Chess Explained: The Nimzo-lndian

Reinaldo Vera

Translated by Phil Adams

[e]A[i!j]B0l1

First published in the UK by Gambit Publications Ltd 2008

Copyright © Reinaldo Vera 2008 English-language translation© Phil Adams 2008

The right of Reinaldo Vera to be identified as the author of this work has been asserted in accor­dance with the Copyright, Designs and Patents Act 1 988.

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 form 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- 1 3 : 978- 1 -904600-72-5 ISBN- 1 0: 1 -904600-72-7

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 582 1 . E-mail: orders@Centralbooks.com

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

Edited by Graham Burgess Typeset by Petra N unn Cover image by Wolff Morrow Printed in Great Britain by The Cromwell Press, Trowbridge, Wilts.

1 0 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 Gennan Editor: Petra Nunn WFM Webmaster: Dr Helen Milligan WFM

Contents

Symbols Introduction

1 Samisch Variation: 4 a3 or 4 f3 2 Capablanca Variation: 4 'iYc2 d5!? 3 Capablanca Variation: 4 'iYc2 0-0 4 Rubinstein System: 4 e3 0-0: Variations with etJe2 5 Rubinstein System: 4 e3 0-0 with �d3 and CLJf3 6 Fianchetto Variation: 4 g3 or 4 etJf3 c5 5 g3 7 Leningrad Variation ( 4 i.g5) and Other Lines

List of Games Index of Variations

4

5

7 20 3 1 51 71 86

97

109 110

Symbols

+ check ++ double check # checkmate !! brilliant move

good move !? interesting move ?! dubious move ? bad move ?? blunder Ch championship corr. correspondence game 1 -0 the game ends in a win for White 11z_11z the game ends in a draw 0- 1 the game ends in a win for Black (D) see next diagram

Introduction

1 d4 ttJf6 2 c4 e6 3 tt'lc3 .ib4 (D)

w

The Nimzo-Indian has arguably the best reputation of all defences against I d4. Most of the World Champions have relied upon the Nimzo at some point in their careers.

What is the reason for this? Basically it is due to the logic and simplicity of this defensive forma­tion, since with his first three moves Black obeys the most important opening principles: develop­ment, control of the centre and preparation for castling, to ensure the safety of the king. Moreover, he does so while retaining great flexibility regarding his eventual set-up and structure.

When Black uses the Nimzo-Indian it is very rare for him to fall behind in development, so that he is almost never overwhelmed by a lightning attack in the opening. The main and in fact only problem for Black is that he has to be willing to give up his dark-squared bishop, which means that his kingside is weakened slightly and White gains the famous advantage of the bishop-pair. How­ever, Black always obtains something in return for this strategic concession, whether it be better pawn-structure (doubled white pawns on the c-file, isolated or hanging pawns in the enemy camp, etc.) a pawn-majority in the centre, or a considerable advantage in development which enables him, with energetic play, to maintain the balance.

Black's best-known and most common playing methods of play and strategies in this defence are: • blockade of the position (to restrict the scope of the enemy bishops, with pawns on c5, d6 and

eS); • attack on the doubled pawn on the c-file ( ... b6, . . . .ia6, . . . tt'la5) ; • creation, blockade and siege of an isolated queen's pawn, or pressure against hanging pawns.

Nevertheless, we shall also see many other modem ideas where the only way to exploit White's lag in development is to play actively, even opening the position, with the aim of inflicting perma­nent weaknesses on White or obtaining some other compensation for the bishop-pair.

White has tried virtually everything to gain some advantage against this solid defence and at var­ious times it has looked as if he had finally achieved this goal with one line or another, but each time the defenders ofthe Nimzo-Indian have come up with a way to neutralize the possible advantage.

6 CHESS EXPLAINED: THE NIMZO-INDIAN

Chapter 1 is devoted to the Samisch Variation (4 a3 or 4 f3), owing to its importance for under­standing the basic ideas in the struggle which takes place between the bishop-pair and the superior pawn-structure or the better development.

There is no doubt that the most reputable lines for White are the Classical or Capablanca line 4 �c2 and the Rubinstein System 4 e3. To each of these systems we have therefore devoted two chapters, each presenting different methods of handling the position for Black.

In Chapter 2 we shall see the most energetic way to play against the Capablanca Variation (4 �c2 d5 ! ?), immediately breaking in the centre and by forceful play creating a whole series of com­plications. Black's main objective here is to punish White for the luxury he has allowed himself of losing several tempi in order to gain the bishop-pair without spoiling his structure.

In Chapter 3 (4 'ifc2 0-0) Black employs a more restrained method of play, giving priority to completing his queenside development before starting action in the centre.

We shall study the Rubinstein System (4 e3) in Chapters 4 and 5. From among the many ways of meeting it we have chosen to focus on the most popular and the hotly debated at the present time, concentrating on the move-order 4 . . . 0-0 followed by . . . d5.

The five games in Chapter 4 show how to play when White employs development systems based on moving his king's knight to e2, whereas in Chapter 5 we shall see games in which White opts for the more natural set-up with lt'Jf3 and ..id3.

The dangerous variation with 4 g3 (or 4 lt'Jf3 and g3) is the theme of Chapter 6. There we shall observe the differences created by the deployment of White's king's bishop on the long diagonal and we shall study two ways for Black to deal with this troublesome variation.

Chapter 7 is devoted to the Leningrad Variation (4 ..ig5) and other minor lines against the Nimzo-Indian Defence, among which the move 4 'i¥b3 stands out. In respect of the Leningrad Vari­ation, we shall study the gambit with . . . c5 and . . . b5, which has become popular again at the present time, as well as more traditional methods, but with an emphasis on putting pressure on the ad­vanced white pawn on d5.

The aim of this book is to offer the greatest possible coverage of the different systems that both sides can employ in the vast Nimzo-Indian complex. I have tried to include as many important vari­ations as possible with different strategic themes, but obviously we would need several volumes to cover the long history of this popular opening.

After studying this book, a player should have enough information to put this opening into prac­tice in tournaments. Furthermore, the reader will have enriched his arsenal of strategic knowledge, since this is one of the defences richest in positional ideas. If this is achieved, the author's objective will have been fulfilled.

GM Reinaldo Vera

1 Samisch Variation: 4 a3 or 4 f3

Introduction

The first chapter of this book is dedicated to the systems with 4 a3 or 4 f3. Why? For two funda­mental reasons. First, these variations present us with the most basic and characteristic struggle of the Nimzo-lndian Defence, where White obtains the bishop-pair or gains control of the centre and Black seeks counterplay against the doubled pawns or contests the aforementioned central occupa­tion. The second reason is that these are the most ambitious lines for White and the most dangerous for Black.

White's idea in both cases is obvious: in the case of 4 f3 the immediate occupation of the centre, and with 4 a3 the gain of the bishop-pair and the subsequent occupation of the centre with f3 and e4. They often transpose, which is why they are being studied in the same chapter.

These variations have always had the support of great players: Botvinnik and Spassky in the past, Shirov later on and at the present time a whole generation of GMs headed by Milov, Yakovich, Volkov and Moskalenko, to mention just a few of those who have investigated this variation, creat­ing many ideas on the white side.

Getting down to details, in the case of the direct 4 a3 we shall look at the system in which Black, after doubling the pawns on the c-file with 4 . . . .ixc3+, goes immediately after the c4-pawn (5 . . . b6, ... i.a6 and . . . tt:'lc6!?), which does not leave White with a great deal of choice. He must either attack on the kingside, in many cases sacrificing the c-pawn, or else the defend the pawn, tying down some of his pieces. Black has sufficient defensive resources in both cases, and there ensues a fight to the death in which the plans are quite clear for both sides.

The Games

In Game 1 (Radjabov-Dizdarevic), we shall see the most obvious ways for White to defend the c4-pawn: with �a4 or .id3 followed by �e2, as well as the cases where White decides to sacrifice the c-pawn for an attack, and we shall see how Black should react in each case.

On move 8 Black chose the wrong plan, ending up in a rather passive position, and his attempt to become active gave White the opportunity to finish off the struggle with an unusual attack on the queen side.

In the encounter Hillarp Persson-P.H.Nielsen (Game 2) we shall study a way of meeting 4 f3 which is closely related to the previous game ( 4 . . . tt:'lc6!?) , since if White plays 5 a3 the same posi­tion would be reached. Following the consistent 5 e4, Black breaks with 5 . . . d5 and after White ad­vances with e5 a position arises where White has a solid but quite extended centre, which gives Black good squares for his minor pieces and possibilities of striking with .. .f6 and the manoeuvre .. . i.a5-b6. In the game, White lost a few tempi and allowed the enemy bishops too much activity, while his king remained in the centre, which cost him material and eventually the game.

In Cebalo-Komarov (Game 3) we shall examine another way of dealing with 4 f3, in this case with the immediate 4 . . . 0-0!?, which has the virtue of lending more force to the break 5 e4 d5 ! , since

8 CHESS EXPLAINED: THE NIMZO-INDIAN

Black' s king is safer and he has the possibility of attacking the base of the pawn-chain (i.e. d4) with . . . c5. For this reason the best continuation for White is 5 a3 ! ?, but then Black has the original ma­noeuvre 6 . . .. �'Jh5 followed by . . . f5, as used by Komarov in the game, obtaining an excellent posi­tion, although mistakes later on allowed White to equalize the struggle with an ingenious pawn sacrifice.

