Chess Explained: the Main-Line Slav

Chess Explained: The Main-Line Slav

David Vigorito

The right of David Vigorito to be identified as the author of this work has been asserted in accor dance with the Copyright, Designs and Patents Act 1988.

All rights reserved. This book is sold subject to the condition that it shall not, by way of trade or otherwise, be lent, re-sold, hired out or otherwise circulated in any fonn of binding or cover other than that in which it is published and without a similar condition including this condition being imposed on the subsequent purchaser.

ISBN-13: 978-1-906454-05-0 ISBN-IO: 1-906454-05-1

DISTRIBUTION: Worldwide (except USA): Central Books Ltd, 99 Wallis Rd, London E9 5LN, England. Tel +44 (0)20 8986 4854 Fax +44 (0)20 8533 5821. E-mail: [email protected]

Gambit Publications Ltd, 99 Wallis Rd, London E9 5LN, England. E-mail: [email protected] Website (regularly updated): www.gambitbooks.com

Edited by Graham Burgess Typeset by Petra Nunn Cover image by Wolff Morrow Printed by the MPG Books Group in the UK.

10 9 8 7 6 5 4 3 2 1

Gambit Publications Ltd Managing Director: Murray Chandler GM Chess Director: Dr John Nunn GM Editorial Director: Graham Burgess FM German Editor: Petra Nunn WFM Webmaster: Dr Helen Milligan WFM Contents

Symbols 4 Dedication 4 Acknowledgements 4 Bibliography 4 Introduction 5

1 Dutch Variation: White Plays 9 'iWe2 8 2 Dutch Variation: White Plays 9 lDh4 29 3 Central Variation: Black Plays 6 ... lDbd7 7 lDxc4 "iIIc7 42 4 Central Variation: Black Plays 6 ... lDbd7 7 lDxc4lDb6 54 5 Central Variation: Black Plays 6 ... e6 66 6 Black Avoids 5 ... i..f5: Bronstein and Smyslov Variations 78 7 White Avoids 5 a4: The Quiet 5 e3 and the Geller Gambit 96

List of Games 109 Index of Variations 110 Symbols

+ check ? bad move ++ double check ?? blunder # checkmate Ch championship !! brilliant move 1-0 the game ends in a win for White good move 112-112 the game ends in a draw !? interesting move 0-1 the game ends in a win for Black ?! dubious move (D) see next diagram

Dedication

To my fiancee Heather Denison, for her love and support

Acknowledgements

I would like to thank James Rizzitano for his frequent advice, Joe Fang for his keen eye, Dean Ippolito for his boundless enthusiasm, Graham Burgess for his constructive guidance, and Dena Vigorito for being my sister.

Bibliography

Books Burgess: The Slav, Gambit, 2001 Cox: Starting Out; 1 d4!, Everyman, 2006 Flear: Starting Out; Slav and Semi-Slav, Everyman, 2005 Kasparov: My Great Predecessors II, Everyman, 2004 Sadler: The Slav, Chess Press, 1997

Periodicals Informator - up to Volume 100 New In Chess Magazine - up to 2008, issue 3 New In Chess Yearbook - up to Volume 86

Electronic Resources ChessBase Mega Database 2008 ChessLecture.com Chesspublishing.com The Week in Chess - up to issue no. 703 I ntrod uction

I am happy to present this book, my first for Gambit, on the Main-Line Slav. The purpose of this work is to give the reader a good understanding of all of the lines arising after 1 d4 d5 2 c4 c6 3 tiJf3 tiJf6 4 tiJc3 dxc4. This is not intended as a theoretical manual, although I shall delve into the com plicated theoretical lines when necessary, especially in the most critical and topical variations. The Main-Line Slav is a classical opening and has been played not only in several World Championship matches, but also by most of today' s top grandmasters. The list of players that have employed the Main-Line Slav includes World Champions Alekhine, Euwe, Smyslov, Petrosian, Kasparov, Kramnik, Khalifman, Ponomariov, Topalov and Anand. The list of top modem players is even more extensive: Akopian, Aronian, Bacrot, Bareev, Beliavsky, Bu Xiangzhi, Carlsen, Gel fand, Grishchuk, Harikrishna, Ivanchuk, Kamsky, Kasimdzhanov, Lautier, Malakhov, Moroze vich, Rublevsky, Sakaev, Shirov, I.Sokolov, Van Wely, Wang Yue and Yakovenko. The really amazing thing about these lists is that every player on them has played the Main-Line Slav with both colours. Not only is the Main-Line Slav undoubtedly sound, but it leads to very interesting po sitions. My personal interest in this opening is long-standing. The Main-Line Slav has been a mainstay of my own repertoire for over a decade, and I firmly believe that an understanding of the structures and piece-play will be the leading factor in the reader's success when employing this opening. Firstly, let's examine how we reach our standard positions.

1 d4 d5 2 c4 c6 (D) Sometimes White will use the tricky move This move defines the Slav complex. 2 ... e6 order 3ltJc3 tiJf6 4 e3. Here the desirable devel is also quite solid, and this move is examined opment 4 ... i..f5 runs into the annoying 5 cxd5 in James Rizzitano's Chess Explained: The cxd5 6 'ilfb3!, exploiting the weaknesses on b7 Queen's Gambit Declined. By playing 2 ... c6, and d5. Nevertheless, 4 e3 is rather committal, Black indicates his desire to develop the c8- as White has also locked in his own queen' s bishop before playing ... e6. bishop. Black has a few ways to play. He can re ply 4 ... e6, heading into the Semi-Slav Defence. Now that White has committed to 4 e3 there are no 5 i..g5 lines to worry about. Another option w is 4 ... a6, with a Chebanenko Slav. Again, be cause White has committed to 4 e3, Black has a smaller set of lines to learn. A less common al ternative is 4 ... g6. This hybrid of the Slav and Grtinfeld is called the Schlechter Defence. Black can also play 3 ... dxc4, giving up the centre im mediately. All four options are playable, and Black's choice is largely a matter of taste. Black must also be ready to face the Ex change Slav, 3 cxd5 cxd5. 3 ••• tiJf6 4 tiJc3 (D) Here, too, White could play 4 e3, when Black 3tiJf3 can head for a Semi-Slav with 4 ... e6 (these lines 6 CHESS EXPLAINED: THE MAIN-LINE SLAV are covered by Reinaldo Vera in Chess Ex plained: The Meran Semi-Slav) or stay in Slav territory with 4 ... .i.f5 (or 4 ... Ji.g4). Compared to W the line mentioned above. 5 cxdS cxdS 6 'iib3 is less dangerous because there is less pressure on the dS-pawn. so Black can defend b7 comfortably with 6 .. :Wic7. Once again. 4 ... a6 is a major option here too. As on the previous move, 4 cxdS cxdS is the Exchange Slav.

B 5 ... ~f5 (D) Black has alternatives to this natural devel oping move. The Bronstein Variation, 5 ... ~g4, is rather risky, while the Smyslov Variation, 5 ... lDa6, is slightly passive. Both moves are ex amined in Chapter 6.

4 ..• dxc4 One of the virtues of the Slav Defence is its flexibility. Both 4 ... e6 (the Semi-Slav), and 4 ... a6 (the Chebanenko Slav) are important options. The immediate 4 ... .i.f5?! again runs into 5 cxdS cxdS 6 'iib3. This book will concentrate on 4 ... dxc4, with which Black cedes the centre in order to de velop his c8-bishop. There is also the possibil ity of playing ... b5, making it difficult for White Now White is at a crossroads. There are two to recover the pawn on c4. main bodies of theory, and the resulting mid It is the position after 4 ... dxc4 (see next dia dlegames that arise are quite different. gram) that constitutes the Main-Line Slav. 5a4 Dutch Variation This is the most important move. By pre venting ... b5, White ensures that he will recover 6e3 the pawn. The solid 5 e3 also enables White to This is the classical choice. White recovers recover the pawn, but following 5 ... b5 6 a4 b4 the pawn immediately. White will have to waste time moving his knight 6 •.• e6 7 Ji.xc4 ~b4 around. The aggressive 5 e4 grabs the centre Black takes advantage of the hole on b4 cre immediately but this is a true gambit, as White ated by White's 5th move and hinders the ad will not get his pawn back. Both of these moves vance e4 at the same time. are covered in Chapter 7. 80-0 (D) INTRODUCTION 7

B B

This is the starting point of the Dutch Varia ... eS, whereupon the play can quickly become tion. Black already has a decision to make re very sharp. This is examined in Chapter 3. garding his move-order: he can play either More solid is 7 ... tLlb6 8 ttJeS as, when Black 8 ... tLlbd7 or 8 ... 0-0. The difference between will usually follow up with ... e6 or ... tLlfd7, de them is discussed in the first two chapters of pending on White's play. This approach is cov the book. Against either move White will have ered in Chapter 4. to make a choice between two principled ways Black's other two methods to fight White's of fighting for the advantage: he can aim to plan begin with 6 ... e6. After 7 f3 Black can play playa quick e4 by choosing 9 'iVe2 (Chapter I) a piece sacrifice with 7 ... ~b4 8 e4 ~xe4 or else or he can try to corral Black's fS-bishop by strike in the centre with 7 ... cS. These methods playing 9 tLlh4 (Chapter 2). Both plans lead to are both discussed in Chapter S. middlegames that require understanding of typ ical positions more than concrete theoretical In general, the play after 6 e3 mostly requires knowledge. a more general understanding of the structures that arise, while 6 tLleS necessitates a bit more Central Variation concrete knowledge. Because of this, the pro spective player of the black pieces is well ad 6 tLleS (D) vised to have something worked out against 6 This is the other main branch of play. tLleS. The sword cuts both ways, however, and This ambitious continuation sees White at White must know quite a few different systems tempt to dominate the centre with f3 and e4. If in order to play 6 tLles himself! White is able to do this without making any concessions, Black's fS-bishop could be locked I hope the reader enjoys this work as much as out of play, so Black must take concrete mea I enjoyed writing it and I wish everyone much sures again this plan. There are four main ways success in this rich opening with both colours! of fighting White's intentions. The first two see Black play 6 ... tLlbd7, chal David Vigorito lenging the white knight. After 7 tLlxc4 Black Lowell, Massachusetts again has a choice. 7 ... "fIc7 intends a quick November 2008 1 Dutch Variation: White Plays 9 iVe2

The four games in this chapter examine the main lines arising after 1 d4 dS 2 c4 c6 3 ltJf3 ltJf6 4 ltJc3 dxc4 S a4 1.fS 6 e3 (the Dutch Variation) 6 ... e6 7 i.xc4 1.b4 8 0-0 where White follows up with 9 'ii'e2. White will aim to advance in the centre with e4, and Black can react to this plan in a few different ways. It is important for Black to be prepared to meet this advance, and the move order that Black chooses will generally commit him to certain middlegame plans. This chapter is fundamental to understanding the Main-Line Slav because both sides develop in a very natural and classical manner, so we see the clash of ideas in its purest form.

The Games

Game 1 (Harikrishna-Vescovi) examines the real 'main line' of the Slav. Black allows e4 by playing 8 ... 0-09 'ife2ltJbd7 (or 8 ... ltJbd7 9 'ii'e2 0-0) and prepares to strike back in the centre with ... cS or ... eS. After 10 e4 1.g6 11 -'.d3 i.hS 12 eSltJdS 13ltJxdS cxdS the pawn-structure is deter mined for the middlegame. The play is surprisingly rich, with both sides having reasonable options on almost every move. In the game White is able to reach an endgame with the bishop-pair and a space advantage. Harikrishna's instructive play shows the long-term chances that are possible for White in these kinds of positions.

Game 2 (Siebrecht-Peralta) sees Black prevent e4 using the move-order finesse 8 ... 0-0 9 'ii'e2 -'.g6. This allows White to grab the bishop-pair with 10 ltJeSltJbd7 I1ltJxg6 bxg6, but Black gets a very solid position in this line. This variation is not overly theoretical, which explains its popular ity. Patience is required by both sides in this type of middlegame. White must hope that he will eventually make something of the bishops, while Black must try to achieve meaningful counter play. In our featured game, there is an exchange of inaccuracies, and later White gets sloppy and Black scores with a kingside attack.

Game 3 (Pelletier-L' Ami) looks at a formerly topical pawn sacrifice. After 8 ... ltJbd7 9 'ii'e2 i.g6, White correctly plays 10 e4 anyway and Black snatches the pawn with 10... -'.xc3. Instead the prudent 10... 0-0 would head for the strategic waters of Game 1 and many players use this move order to do just that. After 11 bxc3 ltJxe4 12 i.a3 Black cannot easily castle kingside. This line has lost much of its former popularity because few players relish the task of withstanding White's at tack, but it still appears viable. Here we see White offer up a second pawn to fuel his initiative, and Black ends up having to bail out by giving up two minor pieces for a rook. Materially speaking, this should be fine, but the rooks have little scope and White narrowly misses winning the ending.

Game 4 (Leko-Gelfand) examines a different approach altogether, with Black playing an early ... 1.g4. After the exchange on f3, White has the bishop-pair, but his central control on the dark squares is diminished. This is a solid, if somewhat passive, line for Black. White introduces the in teresting novelty IS 'iffS and an endgame arises that is very pleasant for White. A struggle revolves around the mobility of White's kingside pawn-majority. Black unexpectedly loses a piece and de spite tenacious resistance, White's technique carries him through in the end. DUTCH VARIATION: WHITE PLAYS 9 'ife2 9

Game 1 Penteala Harikrishna - Giovanni Vescovi FIDE World Cup, Khanty-Mansiisk 2005

1 d4 dS 2 c4 c6 3 lbfJ lbf6 4 lbc3 dxc4 S a4 8 ... lbbd7 9 'ii'e2 0-0 10 e4 iLg6 is an alternate iLfS (D) route to the Main Line examined in this game.

w w

6e3 9 'it'e2 This is the most fundamental move - White Note that 9 'it'b3 leads to nothing after simply prepares to recapture the c4-pawn in the 9 ... 'fIe7, protecting both the b4-bishop and the most natural way. He will choose his middle b7-pawn. (We should note that this is a signifi game plan only after completing kingside de cant difference between the 8 ... 0-0 and 8 ... lbbd7 velopment. move-orders, since 8 ... lbbd7 9 'ilVb3 poses Black 6 ..• e6 7 il.xc4 ~b4 more problems, especially if he wishes to avoid Black takes advantage of the weakened b4- a draw - see the notes to Game 6.) square and at the same time holds back White's 9 lbh4 is the only other move to fight for an e4 advance. Often the pressure on the a5-el di advantage. This is covered in Chapter 2. agonal is a nuisance for White. 9 ••• lbbd7 80-0 Black simply allows White to carry out his White is now ready to play 'ife2, intending plan. This may seem like a concession, but e4. Black can simply allow this or try to prevent Black hopes that he will be able either to exert it in a couple of different ways. pressure on White's centre, or to strike back· 8••• 0-0 (D) with ... cS or ... eS. Black can also prevent e4 This is already an important decision. If with the prophylactic 9 ... ~g6, although this al Black plans to play the main line, as in this lows White to grab the bishop-pair with 10 game, it does not make much difference if he lbeS. This approach is covered in Game 2. An castles or plays ... lbbd7 fIrst. The choice may other idea is 9 ... ~g4. This plan is examined in depend on how Black wants to play against Game 4. lbh4 ideas (covered in the next chapter). How 10e4~g6 ever, if Black wants to prevent e4, he must Not 10 ... il.xc3? because of 11 exfS!. Now, know which position he is aiming for. 8 ... lbbd7 however, Black threatens ... ~xc3, winning the 9 'ife2 ~g6 forces White to offer a pawn sacri e4-pawn. fice if he wants to play e4 (Game 3), while 11 il.d3 (D) 10 CHESS EXPLAINED: THE MAIN-LINE SLAV

This is the best way of dealing with the threat actually worse. White has also tried 13 li)a2, to the pawn. but 13 ... li)d5 14 i.xg6 hxg6 15li)xb4li)xb4 16 11 e5 is a rarely-played alternative. White's li)xd4 l:tc8 looks pretty solid for Black. So centre loses its flexibility without getting any while 11. .. c5 looks a bit risky, it is probably not thing in return. After 11 ... li)d5 it is not easy for as bad as it is made out to be. White to achieve anything on the kingside. 12e5 Black's g6-bishop is like a rock, while the c4- Because Black has spent a tempo moving bishop is not contributing much to White's his bishop from the solid g6-square, White play. After 12 li)e4 (12 li)xd5 cxd5 13 i.d3 is makes his move in the centre. After 12 i.f4 similar to the main game, but Black has not both 12 ... :e8 and 12 ... 'iVe7 prepare ... e5 again. spent time on ... i.h5-g6) 12 ... i.e7 13 :dl White can play 13 e5li)d5 14li)xd5 cxd5, but it %lc8 14 a5 a615 h4 h6, Black's pieces all stand is not clear that f4 is such a great square for well and he can prepare ... c5 at his leisure, White's bishop. For one thing, ideas of f4-f5 Lund-Vigorito, Budapest 2003. are off the table for a long time. 12.•. li)d5 (D)

B w

White has created his ideal pawn set-up. Black has to decide how to fight against White's Now White has a choice to make. centre. 13li)xdS 11 •.• ~h5 White changes the pawn-structure, and clari By pinning the f3-knight, Black prepares to fies the position. He can also maintain the ten break in the centre instead with ... e5. Black can sion with 13 li)e4. This is a totally different also fiddle about with 11. .. h6 or 11. .. :e8 but approach. White keeps the knights on the board, the text-move is more direct. hoping to develop an initiative on the kingside. The immediate 11 ... c5 is considered to be a This is a double-edged approach because Black mistake because of 12 e5!. Now 12 ... ~xd3? 13 has a very strong knight on d5, and he can also 'iixd3li)d5 gets crushed by 14li)g5! g6 15 'fih3 strike in the centre with ... c5. Black has two h5 16li)xe6!, and after 12 ... li)d5 13 li)xd5 exd5 sensible replies: 14 i.xg6 hxg6 Black's pawn-structure looks a) 13 ... c5!? was considered bad for a long very creaky. However, 12 ... cxd4 is not so clear. time, but it has been resurrected by Timman. After 13 exf6 dxc3 14 fxg7 ~xg7 15 %ldl 'iVc7! White has: 16 bxc3 i.xc3 17 i.b2 i.xb2 18 'iixb2+ ~g8 al) 14 li)g3 is inaccurate. After 14 ... ~xf3 19 l:tac 1 li)c5 20 i.xg6 hxg6 21 l:[d4 l:tac8 22 15 'iVxf3 cxd4 16 'fIe4 White wins back his lIh4 White certainly had compensation for the pawn, but he has wasted time with his queen pawn in Smyslov-Penrose, Amsterdam Olym and his centre has been undermined. Then piad 1954, but it is not so clear that Black is 16 ... g6 17 i.h6 (17 'fIxd4 'fIc7 leaves White DUTCH VARIATION: WHITE PLAYS 9 'iVe2 11

struggling to protect his e5-pawn, and indeed that the e6-pawn is a long-term weakness, and 18 .i.h61Uc8 19 l:tac 1 i.c5 saw the pawn fall in if Black opens the position with ... c5 this is Paschall-Donaldson, Lindsborg 2(02) 17 ... :e8 more likely to prove important. In practice 18 'ifxd4 .i.c5 19 'ife4 a6 was quite comfort Black has chosen to manoeuvre with his pieces, able for Black in Kobylkin-Dzhakaev, Krasno but White's space advantage can mean a long dar 2004. and passive defence for Black. 16 It)e4 h6 a2) 14 .i.g5 is considered the critical re (16 ... ':f5 17 a5 g5 18 .i.d2 a6 19 b4 llJf8 20 sponse. Timman's innovation was 14 .....c7! (in .:tabl llJg6 21 g3 It)c7 22 i.e3 h6 23 It)fd2 stead, 14 ...... a5?! 15 .i.b5! is annoying) 15 l:tfcl It)b5 24 It)c4 "dS25 'ii'c2 was Yermolinsky h6! 16 .i.d2 'ifd8!. Now Timman cites 17lt)xc5 Khmelnitsky, USA Ch, Parsippany 1996; Black (instead 17 .i.xb4?! It)xb418lt)xc5lt)xc5 19 has no counterplay and the weaknesses of the ':'xc5 .i.xf3 20 "xf3 'iixd4 21 "e4 'ifxd3 22 c5- and d6-squares are a headache for him) and 1fxb4 b6, as in Wojtkiewicz-Timman, Reykja now 17 a5 is a typical method. White gains vik 2000, gives Black a very comfortable ma space and prevents ... a5, which would secure jor-piece middlegame, because the only real the b4-square for Black's .pieces. 17 ... a6 18 weakness on the board is White's e5-pawn) It)e I! (another typical idea - the knight does 17 ... lt)xc5 18 dxc5 .i.xd2 19 "xd2 .i.xf3 20 very little on f3, and from d3 it will eye the im gxf3 1fc7 21 f4 :ad8 22 :c4 a5 as giving Black portant f4-, b4- and c5-squares) 18 ... 'ii'e8 19 compensation for the pawn. This has not been It)d3 g5 20 b4 ...g6 21 It)dc5 It)xc5 22 bxc5 tested in practice, one reason being that 13 It)e4 ltad8 23 It)d6 .i.xd6 24 exd6 ltd7 25 lta3 was has basically vanished from tournament play. played in OIl-Wojtkiewicz, New York 1994. b) 13 ... .i.e7 is the traditional move. Black The protected passed pawn on d6 gives White a returns the bishop to the kingside and covers lasting advantage. the g5-square. Now 14 .tg5? loses a piece to b2) 15 ... hxg6 is a sounder continuation. 14 ... .txf3, so White usually plays 14lt)g3 .tg6 Black will be able to play a quick c5 to make 15 .i.xg6 (D), which gives Black a decision to room for his pieces. The danger here is that make. White will get a knight to g5 and play 'it'g4-h4, but this is difficult to achieve. Several games by Akopian have shown the soundness of Black's position. 16lt)e4 c5 17 ':'dl (instead 17lt)eg5 B cxd4 18 'ii'e4 :e8 19 'iih4 It)f8 is given by Sadler, and is very solid for Black, while 17 It)c3 "b6!? 18 It)xd5 exdS 19 dxc5 It)xc5 20 i.e3 'it'e6 21 a5 a6 22 l:.ac 1 l:ac8 was fine for Black in Xu Jun-Akopian, Moscow Olympiad 1994; Black has good posts for all of his pieces and the e5-pawn is just as weak as the d5-pawn) 17 ... cxd4 18 ':xd4 ':'c8 19 i.e3 'fic7 20 ':c 1 'ilfb8 and the weakness of the e5-pawn is already giving White problems, Gabriel-Akopian, Baden-Baden 1996. 13.•• cxd5 bI) 15 ... fxg6 is played frequently, and in We have reached a French pawn-structure deed the recapture with the f-pawn is a common (or Caro-Kann, if you prefer, because Black's idea in the Slav. Black opens the f-file for his queen's bishop is outside the pawn-chain) rook and can sometimes even take the initiative which offers interesting strategic possibilities on the kingside. Nevertheless, I believe this re for both sides. White has a space advantage, capture is inaccurate here and gives White better while Black has a solid position and can take chances of retaining an edge. The problem is comfort in the fact that he has exchanged one 12 CHESS EXPLAINED: THE MAIN-LINE SLAV minor piece, which helps give him enough room del Mar 2001) and now 18 f4? (Marra-Bella, to manoeuvre. correspondence game 1981) is bad due to 14 'it'e3! (D) Rogozenko's 18 ... lbb6! intending ... lbc4. If 19 b3, then 19 ... l:.c3 20 a5lbd7 threatens ... .te2!. 18 a5 lbb8! is about equal. a2) By 15 lbel! (DJ White prepares f4-f5 B and the knight may come back into play on d3, where it will control many important squares.

This is the point of White's play started with 12 e5. The queen move does several things. The pin on the f3-knight is now broken, and White immediately creates a threat of lbg5. For exam ple, if Black plays the careless 14 ... lbb8? (a standard regrouping, as we shall see, but here it Now Black has: is poorly timed) White has 15 lbg5!, when a21) 15 ... i.g6 16 i.xg6 fxg6 17 lbd3 (17 15 ... ~g6 loses to 16 i.xg6 hxg6 17 'ii'h3. In lbc2 is also possible, when Black should proba stead, 15 ... h6 does not help either. White can bly just play 17 ... i.e7, instead of 17 ... i.a5 as in still play 16 'ii'h3, or even 16lbh7, with the idea Bacrot-Gustafsson, Bundesliga 2(03/4) 17 ... i.e7 16 ... l:e8 17 lbf6+! gxf6 18 'ii'xh6 with a strong 18 'iVh3 l:f5 leaves Black solid but White cer attack. tainly has some initiative, Smejkal-Portisch, The position after 14 'ii'e3 is very important Budapest 1975. in the Main-Line Slav. Black has a number of a22) 15 ... f5 16 exf6 'iVxf6 17 .i.b5 (17 'iVg3 viable options. l:ac8 18 .i.b5 lbb8 19 lbd3 a6 20 lbxb4 axb5 14... ~e7 21 .td2 bxa4 22 l:.xa4 lbc6 23 lbxc6 :Xc6 24 This is the most common continuation. Black i.c3 'ifg6 25 'ii'xg6 .i.xg6 did not lead to much is ready to capture the knight should it hop to in Van Wely-Pelletier, Merida 2005) 17 ... lbb8 g5. Because of White's direct and strong reply, 18 lbd3 a6 19 lbxb4 axb5 20 a5 lba6 21 lbd3 recently Black has started checking the alterna :fc8 22 i.d2 l:.c2 23 l:.ac 1 :ac8 24 :Xc2 l:.xc2 tives more thoroughly: 25 l:.c 1 l:.xc 1+ 26 lLxc 1 lLg6 27 lbe5 gave a) 14 ... h6 has been a frequent choice, but I White an enduring advantage in Gelfand-Lau think that Black should avoid this weakening tier, Horgen 1994. move, especially as it leaves the h5-bishop out a23) 15 ... lLxe1 16 ':'xe1 (16 'ii'h3 'ii'a5! 17 on a bit of a limb. 'ii'xh5 lLd2 is equal) 16 ...lLg6 17 .i.f1! l:.c8 18 a1) 15 i.d2 'ii'e7 (15 ... 'ii'a5 is the same) 16 .td2 "ike7 19 a5 lbb8 20 :'ec1 'ii'd7 21 .i.b4 .i.xb4 'ii'xb4 17 lbh4! plays against the h5- :'xc 1 22 :'xc 1 :c8 23 i.c5 gave White a space bishop, when after 17 ... l:.ac8 (17 ... f5?! 18 exf6 advantage and the bishop-pair in Beliavsky :'xf6 19 f4lbb8 20 h3 lbc6 21 g4 .i.e8 22lbf3 Ribli, Slovenian Team Ch, Bled 2000. gave White the better bishop and pawn-structure b) 14 ... l:.e8 is a relatively recent finesse. in Werle-Sadykov, World Under-18 Ch, Oropesa Black overprotects e6 and prepares ... lbf8, DUTCH VARlATION: WHITE PLAYS 9 'iWe2 13

which will cover the sensitive e6-, h7- and g6- the accurate 18 .....e719 'ii'xb7 (19 ttlxe6 a6 20 squares. 15 ~1 (15 ttlg5 ~g6 transposes to 'ii'xb7 ltab8 21 "xdS 'ii'xe6 is better for Black 14... ~g6 15 ll)g5 l:eS below, although Black because of his active pieces) 19 ... h6! 20 lLJh3 has avoided 15 ~xg6; another idea is 15 ~d2!? (20 1LJf3lLJxe5! 21 "fIxe71LJxf3+ 22 l:xf3 l:xe7 .i.xd2 16 lLJxd2 lIcS 17 as lLJbS IS ~b5 :tfS is also good for Black) 20... a6! White must head 19 lLJb3, which gave White a slight edge in for a draw with 21 ~e3 :ebS 22 'ikc7 :CS 1/2-1/2 Berkes-Portisch, Hungarian Ch, Heviz 2(03) Gulko-Vigorito, USA Ch, Tulsa 2008, because 15 ... :cSI6f4~xeI17 :xel ~g6isequal. M 21 as l:eb8 22 'ifc6 l:cS 23 'iia4 :c4! gives ter IS ~f1 :c2 19 b3 "as 20 ~b5 (Topalov Black tremendous compensation for the pawn. Kramnik, World Ch rapid playoff (game 1), c3) IS g4 is a sharp try, which commits Elista 2006), 20 ... ~e4! 21 ~xd7 Lg2+ 22 White to a controversial piece sacrifice: IS ... h6! ~f1 :xh2 23 ~xeS :hl+ 24 ~f2 :h2+ is a 19 lLJxe6 (retreating would simply allow Black draw, because 25 ~g3?? 'ifdS! leads to mate. to capture on g4, and 19 gxf5 is worse in view c) 14 ... ~g6151LJg5 (15 ~xg6 fxg6 16 ttlg5 of 19 ... hxg5 20 f6 g4 21 'iig6ltJxf6!, Cordova :eS 17 'iih3 lLJfS does not trouble Black, al A.Kovalyov, South American Junior Ch, Bue though 15 ~e2 could be tried, when Black can nos Aires 2(07) 19 ... ltxe6 20 gxf5:e7 (20 ... ltc6 even respond with the equally coy 15 ... .i.h5) is also possible, one point being that 21 "fIb5 15 ... :eS! 16 f4 ~xd3 (16 ... lIcS 17 g4, as in can be met by the counterattack 21 ... 'iih4!, Pelletier-Deviatkin, Moscow 2003, should be when 22 'ifxb4 ':c2 23 ~d2 as! is winning for met by 17 ... ~xd3 IS .xd3 lLJfS with a solid Black and 22 'iixdS+ hS also looks insuffi position) 17 'ifxd3 f5! (D). cient for White because Black's pieces will be very active) 21 ~hl ~h8 22 ltgl "'fS is given as better for Black by Timman, but matters are still not so clear. w c4) 18 ~h 1 !? is an interesting idea. After 18 ... ltJf8 19 :gl ~e7 20 ltJf3 :c8 21 .i.d2 White intends a quick g4. The bishop may go to c3, or in the case of b3, it will cover the c3- square. 15 ltJg5! This forcing continuation is White's latest attempt to squeeze something from the posi tion. The coming simplifications do not make things as easy for Black as was once thought. Black's defences have held up well against other moves: Black blocks the kingside while he has the a) 15 l2Jel was a favourite of Gligoric, but chance. White can reply: without Black's weakened kingside (compared c 1) 18 iLe3 lLJf8 19 ~h 1 ':'c8 (chasing the to the position after 14 ... h6), it does not look knight away with 19 ... ~e7Iooks safer: 20 lLJf3 too threatening. After 15 ... ~g6 (15 ... .:tc8 is l:c8 21 h3 "c7 22 :tfc 1 'ifd7 23 :tg1 lIc7 24 also logical, because 16 f4 is met by the clever ':'gcl lid7 25 'lib3 1/2-1/2 Yakovenko-Rublev trick 16 ... l2Jc5!) 16 f4 ':'c8 (16 ... ~xd3 17ltJxd3 sky, Russian Ch, Moscow 2006) 20 g4 'ii'd7 21 f5 18 exf6 ':'xf6 19 iLd2 as 20 ':'ael .i.d6 was :gl ~e7 221LJf3 :C4 23 :g2 fxg4 24 ':'xg4 agreed drawn here in Gligoric-Khalifman, Plov :xa4 25 ':'agl g6 26 h4 l:b4 27 h5 'Wb5 with div 19S6, but White has 21 f5 exf5 22 'iff3 sharp play was the famous game Topalov-Kram threatening both 'iixdS+ and .i.g5) 17 g4 .i.xd3 nik, World Ch match (game 2), Elista 2006. 18 ltJxd3 lk4 19 ~d2 'ifb6 20 .i.c3 f5 21 exf6 c2) 18 'iib5 has gone unmentioned by com .i.xf6 22 l2Je5 ~xe5 23 fxe5 ':'cc8 1/2-1/2 was mentators, even though it wins a pawn! After Gligoric-Donaldson, Lone Pine 1981. 14 CHESS EXPLAINED: THE MAIN-LINE SLAV

b) 15 ..td2 is a common and flexible move. counterplay) 18 ... ':'xd8 19 ..td2 ttJc6 20 ..tc3 The bishop supports b4 and can go to c3 if nec :dc8 21 a5 a6 gave Black a solid position in essary to shore up the d4-pawn. 15 ... ttJb8! (the Topalov-Bu Xiangzhi, Bilbao (blindfold) 2007, knight heads to its ideal square on c6) 16 ttJel but it seems White still has a little something (White could also try to seize space on the with his bishop-pair. queenside immediately with 16 b4 or 16 a5) 18..txgS 16 ... ..tg6 17 f4 ttJc6 18 g4 (or 18 ..tc3 "'e8 19 With a space advantage and the bishop-pair, g4 f5 20 g5 Ih-1f2 Gligoric-Donaldson, Vancou White holds a nagging edge in the endgame. ver 2000) 18 ... f5 19 exf6 ..txf6 20 ..tc3 'ii'e8 21 Black has no counterplay. h3 ~h8 22 :dl has been played a few times. 18••• :rc8 White has space while Black is very solid, and After 18 ... l:ac8, 19 ..tb5 is a problem be both sides have pawn weaknesses. cause the desired 19 ... ttJb8 fails to 20 ..te7. The IS.•• ..txgSI6 "'xgS ..tg6 (D) immediate 18 ... ttJb8 19ltfcl ttJc6 is relatively best, at least forcing one of White's pieces to defend the d4-pawn. 19 :rcl (D) w

17..te2! This is the point of White's play - White simply holds on to the bishop-pair. Instead 17 ..txg6 fxg6 18'iVg4 l:f5, as in Bacrot-Akopian, Black already feels very uncomfortable. It is European Team Ch, Gothenburg 2005, does not becoming difficult for him to coordinate his get White anywhere. pieces because he cannot play ... ttJb8-c6. After the text-move, the position is not so 19••• a6 20 h4! easy for Black despite his smooth development. Gaining more space and forcing Black to In addition to the bishop-pair, White has a space make a decision concerning the kingside pawn advantage and the possibility of pressing on ei structure. ther wing. The one weak spot in White's posi 20..• f6 tion is the pawn on d4, but it is difficult for Mter 20 ... h6 21 ..td2 Black is faced with al Black to bring any meaningful pressure to bear lowing h5, gaining more territory and fixing the upon this pawn. kingside pawns on dark squares. If he plays 17 ...... xgS 21. .. h5 himself, his bishop will be tied to the If Black avoids the exchange of queens with defence of the h-pawn for the foreseeable fu 17 ... "'b6, then 18 "'e7! is rather annoying. In ture. stead, rerouting with 17 ... ttJb8 is thematic. 18 21 exf6 "'xd8 (18 "'e3!? ttJc6 19 ..td2 was Shabalov The disturbance in the pawn-structure fa Vigorito, Las Vegas 2005, and here Donaldson vours White, because his bishops can operate suggests 19 .. :ii'b6 20 ..tc3 ttJa5 with some more easily allover the board. DUTCH VARIATION: WHITE PlAYS 9 'ife2 15

21 •• .lLJxf6 unpleasant. Black's pieces are pushed back and Perhaps 21..,gxf6 was a better try, hoping to White's pawns will be closer to queening as the get some counterplay in the centre. position simplifies. 22 f3 i.c2 25 ••• liJd7 26 b5 :'cc8 27 as This temporary closing of the c-file brings Fixing the a-pawn with 27 bxa6 bxa6 28 a5 Black little relief. also looks very good. 23:aJ! ~7 24~2:c6 (D) 27 ••. axb5 28 i.xb5 ltJf6 29 :a2 i.b3 29 ... i.g6 was a better try, but Black was un derstandably loathe to remove the bishop from the battle on the queenside. 30 ':xc8 :'xc8 31 ':'b2 ii.c4 Exchanging rooks with 31. .. :c2+ 32 ':xc2 ii.xc2 does not help after 33 ii.xf6 gxf6 34 i.c6!. 32 ii.xc4 dxc4 33 l:txb7+ ~g6 34 a6 Black's c-pawn is not dangerous, because White controls the queening square. 34... ':'c6 35 a7 l:ta6 36 i.xf6! White simplifies into an easily winning rook endgame. 36••. xf6 37 ~e3 liaS 38 ~d2 :a4 39 c3 h5 40 g4 ~g6 41 g5 h7 42 l:te7! 1-0 25b4! Black is in zugzwang. After 42 ... ~g6 43 f4 This is a typical method in this kind of struc h7 44 d5! exd5 45 f5 Black will soon have to ture. The further advance of the b-pawn is very give up his rook or be mated.

