bb7 1 3 Bd3 g6 14 Qg4 Kg7 1 5 a3 h5 16 g3 h4 with complex play and mutual chances.

In Shirov-Kramnik, Oakham 1992, White tried 7 a3 (instead of 7 e5 seen above). Play continued 7 . . . Bd6 S b4 Nf6 9 b5 Na5 10 e5 Nxc4 I I exf6 gxf6 1 2 Qxd4 Ne5 1 3 Qe4 Bd7 14 Nc3 ReS with a fine game for Black.

S ... Qxd4

6 Nxd4 Bd7 Best. In Speelman-Seirawan, Monaco 1 992, 6 . . . Nf6 was played, but

after 7 e5 Nfd7 S e6 Ne5 9 Na3 Nd3ch 10 Ke2 Bxe6 1 1 Nxe6 Nxcl ch 1 2 Rxcl fxe6 1 3 Nb5 Na6 14 g3 g6 15 Bg2 White held the advantage.

7 Nc3 Or 7 Bxc4 Nc6 8 Be3 (or 8 Nxc6 Bxc6 9 Nc3 e6 10 Bb5 ReB J J

Bxc6ch Rxc6 12 Be3 a6 13 a3 Nf6 14 Ke2 Bd6 15/4 e5 16/5 Ng4 1 7 Bd2 Bc5 with about even chances, Nalbandian-Lputian, Protvino Zonal 1993) S . . . Nf6 9 f3 e6 10 Nb3 ReS! I I Be2 Nb4 12 Na3 a6 1 3 Kf2 b5 14 Rhdl Be7 1 5 g4 e5 with equality, Chemin-Seirawan, Wijk aan Zee 199 1 .

7 • • • Nc6 8 Nxc6 Bxc6 9 Bxc4 e6 10 Ke2 a6 1 1 Be3 Nf6 1 2 13 Nd7 13 a4 Bd6 14 NbS ubS 15 axb5 Ke7 16 bxc6 bxc6 1 7 f4 DeS

With equal play, Korchnoi-Seirawan, Monaco 1 992.

X B2 ( 1 d4 d5 2 c4 dxc4 3 e4 c5)

4 d5

1 52 THE QUEEN'S GAMBIT ACCEPTED

Seizing more central space. Now Black has two replies: X B2 1 4 . . . Nf6 and X B22 4 . . . e6.

X 821

( 1 d4 d5 2 c4 dxc4 3 e4 c5 4 d5)

4 • • • Nr6

Developing while attacking White's newly established central pawn chain.

5 Nc3 b5 Reinforcing c4 while threatening the undermining b4.

6 or4 Or 6 Nxb5?! Qa5ch 7 Nc3 Nxe4 8 QD Nd6 9 Bf4 Nd7 with fully

satisfactory play for Black, Furman-Birkan, USSR 1968.

CHAPTER X 153

Also, after 6 e5 b4 7 exf6 bxc3 8 bxc3 Nd7 9 Qa4 exf6 I 0 Bf4 Qb6 I I Bxc4 Bd6 12 Ne2 0-0 13 0-0 and now Shirov's 1 3 . . . Qc7 gives even chances. In the game Shirov-011, Tilburg 1992, White tried 9 fxe7?! but after 9 . . . Bxe7 10 Bxc4 Nb6 I I Bb5ch the players agreed to a draw.

6 ... QaS Best. After 6 . . . a6 7 e5 b4 8 exf6 bxc3 9 bxc3 Nd7 (as in Belyavsky­

Ehlvest, Belfort 1988) 10 Qa4! White has the upper hand.

7 eS In Belyavsky-Kramnik, Linares 1993, 7 a4 was tried. There followed

7 . . . Nxe4 8 Nge2 Nd6 9 axb5 Qb6 10 Bxd6 (or 10 Ng3 Nd7 1 1 Bxd6 Qxd6 12 Nge4 Qe5 13 Be2 g6 14 0-0 Bg7 15 d6 as in Azmaiparashvili-Granda, Groningen 1 993; now correct is 15 . . . 0-0 16 Bj3 Bb7 1 7 Nxf6ch Nxf6 18 Bxb7 Rad8 19 dxe7 Qxe7 20 Qj3 and now GM Flear's 20 .. . Rd3 is fine for Black) IO . . . exd6 I I Ng3 Be7 12 Bxc4 0-0 13 0-0 Bf6 with a good game for Black.