Obviously there are many other ways for Black to confront the Samisch Variation (4 a3 or 4 f3), but I have chosen these lines on the basis of their soundness and practical successes.

Game 1 [E241

Teimour Radjabov - Emir Dizdarevic Calvia Olympiad 2004

1 d4 l2Jf6 2 c4 e6 3 l2Jc3 i.b4 4 a3 i.xc3+ Naturally, retreating the bishop makes no

sense here. 5 bxc3 (D)

B

From this position the basic plans for both sides are well-established. White's primary ob­jectives are total occupation of the centre with f3 and e4 (taking advantage of the fact that he has not developed his knight to f3) and attack on the kingside. Also among White's possible threats is the pin with i.g5, which can be un­pleasant for Black since he no longer has his dark-squared bishop. White's ideal development would be to form a pawn-centre with f3 and e4, place his bishops on d3 and g5 and his knight on e2 or h3, etc.

What is the fundamental defect of the white position? The weakness on c4 ! Based on that premise, Black should start to lay siege to c4 as soon as possible; the most direct method is to

deploy the bishop to a6 and the knight to aS, which causes White a lot of problems. If the c4-pawn falls and the exchange of the light­squared bishops is achieved without making too many concessions on the kingside, Black will not only have an extra pawn, but will have considerably limited White's attacking poten­tial. A final point: do not be in too much of a hurry to castle as Black! In general, if your op­ponent has the attack on the kingside as his main plan, there is no reason to rush to place your king there. It is preferable to give priority to the attack on the c4-pawn and only later de­termine the best position for the king, which on many occasions finds refuge on the queenside. This rapid concentration of force against the c4-point is one of the best ways of confronting the Samisch Variation.

5 ... b6 The actual move-order in this game was

5 . . . l2Jc6 and then . . . b6 and . . . i.a6. 6 f3 Pursuing the main idea; if White plays 6 e3

instead, then Black would continue with the standard plan of 6 . . . i.a6 and . . . l2Jc6-a5 . White can make immediate use of the pin with 6 i.g5 but after 6 . . . h6! (to have the unpinning move . . . g5 available) 7 i.h4 i.a6! (more accurate than 7 . . . l2Jc6 8 l2Jf3! and 9 l2Jd2! , defending c4 and supporting the advance e4) 8 e4 g5 ! 9 e5 gxh4 1 0 �f3 !? d5 1 1 exf6 i.b7 ! 1 2 �f4 l2Jd7 1 3 l2Jf3 h3 ! 1 4 c5 hxg2 1 5 i.xg2, instead of 1 5 . . . bxc5?! (Muci-Graf, Merida 2001 ), Black can take on f6 immediately by 1 5 . . . 'ihf6! 1 6 ii'xf6 ( 1 6 ii'xc7 l2Jxc5!) 1 6 .. . l2Jxf6 with an extra

SAMISCH VARIATION: 4 a3 OR 4 f3 9

pawn. Note that if Black had castled previ­ously, the advance with . . . h6 and . . . g5 would be extremely risky for him.

6 ... tt:Jc6 7 e4 .ta6 (D)

A key position for this variation. Faced with the threat of ... tt:Ja5 and the capture on c4, White has various plans.

He can defend the pawn with .td3 and �e2 (although this weakens the squares b3 and a4) or with �a4, which would take the queen away from the wing on which White supposedly will attack. He can also dislodge the knight from f6 by playing e5 before deciding how to defend c4.

In the game, White employed one of the most troublesome plans (8 .tg5 !), creating the threat of e5, although Black always has avail­able the response . . . h6 and . . . g5, which avoids loss of material.

8�g5! The most natural defence of the c4-pawn is 8

�d3 ctJa5 9 �e2 (D), but then the queen loses control of the squares b3 and a4.

This allows Black to eliminate the bishop­pair and head for a simple position without many problems by playing 9 . . . ctJb3 !? 1 0 .l::!.b 1 ctJxc l 1 1 .l::i.xc l ctJh5 followed by .. . �e7 !?. However, the ambitious 9 . . . d6! ? with the idea of . . . �d7-a4/c6 is even better.

The advance 8 e5 gains space and forces the knight back. However, after the retreat of the knight Black has .. . �h4+ available, which will be useful in many variations; for example, 8 ... ctJg8 9 ctJh3 (if he now defends c4 with 9 �d3 and �e2 then the check on h4 causes

B

problems: 9 . . . ctJa5 1 0 �e2 c5 ! 1 1 .te3 { 1 1 d5? .txc4! and . . . �4+ or 1 1 ctJh3 cxd4 1 2 cxd4 �4+ } l l .. ..l:.c8 1 2 .l::!.c 1 ctJe7 1 3 ctJh3 cxd4 1 4 cxd4 ctJf5 and Black has pressure on the white pawns) 9 . . . tt:Ja5 1 0 �a4 (D).

B

The e5-pawn gives White extra space on the kingside, allowing him to go on the offensive in that sector. Black should seek the exchange of queens, to eliminate a defender of the c4-square and reduce White's offensive arsenal, and to achieve this Black has two manoeuvres that the reader should remember: bringing the queen to c6 (via c8-b7-c6) or else placing the queen on e8 in order to play . . . d5 or . . . d6.

It is easy to understand why even 1 0 . . . 'it>f8 !? followed by .. . �e8 and .. . d5 ! can be considered in this position. However, the most common continuations are 1 0 . . . h6 !? (controlling the g5-square) and 1 0 ... ctJe7 ! . Let us look at a few of White's attacking ideas after 1 0 . . . tbe7 ! :

10 CHESS EXPLAINED: THE NIMZO-INDJAN

a) I I .id3 h6!? 1 2 lbf4 0-0 13 0-0?! (it is better to advance against Black's king with 1 3 h4, but after 1 3 . . . d6! hitting the white centre, followed by a timely . . . �e8 !, Black is fine) 1 3. .. d6 14 .Ud l �e8 ! 1 5 �c2 .ixc4, Gutman­Karner, USSR 1 975.

b) I I .ig5 0-0 (necessary to be able to ad­vance the d-pawn and to free the e8-square for the queen) 1 2 .id3 and now both 12 . . . �e8 ! and 1 2 . . . h6 1 3 .ih4 �e8 ! put into practice one of Black's main ideas when the white queen is on a4: the threat is . . . d5 or . . . d6, offering the ex­change of queens, which forces White to rush into the attack.

c) The dangerous I I lbg5 is also answered by l l . . .h6 ! 12 lbe4 0-0 and now in reply to Gutman's recommendation 1 3 .if4, 1 3 . . . �e8 !? is again very interesting, and although White has many ways to continue the attack, none seem to give him any advantage. For example, 14 .ixh6 d5!? or 14 �c2 !? lbg6! 1 5 .ixh6 f5 ! 16 exf6 gxh6 and Black holds.

From these variations we can conclude that when the white queen is on a4 and the white pawn has advanced to e5, Black's strategy of breaking in the centre with . . . d6 or . . . d5 and of-fering the exchange of queens with .. . �e8 !? is quite effective.

8 .• .'iVc8?! This is the cause of Black's future difficul­

ties. He allows the set-up with .id3 and �e2, which differs from what we saw earlier (with­out the inclusion of .ig5), since in that position Black was able to eliminate the dark-squared bishop with the immediate . . . lbb3, or attack the c4-pawn with . . . d6, . . . �d7 and . . . �a4.

Instead of bringing the queen to c6 (which is effective when the white queen is on a4), Black should play 8 . . . lba5 ! (D) immediately.

This gives White no time to arrange the de­fence of the c-pawn with .id3 and �e2 and in­vites the complications that follow the advance 9 e5 !?, giving rise to a forced variation which has stood the test of time: 9 . . . h6 ! I 0 .ih4 g5 I I .if2 lbh5 ! 1 2 h4 ( 1 2 c5 .ixfl 1 3 �xfl f5 ! ? does not give White anything, Volkov-Brodsky, Dubai 2003) 1 2 . . .f5 ! . This move of Petrosian's, with the objective of fighting for space on the kingside, allows Black to achieve reasonable

w

counterplay. White has tried various options without gaining any advantage:

a) 1 3 exf6? 'iVxf6 14 c5 .ixfl 1 5 �xfl g4! 16 �d3 0-0 and suddenly Black goes over to the attack, Simagin-Petrosian, Moscow 1950.

b) 1 3 c5 .ixfl 14 �xfl 'iVc8! followed by . . . �a6+ with excellent play for Black, Volkov­Lysy, Tomsk 2006.

c) 1 3 hxg5 'iYxg5 14 lbh3 �g6 15 g4 fxg4! 16 .id3 �f7 1 7 fxg4 lbf4 and Black is fine, Agrest-A.Sokolov, St Petersburg 1993.

d) After 1 3 g3 f4 ! the complications are not unfavourable to Black; for example, 14 hxg5 �xg5 15 lbh3 �g8 16 gxf4 and in addition to 16 .. . 0-0-0, Berkes-Almasi, Kazincbarcika 2005, Black can capture with 1 6 . . . .ixc4 or prevent the f5 advance with l 6 . . . 'iVf7.

All these variations show that Black can ven­ture upon the combative 8 . . . lba5 !, allowing 9 e5. And if White defends c4 with 9 �a4, then the manoeuvre . . . �c8-b7-c6 is fully justified.