Game 2 Sebastian Siebrecht - Fernando Peralta Binissalem 2004

1 d4 d5 2 c4 c6 3 liJf3 liJf6 4 liJc3 dxc4 5 a4 i.f5 6 e3 e6 7 i.xc4 i.b4 8 0-0 0-0 9 'ii'e2 i.g6 This prophylactic move is a solid alternative w to the main line of Game 1. If Black really wants to prevent e4, this is the safest way to do so. Black develops easily and has plenty of room for his pieces. This all comes at a price though, as White can easily grab the bishop-pair while slightly weakening Black's pawn-structure. Black can also choose to stop e4 physically with 9 ... liJe4 (D). Usually White has chosen to meet this with an undoubtedly sound pawn sacrifice (line 'c'), but there are alternatives. a) 10 liJxe4 does not look critical. 10 ... i.xe4 either) 11. .. liJd7 12 ii.d3liJf6 13 i..d2 i.d6 was 11 ltdl (11 liJd2 i.g6 12 liJb3 has also been very comfortable for Black in Sunye-Vescovi, played, but this does not look too dangerous Sao Paulo 2005. 14 CHESS EXPLAINED: THE MAIN-LINE SLAV

b) 15 .id2 is a common and flexible move. counterplay) 18 ... :xd8 19 .id2lLlc6 20 .i.c3 The bishop supports b4 and can go to c3 if nec lldc8 21 a5 a6 gave Black a solid position in essary to shore up the d4-pawn. 15 ... lLlb8! (the Topalov-Bu Xiangzhi, Bilbao (blindfold) 2007, knight heads to its ideal square on c6) 16lLlei but it seems White still has a little something (White could also try to seize space on the with his bishop-pair. queens ide immediately with 16 b4 or 16 a5) 18.i.xgS 16 ... .ig6 17 f4 lLlc6 18 g4 (or 18 .i.c3 'ii'e8 19 With a space advantage and the bishop-pair, g4 f5 20 g5 Ih-1f2 Gligoric-Donaldson, Vancou White holds a nagging edge in the endgame. ver 2(00) 18 ... f5 19 exf6 .i.xf6 20 .i.c3 'ii'e8 21 Black has no counterplay. h3 c.th8 22 l:tdl has been played a few times. 18••• lUe8 White has space while Black is very solid, and After 18 ... :ac8, 19 .i.b5 is a problem be both sides have pawn weaknesses. cause the desired 19 ... lLlb8 fails to 20 .i.e7. The IS ••• .ixgS 16 'ii'xgS .ig6 (D) immediate 18 ... lLlb8 19 l:fcllLlc6 is relatively best, at least forcing one of White's pieces to defend the d4-pawn. 19 :fel (D) w

17.i.e2! This is the point of White's play - White simply holds on to the bishop-pair. Instead 17 .i.xg6 fxg6 18 'iWg41H5, as in Bacrot-Akopian, Black already feels very uncomfortable. It is European Team Ch, Gothenburg 2005, does not becoming difficult for him to coordinate his get White anywhere. pieces because he cannot play ... lLlb8-c6. After the text-move, the position is not so 19... a6 20 h4! easy for Black despite his smooth development. Gaining more space and forcing Black to In addition to the bishop-pair, White has a space make a decision concerning the kingside pawn advantage and the possibility of pressing on ei structure. ther wing. The one weak spot in White's posi 20 ••• f6 tion is the pawn on d4, but it is difficult for After 20 ... h6 21 .i.d2 Black is faced with al Black to bring any meaningful pressure to bear lowing h5, gaining more territory and fixing the upon this pawn. kingside pawns on dark squares. If he plays 17•.• 'ii'xgS 21. .. h5 himself, his bishop will be tied to the If Black avoids the exchange of queens with defence of the h-pawn for the foreseeable fu 17 ... 'iWb6, then 18 'ife7! is rather annoying. In ture. stead, rerouting with 17 ... lLlb8 is thematic. 18 21 exf6 'ii'xd8 (18 'ii'e3!? lLlc6 19 .i.d2 was Shabalov The disturbance in the pawn-structure fa Vigorito, Las Vegas 2005, and here Donaldson vours White, because his bishops can operate suggests 19 .....b6 20 .i.c3 lLla5 with some more easily allover the board. DUTCH VARIATION: WHITE PLAYS 9 'ife2 15

21 •.• ti)xf6 unpleasant. Black's pieces are pushed back and Perhaps 21. .. gxf6 was a better try, hoping to White's pawns will be closer to queening as the get some counterplay in the centre. position simplifies. 22 f3 i.c2 25 ••• ti)d7 26 b5 ':'cc8 27 as This temporary closing of the c-file brings Fixing the a-pawn with 27 bxa6 bxa6 28 a5 Black little relief. also looks very good. 23 1la3! c:j;f7 24 ~2 :c6 (D) 27••• axb5 28 i.xb5 ti)f6 29 :a.z i.b3 29 ... i.g6 was a better try, but Black was un derstandably loathe to remove the bishop from the battle on the queenside. 30 ':'xc8 :xc8 31 :b2 i.c4 Exchanging rooks with 31. .. :c2+ 32 ':'xc2 i.xc2 does not help after 33 i.xf6 gxf6 34 i.c6!. 32 i.xc4 dxc4 33 %lxb7+ c3 h5 40 g4 ~g6 41 g5 ~h7 42 lIe7! 1-0 25 b4! Black is in zugzwang. After 42 ... h7 44 d5! exd5 45 f5 Black will soon have to ture. The further advance of the b-pawn is very give up his rook or be mated.

Game 2 Sebastian Siebrecht - Fernando Peralta Binissalem 2004

1 d4 d5 2 c4 c6 3 ti)f3 ti)f6 4 ti)c3 dxc4 5 a4 i.f5 6 e3 e6 7 i.xc4 i.b4 8 0-0 0-0 9'ilie2 i.g6 This prophylactic move is a solid alternative W to the main line of Game 1. If Black really wants to prevent e4, this is the safest way to do so. Black develops easily and has plenty of room for his pieces. This all comes at a price though, as White can easily grab the bishop-pair while slightly weakening Black's pawn-structure. Black can also choose to stop e4 physically with 9 ... ti)e4 (D). Usually White has chosen to meet this with an undoubtedly sound pawn sacrifice (line 'c'), but there are alternatives. a) 10 ti)xe4 does not look critical. 10 ... i.xe4 either) 11. .. ti)d7 12 i.d3 ti)f6 13 i.d2 i.d6 was 11 Ildl (11 ti)d2 i.g6 12 ti)b3 has also been very comfortable for Black in Sunye-Vescovi, played, but this does not look too dangerous Sao Paulo 2005. 16 CHESS EXPLAINED: THE MAIN-LINE SLAV

b) 10 lLla2 i.e7 (10 ... i.d6 I1lLle5 i.xe5 12 quite well. All of his pieces, especially the dxe5 ltJc5 13 ltd 1 was better for White in g6-bishop, are active, and the attempt to fight A.Rychagov-Guliev, Moscow 1997) 11 i.d3 for the initiative with 14 lba2 is well met by lbg5 12 lbe5 i.xd3 13 1Wxd3 f6 14 lbc4 c5 15 14 ... lbb6 (14 ... .i.a5 15 b4 .i.c7 16 .i.a3lLlb6 17 f4 lbf7 was unusual, but equal, in Lugovoi .i.b3 lLld5 is also quite good) 15 b3 (or 15 lbxb4 Bareev, Russian Ch semi-final, Kazan 2005. 'ii'xb4 16 b3 lLlxc4 17 'ifxc4 'iVxc4 18 bxc4 c) 10 i.d3 (this pawn sacrifice of Euwe's is ltfd8) 15 ... .Jtc5 with a great position for Black. still considered to be important) 10 ... .i.xc3 11 10•.. lbbd7 lllLlxg6 bxc3 lLlxc3 12 'iWc2 .i.xd3 13 'iVxd3 ttJd5 14 White must capture immediately, because 11 .Jta3 :e8 15 l:tabl b6 16 lIfcl a5 17lbe5 "ilc7 %ldl?! lbxe5 transposes to the previous note. (it is too early to return the pawn: 17 ... lbb4? 18 11 •.. hxg6 (D) .i.xb4 axb4 19 lbxc6 lbxc6 20 lIxc6 was much better for White in Euwe-Alekhine, World Ch match (game 17), The Hague 1937) 18 'fib3 (18 e4 allows Black to simplify the position by re w turning the pawn with 18 ... lLlb4 19 i.xb4 axb4 20 l:txb4 c5) 18 ... :d8 19""2 gave White good compensation in Van Wely-Ni Hua, Moscow 2005. d) 10 lbe5!? is the latest try. After 10 ... lbd7 11 lbxd7 'ilxd7 12lLla2 .i.e7 13 f3 lbf6 14 e4 .Jtg6 15 .i.e3 'fIc7 16lbcl! White had a stable advantage in Topalov-Bu Xiangzhi, Sofia 2008. The knight is heading to d3, where it will con trol the important squares b4, c5 and e5. We now return to 9 ... i.g6 (D): White has the bishop-pair, but Black has a very solid position, a slight lead in develop ment, and the chance to play in the centre with w an eventual ... c5 or ... e5 break. Compared to the previous game, White does not have much of a space advantage, but it is also less likely that White will become overextended. An interest ing feature of this line is that it is usually easier for Black to develop constructively - the queen goes to a5 and the rooks come to the centre. White will usually play .i.d2, but then he has to figure out how to arrange his major pieces. Be cause Black is so solid and it is not so easy for White to advance in the centre, recently White has been examining some rather unconventional 10 ttJes! continuations over the next few moves. Instead 10 e4? fails to 10 ... .Jtxc3 11 bxc3 l2l:tdl lbxe4 12 .Jta3 l:te8, when White has very little The immediate 12 e4?! is premature because for his pawn. the d4-pawn is a bit loose. After 12 ... lLlb613 e5 The natural move 10 l:tdl is too tame. After (no better is 13 .Jtd3 'ilxd4 or 13 .Jte3 .i.xc3 14 10 ... lbbd7 it is not easy for White to do any bxc3lbxe4) 13 .....xd414 i.xe6 (14 exf61iVxc4) thing constructive because 11lLle5?! lbxe5! 12 14 ... fxe6 15 exf6 'ii'xf6 16 ttJe4 'ike5 White dxe5 lLld7 13 f4 'ike7 leaves Black standing does not have enough for the pawn. DUTCH VARIATION: WHITE PLAYS 9 'iVe2 17

A relatively new try is the aggressive 12 ltJa2!?, by which White tries to muscle in the e4 advance. After 12 ... i.d6 (12 ... i.e7 is also pos w sible, if somewhat passive) 13 e4 e5 White ini tiates complications with 14 f4! (D).

successfully meet these breaks with d5, Black's position can cave in quickly. White has the bishop-pair, but his dark-squared bishop is passive and his queenside weaknesses make it somewhat difficult for him to advance in the centre. Often the middlegame sees White try 14 ... exd4 (instead 14 ... .,,6 loses a piece, ing to slow the pace down a bit because Black but it still may be playable! 15 ~h 1 exd4 16 e5 can mobilize all of his pieces so quickly. If l:tae8 17 'iWc2 i.xe5! 18 fxe5 liJxe5 19 .i.f4 White can get properly developed without mak liJxc4 20 'iWxc4 l:e4 gave Black reasonable play ing any major concessions, he can look forward in Golod-U.Zak, Tel-Aviv 2(00) 15 e5 i..c5 to gradually exploiting the long-term advan (the pawn fork does not win a piece for White tage of the bishops. after all, but he hopes to use it to generate an 13 i..d2 initiative, possibly with the further advance e6, This obvious developing move is the most cracking open Black's position on the light natural reply to the attack on c3, but because squares; it seems, however, that Black has suf this can render it difficult for White to make ficient resources) 16 ~h 1 tLld5 17 b4!? (White any headway without allowing an exchange of does not have the development to support the e6 dark-squared bishops, White has tried lashing advance just yet) 17 ... .i.e7 (17 ... liJxb4 18 liJxb4 out with a couple of aggressive alternatives: .i.xb4 19 e6 looks dangerous for Black) 18 a5 a) 13 e4 and then: :e8 19 i..d2 tLlfS 20 'ii'e4 'ii'd7 21 'iWxd4 l:.ed8 al) 13 ... .i.xc3 14 bxc3 "xc3 is rare, but was unclear in Van Wely-Z.Almasi, Polanica there is no specific reason why Black should Zdroj 2000. Black is still very solid and although avoid this. 15 .i.d2 'ii'xd4! 16 i.b4 'iWe5 17 White has recovered his pawn and has the .i.xf8 l:xf8 should be fine for Black, with two bishop-pair, all of his pawn advances mean that pawns for the exchange, and 15 ':a3 can be met his position could easily become overextended. by 15 .....b4 or even 15 ... 'ii'a5, when although 12 .. :.aS (D) White certainly has some compensation, Black Black can also play 12 .....e7 but it looks can be satisfied with the opening, having an ex rather passive compared to the text-move. After tra pawn and solid position. Of course this will 13 e4 e5 14 d5 tLlb6 both 15 dxc6 and 15 i.b3 not suit everyone. have scored well for White. a2) 13 ... e5 is more conservative and is the After the text-move, Black has a nice, active most common move in practice. 14 d5liJb6 15 position with good development, but he will i.b3 (15 dxc6 bxc6 16 .i.d3 l:tfd8 is not bad for have to be careful on the light squares, especially Black - he has active pieces and play on the dark after advancing with ... c5 or ... e5. If White can squares and d- and b-files, and the c6-pawn 18 CHESS EXPLAINED: THE MAIN-LINE SLAV covers the bS- and dS-squares) IS ... l:.ad8 16 it open up. Black can play 13 ... eS immediately, ~gS ~xc3 17 bxc3 cxdS 18 'tiDs 'iixbS 19 but this allows 14 dS! l:ad8 (l4... cxdS 15 tiJxdS axbS tiJxe4!? 20 ~xd8 l:.xd8 21 l:.xa7 tiJcs tiJxdS 16 .i.xb4 'iixb4 17 .i.xdS is very nice for gave Black reasonable compensation for the White) IS dxc6 bxc6 16 .i.el e4. Black has exchange in Antic-Sakaev, Herceg Novi 2001. some activity, but with the bishop-pair and b) 13 ttJa2 'ifxa4 (both 13 ... .i.d6 and 13 ... ~e7 better pawn-structure, White looks for choice are safer, but they allow White to play 14 e4) 14 after either 17 l:td4 or Kramnik' s suggestion 17 e4 (14 b3?! does not work after 14 ... 'iiaS IS 'iVc2. tiJxb4 'iixal 16 tiJa2 'iibl! 17 .i.a3 "'fS 18 .i.xf8 l:txf8, when Black was just a pawn up in C.Bauer-Z.Almasi, European Team Ch, Batumi 1999) 14 ... 'iIi'aS IS eS (IS tiJxb4 'iixb4 16 b3 w l:tfe8100ks insufficient for White) IS ... tiJdS 16 h4. White is playing for a direct attack. Black has: bI) 16 .. JUe8 17 l:td3 tiJ7b6 (Hubner's sug gestion of 17 ... .i.e7 18 l:h3 tiJf8 is solid enough, although White still has attacking chances after 19 hS) 18 .i.b3 'iibS 19 ~c2 .i.f8 20 hS gave White the initiative in Gomez Esteban-Y.Geor giev, Elgoibar 1998. b2) With 16 ... cS!? Black strikes at the centre immediately. 17 .i.xdS (the untried 17 tiJxb4!? looks tempting, because 17 ... 'iVxal 18 tiJxdS 14.i.el exdS 19 ~xdS gives White a strong initiative) White drops back his bishop so it does not 17 ... exdS 18 .i.e3 'iWb6 19 liJxb4 'iWxb4 20 hS obstruct the d-file. The immediate 14 e4 can be gxhS 21 'iWxhS l:.fe8 22 lla3 !? gave White some met by 14 ... tiJb6, hitting the c4-bishop as well initiative for the pawn in Nayer-Skachkov, St as the d4-pawn. Now, however, because the Petersburg 2003. d4-pawn is secure, White is finally threatening We now return to 13 .i.d2 (D): to play e4. 1bis is the most popular way to play the position. Other moves: a) 14 ~b3 can transpose to the game after 14 ... tiJb6 IS .i.el, or Black can play the imme B diate 14 ... eS IS dxeS'iixeS. b) 14 tiJa2 ~xd21S 'iVxd2 'ir'xd216 l:xd2 is not too threatening. After 16 ... tiJb6 17 b3 (17 ~b3 as is equal) 17 ... l:td7 18 l:c2 liJxc4 19 l:xc4 l:e8 20 tiJc3 eS 21 tiJe2 exd4 22 liJxd4 tiJe4 Black even developed some initiative and went on to win in Blackburn-Shaw, European Union Ch, Liverpool 2006. c) 14 'ilfel is an odd move. White is not really creating any pressure on the el-aS diago nal, if that is what he was hoping for. After 14 ... eS ISliJe4 (instead IS tiJe2 ~xd2 16'iixd2 13 ••• l:.ad8 (D) 'iixd2 17 l:txd2 exd4 18 tiJxd4 tiJeS 19 l:tc2 This centralization is very logical. The rook tiJxc4 20 l:xc4 l:tdS was equal in J aeschke can help control the important dS-square and Shaw, Bad Wiessee 2003) IS ... ~xd2 16 tiJxf6+ Black will be ready to contest the d-file should tiJxf6 17 'i'ixd2 'iixd2 18 l:txd2 exd4 19 exd4 DUTCH VARIATION: WHITE PLAYS 9 "ike2 19

(keeping the pawns intact with 19 :adl is more solid) 19 ... :d7 20 f3 :fd8 21 :adl a5 Black had a comfortable ending in Zhidkov-Skach- W kov, Naberezbnye Chelny 2006. We now return to 14 i.el (D):

healthier-looking pawn-majority. Black is well centralized, has a very solid position with good dark-square control on the queenside and some prospects for counterplay on the kingside as well by ... ltJg4 or ... g5-g4. White would love to be able to play e4, but this is not easy because 14••• lZ'lb6 the pawn advance will give Black even better Before breaking in the centre, Black takes control of the dark squares in the centre. The control of the d5-square. This is a solid ap course of the middlegame will focus on Black's proach, although Black has also tried to do piece activity - if White can neutralize Black's without this move: counterplay, he can hope to gradually exploit a) 14 ... c5 15 d5 exd5 16 i.xd5 (16 lZ'lxd5 his bishop-pair. i.xel 17 :xellZ'lxd5 18 i.xd5lZ'lf6 19 i..xb7 171i'c2 :d2 gives Black counterplay) 16 ... lZ'lxd5 17 Tickling the g6-pawn. White can also rush to lZ'lxd5 i.xe118 ':'xellZ'lf619 'iVb5 was drawn in simplify himself with 17 ':'xd8 l:xd8 18 l:dl. Beliavsky-Gelfand, Lvov 2000, although White Black usually chooses to keep one pair of rooks still maintains a slight initiative. on the board with 18 ... ':'e8 because White can b) 14 ... e5 15 d5 (this is the most critical, as not penetrate on the d-file anyway. Retaining 15 'ii'c2 {eyeing the g6-pawn} 15 ... exd4 16 one set of rooks also gives Black better chances ':'xd4lZ'le5 17 i..a2 c5 18 l:ddl 'ii'a6! controls of gaining active counterplay at some point. c4 and d3, giving Black a comfortable position, Black will follow with ... a5 or ... g5 and should Atalik-Haznedaroglu, European Team Ch, Plov have sufficient play to hold the balance. div 2003) 15 ... cxd5 (l5 ... e4 16 dxc6 bxc6 trans Prophylactic moves such as 17 g3 and 17 h3 poses to the note to Black's 13th move) 16 are also possible. i..xd5ltJc5 171i'c41ooks a bit more comfortable 17... ':'xdI18 ':xdl gS for White due to his good light-square control. Black removes his g-pawn from attack. Black 15 i.b3 e5 can also get adventurous with 18 ... ltJg4 al Black finally breaks in the centre. After though after 19 h3! (instead 19 g3?! 1i'h5 20 h4 playing this move, Black must always pay at ltJe5 gives Black very good play, and 19 1i'xg6 tention to his g6-pawn (because the f-pawn is 'ifxh2+ 20 c;t>n ltJe5 was equal in Lugovoi pinned by White's bishop on b3), but in prac Skachkov, St Petersburg 1999) 19 ... 'ifh2+ 20 tice this has not been too troubling for Black. ~n 'ifhl+ (20 ... ltJf6 21 'ii'xg6) 21 d2 i..xc3+ 24

30 i.b3+ 1-0 was Werle-L'Ami, Groningen 21 .....e7 2001. Instead 21. .. 'ii'xd4 22 :xd4 gives White just 19 h3 (D) the kind of ending that he is looking for. This just gives something for Black to latch 22~c2:e8 onto. Calmly strengthening the position with By keeping the queens and one pair of rooks 19 'ii'd3! intending 'ii'd4 was stronger, because on the board, Black can hope for counterplay. 19 ... tiJg4 20 g3 does not lead anywhere now 23 "d3 g6 24 'ike2 "'eS 25 "c4 cJ;;g7 26 g3?! that the black queen cannot go to h5. This weakening looks unnecessary. 26 ~b3 'fIe7 is better, with approximate eqUality. White has the bishop-pair but it is difficult to exploit this because White cannot do much without ex B changing dark-squared bishops. This is a typi cal situation with Black's bishop entrenched on b4 opposing White's passive bishop on el. 26..... e7 27 ~b3 :b8 (D)

19 ... aS Black secures both his bishop on b4 and the dark-squares on the queenside. This typical idea is solid enough, but Black could also play 19 ... g4, immediately creating counterplay on the kingside. 20 'ii'd3! tiJbd7 20... g4 is still possible, although now White Suddenly Black has serious kingside play. can force the exchange of queens with 21 hxg4 28 g2 tiJe5 29 'iWe2 g4! 30 bxg4 tiJxg4 22 'ii'd4 because the check on h2 is not 30 h4 tiJf3, with ... g5 coming, also does not fatal and both of Black's knights are attacked. look too appetizing for White. 21 "'d4 (D) 3O •.. tiJfxg4 (D)

B w DUTCH VARIATION: WHITE PLAYS 9 1We2 21

31 !i)e4? Attacking most of White's pieces. This is a blunder, but after 31 -'.c2 or 31 33 -'.c2 l:h2+ 0-1 l:d4, 31. ....g5 heads to the h-file. Black wins after either 34 ~f1 :h1 + or 34 31 .•. -'.xel 32 l:xel 'iIb4! ~gllbf3+.

Game 3 Yannick Pelletier - Erwin L'Ami Bundes/iga 2006/7

I d4 cIS 2 c4 c6 3 lbfJ lbf6 4 lbc3 dxc4 5 a4 -'.f5 6 e3 e6 7 -'.xc4 -'.b4 8 0-0 lbbd7 9 'iWe2 -'.g6 (D) B

the possibility of .i.xe6!. After White moves his c4-bishop, there will be chances to increase the pressure with the pawn advances c4 and a5. Theoretically, Black seems to be doing alright, With this move-order, Black attempts, or at but the position is very dangerous for Black and least pretends to attempt, to prevent both e4 and because of this it is not too popular any more. lbe5. However, White still can, and should, ad Nevertheless, Black has a solid position and an vance in the centre. extra pawn, and the position is not without risks 10 e4! for White either. This pawn sacrifice is the only way to chal 12..... c7 lenge Black's move-order. 10 ':'d1 0-0 11lbe5?! Black prepares to castle queenside. The sec is not dangerous. This transposes to the note to ond pawn is really too hard to digest after White's 10th move in the previous game. 12 ... lbxc3? 13 'ifb2. With 'iixb7 coming, and 10... .i.xc3 the bishop controlling f8, Black's king has no Black can grab the e4-pawn because his good shelter. bishop is not hanging on f5 any more. In prac After the text-move, Black may also consider tice Black often plays 10... 0-0 transposing back grabbing the c3-pawn because b7 is protected, into Game 1. Using this move-order can at least and ... c5 becomes possible, blocking White's make White bum some time deciding to offer a3-bishop. Black's main defensive scheme will the pawn sacrifice. be to play ... lbd6 and castle queenside. He will II bxc3lbxe412 i..a3 (D) then have the possibility to break in the centre White prevents Black from castling kiiig with ... c5 or, more frequently, ... e5. White has a side. There is also some latent pressure on the decision to make. e-file because if the e4-knight moves there is 13 l:fel!? 22 CHESS EXPLAINED: THE MAIN-LINE SLAV

White prefers to use his king' s rook to bring c3) 15 d5 and here instead of 15 ... 0-0 (Van pressure to the e-file, even though the c3-pawn Wely-Hutters, Tastrup 1992) Black can play remains loose. This is a promising way to han 15 ... 0-0-0! threatening both ... tbb6 and ... tbf6, dle the position, but other moves have been winning another pawn. 16 dxe6 is answered by tried as well: 16 ... tbb6 17 'ii'e2 tbxc4 18 'ii'xc4 .i.d3 winning a) 13 'iWb2 is a little inflexible. 13 ... c5 (Black the exchange. can also try 13 ... 0-0-0 and 13 ... tbd6) 14 dxc5 d) 13 ':fcl is White's main alternative. tbexc5 15 i..b5 0-0 (15 ... 0-0-0!?) 16 i..xd7 13 ... 0-0-0 (after 13 ... c5, 14 .i.b5 0-0 15 i..xd7 tbxd7 17 i..xf8 ':xf8 18 a5 tbc5 was Gulko 'ifxd7 16 tbe5 'ifd5 17 f3 tbf6 18 .i.xc5 gave Kreiman, USA Ch, San Diego 2006. Black has White a pleasant advantage in Ikonnikov-Sash a pawn and a very solid position for the ex ikiran, Vlissingen 2005, while 14 ~5 tbxe5 change. 15 dxe5 0-0-0 16 ':a2 also looks good) 14 a5 b) 13 'ife3 again commits the queen rather (D). early. 13 ... 0-0-0 14 a5 ':he8 15 :fcl ~b8 16 tbh4 tbd6 17 i..f1 e5 gave Black good central counterplay in M.Gurevich-Gulko, Brussels (rapid) 1992. B c) 13 tbd2 (D) has enjoyed some popularity, but it clarifies the position rather early, and this helps Black arrange his defences.

This is an important position in the pawn sacrifice line. Both sides have made all natural moves so far. White must find a way to increase the pressure on the queenside, often with a timely push of the c-pawn combined with the a6 advance. Meanwhile Black usually does best to fight back in the centre. From here Black has 13 ... tbxd2 (13 ... tbdf6?! wastes time and jus tried: tifies White's play: 14 tbxe4 tbxe4 15 ':fel dl) 14 ... tbd6 15 .i.b3 .i.h5 16 h3 ':he8 17 0-0-0 16 'iWb2 ':he8 17 f3 tt)d6 18 i..f1 ~b8 19 a6 b6 18 c4 gave White a typical initiative in a5 tbc8 20 i..c5 gave White a strong attack in Ivanchuk-Lautier, Linares 1994. Kasparov-Bareev, Tilburg 1991) 14 'ifxd2 c5 d2) 14 ... ~b8 15 .i.e7!? (this is an interest makes it difficult for White to fuel his initiative; ing manoeuvre; the bishop has done its job on for example: the a3-f8 diagonal, so White looks to bring it to cl) 15 'ife3 0-016 dxc5 l:tfd8 17 ':fdl b6 18 the g3-square to bear down on Black's queen cxb6 tbxb6 was fine for Black in Piket-Anand, side) 15 ... .:de8 16 i..h4 ~a8 17 '6b2 (17 tt)d2 Amber Rapid, Monte Carlo 1993. allows the annoying 17 ... 'iff4 18 tbxe4 'ifxe4 c2) 15 dxc5 O-O-O! 16 'ife3 tbe5 has given exchanging queens) 17 ... f5 18 tbd2 f4 (trying Black good counterplay in a couple of games. to blunt White's dark-squared bishop; now He can follow up with ... :d5 or an invasion on 18 .. :iff4 can be met by 19 tbxe4 'ii'xe4 20 the d3-square. :cbl) 19 f3 tbxd2 20 'ifxd2 l:thfS 21 .i.f1 was DUTCH VARIATION: WHITE PLAYS 9 1i'e2 23

Ehlvest-Schwartzmann, New York 1996. White .i.d3! lDxc3 16 'iid2 .i.xd3 171i'xd3lDxa4 18 has good compensation, although it is probably d6led to trouble in Hiibner-Hertneck, Munich not enough for a real advantage against accu 1994) 15 dxe6 fxe6 16 .i.xe6 :he8 gave Black rate play. counterplay in Gershon-Arutinian, World Ju d3) With 14 ... :he8, Black does not sit idly, nior Ch, Erevan 1999. and instead prepares counterplay in the centre 141fe3 with ... e5. 15lDh4 (White ensures that he will White can also try 14 'iib2!? lDe4 15 a5lDdf6 have two bishops against two knights and gains (15 ... lDd6 16 .i.xe6!) 16 lDe5, as in the game time by chasing off the e4-knight; he can also Hiibner-Beliavsky, Munich 1994. Black is two flick in 15 a6 b6 before playing 16lDh4 but there pawns up, but his position still looks very shaky. is no need to commit the queens ide structure 14••. lDe4 just yet, while 15 'ifb2 ~b8 16 lDh4 is another Black would like to play 14 ... lDd5, but this possibility) 15 ... lDd6 16 .i.b3 (the immediate brings nothing but trouble after 15 .i.xd5 cxd5 16 lDxg6 hxg6 17 a6 b6 18 .i.b3 lLlf5 19 1ff3 16 lDg5! (16 l:ac 1 is also good), threatening to lDf6 20 .i.a4 lId5 21 g3 e5 gave Black coun detonate on e6. terplay in Vigorito-Stamnov, Las Vegas 2001) 15lDe5 16 ... e5 and with 17 1ff3 White hopes to create 15 lLlg5 looks too ambitious after 15 ... lLldf6, some pressure against the f7 -pawn, although when the sacrifices on e6 do not work. after 17 .. .lDf6! Black is well centralized and 15.•• lLlxe5 16 dxe51ib6 should have sufficient counterplay. If Black grabs the pawn with 16 ... 1i'xe5 We now return to 13 l:fel!? (D): White should not try to win a piece with 17 f3 because of 17 .. :ii'c3, but instead play 17 .i.d3 f5 18 .i.xe4 fxe4 19 :lad1 , when Black has a lot of trouble with his king and the presence of op B posite-coloured bishops just helps White's at tack. 17 'iff4 (D)

13••• lLlxc3 Now that Black has covered the b7-pawn, this second pawn-grab is critical. Of course, the natural 13 ... 0-0-0 is also possible. 141fb2 lIhe8 15 a5 e5 16 ]::tabl (with threats involving .i.d6) 16 ... c5 17 .i.f1 gave White compensation in Beliavsky-Akopian, Novosibirsk 1993. Re 17... 0-0-0 treating the c4-bishop to f1 has its points, be White has a strong initiative for the sacri cause White may play g3 and .i.g2, taking aim ficed pawns, so Black correctly offers back at Black's king position. some material. 17 .. :iWd4? 18 'iWh4, threatening Instead 13 ... c5 looks very risky, but it is mate on e7, was crushing in Lutz-Bareev, Mu probably not so bad. After 14 d5 0-0-0 (trying nich 1994. to keep the position closed with 14 ... e5 15 18 :xe4 .i.xe4 19 'ili'xe4 1fd4 24 CHESS EXPLAINED: THE MAIN-LINE SLAV

Black heads for an endgame because his 32 C:;d4 c5+ 33 c3 Itc7 34 ':dl+ cJ;e7 35 king will feel much safer without queens on the g4 h6 36 h4 g6 37 :d2 c4 board, but his position is still rather unpleasant. A risky decision, giving White's king an en 20 'ii'xd4 Itxd4 21 Itcl b6 22 i.d6 ~b7 (D) tryway into Black's queenside. 38 ':'dl ':'c5 39 b4 ':'c7 40 ~c3 ':c5 41 b4 ':'c7 42 i.e4 White correctly declines the repetition. Black w is very solid but he has no counterplay. 42••• f5 43 gxf5 gxf5 44 i.D ~f6 45 h5 ':'g7 46~c5 White decides to go forward with his king. 46 l:td6 was another possibility. 46 ••• c3 47 ':'cl ':c7+ 48 i.c6 b4 49 b6 ':'c8 50 b7 :d8 51 xa6 ':'b8 52 i.b7 ':'d8 (D)

w With a rook and two pawns for the two bish ops, Black does not appear to stand so badly, but it is not easy to create open lines for the rooks or utilize his queenside pawns. White will create pressure against the black king and he can easily bring his own king to the centre. 23 n a6 24 e2 b5 25 i.b3 Keeping lines closed and daring Black to wreck his pawn-structure. 25••• l:td8 26 ~e3 l:t8xd6 This looks like the best practical decision. Af ter 26 .. .l:tg4 27 g3 intending f4, Black's rooks 53 i.c6 would both be locked out of play. 53 b5! :d2 54 a6 would have won rather 27 exd6 ':xd6 28 as ~c7 29 f4 l:td8 30 i.dl easily, but this move does no harm. ~d6 31 i.f3 ':c8 53.•. :b8 54 .i.b7 l:td8 lh,_lh,?? Not 31 ... c5 32 ':dl+ C:;c7 33 ':xd8 C:;xd8 34 A strange repetition and a lucky escape for i.b7. Black. 55 b5 still wins.

Game 4 Peter Leko - Boris Gelfand Tal Memorial, Moscow 2006

1 d4 d5 2 c4 c6 3 liJf3 liJf6 4 liJc3 dxc4 5 a4 double White's pawns. After 12 h3 .i.xf3 13 Jtf5 6 e3 e6 7 .i.xc4.i.b4 8 0-0 liJbd7 'ii'xf3 we have transposed to Kasparov-Bareev Black can also play 8 ... 0-0 9 'ii'e2 Jtg4, but in the note to Black's 12th move. after 10 ltd 1 (10 h3 iLxf3 11 'i'ixf3liJbd7 trans 9 'ii'e2 Jtg4 (D) poses to the main game) 10 ... liJbd7 11 e4 'ii'a5, This uncommon but solid line represents a Black does not get his queen to h5 in time to completely different approach. Now that White DUTCH VARIATION: WHITE PLAYS 9 'ife2 25

bishop to create a safe square for his queen, be cause the immediate 18 ... ttJd5 19 ~g5 'ii'e6? w loses after 20 ttJxb4) 19 ~xd3 ttJd5 20 .i.g5 'ife6 21 ttJxb4 ttJxb4 22 ~c4 ttJd5 23 l:dl l:ae8! allowed Black to develop normally and hold the balance in Topalov-I.Sokolov, Hooge veen 2006. 10... ~xfJ This is Black's idea. Instead 10... .i.h5 is pretty solid. 11 Adl 0-0 12 e4 can lead to similar play to Game 1, but here Black may find it diffi cult to challenge White in the centre, so he could quickly be lacking for space. After 12 .. :ii'e7 (D) White has: has moved his queen, Black tries to slow down White's play by trading off his bishop because White will move his queen again in recaptur ing. The exchange will allow Black to strike w back on the dark squares in the centre with ... c5 or ... e5. Compared to Game 2, Black is able to give up the bishop without compromising his own pawn-structure, but here White will be able to achieve the e4 advance more easily, be cause his queen will defend the c3-knight later ally from the f3-square. 10h3 Posing the question to the bishop is consid ered the most accurate response. White can also play 10 :dl but this allows Black to swing his queen quickly to h5: 10 .. :iia5 11 e4 (the end a) 13 e5lbd5 14 tiJe4 f5!? (14 ... h6 15 ttJg3 game after 11 .i.d2 'iib5 12 h3 .i.xf3 13 'ii'xf3 ~g6 is more solid) 15 exf6 ttJ7xf6 16 ttJg3 'ii'xf3 14 gxf3 is also fine for Black) 11. .. 'ii'h5 .i.xf3 17 'ifxf3 was Elianov-Bu Xiangzhi, Wijk (grabbing the pawn with 11. .. ttJb6 12 ~b3 aan Zee 2007. Black's pawn-structure is a little ~xc3 13 bxc3 'ilxc3 is also possible, but with loose, but all of his pieces are active so he the bishop-pair and a strong centre, White's should have sufficient counterplay. compensation is obvious). White can now al b) 13 g4 ~g6 14 .i.d3 (White gains space Iowan ending or sacrifice a pawn: and keeps the tension in the centre) 14 ... h6 a) 12 h3 .i.xf3 13 .xf3 1i'xf3 14 gxf3 0-0 (14 ... e5 fails to 15 dxe5 ttJxe5 16 ttJxe5 'ifxe5 15 ~e3 :fd8 16 ~e2 a5 was solid for Black in 17 f4 smothering Black's kingside; 14 ... .:tac8 is Kramnik-Bacrot, Turin Olympiad 2006. The more solid, although I prefer White after 15 a5 fact that even Kramnik could not grind out this because of his extra space) 15 ~f4 .:tac8 16 endgame speaks to the soundness of Black's :lacl ~a5 17 .i.g3 .i.b6 18 .i.bl was very position. pleasant for White in Pelletier-Bu Xiangzhi, b) 12 :'d3 is an enterprising choice from the Biel 2007. White has a big space advantage and Topalov laboratory. White keeps the queens on Black lacks counterplay. the board and fights for the initiative by means 11 1i'xfJ 0-0 12 ltdl ':c8 of a pawn sacrifice. 12 ... e5 13 h3 ~xf3 14 ':xf3 The idea of this move is to prepare ... e5 by exd4 15 g4! 'iic5 16 ttJa2 0-017 ':f5 'ike7 18 e5 putting indirect pressure on the c4-bishop. It d3! (Black returns the pawn by diverting White's will thus be more difficult for White to play d5 26 CHESS EXPlAINED: THE MAIN-LINE SLAV because after ... cxdS this bishop will be hang here, because after 14 ... cxdS White must take ing. with the bishop, and 15 i.xd5 'iWb6 16 i.e3 Black can also play 12 .....a5 13 e4 e5 14 d5 ..tc5 is equal. ltJb6 (14 ... i.xc3 15 bxc3 cxd5 16 i.xdSltJxd5 14... 1IVaS (D) 17 ':xd5 gives White the initiative) 15 iLb3 The most active continuation. 14 ... 1IVe7 is ~xc3 16 bxc3 cxd5 17 exdS (D), and now: also possible. White can play 15 dS (or maybe even 15 'iVf5!? again), because after 15 ... ltJb6 (15 ... cxdS 16 ltJxdS) 16 i.b3 cxdS 17 ltJxd5 ltJbxd5 18 i.xdS both 18 ... ltJxdS 19 :xd5 and B 18 ... .i.c5 19 iLg5 give White some pressure.

a) 17 .. .lHc8 doesn't look very logical, be cause the rook has little to do on a8. After 18 c4! ltJbd7 19 i.b2 'ii'b6 20 :abl White had the initiative in Krush-Khuzman, Montreal 2006. b) With 17 ... :ac8, Black hopes to blockade 15"f5 the c- and d-pawns, so White must play very An interesting new try. White brings some energetically, using his bishop-pair and passed pressure to bear upon the centre and creates an d-pawn to disturb Black. Here too 18 c4! is unusual line-up of queens on the 5th rank. Al best: 18 .. .ltfe8 (18 ... ltJxc4 19 iLg5 intending ternatives: d6 leaves Black very off-balance) 19 ..td2 'ii'a6 a) Mter 15 d5, 15 ... iLxc3 16 bxc3 cxdS 17 20 d6ltJbd7 21 i.e3 :c6 22 a5! (increasing the ..txd5 ltJc5 18 c4 turned out poorly for Black in scope of the bishops and setting up various Gulko-Lakdawala, USA Ch, San Diego 2004, pins) 22 ... ':xd6 23 i.a4 :xd1 + 24 1h.d 1 gave but Black can improve by throwing in 15 ... ltJb6 White tremendous pressure for the pawn in first. For example, 16 iLa2 iLxc3 17 bxc3 cxdS Kasparov-Bareev, Novgorod 1994. 18 exdS :xc3!? 19 .i.d2 :Xf3 20 .i.xa5 :f4 13e4 with chances for both sides in the ending. White can also play more simply with 13 b) 15 ltJe2 is interesting. The white knight ltJe4 but after 13 ... ltJxe4 14 'ii'xe4 'iia5 Black is heads to g3, when Black may come under pres well developed, and it is hard for White to cre sure on the kingside. Sliding across the board ate any kind of initiative. One reason for this with 'ilb3 may also become an option. After lies in the queenside weaknesses, most notice 15 ... :fe8 (the immediate 15 ... exd4 16 .i.xd4 ably on b4, which make it difficult for White to ltJe5 is also possible) 16 ltJg3 exd4 17 ..txd4 get his dark-squared bishop quickly into play. ltJe5 18 .i.xe5 :xe5 19 'ifb3 White maintained 15 'ii'c2 ltJb6 16 ..td3 g6 17 e4 ltfd8 was fine some pressure in Sashikiran-Morozevich, Biel for Black in Beliavsky-Ivanchuk, Munich 1994. 2004. 13... e5 14 i.e3 15 .•• exd4 14 dxe5 ltJxe5 15 'iif5 "'a5 gives Black no Black cooperatively heads into the endgame. trouble. The advance 14 d5 is less effective 15 ... g6 just creates weaknesses in Black's DUTCH VARIATION: WHITE PlAYS 9 .e2 27 position, but he could try holding the centre 21 .•. h5 with 15 ... :fe8. Black hopes to fix the dark squares on the 16 'iVxaS .i.xa5 17 .i.xd4 .:tfeS IS f3 (D) kingside. 22 g4 bxg4 23 bxg4 ttJg6 24 l:txdS! ':'xdS 25 g3 Now White is ready to push his f-pawn, so B Black must put the pressure back on e4. 25 ••• l:eS 26 :dl White takes over the d-file after all. 26 ... .i.c7+ 27 ~2 l:te7 28.i.n ttJf4 (D)

White has a pleasant advantage because he has the bishop-pair and more space. Even though it will take some time, it will be easier for White to advance on the kingside than it will for Black to push his queenside pawns, because ... c5 will greatly weaken the d5-square, which can be used by White's pieces. Probably Black should hold positions like this, but it is not a Black has realized his plan of controlling the pleasant task. dark squares, but it is only temporary. IS••• a6 29g5! It was certainly worth considering eliminat Otherwise Black would secure his knight by ing one of White's bishops with 18 ... .i.b6. playing ... g5 himself. 19 ~2 ttJe5 20 .i.e2 29.•• ttJh7? Leko was critical of this move, preferring 20 29 ... ttJ6h5 is better. ~f1 so that he has the option of fighting for the 30.i.xf4? f4-square with ttJe2. Leko missed the chance to play the amazing 20 .•• ':'cdS 21 ~e3 (D) 30 e5!, which breaks Black's control of f4. If 30 ... .i.xe5, then 31 .:td8+ ttJf8 32 .i.c5 wins, and 30 ... ttJe6 31 f4 is awful for Black. 30... ioxf4 31 l:tdS+ ttJfS 32 g6 B White tries a last trick. 32..• b5? 32 .. J:~d7! 33 .:ta8 l:td2+ would give Black enough counterplay. 33 .i.h3! fxg6 34:aS bxa4?! (D) Falling for a trick. 34 ... f7 was the best try, although after 35 a5! Black is still under pres sure. 35 ion! Essentially winning a piece. The rest is not easy, but Leko gets there in the end. 28 CHESS EXPLAINED: THE MAIN-LINE SLAV

6O ••• :c7+ 61 ltJc4+ ~5 62 I:txg3 g5 63 W+ ~e4 64 :f7 Ilc8 65 :g7 ~4 66 ~d3 :d8+ 67 'ite2 :e8+ 68 rM2 :b8 69 :£7+ e2 1-0

Conclusions

The main line of Game 1 can be reached with two different move-orders: 8 ... 0-09 'ii'e2ltJbd7 10 e4 i.g6 or 8 ... ltJbd7 9 .e2 i..g6 (daring White to sacrifice a pawn) 10 e4 0-0. After 11 i.d3 i.h5 12 e5lilit5, 13ltJe4 is not so danger 35.•• ~f7 36 i.c4+ ltJe6 37 ltJe2 i..d6 38 f4 ous, so these days White prefers the forcing ~6 continuation 13 ltJxd5 cxd5 14 1i'e3!. This leads Black gives up the piece immediately to to interesting play with chances for both sides. gather as many pawns as he can and free his po The old response 14 ... i.e7 looks rather un sition. After 38 ... i.c5+ 39 ~f3 ~f6 40 e5+ pleasant for Black after 15ltJg5, so 14 ... i.g6 or ~f7 41 :c8 Black will never escape the pin. 14 ... :e8 should be preferred, with an interest 39 e5+ i..xe5 40 fxe5+ ~xe5 41 i.xe6 he6 ing middlegame in store. 42 :xa6 :b7 43 ltJf4+ ~e5 44 ~e3 c5 45 If Black wants to prevent the e4 advance, ltJd3+ ~d5 46 :as ~d6 47 :xa4 ~dS 48 :a5 8 ... 0-0 9 'ii'e2 i.g6 is the most solid line at his ~d6 49 :&4 ~d5 50 'itdl c4 51ltJb4+ ~e4 52 disposal. After 10 ltJe5ltJbd7 I1ltJxg6 hxg6 12 ~c3 g5 53 ltJc6 g4 54 :xc4+

In this chapter, instead of trying to push in the centre with e4, White seeks to hunt down Black's f5- bishop with the manoeuvre ltJh4. This is most commonly played after 1 d4 d5 2 c4 c6 3 ltJf3 ltJf6 4 ltJc3 dxc4 5 a4.i.f5 6 e3 e6 7 ~xc4 Jtb4 8 0-00-0 (or 8 ... ltJbd7), although White also frequently plays 6 ltJh4, forcing Black to make an early decision. Usually Black cannot avoid the trade of knight for bishop, but he can generally choose how and where it is exchanged. The resulting middlegames are generally very strategic in nature and offer chances to both sides.