In Shirov-Kramnik, Linares 1993, White tried 7 Bd2. There followed 7 . . . b4 8 e5 bxc3 9 Bxc3 Qa6 10 exf6 exf6 I I b3 Be7 12 Bxc4 Qd6 1 3 Ne2 0-0 14 0-0 f5 1 5 Rei Nd7 16 Ng3 g6 17 Rxe7 ! ! ? Qxe7 18 d6 Qh4 19 Qd5 Rb8 20 Nxf5 gxf5 2 1 Rei Qg4 22 D Qg6 23 Re7 Bb7 24 Qd3 Nb6 25 Bal Nxc4 26 Qc3 f6 27 Qxc4ch and now instead of the actually played 27 . . . Kh8?, Black should have chosen 27 . . . Rf7 and he would be winning.

7 ••. Ne4 8 Nge2 Na6 9 f3 Nb4! !

A big improvement over 9 . . . Nxc3 played in Belyavsky-Anand, Linares 1993. In that game play continued I 0 Nxc3 Bf5 1 1 g4 Bg6 1 2 a4 Nb4 1 3 Kf2 with White on top.

10 f:xe4 N dJch 1 1 Kd2 g6! !

Another brilliant resource. We have been following Gelfand-Anand, Linares 1993 (played after the previously mentioned Belyavsky-Anand game). The game continued 1 2 b3 Bg7 13 bxc4 Nxf4 14 Nxf4 Bxe5 1 5 Nfe2 b4 16 Qa4ch Qxa4 1 7 Nxa4 Bxal and Black won the ending.

X B22 ( 1 d4 dS 2 c4 d:xc4 3 e4 cS 4 dS)

4 .. . e6

1 54 THE QUEEN'S GAMBIT ACCEPTED

This doesn't seem to be as good as X B2 1 (4 . . . Nf6).

5 Bxc4 Nf6 After S . . . exdS 6 BxdS Qc7 7 Ne2 Bd6 8 Nbc3 Nf6 9 0-0! Bxh2ch 10

Kh 1 BeS (JO . . . Ng4 JJ /4) I I f4 Bxc3 12 Nxc3 0-0 (12 . . . Bg4 13 Qa4ch Nbd7 14 Nb5 Qb6 15 Bxb 7/) 1 3 fS ! White is much better, Gaisinuk­Barajev, 1988.

6 Nc3 exd5 7 Nxd5

Better than 7 exdS a6 8 a4 and now the game Kakageldiev-Lima, Biel Interzonal 1993, continued 8 . . . Bd6 9 Qe2ch Be7 10 Qc2 0-0 1 1 Bf4 Bd6 1 2 Bg3 ReSch 13 Nge2 Bxg3 14 hxg3 Nbd7 I S Rd1 NeS 16 Ba2 c4 1 7 0-0 bS ! with fine chances for Black.

7 ... Nxd5

8 Bxd5 Be7

9 NO In the game Dejkalo-Karolvi, Prague 1988, White tried 9 Qb3 . There

followed 9 . . . 0-0 10 Nf3 Qb6 1 1 NeS Be6 1 2 Bxe6 fxe6 1 3 Qxb6 axb6 14 0-0 Ra4 I S Rei Bf6 with good chances for Black.

Or 9 Ne2 0-0 10 0-0 Nd7 I I Nc3 Nb6 12 Bf4 BgS 13 Bd6 Be7 14 Bf4 BgS 1 S BxgS QxgS 16 f4 Qe7 17 eS Be6 with even chances, Razuvaev­Psakhis, Rostov 1993.

Finally, the sharp 9 QhS is met by 9 . . . 0-0 10 Nf3 Nd7 I I NgS BxgS 1 2 BxgS Nf6 13 Bxf6 Qxf6 14 0-0 Rb8! as in Adianto-Z. Polgar, San Francisco, 199 1 .

9 . . . 0-0 10 0-0 Qb6!

CHAPTER X lSS

From the game Azmaiparashvili-Schulte-Bartold, Dortmund 1992. The position is equal.