9 .id3! lba5 10 �e2 d6 The plan of I O . . . �b7?! (intending . . . �c6)

now runs into I I e5 lbg8 12 .ie4 ! . ll lbh3 lbd7 12 f4 f6 13 .ih4 �b7 14 0-0

0-0-0 15 llabl (D)

The Azeri GM brings the two pieces that he has on the kingside towards the centre before starting the offensive. The main problem for Black is that he is bogged down on the queen­side and is unable to increase the pressure on the c-pawn.

15 •.• �de8 1 5 . . . �c6 is answered with 16 .l::i.b4 !, and

Black remains immobilized.

SAMISCH VARIATION: 4 a3 OR 4 f3 11

16 ..tf2 h6 17 ..te3 .l:te7 1S lL'lf2 �bS 19 h3 Prophylaxis. As White is in no hurry, after

centralizing his pieces he gives his king an es­cape-square.

19 ... .l:!.c8 If 1 9 . . . g5, then 20 fxg5 fxg5 2 1 ltJg4. 20 lifd1 e5 21 fxe5 fxe5 22 d5 (D)

Keeping the black pieces imprisoned on the queenside.

22 ••• c6?! It is natural for Black to try to open a line on

the queenside in order to bring his dormant pieces back into play, but this move weakens his pawn-chain, especially b6. Here 22 . . . c5!? was best, permanently preventing the threat of White playing c5, but White would retain his space advantage. A good plan would be 23 ttlh I ! ? intending lL'lg3-f5, attacking d6, followed by doubling rooks on the f-file.

23 c5!

Liquidating the doubled pawns and irreme­diably weakening the hitherto secure black for­tress.

23 ••• .ixd3 24 ltJxd3 cxd5 There are not many options. If 24 . . . dxc5,

then 25 lL'lxc5 lL'lxc5 26 .ixc5 .l:td7 27 .ib4 and the knight on aS is lost.

25 cxb6 'l'i'a6 Recapturing on b6 would also lose to the ma­

noeuvre lL'lb4-d5, attacking the rook and the b6-pawn, but now the pawn reaches a7, with check and also protected.

26 bxa7++ �aS 27 '!'i'g4! ltJc4 Little is changed by 27 . . . lL'lf6 28 'i¥'f5 dxe4

(28 . . . ltJxe4 29 .l:tb8+ !) 29 lL'lb4 'l'i'c4 30 �xd6 'l'i'xc3 3 1 �h2 ! , with a decisive advantage.

2S .if2 lL'lf6 29 'iVf5 .l:Iec7 There is no time for 29 . . . dxe4 in view of 30

lL'lb4 'ti'b7 3 1 lL'ld5 and the material losses are substantial. The best try was 29 . . . llf8 ! , but then White still wins by 30 lL'lb4 'i¥'c8 3 1 'i¥'xc8+ .U.xc8 32 lL'lc6 !.

30 ltJb4 (D)

Now the d5-pawn falls and the one on a7 de­cides the game.

30 .• .'i�'a4 31lL'lxd5 ltJxd5 32 .U.xd5 ltJxa3 If 32 . . . 'iixa3, then 33 .l:Ib8+ .l:txb8 34 axb8'1'i'+

�xb8 35 .l.:tb5+ and mate in a few moves. 33 llbS+ The shortest way. White sacrifices his a7-

pawn to leave the black king without any pro­tection.

33 ••• .l:IxbS 34 axbS�+ �xb8 35 11xd6 ltJc4 36 lidS+ �b7 37 'iVfS

12 CHESS EXPLAINED: THE NIMZO-INDIAN

The queen joins in to leave Black's king de- 37 .•• i!Va6 38 .l::tb8+ Wc6 39 i!Ve8+ l:td7 40 fenceless against the combined attack of the .l:.c8+ 1-0 three white pieces. Excellent play by Radjabov.

Game 2 [E20]

Tiger Hillarp Persson - Peter Heine Nielsen Danish League 2001/2

1 d4 l2Jf6 2 c4 e6 3 l2Jc3 .tb4 4 f3 This move-order can also be regarded as part

of the Samisch Variation, or Neo-Samisch, since generally, at some point White plays a3, trans­posing to the positions seen in the previous game. At the start of the 1 990s, this line be­came very popular owing to its adoption by the super-GM Alexei Shirov and now it is very popular once more thanks to the contributions of I vanchuk, Milov and Volkov, to mention just a few illustrious names. White's intention is clear: to take over the centre completely.

4 ... l2Jc6 (D)

With this move, Black allows White tempo­rary domination of the centre, in order to coun­ter-attack there later. This is a way of playing that has given Black excellent results. He gives priority to development and subsequently strikes at the white centre, forcing it to advance so as then to occupy key posts with his minor pieces. Note how many pawn moves White is making, causing a slight lag in his development which Black will exploit with his next few moves.

5 e4

Consistent with his previous move. If he plays 5 a3 instead, then after 5 . . . .txc3+ 6 bxc3 b6 there is a transposition to Game 1 .

s ... ds! It is essential to strike in the centre. 6 cxd5 It seems more logical to advance without

capturing on d5, so that the c8-bishop will re­main shut in, but this has the drawback that Black has available the subsequent . . . dxc4 (at­tacking d4) and after .txc4? there is the re­source . . . lbxe5 ! and . . . 'iVh4+ !, winning a pawn.

For example: 6 e5 l2Jg8 ! (the knight always . comes to this square, in order to initiate the blockade on the light squares with .. . l2Je7-f5), and now:

a) If White plays 7 f4? then after 7 . . . dxc4 ! he cannot recapture on c4 since the d4-pawn drops. Later Black can consolidate the pawn with . . . i!Vd5 or increase his pressure on d4 by .. . l2Jge7 and . . . l2Jf5, with excellent play in both cases.

b) Another inaccurate move is 7 .te3?! in view of 7 . . . dxc4! 8 a3 (if 8 .txc4? then Black replies 8 . . . l2Jxe5 !) 8 . . . �xc3+ 9 bxc3 �d7 ! 10 f4 lDa5 and Black defends the extra pawn.

c) The critical continuation is 7 a3 ! �xc3+ 8 bxc3 dxc4 9 i!Va4! (9 �xc4? l2Jxe5 !) 9 ... f6! ? (9 . . . �d7 !? 10 �xc4 lba5 ! followed by .. . c5 ! deserves consideration) 1 0 exf6 lDxf6 1 1 �xc4 0-0 1 2 lbe2 Wh8 1 3 0-0 e5 14 �d 1 e4! and Black's activity on the kingside compensates for White's bishop-pair, Moskalenko-Barsov, Sitges 2000.

6 ... exd5 7 eS l2Jg8 (D) 8 f4 White finds himself forced to carry on mak­

ing pawn moves in order to complete his devel­opment. This is one of the problems of the

SAMISCH VARIATION: 4 a3 OR 4 f3 13

move 4 f3: it occupies the knight's natural de­velopment square.

8 �b5 has the drawback that after 8 . . . �d7 ! there is a threat to win a pawn with both .. :iVh4+ and .. .lbxe5, so that White is forced to capture on c6, losing control of the light squares.

The most sensible course is to defend against the threat of .. :iVh4+ and . . . 't�Vxd4 with 8 �e3 !? and now I like 8 . . . tbh6! (preventing 9 f4 in view of 9 ... tbg4!) 9 �d2 (9 �xh6?! 'iVh4+) 9 . ..Cbf5 1 0 �g5 f6 ! and Black seizes the initia­tive, Linares-F.Amold, Villa Real 200 I , but also satisfactory is 8 . . . tbge7 9 f4 0-0 I 0 �e2 �f5 I I ctJf3 �e4 ! and Black manages to activate this bishop, followed at an appropriate mo­ment by .. .lbf5 and the manoeuvre . . . �a5-b6 with counterplay against d4, Milov-Aimasi, Corsica (rapid) 2005 .

8 ••. tbh6! Not only preparing to blockade on f5 but

also ready to jump to g4 in reply to �e3. 9�e2 The purpose of this move is to prevent the

pin on the knight, but in view of what happened in the game it seems preferable to play 9 ctJf3 �g4, and now there are two ways:

a) 10 �e3 tbf5 1 1 �f2 �xf3 ! (destroying White's structure and securing f5 for the black pieces) 12 gxf3 0-0 and now, instead of 1 3 l:tgl ?! lt:Jcxd4!? 1 4 �xd4 'iVh4+ 1 5 �f2 'iVxf4, when Black had a strong attack in Shirov-Tivia­kov, Wijk aan Zee 200 I , Tiviakov recommends 13 'i¥d3 tbce7 14 �h3 g6 1 5 .l:!.g1 �h8 (D).

However, Black has little to fear, because the blockade on f5 is practically impossible to

break, and this keeps the f2-bishop completely paralysed and out of play.

b) It seems more logical to avoid the dou­bling of the pawns with I 0 �e2 tbf5 1 1 0-0 �a5 !? (a typical manoeuvre in this variation; the bishop goes to b6 to reinforce the pressure on d4, but the simple I l . . .'iVd7, followed by . . . 0-0-0, is also sufficient) 1 2 tbg5 �xe2 1 3 tbxe2 �b6 1 4 tbf3 h5 (D).