The Games

Game 5 (Moiseenko-Swathi) examines the structure arising after 1 d4 d5 2 c4 c6 3 ltJf3ltJf6 4 ltJc3 dxc4 5 a4 Jtf5 6 e3 e6 7 ~xc4 Jtb4 8 0-00-09 ttJh4 ttJbd7 10 ttJxf5 exf5. In the game Black plays a quick ... c5 to discourage e4, but White manages to play it anyway, based on a typical pawn sacrifice to open the position for White's pair of bishops. Black declines the pawn, but his pieces lose their coordination. After White gets Jth6 in, Black neglects his back-rank problems and loses material to some rather elementary tactics.

Game 6 (Sakaev-Kasparov) sees White play ttJh4 before Black castles. White is wary of playing ttJh4xg6 because it will open the h-file for Black's rook. When Black decides to preserve the bishop with ... .i.h5, White does not respond critically with g4 and he quickly loses the initiative. Kaspa rov's forceful play gives him a better ending, which he duly converts into victory. In the notes to this game we also examine lines with a quick 'ifb3 by White.

Game 7 (Carlsen-Gelfand) investigates the immediate 6ltJh4. Black responds with the provoca tive 6 ... .i.g4. After 7 h3 Jth5 8 g4 .i.g6 9 ltJxg6 hxg6 we have a typical case where White has the bishop-pair and space while Black has good development and a solid position. When White errs with 12 Jtn?, Black quickly seizes the initiative. White is quickly saddled with structural weak nesses and he is overrun by Black's active knights and rooks.

Game 5 Alexander Moiseenko - Ghate Swathi Montreal 2006

1 d4 dS 2 c4 c6 3 ltJf3 ltJf6 4 ltJc3 dxc4 5 a4 does, arise via the move-order 8 ... ltJbd7 9ltJh4 JtfS 6 e3 e6 7 Jtxc4 ~b4 8 0-0 0-0 9ltJh4 (D) 0-0 as well. This is a completely different approach from 9 ••• ttJbd7 the one we saw in Chapter 1. White is less inter Black allows the exchange of minor pieces ested in advancing in the centre and he grabs the on f5, drastically changing the pawn-structure. opportunity to corral Black's bishop. The posi Here the retreat 9 ... ~g6 is rather compliant. tion after Black's next move can, and frequently After 10 ttJxg6 hxg6 White can play 11 'it'c2! 30 CHESS EXPLAINED: THE MAIN-LINE SLAV

B B

lL'lbd7 12 Adl (D) with good chances for an ad- 13 e4 the position is unclear. White has the vantage. bishop-pair and space, but the g4-pawn looks rather out of place. Black can complete his de velopment with 13 ... lL'lbd7 or strike immedi ately at the centre with 13 ... c5. B 10 lL'lxfS exfS (D)

If we compare this position to Game 2, White's queen is on c2 instead of e2. This is a subtle improvement, because the c3-knight is al ready protected, and in fact we saw earlier that White often spends a tempo playing 'ii'c2 in the 9 This is an important pawn-structure in the 'ii'e2 lines. Also note that ... e5 can be difficult to Main-Line Slav. At first it looks like Black has achieve because then White could play 'ii'xg6. just ruined his pawn-structure and given White 9 ... ..tg4 is a popular alternative. After 10 f3 the bishop-pair as well, but matters are not so ..th5 (lO ... liJd5 11 fxg4 'ii'xh4 12 'ii'f3 lL'lxc3 13 simple. The f5-pawn helps Black to control e4, bxc3 ..txc3 is met by 14 ..ta3!, when Black's and in some cases it may advance to f4 to dis king is very uncomfortable) 11 g4 (D) we have rupt White's control of d4. If Black can get a a common situation in the liJh4 lines. bind on the e4- and d5-squares, he might even Black will acquiesce to the exchange of his be able to play for a kingside attack. It is also bishop, but he wants to lure White's pawns for not always so easy for White to develop his cl ward first. This is double-edged - White gains a bishop, in part because of the queenside weak lot of space, but he may become overextended. nesses created by 5 a4. White will often try to After 11. .. ..tg6 (l1. .. lL'ld5 12 lL'lg2 ..tg6 13 lL'la2 play f3 and e4, even as a pawn sacrifice. This ..te7 14 e4liJb6 is also possible) 12liJxg6 hxg6 can lead to a very strong initiative for White, DUTCH VARlATION: WHITE PLAYS 9 li::Jh4 31 but his 'hanging' central pawns may also be ... 'iWd7 to protect the vulnerable pawn on f5. come vulnerable. Because Black is slightly Now: ahead in development, sometimes he will be a) 12 .i.e2 is a bit passive. 12...... d7 13 ttJa2 the one trying to open the position with ... c5. (13 a5 ttJbd5 14 a6 b6 15 .td2 ':'ac8 16 :lfcl The structure is quite delicate for both sides - l:.fe8 17 ttJxd5 cxd5 18 "'d3 .ixd2 19 'ii'xd2 White may break with e4 or d5, while Black can h5 is quite solid for Black, and he even went on strike with ... c5 or ... f4. A lot depends on the spe to win in Filippov-Wang Vue, Moscow 2(05) cifics of each position, and both sides must take 13 ... .i.d6 14 b4 l:tac8 15 .td2 ttJe4 16 ttJc 1 extreme care with their next few moves. ttJd5 17 ttJd3 'ii'e7 18 :tabl .:tfe8 19 g3 g620 II1IVc2 ttJc5 h5 was Pomerleano-Vigorito, Philadel White attacks the f5-pawn and covers c3. He phia 2006. This is an ideal set-up for Black. He may then continue with l:.dl or f3, or both. In has a grip on the centre and attacking chances stead 11 'fin is another way of hitting f5, but it on the kingside. is much less flexible, because White cannot ex b) After 12 .td3 Black will indeed be forced pand in the centre. After 11. .. g6 12 h3 "'a5 13 to play 12 ... g6, but from d3 the bishop is less ttJe2 l:tae8 14 b3 ttJe4 Black had a nice grip on active (there is no more pressure on f7) and the the central squares in Polugaevsky-Ivanchuk, d4-pawn will be more vulnerable, making it Bie11989. more difficult for White to engineer an advance II••• g6 of his e-pawn. The bishop may also be mis This obvious defence of the f5-pawn is by placed because ... ttJd5-b4 becomes an addi far the most common move. Having given up tional possibility. 13 a5 ttJbd5 14 a6 b6 15 .i.d2 his light-squared bishop, Black does not object l:c8 16 'iib3 .td6 17 ttJxd5 cxd5 18 lUc 1 'iWd7 to erecting a light-squared pawn-chain. Never 19 g3 'iWe6 20

This is a typical idea for White to gain space and prevent ... a5. The pawn may even advance to a6 to weaken the light squares in Black's This move avoids creating weaknesses on camp, while .ta4 becomes a distinct possibil the kingside because Black will be able to play ity as well. 13 ... ttJbd5 14 f3 (14 .td2 :lac8 15 32 CHESS EXPLAINED: THE MAIN-LINE SLAV

ttJa4 ~xd2 16 'ifxd2 l:tfe8 17 ttJc5 ttJe4 18 under tremendous pressure. However, 12 ... 'ii'b6 ttJxe4 ':'xe4 19 g3 :'ce8 was solid enough for and 12 ... ttJb6 both put some pressure on d4 and Black in Ehlvest-Donaldson, Stratton Mountain are worthy alternatives. 2(03) 14 ... 1tfe8 15 ttJxd5 ttJxd5 (15 ... cxd5? 16 ~a4) 16 e4 (White tries to seize the initiative) 16 ... ttJf6! (Black keeps the f-file closed) and now White has tried: w cl) 17 ~e3 hl 'ifxe4 26 'iWxe4 ':xe4 27 ~xf7 .:te2 28 b3 ~e7 was soon drawn in Onishchuk-Rublevsky, Poikovsky 2007. c2) 17 1i'c4 .i.f8 18 .i.g5 (after 18 exf5 ':ad8 19

h 1 :e7 20 'ific2 a draw was agreed in Kramnik-Anand, Linares 2(00) 18 ... fxe4 19 13 hl fxe4 l:txe4 20 ~xf6 gxf6 21 :adl l:tae8! and White tucks his king away in anticipation of Black's piece activity maintained the balance in the centre opening up. The immediate 13 e4?! Lautier-Bareev, Enghien-Ies-Bains 2003. would be met by 13 ... ttJb6, attacking both c4 12 f3 (D) and d4. This double attack is frequently part of Black's defence in many positions in the Dutch Variation. Thus 13 .ta2 is another prophylactic move that White can try. He also has a couple of B aggressive options: a) 13 'ii'f2 covers the d4-pawn and swings the queen over to the kingside. After 13 ... c5, White removes his bishop from the glare of the c8-rook with 14 ~a2 (D).

This is the most ambitious. White hopes to make use of the rook on the f-file, where it may assist in an attack on the kingside. 12•• .l:tc8 (D) Black logically lines up his rook against White's queen in preparation for ... c5. Instead 12 ... :e8? allows 13 ~xf7+

but Black must be careful, because the advance 14•.• lLlb6 may come in the fonn of a sacrifice. For exam 14 .....b6 is another idea. White can retain ple, 14 ... a6 15 ~hl 'iWb6 16 'ii'h4 lUe8 (now the tension with 15 :d1 or 15 'ii'f2, or release it may be the time for 16 ... cxd4) 17 dxc5 lLlxc5 by 15 dxc5 lLlxc5 16 e4!? fxe4 17 .tg5 with 18 e4! and White developed a strong attack in sharp play. I.Sokolov-De Vreugt, Amsterdam 2001. 15 dxc5 b) 13 :dl would seem to indicate that White If White plays 15 'iff2 now, Black might wants to play in the centre. After 13 ...... e7, 14 even try 15 ... f4!? to fight for the initiative. e4!? is a sharp try by Onishchuk. White offers 15.. ':xc5 (D) up a pawn even though his rook has left the f file. Black can play: bl) 14 ... fxe4 15 lLlxe4 lLlxe4 16 fxe4 c5 (Black fixes White's pawn-centre, but the b4- w bishop is locked away from the kingside) 17 d5 'ii'd6 18 a5 a6 19 :n ':'ce8 (Black surrenders the exchange because 19 ... l2Je5 20 ~h6 gives White the initiative) 20 ~h6 lLle5 21 g3 f6 22 ~xf8 l:.xf8 23 :f4left Black with too little for the ex change in Onishchuk-Shirov, Poikovsky 2006. b2) 14 ... ~xc3 15 bxc3 fxe4 16 fxe4 and now 16 ... 'ii'xe4 17 'ii'f2 gives White compensa tion according to Onishchuk. This may not be so bad for Black, but there is also 16 ... lLlxe4(?), when Onishchuk claimed 17 :e1 (?) as winning for White, although 17 ... l:tfe8 holds (e.g., 18 Black's pieces are coming into play very ~d3?! 'iih4!). However, 17 ~a3 c5 18 l:.el quickly, but White's next move shows that :tfe8 19 .td3 looks strong; e.g., 19 ... lLldf6 20 Black will not be so well coordinated when the lte2 intending l:.ael. position opens up. 13••• c5 14.ta2 (D) 16 e4! 'ii'e7 White must meet the threat of ... cxd4. In Because recapturing with the rook on f5looks stead 14lLla2 ~a5 (14 ... lLlb6 is also good) 15 a bit funny, 16 ... fxe4 may seem more natural, dxc5 ':'xc5 (15 .....e7 16 b4 Ji.xb4 17 lLlxb4 but it is risky to open the f-file. After 17 fxe4 'iixc5 is similar) 16 b4 .txb4! is a typical trick. (17 Ji.h6!? also looks dangerous for Black) Mter 17 lLlxb4 'ikc7 Black wins back the piece, 17 ... lLlg4 18 h3 'ii'h4 (Babula-Haba, Karlovy and 18 .txf7+ :Xf7 is about equal. Vary 2005), Babula suggests 19 'iVe2 because 19 ... l:.h5? loses to 20 l:.f4. Instead 16 ... Ji.xc3 17 bxc3 "fIIc7 is also possi ble, when Babula gives 18 exf5 ':'xf5 19 .te3, B heading for d4. White clearly has some com pensation for the pawn, but Black's pieces are all reasonably well placed. 17"f11b3 Before capturing on f5, White tries to disturb the coordination of Black's pieces yet further. 17••• 1:.aS This is a strange way to protect the bishop. Instead 17 ... Ji.a5, or even 17 ... .txc3 18 bxc3 'ii'c7 transposing to the previous note, comes into consideration. 34 CHESS EXPLAINED: THE MAIN-LINE SlAV

18 it.h6 ':c8 19 exf5 .txc3 23 ••. 4Jce3 (D) Black is forced to part with the bishop, be 23 ... 'irc7 fails to the simple 24 "'xc4, and af- cause after 19 ... .:.xf5? 20 lIfel, Black's queen ter 23 ...... d7 24 'irxc4 ':'xc4 25 .li.xc4 White is overloaded trying to protect both the f7- will win the knight and be left with rook and square and the b4-bishop. two bishops for the queen. Therefore Black 20 bxc3 :xf5 21 ':'adl (D) must walk into a pin.

B w

White's bishops generate strong pressure 24':'xd5 around the king, so Black must be careful. 24 .li.xe3 4Jxe3 25 :d3 is even simpler be 21 .•• 4Jfd5? cause after 25 ... ':'e5 White has 26 :dxe3! l:xe3 This is a blunder. It was better to chase off 27 ':'xe3'irxe3 28 'iVxf7+ ~h8 29 "f6#. White's bishop with 21. .. .:h5 to prepare ... 4Jfd5 24 .. J~xd5 25 ':'xe3 'ii'a3 or ... 4Jc4. The knights need firm outposts if Black tries some back-rank tricks himself, they are to compete with the bishops. but they are not so effective. 22 as 4Jc4 26 h3 'ii'xb3 27 .txb3 Probably Black saw 22 a5 and was counting The black rooks are no match for White's on this, but his pieces are not as stable as they rook and two bishops. may appear to be. 27 ...1:.h5 28 .tf4 ':'b5 29 c4 ':'xaS 30 l:e7 23 l:fel ':f5 31 .li.g3 as 32 .:txb7 a4 33 .li.al ':'d8 34 Black has tactical problems because of his ':a7 1:.dl + 35 ~h2 ':'d2 36 lIxa4 l:xf3 37 c5 weak: back rank. cj;g7 38 c6 f6 39 c7 l:c3 40 ':'c4 1-0

Game 6 Konstantin Sakaev - Garry Kasparov European Clubs Cup, Rethymnon 2003

1 d4 d5 2 c4 c6 3 4JO 4Jf6 4 4Jc3 dxc4 5 a4 bishop around, because Black has not yet castled .tf5 6 e3 e6 7 .li.xc4 .li.b4 8 0-0 4Jbd7 9 4Jh4 and ... hxg6 would open the file for the h8-rook. If either the main line of Game I or the pawn 9 'ilfb3 (D) is another possibility for White. sacrifice of Game 3 lacks appeal to White, then 9 This is an attempt to exploit the fact that Black 4Jh4 is a popular alternative. However, in com has played ... 4Jbd7 instead of ... 0-0 (after 8 ... 0-0, parison with 8 ... 0-0 9 4Jh4 lines, here White 9 'ifb3 is relatively harmless as Black can safely may need to use a little more care chasing the reply 9 .....e7). DUTCH VARIATION: WHITE PlAYS 9 li:Jh4 35

so Black should consider keeping the bishop with 11. .. ..th5!? B b22) 10 ... .ltg4 11 f3 ~h5 12 g3 (12 g4 .ltg6 13 g5 ltJd5 14 e4 ltJ5b6 was Ki.Georgiev Bacrot, Turin Olympiad 2006; these positions are often difficult to assess - White has a lot of space, but his position could prove to be a bit loose) 12 ... 0-0 13 e4 tDb6 (the d4-pawn is hanging, so Black snags the white bishop) 14 .lte3 tDxc4 15 ~xc4ltJd7 is equal. Now we return to 9 tDh4 (D).

Now 9 ... 'ii'e7?! can be met by 10 tDa2! be B cause the b7-pawn is loose. Therefore Black has two options: a) 9 ... 'ifb6 is risky. 10 e4 (10 tDh4 is also possible, but the text-move is more critical) 10 ... .ltg4 (10 ... tDxe4? 11 tDxe4 .ltxe4 12 ~xe6! is good for White, but 10 ... .ltg6!? 11 ~xe6 fxe6 12 a5! .ltxa5 13 'ii'xe6+ d8 14 e5, while highly risky for Black, remains unresolved the oretically) 11 tDe5 tDxe5 12 dxe5 .ltxc3 13 'ii'xc3 (13 'iixb6 axb6 14 bxc3 tDxe4! 15 f3 b5 16 .lta2ltJxc3 17 fxg4 Ilxa4 18 lH2 b4 19 Ilc2 e7! gives Black more than enough for the 9 ••• .ltg6 piece) 13 ... tDxe4 14 'ii'e3! gives White good A common alternative is 9 ... 0-0 10 tDxf5 compensation for the pawn. Black's e4-knight exf5, transposing to the previous game. If Black is very unstable. does not mind this kind of structure, he can play b) 9 ... a5 is supposed to allow White to force 8 ... 0-0 or 8 ... tDbd7 to reach it against various a draw, but this may not be so. White has: ltJh4 lines. bl) 10 ltJa2 and now: lOh3 bll) 10 ... .i.e7 allows White to head for a This is to some extent a waiting move. White draw after 11 'iixb7 :b8 12 'iia6 :a8 13 ~xc6 secures his kings ide a bit before committing to ':'c8 with a perpetual attack on White's queen. the exchange on g6. There are several other In practice, White usually prefers 11 tDh4..te4 ideas: (11. .. .i.g6 is also fine) 12 ltJc3, when both a) 101ib3 transposes to note 'b21' to White's 12 ... .i.d5 and 12 ... ltJb6 have done alright for 9th move after 10 ... a5, while Black can also Black. play 10 ... 'iib6. b12) Black can try 10 ... c5!? to avoid the b) 10 f3 tDd5 attacks both c3 and h4, forcing draw. This is not such an unusual idea, because White to capture on g6. 11 tDxg6 bxg6 should after 11 ltJxb4 cxb4 Black has a good grip on be fine for Black, because White will be com the centre and White's dark-squared bishop is pelled to play g3 or h3 soon, further weakening difficult to activate. his kingside. b2) 10 tDh4 is often preferred if White wants c) 10 f4 looks ugly to me, but it is not so bad. to fight. Black has a choice: After 10 ... 0-0 I1ltJxg6 bxg6 we have a typical b21) 10 ... i.g611 g3 and now 11...1!ib6?! 12 structure, but with the move f4 thrown in. This tDxg6 hxg6 13 :dl has scored well for White, prevents ... e5, but also loosens White's position, 36 CHESS EXPLAINED: THE MAIN-LINE SLAV

so Black should aim for ... cS. A good example of matter), he must be well prepared to meet dS. this plan for Black will be seen in the next game. That is not the case here: IS dS! ~S 16 i..e2 d) 10 g3 is a typical waiting move. White exdS 17 ~xdS ~xdS 18 lIxdS was clearly protects the h4-knight 'just in case' and also better for White in Kasparov-Anand, Linares shields the h2-pawn along the h2-b8 diagonal 1993. White has the bishop-pair and the initia in anticipation of capturing on g6. Black can tive, while Black is especially soft on the light castle, wait with 10 ... aS, or preserve his bishop squares. with 10 ... iLhS. 11'iVb3?! e) Opening the h-file by 10 ~xg6 appears to The safe 11 i..e2 iLxe2 12 ~xe2 0-0 13 b3 be rather cooperative, but White is not really ':c8 14 ~b2 'iVaS IS ~f3 :fd8 resulted in a likely to get mated on h2. After 10 ... hxg6 11 h3 sterile equality in Zilberstein-Vigorito, Reno Black could castle queenside after 11. .. 'ii'c7 200S. with complicated play, or just play the simple 11 g4 is sharper. Now 11. .. ~dS hits the h4- 11. .. 0-0 considering the move h3 a somewhat knight, and then 12 ~g2 iLg6 leads to a com wasted tempo. Play would then be similar to plicated fight. There will be a typical battle of that of Game 2. White's space and initiative against Black's t) 10 i..e2 prevents any ... iLhS ideas, but is a compact, counterpunching position. White has bit slow. The bishop was certainly more active tried 13 i.d2, 13 ~a2 and 13 f4!? with varying on the c4-square. Mter 10 ... 0-0 11 ~xg6 hxg6 success. 12 'ii'c2 (D) we have a familiar structure. 11 ••. a5 (D)

B w

This position is similar to Game 2, except 12g4 that White has played 'ii'c2 and ~e2 instead of This seems a bit inconsistent after deciding 'iVe2 and :d1. After 12 ... .:c8, Black has a solid against this double-edged advance on the previ position and he can follow up with ... 'iVaS or ous move. However, it is not easy for White to ... 'iVe7, aiming for the usual ... cS or ... eS breaks. find a good plan, and the alternatives are not 10.•• i..h5! without risks: The compliant 10 ... 0-0?! 11 ~xg6 hxg6 12 a) 12 f4 is an advance that must always be 'iWc2! is a little better for White. His bishop is made with very careful consideration. The e4- still active on c4, so it is difficult to play ... eS square is weakened and a ... cS break will leave because of 'iVxg6. Black has trouble finding White's structure looking very shaky. After counterplay because it is not easy to break in 12 ... 0-0 13 ~f3 (13 g4 ~xg4 14 hxg4 1i'xh4 IS the centre. A very high-level example contin gxhS 1i'g3+ leads to a draw) 13 ... ~b6 14 i..e2 ued 12 .. .llc8 13 ':dl Wfb6 14 e4 cS?!. When cS! IS ~a2 lIc8 16 ~xb4 cxb4 Black's active ever Black plays this advance (or ... eS, for that pieces and central control prevented White from DUTCH VARIATION: WHITE PLAYS 9 liJh4 37 enjoying his bishop-pair and 'better' pawn structure in Piket-Anand, Amsterdam 1993. b) 12lLla2 i..e7 (both 12 ... ..td6 and the un B tried 12 ... ltJd5 are also attractive possibilities) 13 "xb7 :b8 14 'il'a6 (14 'ifxc6? :b6) may be playable for White, but it looks very risky. 12•.• i..g6!? Already there was a draw to be had after 12 ... ~xg4 13 bxg4 'ii'xh4 14 gxb5 'iVg4+ with perpetual check. Perhaps Sakaev was hoping for this, but Kasparov plays for a win. 13~g2 This reluctant retreat is far from ideal, but 13 ~xg6 hxg6 is not very attractive with the weak nesses on the h-file. 18... ~cS! 13••• 0-0 14 ~f4 eS! This fine move keeps White off-balance. Black is ahead in development and White 19lhd8 has several pawn weaknesses, so Black opens If White keeps the queens on with 19 'fIc2, up the position. The fact that White can obtain Black will start an attack with 19 ... 'ii'h4. After the bishop-pair at any time is of no comfort be 20 g2, both the straightforward 20... :ad8 and cause he has trouble developing properly. the sharper 20 ... l:tae8, intending to engineer ... f5 15 dxeS or ... ~e6-g5, are good for Black. White's king This activates Black's pieces. It was safer to position is just too loose. Exchanging queens play 15 ~xg6 hxg6 16 :dl, when White can may look safe enough, but now White comes shore up his king's position with i.f1 if neces under fire on the other side of the board. sary. 19••• ~xb3 20 l:txa8lha8 21 l:tbl :d8 (D) IS ••• ~xeS 16 i.e2 ~fd717lLlxg6ltJxg6! The knight eyes the h4-square and makes way for its d7 counterpart. Instead 17 ... hxg6 is met by 18lLle4, preventing ... lLlc5, and getting w fancy with 17 ... ~c5 18 'ii'c2 fxg6 allows 19 ~a2. 18 :dl (D) White can also play 18 ~e4 "h4 19 ~h2 (19 ~g2 'fIe7 and Black will play ... ltJh4+ with tempo) although his position still looks shaky af ter the aggressive 19 ... h5 or the calm 19 ... l:tad8. On the surface, White's position after the text-move looks quite reasonable. He has the bishop-pair and his weaknesses do not look too serious. Nevertheless, grabbing the bishop-pair has cost White time and the structural weak The queenside weaknesses created by White's nesses cannot be repaired. More importantly, 5th move are quite telling. White can hardly Black has a chance to prevent White from de move. This game is a great illustration of what veloping properly. Positions like this in the Slav may happen if White is unable to develop are not unusual and they have a common quickly. Despite the bishop-pair, the pressure theme: Black needs to cause White trouble be on the a5-el diagonal and d-file means that fore he gets organized. Often small tactics are Black is happy to play with knights in an open needed to prevent White from catching up. position. 38 CHESS EXPLAINED: THE MAIN-LINE SLAV

22 iLdl 22 ttJe4 ~e5 is no improvement for White. 22 ... ~xel 23 ':'xel ttJe5 24 ~b3? Instead, 24 ~e4?! ttJd3 25 lIbl ~xb2 26 iLb3 Sl.a3 27 ~g5 1lf8 does not help much, but 24 iLc2 keeps White in the game. 24 ... ~4 (24 ... Sl.xc3 25 bxc3 lId2 26 ~e4 is certainly better for Black, but White can probably hold) 25 :dl! :b8 26 ~e4 (26 :bl ~xb2) 26 ... ~xb2 27 :d7 gives White some counterplay. 24 ..• :d2 25 %te2 ~xe3! 26 bxc3 ~f3+! 27 ~n~(D) 28l:tel There is hardly any other move. Black's rook and knight coordinate wonderfully in the end With an extra pawn and the better minor ing. piece, the rest is simple for Kasparov. 28 ... ~g5! 29 :e2 32 ~d2 ~d5 33 ~b3 ~e7 34 ~d3 ~d6 35 After 29 el Ilb2 Black wins the h3-pawn. ~d4 f6 36 h4 h6 37 h5 ~e7 38 f4 ~e6+ 39 29••• ~e4 30 ~el Ilxe2 31 ~xe2 fuc3 ~e4 ~c5 40 ~e2 fua4! 0-1

Game 7 Magnus Carlsen - Boris Gelfand Tal Memorial, Moscow 2006

1 d4 d5 2 e4 e6 3 ~f3 ~f6 4 ~c3 dxe4 5 a4 a) 6 ... e6 usually leads to the position reached ~f56~h4(D) in Game 5 after 7 ~xf5 exf5 8 e3 ~b4 9 ~xc4 0-0 lO 0-0 ~bd7. This is a safe choice if Black feels comfortable in this kind of structure. b) 6 ... Sl.g6 is too compliant. After 7 ~xg6 B hxg6 White has a pleasant choice between 8 e3 and 8 e4. c) 6 ... iLc8 invites a repetition after 7 ~f3, but White can also play 7 e3 e5 8 .i.xc4 exd4 9 exd4 with a structure more reminiscent of a Petroff or Queen's Gambit Accepted. d) 6 ... ~d7 (D) is one of Black's more ambi tious continuations. The bishop may be a bit passive here, but it is still developed and may help support ... b5. The white knight also re mains out on a limb. White has tried: dl) 7 e4 e6 S ~f3 (ifS ~xc4, then S... ~xe4 This sortie is quite popular. White forces a takes advantage of the loose knight on h4) S... b5 decision from Black regarding his bishop. 9 e5 ~d5 10 ~g5 ~b4 gave Black counterplay 6 ••• ~g4!? in Ftacnik-Azmaiparashvili, European Ch, Sil This provokes White to advance his pawn in ivri 2003. order to justify the knight's lunge to h4. Black d2) 7 e3 is more restrained, but it is also has several other moves as well: more stable. White will play for ttJe5 and 'iff3, DUTCH VARIATION: WHITE PlAYS 9 'tJh4 39

w w

exerting pressure on both the h I-aS diagonal he could also strike in the centre with 10 ... e5!? and f7. After 7 ... b5 S ttJf3 e6 9 ttJe5 a5 10 'iVf3 immediately) 11 ..txc4..tb4 12 f3 c5! 13 dxc5 :a7 11 g4 h6 12 h4 White had an initiative for 'tWa5 14 ..tb5+ ttJfd7 15 ..te3 0-0 16 "'e2?! the pawn in Moiseenko-Gagunashvili, Euro ..txc5 was at least equal for Black in Ward pean Ch, Silivri 2003. Hillarp Persson, St Helier 1999. d3) 7 g3 is another idea. White supports the b) 10 g5 ttJd5 11 e4 ttJb4 12 ..te3 e5 13 h4-knight and hopes to develop Catalan-like ..txc4 (after 13 dxe5 ttJd7! both 14 ..txc4 compensation. Black's most aggressive counter ttJxe5 and 14 f4 .i.c5! give Black good play) is Ivanchuk's 7 ... e6 S ..tg2 c5!. After 9 d5 (the 13 ... exd4 14 'iib3 'iic7 (14 ... ttJd3+ is interest point is that 9 .i.xb7 ..tc6 10 ..txaS ..txaS gives ing) 15 ..txd4 ttJSa6 16 .i.xa6 ttJxa6 17 0-0-0 Black good counterplay) 9 ... exd5 10 ttJxd5 ttJc6 ttJc5 IS 'iic2 ttJe6 19 ..te3 ..td6 20 ttJe2 a6 11 'iic2 ttJxd5 12 .i.xd5 ttJb4 13 'iie4+ ..te7 14 liz-liz Ponomariov-Grishchuk, Calvia Olym ..txc4 0-0 Black was better in Nyback-Ivan piad 2004. chuk, European Ch, Antalya 2004. 10... e6 11 ..txc4 7h3 White would like to fianchetto his bishop by This is the most consistent choice and is by 11 ..tg2 with the hope of recapturing the pawn far White's most common move. He avoids with 'ifie2xc4, but Black can immediately head weakening his kingside and creates the possi for the weakened d3-square: 11. .. ttJa6! 12 'iVe2 bility of developing his bishop on the long di ttJb4 13 0-0 (not 13 'iixc4? ttJc2+) 13 ... ttJd3 agonal. After 7 f3 Black can play 7 ... ..th5 S g4 and Black will play ... ..tb4 with a good game. ..tg6 or consider retreating the bishop with 11 ... ..tb4 (D) 7 ... ..tcS or 7 ... ..te6!? to leave the h4-knight offside. 7••• ..thS 8 g4 ..tg6 9 ttJxg6 White could also delay this exchange with 9 w ..tg2 e6 10 0-0, although it is not apparent how White derives any real benefit from this. 9 ...hxg6 (D) 10e3 A solid choice, and the most popular move. White avoids any more weakening pawn-pushes and prepares to recapture the pawn on c4. Alter natives: a) 10 e4 is more aggressive, but it looks rather loose. 10 ... e6 (Black develops normally; 40 CHESS EXPLAINED: THE MAIN-LINE SLAV

Black can be pretty satisfied with the out come of the opening. Because of White's g4 lunge, it is not easy for him to castle in this oth w erwise typical Slav structure. Black can con tinue with ... 'ii'a5 and ... lbbd7 before deciding where to put his king. 12.tO? Perhaps fearing his opponent's preparation, Carlsen deviates from his previous games. How ever, this artificial move is simply too ambi tious. White wants perfect development with .i.g2 and 0-0, but after Black's reply White cer tainly regretted this decision. Other moves are more sensible: a) 12 ~f1 makes sense - White tucks his 18 gs?! king away. 12 ... lbbd7 13 g2 'ii'a5 is about This does not help. Instead of pushing pawns, equal. White should at least develop with 18 'iVe2 b) 12 'ii'f3 lbbd7 13 f1 "a5 14lbe4 tDxe4 IUd8 19 .i.d2, even though Black holds the ini 15 'i¥xe4 0-0-0 gave Black an active position in tiative here as well. Shipov-Sakaev, Russian Team Ch, Sochi 2004. 18••• :rd8 19 'ii'e2 .i.xc3 20 bxc3 lbds (D) c) 12 .i.d2 lbbd7 (or 12 .....a5) 13 g5 lbd5 14 e4lb5b6 15 .tb3 a5 16 "e21eads to unbal anced play. This position had been tried a cou ple of times by Carlsen. w 12.•• 'ii'ds! White is immediately punished for his back ward development. 13f3 After 13 :g1 White will manage to play .i.g2, but his king will not find a happy home. 13.•• 'ii'as 14 .tg2 cs! Gelfand is not interested in grabbing pawns. Instead, he develops rapidly, bringing his knight to its most active post on c6. This game is a good example of the side with the knight-pair opening the position quickly to exploit a lead in A positional nightmare. White's queenside development. pawns are split and the black knights are tremen 15 dxeS tDc616 0-0 0-017 f4 "xeS (D) dous. Black's development advantage means Black has beautiful classical development. that his rooks will take the c- and d-files, and his White still must find a way to get his queen king is much safer than White's. White's posi side pieces out. Just as in the previous game, tion creates a good illustration of the dangers of White's acquisition of the bishop-pair has left pushing so many pawns - if you lose control of him overextended and behind in development. the space gained by the pawn advances, numer This is not to say that grabbing bishops is a bad ous weaknesses may remain behind them. idea. Often the side with the bishop-pair must 21 'iVbs 'slow things down' in order to avoid an early White strains to activate his pieces. Both 21 accident. Only when the position is stabilized c4 lbc3 and 21 .i.d2 lba5 (heading to c4) are can the bishop-pair's latent power hope to be awful for White. exploited. 21 .....xc3 22 :bl DUTCH VARIATION: WHITE PLAYS 9 l'iJ.h4 41

Perhaps 22 ':'a3 1i'c2 23 'ifb3 (23 'iVxb7? 31 ••. a6 loses the queen to 23 ... :ab8 24 'iVa6 lbdb4) In a dominant position, Black takes yet an would have been a better chance. At least White other square away from White's floundering can move. king's bishop. Still, 31. .. ltJb3 is moreprecise. 22••• ltJd.b4 23 'ii'e2 l:.ac8 24 ~e4 ltJaS 2S After 32 l:.bl (or 32 ~b5 'ifc5) 32 ... ltJxe3! 33 ~b2 'iic4 26 'iVg2 (D) 'ii'xe3 l:.8d3 White can resign. 32 l:.f2 ltJb3 33 ':'cfi ltJb4 33 .. :iffc5 is even quicker. 34 'ii'e4 'it'xc6 3S 1i'xb4 l:.xf2 36 :xf2 'ikcl + B 37 g2 1i'xe3 38 1i'b7? 38 ~d6 was the only move to play on, but White's position is lost in any case. 38••• ltJd2 0-1 There is no good defence to ... ltJe4.