1 56 THE QUEEN'S GAMBIT ACCEPTED

X C ( 1 d4 d5 2 c4 dxc4 3 e4)

3 ... Nf6

Hitting back at White's newly exposed e-pawn. Now, after 4 e5 Nd5 the d5 square will figure in Black's counterplay. If White plays 4 Nc3 then Black should play 4 . . . e5 ! . For example, 4 . . . e5! 5 NO exd4 6 Qxd4 Bd6! 7 Bg5 (poor is 7 e5? because of 7 . . . Nc6 8 Qe3 0-0 9 exf6 ReB 10 Ne4 B./5 1 1 Nfd2 Bxe4 1 2 Nxe4 Bb4ch) 7 . . . Nc6 8 Qxc4 h6 9 Bh4 Be6 I 0 Qa4 0-0 1 1 Be2 Ne5 12 Nd4 Bd7 1 3 Qc2 Ng6 14 Bg3 Qe7 1 5 0-0 Bxg3 16 hxg3 Rfe8 17 Rfe 1 with equal chances, Goldin-Balashov, Irkutsk 1986.

4 e5 Nd5

5 Bxc4 Nb6 Now White has two Bishop retreats: X C l 6 Bd3 and X C2 6 Bb3.

X CI ( 1 d4 d5 2 c4 dxc4 3 e4 Nf6 4 e5 Nd5 5 Bxc4 Nb6)

6 Bd3 (See next diagram)

CHAPTER X 157

The "traditional" diagonal for the KB, which has the advantage of preventing . . . Bf5 for Black. However, the exposed position of the Bd3 is utilized by Black to obtain equal chances.

6 ... Nc6 7 Ne2

The game Timman-Salov, Sanghi Nagar (4th matchgame) 1994, continued 7 Be3 Nb4 8 Be4 f5 9 exf6 exf6 10 Nc3 f5 I I Bbl N4d5 1 2 Nf3 Bd6 1 3 Bg5 Qd7 I4 Qe2ch Qe6 I 5 Ne5 0-0 I6 0-0 Nxc3 I7 bxc3 Bxe5 I8 dxe5 Qc6 with a very satisfactory position for White.

7 ... Nb4! An innovation by GM Short. After 7 . . . Bg4 8 f3 Be6 9 Nbc3 Qd7 IO

Ne4 Bd5 I I Nc5 Qc8 I2 Be3 e6 13 Rei ! Nb4 14 a3 ! Nxd3ch I 5 Qxd3 Nd7 I6 Qe4! Qd8 I 7 0-0 Ba7 I 8 f4 White has the edge, Meulders-van der Sterren, Lyon Zonal I990.

8 Be4 f5! 9 e:d6

On 9 Bf3 Black plays 9 . . . f4 ! since 10 Bxf4 Qxd4 I I Qxd4 Nc2ch I 2 Kd2 Nxd4 1 3 Be4 Bf5 i s clearly i n Black's favor.

9 . . . e:d6 10 Nc3 f5 1 1 Bf3 N4d5 12 Bd2 Be6 13 Nge2 Qd7 14 0-0 0-0-0 15 Ret Rg8! 16 Bg5 ReS

158 THE QUEEN'S GAMBIT ACCEPTED

1 7 Nf4 N:d4 18 B:d4 g5 19 Be5 Bg7

As in Karpov-Short, Candidates Matchgame, Linares 1992. Black stands better.

X C2 ( 1 d4 d5 2 c4 dxc4 3 e4 Nf6 4 e5 Nd5 5 Bxc4 Nb6)

6 Bb3

This is White's best choice.

6 ... Nc6 Also possible is 6 . . . c5 !7. For example, the game Karpov-Short,

Linares 1 992, continued 7 dxc5 Qxdl ch 8 Kxdl N6d7 9 e6 fxe6 10 Bxe6 Na6 I I c6 bxc6 1 2 Be3 Nc7 13 Bb3 Nd5 14 Nf3 Nxe3 1 5 fxe3 g6 16 Nbd2 Nc5 and after a long struggle the game wound up in a draw.

7 Be3 In Epishin-Kharlov, Moscow 1992, White played 7 Ne2. There

followed 7 . . . Bf5 8 Nbc3 e6 9 a3 Be7 10 0-0 0-0 I I Ba2 Qd7 1 2 Ba2 Rfd8! 1 3 h3 (or 13 Qcl Bj8 14 Rd1 Ne7.1 15 Ng3 Bg6 16 Nce4 Nbd5 1 7 Nc5 Qc8 18 Bb1 N./5 19 Bg5 ReB with approximately equal play, V. Ruban­Yakovich, St. Petersburg Zonal 1993) 1 3 . . . h6 14 Qc l Bts 15 Rd l and now, instead of the actually played 1 5 . . . Na5?, correct was 1 5 . . . Ne7! with sharp play and mutual chances.