Black has achieved a perfect blockade on the light squares, although White's control of more space means that the chances are balanced. White's position suffers from a certain paraly­sis; practically his only plan is to play a3, with the idea of b4. In the two available practical ex­amples, the struggle was balanced after 1 5 'iVd3 't�Vd7 16 a3 a5 ! . Curiously, in one of these Black sent his king to the queenside ( . . . �d8!?-c8-b8), while in the other the king took refuge on the kingside with . . . g6 and . . . �f8-g7, maintaining

14 CHESS EXPLAINED: THE NIMZO-lNDIAN

the blockade and keeping the rooks on their original squares.

9 ••• lLlf5! Much better than Tiviakov's recommenda­

tion 9 . . . .i.g4. With the sequence in the game ( . . . lLlf5 ! and . . . .i.a5 !) Black makes use of his dynamic resources (better development and the initiative) to contest White's occupation of the centre.

10 lZ'lf3 .i.a5! (D)

w

A recurrent idea which is very effective. White must move the e2-bishop again, which demonstrates that his strategy has failed.

ll.i.b5 A pretty and simple variation arises in the

case of I I g4? lZ'lfxd4 ! 1 2 lZ'lxd4 �h4+ 1 3 �fl lZ'lxd4 and the white position falls apart, since i t i s not possible to capture on d4 because of . . . .i.b6.

1 1 ... 0-0! Black's lead in development is becoming

alarming. 12 a3?! Another wasted tempo. It was necessary to

accept a slightly inferior position with 1 2 .i.xc6 bxc6 1 3 �a4 .i.b6 14 �xc6 l:l:b8! 1 5 �xd5 lLlxd4 1 6 lLlxd4 .i.xd4 1 7 �xd8 l:l:xd8 1 8 lZ'le2 .i.c5 and the activity of all the black pieces, es­pecially his bishop-pair, clearly compensates for the missing pawn.

12 ••• .i.b6 13 .i.xc6 bxc6 14 �d3 In the hope of keeping things blocked up after

.i.e3, followed by queenside castling, but Black will not allow him the time for this. Instead, 1 4

0-0 was better, although after 1 4 . . . c5 ! Black breaks open the centre.

14 ... f6! (D)

I t is essential to contest White's central dom­ination. In addition, the f-file is opened so that the rook protects the knight, freeing the c8-bishop from that task.

15 .i.e3 fxe5 16lZ'lxe5 After 1 6 fxe5 .i.a6 ! 1 7 �d2 lLlxe3 1 8 �xe3

the exchange sacrifice 1 8 . . . l:!.xf3! is decisive, since the pawns at d4 and e5 drop.

16 ••• .i.a6! Black continues energetically. 17 �d2lZ'lxe3 18 �xe3 c5! (D)

After this break the position completely opens up in Black's favour.

19 dxc5 d4! 20 �e4 dxc3 Now the white king is condemned to remain

in the centre.

SAMISCH VARIATION: 4 a3 OR 4 f3 15

21!.1d1 'iVf6! Attacking the last soldier of the once-proud

white centre! W 22 g3 After 22 cxb6 'iVxf4 23 'iVxf4 .l:!.xf4 the e5-

knight is left without support. 22 •• J:tae8 Here 22 . . . lhd8 ! would have been immedi­

ately decisive, because not only would the d5-square be protected but also the black queen would penetrate along the d-file: 23 cxb6 (23 .l:!.xd8 'iVxd8 ! ) 23 ... .U.xdl + 24 �xd l 'iYd6+ and the queen enters at d2.

23 cxb6 ..ib7! 24 'iVc4+ Here 24 'iVxb7 �xe5+ ! leads to an elegant

mate. 24 .. .<.t>h8 25 0-0 �xb6+ 26 't!Vd4 If 26 .l::tf2, then 26 . . . .l:txe5 ! . 26 ... c2 27 't!Vxb6 cxd1 't!V 28 'iVxb7 �d4+ (D) Black is the exchange up and has the safer

king, so the struggle is decided. The rest is quite clear although apparently both players were in time-pressure.

29 '.t>g2 .l::tb8 30 'J/IIxc7 �xb2+ Better was 30 . . . Ii.xb2+! and . . . 'iVd2.

31 �h3 'i¥xa3 32 i:Id1 �a6 33 l:td7 't\Ye6+ 34 �h4 't\Yh6+

34 . . . �f6+! . 35 �g4 �xh2 36 lt:Jf7+ After 36 .l::!.xg7 h5+! 37 �g5 'iYxg3+ the

white king will not survive. 36 ... �g8 37 lt:Jg5 h5+ 38 '.t>f3 l:tb3+ 39 �e4

.l::!.e8+ 40 �f5 0-1 White lost on time before being mated. An

entertaining game, illustrating the problems that can be caused by falling behind in development.

Game 3 [E27]

Miso Cebalo - Dimitri Komarov Bratto open 2006

1 d4 lt:Jf6 2 c4 e6 3 lt:Jc3 ..ib4 4 f3 0-0!? (D) This move has, in principle, the same idea as

4 .. . lt:Jc6. Black makes a developing move and waits for White to play 5 e4. Then Black strikes in the centre immediately with . . . d5, so that he can exploit his development advantage.

5 a3 The central strike 5 e4 d5 ! is even better now

than when the black knight is on c6 (Game 2), because Black has the immediate central coun­ter-attack . . . c5 ! available. White must capture with cxd5, either immediately or after 6 e5, since otherwise Black will take on c4 and de­fend the pawn with .. . b5 or . . . lt:Jb6. For exam­ple, 6 e5 lt:Jfd7 7 a3?! ..ixc3+ 8 bxc3 dxc4! and now 9 ..ltxc4? is not possible in view of the fa­miliar idea 9 . . . lt:Jxe5 ! .

w

After 6 cxd5 exd5 7 e5 lDfd7 ! White has problems in supporting his advanced pawn­centre. If he tries to bolster d4 with 8 a3 ..ixc3+

16 CHESS EXPlAINED: THE NIMZO-IND/AN

9 bxc3 Black breaks with 9 . . .f6 ! , and the white centre vanishes, because I 0 f4? is not possible in view of 1 0 . . . fxe5 I I dxe5 lt:Jxe5 ! and the de­velopment advantage makes itself felt. White must play more modestly with I 0 exf6 �xf6 I I �d3 lt:Jb6 1 2 lt:Je2 �f5, but after the exchange of the light-squared bishops the outpost on c4 guarantees equality, Moskalenko-J.Gonzalez, Mataro 2004. If White plays 8 f4 then after 8 . . . c5 ! the centre explodes and Black can play sacrificially to smash open lines towards the white king in the centre. For example:

a) 9 a3 �a5 ! 1 0 lt:Jf3 cxd4 II lt:Jxd4 lt:Jxe5 ! (D).

w

A typical sacrifice in this variation. Black · exploits his advantage in development and the

exposed position of the white king. 1 2 fxe5 �h4+ 1 3 '>t>e2 �b6! (the key move: the bishop comes to the a7-gl diagonal with devastating effect) 1 4 �e3 �g4+ 1 5 lt:Jf3 d4 ! 1 6 �xd4 �xf3+ 1 7 '>t>xf3 i.xd4 and White is destroyed, Volkov-Vekshenkov, Togliatti 2003.

b) 9 lt:Jf3 cxd4 !? 1 0 �xd4 (after 10 lt:Jxd4 once again the sacrifice works: 1 0 . . . lt:Jxe5 ! I I fxe5 �h4+ 1 2 '>t>e2 �c5 !) 1 0 . . . lt:Jc6 I I �xd5 lt:Jdxe5 ! 12 �xd8 lt:Jxf3+ 1 3 gxf3 .l:!.xd8 14 �e3 �f5 and White stands badly, because his king remains in the centre and his pawns are weak, Kadziolka-Alexandrov, Warsaw 2005 .

5 ... �xc3+ 6 bxc3 (D) 6 ••. tt:Jh5!? Capablanca's manoeuvre 6 .. . lt:Je8 !? is also

well-known and popular here. Black avoids the pin and plays . . . b6, . . . �a6 and . . . lt:Jc6-a5 . Also

the e8-knight can go to d6, reinforcing the at­tack on c4, while the f-pawn serves to restrict the white forces.

TLe move 6 . . . lt:Jh5 !? gains time for moving the pawn to f5 and subsequently the queen to h4 (important!), thus hindering White's advance in the centre. Black's general idea consists of con­testing control of the centre with the above­mentioned advance . . . f5 and later . . . d6 and .. . e5. In this way Black restricts White's aggressive intentions on the kingside. Development is com­pleted with . . . lt:Jc6 and . . . i.d7. If White can be prevented from advancing in the centre, Black will later tum his attention to White's weak­nesses on the queenside ( c3 and c4) or else gain space with . . . e4 or . . . f4, with possibilities of at­tacking on the kingside.

7 lt:Jh3 The most common move, which is designed

to protect against the queen check on h4. Of course, 7 e4? is bad in view of 7 . . . �h4+ 8 g3 lt:Jxg3 9 hxg3 �xh I with a decisive advantage. Grandmaster Sergei Volkov experimented suc­cessfully with 7 g3 !? f5 8 e4 but, instead of 8 . . . b6, Black should remain consistent with the general idea of the variation by playing 8 . . . d6!?, followed by . . . e5, since it is better to develop the bishop along the c8-h3 diagonal, to make it difficult for White to develop his knight.

7 ... f5 The point of . . . lt:Jh5. 8 e3 The simplest way of bringing the king's

bishop into play and castling quickly.