Conclusions The structure that arises. after the exchange 10 ltJxf5 exf5 is quite tense and gives chances to both sides. After 11 'ifc2, 11. .. g6 is the most common way to defend the f5-pawn, but Black It is rare for two bishops to be so outclassed must be very careful because if the position by two knights, especially in such an open posi opens up, his kingside weaknesses could be tion. The bishops occupy the long diagonals, telling. The manoeuvre 11. .. ltJb6 followed by but they attack nothing while Black's knights 12 ... 1i'd7 has become more popular recently, and rooks are ready to pour themselves into and this solid line would be my choice. White's position. If Black plays the move-order with 8 ... ltJbd7, 26 ••• ltJb3 then 9 ltJh4 is somewhat less forcing than it is Black plays for maximum activity. He could after 8 ... 0-0, because White is in less of a hurry also consolidate his extra pawn with 26 ... b6 or to capture on g6, thus opening the h-file for 26 ... ltJd5. Black. This gives Black the option of playing a 27 ~xb7 :d2 28 'ii'fJ :cd8 29 ~eS ltJcs quick ... i.h5, maintaining some tension. White White's light-squared bishop is almost out of can still hunt down the bishop with a timely g4 squares. advance, but this is double-edged because it 30 l:.bclltJc2 31 ~c6 (D) weakens the white kingside. When White plays a quick 'ifb3, the position becomes tense across the board, and the position can be difficult to evaluate. If Black does not like these murky po sitions, he can always leave the bishop on f5, heading to the structure of Game 5. When White plays 6 ltJh4, Black can again allow the capture on f5 with 6 ... e6 if he prefers that type of pawn-structure. Instead 6 ... i.d7 aims to hold on to the c4-pawn, while 6 ... ~g4 is a good choice. White must weaken himself with g4 in order to round up Black's bishop, and then the h-file is opened immediately. Black gets a very solid position without having to know too much theory. 3 Central Variation: Black Plays 6 ... ttJbd7 7 ttJxc4 iJlc7

This chapter introduces the Central Variation, 6 ttJe5. This is a very different approach from the classical development of the Dutch Variation. White would like to build a broad pawn-centre with f3 and e4. This is a very ambitious plan, and White may lag behind in development. The games in this chapter and the next see Black challenging the white knight immediately with 6 ... ttJbd7 (for 6 ... e6, see Chapter 5). After 7 ttJxc4, this chapter examines 7 .. :fic7, by which Black intends to open the position with ... e5. Usually White allows this by playing 8 g3, preparing not only the fianchetto of the fl-bishop, but also the development of White's queen's bishop to f4, where it may annoy Black's queen. The play in the Central Variation is usually very concrete, and this chapter is no ex ception.

The Games

Game 8 (Aronian-I.Sokolov) examines the older variation with I d4 d5 2 c4 c6 3ltJf3 ttJf6 4 ttJc3 dxc4 5 a4 .i.f5 6 ttJe5 ttJbd7 7 ttJxc4 'WIc7 8 g3 e5 9 dxe5 ttJxe5 10 .i.f4 ttJfd7 II 1.g2 f6. This line was first given a serious examination in the 1937 World Championship match between Alekhine and Euwe. Both players were happy to play either colour and refinements of both sides' play were developed. After 120-0, in our main game Black plays the modem 12 ... ttJc5, which had been championed by Morozevich. Aronian shows his deep preparation and sacrifices the exchange to generate a promising initiative. Although Black may be alright theoretically in this line, with the clock ticking, Sokolov was unable to deal with the mounting pressure and he quickly succumbed to Aronian's purposeful play.

Game 9 (Ki.Georgiev-Bu Xiangzhi) introduces Morozevich's brainchild 11. .. g5!?1t is mostly due to this tactical counterblow that 7 .. :fic7 has seen such a revival in the past decade. This move was quite shocking when Morozevich unleashed it against none other than Kasparov in 2000. Nowadays this move is not so surprising, but the final word has hardly been spoken. Our featured game sees Georgiev sticking to Kasparov's original antidote 12 ltJe3. This positional approach leads to very unusual positions where both sides have their trumps. After 12 ... gxf4 13 ttJxf5 0-0-0 14 'WIc2, Bu plays the interesting 14 ... ttJg6 and gradually assumes the initiative. This game serves as a good example of Black maintaining pawn tension while keeping a watchful eye on the centre. After one slip from Georgiev, Bu is relentless and wins a fine game.

Game 10 (Jobava-Grishchuk) delves into the tactical ramifications of 12 ~xe5 and 12 ttJxe5 in Morozevich's line. While the latter is not looking very dangerous, the former move especially leads to very complex play where both players must be well prepared. 10bava is a player re nowned for his home preparation, and Grishchuk is soon caught in a web of tactics. With oppo site-side castling, one slip is all it takes and 10bava crashes through with a direct attack on Black's king. CENTRAL VARIATION: BLACK PLAYS 6 ... 'tJbd7 7 ttJxc4 'iic7 43

Game 8 Levon Aronian - Ivan Sokolov Wijk aan lee 2006

1 d4 d5 2 e4 e6 3 tDf3 tDf6 4 tDc3 dxe4 5 a4 tDbd5 12 e4 tDb4 13 'ita (if 13 exf5 then .i.f5 6 tDe5 (D) 13 ... l:txd4 14 'ite2, when both 14 ... .i.d6 and 14 ... exf5 give Black good play) 13 ... l:xd4!, which gave Black the initiative in Muham mad-Khachian, Lindsborg 2004. B

This continuation generally leads to sharper play than 6 e3. White intends to play the ambi tious f3 and e4, creating a broad pawn-centre while locking Black's f5-bishop out of play. 8 ... e5 9 dxe5 tDxe5 10.i.f4 Black can allow this plan and hope to fight back White pins the knight in order to maintain tactically by playing 6 ... e6, which is considered some tension. Instead 10 tDxe5 "ikxe5 11 .i.g2 in Chapter 5, or he can challenge White in the i.c5 is very comfortable for Black. centre immediately, as he does here and in 10... tDfd7 Chapter 4. Black acquiesces to moving the knight again 6 .•• tDbd7 7 tDxe4 "ike7 because after he secures the e5-knight, he may Black intends to play ... e5, opening the posi follow with ... J:d8 and ... tDc5. White's queen tion. A solid alternative is 7 ... tDb6, which is may feel uncomfortable because the b3-square discussed in the next chapter. is vulnerable to a knight invasion. 8 g3 (D) Black can also play 10 ... l:td811 'itcl i.d612 Because White has moved his king's knight tDxd6+ 'itxd6 13 .i.g2 and then either 13 ... 0-0 three times already, he must be wary of falling or 13 .. :ife7. Black has a solid position, but behind in development. This move allows White White can be happy with his effortless acquisi to fight for the initiative by preparing both .i.f4 tion of the bishop-pair. and the fianchetto of the king's bishop. Instead ll.i.g2 (D) 8 f3 e5 and 8 .i.g5 e5 are considered to be fine The tricky 11 e4 .i.e6 12 'ili'h5 can be met for Black. An interesting idea is 8 "ikd2 with the with the clever 12 ... g5!, exploiting both the point that 8 ... e5? loses to 9 dxe5 tDxe5 10 "ikf4! hole on f3 and the loose c4-knight. with a double attack on e5 and f5. Instead Black Suddenly White is the one with a develop should prefer 8 ... e6 or 8 ... tDb6, when the white ment lead, but Black's position is quite solid queen may end up looking silly on d2. One ex and White lacks an obvious pawn-break. ample is 8 ... tDb6 9 tDe5 e6 10 f3 O-O-O! 11 a5 11... f6 44 CHESS EXPLAINED: THE MAIN-LINE SLAV

the pieces he does have out are active and he may play ... 1ldS with tempo. White must play B very deliberately and strive for active play, since otherwise his queenside weaknesses may cause him some problems. Other moves have proved somewhat lacking: a) 12 ... iLe6 (there is no need for this volun tary retreat) 13liJxe5 fxe5 14 iLe3 iLc5 15 'ifcl i..xe3 16 'ifxe3 is pleasant for White. b) 12 ... :dS 13 'ifc1 JLe6 (13 ... i.e7 14lbxe5 lbxe5 allows 15 lbd5! and 13 .. :iVbS? 14 ttJe4 iLe7 15 'ii'c3 0-0 16 :adl iLe6 17lbxe5liJxe5 IS lbg5! gave White a big advantage in Euwe Alekhine, World Ch match (game 20), Amster This is a very sturdy continuation. Black sim dam 1935) 14lbe4! iLb4 (both 14 ... iLe7 15 a5 ply reinforces the e5-knight. Instead 11 ... iLe6 a6 16 :dl 0-0 171Wc3 and 14 ... i.xc4 15 'ifxc4 12 liJxe5 liJxe5 13 0-0 .i.e7?! (in an earlier lbxc4 16 iLxc7 ':'cS 17 i.f4 are also very pleas game, Alekhine had played 13 ... f6 14 "c2"a5 ant for White) 15 a5 0-0 16 a6 gave White the as Black, and this gives Black chances to equal initiative in Euwe-Alekhine, World Ch match ize according to Kasparov) 14 1Ii'c2! :dS 15 (game 1), The Hague 1937. :fdl 0-0 16 liJb5 gave White the initiative in We now return to 12 ... liJc5 (D): Alekhine-Euwe, World Ch match (game 1), Amsterdam 1935. The sharp alternative 11. .. g5!? is considered in the next two games. w 120-0 Resolving the tension with 12 liJxe5 should not be dangerous. Black has: a) 12 ... liJxe5?! 13 a5 a614 0-0 i..d6 15lbe4 .i.e7 16 'Wb3 iLg6 17 ':'fdl i..f7 IS 'ii'c3 0-019 lbg5! gave White a large advantage in Van Wely-I.Sokolov, Dutch Ch, Rotterdam 1995. We shall see this trick again. b) Gaining time by 12 ... fxe5! looks better. Black's remaining knight can still head to c5 and his dark-squared play will at least make up for the slight weakness in his pawn-structure. Black is ready to play ... :dS, when the white This structure can arise in many different posi queen will not feel very comfortable. However, tions, and in general Black is active enough to it is White's move and he can hope to exploit hold the balance. 13 iLg5 (tIying to provoke his development lead before Black gets prop Black, though 13 iLe3 is also possible) 13 ... ltJc5 erly organized. 140-0 iLe7 (14 ... a5 is also OK) 15 iLxe7 'ilxe7 13lbe3 16 :a3 a5 17 'fIc 1 0-0 IS 'iVe3 ':adS was fine White focuses on piece-play. 13 e4 i.g6 is for Black in Kasimdzhanov-Sashikiran, Istan also important: bul Olympiad 2000. a) 14 'ife2 .i.e7 (14 ... i.f7 also looks fine) 15 12... liJc5 ':'adl 0-0 did not pose Black any particular prob This is the modem interpretation of 11. .. f6 lems in Bellon-Nogueiras, Santa Clara 1995. which was used a lot by Morozevich before his b) 14lbxe5 fxe5 15 JLe3 i.e7 16 b4 :dS 17 11 ... g5 creation. Black lags in development, but 'ifc2 (17lbd5 cxd5 IS iLxc5 iLxc5 followed by CENTRAL VARIATION: BLACK PLAYS 6.. .li:Jbd7 7 li:Jxc4 1kc7 45

... dxe4 is fine for Black) 17 ... lL1e6 gave Black good play in Iskusnykh-Morozevich, Russian Clubs Cup, Maikop 1998. He has a nice grip on w the d4-square. c) 14 i.xe5 gives up the bishop-pair to try to exploit the lead in development. This is risky because White will be weak on the dark squares. 14 ... fxe5 15 f4 (the consistent move, but it weakens the dark squares further and Black may play ... i.c5 with tempo) 15 ... :d8 (the immediate 15 ... exf4 has also scored well) 16'ifg4?! (16'ii'e2exf417 gxf41L1b318:adl i.c5+ 19 ~h 1 lL1d4 is also very comfortable for Black) 16 ... exf4 17 gxf41L1d3 18 rj;h 1 i..c5 was already much better for Black in Sakaev 15 ... a6! he will not have to worry about White Morozevich, Kishinev 1998. White has too opening lines with b5. After 16 .a4 i..d6 17 many weaknesses on the dark squares. lL1ed5 'fIf7 Black was fine in Aronian-Gelfand, 13 ... i..g6 Wijk aan Zee 2006. Black keeps his bishop on the bl-h7 diago Playing the immediate 15 b5 certainly looks nal so that White cannot play 'ffbl or 'ii'c2. the most consistent, but this does weaken the Trying to throw in 13 ... :d8 is not so good be dark squares and Black may be able to take ad cause after 141L1cd5! (a typical trick) 14 ... 'ifi'd7 vantage of this. After 15 ... lId8 16 ""3 Black 15 lL1xf5 'iVxf5 16 lL1e3 the bishop-pair gives has: White some advantage. a) 16 ... i.f7 is too slow. 17 i.xe5 fxe5 18 Retreating with 13 ... i.e6 is feasible though. bxc6 bxc6 19 ltabl gives White some initiative. After 14 b4 :d8 15 'ifbl (15 'ii'c21L1a6 16 b5 The position has opened up, and Black's pieces 1L1b4 17 'iVe4 i..c5 18 bxc6 bxc61eaves Black's are just not active enough. For example, after pawn-structure a little shaky, but all of his 19 ... lL1c5 20 'iWb2 i.g6 21 lIbdl i.e7 22 lIxd8+ pieces are very active) 15 ... lL1b3 (15 ... lL1a6 is Ji.xd8 231L1b5! cxb5 24 "xb5+ lL1d7 25 lIdl a6 also playable, though 16 b51L1b4 does not come 26 'ii'c6 Black could hardly move in Halkias with tempo) 16 lL1cd5! cxd5 17 'ifxb3 looks a Drenchev, Kavala 2005. little better for White. b) 16 ... lL1d417 'iVb2 i..c5 18 :fcl (note the 14 b4! trap 18 lL1e4? i..xe4 19 i.xe4 g5, when Black White must fight for the initiative. His devel wins a piece, while 18 i.xe5 fxe5 19 lL1e4 opment lead has grown, but he needs a way to should be met by 19 ... i..e7) 18 ...... e7 (Black re open the position for his pieces. Black's tempo moves the queen from both the c-file and the rary lack of control of b4 gives White the lever pin of White's f4-bishop) 19 bxc6 bxc6 20 lbe4 he needs on the queenside. i.b4 211L1c41L1xc4 22 ltxc4 c5 23 e3 lbe6 24 14... lL1e6 (D) lL1c31L1xf4 25 i.c6+ f8 26 ':xf4 'iVe5 27 ':c 1 ? Again, 14 ... :d8 can be met with 151L1cd5!. (27 ':xb4 cxb4 28 'fIxb4+, with compensation, After 15 ... cxd5 16 bxc5 i.xc5 171L1xd5 White was better) 27 ... a5 and the powerful bishops has the initiative. and control of the d-file gave Black excellent 15 'iWb3 play in Bacrot-Gelfand, Wijk aan Zee 2006. An interesting moment. White attacks the IS... i.f7 knight on e6 and avoids ... ':d8 ever coming After the obvious 15 ... lL1d4 White has 16 with tempo. By delaying b5, White tempo 'fIb2, when the knight is not so stable; for ex rarily denies the black bishop the c5-square.1t ample, after 16 ... .:d8? White has the typical is much less dynamic to advance with 15 a5, trick 17 lL1cd5!. The immediate 15 ... lL1xf4 16 because after Black fixes the queenside with gxf4 leaves Black facing an unpleasant check 46 CHESS EXPLAINED: THE MAIN-LINE SLAV

on e6 combined with the b5 advance. Black is Black has also tried 19 ... ~c5 here. After 20 too far behind in development. bxc6 0-0 (20 ... ~xe3 21 fxe3 'iixe3+ 22 ~hl 16 'fIbl! (D) 0-0 23 cxb7 is no improvement for Black) 21 cxb7 the monster b7-pawn is the most impor tant factor in the position, Zhao Xue-Gerasimo vich, European Clubs Cup (Women), Kerner B 2007. 20 ':'xd6! White plays with great energy. Every move adds fuel to the fire. 20••• 1i'xd6 21 bxc6 (D)

16••• tLlxf4 This is very natural, but it appears to be a mistake as White's initiative develops quickly. Alternatives: a) 16 ... ~e7 17 tLlf5 tLlxf4 18 gxf4 tLlg6 (18 ... tLlc4!?) 19 e3 (if 19 tLlxg7+, then 19 ... ~f8 20 tLlf5 tLlxf4) 19 ... 0-020 b5 l:tad8 21 ltcl 'ii'd7 22 as gives White a strong queenside initiative, E.Atalik-Haznedaroglu, Turkish Ch, Istanbul 21 ••. 0-0? 2006. It is understandable that Black wanted to get b) 16 ... :d8 17 lIcl (17 b5 ~c5!) 17 ... tLlxf4 his king to safety, but this allows his queenside 18 gxf4 tLlg6 19 tLlcdS and now 19 .. :iVd7?! 20 to be destroyed. It was better to try to hold 'ili'e4+ ~e7 21 tLlxe7 'iixe7 22 'ii'xe7+ tLlxe7 23 things together with 21. .. bxc6, when White b5 gave White a clear initiative in the endgame has: in Khuzman-Arencibia, Montreal 2006, but a) 22 tLlf5 'iVd7 23 'iib4 tLle7 24 tLld6+ (24 19 ... ~xdS 20 tLlxdS 'ii'd6! looks fine for Black. tLlxe7 'iixe7 25 .i.xc6+

The extra exchange is little consolation for Black. The b7-pawn and White's nimble minor pieces soon overrun Black's position. B 23••• ':abS 24 'iff5 (D) 24••• ltJeS Both 24 ... 'iib6 and 24 ... 'ifb4 are better tries, but Black was already in serious time-pressure. 25 'ife2! 'ifb6 Aronian indicates that 25 ... 'ii'b4 is better, but after 26 l:.bl 'ifaS 27 ltJd6 the position is still pretty miserable. 26 'ile7 l:[fd8 27 .:tel Threatening 'ilxbS followed by ':cS+. 27 .•• 'ifa6 2S ltJfS 'ilb6 If 2S .....xa4 White has 29 ltJe7+ followed White mops up. by ltJcS. 31 •.• iLeS 32 ':xdS ':xdS 33iLdS+ iLf7 34 29 'ifxbS ':xbS 30 ':eS+ 'ifdS 31ltJxa7 ltJe7 + f8 35 ltJee6 1-0

Game 9 Kiril Georgiev - Bu Xiangzhi Gibraltar 2008

1 d4 dS 2 e4 e6 3ltJf3ltJf6 4ltJc3 dxe4 5 a4 on c4. White already has a difficult decision to iLfS 6ltJeS ltJbd7 7 ltJxe4 'ife7 S g3 eS 9 dxeS make. ltJxe5 10 iLf4ltJfd7 11iLg2 gS!? (D) 12ltJe3 This positional move was Kasparov's origi nal choice. By removing the knight from attack with tempo, White forces Black's response. At w first this looks like a positional refutation, be cause White's knight will be very strong on f5 and Black's kingside pawns will be split. How ever, White has weaknesses of his own, espe cially on the dark squares on the queenside. The exchange of opposite-coloured bishops cre ates an unusual positional situation. Each side will be very strong on the colour of their re maining bishop, and this is enhanced by the pawn-structure. Both sides have their trumps and an interesting middlegame will ensue with both sides fighting for the initiative. This surprising move was unleashed by Al The tactical alternatives 12 iLxe5 and 12 exander Morozevich against Kasparov in 2000. ltJxe5 are considered in the next game. Although he lost the game, many strong players 12.•• gxf413ltJxfS 0-0-014 'ife2 (D) took notice, and Morozevich' s line has been White removes his queen from the d-file. Af taken up by such players as Shirov, Gelfand, ter the immediate 14 0-0, 14 ... ltJc5 15 'ii'c2 Grishchuk, Bareev and Van Wely. The tactical transposes to the note to Black's 14th move, justification of 11. .. g5 lies in the loose knight and 14 ... ltJg6 15 'ifc2 to note 'b' to White's 48 CHESS EXPLAINED: THE MAIN-LINE SLAV

15th move. Black could also try 14... fxg3 15 c) 16 ltJe4 fxg3 17 hxg3 ltJg7 18 ltJxg7 hxg3 h5!? now that White has committed to ~xg7 19 b4 h5 20 b5 h4 was unclear in Gyi castling kingside. mesi-Acs, Balatonlelle 2002. It seems that both sides have good play! d) The centralization of the white queen by 16 'iVe4! looks the strongest. 16 ... fxg3 17 hxg3 B a5 (D) leads us to what was once considered a critical position:

Now each side's dominance over a colour complex is clearly apparent. White has a firm grip on light squares such as f5 and e4, while Black controls c5 and b4. Black's pawn-struc ture may look a bit ragged, but a timely ... fxg3 and ... h5-h4 can do a lot to eradicate Black's However, a discovery by Nielsen, which was weaknesses. introduced by Johannessen, appears to have 14... ltJg6!? closed the door on this particular variation: 18 This flexible move is a relatively new devel ltJb5!! cxb5 19 axb5 ltJc5 (l9 ... ltJd7 20 ltxa5 opment. From g6 the knight can help support ~b8 21 l:.fal ltJb6 22 l:.a7 does not help, but the advance of the black h-pawn to h4, and 19 ... b6 20 ':'fcl ltJc5 may be a better try, al Black may be able to challenge White's hold on though White still has a strong attack) 20 'iie3 f5 with ... ltJe7 at some point. ltJg4 (after 20... ltJcd7, 21 'ifa7 ltJc4? 22 :tacl 14 ... ltJc5 15 0-0 ltJe6 (l5 ... fxg3 16 hxg3 a5 was winning for White in S.Porat-Gonda, Bu is also possible) reaches a common position. dapest 2005, and 21 l:fcl ~c5 22 b6 'irxb6 23 White has: ltxc5+ ltJxc5 24 1i'xe5 'irc7 25 lDe7+ ~b8 26 a) 16 a5 is always a double-edged idea. Often ltJc6+ bxc6 27 "xc5 also looks very good) 21 Black will play ... a5 himself to solidify his grip 1i'c3 ltJe4 (21. .. 'ife5 22 'ilff3 'iffe4 23 'irxe4 on the b4- and c5-squares. By pushing his pawn ltJxe4 24 ~xe4 left White with a healthy extra to a5, White gains some space and increases the pawn in L.Johannessen-Shirov, Bundesliga scope of his aI-rook, but the pawn itself may be 2004/5) 22 ~xe4 "xc3 23 bxc3 b6 24ltJd4 was come weak. 16 ... a6 (Black should almost always Shirov-Harikrishna, Foros 2006. Again, White react in this way both to fix the white a-pawn is a healthy pawn up. and to prevent its further advance, which would Black will have to find something here if greatly soften up the light squares around the 14 ... ltJc5 is to remain a viable option. black king) 17 ':'a4 fxg3 18 hxg3 b8 is unclear. lSl:cl b) 16 l:tadl ~c5 17 ltJe4 iLb4 18 'ii'cl White tries to create more pressure against (White wants to clarify the tension on the king the black queenside. Alternatives: side, but this costs a tempo) 18 ... fxg3 19 hxg3 a) 15 a5 is possible, but again, although ':'xdl 20 ':xdl :d8 21 ~h3 c;t>b8 22 'it>g2 a5 23 this advance is often tempting, it actually re b3 112-112 Kramnik-Morozevich, Astana 2001. duces White's options. Advancing with b4-b5 CENTRAL VARIATION: BLACK PLAYS 6 ... tiJbd7 7 It:ttc4 'ili'c7 49 and sacrifices with ltJbS are no longer realistic. IS ... a6 160-0 fxg3 17 hxg3 hS 18 l:fcl b8 is nice for Black. White played the desperate 19 w ltJbS? axbS 20 a6 in Rahman-AI Modiahki, Turin Olympiad 2006. Here the simplest is 20 ... ltJdeS, when Black should win. b) IS 0-0 as!? (1S ... fxg3 16 hxg3 should be compared with the note to Black's 16th move) 16 l:fdl b8 17 l:acl transposes to the main game. c) IS 0-0-0 is a completely different idea. White avoids the race of opposite-side castling and looks to give the play a more positional vein. IS .. .'~b8 16 ltJe4 i..b4 17 .:td4 as 18 :hd1 ltJdeS 19 ltJfd6 'ii'b6 20 ..th3 h5! gave This position is rather typical for the 12ltJe3 Black sufficient counterplay in Gelfand-Moro line. White totally dominates the light squares zevich, Russian Team Ch, Sochi 2004. while Black controls the dark squares. Black 15 ••• li>b816 0-0 (D) will normally play ... fxg3 and ... hS-h4 both to eliminate some weaknesses and to bother the white king. The next few moves see both sides strive for the initiative. B 19 l:.d4ltJde5 20 ltxd8+ ':'xd8 21 ltdl h5 22 e3 fxe3 Note that it is only because Black kept the tension between the f4- and g3-pawns that he has this possibility now. 23 ltJxe3 h4 24 ltJf5 hxg3 25 hxg3 ltJe7! By eliminating the fS-knight, Black is gradu ally able to reduce the effect of White's light square control while retaining his own trumps. 26 ltJxe7 iL.xe7 27 l:.xd8+ 'iixd8 28 ltJd2 'ii'd4 (D)

16•• .85 Black can also release the tension immediately with 16 ... fxg3 17 hxg3 hS. After 18 b4 (18 W ltJbS 'ifb6 19 ltJbd4 ltJdeS 20 .:tfd 1 h4 gave Black counterplay in Banikas-Wang Yue, Gi braltar 2(08) 18 ... h4 19 bS cxbS 20 "bl?! (20 ltJxbS 'ii'xc2 21 l:.xc2 is pretty even) 20 ... hxg3 21 ltJxg3 ltJf4 22 l:.fdl i..cs Black had the stronger attack in Pogorelov-Handke, Barce- lona 2004. 17 :Cdl 'iVb6 17 ... iL.cS!? is very solid. After 18 b4?! axb4 19ltJe4 i..a7 20 ltJfd6?! fxg3 21 hxg3 Black de fused White's play with 2L.. ltJcS! in Wood- ward-N.Pert, British League (4NCL) 2004/S. White must already be careful because Black 18 ltJe4 iL.b4 (D) is so well centralized. 50 CHESS EXPLAINED: THE MAIN-LINE SLAV

29liJf3?! 29 .i.e4 maintains approximate equality. 29••• 'fid3! Suddenly White's queens ide pawns are look ing vulnerable. 30 'fic3 30 'fixd3 liJxd3 31 b3 liJc 1 32 liJd2 .ltb4 costs White a pawn. 30... 'fidl+ 31 h2?! Perhaps 31liJel f6! 32 'ii'xa5liJd3 33 .ltxc6! is a better try because 33 ... 'fixel + 34 'fixel liJxel 35 .lte4 corrals the black knight. White cannot win it because Black has ... .ltb4, but it will be very difficult for Black to win the game. Instead 33 ... liJb4 is better, although after 34 42.•• b6 43 'ii'el cj;c7 44 'fie2 ~d6 45 'ilc2 .ltf3 'fixe 1+ 35 b4 52 ~h7 cS 53 'iln and-knight tandem gives him a strong attack. If 53 ~c2 then 53 ... liJc6 followed by ... liJd4 37••• 'ilb4 38 liJxd6 'fixd6 39 'ilxa5 liJg4+ and even ... ~c3! will win the b3-pawn. 40 h3 liJe5 41 ~e4 (D) 53... 'ild2+ 54 cat>gl 'fie3+ 55 c;t;g2 'iixb3 56 41 •.• 'fid4 'fif4+ liJc4 57 .ltg8 'ilb2+ 58 ~n 'fial+ 59 41. .. 'ili'd7 + 42 ~g2 'iVd4 is even better. ~g2? 42b3 This is just a blunder, but the position was Now Black cannot take the bishop because lost anyway. of perpetual check. 59...... a8+ 0-1

Game 10 Baadur Jobava - Alexander Grishchuk Calvia Olympiad 2004

1 d4 d5 2 c4 c6 3liJf3liJf6 4liJc3 dxc4 5 a4 .ltb4 18 'iWf6 .c2!, when White will start shed ~f5 6 liJe5 liJbd7 7 ltJxc4 'fic7 8 g3 e5 9 dxe5 ding material and his king and hI-rook are both liJxe5 10 ~f4liJfd7 11 .ltg2 g5 12 .i.xe5 poorly placed) 15 ... .i.d6 (D). This is White's sharpest continuation. White Black has good compensation for the pawn has another tactical alternative in 12 liJxe5, al due to his excellent development. His initiative though this is not looking too dangerous these often persists into the endgame, so Black fre days. Mter 12 ... gxf4 13 liJxd7 O-O-O! 14 'iWd4 quently offers an exchange of queens. Some ex (if White does not play this, Black will stand amples: well with his bishop-pair and good develop a) 16 'iWd4 ~c7! 17 'ilxd7+ :xd7 18 .i.e4 ment) 14 ... 'fixd7 15 'iWxf4 (the point of Black's ~xe4 19 liJxe4 and here both 19 ... :d4 (Rogo play is that 15 'ilxh8 leads to difficulties for zenko-L.lohannessen, Gothenburg 2004) and White after 15 ... 'iid2+ 16 ~f1 'ilxb2 17 l:.el 19 ... .lte5 (Erdos-N.Pert, Budapest 2(03) offer CENTRAL VARIATION: BLACK PLAYS 6.. .ti:Jbd7 7 tDxc4 'ii'c7 51

White must play sharply to justify giving up his bishop. Queenside caStling looks risky, but w White must develop quickly to fuel his initia tive. Other moves have not troubled Black: a) 14 tOe4 lLlxc4 15 'iWxc4 "ika5+ 16 tOc3 0-0-0 17 0-0 'ifb4 gives Black the initiative. b) 14 tOxe5 "xe5 15 'iixe5+ fxe5 16 0-0 0-0-0 17 J..e4 J..e6 was Wang Lei-Zhukova, Women's World Clip, Shenyang 2000. White controls the e4-square, but Black's bishop-pair more than makes up for this. White also has weaknesses on the queenside. b) 14 tOe3 J..g6 (14 ... J..e6 also looks OK) 15 0-0-0 'itb6 16 tOe4 J..e7 17 f4 'iWxd4 18 Black good compensation in return for the sac Itxd4 gxf4 19 gxf4 tOd7 20 tOd6+ J..xd6 21 rificed pawn. ltxd6 0-0-0 22 l::thdl l:he8 23 J..h3 ltxe3 24 b) 16 'iicl b8 17 0-0 'ii'e6! 18 'ii'xe6 J..xe6, 14•.. J..e7 Jobava-Khalifman, Bled Olympiad 2002. Here Grishchuk plays a solid developing move, too Black's bishop-pair combined with White's protects the f6-pawn, and prepares to castle. queenside weaknesses gives Black excellent Other moves are not without risk: compensation for the pawn. a) 14... tOxc4 releases the tension and gives 12•.• tOxeS 13 'iWd4 (D) White easy play. 15 'iixc4 J..d6 and now 16 J..e4 J..h3 17 'iWd3 J..e5 18 J..f5 J..xf5 19 'iWxf5 'ifc8 20"ike4 0-0 21 h4 gave White some initia tive in Jobava-Kiilaots, Istanbul 2004. White B could consider varying with either 16 h4 (to stop ... J..h3) or 16tOe4!?, when White's devel opment lead and light-square play give him a pleasant initiative. b) 14 ... :d8 involves a pawn sacrifice, but it is playable. 15 'iWe3 J..e7 16 ':'xd8+ (not 16 tOxe5 'iWxe5 17 'fIxa7?? 1ffxc3+! 18 bxc3 J..a3#) 16 ... ~xd8 17 tOxe5 'iYxe5 18 'iWxa7 J..a3! 19 ltd 1+

15... i..e6 Although this move is quite playable, Black w has also investigated 15 ... i..g6. From here the bishop takes the e4-square away from the white queen and Black often follows with ... 1fb6. The g6-bishop may be vulnerable to an h4-h5 ad vance, but in practice Black has not done badly. White has tried: a) 16 i..e4 1fb6 17 i.xg6+ hxg6 IS 'iVe4 'iVb4 (the immediate IS .. :ii'c5 is also possible) 19 lId4 was Adianto-Zhou Jianchao, Asian Ch, Cebu City 2007. Here 19 .. :.c5 looks alright for Black. b) 16 ttJe4 f5 (Black holds his own in the cl) 20 a2?! ttJe3 21 i..h3 tiJxdl 22 lIxdl coming tactical melee, although 16 ... 'iIb6 looks :d8 23 lIfl 'iVc4+ 24 al 'iWc5 25 i.e6?! like a reasonable alternative) 17 ttJc4 fxe4 18 (better is 25 'ii'g8 with the idea 25 .. :ii'e7 26 "'xe5 "'xe5 19lbxe5 i.fS 20 g4 i.e6 21 i.xe4 ltxf8+! forcing perpetual check) 25 ... 'fie7 26 i.f6 22 liJf3 i..xg4 23 lIdg 1 i.xf3 24 i.xf3 i..f5 i..xf5 27 l:.xf5 'fIg7 2S 'fixg7 i.xg7 gave 0-0-0 was equal in Ris-Al Modiahki, Gibraltar Black a better ending in Gelfand-Morozevich, 2007. Amber Blindfold, Monte Carlo 2005. c) 16 i.h3 is not very challenging. 16 ... 1fb6 c2) 20 al is a subtle improvement, avoid 17 "'xb6 axb6 was at least equal for Black in ing any checks on the a2-gS diagonal. 20... ttJe3 Izoria-Akopian, European Clubs Cup, Saint 21 l:td2!? (21 :hfl is also possible, when in Vincent 200S. stead of 21. .. ttJxfl 22 :Lxfl with an attack, d) 16 h4!? 'i!ib6 17 'ii'xb6 axb6 IS h5 i.f7 Black could try 21 ... ttJc2+ 22 ~a2 "c4+ 23 b3 19 f4 gxf4 20 gxf4liJd7 21lbfS ltJcs with an 'iic5) 21. .. ttJc2+? (21. .. ttJxg2 22 l:[d4 'iVf5 had unclear ending, Salvador-Abdulla, Lodi 2006. to be tried) 22 l:.xc2 i.xc2 23 :fl gave White 16 'iVe4 i..b317 :d2 0-0 too strong an initiative in Jobava-Carlsen, Eu Black 'castles into it'. This is the correct de ropean Ch, Warsaw 2005. cision because Black is able to activate all of his ISlbe3 (D) pieces. Instead 17 ... ttJd7 IS liJf5 liJcs 19 'ii'e3 Instead 15 ttJxe5 "xe5 16 "xe5 fxe5 is 'ife5 20 ttJe4 (20 i..e4 ttJxe4 21 tiJxe4 i..d5 22 about even, and 15 tiJe4 lidS 16 'ii'c3 ttJxc4 17 ttJed6+ i..xd6 23 liJxd6+ q;e7 got White no 'ifxc4 'iie5 IS :xdS+ ~xdS 19 lid 1+ xe6 27 :hdlleft White with a nagging edge in Sargissian-Bu Xiangzhi, Tegernsee 2007 due B to his control of the d-file and Black's pawn weaknesses and exposed king. 18h4gxh4? But this is taking things too far. Black should play the active IS ... i..cS! 19 bxg5 (after 19liJg4 Black should just play 19 ... liJxg4 20 'ii'xg4 i..xf2 with an unclear position, as 21 bxg5 can be met by 21 ... f5 and 21 lId7 allows 21 ... 1Wxg3) 19 ... fxg5 20 ..th3 (D). This is a critical position for the evaluation of 12 i.xeS. Black's king position looks shaky, CENTRAL VARIATION: BLACK PLAYS 6 ... lCJbd7 7 4:1xc4 1i'c7 53

B w

but his pieces are very active and he has coun 23... l:txg7 24 ~xh7 f5 25 'ii'h5 ~h4 26 terplay on the dark squares. Black has: i.xf5+ 1-0 a) 20 ... ~h8?! 21 ~f5 ~xe3 and here instead of 22 fxe3 ~g8, as in Aronian-Bu Xiangzhi, Conclusions Stepanakert 2005, White should play 22 :'xh7+! 'ii'xh7 23 i.xh7 ~xd2+ 24 ~xd2 with a big ad The old method of play with 11 ... f6 remains a vantage. solid option for Black. White must play with b) 20 ... :f7 21it)f5 (21 i.f5 ~xe3 22 fxe3 great energy in order to make anything of his :'af8 23 g4 looks about even) 21. .. ~h8 22 f4 development advantage. A typical idea that gxf4 23 gxf4it)c4 24 :d4it)b6 was unclear in both sides must watch for is the pseudo-sacri Vitiugov-Sakaev, FIDE World Cup, Khanty fice ~! (the knight could come from c3 or e3 Mansiisk 2007. in response to a poorly-timed ... :d8). White c) 20... lhf2 was suggested by Scherbakov will usually have to try to play b4-b5 in order to and looks best. Not only does Black capture a open the position, but the timing of this break is pawn, but the e3-knight is also undennined. 21 critical. ~f5 h6! 22it)g4 (22 :'xh6 is met by 22 ... ~xe3, Morozevich's 11. .. g5 is alive and kicking. while 22lt)edl :'xf5 23 "xf5 IU8 241i'e41U6 12 ttJe3 gxf4 13 ttJxf5 0-0-0 14 'ii'c2 leads to gives Black excellent play for the exchange) very rich play. Both sides dominate one col 22 ... it)xg4 23 i.xg4 'ii'f7 24 :d7 :e8 25 :'xf7 our-complex, and with frequent opposite-side :'xe4 26 lhf2 lhg4 27 :'f3 ~g7 and Black has castling the play is very interesting strategi compensation in the endgame. cally. Right now 14 ... ttJg6 looks like a solid 19'ii'xh4 option for Black. White just has a strong attack for nothing. 10bava favourite, 12 i.xe5, remains White's 19•• ':f7 sharpest option. Giving up this bishop combined It is too late for 19 ... i.c5. After 20 it)g4 with queenside castling looks rather suspect it)xg4 21 "'xg4+ 'ii'g7 22 'ilf5 attacks the c5- positionally, so White must fight vigorously for bishop and threatens :d7. the initiative after 12 ... ttJxe5 13 'ii'd4 f6 14 20 lbf5 ~h8 21 i.e4 0-0-0. Following 14 ... i.e7 15liJe3, both 15 ... i.e6 White's pieces flood into Black's position. and 15 ... i.g6, while complicated, look ade 21 •. AafS 22 f4 it)c4 (D) quate for Black. 23it)g7! White's other option, 12 ttJxe5, is not look This clearance sacrifice destroys Black's ing too dangerous if Black plays accurately and defences. is willing to sacrifice a pawn. 4 Central Variation: Black Plays 6 ... ttJbd7 7 ttJxc4 ttJb6

In this chapter we look at the solid 7 ... tt)b6. Black avoids the complications of the previous chapter and aims for a more positional game. This approach has been championed by I.Sokolov and has been played by the likes of Anand, Shirov, Kasparov, and several of the new generation of Chinese grandmasters. Black chases White's centralized knight, and he will play ... e6 or even ... e5 depend ing on White's reaction.