7 ... Bf5

CHAPTER X

8 Ne2 e6 9 Nbc3 Nb4

159

Interesting is 9 . . . NaS which is best answered by 10 Ba4ch! . Karpov­Spee1man, Brussels 1988, then continued IO . . . c6 (JO . . . Nxa4 1 1 Qxa4ch Nc6 12 0-0 a6 13 Ng3 Bd3 /4 Rfe 1 slightly favors White) 1 1 Bc2 Bg6 (maybe better is J J . . . Bxc2!? 12 Qxc2 Nac4 /3 0-0 Nd5 as in Kolev-Maljanovic, Vmjacka Banja 1 990) 1 2 0-0 Nac4 13 Qc l !? Be7 14 Bxg6 hxg6 I S Ne4 with a small advantage for White.

10 0-0 Be7

1 1 Nf4 Worth consideration is 1 1 Ng3 !?.

11 .. . Qd7 l2 Rcl

Other untested possibilities are 12 g4!? or 12 a3!?.

12 ... c6 13 QIJ N4d5 14 Nh5 0-0 15 Ne4 Rae8 16 Nhg3 Bg6 1 7 Qd1 Rd8 18 a3 Qd7

As in Agrest-Zvjaginsev, Orel 1992. Play is about even.

160 THE QUEEN'S GAMBIT ACCEPTED

X D ( 1 d4 d5 2 c4 dxc4 3 e4)

3 ••. Nc6

Attacking d4 and preparing . . . e5. This line is double�ged, leading to complex strategy, and tactics suitable to those who want decisive results.

The obvious reply to 3. . . Nc6, 4 d5, is nothing for Black to worry about. For example, the game Romanishin-Magem, Barcelona 1 992, continued 4 d5 Ne5 5 Nc3 Nf6 6 Qd4 Ng6 7 NO e5 8 Qxc4 a6 9 h4 Bg4 10 Be2 Bd6 1 1 g3 h6 12 Be3 0-0 1 3 Rd1 Qb8 14 Nd2 Bd7 1 5 Nb3 c6 16 Qd3 cxd5 17 exd5 and the players agreed to a draw. In the game Romanishin­Sadler, Altensteig 1992, White varied from 7 NO with 7 Qxc4, but after 7 . . . e5 8 Bg5 Bd6 9 Qa4ch Kf8! 10 NO h6 1 1 Bd2 a6 1 2 Be2 Bd7 1 3 Qc2 Qe7 14 0-0 Nf4 Black was fine (the game was drawn).

So, after 3 . . . Nc6 we analyze: X 01 4 Be3 and X 02 4 NO.

X D1 ( 1 d4 d5 2 c4 dxc4 3 e4 Nc6)

4 Be3 (See next diagram)

CHAPTER X 161

Some theorists prefer our next line (X 02) with 4 Nf3. Current theory indicates that X 02 is creating problems for Black.

4 ... Nf6 S Nc3 eS

Another possibility is 5 . . . Ng4. In Vyzmanavin-Baburin, Gorky 1 989, there followed 6 Bxc4 e5 7 Qb3 Qd7 8 0-0-0 exd4 9 Nf3 Bc5 10 Qb5 Bd6 1 1 Bxd4 0-0 1 2 h3 Nxd4 1 3 Qxd7 Bxd7 14 Rxd4 Nxf2 15 Rfl Be6 16 Bxe6 fxe6 1 7 Rxf2 Bc5 18 Rfd2 Bxd4 19 Nxd4 with White on top.

6 dS Ne7

7 Bxc4 a6 In Timman-Lautier, Candiates Match at Wijk aan Zee 1994, Game #

3, Black played 7 . . . Ng6 but after 8 Bb5ch Nd7 (after 8 . . . Bd7 the exchange of light-squared Bishops favors White. For example the game Shirov-P. Nikolic, Groningen 1993, continued 9 Qb3 b6 10 j3 Bd6 1 1 Nge2 0-0 12 0-0 a6 13 Bxd7 Qxd7 14 Ng3 Ne 7 15 a4 with a slight positional pull for White) 9 Qd2 a6 l O Bd3 Bd6 I I Nge2 Qe7 12 f3 b5 l 3 0-0 0-0 14 Rac1 Nc5 1 5 Bc2 Bd7 16 Ng3 Rfc8 17 Nf5! and White had a clear positional advantage. However, with 7 . . . a6! Black improves by preventing the Bishop check on b5. This improvement comes in a game played a few weeks later by . . . Lautier! .