SAMISCH VARIATION: 4 a3 OR 4 f3 17

The brilliant Bronstein employed the ener­getic idea 8 e4 !?, which gave him an excellent position in Bronstein-Szabo, Candidates tour- W nament, Budapest 1950 after 8 . . . c5?! 9 e5 ! l2Jc6 10 f4 g6 II i.e2 b6 1 2 0-0 l2Jg7 1 3 i.e3 cxd4 14 cxd4 i.a6 15 'ifa4, and White had the bishop-pair and a strong centre, while Black's g7-knight is out of play. Knaak's recommendation 8 .. .fxe4?! 9 i.g5 l2Jf6 1 0 fxe4 h6 II i.h4 g5? does not seem beneficial, since the insecurity of the black king becomes clear after 1 2 i.g3 l2Jxe4 1 3 'ifc2 ! d5 I4l2Jf2 ! . It is best for Black not to deviate from the general plan and so to continue with 8 . . . d6 ! followed by . . . e5 and in some cases .. .f4.

A further option for White is to develop the queen's bishop before playing e3, but this does not alter the character of the position much: 8 i.g5 !? 'ife8 9 e3 and Black should continue with the standard plan: . . . d6, . . . l2Jc6 and . . . e5.

8 ... d6 After hindering White's advance in the cen­

tre with .. .f5, Black prepares to complete the mobilization of his queenside with . . . e5 and .. . l2Jc6, followed by developing the bishop on d7.

9l2Jf2 After 9 i.d3 the best move is 9 . . . 'ifh4+ ! ,

which forces the white knight to a more passive position on f2. Furthermore, when White cas­tles, it will be difficult to dislodge the queen from this square, since the move g3 can be an­swered with the sacrifice . . . l2Jxg3 !, dismantling the white fortress.

9 .•. l2Jc6 It is also possible to play 9 . . . e5 !?, in accor­

dance with the general plan. After 1 0 f4l2Jf6 I I fxe5 dxe5 1 2 c5 l2Jc6 1 3 �c4+ �h8 14 0-0 b6 !?, White has difficulty in bringing his queen's bishop into play, Volkov-Izoria, European Clubs Cup, Saint Vincent 2005.

10 �d3 e5 11 0-0 If I I 'ifc2 then the best move is once again

l l . . .'ifh4!?, posting the black queen on its ideal square .

11 ... 'ifh4! (D) From this menacing square, the queen exerts

pressure on d4, and at the same time restricts White's actions on the kingside. Black's plan is

quite simple: develop with . . . �d7 and . . . .l:i.ae8 and then advance in the centre or attack on the kingside.

White has little activity and his best plans are to exert pressure on the b-file, and try to ad­vance with c5, given that it is not sensible to play e4 because Black can respond with . . .f4 and then transfer a rook to h6 or g6.

12 'tiVe1 With the objective of dislodging the black

queen from h4. It seems more logical to play ei­ther 1 2 'ifc2, exerting pressure on f5, or l 2l!b I, making it difficult for the black bishop to be de­veloped.

12 ••• i.d7 13 g3!? 'ife7 It is not possible to play l 3 . . . l2Jxg3? be­

cause of 1 4 lLlh l ! winning a piece, but it was worth considering keeping the queen on the kingside with l 3 .. .'�!Vg5 ! ?, renewing the threat of . . . l2Jxg3.

14 'ifd1.l::i.ae8 15 l:!.a2l2Jf6 (D)

w

18 CHESS EXPlAINED: THE NIMZO-INDIAN

The knight returns to f6 to support the ad­vance . . . e4, which will leave White with weak­nesses on the kingside and a wretched bishop on c l . This is one of Black's basic ideas in this variation: to maintain the blockade and at some point to advance the e- or f-pawn, gaining space.

16 f4 Cebalo prefers Black's . . . e4 advance to take

place without i t attacking the f3-pawn, so that he will not be forced to capture on e4, which would allow the d7 -bishop into play.

16 ... e4 17 .i.e2 �f7! Now Black puts pressure on c4 in order to

force the advance of the white pawns, which would result in more weaknesses in the white camp. It is worth noting this formation for at­tacking c4: . . . Wiif7, . . . lt::!a5 and . . . .i.e6 if neces­sary.

18 a4 lt::!a5 19 .l:ta1 !? It would be much worse to defend the pawn

by playing 19 d5, leaving himself paralysed and handing over the c5-square to Black, who would respond with the manoeuvre ... b6, ... lt::!b7, . . . aS and ... lt::!c5, with a clear positional advan­tage.

19 ••. ltJxc4 (D)

w

It would seem that Black has his opponent completely under his thumb, since as well as having an extra pawn, he exerts a degree of con­trol over the light squares and the opposing pieces lack activity. However, the experienced Croatian GM finds a way to complicate his op­ponent's task.

20 d5! lt::!a5?!

This allows the c 1-bishop to come back to life. It was better to maintain the blockade on the light squares by 20 . . . b5 ! 21 axb5 .ixb5 22 .l:txa7 �xd5! 23 l:!.xc7 l:!.c8, with a clear posi­tional advantage due to the passivity of the white pieces and Black's pressure on c3 and e3.

21 c4! (D)

B

Now White's bishop-pair and command of the a l -h8 diagonal provide some compensation for the pawn. In this defence it is essential to take particular care not to allow the bishop-pair any activity, especially the dark-squared bishop when it has no opposite number.

21 ... b6 22 .ib2 lt::!b7 23 'it>h1! (D)

B

White prepares the opening of the g-file, to exert pressure on g7 in coordination with the bishop on b2.

23 .•. ltJc5

SAMISCH VARIATION: 4 a3 OR 4 f3 19

Now White exchanges his weak a-pawn and activates his queen' s rook, but if Black prevents the advance with 23 . . . a5, White would break immediately with 24 g4 !, obtaining sufficient counterplay.

24 aS! .l:!.a8 25 .l:!.gl Patiently preparing the break, since his op­

ponent lacks active play. 25 •.. c6 26 g4! Better than regaining the pawn with 26 dxc6

.ixc6 27 'iVxd6 'iVe8 !, since the threat of .. .'�Jd3 gives Black the initiative.

26 .•. t2Jd3!? (D)

Creating more complications, to distract White from his positional attack.

27 t2Jxd3 exd3 28 'iVxd3 fxg4 Now White will simplify in order to regain

the pawn. It was worth considering 28 .. . t2Jxg4!? but after 29 .l:!.g2 .l:!.ae8 30 axb6 axb6 31 e4! White has sufficient counterplay because the position of the knight on g4 is unstable.

29 axb6 axb6 30 .l:!.xa8 .l:!.xa8 31 .i.xf6 'iVxf6 1/z-lfz

The position is fairly equal and since both kings are exposed, the chances of perpetual check are high.

Conclusions

When White gains the bishop-pair by playing 4 a3 right away, the plan of an immediate siege of the c4-pawn with . . . b6, . . . .i.a6 and . . . t2Jc6-a5 can be strongly recommended, as Game 1

demonstrates. Black must do this before cas­tling, so that against the .i.g5 pin he has the re­sponse . . . h6 and . . . g5 available, if required.

It is essential to remember important queen manoeuvres such as . . . 'iVe8 (especially when the white queen is on a4) in order, after the ad­vance of his d-pawn, to offer the exchange of queens, which would leave the c4-pawn with one less defender. Also, . . . 'iVd7 and . . . 'iVa4 or . . . 'iVc8-b7-c6 share the same objective of at­tacking the c4-pawn and activating the black queen. Dizdarevic 's error in Game I was to em­ploy the last of these queen manoeuvres, di­rected against the 8 .i.g5 pin, when he should have played 8 . . . t2Ja5 !, entering the complica­tions that arise after 9 e5 h6 ! 1 0 .i.h4 g5, or else forcing the white queen to defend the c4-pawn from a4.

Against the immediate 4 f3 Black has a rea­sonable option in 4 . . . t2Jc6! ?. The positions that arise after 5 e4 d5 6 cxd5 exd5 7 e5 t2Jg8 are perfectly playable for Black, who achieves a good blockade on the light squares and has ac­tive play with . . . f6 and the manoeuvre . . . .i.a5-b6. Game 2 is an example of how to punish in­accurate play and a delay in castling by White.

I think that White should handle this position by playing the immediate 6 e5 followed by 7 a3 ! , so that the c8-bishop remains locked be­hind its own pawns, and although so far Black has achieved sufficient counterplay, I expect that this line will be the most problematic in the future.

The encounter Cebalo-Komarov (Game 3) featured 4 f3 0-0! ? and, as the analysis demon­strates, Black has nothing to fear in the event of 5 e4 d5 ! because after the advance 6 e5 the breaks . . . c5 ! and/or . . .f6 ! , as appropriate, make it very difficult for White to maintain his centre. The critical line begins with 5 a3 but by employing the formation seen in the game (6 .. . t2Jh5 fol­lowed by ... f5 and the central advance . . . d6 and . . . e5), Black has no serious problems neutraliz­ing White's initiative in the centre and on the kingside. It should be emphasized that in this variation the black queen's best square is h4. It is to be expected that this line will gain popularity in the future owing to the simplicity of its ideas.

2 Capablanca Variation: 4 �c2 d5!?

Introduction

The classical 4 �c2 is one of the most solid and reputable responses to the Nimzo-Indian, based above all on its logic: White wants to obtain the bishop-pair without spoiling his pawn-structure.