The Games

Game 11 (Peng Zhaoqin-Liu Xianglin) sees White playing (after 1 d4 d5 2 c4 c6 3 tt)f3 tt)f6 4 tt)c3 dxc45 a4 ii.f5 6 tt)e5 tt)bd7 7 tt)xc4 tt)b6) the direct 8 tt)e5 a5 9 f3. Black challenges White with the obligatory 9 ... tt)fd7. After 10 lbxd7 tt)xd7 White has several approaches, most of which allow a quick ... e5 with good counterplay on the dark squares. Here White tries the radical 11 f4, which prevents ... e5 but weakens the e4-pawn. Black grabs the central pawn but never manages to get castled and White's escalating pressure eventually crashes through.

Game 12 (Bhat-Peralta) examines the traditional 9 g3. White plays for a modest edge, first com pleting his development before advancing in the centre. This approach has been considered to give White some advantage, but it is not looking too threatening these days. After an exchange of knights, Black gradually equalizes and eventually takes over the initiative when White rushes to re lease the tension in the position.

Game 13 (lbrahimov-Moradiabadi) looks at the modem 9 ii.g5. It is not so common for White to develop this way in the Slav. Both sides complete their development and an endgame is quickly reached which looks equal at first. Mter White's clever 22nd move (which had been invented by Gelfand) it becomes clear that White has a slight but nagging initiative. After a couple of careless moves by Black, he is doomed to passive defence and eventually succumbs to the pressure.

Game 11 Peng Zhaoqin - Liu Xianglin Jinan 2005

1 d4 d5 2 c4 c6 3 lbf3 tt)f6 4 tt)c3 dxc4 5 a4 Most of the lines are fairly calm, and while ii.f5 6 lbe5 tt)bd7 7 tt)xc4 tt)b6 (D) Black's position is very solid, there are also op If Black wants to avoid the complications of portunities to play for a win without taking big the Morozevich Variation and the piece sacri positional or tactical risks. fice of Chapter 5, this is a good line to play. 8 tt)e5 CENTRAL VARIATION: BLACK PLAYS 6.. .ti:Jbd7 7 li:Jxc4 l'Db6 55

10 as (10 llJxd7 llJxd7 11 e4 is also pleasant for White) and 9 as llJbdS 10 a6!? llJb4 11 axb7 w ltb8 12 f3! llJc2+ 13 ~ lDxal 14 e4 with a dangerous initiative. 9f3 White consistently tries to build a broad pawn-centre. The alternatives 9 g3, 9 i..gS and 9 e3 are considered later in this chapter. 9 ••• llJfd7 (D)

White keeps his centralized knight. Other moves are not very dangerous: a) 8 e3 e6 (8 ... lDxc4 9 ~xc4 e6 is also fine, but 8 ... aS?! 9 'iVb3! is troublesome) 9 as (it is rather late for 9 lDeS because Black can de velop comfortably with 9 ... ..tb4) 9 ... llJxc4 10 i..xc4 ..tb4!? is a logical suggestion by Atalik. The position resembles a 6 e3 Slav, but here Black has exchanged a pair of knights, so White's slight space advantage is not too trou blesome. This is the point of Black's play. It is impor b) 8 lDe3 i..g6 9 g3 (9 as can be met by the tant to challenge White in the centre before he simple 9 ... lDbd7 10 g3 e6 or the enterprising can play e4. This move again tries to remove 9 ... llJbdS 10 lDexdS lDxdS 11 1ffb3 e6 12llJxdS White's strong knight from eS, and Black may cxdS 13 'ii'xb7 :b8 14 "'c6+ 'ifd7 IS 'ii'xd7+ also get the opportunity to strike back in the '1t>xd7 with good compensation for the pawn) centre with ... eS himself, challenging White on 9 ... aS 10 i..g2 e6 110-0..tb4 12 f4 "'d7 is also the dark squares. fine for Black. A nice example was 13 i..h3 10 llJxd7 llJbdS 14 llJcxd5 llJxd5 IS llJc4 i..e4 16 llJeS White gains time to play e4 and avoids mak 'ifd6 17 llJd3 lDf6 18 lDf2 l:.d8 19 llJxe4 llJxe4 ing any concessions with his pawn-structure. 20 i..g2? llJd2! and Black wins a pawn, Kem There are a few alternatives, of which only one pinski-Zhang Pengxiang, Linares 2002. is dangerous: 8 ••• a5 a) 10 llJd3 is very tame. Black can choose Black holds up the white a-pawn to secure his between 10 ... i..xd3 11 "'xd3 eS and 10 ... eS 11 b6-knight. Continuing the chase with 8 ... llJbd7? e4 exd4 12 llJe2 i..e6 13 llJxd4 i..c4, as in is a well-known mistake due to 9 "'b3!, when Bozinovic-Atalik, Biel2oo6, in both cases with the double attack on b7 and f7 means that good counterplay. White can at least win a pawn. After 9 ... lDxeS b) 10 i..f4llJxeS 11 i..xeS f6!? (Black elimi 10 dxeSllJg4 11 "'xb7lDxeS 12 f4lDg6 White nates the pressure on g7 and is able to develop played a vigorous sequence to gain even more: harmoniously; 11 ... lDc4 is another possibility) 13 e4! ..td7 14 fSllJeS IS ..tf4 f6 16 i..xeS fxeS 12 i..g3 i..g6 13 e4 e6 14 i..e2 i..b4 IS 0-00-0 17 l:.d 1 gave White a crushing position in Kas 16 i..f2 fS 17 1i'b3 ~h8 18 eS (18 "'xe6?! i..f7 parov-Timman, Riga 1995. 19 "'eS i..c4 20 i..xc4 llJxc4 gives Black the Black can also play 8 ... e6, but this gives initiative) 18 ... lDdS was equal in P.H.Nielsen White a promising choice between 9 f3 llJfd7 I.Sokolov, Reykjavik 2001. 56 CHESS EXPLAINED: THE MAIN-LINE SLAV

c) 10 e4 is White's best alternative. Black a) 14 .. :~e7 IS tiJa2 .i.d6 16 'iVb3 .i.c7 17 has: l:ac 1 :ab8 18 l:tfdl gave White a comfortable cl) 10 ... .i.g6 is unusual, but it is sensible. edge in the game Navara-A.Ledger, British 11 '5b3 (11 tiJxg6 hxg6 12 ~e3 eS is pleasant League (4NCL) 200S/6. for Black, while 11 tiJxd7 tiJxd7 transposes to b) 14 ... l:c8 prepares ... cS, which will give the main game) 11. .. tiJxeS 12 dxeS e6 13 Jte3 him some counterplay against White's a-pawn. .i.b4 14 %:tdl tiJd7 was fine for Black in Jans After IS tiJa2 .i.e7 16 tiJc 1 cS 17 dxcS .i.xcs 18 sen-I.Sokolov, Dutch Ch, Leeuwarden 2002. .i.xcs ':'xcS 19 tiJb3 l:c8 20 'ifd4 .i.xa4! 21 c2) 10 ... tiJxeS (D) and now: ':xa4?! (21 'ii'xd8 :fxd8 22 tiJxaS is even) 21. .. tiJxa4 22 'ii'xa4 'ii'b6+ 23 1H2 l:c2 24 'ii'bS 'ii'e3 2S 'ii'd3 ':'xe2 26 'ii'xe2 'iWxb3 Black was a pawn up for nothing in Sarkar-Boumival, w Sturbridge 2007. 11 e4 .i.g6 (D)

c21) 11 exfS tiJed7 12 .i.gS g6 13 'iVc2 .i.g7 140-0-0 tiJf6 IS h4?! tiJfdS 16 tiJxdS 'ii'xdS 17 l:.el 0-0-0 18 Jtxe7 'iVa2! was perilous for White in Kuzubov-Rublevsky, European Ch, Warsaw 2OOS. c22) 11 dxeS 'ii'xdl + (11. .. ~e6!?) 12 ~xdl White has built up his pawn-centre in the .i.e6 13 <;t;c2 f6 (13 ... .i.c4 is also possible) 14 most thematic way - the purpose of the Central :bl! (a nice move to fight for the initiative; in Variation. Black has achieved something too - stead 14 exf6 exf6 is very comfortable for Black) namely the exchange of a piece - and now he is 14 ... fxeS (Black could try to resolve the tension ready to strike back in the centre with ... eS, if in the centre in a more favourable way with allowed. 14 ... tiJd7) IS b4! tiJc4 16 .i.xc4 (16 bxaS tiJxaS 12 f4!? 17 .i.e3 gives White compensation for the pawn This direct move has become popular. White but probably no more than that, Berlces-P.Kiss, prevents ... eS and seizes space. Others: Hungarian Team Ch 2004/S) 16 ... .i.xc4 17 bxa5 a) 12 h4 should be met with 12 ... hS!, when 0-0-0 18 ':d 1 l:xd 1 19 tiJxd 1 e6 20 ttJe3 .i.a6 it is not clear what White has achieved. Now 13 21 tiJg4 .i.d6 22 .i.b2 l:.d8 23 l:.dl hS 24 tiJxeS f4 (this further weakens the g4-square) 13 ... e6 .i.xeS 2S .i.xeS l:txdl 26 <;t;xdl g6 was soon 14 ~c4 tiJf6 IS fS? really takes things too far. drawn in Vallejo-Kasparov, Linares 2003. IS ... exfS 16 eS tiJg4 17 Jtf4.i.b4 was V.Mikh 10... tiJxd7 alevski-I.Sokolov, Vlissingen 2000. Although The alternative recapture 10 ... .i.xd7 is very the g6-bishop is immobile, White does not have rare, but it is playable. Black foregoes any ... eS enough for the pawn because of the strong plans, but his position remains quite solid. Af knight on g4. ter 11 e4 e6 12 .i.e3 Jtb4 13 ~e2 0-0 14 0-0 b) 12 dS is a radical attempt to disturb Black has: Black's development. After 12 ... eS (this is a CENTRAL VARIATION: BlACK PlAYS 6 ... liJbd7 7 ltJxc4 liJb6 57 typical idea - Black does not worry about his c12) 14 f4 (White need not avoid the ex pawns being split, but 12 ... 1Vb6 is also possi change of queens) 14 .....xd1 + IS ':xd1 ttJd7 ble) 13 dxe6 fxe6 Black's pawn-structure is 16 i.e2 f6 17 i.g4 lId8 180-0 i.e7 19 :d2 hS somewhat weakened, but he has good piece 20 i.h3 i.cs 21 %leI was a little bit better for play, especially on the dark squares. White has White in Nakamura-Bu Xiangzhi, Reykjavik tried: 2004. b1) 14 i.c4 'iVb6! (Black offers a pawn and c2) 12 ... 'ifb6 looks sufficient. Black avoids seizes the gl-a7 diagonal) IS i.xe6 ltJes 16 the exchange of queens and prepares to castle iie2 .:td8 17 i.e3 i.cs 18 i.xcs 'iixcs 19 queenside to bring further pressure to bear upon i.a2?! (this ends disastrously) 19 ... :d3 20 ~f1 the centre. 13 iid2 eS! and here: l:.f8 21 ttJd1 i.hS 0-1 Beliavsky-Tukmakov, c21) After 14 ':d1 O-O-O! Black puts imme Slovenian Team Ch, Bled 200 1. Detonation will diate pressure on the centre, because IS d5 i.cs follow on f3. 16 i.xcs ttJxcS gives Black good play, as shown b2) 14 i.e3 i.cs IS i.xcs ttJxcS 16 i.c4 in several games. 'iixd1+ 17

B B

Black should investigate the more solid alterna consistent) 19 lIabl! (not 19 c4? i.xc4 20 i.xc4 tives: 'iixd4+; perhaps Black was seduced by this tac a) 13 ... ltJf6 14 'ii'd3!? (14 e5 ltJe4 15 0-0 tical possibility) allowed White strong pressure i.b4 16 ltJe2 0-0 17 i.e3 'fIIe7 is pretty solid on the b-file and Black had problems surviving for Black, Beliavsky-Acs, Paks 2004) 14 ... h6?! in Shulman-Wojtkiewicz, Stillwater 2005. (14 .. :ifb6 looks like a better attempt to stir up 18:t'2 trouble) 15 f5 exf5 16 exf5 i.h7 17 0-0 was White takes a moment for some prophylaxis better for White in Cox-Shaw, British League by covering g2 and prepares 'it'e2 by avoiding (4NCL) 2005/6. any ... i.d3 skewers. b) 13 ... ltJb6 14 i.b3 'ii'h4+ 15 g3 'ifh3 16 18 ... h5 f2 i.b4 is interesting, Y.Vovk-Ni Hua, Cap Black fortifies his position on the kingside pelle la Grande 2007. light squares. 18 ... 0-0-0 is also possible, but 14 O-O! i.xc3 Black's king position is still not very secure - Declining the pawn with 14 ... ltJb6 15 i.b3 White can even swing a rook to b2 to put pres 'it'd7 is also uncomfortable for Black after 16 sure on Black. 'iff3! f5 (16 .. :1!i'xd4+ 17 i.e3 'it'd8 18 f5) 17 19 'ife2 'ii'f5 exf5 exf5 18 l:.e1 +, Trent-Shaw, British League Black is trying to play actively. 19 ... i..f5 is (4NCL) 2004/5. safer, even though it is not easy to see where 15 bxc3 i.xe416 i.a3 Black is heading. It is difficult to get his rooks This gives White good play, but he could into action, while White's play is very easy. also choose 16 'ii'e2! ltJb6 (16 ... i.d5 17 i.d3) 20 ':'el i.d3 21 'ii'b2! (D) 17 i.xe6 fxe6 18 'iixe4 'ii'd5 19 Ae1 'ii'xe420 ':'xe4 f7 21 IIbl with a strong initiative, J.Eriksson-Welin, Malmo 2003. 16••• ltJb617 i.b3 (D) B White has good compensation for the pawn. His piece placement is similar to that in Game 3, but here Black's king is not likely to find safety on the queenside. 17.•. 'ii'f6 Another idea is 17 ... ltJd5 intending ... f5. Black has also tried 17 ... h5 (by preventing g4, Black secures f5 for his bishop) 18 'iie2, when 18 ... i.d5?! (this gives up control of b1 and weakens the f5-square; 18 ... i.f5 is more CENTRAL VARIATION: BLACK PLAYS 6... l"i:Jbd7 7 tDxc4 l"i:Jb6 59

Threatening .i.xe6. 21 ..• lDd7 22 c4 1Iff6 Black tries to hold up a d5 breakthrough by W pinning the d4-pawn, but this fails. 22 ... lDf6 is well met by 23 :e5 'iih7 24 d5 with a strong initiative. 23"d2! ~f5 23 .....xd4 24 l:dl wins for White. 24~d6 White is patient. The immediate 24 d5 is also very strong. 24••• h4 2S c5 'ii'g6 Black should have prevented the coming breakthrough with 25 ... h3 26 g3 ~d8 although the position remains depressing. Giving White access to g3 only makes things 26 d5! l:.h5 27 dxe6 fxe6 28 l:.fe2lDfS (DJ worse. 29 i.c4 31 :g3! l:td8 32 'ii'b2 'iVf5 33 'iVxg7 White calmly improves his position and may 33 'ii'xb7 is also crushing. swing over to the b-file. The immediate 29 33.•. ':'xd6 34 cxd6 'ii'c5+ 35 l:tge3 :h7 36 ~xf8 xf8 30 ~xe6 iLxe6 31 ':'xe6 is also 'ii'xg4 'ii'xc4 37 l:.xh3 strong, but there is no hUrry. 37 ':'xe6+ lDxe6 38 ltxe6+ ~f8 39 'ii'f5+ 29••• ~g4 30 l:te3 ':'f7 40 ':'e8+! is even quicker. 30 f5! ~xf5 31 i.xf8 ~xf8 32 ':'xe6 i.xe6 37•.. ':'xh3 381i'xh3 "d4+ 39 "e31i'xd6 40 33 i.xe6 is instantly decisive. f5 hl 3O ••• h3? 'ili'd7 44 "a8+ 1-0

Game 12 Vinay Bhat - Fernando Peralta Badalona (open) 2006

1 d4 d5 2 c4 c6 3 lDfJ ltJf6 4 ltJc3 dxc4 5 a4 The fianchetto of White's bishop has long i.fS 6 ltJe5 ltJbd7 7 ltJxc4ltJb6 8 ltJe5 as 9 g3 been held in high regard and is considered to (DJ give White a slight advantage. In recent years Black has been refining his defensive methods and now it is difficult for White to obtain any meaningful initiative. B 9 ••• e6 This is by far the most popular move, but Black can also challenge White's e5-knight im mediately with 9 ... ltJfd7, as he did in the previ ous game. Mter 10 ltJxd7 'ili'xd7 (l0... ltJxd7 11 iLg2 e6 12 0-0 ~b4 13 e4 compares unfavour ably with the main game because White can play e4 in one go) 11 e4 (instead, 11 i.g2 .i.h3 12 ~xh3 'iVxh3 13 'ii'b3 :a6 14 d5 e6! is fine for Black, because 15 dxe6 'iVxe6 is even and 15 dxc6 can be met by 15 ... 'iii' g2 followed by 60 CHESS EXPlAINED: THE MAIN-LINE SlA V

.. :ii'xc6) 11 ... ~h3 12 ~xh3 'ifxh3 13 'ii'b3 l:ta6 b) 12f30-0(12... iLc2!?) 13e4~h714~e3 (D). tDfd7 is about even. After 15 'ii'e2 tDxe5 16 dxe5 tDd7 17 l:fdl 'fIe7 18 f4 l:fd8 19 1If2?! f6! Black even took over the initiative in Ves covi-Anand, Sao Paulo (rapid) 2004. w c) White should probably just continue 12 e3, which would transpose to the game after 12 ... 0-0.

Black's pieces are awkwardly placed on the queenside. but he intends ... e6 and ... ~b4. and White must act quickly if he is to achieve any thing meaningful. White has tried: a) 14 ~e3 e6 150-0-0 (15 dS ~b4 16 dxe6 'irxe6 17 'ilxe6+ fxe6 was equal in Gyimesi-Bu Xiangzhi, Bundesliga 2006n) 15 ... ~b4 16 d5 12e3 exdS 17 exd5 0-0 18 dxc6 bxc6 19 'ii'c2 ~xc3 Instead 12 iLg5 h6 13 iLxf6 'ifxf6 is not dan 20 bxc3 tDdS 21 ~c5 l:te8 was fine for Black in gerous, and 12 e4 iLxc3 13 exf5 iLxd4 costs Flumbort- Tischbierek, Zehlendorf 2007. White a pawn. b) 14 ~f4 e6 15 ~e5!? ~b4 (Black develops 12 ••• h6 aggressively; instead, 15 ... f6 creates a weak Black can also consider challenging White's ness on e6 for White to target by 16 iLc7 iLb4 knight immediately with 12 ... tDfd7!? After 13 17 0-0-0 with the idea tDe2-f4) 16 iLxg7 :tg8 e4 iLg6 14 tDxg6 hxg6 White has tried: 17 ~e5 'iWg2 18 0-0-0 iLxc3 19 bxc3! (19 'ifxc3 a) 15 e5 c5! 16 ~xb7 l:b8 17 iLg2 cxd4 18 'ii'xe4 20 :hel 'ii'd5 was unclear in Aronian 'ii'xd4 tDc5 gave Black good compensation for Carlsen, Candidates match (game 6), Elista the pawn in Gyimesi-Gagunashvili, European 2007 - Black has play on the light squares, but Team Ch, Plovdiv 2003. his rooks do not coordinate well) 19 ... 'ifxe4 20 b) 15 'ii'e2 e5 16 ~e3 exd4 17 iLxd4 'ire7 18 c4 'ife2 21 c5 tDd5? (21...'ifc4+ 22 'irxc4 tDxc4 %tfdl l:fe8 19 'ii'c2 "'e6 (both 19 ... iLc5 and is better, although even here White is for choice 19 ... :ad8100k sensible) 20 liJe2 'fi'c4 21 'fi'xc4 after 23 iLf4) 22 l:del! 'ii'h5 (22 .. :ii'xf2 23 tDxc4 and here instead of 22 b3? tDd2!, as in :hfl does not help Black) 23 'ilxb7 was crush Jobava-Felgaer. FIDE Knockout, Tripoli 2004, ing for White in Wang Hao-Caruana, Reykja when 23 .i.e3 fails to 23 ... tDxb3, White could vik 2008. have played 22 l:ac 1 ! with some advantage be 10 iLgl ~b411 0-00-0 (D) cause now 22 ... tDd2? 23 .i.e3 wins for White Black can also try 11 ... h6, which has been because there is no rook hanging on al. played by Anand. Then: 13 'fi'e2 i.h714 :dl (D) a) 12 e4 iLh7! looks OK for Black because 14••• tDfd7 it is not easy for White to set up his major It is natural to try to exchange off the strong pieces effectively due to the pressure on the d4- e5-knight, but Black could also play 14 ... 'fi'e7 and e4-pawns. 15 e4 l:fd8 (15 ... tDfd7 16 tDxd7 transposes to CENTRAL VARIATION: BLACK PLAYS 6 .. .t?Jbd7 7 '£Jxc4 ltJb6 61

B B

the main game) 16 .i.f4 :ac8. One possibility and he may break in the centre with a timely is 17 iLld3 ':'xd4!? 18 .i.e3 eS with interesting ... eS or ... cS. Firstly, he must decide how to ar play. range his rooks. 15 iLlxd7 17•• .Md8 White avoids losing time by exchanging This is by far the most common move, but knights, and he is also able to seize the centre. 17 ... :fe8 and 17 ... :ad8 have also been played. However, this gives Black enough room for his 18 d5 (D) pieces, and White should consider keeping White tries to blast open the position. After more tension with the retreat IS iLld3!? Then the more solid 18 ltac 1, Black can develop with Black has tried: 18 ... :ac8 or try to exert pressure on the e4- a) IS ...... e7 16 e4 eS 17 dxeSiLlxeS 18iLlxeS pawn with 18 ... iLlf6, which essentially forces 'ii'xeS 19 .i.e3 .i.cs 20 :acl .i.xe3 21 'ii'xe3 19 f3. lLlc4 22 'ii'd4 'iixd4 23 ':xd4 iLlb6 24 b4 gave White pressure in Gu Xiaobing-Li Shilong, Shandong 2007. b) IS ... .i.xd3 16 'ii'xd3 (16 :xd3 'iVe 7 17 f4 B f6 18 e4 eS 19 .i.e3 exd4 20 .i.xd4 iLlcs and now 21 .i.xcs \i'xcS+ 22 ~hl l:tad8 was fine for Black in Hillarp Persson-Sakaev, European Ch, Dresden 2007, but White could have tried 21 :ddl with a small edge) 16 ... 'ii'e7 and here both 17 \i'c2 eS 18 dxeS "xeS 19 e4 'ife6 20 .i.e3 ':fd8 21iLle2 'iVc4 22 :dc 1 'iVxc2 23 I1xc2 (Kacheishvili-Zhou Jianchao, Jinan 2OOS) and 17 e4lUd8 18 'iie2 eS 19 dS iLlf6 20 .i.e3 .i.cs 21 %:tacl .i.xe3 22 'iVxe3iLlc4 23 'ike2 cxdS 24 exdS iLld6 2S :el :e8 26iLlbS (Ki.Georgiev Megaranto, FIDE World Cup, Khanty-Mansiisk 18••• .i.c5 2007) leave White with some advantage. Black keeps the tension and refuses to ex 15••• iLlxd7 16 e4 "ike717 .i.e3 (D) tend the scope of White's g2-bishop. Both sides have played natural moves and The alternative is 18 ... exdS 19 exdS, and developed in a harmonious way. White has com then: pleted his development and advanced in the a) 19 ... ..txc3 20 dxc6!? .i.f6 21 cxd7 :xd7 centre. Black has a little less space, but he has and now 22 l:txd7 'ii'xd7 23 'ii'bS allowed White managed to exchange off White's eS-knight to maintain some initiative in Ruck-EBerkes, 62 CHESS EXPLAINED: THE MAIN-LINE SLAV

Hungarian Ch, Heviz 2003 because of the pres 22 ••• iLg6 sure on the b7-pawn, while 22 'ifb5 may be This slightly improves the position of the even better. bishop and also gives Black's king a little more b) 19 ... ':'ac8 and after 20 l:ac 1 liJe5 21 h3 room. ':'e8 22liJe4 (Bareev-I.Sokolov, Wijk aan Zee 23 :acl (D) 2004) Bareev points out that Black could play 22 .. .llcd8 23 dxc6liJxc6 with equality. Instead 20 dxc6 bxc6 21 .i.h3looks pleasant for White, even after Bareev's indicated 21. .. f5 22 i..g2 B i.g6. 19 dxe6 After 19 i.xc5 both 19 ... 'ifxc5 and 19 ... liJxc5 are about equal. 19..• fxe6 (D)

23 ••• liJb3 The porosity of White's queenside begins to show. 24 ':'xd8+ ':'xd8 25 l:dlliJd4 26 'iic4 b5! Black activates his dormant queenside majority. 27 axb5 cxb5 28 "d3 Not 28 liJxb5? liJf3+, winning the dl-rook. 28 ..• b4 29 liJb5 e5 We have seen this type of structure before in Black seizes the dark squares. the Central Variation. The slight weakness in 30 :al ~h7 (D) Black's pawn-formation is counterbalanced by the activity of all of his pieces. Black also has good control of the dark squares, which will be accentuated by the exchange of dark -squared bishops. 20.i.h3 White cannot develop an initiative, so he tries to put pressure on the e6-pawn, even though this does not lead anywhere. Instead 20 :ac1 iLxe3 21 'ifxe3 liJc5 22 iLfl liJb3 23 l:bl 'ifc5 24 'ifxc5liJxc5 25 f3 i.g6 26 ~f2 e5 27 i..c4+ iLf7 28 i.xf7+ xf7 gave Black a comfortable end game in Nayer-Wang Vue, China-Russia match, Ergun 2006. 20 ..• i.xe3 21 'ii'xe3 liJc5 22 "e2 Note that advancing with 22 e5 gives Black's This is a nice little move, removing the king remaining bishop more scope than it does from any checks on the a2-g8 diagonal or the White's. back rank. CENTRAL VARIATION: BLACK PLAYS 6 .. .11Jbd7 7 'l:ttc4 tiJb6 63

31.i.g2 White keeps getting pushed back. Not 31 :'xa5? liJc6. 33 'ife2 ':'c4 34 b3 ':'cl+ 35 ':'xcl 'ili'xcl + 36 31 .....c5 32liJxd4 ':'xd4 ~n a4 37 h4 axb3 38 h5 ~f7 39 'ii'f3 g8 0-1

Game 13 Rasul Ibrahimov - Eishan Moradiabadi Abu Dhabi (open) 2007

1 d4 d5 2 c4 c6 3 tt)f3 liJf6 4 tt)cJ dxc4 5 a4 We now return to 9 ~g5 (D): .i.f5 6 lLle5 liJbd7 7 tt)xc4 liJb6 8 lLle5 a5 9 ~g5 Because Black has been holding finn against 9 f3 and 9 g3, recently players of the white B pieces have tried a couple of different ap proaches. The text-move is logical enough - White develops his queen's bishop before deter mining how he will arrange his centre. Another fresh continuation is the apparently modest 9 e3. This looks harmless, but White has the pos sibility of pestering Black on the kingside with g4. Black has: a) 9 ... liJfd7 allows 10 liJxf7 xf7 11 'ili'f3 e6 12 g4, although even this is not so clear. b) 9 ... liJbd7 10 'iib3 liJxe5 11 dxe5 tt)d7 12 e4 ~e613 'ii'xb7 is not clear, but White's game 9 ••• h6 looks preferable. Black chases White's bishop and gives his c) 9 ... h6 is solid. After 10 g4 ~h7 11 ~g2 own a haven on h7. Instead 9 ... lLlfdS looks risky e6 120-0 .i.d6 13 f4 0-0 Black has a reasonable in view of 10 e4!? lLlxc3 11 bxc3 ~xe4 12 'iVb3 position. ~d5 13 c4 ~e4 14 'ife3, with a nice initiative d) 9 ... e6 is very natural, but the play can for White. The combative 9 ... g6 is also possi sharpen considerably. Mter 10 g4 ~g6 11 h4 ble, but I suspect White is doing well after 10 ~d6! 12 h5 ~e4 White can play 13 f3 .i.xe5 f3!. 14 fxe4 ~g3+ 15 d2 (Milanovic-Ve1cheva, The text-move looks the most reliable be Kavala 2(07) or 13liJxe4liJxe4 14 ~g2 (Mched cause White does not really want to take on f6 lishvili-I.Sokolov, Villarrobledo (rapid) 2007) anyway. By flicking in this move, Black creates in both cases with an unclear game. a haven for his bishop on h7 and he can also e) 9 ... g6 is Black's main continuation. It is play ... g5 if necessary. not very common for Black to fianchetto his 10~h4e6 king's bishop in the Slav, but here it is quite rea 10 ... liJbd5 is a solid alternative favoured by sonable. 10 .i.d3 .i.xd3 l1liJxd3 ~g7 12 'iVb3 Chinese players. In practice, White has headed 0-0 13 0-0 liJfd7 14 lLle2 "fic7 (better than for an ending with 11 'iVb3 'ili'b6 12 'ii'xb6 14 ... e5 15 dxe5 liJxe5 16 liJxe5 ~xe5 17 e4 lLlxb6 13 f3, although this does not look too 'ili'c7 18 ~e3 liJd7 19 f4 ~g7 20 e5 with a large threatening. 13 ... lLlfd7 (13 ... .i.c2 14 e4 e6 is advantage for White in Jobava-Nogueiras, Ha also reasonable) 14liJxd7 (14 e4liJxe5 15 dxe5 vana 2005) 15 e4lLlc8 16 f3 ':d8 was played in .te6 intending ... lLld7 looks OK for Black) Jobava-Shirov, Russian Team Ch, Sochi 2007. 14 ... liJxd7 15 e4 ~g6 16 d5 e5 170-0-0 Ji.c5 18 Then 17 Ji.e3 looks about equal. ~c4 f6 19 ~bl, Gelfand-Bu Xiangzhi, World 64 CHESS EXPLAINED: THE MAIN-LINE SLAV

Team Ch, Beersheba 2005, and now 19 ... 'it>e7 is equal according to Gelfand. 11 e4 ~h7 12 f3 B With ... gS in the air, White must safeguard his e4-pawn. 12••• ~e7 13 ~f2 0-0 The immediate 13 ... lbfd7 14 lbd3 0-0 IS i..e2 transposes to the game. 14 i.e2lbfd71Slbd3 (D)

An interesting idea, introducing the possibil ity of b4. The direct 22 c;t>f2 fS (Black is ready to play ... lbd7-cS) 23 ':'xdS+ ':'xdS 24 ':'a3 (threatening ':'b3) can be met by 24 ... :td4!, as pointed out by Gelfand. 22 ••• fS! Black must get his h7-bishop into play. The inferior 22 ... q.,fS 23 ':'xdS+ ':'xdS 24 b4! ':'d4 2S bxaS lbxa4 26 ttJd I! leaves Black in major difficulties, Gelfand-Felgaer, FIDE World Cup, White has a strong centre, but Black is Khanty-Mansiisk 200S. Black is essentially a slightly ahead in development and he can pre piece down in trying to defend his queenside. pare to strike in the centre. 23 ~2 fxe4 24 fxe4 (D) IS•• J:tc8 Black prepares ... cS. It is also possible to prepare ... eS with IS ... i.d6 160-0 'iJlic7. After 17 g3 eS IS ':'c 1 exd4 19 lbbS 'ii'bS 20 lbxd4 B lldS 21 b3 lbeS 22 'iffc2lbxd3 23 i.xd3 lbd7 Black had a solid, if somewhat passive position in Kasimdzhanov-Ki.Georgiev, FIDE World Cup, Khanty-Mansiisk 2007. 16 0-0 c5 17 dxc5 lbxc5 18 lbxc5 ..ixcs 19 ..ixcs :Xc5 20 'iffxd8 ':'xd8 21 lUdl (D) Both sides have played logically to reach this position. Despite the considerable simplifica tions, Black must be careful. The b6-knight is awkwardly placed and the h7 -bishop needs to get into play. 21 .•• ':'cc8 24•• ..:.t'8+ After 21 ... .:.xd 1+ 22 :xd1 ~fS 23 ':'d6 ':'c6 This is a somewhat odd decision. The rook 24 :dS+ ~e7 2S :bS llc7 26 ~f2 White re will have little to do on the f-file. Instead tained a nagging edge in Zhao Xue-Ruan Lufei, 24 ... 'itfS 2S llxdS+ llxdS 26 q.,e3 (26 b4 should Women's Zonal, Beijing 200S, although even lead to a draw after 26 ... ':'d4 27 bxaS lbxa4 2S here Black's position is hard to crack. lbxa4 ':'xa4 29 ':'xb7 ':'xaS) 26 ... eS 27 ~b5 2211abl! ':'d4 2S lln + 'ite7 29 ':'f3?! was agreed drawn CENTRAL VARIATION: BlACK PlAYS 6.. .llJbd7 7 ttJxc4 tiJb6 65

here in A.Grigorian-Kuzubov, Kirishi 2007. There was no hurry to play this. The prophy Black could have played 29 ... ii.xe4! 30 ltg3 lactic 36 :d3 was called for. (30 lDxe4 lDd5+ is Black's point) 30... lDd5+ 36 ••• ii.h7? 31 lDxd5+ ii.xd5 32 l:txg7+ CJtf6 when Black A better try is 36 ... ltc5, taking advantage of has some initiative. the unprotected e5-pawn in order to get in ... b5, 25 e3 lDc4+ with counterplay. This also looks suspicious. White's flexible 37 l:td6! knight will prove to be more nimble than Black's Now the rooks leave the board and the knight bishop. shows its superiority. 26 i.xe4 ltxe4 27 %:td6 37••• l:xd6 38 exd6+ d7 39 ~d4 i.e2 40 White's rook now causes some trouble in the lDxb6+ c3;xd6 41lDe4+ c6 42 g3 black camp. White is patient. The a5-pawn is not going 27••• lte8 28 :b6 ltb4 29 ltb5 anywhere. This is good enough to keep an edge, but 29 42 ••• g5 43 bxg5 bxg5 44 liJxaS+ ~d7 45 ltxb4 axb4 30 lDe2 looks even stronger. Black CJte5 i.dl 46 CJtd4 ii.e2 47 c3 i.dl 48 liJe4 will have trouble holding on to his front b CJte6 49 tLle3 ii.e2 50 ~d4 CJtd6 51liJe4+ We6 pawn. The h7-bishop simply does not partici 52 b41-0 pate in the fight. 29••• ltxb5 30 lDxb5 lte8 31 tLld6 l:tc6 Conclusions After 31. .. ltc2 32 b4! axb4 33 l:txb4 l:xg2 34 l:xb7 the black king and bishop are both un The solid 7 ... lDb6 SlDe5 a5 is holding up well. comfortable and the white a-pawn is danger The critical 9 f3 lDfd7 10 liJxd7 tLlxd7 11 e4 ous. ii.g6 can lead to very sharp play where Black 32 ltdl b6 33 b3 ~ 34 tLle4 c3;e7 35 h4 (D) seems to be holding his own. 12 f4 e613 ii.c4 is an interesting try. Here Black should probably investigate 13 ... liJf6 and 13 ... tLlb6, because the pawn-grab with 13 ... ii.b4looks too risky. B If White plays a kingside fianchetto with 9 g3, Black is advised to develop in a classical vein with 9 ... e6, because 9 ... tLlfd7 10 lDxd7 'iVxd7 11 e4 i.h3 12 ii.xh3 'ii'xh3 13 ""3 lta6 14 ii.f4 e6 15 ii.e5 makes Black's natural devel opment very difficult. After 9 ... e6 10 i.g2 i.b4 11 0-00-0 12 e3 h6 13 'ii'e2 i.h7 14 ltdlliJfd7 Black seems to be holding up nicely against 15 tLlxd7, so maintaining the tension with 15 tLld3 looks like the best try for a small edge. The modem attempts 9 e3 and 9 i.g5 lead to interesting play, but should not threaten the via Black's position remains unpleasant. bility of Black's system. 9 e3 gives Black a few 35••• ii.g8 options, with 9 ... h6 and 9 ... g6Iooking the most Black hopes to get in ... e5 somehow in order solid. 9 i.g5 is best met by 9 ... h6, and after 10 to get counterplay against White's queenside i.h4 both 10 ... tLlfd5 and 10 ... e6 give Black ex- pawns. cellent chances for gradually equalizing the 36 e5?! game. 5 Central Variation: Black Plays 6 ... e6

In recent years, the variation with 6liJeS e6 has been very much overshadowed by Morozevich's 11. .. gS in the 6 ... liJbd7 7 liJxc4 "fIc7 variation as well as the solid 7 ... liJb6 from the previous chapter. Nevertheless, 6 ... e6 is a very important move and can lead to a few distinct types of middlegames. In order to combat White's plan of expansion in the centre following 7 f3, Black must take very concrete measures in order to get a playable game. Historically, the 'piece sacri fice' variation of Game 14 was the main line of 6 ... e6 because it seemed like a counterattacking system, but White has refined his play and Black nowadays faces a rather thankless task to draw. The two systems beginning with 7 ... cS aim to equalize more gradually. The fact that it is difficult for Black to play for more than a draw in the 6 ... e6lines is one of main reasons for the move's de cline in popUlarity. However, even if the main lines in this chapter are not as common as they once were, they are still important for theory and anyone looking to play White in the Central Variation would be well advised not to take any of these lines too lightly.

The Games

Game 14 (Onishchuk-Grishchuk) examines the sharp I d4 d5 2 c4 c6 3 liJf3 liJf6 4liJc3 dxc4 S a4 .tfS 6ltJeS e6 7 f3 i.b4 and the piece sacrifice that follows when White plays the critical move 8 e4. lIDs was once an extremely popular line, but nowadays top players shy away from this varia tion. Following the almost forced sequence 8 ... .txe4 9 fxe4liJxe4 10 i.d21i'xd4 lIliJxe4 1i'xe4+ 12 "fIe2 .txd2+ 13 xd21i'd5+ 14 c2liJa6 ISlLJxc4, the tacticalIS ... bS is looking dubious, so Black usually castles. Both IS ... O-O-O and the game's IS ... O-O are pretty solid, but the general con sensus is that Black is suffering a bit to achieve nothing more than a draw. It is more fun to playa piece up than a piece down! In this game Black is on the cusp of equality when a single mistake al lows White to seize the initiative for good.