8 0 Ng6 9 Nge2 Bd6 10 Qd2 Bd7 1 1 Ng3 hS 12 0-0-0 h4 13 Nf5 BxfS

162 THE QUEEN'S GAMBIT ACCEPTED

14 exf5 Ne7 We have been following Belyavsky-Lautier, Linares 1994. The game

continued 1 5 g4 hxg3 16 hxg3 Rxh1 17 Rxh1 Nxf5 18 Bg5 ! Qd7 19 Bfl Nd4 (19 . . . 0-0-0 20 Bxf6 gxf6 21 Bh3 Rh8 22 Qdl Kd8 23 Bg4) 20 Rh8ch? ­the right way was 20 f4 NO 2 1 Qe2 Nxg5 22 fxg5 Ng8 23 Rh8 KfS 24 Bh3 Qe7 25 Ne4 with a winning position. The actual game ended in a draw after further vicissitudes.

X D2 ( 1 d4 d5 2 c4 dxc4 3 e4 Nc6)

4 NO

This is the most natural response.

4 .. . Bg4 Continuing to pressure d4 indirectly by the pin. Now there are three

choices: X D2 1 5 Be3, X D22 5 Bxc4 and X 023 5 d5.

X 021 ( 1 d4 d5 2 c4 dxc4 3 e4 Nc6 4 NO Bg4)

5 Be3 (See next diagram)

CHAPTER X 163

This line is giving Black problems currently. But we expect improvements will be found.

5 ... e6 6 Nc3

In Kouatly-Piket, Cannes 1992, White played 6 Bxc4. After 6 . . . Bxf3 7 gxO Qf6 8 d5 exd5 9 exd5 0-0-0 10 Nd2 Ne5 I I Bb3 Bb4 12 f4 Bxd2ch 1 3 Kxd2 Black was clearly on top.

6 .. . BxfJ Qf6

A strong novelty.

7 gxfJ 8 Qa4!

8 ... Bb4 9 0-0-0 Nge7 10 Rg1 0-0 1 1 Rg3 h6 12 e5 Qf5 13 Bh6

White has the advantage, M. Gomez-Magem, Spain 1993.

X D22 ( 1 d4 d5 2 c4 dxc4 3 e4 Nc6 4 NIJ Bg4)

5 Bxc4 (See next diagram)

164 THE QUEEN'S GAMBIT ACCEPTED

Developing and threatening 6 Bx17ch Kx17 7 Ng5ch and 8 Qxg4.

s . . . e6

6 BbS After 6 d5 the game Speelman-Sadler, Hastings 1992/93, continued

6 . . . exd5 7 Bxd5 Nf6! (not 7 . . . Qd7 8 0-0 0-0-0 9 Qb3 Nf6 10 Ne5! winning on the spot, from Gartner-Mainka, Zurich 1 988) 8 0-0 Be7 9 Bxc6ch bxc6 10 Qc2 0-0 1 1 Ne5 Bd6! 1 2 Bf4 ReS 13 Nxc6 Qd7 14 e5 Nh5 1 5 Be3 Bxe5 16 Nxe5 Rxe5 17 Nd2 Nf6 18 Nc4 Rd5 19 D Bf5 with equal play.

6 ... Ne7 7 Nbdl a6!

An improvement over 7 . . . Qd6 8 a3 ! 0-0-0 (or 8 . . . a6 9 Be2 0-0-0 1 0 Nc4 Qdl l l Nce5 Nxe5 1 2 Nxe5 Bxe2 1 3 Qxe2 Qe8 with a small plus for White, Bareev-Speelman, Hastings 1990/9 1) 9 Nc4 Qxd4 10 Nxd4 Bxd1 1 1 Nxc6 Nxc6 1 2 Bg5 f6 13 Bxc6 bxc6 14 Bxf6! gxf6 15 Rxd1 Rxd 1ch 16 Kxd I Bc5 1 7 Ke2 Rg8 with a small but clear positional advantage for White, Salov-Magem, Madrid 1992.