However, its fundamental drawback is that it uses several tempi to this end, which means that White lags behind in development and the queen remains rather exposed to attack by the black pieces.

The move 4 �c2 was very popular from the 1 920s to the 1950s owing to its inclusion in the rep­ertoires of players such as Capablanca, Alekhine and Euwe. Later it fell into disuse until the 1 980s, when Seirawan obtained excellent results with it. Later he was joined by Kasparov, I.Sokolov, Bareev and M.Gurevich, who contributed to its success with new ideas and splendid results. Today it is one of the most common and deeply-investigated variations.

The variation that we shall study in this chapter, 4 1i'c2 d5 !?, aims to take advantage of White's lack of development by immediately opening the position with central breaks such as a well-timed . . . c5 or even . . . e5.

Black's philosophy is very clear: to give priority to active piece-play above any other positional or material considerations, so that White does not have time to use his bishop-pair. That is why in this variation Black even delays his own castling, allows the destruction of his pawn-structure or sacrifices a pawn, all with the objective of mobilizing his queenside and creating threats against the insecure queen and the enemy king, which needs several tempi to castle on the kingside and which would be too exposed on the queenside.

The Games

In Game 4 (Bareev-Ivanchuk), we shall see the line 5 a3 ! ?, in which White persists in his idea of gaining the bishop-pair without attaching too much importance to the tempi consumed. Black is forced to give up his bishop, but with the continuation ... lt:Je4 and . . . c5 ! he obtains very good counterplay along the a5-e I diagonal, or full compensation for the pawn if White hangs on to the material, as in the game. We shall also see that the variation with . . . lt:Je4 followed by . . . e5 is some­what better for White, who manages to complete his development and maintain the bishop-pair plus a stable centre. In the notes to this game we shall also see a less risky alternative (6 .. . dxc4 and . . . b6 !?) with the aim of exchanging one of White's bishops and trying to equalize quietly. In the game, Ivanchuk achieved excellent compensation for the sacrificed pawn in the shape of a lasting initiative, which finally grew into a strong attack owing to some slight inaccuracies from Bareev.

In Ibrahimov-Mamedyarov (Game 5), we shall analyse the variation where White takes on d5 and then plays 6 .ig5 and we shall see how the response 6 . . . c5 ! produces an extremely complicated position, where both players are walking a tightrope. White chose the main move 7 dxc5 ! ? and Black put immediate pressure on the c3-knight with . . . h6, . . . g5, followed by . . . lt:Je4 and .. .'iVa5, which is the most popular manoeuvre these days and which reduces White's options. As the reader will see, the struggle was very complicated, with both kings exposed, but it took only one mistake

CAPABLANCA VARIATION: 4 Wfc2 d5!? 21

by White on move 1 8 for Black to launch a very strong offensive, which ended in a mating attack on move 3 1 .

Game 6 (Van Wely-Topalov) shows us all the alternatives to the move 7 dxc5 of the previous game and in particular 7 a3, which was the move used by Van Wely and which is quite popular at the time of writing. With this move White obtains the bishop-pair and the better pawn-structure, but Topalov demonstrated how Black's development compensates for these factors, even in the end­game.

Game 4 [E37]

Evgeny Bareev - Vasily lvanchuk Havana 2006

1 d4 tbf6 2 c4 e6 3 tbc3 i..b4 41Vc2 (D)

The basic idea of this variation is to obtain the bishop-pair without any damage to the pawn­structure.

4 ... d5 Black immediately strikes in the centre, con­

trolling e4. 4 ... 0-0 is also possible, as we shall see in the next chapter, as is the immediate at­tack on the centre with 4 . . . c5.

5 a3 White carries on with his initial plan. We

shall see 5 cxd5 in the next game. If 5 i..g5? ! Black captures by 5 . . . dxc4! attacking d4 and af­ter 6 e3 or 6 tbf3 he defends the pawn with 6 . . . b5, and . . . c6 if necessary.

s ... i..xc3+ The retreat 5 . . . i..e7?! makes no sense here,

since White continues with the natural develop­ing moves tbf3, i..f4 or i..g5, followed by e3,

and the move a3 is useful for responding to the central attack . . . c5 with dxc5 followed by b4. It is also possible to answer with 6 cxd5 followed by i..g5, reaching the structure of the Exchange Variation of the Queen' s Gambit, with the fa­vourable difference that the move a3 supports the advance b4, which is an essential part of the typical minority attack.

61Vxc3 (D)

White has invested his last two moves in the acquisition of a stable and lasting advantage -the bishop-pair - but he has fallen behind in de­velopment and his queen is rather exposed. It is important to understand that if Black quietly develops his pieces without disturbing White, allowing him to complete his development, he will be condemned to a passive and unpleasant defence in the middlegame and even in the end­game. What can Black do? One plan consists of

22 CHESS EXPLAINED: THE NIMZO-INDIAN

opening the centre as quickly as possible (even if it means sacrificing one or more pawns) to try to exploit his advantage in development. An­other plan is to try to exchange one of White 's bishops. In the game the first method is used, while the attempt to exchange a bishop is ex­plained in the next note.

6 .. .lbe4 The knight occupies an important post in the

centre, and at the same time attacks the queen and prepares a central break. The alternative is 6 . . . dxc4 7 �xc4 b6 ! ? to play . . . .i.a6, attacking the queen, and to exchange the light-squared bishops as soon as White moves his e-pawn. However, I have the impression that after seiz­ing the centre and castling on the queenside the white position is slightly better. In contrast, if White tries to retain his bishop-pair by develop­ing the fl -bishop to g2 then Black first harasses the queen with . . . .i.b7-d5 or . . . .i.a6, and then follows up by . . . lbbd7, . . . .l:tc8 and . . . c5, with very good piece coordination. Let us see an ex­ample: 8 CtJf3 0-0 9 .i.g5 (this pin is logical, given that Black has no dark-squared bishop; if 9 g3 .i.b7 10 .i.g2, then I O . . . .i.d5 followed by . . . lbbd7, . . . c5 and . . . .l:i.c8, with excellent play) 9 . . . .i.a6 10 �c2 (inadvisable is 1 0 .i.xf6?! 'iixf6 1 1 'iixc7 .l:tc8 and the black pieces are very ac­tive) 1 0 . . . lbbd7 1 1 e4 .i.xfl 1 2 llxfl 'iic8 !? 1 3 0-0-0 �b7 1 4 �fel �ac8 1 5 Wbl l:Ife8 fol­lowed by ... c5 with equality, Golod-S.Feller, Cappelle Ia Grande 2007.

7 Wic2 c5! At the time of writing, this break has the best

reputation. The two alternative lines for break­ing in the centre are connected with the advance . . . e5:

a) 7 . . . lbc6 8 e3 e5 !? is one way, but after 9 f3 ! lbf6 1 0 dxe5 lbxe5 1 1 cxd5 'ifxd5 1 2 e4 we reach the type of position that Black must avoid, because his knights lack outposts in the centre and he has no active breaks, while the bishop­pair and central superiority give White the ad­vantage.

b) With 7 . . . e5 !? Black intends to answer 8 cxd5 with the active 8 . . . .i.f5 ! ? (D) (instead of 8 . . . Wixd5? ! 9 f3 followed by 1 0 dxe5 and 1 1 e4 and Black is condemned to passive defence, Kasparov-Adams, Linares 2005).

White now has various ways to win material but Black has obtained sufficient compensation in the majority of cases. For example, 9 g4 (9 dxe5 c6! ?) 9 . . . .i.g6 1 0 'iib3 exd4 1 1 lbf3 0-0 12 .i.g2, Nechepurenko-Khairullin, Russian Ch, Krasnoiarsk 2007, and now 1 2 ... d3! gives Black the initiative.

Regardless of their soundness, the moves 7 . . . e5 !? and 8 . . . .i.f5 ! ? are in the spirit of how Black should operate to counteract White's bishop-pair .

8 dxc5 White must not accept a position where he

has an isolated pawn without any attacking pos­sibilities, i.e. 8 e3? cxd4 9 exd4 lbc6 10 lbf3 'i!Va5+.

8 ... lbc6 9 cxd5 Protecting the c5-pawn and giving his oppo­

nent an isolated pawn, but the queen's bishop gains the freedom to reach f5 (or g4), creating various tactical possibilities at the white queen's expense.

Attempts have been made to gain an advan­tage without exchanging on d5, so as to keep the black bishop shut in at c8, but then the check on a5, followed by the elimination of the dark-squared bishop, equalizes:

a) 9CtJf3 �a5+! 1 0 .i.d2 ( 10CtJd2!? is trick­ier, but the complications after 10 .. . lbd4! 1 1 �dl !? {or 1 1 �d3 e5 } l l . . . .i.d7 are no worse for Black) 10 .. .'�xc5 I I e3 lbxd2 1 2 'i!Vxd2 dxc4 with equality.

b) 9 e3 �a5+! 1 0 .i.d2 lbxd2 ! 1 1 'i!Vxd2 dxc4! 1 2 'ifxa5 lbxa5 1 3 lie I b5 ! (giving up a pawn, but once the files are open, the queenside

CAPABLANCA VARIATION: 4 'iVc2 d5!? 23

pawns will be subject to attack) 1 4 cxb6 .i.b7 15 ctJf3 �e7 16 bxa7 and now the most appro-priate course is to eliminate the f3-knight, and W exert immediate pressure on the white pawns along the b-file.