Game 15 (Alekseev-Wang Yue) begins our coverage of 7 ... cS. After 8 e4 this game examines 8 ... cxd4, which was promoted by Sadler. Historically, this line has been viewed as something of a drawing attempt following 9 exfS.i.b4 (9 ... ltJc6Ieads to a very unpleasant endgame for Black) 10 i.xc4 dxc3 II 1i'xd8+ e2 cxb2 13 .i.xb2

Game 16 (Ponomariov-Rublevsky) covers Black's latest anti-6 lLJeS variation - Kramnik's super-solid 8 ... .tg6 (after 6 ... e6 7 f3 cS 8 e4). Now White may try to avoid an endgame with 9 dS, but this has not brought him anything special so far. Theory had always considered the position arising after 9 i.e3 cxd4 10 'ii'xd4 'iixd4 II .txd4 to favour White, and for many years, that was that. Then in his 2006 World Championship match with Topalov, Kramnik defended this endgame with surprising ease. In this game we take a look at the developments in the endgame that follow 11. .. li)fd7 12lLJxd7li)xd7 13 .i.xc4. Kramnik's 13 ... a6 remains solid and the game's 13 ... :c8 has been holding up as well. CENTRAL VARlATION: BLACK PLAYS 6 ... e6 67

Game 14 Alexander Onishchuk - Alexander Grishchuk Biel2007

1 d4 dS 2 c4 c6 3 lLlfJ lLlf6 4 lLlc3 dxc4 S a4 i.xc5 16 a5 is a little better for White - the h7- .tfS 6 lbe5 e6 (D) bishop is out of play. a2) 10 ... 'ii'a5 is more active: 11 'ii'd4 ~6 12 lLlxc6 bxc6 13 e4 ~xc5 (13 ... i.g6 14 ~f2 0-0 15 i.e2 %:tfd8 16 'ifxc4 lLld7 17 0-0 ~xc5 is w also reasonable) 14 'ii'xc4 ~g6 15 "'a6 'ii'xa6 16 i.xa6 Ab8 17 i.xf6 (after other moves White risks being worse) 17 ... gxf6 18 0-0-0 is about even. b) 8 lLlxc4 0-0 (8 ... c5 9 dxc5 'ii'xdl+ 10 xdl is better for White after either 10 ... i.xc5 11 e4 i.g6 12 lLlb5 or 10 ... 0-0 11 e4 i.g6 12 lLld6) 9 i.g5 (9 e4? lLlxe4 10 fxe4 'iWh4+ 11 d2 'ii'xe4 12 'iff3 'ii'xd4+ 13

This natural move looks very solid, but it can lead to very wild play. Instead of challenging White's knight with 6 ... lLlbd7, Black intends to W develop rapidly. He must often rely on tactics to justify his play; otherwise White will dominate the centre. 7fJ White continues with his plan. 7••• .th4 This is Black's sharpest response, although it essentially commits Black to a piece sacrifice. The alternative 7 ... c5 is covered in the next two games, while other moves would simply be met by 8 e4 'for free' . 8e4 bI) 12 xdl (this time White's develop White accepts the challenge. There are a few ment is too slow to justify this) 12 ... l:.d8+ 13 ways to avoid the coming complications, al xdl lLlbd7 12 lLlxd7 equal too) 14 i.xf6 (the immediate 14 l1al?! 0-0-0 13 e4 ':'xd7+ 14 c2 i.h7 15 i.xc4 i.b3 gives White nothing after 15 i.xf6 i.xc4, 68 CHESS EXPLAINED: THE MAIN-LINE SLAV while 15 lDb6lDd5! is even worse) 14 ... gxf6 15 14••• lDa6 ':a1 .i.b3 16 lDb6 lDc6 17 lDxa8 ':xa8 18 e3 Instead 14 ... b5 is supposed to lose to 15 I:.d1 ~xc5 gave Black good compensation for the 'ii'c5 16 axb5, although Cox points out that after exchange in Beliavsky-Bareev, USSR Ch, Kiev 16 ... cxb5, if White plays 17 'ii'e4 0-0 18 'fIixa8 1986. 'ii'xe5 19 'fIif3lDa6, matters are not easy for him These lines allow White to play with fewer at all. Also note that 17 lDxf7? ~xf7 18 'ii'f3+ risks than in the game (aside from the drawing fails to the cross-check 18 ... 'fIif5+. Therefore, I option at move 10 below), but Black is able to would suggest 17 g4!. This move covers the strike in the centre effectively leading to equal f5-square and prepares .i.g2. White's extra piece play. and initiative look worth more than Black's 8 ..• .i.xe4 pawns. Instead 8 ... lDxe4 ? 9 fxe4 'ii'h4+ 10 ~e2 does 15lDxc4 (D) not give Black enough, and 8 ... .i.g6? 9.i.xc4 gives White exactly what he wants. 9 fxe4 lDxe4 (D) B w

This is without doubt a critical position for the whole line with 8 e4. This unbalanced posi tion has been seen hundreds of times in master 10.i.d2 play, and poses challenges for both sides. Black This is the only try for an advantage, although has three pawns for the piece and White's king Black should certainly be aware of White's is exposed, but often White can successfully other options. 10 "'f3 'ii'xd4 11 'ii'xf7+ d8 12 fight for the initiative. Black must also be care .i.g5+ lDxg5 13 'fIixg7 .i.xc3+ 14 bxc3 'ii'xc3+ ful about what type of endings he goes into. If 15 ~e2 "'c2+ 16 e1 'fIic3+ is a well-known he cannot get his pawns moving, White's extra drawing line. 10 :a3?! is a tricky move, but after piece will be more important. Often White's 10 ... 'ii'h4+ 11 g3lDxg3 12 hxg3 'fIixhl 13 'ii'g4 king can play a vital role in the endgame too. 0-0 14 :tal Hubner's suggestion 14 ...... h2! Black has to be particularly careful about his leaves White struggling. queenside falling, because White can focus all 10••• 'ii'xd411lDxe4 'iixe4+ of his pieces (induding this king) in that direc 11.. ..i.xd2+ 12 lDxd2 'fIixe5+ 13 .i.e2 b5 is tion. considered to be insufficient after either 14 axb5 Nevertheless, it is not all doom and gloom cxb5 15 :a5 or even the simple 14 0-0. The for Black - the white position is not so easy to text-move forces White to displace his king. handle either. Any mistake could prove very 12 'fIie2 .i.xd2+ 13

15 ••• 0-0 lig3 was Kramnik-Shirov, Dortmund 1996. This is probably the most solid move. Black White will have to watch out for tricks with his gets his own king to safety. There are two com king so exposed, but an extra rook is an extra mon alternatives: rook. a) 15 ... 0-0-0 (D) was the main line for a b) 15 ... b5 tries to solve Black's problems long time. with tactics. It would be nice if this move worked, but it appears to go too far. 16 axb5 lbb4+ 17 ~c3 cxb5 18 :dl! (D) (instead 18 lbb6 'iic5+ 19 b3 ·lbc6 20 "xb5 'iie3+ 21 w c2 'ili'f2+ is a draw, as is 18 xb4 bxc4 19 'ii'xc4 :b8+ 20

Black gets his queen's rook into play imme diately. If Black can exchange queens, his king will help protect his queenside pawns. How ever, the black king is less safe on the queenside if queens remain on the board. White has: al) 16 'iie5 f6 17 "e3 (17 'iixd5 cxd5 com pares unfavourably with the main game because Black's queenside pawns are not so vulnerable) bl) 18 ... bxc4 19 ':xd5 lbxd5+ is a rather 17 ... b8 17 .1i.e2 'ii'xg2 (17 ... ~a8 18 puters think it is equal) but I strongly got the g4!) 18 lthgl 'ii'xh2 19 ':xg7 also favours sense that White would eventually queen the White. b2-pawn, and that was that. a23) 16 ... lbc5 17 .1i.e2 'ii'xg2 18 ':hgl 'iixh2 b3) 18 ... lba2+ is an idea from Sammour 19 ':'xg7! (long ago, this move was dismissed Hasbun that makes a fight of it, although it too as a blunder) 19 ... :d4 (19 ... :d3 20 lihl! is the seems insufficient after 19 ~b3 bxc4+ 20 c;t>xa2 tactical point) 20 'iixd4 'iixe2+ 21lbd2 ':d8 22 'iia5+ 21 c;t>bl 0-022 'iixc4 :ab8 and now ei 'ii'xc5 ':'xd2+ 23 ~b3 ':'xb2+ 24 ~a3 %:td2 25 ther 23 'iia2 (Krush-Pang, Philadelphia 2001) 70 CHESS EXPLAINED: THE MAIN-LINE SLAV

or 23 ~d3 (Sarkar-Vigorito, USA Ch, San 24 ~d2 d3 when Black had good counterplay Diego 2006) gives White the better chances, al in Babula-Haba, Czech Ch, Luhacovice 2003. though Black is not without tricks. If 25 i.f3, then 25 ... ~5 26 ltxd6 lbc4+ 27 16 'ifeS ~xd3 lbxd6 is much better for Black. With the black king far away from the vulner 17... f6 able black queenside pawns, White is happy to Black forces the exchange on d5. He could exchange queens. Instead, 16 'iff3 'ifc5 17 ~b1 also play 17 ... :fd8 18 ~e2 f6 191i'xd5 cxd5 20 lbb4 18 i.e2 :ad8 19 l:c 1 :d4 20 lta3 :fd8 lbd2 (the a5-square is no longer available, but gives Black sufficient counterplay, Ki.Geor 20 lbe3 is also possible), which has similarities giev-Shirov, Sarajevo 2000. to the game. 16••• :ab8! (D) The logical 17 ... lbb4+ 18 b1lbb4 21 :a4lbc6 22lbb3!? q;f7 23 ~e2, Vescovi-Gelfand, Bermuda 2004. Just like in the main game, Black had his chances to equal ize, but after a couple of mistakes White pre vailed. 19.• .:tc8+ Now that the b7-pawn is unlikely to come under attack, 19 ... :bc8+ also has its logic, al though after 20 'it>b 1 lbc5 21 :a3 f5 22 ~e2 17 as lbe4 23 :d1 White still has some initiative. Fol This is a typical way to gain space and cramp lowing 23 ... :c7 24lbc2 r;j;f7 25lbd4lbd6 26 a6 the black queenside in the endgame. b6 27 l:1e3 :Le8 28 :e1lbe4 29lbb5 :d7 30 1%c1 Instead 17 :d1 allows 17 ... lbb4+ 18 ~c1 lbc5 31 lbd4 :a8 32 b4lbe4 33 i.b5, Black's lba2+ 19 Q;c2lbb4+ with an immediate draw. position soon fell apart in Anand-Khalifman, 17 i.e2 also comes up short: 17 ... lbb4+! 18 FIDE Knockout (rapid), New Delhi 2000.

After 27 ... g6, 28 l:.h3 creates big problems for Black, because 28 ... .:c7 can be met by 29 w gS+!' The best chance was probably 27 ... .:te8 although after 28 gxfS exfS 29 l:.h3 l:.xaS 30 l::txh7 White is much better. 28 gS+ cJ;;e7 29 l::th3 :h8 30 ~g4 (D)

This is a good move, activating the rook while eyeing White's as-pawn. 24 tlJc2 eM7 2S ltJd4 f7 33 ltJfS does not help Black. a draw. 33 ltJfS+ d8 34 g6 ltJf6 27 g4! If 34 ... ltJf2, then 3S ':xh7 :g8 36 ltJd6ltJxg4 Now Black's pawn-chain will be compro (36 ... :xg6 37 :d7#) 37 ':cl mates. mised and the presence of White's extra piece 3S g7 l:.e8 makes itself felt very quickly. 3S ... l:.g8 is also met by 36 il.hS. 27..• f4 36~hSl-0

Game 15 Evgeny Alekseev - Wang Vue Russia-China match, Nizhny Novgorod 2007

1 d4 dS 2 c4 c6 3 ltJf3 ltJf6 4 ltJc3 dxc4 S a4 draw (although nothing more that that, really). i..fS 6ltJeS e6 7 f3 cS (D) 7 ... cS does not have to be abandoned just yet This logical move introduces a completely though, and Kramnik's solid alternative 8 ... i..g6 different approach from 7 ... i..b4. Black imme is covered in the next game. diately strikes in the centre, hoping to dissuade 9 exfS ~b4 White from his plan of central domination. Nat 9 ... ltJc6 is also not very much fun for Black urally, White is consistent. after 10 ltJxc6 bxc6 11 fxe6 fxe6 12 't'ie2 (12 8 e4 cxd4?! i..xc4 dxc3 13 bxc3 is another good line for This counterattack has been put under a cloud White, one point being that 13 ... 'iYaS can be by this game. For a long time this line was con met by 14 "e2!! 'ifxc3+ IS ~f1 'ifxal 16 sidered to be a reasonable way to play for a 't'ixe6+ ~d8 17 ~e2! with a winning attack, Van 72 CHESS EXPLAINED: THE MAIN-LINE SLAV

w w

der Sterren-Petursson, San Bernardino 1992) A position is reached where in general Black 12 ... dxc3 13 'ii'xe6+ "fIe7 14 ~xc4 'ifixe6+ 15 had been holding his own ... until this game. ~xe6 cxb2 16 ~xb2 .i.b4+ 17 ~e2 ~e7 18 12 ~xf7+! .i.c4 when the bishop-pair gave White a very At a glance, this move looks obvious, but the pleasant endgame in Topalov-Gelfand, Dos impending problems on the e I-aS diagonal had Hermanas 1996. scared White off. Before this discovery, White 10.i.xc4 generally played 12 ~e2 (with the queens off 10 fxe6 fxe6 11 .ltxc4 dxc3 12 'ii'xd8+ ~xd8 the board, this is more logical than 12 0-0) 13 ~e2 cxb2 14 ~xb2 ~e7 leads to an end 12 ... cxb2 13 .ltxb2 ~e7 and here: game discussed in note 'a' to White's 12th a) 14 fxe6 fxe6 15 l:.hbl (after 15 ~d3 Sad move, which is considered to be defensible for ler's suggestion 15 ... .:c8 looks sufficient, and Black. 16 l:tac 1 a5 17liJxb4?! axb4 18 i.b5 ~bd7 was Before the discovery of White's coming sac already better for Black in P.H.Nielsen-Bara rificial play, 10 .lte3 was briefly considered to midze, European Ch, Dresden 2007) 15 ... aS! be promising. After 10 ... dxc3 11 'ii'xd8+ ~xd8 16 .i.a3 It)a6 17 .i.xa6 :xa6 18 ~xb4+ (here 12 0-0-0+ q;e7 13 bxc3, 13 ... .i.a3+ 14 Q;c2 18 ':xb4 axb4 19 ~xb4+ ~e8 20 lIc 1 ':a8 21 liJbd7 15 liJxd7 liJxd7 16 fxe6 fxe6 17 i.xc4 l:tc7 ~d5 22 ':'xb7 ~xb4 23lt)f7 does not work gave White a large endgame advantage in Bu duet023 ... 1H8-compareline 'b') 18 ... axb419 Xiangzhi-Atalik, Ottawa 2007, but instead the l:txb4 b6 leads to a draw, as shown in several cold-blooded 13 ... i.xc3! 14 .i.c5+ ~e8 15 games. liJxc4 liJbd7 16 .i.a3 as! leaves White with b) 14 ':hb1 (delaying the exchange on e6 has nothing better than perpetual check after 17 a small point, but it may not make much differ liJd6+ Q;e7 18 ~c8++ ~e8 19 ~d6+

18 i.b5+ lDbd7 19 i.c6! ':c8 19 ... ':b8 looks more natural, but after 20 B ~d3! i.f2 21 lDe5 White starts to cash in. 20 lDd6+ e7 20 ... ~d8 21 lDxc8 :e8!? (after 21. .. ~xc8 22 exd7+, 22 ... lDxd7 is not possible because of 23 i.xd7+ xd7 24 :dl) 22lDxa7 ':xe6+ 23 f1 is not much of an improvement for Black. 21lDxc8++ ~xe6 22lDxa7 (D)

a) 15 ... ~g6 16 i.d3+ ~h5 17 h4lDc6 (after 17 ... i.xal 18 iLf5! White will mate with g4) 18 ':bl l:.ae8 19 :tb5+ i.e5 20 g4+ lDxg4 21 fxg4+ ~xg4 22 l:.gl + ~h5 23 ':xe5+ lDxe5 24 :tg5+ ~xh4 25 ':xe5 is winning for White. b) 15 ... ~e8 16 ':bl lDc6 17 ':xb7 a6 18 i.a3 leaves Black all tied up. 15 fxe6 This works out well, but 15 lDxh8 exf5 16 lDf7lDc6 17lDd6+ ~d7 18lDxf5 g6 19 :dl+ ~c7 20 lDg7 is also very strong. Now instead of being a rook up, Black is a 15 ••• l:.g816 i.a3 i.d417 :b1 (D) pawn down! A rook down, White plays a rather calm 22••• lDe5 23 :d1 move. Also possible is 17 :dl (or 17 lDd6+ 23 i.b5 lDd5 gives Black counterplay, but 23 d8 18 l:.dl) 17 ... lDc6 18lDd6+ ~d8 19lDb5, iLb7 was possible. winning back some material while maintaining 23 .•. lDxc6 24 lDxc6 iLe5 25 lDd8+ ~5 26 the initiative. g4+ ~g6 27 h4 h6 28 h5+ h7 29 ~2 (D)

B B

17 •.• b6 White has an extra pawn and a positional ad If the white rook gets to b7, Black will never vantage. Alekseev smoothly converts this into untangle. victory. 74 CHESS EXPLAINED: THE MAIN-LINE SLAV

29 ••. :e8 30 ttJc6 ~f4 31 ~d6':'e6 ttJd5 39 ~d2 ttJc7 40 ':'f4+

Game 16 Ruslan Ponomariov - Sergei Rublevsky Candidates match (game 1), Elista 2007

1 d4 d5 2 c4 c6 3 ttJf3 ttJf6 4 ttJc3 dxc4 5 a4 development in Vitiugov-S.Kaplan, BieI2007) ~f5 6 ttJe5 e6 7 f3 c5 8 e4 i.g6 (D) 13 ... fxe6 14 'iixe6+ 'fIIe7 15 'ifxe7+ ~xe7 16 ~xc4 ttJde5 17 i.d5 (17 ~e2 0-0-0 intending ... ttJb4 is better for Black) 17 ... ttJd3+ 18 ~e2 ttJxc 1+ 19 .:taxc 1 ttJd4+ 20 ~f2 0-0-0 was w equal in Shirov-Gelfand, Moscow 2006. a2) 11. .. ttJh5 looks promising: a21) 12 'ifxdS 'iih4+ 13 g3 ttJxg3 14 bxg3 'iVxg3+ 15

w B

ltJc2?! 26 l:dl 'fIe7 27 eS! saw White break 13.. Jtc8 through in Radjabov-Bareev, Odessa (rapid) This is a recent wrinkle in Black's defensive 2007. scheme. Usually Black prevents any unwelcome b2) 20 :ael g6 21 i..g4 rj}g7 22 ~hl hS! 23 visitors from appearing on bS with 13 ... a6. i..h3 b6 24 :d1 'fIc7 2S g3 a6 26 i..g2 .:tfe8 White has then tried: was fine for Black in R.Ruck-P.Kiss, Hungar a) 14

dull, but instructive: IS ... .i.d6 (Black retains the 14... a5!? (D) bishop in order to keep c7 under control) 19 .i.e3 lbeS 20 lbf4 l:xc1 21lhc1 .i.h7 22 ~f1 .:tbS 23 ~e2 ~S 24 g4 ~e7 2Slbg2 f6 (all part of the plan) 26 hS .i.gS 27 ii.a7 l:laS 2S ii.b6 ii.f7, W Aronian-Kramnik, Wijk aan Zee 2007. Black is ready to play ... i.eS with complete equality. c) 14.:tc 1 is a relatively new try. White plays very concretely and fights for the c-file. 14... l:lcS IS lbe2 (D) (15 i.a2 .:tgS 16 ~e2 lbbS 17 lthdllbc6 IS i.b6 ii.b4 19 .:td2 ~e7 20 ltcdl lbbS 21 l:ld4 occurred in Zhang Zhong-Mega ranto, Jakarta 2007, when 21...i.cS 22.i.xcS+ ltxc5 would have held White to a very slight edge) and now:

This move, combined with 13 ... ':'cS, consti tutes Black's plan. Instead of covering the bS B square with ... a6, Black completely gives it up in order to fix the dark squares on the queen side. 15h4 It looks more accurate to play IS ~e2 fIrSt. After 15 .. J:~gS 16lbbS i.c5 17 i.c3 b6 IS h4 h5 19 l:hdl ~e7 20 .:td2 f6 21 l:adl :gdS 22 ltd3 .i.eS (22 ... lbfS fails here to 23lba7!, when 23 ... ltxd3? is not check but 24lbxcs+ is - com pare this to the main game) 23 lbd6 i..xd6 24 lhd6 .i.f7 2S l:ld2 White's position was cer tainly more pleasant in Wojtaszek-Wrrig, Calvia cl) 15 ... .:tgS 16 h4 h6 17 ~f2 (Black has 2007, although that did not prevent Black (who problems developing because of the unprotected was almost 200 points lower rated than his op rook on gS!) 17 ... l:c6 (17 ... i.cS IS ii.xc5 :xc5 ponent) from eventually holding the draw. As 19 b4! .:tc6 20 ii.xa6! highlights Black's tactical we shall see, the insertion of the moves of both problems) IS b4! dS (1S ... ii.xb419 i.xa6!) 19 h-pawns means that Black can sometimes ig h5 .i.h7 20 b5 axb5 21 i.xb5 l:txcl 22 l:xc1 nore the threat to his g-pawn. gave White an overwhelming position in Gyi 15 ... h5 mesi-Bindrich, Bundesliga 2007/S. Of course IS ... h6 is also feasible, but lately c2) Perhaps Black should try IS ... .i.d6 with Black has preferred not to give up any more the idea 16 .i.xg7 l:gS 17 i.d4 lbe5!, when space in these positions. Black will grab one of White's bishops and 16ltcl have some prospects of counterplay. It seems that because White doubles on the 14 i.a2 d-file anyway, this may just constitute a loss of After 14 ii.b5 a6 15 i.xd7+ xd7 16 ~e2 f6 an important tempo compared to the Wojta 17 lthd 1 ~eS IS i.b6 i.c5 19 ii.xcs :Xc5 it is szek- Wirig game mentioned above. However, difficult for White to make headway. Black held with the h-pawns already advanced, 16 ~e2 without any particular difficulty after 20 :d6 can be met by 16... i.c5 immediately, because cJ;e7 21 l:adl i.eS 22 l:6d3 a5 in Bocharov 17 lbbS (17 .i.xg7 l:lh7! traps the bishop) Amonatov, Voronezh 2007. 17 ... ii.xd4 IS lbxd4 c3i;e7 is equal. CENTRAL ltARIATION: BLACK PLAYS 6 ... e6 77

16••• :tg8 Again, Black could have actually saved a tempo with the clever 16 ... ~c5, because after B 17 ltJb5 ~e7 18 ~xg7 Black has the tricky 18 ... ~d4 19 l%xc8 ':xc8. The possibility of a skewer on the back rank forces 20 ~h6, when 20... ~xb2 is equal. 17 e2 ~cS 18 ltJb5 cj;e7 19 ~c3 b6 20 l:hd1 f6 21 ':d2 Trying to steer the game towards Wojta szek-Wirig above with 21 :d3 can be met by 21 ... ltJe5, when Black has no problems. 21 .. .:tgd8 22 ':ed1 (D)

with g4 at some point. Keeping the rooks on the board makes it more difficult for White to play B with such a free hand. 31 i..d2 ltJg6 32 ltc1 l%d8 33 g3 ltJe5 34 ~c3ltJc6 Now it is clear that White has nothing. The knight grips the dark squares on d4 and b4, which is very consistent with Black's plan be ginning with 14 ... a5. 35 ~c4 ltJd4+ 36 i..xd4 l:txd4 37 b3 g6 lh,_lh,

Conclusions

22••• ltJrs The piece sacrifice starting with 7 ... ~b4 is not Black is just in time to cover the d6-square. in such bad standing theoretically, but most 23ltJa7 players have moved away from this variation. It This sortie is not so scary now. is easier to play the endings that usually ensue 23••• l%xd2+ with an extra piece, so the fashion has turned to Check! Compare this to the note to White's wards fresher systems of defence. The piece 15th move. There White's rook was on d3, so sacrifice still requires Black to know quite a bit, the equivalent move ... l:xd3 would have al and if one is willing to do some homework, the lowed ltJxc8+. lines given in the previous two chapters lead to 24 i..xd2 l:tc7 25 ltJb5 ::'c8 26 ltJa7 ':c7 27 richer playas well as better winning chances. ltJb5 ltc8 28 ~f4 (D) The systems with 7 ... c5 have been played White avoids repeating moves, but Black has more with the intention of equalizing the game a couple of ways to equalize. slowly rather than striving for counterplay. Af 28•.• ~e8 ter 8 e4, the 8 ... cxd4line is looking like a dead It is also possible to play 28 ... e5, but Black variation. Kramnik's 8 ... i..g6 is more solid, no prefers not to cede the d5-square just yet - he has doubt, but it is hardly an inspiring variation. correctly assessed that giving up the bishop-pair White can play for an unbalanced middlegame is safe enough. with 9 d5 exd5 10 exd5, even though Black 29 ltJd6 l:.d8 30 ltJxe8 ':xe8 looks alright here. More often White heads for This is better than 30... ::'xdI31 ~xdl ~xe8, the endgame with 9 .i.e3, when 15 ~e2 is look when White can probe around on the kingside ing like White's best hope for an edge. 6 Black Avoids 5 ... ~f5: Bronstein and Smyslov Variations

The five games in this chapter tackle the lines where, after 1 d4 d5 2 c4 c6 3ll:)f3 ll:)f6 4ll:)c3 dxc4 5 a4, Black does not play the main system with 5 ... ~f5. The Bronstein Variation is defined by the aggressive 5 ... ~g4. Mter 6ll:)e5 ~h5 the pin on the e2-pawn means that White already has a decision to make about how to complete his development. The Bronstein can be quite risky for Black if White plays the most principled lines, but some play ers are attracted to it because of the complications that often arise. The Smyslov Variation is characterized by 5 ... ll:)a6. This quiet move aims for the b4-square while waiting for White to commit his e-pawn before playing the ... i.g4 pin. The positions that arise are a little passive, but Black is solid and there are methods for achieving counterplay, espe cially if White overextends himself.

The Games

Game 17 (D.Femandez-R.Gonzalez) begins our coverage of the Bronstein Variation. After 6 ll:)e5 i.h5 White plays the main line, 7 f3, which prepares both to build a pawn-centre and to bother Black's bishop with g4 and even h4. Black plays the most common response, 7 ... ll:)fd7 8ll:)xc4 e5, striking back in the centre. White has several moves here, and he chooses the odd-looking but ef fective 9 i.e3. Soon White's kingside pawn-storm gives him a strong attack and Black is unable to cope with the problems that arise.

Game 18 (Riazantsev-Zhao Jun) examines 7 h3. This little move is trickier than it first ap pears. After the critical7 ... ll:)a6 8 g4 i.g6 9 ~g2ll:)b4 100-0 i.c2 11 'ifd2 i.b3, a very unusual position is reached. The next few moves see both sides fighting for the initiative. After a couple of errors by Black, White reaches a very pleasant endgame and his purposeful play brings home the point.

Game 19 (Kaidanov-Zhang Pengxiang) covers the safe 7 g3. This continuation is hardly an at tempt to refute Black's play, but it is a reasonable choice for players not ready to enter into the com plications covered in the first two games. After a slight hesitation on White's part, Black grabs the initiative. The position is still close to equality when White impatiently lashes out in search of counterplay. This rebounds badly and Black quickly wraps up the game.

Game 20 (Avrukh-Kogan) looks at the Smyslov Variation with White playing the aggressive 6 e4. Black takes the opportunity to double White's pawns with 6 ... i.g4 7 i.xc4 i.xf3 8 gxf3. When White quickly brings a rook to the g-file, Black chooses to castle queenside. The king is not per fectly safe here either, and White gets a comfortable edge. Complications follow, and both sides make mistakes, but the end result is a winning endgame for White. He makes a nice breakthrough which culminates in a clever zugzwang, forcing resignation. BLACK A VOIDS 5... i..j5: BRONSTEIN AND SMYSLOV VARIATIONS 79

Game 21 (Schekachev-Gozzoli) sees White meet the Smyslov with the calm 6 e3. Although not as aggressive-looking as 6 e4, this little move also gives White good chances for establishing some advantage. After 6 ... ~g4 7 .i.xc4 e6 8 h3 ~h5 9 0-0 lbb410 'ii'e2 ~e7, White is ready to advance with e4 and g4, gaining a lot of space. Black manages to fight back from his cramped position, but a couple of inaccuracies land Black in big trouble in the endgame that ensues.

Game 17 Daniel Fernandez - Renier Gonzalez USA Ch, San Diego 2006

1 d4 dS 2 c4 c6 3 ~f3 lill'6 4 lbc3 dxc4 5 a4 play. However, 8 e3 ~gf6 9 .i.xc4 e6 100-0 .i,g4 (D) gives White a normal position similar to the 5 ... .tf5 6 e3 lines of the frrst two chapters. In this case White has effortlessly grabbed the bishop pair and should enjoy a pleasant advantage. w 7 f3 (D) White intends to build a big pawn-centre. Black's h5-bishop may get locked out of play, and an advance with g4 and h4 is another possi bility, when the bishop could be in trouble. The sharp 7 h3 and the solid 7 g3 are consid ered in the next two games. White can also play 7 ~xc4, which has been mostly ignored by ev eryone except Burgess. The critical line is 7 ... e6 8 'ifb3 ~6 9 e4 .i.b4 10 f3 lbxe4 11 fxe4 'iVh4+ 12 g3 'ii'xe4+ 13 ~f2 'ifxhl 14 .i.g2 'i!fxh2 15 .i.e3 .i.f3!? 16 ~xf3 'ffh5+ 17 g4 The Bronstein Variation is a sharp alterna 'i!fh4 with an unclear position that looks some tive to the main lines of 5 ... .tf5. Black allows what favourable to White. So it looks like 7 ~5 with tempo and hopes that the pin on the ~xc4 is yet another line Black needs to worry white e-pawn will cause some discomfort. If about. White hopes to punish Black's provocative play, he will have to go into some rather complicated variations. Black will hope to get sufficient counterchances in these lines, but if White is B well prepared, he has excellent chances of gain ing a real advantage. 6~eS Of course White grabs the chance to play this with tempo. Self-pinning with 6 e3 or 6 e4 can be met solidly by 6 ... e6 or sharply with 6 ... e5!? 6 •.. .i.hS 6 ... ~bd7 is an unusual sideline. After 7 ~xg4 ~xg4, 8 e4 is not so clear because of 8 ... e5!, when 9 'i!fxg4 exd4 is dangerous and 9 d5 ~gf6 gives Black reasonable dark-square 7 ••• ~fd7 80 CHESS EXPLAINED: THE MAIN-LINE SLAV

This is Black's principal continuation. Black 11 ... i.xh4+ 12 We2 i.g6 13 exd5 cxd5 is an challenges White's strong knight and prepares unclear piece sacrifice. the counter-attack ... e5. The apparently passive b3) 8 ... .tg6 is supposed to be bad, but mat 7 ... e6 can also lead to some sharp play. White ters are not so clear. 9 h4 c5 10 h5 (10 dxc5ltJc6 has: 11 ltJxc6 'ii'c7! 12 'iVd4 'ii'xc6 13 'ii'xc4 i.xc5 a) 8 e4 i.b4. Now 9 i.xc4? fails to 9 ... ltJxe4! was OK for Black in Rodshtein-Gupta, World so White's options are: Junior Ch, Erevan 2006) 10 ... i.d3! (a funny al) 9 g4 and now 9 ... ltJxe4 10 fxe4 'ii'h4+ move) 11 exd3 (if 11 dxc5 then 11. .. 'ii'd4!, but 11 Wd2 ~f2+ 12 'ife2 'ii'xd4+ 13 ~c2 'ifxe5 14 perhaps White should go for this) 11. .. cxd4 12 gxh5 ltJa6 15 'iff3 was insufficient for Black in dxc4 'ii'd6 13 'ife2 dxc3 looked fine for Black in Gyimesi-Ilincic, European Clubs Cup, Eupen Khismatullin-Chuprikov, Voronezh 2007. 1999. Perhaps the simple 9 ... i.g6 should be White must play very sharply and be well preferred. After 10 h4ltJbd7!? both 11 h5 i.xe4 prepared to get anything against 7 ... e6. and IlltJxd7 'ii'xd7 12 h5 i.xe4 are unclear. SltJxc4 e5 (D) a2) 9 .i.e2 b5 10 i.e3 ltJfd7 11 g4 i.g6 12 h4 ltJxe5 13 dxe5 was Aronian-Bu Xiangzhi, World Team Ch, Beersheba 2005. Bu gives 13 ... 'ifxdl + 14 Wxdl i.xc3 15 bxc3 h5 16 g5 w a6 as being better for Black, although White certainly has chances here as well. b) 8 g4 (D) is critical:

This is the point of Black's play. Without this counterblow, 7 ... ttJfd7 would make little sense. Now 9 dxe5? loses to 9 ... 'ifh4+, and 9 ltJxe5 ltJxe5 10 dxe5 ltJd7 makes it difficult for White both to hold on to the pawn and to develop nor mally. For a long time White's main response was 9 ltJe4, hoping to exploit the weak d6- bI) 8 ... ltJfd7?! 9 ltJxd7 (9 h4 .i.e7 10 i.e3 square, but in recent years White has discov ltJxe5 11 dxe5 .i.xh4+ 12 l:txh4 'ii'xh4+ 13 .i.f2 ered several other promising ideas. 'iVh2 was unclear in Beliavsky-Comette, Linares 9 i.e3!? 2003) 9 ... 'ii'h4+ 10 ~d2 ltJxd7 11 gxh5 ltJc5 White guards against the check on h4 and (this looks fun, but the complications will fa now he is ready to play g4, when he can both vour White) 12 c2 ttJb3 13 ~bl 0-0-0 14 e3 capture on e5 and play h4 to harass Black's e5 15 i.xc4! gave White a strong initiative in bishop. Often White can whip up a strong king Aseev-Ulybin, Russian Ch, Samara 2000. side attack very quickly. There are other impor b2) 8 ... ttJd5 9 h4 (bad is 9 gxh5? 'ii'h4+ 10 tant options: ~d2 'ii'xd4+, but 9 e4 'ii'h4+ 10 ~e2ltJxc3+ 11 a) 9 e4 is unusual, but it holds a certain ap bxc3 i.g6 12 :bl probably favours White) peal. 9 ...... h4+ (neither 9 ... .i.b4 10 dxe5 nor 9 ... .tb4!? 10 i.d2 .te7 11 e4 (11 ~f2ltJb4 12 9 ... f6 10 .i.e3 seems sufficient for Black either) .tf4 f6 13ltJxc4 .tf7 is given by Bu Xiangzhi) 10

e8 15 e6lLlc5 space advantage and the initiative in the model is perhaps less clear, but also good for White af game Kramnik-Damljanovic, Moscow Olym ter 16 ~c4 .i.xh 1 17 lLlxh8 or 16 :g1!?) and piad 1994. now: b22) 14 ... d8! 15 h4 (15 liJxb7+ ~c7 16 al) It should also be noted that 10... exd4 11 liJba5 ltJa6 17 e3 dxe3+ 18 lLlxe3 liJb4 gives 'iflxd4 ~c5 is not good because of 12 lLld6+ Black good counterplay) 15 ... f6 16 g4 .i.e8 17 f8 13 1ixg7+! rj;xg7 14lLlf5+. A great trap. ttJxb7+ ~c7 18liJbd6 ttJb6 was equal in Cha a2) 10... ~b4 could use a few tests. 11 dxe5, banon-Hear, Montpellier 2006. 11 .i.e3 and 11 g4 all are interesting tries for c) 9 g3 is another testing line. White simply White. prevents the check on h4. Because 9 ... exd4 10 b) 9 lLle4 is the traditional main line for 'itxd4 leaves d6 weak, Black has tried: White. 9 ... .i.b4+ (9 ... 'fIc7? fails to 10 dxe5 cl) 9 ... ~b4 10 dxe5 0-0 11 .i.h3 'ifle7 12 f4 lLlxe5 11 lLlcd6+) 10 ~d2 'fIe7 (D) (this is :d8 13 'ii'c2 f6 14 e6 ttJc5 and now 15 ttJe3 better than 10... ~xd2+ 11 'iflxd20-0 12 dxe5) liJxe6 16 liJf5 'itf7 was solid enough for Black reaches an important position: in Gladyszev-Flear, Montpellier 2007, but Hub ner's idea 15 f5 :d4 16 ttJd2looks much more testing. c2) 9 ... 'fIe7 10 ~h3 .i.g6 11 0-0 exd4 12 w 'iWxd4 ttJc5 was Khalifman-Rogers, Biel Inter zonal 1993. Now 13 :dl! looks very good for White, because 13 ... liJb3 fails to 14 "'f4 with an invasion on d6 coming; for example, 14..... f6 15 'ite3+ .i.e7 16lLld6+ wins for White. c3) 9 ... f6Iooks the most solid. Now 10 1i'b3 and 10 .*.h3 are possible, but White almost invariably chooses 10 dxe5 liJxe5 11 'iflxd8+ ~xd8 12liJxe5 fxe5, when either 13 ~g5+ or 13 ~h3 gives White a nice endgame initiative. We now return to 9 .i.e3!? (D):

White has: bl) I1lLlxe5 is not so common, but it looks like a good try for a safe edge. 11 ... lLlxe5 12 B dxe5lLld7 13lLld6+! (13 .i.xb4 1Vxb4+ 14 ii'd2 ii'xd2+ 15 ~xd2 0-0-0 16lLlg3 ~g6 17 e4 l:he8 18 .i.e2 l:txe5 19 ~e3 f6 was fine for Black in Harikrishna-Acs, Paks 2006) 13 ... .i.xd614 exd6 'itxd6 15 .i.c3 gives White a pleasant edge in the middlegame after 15 ... 1i'e7 16 ii'd4 f617 e4 or the endgame in the case of 15 ... 1i'xdl+ 16 ':xdl f6 17 e4 because of the bishop-pair. b2) 11 .i.xb4leads to an endgame that Black seems to have worked out after 11. .. 'iflxb4+ 12 1i'd2 "xd2+ 13 e7?! 15 lLlf5+ (15 lLlxb7 is also An early 9 ... exd4 is usually a mistake, be possible) 15 ... ~f6 16lLlxd4 :d8 17 e4lLlc5 18 cause after 10 'itxd4 White has a strong initia ~c3 :Xd4 19 xd4liJb3+ 20 ~c3 liJxal 21 tive because of the vulnerable d6-square. Ideas .i.e2 ~e7 22 l:txal liJd7 23 b4 gave White a like 0-0-0, 1i'e4+ and lLle4 are in the air. 9 ... f6 is 82 CHESS EXPlAINED: THE MAIN-LINE SlAV worse here than it is against 9 g3 because White 12••. h6 can quickly castle queenside. 10 dxe5 ~xe5 11 Instead 12 ... hS allows White to open lines 'iixdS+