8 Bel Ng6 Focusing on the dark squares at e5,f4 and h4.

9 h3 BxfJ 10 NxfJ Bb4ch!

Accentuating the dark-squared weakness by trading the same-colored Bishops.

1 1 Bdl 12 Qxdl 13 Nxh4 14 Qe3

Bxdlch Nh4 Qxh4 0-0-0

CHAPTER X

15 Rd1 Qf6 With excellent play for Black, Begovac-lbragimov, Berne 1992.

X D23 ( 1 d4 d5 2 c4 dxc4 3 e4 Nc6 4 NO Bg4)

5 d5

165

White decides the best "defense" of the d-pawn is to advance, but Black has decent resources here.

5 .. . Ne5 6 Bf4

After the sharp 6 Qd4 Black can play 6 . . . Nxflch 7 gxfl Bxfl. Now if 8 Bxc4 ( to answer 8 . . . Bxhl? with 9 Bb5ch c6 10 dxc6! Qxd4 1 1 cxb 7ch winning) then 8 . . . e5! is fine. For example, 8 . . . e5 ! 9 Qc3 Qf6 10 Rg i Bxe4 I I Bg5 Qd6 I 2 Nd2 Bf5 13 Bb5ch Bd7 14 Nc4 Qb4 I 5 Bxd7ch Kxd7 I6 Nxe5ch Ke8 and Black wins a pawn, Rajkovic-Z. Djukic, Cetinje I 993 .

6 ... Ng6 7 Bel Nf6!?

Not the only move; after 7 . . . e5 8 Bxc4 Black can try 8 . . . Bd6!? and if 9 Bb5ch then 9 . . . Kf8!? (avoiding the trade of light-squared Bishops by 9 . . . Bd7?!) with unclear chances. After 7 . . . e5 8 Nxc4 Black should avoid 8 . . . a6 9 h3 Bd7 IO h4 Bd6 I I h5 N6e7 12 h6 with a pull for White, Altennan-Nikolic, Zagreb 1993.

8 NcJ e5

1 66 THE QUEEN'S GAMBIT ACCEPTED

After 8 . . . e6 the game M. Gurevich-Ivanchuk, Reggio Emilia I 99I/92, continued 9 Qa4ch Qd7 10 Qxd7ch Kxd7 I I Bxc4 exd5 12 exd5

Bxf3 I3 gxf3 a6 14 Ne4 with White slightly better.

9 Bxc4 a6 10 Be2 Bd6 1 1 Nd2 Bxe2 l2 Qxe2 0-0 13 0-0 Qd7 14 fJ Rfc8 15 Rfd1 Ne8

White is slightly better - though play is complex (from the game Azmaiparashvili-Piket, Wijk aan lee 1993). Black eventually won in a complicated ending after 16 Nb3 h6 17 Rac1 Kh7 18 Rc2 Rcb8 19 Rdc1 b5 20 Na5 b4 2 1 Nb1 f5 22 Nd2 Rb5 23 Nac4 fxe4 24 Nxe4 Rxd5 25 Rd2 Be7 26 Rcd1 Rxd2 27 Qxd2 Qxd2 28 Rxd2 Nf6.

CHAPTER XI Unusual White 3rd Move Alternatives

(1 d4 dS 2 c4 dxc4)

167

From this fundamental position we now examine two rare 3rd move choices for White: XI A 3 Nc3 and XI B 3 e3.

XI A ( 1 d4 dS 2 c4 dxc4)

3 Nc3

168 THE QUEEN'S GAMBIT ACCEPTED

This is a natural developing move, but unlike 3 Nf3 it doesn't prevent

Black from striking back with 3 . . . e5.

3 . . . eS! 4 Nf3

Poor is 4 dxe5. For example, 4 . . . Qxd lch 5 Kxd 1 (or 5 Nxdl Nc6 6 e4 Nxe5 7 B/4 Bd6 8 Bxe5 Bxe5 9 Bxc4 with equality - Alekhine) 5 . . . Be6 6 f4 f6 ! 7 Nf3 Nc6 and Black's position is preferable, Gurgenidze-Suetin, USSR 1960.