9 ••. exd5 10 ctJf3 (D)

B

10 .• Ji'f6!? Hindering the b4 advance and its further

consequences of White developing his dark­squared bishop actively.

The last word has not yet been said regarding 10 .. . .i.f5 !?, but the latest games favour White: 1 1 b4! d4!? ( l l . . .ctJg3?! 12 �b2 ctJxh l 1 3 �xg7 .l:!.f8 14 �h6 is no use; nor is 1 1 . . .0-0 1 2 �b2 ctJg3?! 1 3 �c3 d4 14 ctJxd4 ctJxd4 15 fxg3 ! with a decisive advantage) 1 2 g4 �g6 1 3 'iVb2 0-0 14 .i.g2 Ii.e8 15 0-0 ltJc3 1 6 .U.el h5 1 7 g5 i.e4 1 8 �d2 and White completes his develop­ment while retaining the extra pawn and keep­ing pressure on d4, Ivanchuk-Anand, Monte Carlo (blindfold) 2007.

11 e3 .i.g4 12 .i.e2 0-0 (D) 13 0-0 Natural, but inaccurate. Now White will find

it difficult to bring his queen's bishop into play. It is better to play 1 3 .i.d2! and now Black should exert pressure on the c-file by 1 3 . . . .l::tac8 ! 14 0-0 b6! with counterplay, Hillarp Persson­J.Hammer, Copenhagen 2007.

13 ••• I:i.fe8! 14 .i.d2 d4! The opening of the centre enables the rooks'

pressure on the central files to make itself felt and highlights the passivity of the white pieces.

15 I!ad1 ctJxd2 16 �xd2

It is more prudent to return the pawn imme­diately with 16 .l:txd2 dxe3 1 7 fxe3 M.xe3 with approximate equality.

16 ••• .l::tad8 (D)

w

This position exemplifies the kind of com­pensation that the pawn sacrifice provides: pres­sure on the centre files and more active pieces.

17 'iVc1 1 7 exd4 is met by 1 7 . . . ctJxd4 ! , when 1 8

ctJxd4? .i.xe2 costs White the exchange. Now it seems that White has solved his problems, but lvanchuk finds an energetic continuation that retains the initiative.

17 ••. d3! A further temporary pawn sacrifice which

brings about a weakening of White's castled position.

18 I!xd3 .l:Ixd3 19 .i.xd3 .i.xf3 20 gxf3 ctJe5 21 .i.e4 ctJxf3+ 22 .i.xf3 'iVxf3 23 �d1 �f5 24 b4 h6!

24 CHESS EXPLAINED: THE N!MZO-INDIAN

B

Now the black rook is free to join the attack. 25 l:te1 aS!? To weaken b4. 26 e4! (D)

Bareev returns the pawn in order to simplify the position and prevent the combined attack of the black queen and rook on his weakened king.

26 • • .'i!Vg6+ 27 Wfl axb4 28 axb4 .l:lxe4 29 �b1?

It was better to play 29 l!xe4! �xe4 30 �d8+ �h7 3 l li'd7 ! f5 32 b5 f4 33 li'g4! with good drawing chances.

29 .• .'i!Va6+ 30 'lt>g2 After 30 b5 .l:Ixel + 3 1 �xel 'iVe6+ 32 �fl

�c4+ 33 �gl 'i¥xc5 the extra pawn seems suf­ficient for a win.

30 .•. .l:tg4+ 31 �h3 I!.g6 32 "Wif5? The decisive mistake. It was necessary to

rescue the king by means of the manoeuvre 32 .l:!.e3 ! and l:tg3.

32 ••. �c6 33 .l:!.e4 There is no salvation in 33 f3 .ilg5 34 �f4

�d7+ 35 �h4 .l:!.g2 with the threat of . . . g5+. 33 ••. .l::tg5 The mating-net tightens around the unfortu­

nate king. 34 li'f3 'i¥d7+ 35 Wh4 l:tg1 36 �e3 If 36 l:te5, then 36 . . . 'i¥d8+ 37 �h3 g5! and

Black wins. 36 • . • �f5 37 .l:!.e8+ Wh7 38 �e4 g5+ 0-1 Bareev resigned in view of the unusual mate

that occurs after 39 Wh5 'i¥g6+! 40 "Wixg6+ fxg6#.

Game 5 [E35]

Rasul lbrahimov - Shakhriyar Mamedyarov President's Cup, Baku 2006

1 d4 lt:Jf6 2 c4 e6 3 lt:Jc3 ii.b4 4 li'c2 d5 5 cxd5 exd5

5 . . . 't!Vxd5 is a major alternative, but we shall not be examining it in this book.

6 ii.g5 c5! (D) It is very logical to attack d4 and open the c­

file when the queen is on c2 and is no longer pro­tecting d4 or putting pressure on d5. Previously preference was given to 6 . . . h6, intending 7 ii.h4 c5! , but this move-order allows the variation 7 ii.xf6!? 't!Vxf6 8 a3 i.xc3+ 9 'i¥xc3, which leads to rather tedious positions where Black is in little danger but is just playing to neutralize White's minority attack on the queenside.

7 dxc5 The most demanding variation for Black. We

shall see the topical 7 a3 !? together with other alternatives in the next game.

7 ••• h6 8 ii.h4 Now 8 i.xf6 !? gains time for completing de­

velopment but gives up the bishop-pair with the centre open, so that Black does not have many

CAPABLANCA VARIATION: 4 Wfc2 d5!? 25

problems equalizing. Black will have compen­sation for the isolated pawn on d5 in the form of pressure on the c-file. In addition, Black can capture on c3 and leave his opponent with a backward pawn, or even retain the bishop-pair in some variations.

8 • . . g5! The f6-knight is needed to increase the pres­

sure on c3. This advance certainly weakens the kingside, but sooner or later it will be essential to break the pin. Note that here and on the fol­lowing moves the advance 8 . . . d4? does not work because with 9 0-0-0 ! White gets out of the pin and in turn pins the d-pawn.

The move-order 8 . . . tZ'lc6?! 9 e3 g5 1 0 ..ig3 CLJe4 gives White the possibility of putting the question to the bishop immediately with 1 1 a3 ! "i'a5 12 tZ'le2 ..if5 and then after 13 "i'c I ! the fol­lowing semi-forced variation arises: 1 3 . . . tZ'lxc5 ! 14 axb4 tZ'ld3+ 15 �d2 "i'xb4 ( 1 5 . . . 4Jxc l ? 16 l:txa5 !) 16 l:ta4 tZ'lxc l ( 1 6 . . . "i'b3? 17 �bl ! ) 1 7 l:txb4 tZ'lxe2 1 8 l:txb7 tZ'lxg3 19 hxg3 d4! 20 exd4 tZ'lxd4 2 1 tZ'ld5 ! (D).

Black is suffering in the endgame, as several master games have shown. As we shall reiterate in the notes to Black's 1Oth and l i th moves, whenever Black includes the move . . . tZ'lc6? ! White can exploit the tempo to reach this ad­vantageous variation.

9 ..ig3 tZ'le4 10 e3 It is essential to try to bring the kingside

pieces into play. It looks as if with 1 0 ..ixb8 White would force Black to simplify, because I O . . . l:txb8? I I �a4+ costs him a piece, but the

surprising 10 . . . �f6 ! works perfectly: 1 1 ..ig3?! (it i s better to solve the development problem with 1 1 tZ'lf3 ! ..ixc3+ 1 2 bxc3 l:txb8 1 3 e3 ! 0-0 14 ..id3 tZ'lxc5 1 5 0-0 ..id7 and both sides have weaknesses, H.Rau-Babula, Bundesliga 200617) l l . . . tZ'lxc3 1 2 a3 ..if5 ! 1 3 �d2 ..ia5 14 b4? tZ'le4 1 5 �c I l:tc8 ! ! , as in the famous game I.Soko­lov-Aronian, Turin Olympiad 2006, which ended in a quick victory for Black, because the undeveloped white forces could not cope with the coordinated attack of the black army.

10 ... �a5! (D) In this variation the most advisable policy is

to apply immediate and continuous pressure on the c3-knight. Remember that 1 O . . . tZ'lc6?! is an­swered with I I a3 ! transposing to the variation mentioned in the note to Black's eighth move.

11 tZ'le2 White does not have time to play 1 1 ..ie5 0-0

1 2 ..id3 in view of 1 2 . . . tZ'lc6 ! 1 3 ..ixe4 tZ'lxe5 14 ..ixd5 ..ig4! 1 5 tZ'lf3 ..ixf3 1 6 ..ixf3 tZ'lxf3+ 1 7 gxf3 l:tac8 and Black will soon regain the pawn with a good endgame. Nor can White expect a quiet life after 1 4 ..ih7+ �g7 1 5 ..id3 d4! 1 6 exd4 ..ixc3+ 1 7 bxc3 tZ'lxd3+ 1 8 �xd3 ..if5 ! 19 'ifc4 �fe8+ with an attack, I.Sokolov-Short, Wijk aan Zee 1 995.

ll ... ..if5! l l . . .tZ'lc6? ! prevents ..ie5, but once again 1 2

a3 ! gives White the advantage. 12 ..ie5 Dreev's idea 12 ..ixb8 is not as strong:

1 2 . . . �xb8 1 3 tZ'ld4 ..id7 ! 1 4 tZ'lb3 ..ixc3+ 1 5 bxc3 �xc3+ 1 6 'ifxc3 tZ'lxc3 1 7 f3 �e7 and

26 CHESS EXPlAINED: THE NIMZO-INDIAN

after . . . tZ:la4 and . . . .l::tc8 Black' s pressure on c3 compensates for his weaknesses on the king­side and the isolated queen's pawn.