It looks like Black is developing counter play, but White refuses to let go of the initiative. It may seem somewhat surprising that White 17 fS! ~b3 can attack like this, but his king is pretty safe Instead 17 ... .i.h7 looks dreadful, but the and he controls a lot of space. threats of ... ~b3 and ... ~bd7 force White to act BLACK A VOIDS 5... i..j5: BRONSTEIN AND SMYSLOV VARIATIONS 83

immediately. After 18 f6 'iff8 19 0-0 ~b3 20 'liel ~xal 21 'lixal White has excellent com pensation for the small material investment, but w Black can certainly fight. 18 'libl ~xal? 18 ... .i.xf5 was a better practical try. 19 fxg6 ~b3 20 O-O:rs (D) After 20 ... fxg6 21 'ifxg6 Black cannot deal with the dual threats of 22 .i.e6+ and 22 .i.f5. 21 :f6! There were other ways to win, but this is cer tainly the flashiest! 21 ... .i.c5 Also losing are 21. .. gxf6 22 gxf6 'lid8 23 gxf7 + and 21. .. ~c5 22 gxf7 + ':'xf7 23 ':'xf7 22 gxf7+ %:txf7 23 'ii'g6 'ifxf7 24 ~d6 'fIe7 25 'ifg6, threatening both Now there is no defence against i.e6. 26 ~f5 and 26 i.f5. 23 ... ~a6 24 :xf7 'iVxf7 25 i.e6 1-0

Game 18 Alexander Riazantsev - Zhao Jun Aeroflot Open, Moscow 2006

1 d4 d5 2 c4 c6 3 ~f3 ~f6 4 ~c3 dxc4 5 a4 7 ... ~bd7 8 ~xc4 e6 9 g4 i.g610 .i.g2 i.b411 i.g4 6 ~e5 i.b5 7 h3 (D) 0-0, when White has a kind of better version of Game 12, which should suffice for an edge. 8 g4 .i.g6 9 i.g2 Instead 9 e3 is not very threatening. 9 ... ~b4 B 10 i.xc4 e6 II ~xg6 hxg6 looks an awful lot like Game 7, which was fine for Black. 9••• ~b410 0-0 After 10 ~xg6?! hxg6 Black has a very solid position and White will have trouble recovering the c4-pawn. 10.•. i.c211 'iVd2 .i.b3 (D) A very unusual position. It is not so easy for White to win back the c4-pawn and his queen looks a bit in the way on d2. However, Black is behind in development, and the b3-bishop looks rather strange. With White's g2-bishop and e5- This little move can lead to surprisingly sharp knight bearing down on c6, Black's queenside play. White will play g4 while avoiding any an may suddenly become somewhat fragile. noying checks on h4 which can occur after 7 f3. 12~e4 Black must play concretely to avoid slipping This is the sharpest continuation. White into a bad position. threatens the b4-knight as well as ~g5, when 7 ... ~a6 f7 cannot be defended. The positional 12 a5 is This is Black's sharpest response. Black aims also possible, when the further advance of the for the c2-square. It is also possible to play a-pawn is in the air. 84 CHESS EXPLAINED: THE MAIN-LINE SLAV

w w

After the text-move, the obvious 12 ... 1£1xe4? represents a serious material investment but 13 'ii'xb4l£ld6 is refuted by the not-so-obvious Black's undeveloped position is difficult to de 14 'ii'xb7! (D). fend. 14 a5, taking the b6-square away from Black's queen, is another interesting idea. Then 14 ... e6 (l4... 1£1xal 15 l£lxb7 'iWc7 16 .i.xc6+ 1i'xc6 17 B l£lxc6 :xc6 18 1i'f4 is much better for White, because Black's pieces are so scattered) 15 l£lxb7 (15 a6 can be met by 15 ... b5, when it is not clear how White should continue) 15 .....xd4 16 i.xc6+ :'xc6 171£1xc6 'ifxd2 18 .i.xd21£1xal 19 l:Lxal looks better for White due to Black's poor development and the weak a7-pawn. 14..... xd4 This appears to be the best try. Other moves: a) 14 ... 1£1xd4 15 e3ltJe6 161£1xb7'ifb6 (or 16 .. .'ii'xd2 17 i.xd2) 17 lDa5 .i.xa4 18 l£laxc4 gives White a big advantage. Therefore Black must head to c2 for more b) 14... e6 15 l£lxb7 'ifb6 16 l£la5 i.b4 17 adventures. l£lexc4! (instead 17 l£laxc4 'ii'xd4 18 'ifxd4 12••• l£lc2 13ltJcS l£lxd4 19 e3 i.c2 20 exd4 i.xbI 21 .i.d2 i.xd2 This is the point of White's play. Black is 22l£ld6+ ~e7 231£1xc8+ ':'xc8 24 :txbll£ld5 vulnerable on the light squares on the queenside. was level in I.Farago-Pitschka, Salzburg 1999) 13••• :c8 (D) 17 ... 'ifxd4 18 'ii'xd4 lDxd4 19 l£lxb3 l£lxb3 20 Of course 13 ... 1£1xal? is impossible because :dl is better for White because Black's pawns of 141£1xb7 and 15 .i.xc6+. are more vulnerable. 13 ... 'ii'xd4?! also looks too risky: after 14 15 l£lxb7 "xd2 l£lxb7 l£ld5 15 1Vg5! e6 (both 15 ... f6 16 'ilh5+ After 15 ... .i.a2, 16 l£lxc6 "xd2 17 .i.xd2 g6 17 l£lxg6 hxg6 18 "xh8 and 15 ... 'iVb6 16 .i.xbl 18 :'xbl is much better for White ac 'iWf5 'iWxb7 17 'ii'xf7+ d8 look too dangerous cording to Riazantsev, while 16 .i.xc6+ :'xc6 as well) 16 .i.e3 "xb2 17 .i.c5 h6 18 .i.xd5 17 l£lxc6 1i'd5 18 :dI also looks very strong. cxd5 19 "'f4 White has a strong attack. However, Black could try either of Riazantsev's 14%tbl suggestions of 15 ... 1£1d5 or 15 ... i.xa4, both of White could try the speculative 14 l£lxb7 which look unclear. 'iVb6 15 l£lc5l£lxal 16 'iff4 intending g5. This 16 i.xd2 .i.xa4 (D) BLACK A VOIDS 5 ... ~5: BRONSTEIN AND SMYSLOV VARIATIONS 85

w B

White is two pawns down, but c4 will fall .:tal, but 23 ... i.bS 24:'fcl i.d3 2S :'b2%:tb8 26 and Black is well behind in development. l:tbxc2 i.xc2 27 %:txc2 :'xb3 is a better attempt 17e3 to resist. At least Black may get counterplay White cuts off one retreat of the c2-knight. 17 with his queenside pawns. :'fc IlDds 18 i.e4 also looks good; for example, 24 .:tbel i.xg2 25 cj;xg2 liJaJ (D) 18 ... f6 (18 ... llki4 loses to 19 :'xc4liJxe2+ 20 ~f1, while 18 ... liJdb4 19liJxc4 and 18 ... liJcb4 19 :'xc4 are also insufficient for Black) 19 i.xc2 i.xc2 20 :'xc2 fxeS 21 .:txc4 and White w has more than enough for the pawn. 17••• e6 Riazantsev gives 17 ... liJdS 18 .:the 1 (better than 18 ':fcl i.b3 19liJaS? i.a2, trapping the rook) 18 ... i.b3 (18 ... e619liJxc4 :'b8 is a better try to hold on) 19 liJaS e6 20 liJexc4 i.xc4 21 liJxc4ltJcb4 22 .:tal!. White threatens the a7- pawn and :a4 is also in the air, when the b4- knight will start looking rather unstable. 18 liJxe4 :'e7 19 liJbd6+ White grabs the bishop-pair. 19••• i.xd6 20 liJxd6+ ~e7 21liJe4 (D) 26 i.b4 White's pieces coordinate very well while White wins back the pawn and keeps a big Black's forces are scattered. For starters, ltJcs is advantage. threatened. 26•.• liJbS 27 i.xeS+ cj;d7 28 :'fdl+ ~e8 21 ••• cS 29 i.e7! fS 30 ':xe7+ liJxe7 31 gxfS exfS 32 After 21 ... liJxe4 22 i.xe4 the c2-knight is in cj;f3 trouble. White has the better pawn-structure, and all 22liJxf6 of his pieces are better than Black's, including Even better is 22 :tfc 1 :'d7 (22 ... :tb8 23 the king. i.c3) 23 i.aS, threatening b3. Black's queen 32•. ':e8 33 i.f6 :'e6 34 l:td8+ ~b7 35 i.d4 side pieces are a mess. ':h636:d7 22•.. gxf6 23 b3 i.e6?! This wins a pawn. White's advantage becomes clear after this 36•••

White is happy to exchange the b3-pawn for liJg3+ 50 f4 liJe2+ 51 ~g5 ~g3 52 i.e7 h4 f5 because he will have two connected passed 53 g4 !De2 54 <;itxh4 e6 55 -.t>g4 -.t>f6 56 pawns. Black's h-pawn is of little consequence. ~a5 d5 45 l:txf5+ lhf5+ 46 ~d5 60 f4! 1-0 xf5 liJd2 47 ~h4 h5 48 ~d8 liJe4 49 f3 The knight is trapped.

Game 19 Gregory Kaidanov - Zhang Pengxiang World Team Ch, Beersheba 2005

1 d4 d5 2 e4 e6 3 liJo ltJf6 4 ~c3 dxe4 5 a4 ~g4 6liJe5 ~h5 7 g3 (D) w

pressure on the pinned c3-knight with 9 ... ~d5. This is another option that Black usually does not have when White fianchettoes. White has tried: This calm continuation is surprisingly popu a) 10 'ii'c2 allows Black to simplify with lar. White makes no attempt to punish Black for 10 ... liJb6 because the d4-pawn is hanging. Note his provocative play and simply gets on with his that the greedy 10 ... liJxc3 11 bxc3 'ii'xd4?! is development. It may seem as if the play would too risky in view of 12 cxb4 'ii'xal 130-0 with end up being similar to that in Game 12, where i..b2 or ~d6+ coming. White also adopts a fianchetto, but here it is de b) 10 tid3 i.e7!? 11 0-00-0 12 e4liJb4 13 cidedly less threatening. It is difficult for White 'ii'd2 tic7 (Black wants to exert pressure on to push his e-pawn, and this in tum makes it the d4-pawn as quickly as possible; instead harder for White to find a good square for his 13 ... liJSa6 14 b3 'ii'c7 15 liJe2! :adS 16 liJf4 queen because the d4-pawn is loose. Ideally, i..g6 17 h4 gave Black problems with his bishop White would like to play 'iWe2 (as in Game 12), in Tukmakov-Oll, Rostov-on-Don 1993, but but in order to do this he would have to spend 13 ... liJd7 is a sensible alternative) 14 h3 l:tdS 15 time playing h3 and g4 before e3 is possible. g4 i..g6 16 l%dl i..f6 17 e5 i.e7 lSliJe4 i.xe4 Although this approach is fairly safe for White, 19 i.xe4 c5 20 liJd6 i..xd6 21 exd6 ':xd6 22 he has much less chance of gaining the initia 'ii'c3liJd5 gave Black good play in Onishchuk tive than after 7 f3 or 7 h3 (or even 7 liJxc4). R.Gonzalez, Lindsborg 2005. 7 ••• e6 8 ~g2 i..b4 (D) c) 10 'iVb3 0-0 and here: 90-0 cl) 11 0-0 loses a pawn to 11. .. i..xc3 12 The immediate 9 liJxc4 is also possible, but bxc3 i.xe2 although after 13 :e1 ~xc4 14 this gives Black the option of putting immediate 'ii'xc4 White has decent compensation. BlACK A VOIDS 5 ... i..f5: BRONSTEIN AND SMYSLOV VARIATIONS 87

c2) 11 e3 a5 12 0-0 liJd7 13 e4 ttJ5b6 14 lDxc3 16 iLxd8 liJxdl because after 17 ~e7 liJxb6 'ii'xb6 15 ~e3 iLe7 gave Black a solid :fe8 (no better is 17 ... ~d3 18 ~xf8 iLxfl 19 position in Bocharov-A.Andreev, Tomsk 2003. iLxfl) 18 iLa3 ~d3 19 ltJd6 White wins mate c3) 11 e4liJxc3 12 'iVxb4 (12 bxc3 'iVxd413 rial. cxb4 {13 'iWxb4?? "dl#!} 13 .....xal 14 0-0 14 iLh4 "f6 is probably insufficient for White because White hopes to gain control of the d6-square. there is no check on d6) 12 ... 'iixd4! and Black 14..• liJb61S iLg3 "dS threatens mate on dl. After the somewhat It seems as though Black has wasted a lot of forced sequence 13 iLe3 'iVd3 14 ~fl "xe4 15 time, but White did spend several tempi bring :gl liJd5 16 'iVxb7 Black has 16 ... liJa6!, with ing his bishop to g3. the idea 17 'ii'xa6 liJb4. 16liJeSliJfd7 (D) c4) 11 iLd2 a5 12 e4 liJb6 13 liJxb6 'iVxb6 14 iLe3 c5! 15 d5 (15 dxc5 'ifa6) 15 ... liJd7 16 0-0 liJe5 17 h3 liJf3+ 18 Wh 1 liJd4 gave Black very good play in Leitao-Beliavsky, Erevan w Olympiad 1996. 9 ..• 0-0 After 9 ... liJdS I think White should play 10 iLd2 (instead of committing the queen with the more common 10 'iVc2) with the idea 10 ... liJb6?! 11 as 'ili'xd4? 12 axb6 'iVxe5 13 :'xa7!. 10 liJxc4 as (D)

w 17liJdJ After this passive retreat Black seizes the ini tiative. Better is 17liJxg6 hxg6, when Black has a very solid position, while White can be satis fied with the bishop-pair. 17... liJc41S :a2 This awkward move is the only way to pro tect the b-pawn without walking into an annoy ing pin. IS... 'iVb6 19 liJf4 eS 20 ttJxg6 So White grabs the bishop after all, but it has Black prevents· as by White and solidifies his cost him valuable time and Black has taken grip on the b4-square. White's e-pawn is pinned, control of the centre. which makes it hard for him to find a construc 20••. hxg6 21 dxe5liJdxeS 22 'ii'c2 :adS tive plan. With his next move White plans the g4 Black's pieces are all coming into the centre advance to kick away Black's annoying bishop. very quickly. 23 ... l:d2 is a threat. 11 h3 liJbd7 12 g4 iLg6 13 iLgS 23liJbl (D) If White plays 13 e4, Black can safely grab This is rather passive. Instead 23 liJe4 f5 24 the e-pawn with 13 ... iLxc3 14 bxc3 ~xe4, so b3! liJd2 (not 24 ... liJa3 25 'ii'b2) 25liJxd2 :'xd2 White pins the f6-knight. 26 'ii'bl is preferable, when Black's weakened 13.•• 'iIi'c7 kingside gives White hopes of achieving coun After 13 ... iLe7 White can play 14 e4, not terplay, while 23 iLf4 ltd4 24 ttJe4 f5 25 ~xe5 fearing the tactical sequence 14 ... liJxe415 iLxe7 liJxe5 26 'ii'b3+ is also sensible. 88 CHESS EXPLAINED: THE MAIN-LINE SLAV

After 28 "'f5 "'d8 Black intends to evict the queen with ... g6 and 29 ~xe5 can be met by B 29 ... l:txe5!. 28...... d8! Black regroups his pieces. 29 :txd7 t;)xd7! Instead 29 .....xd7 30 "f5 gives White some counterplay. 30 'ii'c2 'ii'f6 31 e3 t;)cS 32 "c4 ltd8 White's rook and bishops are all ineffective, but White's next move makes things worse. 33 f4 gxf4 34 exf4 "d4 34... :d4!, intending ... t;)d3, is even better. 35.tn? 23 •• .:t"e8 24 b3 t;)d2 2S t;)xd2 :txd2 26 After 35 "xd4 lhd4 Black is better, but 'ii'bl (D) White can fight on with 36 lte2. 35••• 'ifdl! (D)

B w

26•• J:td7 Black keeps the rooks on the board because Suddenly the b3-pawn is falling and White's it is difficult for White to get the a2-rook into king is in trouble. play. 36 lte2 t;)xb3 37 J.h4 t;)d2! 27~h2 White cannot defend everything. White intends f4, with counterplay. 38 .txd8 t;)xc4 39 :e8+

Game 20 Boris Avrukh - Artur Kogan Maalot-Tarshiha 2008

1 d4 d5 2 c4 c6 3 t;)f3 t;)f6 4 t;)c3 dxc4 5 a4 5 ... J.f5. Black's knight eyes the weakened b4- ltJa6 (D) square and waits for White to push his e-pawn The Smyslov Variation is a solid choice for before developing his bishop from c8. The those wishing to avoid the bulk of theory after Smyslov is certainly safer than the Bronstein BLACK A VOIDS 5 ... i.J5: BRONSTEIN AND SMYSWV VARIATIONS 89

a) 8 ... iLe7 9 0-0 0-0 10 h3 ..th5 11 g4 iLg6 12 tLJe5 lLIb4 and now both 13 lLIxg6 and 13 w 'ii'e2 give White a secure plus. b) 8 ... iLb4 seems unnatural to me - the a6- knight looks silly. 9 'ii'd3 (9 'ii'c2 i.xf3 10 gxf3 is similar) 9 ... i.xf3 10 gxf3"a5 (l0... lLIc7 has also been tried, but the knight looks silly here too) and here both 11 l::tgl and 11 ~e2 give White a nice position, with his strong centre and bishop-pair. c) 8 ... lLIb4 9 0-0 (White can also play 9 a5 if he is concerned with an early ... a5, but I do not think this is necessary, as we shall see) and Black has: Variation, but it is also easier for White to get cl) 9 ... a5 looks dubious to me. The tempo an edge with fairly straightforward play. spent securing the queenside certainly counts 6e4 for something. After 10 h3! ~h5 (in the case of This is White's most ambitious continuation. 10 ... iLxf311 ..xf3lDc212I:tadllDxe313fxe3 He grabs the centre and is unconcerned with the Black is not ready to castle and e5, or even d5, possibility of taking on doubled pawns. The is coming) 11 g4 i.g6, with his development quiet alternative 6 e3 is considered in the next lead, White has many tempting continuations game. such as 12lDd2, 12lDe5 or the line-opening 12 6 ••• ~g4 7 .ltxc4 d5. 7 iLe3, trying to avoid doubled pawns, is in c2) 9 ... i.e7 (D) gives White a choice: accurate because of 7 ... "'a5, putting pressure on the e4-pawn. 7 ••• ~xf3 This is the most forcing and principled move. w Black should avoid 7 ... lLIb4 because of the tacti cal trick 8 tLJe5! (threatening mate on f7) 8 ... e6 (8 ... i.h5 9 "xh5!) 9 f3 with a big advantage to White. Instead 7 ... e6 is popular. After 8 i.e3 (D) (8 0-0 is also possible and is likely to transpose) Black generally places a piece on b4:

B c21) 10 h3 and after 10 ... ..th5 11 g4 ..tg6 both 12 lLId2 and 12 lLIe5 look promising for White, but instead 10 ... i.xf3 11 'ifxf3 lDc2 12 l:ladl 0-0 looks pretty solid. c22) 10 iLe2 0-011 a5 b5 12 axb6 (ignoring the pawn with 12 h3 or even 12 a6!? may be more promising) 12 ... axb6 13 'it'b3 gives White a slight edge at best. c23) 10 l:lcl 0-0 11 h3 is interesting, not al lowing any ... lDc2 ideas. 90 CHESS EXPLAINED: THE MAIN-LINE SLAV

8 gxf3liJb4 so ... as is not such a threat. White's play is on Black can also play the immediate 8 ... e6 9 the kingside and in the centre, so there is no rea 0-0 (Black need not fear the doubling of his son to give Black ... bS possibilities. 10 ... i.e7 pawns after 9 i.xa6 bxa6 10 'ifd3 as, when the 11 ~hl 0-0 12 :gl ~h8 13 f4 bS! (we shall see pressure on the b-file gives Black even chances) more of this idea in the next game) 14 axb6 9 ... i.e7 10 ~hl (as in the main game, White axb6 IS lIxa8 1i'xa8 16 1i'f3 cS 17 dxcS i.xcs does not hurry to play i.e3, preferring kingside 18 fS i.d4 gave Black enough play in Oni play) 10 ... 0-0 11 :gl g6 12 f4 'ifas 13 'iff3 shchuk-Zhang Pengxiang, Manila 2006. l:ad8 14 i.e3 'ifb4 IS 'ife2 liJc7 16 as! (White 10...... a5 threatens 17 lIa4, winning material) 16 ... liJbS It looks a bit risky to delay development of 17 eS liJdS 18 i.xdS exdS 19 liJxbS 'ifxbS 20 the kingside, but Black intends to castle queen 'ifxbS cxbS 21 a6 gave White an enduring end side. White also gets pressure after 10 ... i.e7 11 game initiative in Rogozenko-Shirov, Bundes l::tgl g6 (after 11. .. 0-0 12 eS the possibility of liga 2000/1. 13 i.h6 forces the inconvenient 12 ... ltJe8 al 90-0 e6 (D) though the knight will emerge via c7 and dS or even g7 and fS after ... g6) and here: a) 12 i.e3 a613 as "'c7 14 'iVb3 0-0 ISltJe2 (another idea is IS :a4 cS 16 dxcS ltJc6 17 w "'a3) IS ... cS 16 dxcS i.xcs 17 i.h6lHe8 18 'ikc3 i.e7 and now White won a pawn with the unexpected 19 i.xe6! "'xc3 20 bxc3 fxe6 21 cxb4 in Lalic-Shaw, Liverpool 2007. b) 12 i.h6 i.f8 13 i.e3 (White has gained a whole tempo) 13 ... i.e7 14 f4 "as IS lIgS 'ilic7 16 'ilif3 ltJhS?! 17 ':xhS gxhS 18 "xhS gave White a powerful initiative in Izoria-Vunder, St Petersburg 2002. 11 lIg1 h6 This looks a bit loosening and it does not stem the pressure on the g-file. After 11 ... g6 12 10~hl l::tgS 'ifc7 13 as White is certainly better, but White quickly brings a rook to the g-file. Black's position remains pretty solid. Often White plays 10 i.e3 first, or gets there 12 i.e3 0-0-0 (D) by transposition. This also gives White good chances after 10 ... i.e7 11 ~hl 0-0 12 :gl ~h8 13 f4! (131i'e2 'ifc7 14 ':aclliJhS IS eS :ad8 16 i.b3 'ifas 17 ltJe4 ltJdS gave Black a w solid position in I.Sokolov-Conquest, Reykja vik 2000) 13 ... cS 14 dxcS'ifc8 (14 ... 'ifc7 IS eS ltJd7 16 liJe4 liJxcs 17 ltJd6 l::tad8 18 i.d4! f6 19 'ifhS also gave White the initiative in Stef ansson-Gretarsson, Icelandic Ch, Arborg 1998) IS eS ltJd7 (Mikhalevski suggests IS ... liJfdS but after 16 i.xdS exdS 17 IIc 1 Black will re main a pawn down) 16 ltJe4 was difficult for Black in V.Mikhalevski-Al.Rabinovich, Dieren 1999. 10 as is another idea, but this seems unnec essary to me. The b4-knight is secure enough, 13 'iVb3 ~b8 14 i.e2ltJe8 BLACK AVOIDS 5 ... i..f5: BRONSTEIN AND SMYSLOV VARIATIONS 91

This is not an uncommon manoeuvre. The 19.• Ld4 knight does little on f6 now, and from here it An interesting practical decision. Black hopes may come to d6. If White plays e5, the knight to break White's initiative for a relatively small may head for d5 via the c7 -square. material investment. However, White does not 15 l:tgc1 Ad7 (D) have to bite. 20 tbeS! :f8 21 a6 b6 Instead 21. .. tbcxa6 22 .i.xa6 tbxa6 loses to 23 tbxc6+. w 22 'ifc3 This preserves the advantage, but 22 Aa4 is even stronger. After 22 ... f6 23 :xb4 fxe5 24 l:txb6+! White wins quickly. 22•.. cS 23 .i.xd4 'ii'xd4 If 23 ... cxd4, then 24 tbc6+ tbxc6 25 'ilxc6 'iVc8 26 ':'a4! intending l:tac4, is strong. 24 'iVxd4 cxd4 25 l:tc4 Of course 25 tbd7 + c;1i>c8 26 tbxf8 .i.xf8 should be winning for White, but he probably preferred to avoid any blockading possibili ties. This almost imperceptibly weakens the back 2S •.• c;1i>c8 rank. Avrukh does not miss his chance to ex The only way Black can avoid various knight ploit this with some neat tactics. forks. 16 tbbS! 26 l:td1 .i.c5 (D) This is not a knockout blow, but it allows White to reposition his pieces very effectively. The knight was not doing much on c3, so it heads for c4 where it will have greater influ w ence. 16.. :ii'd8 After 16 ... cxb5 17 axb5 'ifd8 18 .i.f4+ c;1i>a8? (18 ... .i.d6 19 .i.xd6+ and 20 'iixb4) 19 b6! a6 20 'iic3 Black cannot defend his back rank be cause 20 ... tbc6 21 1:txa6+! leads to mate. 17 tba3! tbc718 as .i.e719 tbc4 (D)

B It is not so easy to breach Black's position, so White sacrifices the exchange back to seize the initiative. 27 l:txb4! .i.xb4 28 tbc6 .i.d6 29lh:d4 c;1i>d7 30 tbxa7 eS 31 l:.d1 1%a8 32 tbbS tbxbS 33 .i.xbS+ c;1i>c7 White is clearly better but it is not obvious how he should break through. Avrukh finds an elegant solution. 34 f4! f6 35 fxeS fxeS 36 f4! 92 CHESS EXPLAINED: THE MAIN-LINE SLAV

Again! The white rook will penetrate now. White can win a piece with l:txc7 and a8'iW at 36.•• exf4 37 e5 i.xe5 38 l:.d7+ g2! 1-0 xf3 b5 50 :txb5, which is sim 39.•. xf3 b5 50 a7 i.e7 43 i.e6 l:.e8 :xb5 and White will win Black's rook.

Game 21 Andrei Schekachev - Yannick Gozzoli Paris 2003

1 d4 d5 2 e4 e6 3 ttJf3 ttJf6 4 ttJc3 dxe4 5 a4 ttJa6 6 e3 (D) w

playing e4 (with bishops on b4 and f5 this would be harder) and ttJe5 is a possibility, be cause Black's queen's knight is on b4 instead With this modest continuation, White steers of d7. In order to advance in the centre, White the game towards positions similar to those dis will have to play g4 at some point. Black's hope cussed in Game 4. In this case Black is able to is that this loosening of White's position, com pin White's f3-knight immediately, but White bined with Black's inherently solid position, hopes to demonstrate that the a6-knight will will give Black enough counterplay. All in all, have less influence on the position than it would it seems that White's growing space advantage ond7. will secure him better chances, but he must be 6 ..• i.g4 7 i.xe4 e6 8 h3 i.h5 careful not to overextend. After 8 ... i.xf3 9 'ilixf3 ttJb4 100-0 i.e7 11 11l:.dl l:tdl 0-0 12 'ilie2 White has a pleasant edge, so The immediate 11 g4 is also dangerous. Black maintains the tension. White hopes to use the rook on the f-file. After 90-0 ttJb410 'ilie2 i.e7 (D) 11. .. i.g6 12 e4 ttJd7 (12 ... 0-0 13 i.f4 h6 14 The position is similar to those in Chapter 1, :adl i.h7 15 i.g3 'it>h8 16 ttJe5 tiJd7 17 ttJxd7 but here Black's minor pieces are arranged a lit 'ilixd7 18 d5! favourably opened the position for tle differently. His knight is on b4 instead of his White in Gasanov-Arzumanian, Istanbul 2(07) dark-squared bishop, and his other bishop is on 13 i.f4 0-0 14 l:.adl :e8 15 i.g3 h6 16 ttJel! h5 instead of f5. Here White has little trouble i.h7 17 f4 ttJb6 18 i.b3 c5 19 dxc5 i.xc5+ 20 BLACK AVOIDS 5 ... Ji.f5: BRONSTEIN AND SMYSLOV VARIATIONS 93

~h2 'ile7 21 f5 White kept up the pressure in somewhat off-balance in Bu Xiangzhi-Gretars Sorokin-Bareev, Russian Ch, Elista 1996. son, Reykjavik 2000. 11 ••• 0-012 g4 13••• l£ld7 (D) The immediate 12 e4 can be answered by 12 ... ~xf3 because 13 'ilxf3 allows 13 ... ttJc2 14 :bll£lxd4. Instead 12 as is a standard idea, by which w White hopes to cramp Black on the queenside. Black then has: a) 12 ... :'cS 13 ~b3 c5 14 l£lb5 ~xf3 15 gxf3 a6 16 dxc5 l£lbd5 17 l£ld6 ~xd6 IS cxd6 'ifxd6 19 :a4 l:tc5 20 f4 was Bacrot-Smyslov, Match (game 6), Albert 1996. This has been considered to be clearly better for White, but I think this is due to the result of the game. I do not think Black's position is as grave as has been suggested, and one possible improvement is Hubner's idea 20 ... :fcS 21 ~d2 b5 22 axb6 'ilxb6, when Black's counterplay certainly looks White has a nice space advantage, but Black's sufficient. position remains very solid. b) 12 ... b5! is a good way to gain counter 14.i.f4 play. 13 .i.b3 l£ld7 14 e4 'ilc7 15 .i.e3 a6 16 White takes away the c7-square from Black's l:tdcl 'ilb7 (D). queen. Kramnik has preferred the aggressive 14 l£le5. Then 14 ... lDxe5 15 dxe5 'ilas 16 f4 %ladS 17 ~e3 h61S ~g2 ~h719 l:txdS :'xdS 20 l:tdl g5! gave Black counterplay in Kramnik-Ivan chuk, Amber Blindfold, Monte Carlo 1996. 14.• Jte8 Black plays a useful waiting move and dis courages any d5 ideas by White. 15.i.b3 Instead 15 as is met with the usual 15 ... b5 16 .i.b3 'ifcs intending ... 'iVb7 and ... a6. White has also tried 15 i.g3; for example, 15 ... 'iIi'aS 16 h4!? h6 (after 16 ... h5 17 g5 Black is rather cramped, but this is safer nonetheless) 17 ~g2 it.h7 IS g5! h5 (following IS ... hxg5 19 hxg5 .i.xg5 20 lDxg5 'ili'xg5 21 %lhl, White's space, This is a typical method. The queen is snug bishop, and kingside initiative give him good on b7 and Black will soon be ready to break out compensation for the pawn) 19l£le5 l£lxe5 20 with ... c5. Mter 17 l£la2 l£lxa2 IS .i.xa2 %lacS dxe5 'ili'c5 21 f4 %ladS 22 i.f2 'ili'as 23 'ifxh5 19 g4 ~g6 20 l£ld2 c5! Black took over the ini and White was a pawn up in Ovechkin-Rublev tiative in N.Berry-Shaw, Edinburgh 2000. sky, Russian Ch, Tomsk 200 I. 12••• it.g6 13 e4 15.•• a6 Another idea is 13 l£le5 c5 (13 ... l£ld7 14 This looks odd. Perhaps 15 ... 'ifaS is a better l£lxg6 hxg6 15 as looks a little better for White) try. 14 h4 (White wants to harass the b6-bishop) 16.i.g3 14 ... cxd4 15 exd4 l£lfd5 16 h5 it.c2 17 :d2 Now 16 as! looks strong, because ... b5 is l£lf4 IS 'iff3 ~g5 19 l£la2, which left Black much less appealing for Black. 94 CHESS EXPlAINED: THE MAIN-LINE SlAV

16..• 'iYa5 (D) The knight cannot be tolerated; for exam 16 ... b5 looks more consistent, but White ple, 20 ... l:ted8 21 h5 ~h7 22 liJxf7! xf7 23 can ignore this advance. This plan is better for ~xe6+! f8 (23 ... xe6 24 'iVc4+ f6 25 e5+ Black when White has played a5 first, because g5 26 'ii'e2 is out of the question for Black) 24 then Black can prepare ... c5, whereas here the e5 and White has a very strong attack. pressure on the b5-pawn makes the freeing ad 21 ~xd6 c5!? vance of the c-pawn more difficult. Black is hoping to get some counterplay. 21 ... a5 is very passive, but more solid. 22 l:tacl cxd4 23 ~xb4 'iYb6 (D) Black cannot play 23 ... dxc3 because of 24 ':'xd7.

17liJd2! White takes aim at the d6-square. 17... 'ii'b6 This looks rather odd, but Black's position is not easy. 24liJb5 18 liJc4 'ii'a7 19 h4 This enables White to keep his bishop-pair. White could also clamp down on the queen 24..• axb5 25 a5 'ii'a7 26 'iYxb5 liJf6 27 f3 side with 19 a5 because 19 ... b5 20 axb6liJxb6 The position appears to be very unpleasant 21liJe5 is very nice for White. for Black, but after his next move it becomes 19••• h6 20 liJd6 (D) apparent that matters are not so clear. 27 •.. 'i¥b8! The queen wiggles back into the game. Black would like to play 27 ... e5 but 28 h5 ~h7 29 B l:tc7 is too strong, so Black looks for counter play against White's weakened kingside. 28e5 28 g2? fails to 28 ... liJxg4! 29 fxg4 ~xe4+ 30 h3 'ii'f4, when White is in big trouble. 28 ••• liJd5 29 ~d6 'iYd8! 30 .txd5 :xa5 31 'iYxb7 ':'xd5 32 h5 .th7 This is fine, but it was simpler to play 32 ... 'iYh4 with the idea 33 g2 i..e4! 34 fxe4 'iWxg4+ with perpetual check. 33 g2 'iYa8 Now 33 ... 'ii'h4 can be met with 34 ~e7! White grabs the bishop-pair. trapping the black queen. Taking the bishop al 20••• ~xd6 lows back-rank mate. BLACK AVOIDS 5 ... #5: BRONSTEIN AND SMYSLOV VARIATIONS 95

34 'ii'xa8 l:.xa8 35 l:.c4 d3 36 l:tdcl (D) advantage. The modern tries 9 i.e3 and 9 g3 look especially dangerous for Black, although the positions that arise are quite complicated. The deceptively quiet 7 h3 also leads to great B complications if both sides head down the main line. Black has to avoid some tricks, but even if he successfully navigates the right path, it is not easy to equalize. Probably Black will have to investigate options at move 14 or 15 which are mentioned in the notes to Game 18. If all of these possibilities are not enough to worry Black, the rare 7 4Jxc4 should also give Black something to think about. If White plays quietly with 7 g3, Black should equalize without too much trouble. It is easier for Black to develop and create counterplay Black must be careful because of his back than in similar positions with Black's bishop on rank. f5. 36••• ~h8 Smyslov's 5 ... 4Ja6 is much easier to learn. 36... f6? allows 37 l:.c8+ ltxc8 38 l1xc8+ Black may not equalize, but the positions are ~ 39 :tf8#, but 36 ... g5, creating some breath quite solid and may suit positional players ing room, is sensible. without the time or inclination to study the 37 ~f2 d2?! 38 l:.dl main lines with 5 ... i..f5 or the complexities of Black will not be able to hold on to the pawn. 5 ... .i.g4. The ambitious 6 e4 is still a good try 38••. f6 39 f4?! for an advantage. Black has to decide whether 39 :tc5, with the idea 39 ... l:.d3 40 ~e2, is to double White's kingside pawns. In any case, more effective. White's space advantage gives him a pull and 39••. g6? Black will have to hunker down and hope to get This does not help. After the paradoxical some eventual counterplay. Black must be care 39 ... ~g8 Black could fight on. ful not to fall too far behind in development, be 40 i..e71-0 cause White may open the position with d5 or Perhaps Black lost on time here, but his posi attack along the g-file. tion has become pretty difficult in any case. The simple 6 e3 also gives White good chances of an edge. Here, too, Black must be Conclusions patient and wait for his chance. I think White should refrain from advancing with a5, because The Bronstein Variation. 5 ... .i.g4. is looking this can give Black the counterplay he seeks quite risky. In the main line with 7 f3 4Jfd7 8 with ... b5 and ... a6, followed by bringing the 4Jxc4 e5, White has several promising options. queen to b7 in preparation of ... c5. If White The traditional 9 4Je4 does not look too bad for leaves the queenside alone, Black may play Black if White heads into the endgame with 11 ... a5, which is solid but does not actually create i.xb4, but 11 4Jxe5 looks good enough for an any active play for Black. 7 White Avoids 5 a4: The Quiet 5 e3 and the Geller Gambit

In this chapter we examine lines where, after I d4 d5 2 c4 c6 3 liJf3 liJf6 4liJc3 dxc4, White avoids the main continuation S a4. Our main topics are S e3 and S e4. These two continuations may appear to be similar, but they lead to very different types of games. 5 e3 is a calm move that tends to lead to a quiet middlegame that resembles some lines of the Queen's Gambit Accepted. Black will usually develop his queen's bishop to a6 or b7 instead offS or g4. White can recover the pawn quickly in the S e3 line if he so chooses. On the other hand, S e4 (the Geller Gambit) is a real pawn sacrifice. White bets everything on his attacking chances, which although dangerous, should not trouble Black if he knows his stuff.

The Games

Game 22 (Grivas-Li Shilong) covers the line S e3 bS 6 a4 b4 7liJbl. White does not immediately recover the pawn, instead opting to develop his kingside. Black is able to develop naturally enough while holding on to the pawn even though White can eventually win it back by force. Nevertheless, eventually recovering the pawn with the typical manoeuvre liJbl-d2xc4 takes time and Black de velops with no problems. When White forces the pace in an even middlegame, Black gradually takes over the initiative and wins with a direct attack on White's king.

Game 23 (Nikolic-Atalik) sees White play S e3 bS 6 a4 b4 7liJa2, when the double attack on the b4- and c4-pawns immediately restores material equality. Again, the time spent by White with his queen's knight (in this case liJcl-b3) allows Black to develop with no problems. In a dynamically balanced middlegame, Black pushes a little too hard and forces an endgame where White's domi nance of the c-file gives him the better chances.

Game 24 (Wojtkiewicz-Donaldson) examines the main line of the Geller Gambit with S e4 bS 6 eSliJd5 7 a4 e6 8 axbS. After 8 ... liJxc3 9 bxc3 cxbS 10 liJg5 i.b7 II 'ifhS g6 12 -'g4 i.e7 13 i.e2 Black has two good defences to choose from: 13 ... liJd7 (as played in the game) and 13 ... i.d5 (which may be even better). White forces matters too quickly and ends up in a bad endgame. After mutual inaccuracies, White suddenly resigns in a position that offers chances to hold. The options for Black discussed in this game have gone a long way towards discouraging White from playing the Geller Gambit in general, and 8 axbS in particular.