The space-gaining 4 d5 is harmless. For example, the game Terterians-Savon, Podolsk 1991 , continued 4 . . . Nf6 5 e4 c6 6 Bxc4 Bb4 7 Qb3 Qe7 8 Bg5 Bxc3 9 bxc3 Nbd7 10 d6 Qxd6 1 1 Bxf7ch Kffl 1 2 Rdl Qe7 1 3 Bc4 Nc5 with the upper hand for Black.

4 • . •

5 Qxd4 6 Nxd4 7 e4 8 Be3

exd4 Qxd4 Nf6 Bc5

In Zilbennan-Bagirov, USSR 1985, White tried 8 Ndb5 but after 8 . . . Na6 9 Bf4 Be6! 10 Bxc7 0-0 1 1 Bd6 Rfd8 1 2 Bxc5 Nxc5 1 3 Nc7 Rac8 14 Nxe6 Nxe6 1 5 f3 Nd4! 16 0-0-0 b5 Black was somewhat better.

8 ... Ng4

9 Nd5 Nxe3

10 fxe3 Na6 1 1 Bxc4 c6 12 Bxa6 cxd5 13 Bb5ch Ke7

14 exd5 Bxd4

15 exd4 And a draw was agreed in the game Michalchisin-Gulko, USSR

1985.

XI B ( 1 d4 d5 2 c4 dxc4)

3 e3 (See next diagram)

CHAPTER XI 169

The defect of this continuation is the same as 3 Nc3 - it doesn't prevent an equalizing 3 . . . e5.

3 .. . e5 Also of interest is 3 . . . c5 !?. For example, the game Gabriel-Sermek,

Vidmar Memorial 1993, continued 4 Bxc4 cxd4 5 exd4 Qc7 6 Be2 e6 7 Nf3 Nf6 8 0-0 Be7 9 Nc3 0-0 10 Bg5 Nc6 1 1 Rc1 Rd8 12 a3 Qa5 1 3 b4 Qf5 14 Bd3 Qg4 1 5 h3 Qh5 16 Ne2 Nd5 17 Ng3 Qxg5?! ! 18 Rc2 Nxd4 19 Bxh7ch Kxh7 20 Qxd4 Ne3 2 1 Qxd8 Bxd8 22 fxe3 f5 and though White has a material advantage, Black managed to draw.

4 Nc3 A slightly different move order was played in the game Gelfand­

Adams, Candidates Match, Wijk aan Zee 1994, Game 2. That game continued 4 Bxc4 exd4 5 exd4 Nf6 6 Nf3 Be7 7 0-0 0-0 8 h3 Nc6 9 Nc3 Na5 10 Bd3 Be6 1 1 Re 1 and now, instead of 1 1 . . .Nc6?!, correct was 1 l . . .c5! (which, by the way, transposes into a position known from the Petroff Defense!).

4 ... exd4 After 4 . . . Nc6 5 Nf3 exd4 6 exd4 Bg4 7 Bxc4 Qe7ch 8 Be3 0-0-0 9

0-0 Nf6 10 Rei White held a slight edge, Browne-Shirazi, U.S. Championship 1984.

5 exd4 Nf6 6 Bxc4

In the game I. Sokolov-Piket, Corfu 199 1 , White tried 6 Nc3 and after 6 . . . Be7 7 Nf3 0-0 8 h3 Nbd7 (8 . . . Nc6 would transpose into a Petroff) 9 0-0 Nb6 10 Bb3 c6 1 1 Ne5 Nfd5 12 Ne4 Be6 1 3 a3 Qc7 14 Rei Rad8 1 5 Qf3 Qc8 16 Bc2 f6 17 Qh5 ! with White on top.

6 ... Be7

1 70 THE QUEEN'S GAMBIT ACCEPTED

7 NfJ 0-0 8 0-0

On 8 h3 best is 8 . . . Bf5 with even chances.

8 . . . Nc6 The most accurate. After 8 . . . Nbd7 the game Vaganian-Hubner,

Tilburg 1 983, continued 9 Rei Nb6 10 Bb3 c6 1 1 Bg5 Bg4 1 2 Qd3 Bxf3 1 3 Qxf3 Nfd5 14 Bxe7 Nxe7 15 Re5 with White slightly better.

9 Rei Bg4 10 BgS NdS I I Bxe7 Ncxe7 12 h3 BhS 13 g4 Bg6 14 NeS c6 IS QfJ Kh8 16 h4 f6

As in Hubner-Nikolic, Wijk aan Zee 1984. Play is equal.