Game 25 (Vitiugov-Amonatov) takes a look at the complicated 8liJgS. This leads to murky po sitions where White's compensation is rather vague. There are still unanswered questions here and the complicated positions will appeal to some players. Black plays the most forcing line, 8 ... h6 9 liJge4 b4 10 liJbl i.a6 llliJbd2liJf4. The strategic and tactical complications arising here lead to interesting play. In this game Black plays an interesting novelty and after both sides err in the com plications, Black gets a better double-rook ending which he eventually converts to victory. WHITE A VOIDS 5 a4: THE QUIET 5 e3 AND THE GEILER GAMBIT 97

Game 22 Efstratios Grivas - Li Shilong Wijk aan lee 2008

1 d4 dS 2 c4 c6 3 lbfJ lbf6 4 lbc3 dxc4 5 e3 6a4b4 White has two other continuations here that Black cannot hang on to the extra pawn, but are seen from time to time. Both involve a he can gain time to develop by attacking the kingside fianchetto for White: c3-knight. White now has a decision to make. a) 5 lbe5 can be met by either 5 ... lbbd7 6 7lbbl lbxc4 b5 7lbe3 iLb7 8 g3 c5 9 d5 a6 10 iLg2 g6 This looks passive, but the knight may come 11 0-0 JLg7 with a decent, though unusual posi quickly to d2 to attack the c4-pawn. The alter tion, or with 5 ... b5, which transposes to line 'b' native 7 lba2 is also popular. This is considered after 6 g3 iLb7 7 iLg2. in the next game. b) 5 g3 b5 (less good is 5 ... iLf5 6 JLg2 e6 7 .. .iLa6 because after 7 lbe5 White recovers the pawn Again, Black holds on to the pawn for as without the weakening a4 advance; however, long as he can. White can win it back, but this 5 ... g6 transposes to a decent line of the Grtin will cost some time. feld after 6 JLg2 iLg7 7 0-0 0-0, one point being SiLe2 that 8lbe5 can be met by 8 ... lbg4!) 6 JLg2 iLb7 The main alternative is 8 'ifc2. White wants 7 lbe5 a6! should be fine for Black. In similar to regain the pawn immediately, but avoids mov Catalan positions, ... e6 is played very early. Here ing his knights around to do so. Now Black can Black can delay this move, instead spending the equalize easily enough with simple play; for tempo consolidating his queenside pawn-mass. example, 8 ... e6 9 iLxc4 iLxc4 10 'ifxc4 'ii'd5 11 We now return to 5 e3 (D): lbbd2, when both 11. .. c5 and 11. .. lbbd7 should be fine for Black. However, the real test is the ambitious 8 ... b3! (D).

This quiet continuation avoids a lot of theory, but it gives White little chance of achieving an advantage. This makes it more difficult for White to re S••• bS capture the pawn. Not only does Black win Black must protect the pawn, even if he can time, but he gains b4 for his pieces and cramps not hold it. After 5 ... iLf5?! 6 iLxc4 White has a White's queenside. The risk of playing this way Dutch Variation without having spent a tempo is that if c4 does fall, the b3-pawn could prove on the weakening a4 advance. to be vulnerable. Practice has shown that if 98 CHESS EXPLAINED: THE MAIN-LINE SLAV

Black keeps playing vigorously, he gets excel lent chances. After 9 'it'dl e6 White has a few tries: w a) In reply to 10 ttJbd2, Black has the typical idea 10 ... 'it'd5 !. Black holds on to the pawn and the queen influences the light squares all over the board. After 11 i.e2 ttJbd7 120-0 (not 12 ttJxb3? cxb3 13 .i.xa6 'it'a5+) 12 ... l:tb8! White has nothing better than the artificial 13 ttJb 1, but both 13 ... ttJe4 and 13 ... ~b4 discourage ttJc3 and leave Black with better chances. b) 10 ttJe5 weakens the d4-square and Black can quickly seize the initiative with 10 ... c5!, when 11 i.xc4? .i.xc4 12 ttJxc4 'it'd5, hitting c4 and g2, is a disaster for White. Instead 11 ttJc3 make it easy. 13 i.a3 c5 14 i.xa6 ttJxa6 15 is more solid, but after 11. .. cxd4 12 exd4 both ttJe5?! (safer is 15 'iWd3 lDd5 16 %:tcl with 12 ... 'it'c7 13 'it'f3 .i.b7 14 'it'e2 .i.b4 (Burgess) equality) 15 ... l:tc8 16 'fib3?! (Black already has and 12 ... ~b4 13 ~xc4 .i.xc4 14 ttJxc4 ttJc6 a slight initiative, but this just does not work look good for Black. out) 16 ... cxd4 171ib7 l:tc7 18 'iWa8+ i.d8 19 c) 10 ~e2 c5! 11 0-0 ttJc6 12 ttJe5 l:tc8 13 .i.d6 'ifd5 and the complications favoured Black ttJa3 cxd4 14 ttJaxc4 'it'd5! was better for Black in Nikolic-Bareev, European Team Ch, Pula in Timman-Kasparov, Match (game 2), Prague 1997. 1998. One point is that 15 ttJd6+ ~xd6 16 10 lDbd2 ttJc6 i.xa6 can be met by 16 ... 'iVxe5 17 f4 'iVc5 18 Instead 10 ... c3 11 bxc3 bxc3 12 ttJbl 'ifa5 13 i.xc8 dxe3 19 ~hl 0-0 20 .i.a6 lDd5, when i.a3 ~e7 would transpose to the previous note. Black is much better (Kasparov). I1lDxc4 8.•. e6 White could also play 11 dxc5, when Black The immediate 8 ... c5 is also possible. One can respond with either 11. .. lDa5!?, transpos possibility is 9 0-0 (more solid is 9 ttJbd2 c3 10 ing to Sadler-Bareev above, or the straightfor bxc3 bxc3 11 lDbl 'iia5 12 0-0 e6 13 .i.a3 ward 11. .. .i.xc5. ~e7, transposing to the note to Black's 9th 11. .. cxd412 exd4 ~e7 (D) move) 9 ... ttJc6 10 dxc5 and now the aggressive 10 ... ttJa5! is the point of Black's early queenside play. After 11 ttJbd2 e6 12 b3 i.xc5 13 .i.b2 c3 14 .i.xa6 cxb2 15 i.b5+ r3;e7 16 w %:tbl a6 17 .i.e2 'iWb6 18 %:txb2 ttJd5 Black was very comfortable in Sadler-Bareev, Hastings 1992/3 because of the juicy outpost on c3. 90-0 c5 This is Black's most aggressive response. Black must still be a bit careful because his kingside development is lagging. 9 ... i.e7 is a very sound alternative. Then 10 ttJbd2 c3 (this is a typical idea to disrupt White's position, but 10 ... 0-0 is playable as well) 11 bxc3 bxc3 12 ttJbl 'iVa5 (D) brings about a position that can be reached in various ways. Black has a very acceptable isolated queen' s White should be able to win back the pawn pawn position. All of his pieces are active and without trouble, but Black does not need to the d4-pawn is well blockaded. WHITE AVOIDS 5 a4: THE QUIET 5 e3 AND THE GElleR GAMBIT 99

13 ttJee5 The simplifications do not help White to achieve anything. Instead 13 ttJd6+? can be met w by 13 ...... xd6 14 .i.xa6 ttJxd4 winning a pawn, but White could keep the tension with 13 .i.e3 0-014 :tc1, when the game is about equal. 13••• .i.xe2 14 'ii'xe2 ttJxe5 15 dxe5 ttJd5 (D)

21 'iVe4 'ii'g5 22 g3 'ii'g6 Black offers up an endgame. Keeping the tension with 22 ... :tfd8 was a good option. 23 'ii'f3?! White should certainly exchange queens, when after either recapture Black has only a slight advantage. 23 .•. f6! (D) White has got rid of his isolated pawn, but it could become weak on e5 as well. White's bishop is obstructed by his own e-pawn, while the exchange of light-squared bishops reduces w White's attacking chances. The simplifications also have left Black with the strongest minor piece - the knight which is cemented on d5. 16 ttJd2 The endgame arising after 16 'ii'b5+ 'ii'd7 also leaves Black with good chances. 16... 0-017 ttJe4 :teSlS b3 This enables Black to seize the initiative. In- stead 18 .i.d2 is solid enough, although the sim plifications following 18 ... ttJb6 19 ttJxb6 'ii'xb6 20 .i.e3 .i.c5 21 .i.xc5 l:txc5 still slightly favour Black - the e5-pawn is more likely to be a Black is alert and transforms one advantage weakness than a strength. into another. Even though this allows White to IS•.. ttJc3! activate his bishop, Black will quickly get a This temporary invasion disturbs White's strong attack along the f-file. piece coordination. 24 exf6 ttJxf6 25 'ii'e2 19 'ii'e2 .i.e5 20 .i.b2 Trying to exchange to safety with 25 .i.xf6 If 20 .i.d2, Black can play 20 ... ttJe2+ 21 ~h1 does not solve White's problems after 25 ... .:xf6 ttJd4 with the idea ... ttJf5 with some initiative. 26 'ii'e2 :cfS 27 ttJe3 ~e4 28 :ae1 :f3, when 20 ... ttJd5 (D) White is tied up. Now that White has committed his bishop to 25 •.. ttJe4 26 ttJe3? (D) the passive b2-square, the knight simply returns This just loses. White had to concede a pawn to d5. with 26 ~g2 ttJxf2 27 ttJe5. 100 CHESS EXPLAINED: THE MAIN-LINE SLAV

26••• ~g5! Not only is White vulnerable on the light B squares around his king, but the c2-square will prove to be decisively weak. 27~dl It is too late for 27 ~g2 due to the reply 27 .....e4+. 27..• .i.xf2+ 28 ~xf2 llc2! 29 "e3 Itfxf2! 30 lhf2 ~h3+ 31 ~g2 lhf2+ 32 xh3 32

Game 23 Predrag Nikolic - Suat Atalik Sloven ian Team Chi Bled 2002

1 d4d52c4c63~f3~f64~c3dxc45 e3 b5 White can also play 10 'iWe2 immediately. 6 a4 b4 7ltJa2 (D) After 10 ... Ji.e7 (I would prefer 10 ... a5, prevent ing the coming exchange and heading towards the continuation of the main game) White ex changed bishops with 11 Ji.a6 Ji.xa6 12 'ii'xa6 B 'ii'b6 13 'ii'c4 0-0 14 Ji.d2 a5 15 ~c1 c5 16 ~b3 l:.fc8 17 l:.fc1 in Serper-Bacrot, Cap d' Agde 2000. Black should be alright here, but he needs to be careful on the queenside light squares. 10••• aS 11 ~cl (D)

This leaves both the b4- and c4-pawns at tacked and thus ensures that White will restore material equality immediately. However, the a2-knight is not well placed, and the time White spends getting this piece back in the game will allow Black to develop easily and achieve the ... c5 break. 7 ••• e6 8 .i.xc4 ~bd7 Black has several possibilities such as 8 ... a5, 8 ... .i.e7 and 8 ... .i.b7, but this is the most flexi Black should not experience any trouble get ble continuation. ting in the ... c5 break. Now he can arrange his 9 0-0 .i.b7 10 .i.d2 pieces in two different ways. WHITE A VOIDS 5 a4: THE QUIET 5 e3 AND THE GELLER GAMBIT 101

11 •• ~d6 18'it'bS Black intends to place his queen on e7. The If White wants to avoid weakening his king alternative is 11. .. iLe7. Black will play ... 1fib6, side, then this is the only move. IS .i.e 1? is a ... 0-0, ... :idS and ... :acS. After 12ltJb3 0-0 13 mistake because IS ... 'ifd6 19 h3 iLxd4 20 exd4 .e2 c5 14 :tfdl 'irb6 15 ~el l:tacS 16 dxc5 (20 l:tdl? 'iVc6) 20... 'iVxd4 gives Black a healthy ltJxc5 17 ltJfd4 l:.fdS 18 f3 ltJxb3 19 iLxb3 .i.c5 extra pawn. 20 ~f2 iLxd4 21 exd4 ~d5 Black had a good 18••• iLxd4 position in Czamota-Fressinet, European Ch, Black exchanges bishop for knight in order Kusadasi 2006. to create some imbalance. IS ... ltJd5 is a good 12 'it'e2 c5 13 ltJb3 0-0 14 :rc1 alternative. Then the d4-knight is still pinned, It seems rather strange to place the king's and Black has ideas such as ... iLa6 or .. .l:UdS. rook here. A more logical development is 14 One tactical point is that 19 f3 can be met by :tfdl 'fIe715 .i.el l:tfdSI6l:acl cxd417ltJfxd4 19 .. :it'f6!, winning material. ltJc5 with an equal position, Iotov-Atalik, Euro 19 exd4 iLdS 20 ~d1 pean Ch, Dresden 2007. Of course White hangs on to his light-squared 14.••• e7 bishop. The last thing White wants is a knight Black has developed harmoniously and is on d5 against his inferior dark-squared bishop. ready to bring his rooks to the centre. White re 20••. ltJe4 21 .i.e3ltJd6 22 'iVd3ltJfS 23 :c5 leases the central tension in order to give his l:fd8 24 !:tac1 (D) pieces some scope. IS dxcSltJxc5 16ltJfd4 White intends to play ~e 1, f3, iLf2 and e4 to gain space and limit the scope of the b7-bishop. B This is a slow but dangerous plan for the un wary. Black immediately takes concrete mea sures against this idea. 16••• ltJxb3 17 .i.xb3 After 17ltJxb3, 17 ... ltJe4 with the idea ... 'ii'h4 gives Black some initiative. All of his pieces are pointed at the white king. 17 ••• iLeS (D) Black creates the threat of ... 'ifd6, forking h2 and d4. The d4-knight cannot move because b2 is hanging. 17 ... ltJe4 IS iLel l:fdS was a good alternative. Black has the better pawn-structure and his minor pieces are good, but White is not worse thanks to his control of the c-file and pressure against the as-pawn. w 24•. :iff6 2S ':'c7 'ii'g6 26 g3 This weakens the long diagonal, but White correctly assesses that this is not fatal. 26 f3? can be met by 26 ... e5 !, because 27 dxe5 .li.xf3! wins for Black. 26... ltJxe3?! Now 26 ... e5? 27 dxe5 iLf3? loses to 2S %:td7 because there is no mate on g2. However, keep ing queens on with 26 ... 'iff6 would hold the balance. 27 'iVxg6 hxg6 28 fxe3 eS 102 CHESS EXPLAINED: THE MAIN-LINE SLAV

Black is playing aggressively, hoping to break 36••• I:.xb5 37 i.c6 :bb8 38 i.xe8 ':xe8 39 into White's position. Nikolic responds coolly. I:.b7 f5 40 I:.xb4 g5 41 :b5 g6 29 f2 A better try was 41 ... I:.f8 to try to push the f 29 dxe5 i.e6 30 i.f3 I:.a6, intending ... :d2, pawn to get some counterplay. was Black's idea. 42 b4 eM7 43 I:.e5 I:.b8 44 b5 f6 45 ':cS f4 29 •.• exd4?! 46 e4 I:.e8+ 47 l:.e5 lth8 (D) This allows the white king to come to e3. Black should have kept the tension. Atalik gives 29 ... i.e6 30 i.f3 I:.ab8! 31 I:.1c5 exd4 32 ::'xa5 d3! with counterplay. w 30 exd4 i.e6 31 i.f3 I:.ab8 The active 31. .. I:.a6 can be met by 32 d5!, when Atalik gives 32 ... i.d7 (32 ... i.xd5? 33 ltc8 :ad6 34 I:.xd8+ I:.xd8 35 I:.d1) 33 i.d1! f8 34 I:.1c5 e7 35 I:.5c6! with a big advan tage for White. 32 e3 i.b3 33 I:.lcS ::'e8+ 34 d3 (D)

B 48 gxf4 It was more accurate to play 48 b6 fITst. Then 48 ... I:.xh2 49 gxf4 gxf4 can be answered by 50 ':b5!. 48 ••• gxf4 49 b6 f3 This trick prolongs the game, but doesn't change the result. 50 xf3 I:.h3+ 51 e4 :b3 52 d5 I:.xb6 53 I:.e2 ':a6 54 c5 :a5+ 55 c6 :a6+ 56 ~b5 Black does not have enough room to keep White's active rooks give him good winning checking. chances. 56••• :a3 57 d5 g5 58 d6 I:.b3+ 59 ~c6 :c3+ 34•.• i.xa4 35 I:.xa5 i.b5+ 36 ':'xb5! 60 ~b7 ~f7 61 d7 I:.b3+ 62 ~c7 :c3+ 63 White simplifies into a rook ending. His ac d8 I:.cl 64 I:.e7 + f8 65 I:.e5 I:.gl 66 :t5+ tive king and passed d-pawn give him a decisive g71-0 advantage. 67 e7 l:.e1 + 68 d6 ltd 1+ 69 ltd5 wins.

Game 24 Aleksander Wojtkiewicz - John Donaldson Continental Open, Las Vegas 2001

1 d4 d5 2 c4 c6 3 ltJf3ltJf6 4ltJc3 dxc4 5 e4 (D) Black stands quite well, but White is not with The Geller Gambit is White's most aggres out attacking chances and Black must be very sive response to the Main-Line Slav. Unlike 5 careful in the opening. e3, 5 e4 constitutes a real gambit. Theoretically, 5 .•• b5 WHITE A VOIDS 5 a4: THE QUIET 5 e3 AND THE GELLER GAMBIT 103

and now White should avoid 14 tt)f6+? gxf61S 'fixh8 tt)xd4!, when Black is much better, and B prefer 14 tt)xfS :xfS, when he has won his pawn back and has the two bishops to boot. 12 'ii'g4 i.e7 13 .lie2 White has an alternative in 13 h4. The threat of hS forces Black to react. He can try: a) 13 ... hS is the nonnal move. Black stops the h4-pawn in its tracks and gains time by at tacking the white queen. 14 'fif4it.xgS IS hxgS and now: al) IS ... JldS?! 16 :h4 :g8 17 g4 hxg4 18 :h7 'ii'd7 19 i.a3 gives White good attacking chances. Black defends the c4-pawn and threatens a2) IS ... tt)c6 16 :h4 tt)e7 17 g4 tt)d5 trans ... b4, undennining the defence of White's e4- poses to line 'a3'. pawn. a3) IS ... tt)d7 16 lth4!? (White wants to 6e5 play g4 without allowing ... h4, blocking the Spassky used to have an affinity for 6 'ilc2 kingside) 16 ... tt)b6 17 g4 tt)dS 18 'fig3 'fIIc7 but this is not considered dangerous. Black can (18 ... «iti>d7 is another idea) 19 gxhS gxhS 20 g6 develop nonnally while holding on to his extra fxg6 21 'ii'xg6+ 'iff7 22 'iVg3 :g8 23 'ifh3 pawn as long as he is wary of a dS break. 0-0-024 ':xa7 ':df8 gave Black the initiative 6 •.• tt)d5 7 a4 e6 (D) in Feher-Hajnal, Fuzesabony (rapid) 2004. b) 13 ... h6!? 14 tt)e4 'ifdS IS f3 as is inter esting. It is not so easy for White to breach Black's defences, and trouble is brewing on the w queenside. 16 hS gS 17 ~e2 tt)d7 18 0-0 b4 was better for Black in Olsson-N.Pert, European Team Ch, Gothenburg 2OOS. We now return to 13 ~e2 (D):

Black has tried other moves, but this move is the main line and it is rightfully more popular than all other tries combined. 8axb5 White clarifies the position. The murkier alternative 8 tt)gS is considered in the next game. 8 ... tt)xc3 9 bxc3 cxb5 10 tt)g5 White threatens both 'fif3 and tt)xf7. This is the main position for the Geller Gam 10... it.b7 11 'iVh5 g6 bit. White has ideas such as tt)e4, .lif3, .lia3 or 11...'ii'd7 12 Jle2 (12 tt)xh7?! ~6 13 ~xfS? h4. Black would like to stabilize the position 'fixd4! is a clever trap) 12 ... ~dS 13 ~xh7 tt)c6 and secure his king's safety. Black has two 104 CHESS EXPLAINED: THE MAIN-LINE SLAV well-tested schemes of defence, so he has a ma For a period in the 1980s, this continuation jor decision to make here. was thought to be promising. White attacks 13••• ttJd7 the b5-pawn and seems to threaten 15 .i.xc4 as 13 ... .i.dS is just as popular. Black anchors well. It turns out that the latter threat is not his bishop in the centre and overprotects the real, so White should consider the alterna e6-pawn. White has: tives: a) 14 ttJe4 intends .i.h6, which both blocks a) 14 h4 h5 15 'iVg3 ttJb6 (15 ... .i.d5 is also Black's h-pawn and prevents him from castling. possible) 16 .i.f3 (both 16 0-0 a5! and 16 .i.a3 14 ... h5 15 'ii'f4 ttJc6 16 :a6 (16 .i.f3 transposes .i.xa3 17 :xa3 a5 favour Black) 16 ... .i.xf3 !? to 'bl' below) 16 .. :ifc8 17 ttJf6+ .i.xf6 18 'ii'xf6 (16 ... .i.dS) 17 'fIxf3 0-0 is interesting. One point .:tg8 19 :al 'ii'd8 20 'ii'f4 and now 20 ... a5?! 21 is that 18 g4 can be met by 18 ... .i.xg5 19.i.xg5 0-0 was unclear in Porper-Solak, European "dS!. Ch, Dresden 2007. I prefer the cold-blooded b) 14 .i.f3 (D) is the main continuation. 20 ... .i.xg2! with the idea 21 J:.gl.i.d5 22 .i.xh5 b4! with good play for Black. b) 14 .i.f3 and here: bl) 14 ... h5 15 'ii'g3 ttJc6 160-0 (16 ttJe4 b4 B 17 .i.g5 .i.xg5 18 ttJxg5 ttJe7 19 'ii'f4 0-0 20 cxb4 'ifb6 21 0-0 .i.xf3 22 ttJxf3 :fd8 23 l:.fd 1 c3 was good for Black in ReefschUiger-Htibner, Bundesliga 1983/4, but White may improve somewhere here) 16 ... b4! 17 :a6.c8 18 .i.xdS exd5 19 e6 'ii'xa6 20 exf7+~! 21 :el 'ii'b7! is probably insufficient for White, as pointed out by Burgess. b2) 14 ... ttJc6 150-0 a5 16 ttJe4 h6 17 ttJd6+ .i.xd6 18 .i.xd5 exd5 19 exd6 'ii'xd6 20 .i.a3 b4 21 cxb4 ttJxb4 left White with insufficient compensation for the pawn in Kahn-Portisch, Then 14 ...•c8! is a key move. Black con Hungarian Team Ch 2002/3. Black's king can tests control of the long diagonal and the subtle scoot to g7 if necessary. coverage of the e6-pawn will give Black the de We now return to the position after 13 ... ttJd7 fensive resource ... f5 in many lines. White has: (D): bI) 15 .i.a3 fails to 15 ... .i.xf3 16 ttJxf3 .a6!. b2) 15 .i.xb7 .xb7 16 tiJe4 O-O!? looks fine for Black. b3) 15 tiJe40-0 16 h4 (16 .i.h6 f5!) 16 ... f5 w 17 exf6 ttJxf6 18 ttJxf6+ ':'xf6 gives White a lit tle bit of compensation, but objectively it looks insufficient. 14••• a6! 15 .i.xc4 It is probably better just to play 15 .i.f3 but the insertion of 14 l:.bl and 14 ... a6 is likely to be in Black's favour. For one thing, .i.a3 ideas are off the table. 15..... c7! (D) 16.i.xe6? This should lose quickly. 16 ttJxf7? .xc4 17 ttJxh8 is no better because of the tactical 14:bl shot 17 ... ttJxe5 !. WHITE A VOIDS 5 a4: THE QUIET 5 e3 AND THE GELLER GAMBIT 105

18..... c6 19 f3 lle8 20 cxb5? (D) w

16 .i.e2 is relatively best for White, although 16 ... 'il'xc3+ 17 .i.d2 'it'c2 18 ':c 1 'ifb2! leaves White struggling. For example, 19 liJxf7 r3;xf7 Exchanging queens is the only way to pro 20 Ilc7 1IIb1+ 21 .i.d1 'iVe4+ 22 "xe4 SLxe4 long the fight. (Burgess) is better for Black because of his 20.•• axb5? good development and two connected passed Black returns the favour. 20 ... 'ii'c3+ 21 ~d2 pawns. 'it'xd4 wins; after 22 bxa6.i.d5 Black has an ex 16••. fxe6?! tra piece and the initiative. Even stronger is 16 ... liJxe5!, when White is 21 0-0 'iixe6 22liJxe6 lU7 23 l:tdl at a loss for a move. 23 l:[xb5? SLa6 costs White the exchange. 17 '6'xe6:t'8 (D) 23••• b4 24 .i.d2 ~d5 Instead 24 ... :c4 25 I:tdc1 Ilxc1 + 26 ':xc1 liJfS 27liJc7+ ~d7 28 d5 is not so clear. 25liJf4 (D)

White may have a couple of pawns for the piece, but with no light-square control, it will be difficult to prevent Black from quickly tak ing over the initiative. White is starting to get some counterplay, so 18e4 Black sacrifices the exchange. Relatively best was 18 0-0 although after 25 ...:xt4 18 .....c6! 19 "xc6 SLxc6 20 liJe6 liJb6 21 This regains the initiative, but Black is start liJxfS xfS Black's minor pieces will dominate ing to get low on pawns. the rooks. 26 .i.xf4 b3 27 :d3 liJf8 28 l:dxb3 106 CHESS EXPLAINED: THE MAIN-LINE SLAV

This appears desperate, but it is probably the Resignation seems a bit premature. After 31 best chance. The real surprise comes in just a l:.d3 i.c5 White's pawns will begin to fall, but couple of moves. the limited number of pawns would still give 2S .•. i.xb3 29 ':'xb3 ~e6 30 i.e3 l:tdS 0-1 him some chances to hold.

Game 25 Nikita Vitiugov - Farrukh Amonatov Russian Team Ch, Sochi 2006

1 d4 dS 2 c4 c6 3 ~f3 ~f6 4 ~c3 dxc4 5 e4 bS ~b6 16 i.xe6! fxe6 17 :h3 'ifd5? (17 ... 'iVd7) 6 eS ~dS 7 a4 e6 S ~gS (D) 18 'ifg6 gave White a decisive attack in S.Wil This aggressive lunge is less explored than 8 liams-Collins, Bunratty 2006. axb5. 9 ~ge4 b410 ~bl i.a611 ~bd2 (D)

B B

White maintains the tension on the queen 11 .•. ~f4! side and hopes to lure Black into weakening his This is the most aggressive response. Black kingside. White threatens 9 axb5 ~xc3 10 bxc3 heads for the soft d3-square. 11. .. c3 has also cxb5 11 'iYf3, so Black must drive away the been tried, but after 12 ~c4 i.xc4 13 i.xc4 knight. cxb2 14 i.xb2 i.e7 15 0-00-0 16 'ii'g4 White S••• h6 has reasonable compensation for the pawn. This forcing move may seem to fall into Markov & Schipkov's old suggestion 11. .. ~b6 White's plans, but it is a good way to fight for seems reasonable, even if it has not found any the initiative. The main alternative is 8 ... i.e7, adherents. with which Black hopes to avoid weakening his 12'ifg4 kingside. A couple of examples from many: The untried 12liJxc4 ~xg2+ 13 i.xg2 i.xc4 a) 9 ~ge4 b4 10 ~bl f5 has been considered is widely considered to be much better for Black. good for Black, but 11 exf6 ~xf6 12 ~xf6+ I actually think White still has some compen i.xf613 i.xc4 'ifxd414 'ifh5+ g615 'ife2 'ife5 sation here, although Black certainly has re 16 'ifxe5 i.xe5 17 0-0 0-0 18 ~d2 gave White sources as well. reasonable compensation for the pawn in Aron 12.•. ltJd3+ 13 i.xd3 cxd3 140-0 ian-Kramnik, Amber Rapid, Monte Carlo 2007. 14 ~d6+? i.xd6 15 'ifxg7 is refuted by the b) 9 h4 h6 10 ~ge4 b4 11 ~bl i.a6 12 positional exchange sacrifice 15 ... i.f8! 16 'ifxh8 'ii'g4! ~f8 13 ~bd2 c3 14 ~c4 i.xc4 15 i.xc4 'iVxd4 17 0-0 ~d7 18 ~f3 'ifg4! 19 i.xh6 WHITE A VOIDS 5 a4: THE QUIET 5 e3 AND THE GELLER GAMBIT 107

'ifh5! 20 i..g7 jfxh8 21 i..xh8 Jth6 (D), when gives some play for the pawn, although the d3- Black's pawns are too strong. pawn remains a trump for Black. 16... c5! 17 dxc5 lDxe5 18 lDe3 Instead 18 'iig3 lDc6 leaves White strug gling to justify his play. For example, 19 lDd6+ w i..xd6 20 cxd6 (20 'ii'xg7? i..e5) 20 ... 0-0-0!? is one possibility. Nevertheless, this was probably White's best try. 18.•• lDxg4 19 lDxd5 (D)

22 g4 (no better is 22 i..f6 lDxf6 23 exf6 d2, Van Wely-Van der Sterren, Dutch Ch, Eind hoven 1991) 22 ... rt;e7 23 g5 ':'xh8 24 gxh6 c5 25 l:Ue 1 c4 gave Black a winning position in Fiorito-Smyslov, Buenos Aires 1990. 14••. 'ii'd5!? A fresh and logical idea. Alternatives: a) 14 ... lDd7 15 lDd6+ .i.xd6 16 'ii'xg7 ':'f8?! 19•.• 0-0-0?! (16 ... i..f8 is not possible now because Black The cold-blooded 19 ... exd5! is much stron cannot take on d4 with his queen, but Burgess's ger. After 20 lDf6++ ~d8 21 l:.e8+ (21 lDxg4 suggestion 16 ... i..xe5!? 17 dxe5 ~e7 looks i..xc5 is also much better for Black) 21 ... ~c7 good) 17 exd6 'iVg5 181i'xg5 hxg5100ks better 22 ':xa8 (insufficient is 22 lDxd5+ ~b7 23 c6+ for White. Sacrificing the exchange with 19 ~xc6 24 ':xa8 ~xd5) 22 ... lDxf6 23 ':'xa7+ ttJe4!? d2 20 Jtxd2 .i.xfl 21 ':'xfl is one ap i.b7 24 c6 ~xc6 25 i.e3 lDe4 26 l:.cl+ i.c5 pealing option. White runs out of steam. b) 14 ... h5 15 'ii'f4 Jte7 16 lDf3!? (16 lDc5 20 lDxb4 i.b7 21 i.d2 ltd4 22 c6 i..xc5 17 dxc5 'ii'd5 with the idea ... lDd7 and It was worth considering 22 lDd6+ i.xd6 23 ... 0-0 should be at least equal for Black) 16 ... d2 cxd6, when White will get some counterplay on 17 Jtxd2 i..xfl 18 l:.xfl gave White some posi the c-file and the dark squares. tional compensation in Aronian-Arutinian, Eu 22•.. i.xb4 23 cxb7+ ~xb7 24 i..xb4 ':xb4 ropean Ch, Batumi 2002. 25 lDc5+ ~b6 26 lDxd3 ':b3 (D) c) 14 ... .:h7!? is a cute little computer sug Black has some pressure, but White should gestion. Black overprotects the g7 -pawn to dis hold. courage lDd6+. 27 h3?! 1511el The more active 27 ':e4 lDf6 (27 ... :xd3 28 After 15 lDd6+ Jtxd6 16 'ii'xg7 l:.f8 17 exd6 ':'xg4 gives White enough play) 28 ltd4 gives lDd7 Black is ready to castle queenside, with an Black nothing concrete. unclear position. 27.. ':xd3 28 hxg4 l:hd8 15 ... lDd7 16 lDn? Now White has a pawn on g4 (instead of an This allows Black to break up White's centre. active rook!) and Black has the more active 16 lDf3 is better. After 16 ... c5 17 dxc5 lDxc5 18 rooks and king. lDxc5 'iYxc5 19 i.e3, White's development lead 29:acl 108 CHESS EXPLAINED: THE MAIN-LINE SLAV

c3 w l:.g5 53l:.hS l:.xg2 54 l:[xh4 :f2 55 .:td4+ ~e5 56 l:!e4+ ~5 57 lh4 :xf3+ 5S

Conclusions

The quiet 5 e3 is a very solid continuation, but it is not dangerous for Black if he knows a few useful schemes of development. After 5 ... b5 6 a4 b4 7 ltJbl ~a6, the older continuation 8 "c2 is well met by 8 ... b3!. Here Black may White sheds a pawn in hopes of getting not hold on to his extra pawn, but with accu counterplay. 29 l:e4 l:.8d4 30 :ael a5 with the rate play he has excellent chances to seize the idea ... c5 is also difficult for White. initiative. 8 ~e2 is more solid, but Black can 29•• A3d4 30 l:.c3 l:[d1!? 31 l:tb3+ aS 32 equalize here with either the disruptive 8 ... e6 9 xe1l%d7 0-0 ~e7 10 lDbd2 c3 or he can play the aggres Black wins a pawn. sive 8 ... c5. 34 :b8 ~xa4 35 c;t>e2 ~aS 36 ~e3 Clta6 37 The continuation 7 ltJa2 leads to slower l:.gS g6 (D) play. White regains the pawn immediately but does little to disturb Black's smooth develop ment. Black can set up his pieces with ... ~e7 and ... 1Wb6 or with ... i.d6 and .....e7. In both w cases, Black achieves the ... c5 break effort lessly. The Geller Gambit leads to much sharper play, but White risks at least as much as Black does here. In the main line with 8 axb5 White gets certain attacking chances, but if Black knows the defensive schemes with either 13 ... ltJd7 or 13 ... i.d5 he has a solid position, an extra pawn, and ready-made counterplay on the queenside. The modem 8 ltJg5 is both more interesting and less explored, and it leads to complicated 3S ':'cS positions. 8 ... h6 remains the most reliable line After 38 l:!h8 b5 39 :xh6 ~c4 40 :h8 for Black, and after 9ltJge4 b4 10 ltJbl ~a6 11 l:[b7 Black will gobble up the b2-pawn. The rest ltJbd.2 Black has more than one viable continua takes some time, but Black gets there in the end. tion. Again, White has some attacking chances 38••• %:tb7 39 ':'c2

Evgeny Alekseev - Wang Yue, Russia-China match, Nizhny Novgorod 2007 71 Levon Aronian - Ivan Sokolov, Wijk aan Zee 2006 43 Boris Avrukh - Artur Kogan, Maalot-Tarshiha 2008 88 Vinay Dbat - Fernando Peralta, Badalona (open) 2006 59 Magnus Carlsen - Boris Gelfand, Tal Memorial, Moscow 2006 38 Daniel Fernandez - Renier Gonzalez, USA Ch, San Diego 2006 79 Kiril Georgiev - Du Xiangzhi, Gibraltar 2008 47 Efstratios Grivas - Li Shilong, Wijk aan Zee 2008 97 Penteala Harikrishna - Giovanni Vescovi, FIDE World Cup, Khanty-Mansiisk 2005 9 Rasullbrahimov - Elshan Moradiabadi, Abu Dhabi (open) 2007 63 Baadur Jobava - Alexander Grisbchuk, Calvia Olympiad 2004 50 Gregory Kaidanov - Zhang Pengxiang, World Team Ch, Beersheba 2005 86 Peter Leko - Boris Gelfand, Tal Memorial, Moscow 2006 24 Alexander Moiseenko - Ghate Swathi, Montreal 2006 29 Predrag Nikolic - Suat Atalik, Slovenian Team Ch, Bled 2002 100 Alexander Onishchuk - Alexander Grishchuk, Bie12007 67 Yannick Pelletier - Erwin L'Ami, Bundesliga 200617 21 Peng Zhaoqin - Liu Xianglin, Jinan 2005 54 Ruslan Ponomariov - Sergei Rublevsky, Candidates match (game 1), Elista 2007 74 Alexander Riazantsev - Zhao Jun, Aeroflot Open, Moscow 2006 83 Konstantin Sakaev - Garry Kasparov, European Clubs Cup, Rethymnon 2003 34 Andrei Schekachev - Yannick Gozzoli, Paris 2003 92 Sebastian Siebrecht - Fernando Peralta, Binissalem 2004 15 Nikita Vitiugov - Farrukh Amonatov, Russian Team Ch, Sochi 2006 106 Aleksander Wojtkiewicz - John Donaldson, Continental Open, Las Vegas 2001 102 Index of Variations

1 d4 d5 A) 2 c4 c65 3 ~f3 ~f6 6 ~e543 4 ~c3 dxc46

6 ••• ~bd7 5 a4 6 ... e6 677 f3: Or: a) 7 ... i.b4 67 a) 5 ~e5 97 b) 7 ... c5 71 8 e4: b) 5 g3 97 bl) 8 ... cxd4?! 71 c) 5 e3 97 5 ... b5 6 a4 b4: b2) 8 ... i.g6 74 cl) 7 ~bl 97 7 ~xc4 c2) 7 ~a2100 d) 5 e4102 5 ... b5 6 e5 ~d5 7 a4 e6: dI) 8 axb5 103 d2) 8 ~g5106 B 5 .•• i.f59 a) 5 ... ~a6 88: al) 6 e4 89 a2) 6 e3 92 b) 5 ... i.g4 79 6 ~e5 i.h5 79: bI) 7 g3 86 b2) 7 ~xc4 79 b3) 7 f3 79 b4) 7 h3 83 Now (after 5 ... i.f5): A: 6~e5 B: 6e3 7 ••• 'Wic743 7 ... ~b6 54 8 ~e5 a5 55: 6~h438 a) 9 e3 63 INDEX OF VARIATIONS 111 b) 9 f3 55 8 0-0 0-09 c) 9 g3 59 Or8 ... ~bd7: d) 9 .i.g5 63 a) 9 'i!Vb3 34 8 g3 eS b) 9 'ife2: 9 dxeS ~xeS43 bI) 9 ... .i.g621 10 .i.f4 ~fd7 b2) 9 ... i..g424 11 i..g2 c) 9 ~h4 34: cl) 9 ... 0-029 c2) 9 ... .i.g635

11 000 gS!? 47 11. .. f643 Now (after 11. .. g5!?): 9 'iWe29 12 ~e3 47 9~h429: 12 .i.xe5 50 a) 9 ... ~bd7 29 12 ~xe5 50 b) 9 ... i..g629 c) 9 ... .i.g430

B) 9 000 ~bd7 9 ... ~e415 6 e39 9 ... .i.g424 9 ... .i.g615

B w

6 000 e6 7 .i.xc4 .i.b4 10 e4 .i.g69