12 ••• 0-0! 13 tZ:ld4 .l::te8! (D)

w

Short's idea, to involve this rook in the fight as soon as possible.

14 .i.xb8 There is no advantage for White in 14 l2lxf5

.U.xe5 1 5 tZ:lxh6+ �g7 16 tZ:lg4 and now instead of 16 . . J:te7 (Biriukov-Emelin, St Petersburg 2005), Black should play 1 6 . . . lle6! , which de­fends the king side and is ready for the attack if White castles on the queenside.

14 ••• tZ:lxc3! Better than 14 . . . .i.g6? ! 1 5 .i.d3! .l:i.axb8 16

0-0 and, now that he is castled, White goes over to the attack on Black's weaknesses in the cen­tre and on the kingside, Atalik-Short, Sarajevo 2004.

15 tZ:lxfS The knight is more effective than the queen

on f5 . 15 .•. tZ:le4+ 16 �e2 White wants to play f3 and shelter the king

on f2. An alternative worth consideration is 16 �d l ! ? , so as not to obstruct the bishop and in some cases bring the king to the queenside, but after 16 .. Jhxb8 1 7 f3 tZ:lf6! 1 8 l2lxh6+ �f8 ! the struggle remains wide open. White will try to penetrate on the king side, but the black forces occupy more active positions and after . . . l:tbc8 and . . . .U.xc5, the white king also finds itself in danger.

16 .. :�a6+!? 17 �d1 "iff6 18 f3?

Overlooking Black's strong reply. It was es­sential to maintain the knight on f5 and start the counter-attack with 1 8 g4! l:Iaxb8 1 9 h4.

18 •• .'*1Vxf5!? 19 fxe4 dxe4! (D)

The unprotected white king comes under fire from the opponent's major pieces.

20 .i.d6 l:Ied8! Threatening . . . .i.xc5. Black's other rook is

needed on the c-file. 21 �e2 An ugly but unavoidable move. The king

would not be safe on the other wing either: 2 1 �c I .l::tac8 22 i.c4 .i.xc5 23 i.xc5 WUxc5 24 .i.xf7+ �xf7 25 ltfl + �e6 26 WUxc5 :Ixc5+ 27 �bl l:td2 and the rooks invade the seventh rank.

21. •• .l::tac8 Threatening . . . .l::i.xd6! . 22 'it'a4 It was better to play 22 l:Id l WUg4+ 23 �f2

i.xc5 24 i.c4 .U.xd6 25 �xd6 i.xd6, although with an extra pawn and the safer king, Black's advantage would be decisive.

22 ••• .l::txc5! (D) 23 'it'xb4 l:Ixd6 The king stuck in the centre and the total lack

of coordination among White's forces outweigh the sacrificed piece.

24 �e1 aS! Deflecting the queen from the d2-square. 25 'it'xb7 After 25 WUa4 .l::tf6 26 iYd4 WUf2+ 27 �dl

.U.cc6, followed by .. . l:i.fd6, Black wins the queen and the game.

CAPABLA.NCA VARIATION: 4 "YJHc2 d5!? 2 7

w

25 .• JU6 26 �b8+ �g7 27 �g3 The queen has managed to defend against

penetration at f2, but the final attack will come from the other side of the board.

27 .• J:tc2 28 i.e2 a4! (D)

Vacating the a5-square for the queen. 29 b4 axb3 30 axb3 �d5 0-1 White resigned in view of the imminent mate.

It is significant that White did not manage to move either of his rooks in the 30 moves.

Game 6 [E35]

Loek van Wely - Veselin Topalov Wijk aan lee 2004

1 d4 ctJf6 2 c4 e6 3 ct:Jc3 Jl.b4 4 �c2 d5 5 cxd5 exd5 6 i.gS c5 (D)

7 a3 Apart from capturing on c5 (which we ex­

amined in Game 5) and playing a3, which is the main line of this game, what else can White do? Defending the pawn with 7 e3 is not very ambi­tious and will lead to a symmetrical position in

the centre where Black does not have anything to fear, since the position of the queen on c2 fa­vours Black because it does not defend d4.

The creative lvanchuk tried to make use of having brought the queen to c2 by playing 7 0-0-0! ? but 7 . . . ..txc3 ! is a typical course of ac­tion when White castles queenside, since it eliminates the knight that defends the castled position and gains time for seizing the c-file. After 8 �xc3 cxd4 9 ..txf6 dxc3 I 0 i.xd8 cxb2+ I I �xb2 �xd8 Black is fine since he has no weaknesses and has a queenside pawn­majority, Ivanchuk-Gashimov, Havana 2007.

It is more logical to defend d4 with 7 ctJf3, but Black can either capture on d4 and play . . . ctJc6 ! , when after the exchange on c6 the po­sition is fairly equal, or else insert 7 . . . h6 ! ?, and although this allows White to gain time to com­plete his development with 8 Jl.xf6 �xf6, he has obtained no advantage with any of the fol­lowing lines:

a) 9 a3 ..txc3+ (the simplest) 1 0 �xc3 ctJd7 I I e3 0-0 I 2 ..tb5 c4 I 3 i.xd7 ..txd7 14 ctJe5

28 CHESS EXPLAINED: THE NIMZO-INDIAN

.i.f5 1 5 b3 b5 1 6 0-0 a5 1 7 bxc4 bxc4 and the passed pawn gives Black a slight advantage, Wang Yue-Asrian, Taiyuan 2006.

b) 9 0-0-0?! .i.xc3! (as soon as White cas­tles queenside, Black captures the knight which attacks d5 and protects the white king) 1 0 �xc3 c4 1 1 e4 0-0! 1 2 exd5 b5 ! and Black's attack on the queenside arrives more quickly, Lopez Co­Ion-Vera, Canaries Team Ch, Las Pal mas 2007.

c) 9 dxc5 0-0 1 0 e3 .i.e6 1 1 l:tc l lt:Jc6 1 2 .i.b5 l:i.ac8 1 3 0-0 .i.xc5 with the bishop-pair and tremendous development, Reinderman­Michalczak, Gausdal 2005 .

7 ••• .i.xc3+ The only move, to obtain a development ad­

vantage in return for the bishop-pair. 8 �xc3! (D) White is very ambitious; now he has the

better structure and the bishop-pair, but the de­velopment factor and the exposed position of the queen work in Black's favour.

The best way to deal with 8 bxc3!? is by 8 . . . lt:Jbd7 !?, with a clearly defined plan of cre­ating problems for White by the manoeuvre . . . �a5 and . . . lt:Je4: 9 e3 �a5 ! 1 0 .i.d3 c4 1 1 .i.f5 lt:Jc5 ! 1 2 .i.xc8 lt:Jd3+ 1 3 'it>e2 .l:i.xc8 14 .i.xf6 gxf6 and the knight on d3 provides com­pensation for Black's poor kingside structure, Drasko-M.Roder, Cutro 2007.

B

8 . . . cxd4!? Obtaining a queenside majority with 8 . . . c4 !?

is a perfectly playable idea. After 9 lt:Jf3 the flexible 9 . . . 0-0! , without determining how to develop the queenside, is the most accurate. If

White replies 1 0 g3, Black can continue with 1 0 . . . lt:Jbd7 and . . . h6, forcing White to capture on f6 when Black will recapture with the knight, followed by . . . .i.f5 with harmonious develop­ment. If White brings his bishop into play with 1 0 e3 instead, then 10 . . . lt:Jc6 and . . . .i.f5, fol­lowed by . . . b5 and .. . a5, is sufficient to equal­ize.

9 �xd4 lt:Jc6 10 �f4 1 0 .i.xf6 ?? loses material after 10 . . . lt:Jxd4 I I

.i.xd8 lt:Jc2 + 1 2 'it'd 1 CDxa 1 . 10 ... d4! It is possible to continue developing with

1 0 . . . .i.e6 l l CDf3 but now l l . ..d4! is essential in order to gain space and prevent White from de­veloping easily with e3. 1 2 b4 �d5 ! (D) is an idea that works very well against the b4 ad­vance.

Black sacrifices a pawn and even a rook, but obtains good counterplay by penetrating with the queen on b3, which creates problems not only for the insecure white king but also for the rook and the white pawns at a3 and b4. For ex­ample, 1 3 .i.xf6 gxf6 14 g3 ( 1 4 �xf6 �b3 ! 15 �xh8+? 'it>e7) 14 . . . �b3 15 �d2 0-0-0 1 6 .i.g2 d3 ! and Black's initiative compensates for his weaknesses, Mena-F.Corrales, Havana 2006.

ll .i.xf6 In response to 1 1 b4 Black can choose be­

tween the thematic 1 1 . . . �d5 !? (to penetrate on b3 or c4, without worrying about 12 .i.xf6 gxf6 1 3 �xf6 �c4 ! 14 �xh8+ 'it>e7, when curiously White cannot avoid the perpetual), and the sim­plifying 1 1 . . .lt:Jd5 !? 1 2 .i.xd8 CDxf4 1 3 .i.c7

CAPABIANCA VARIATION: 4 '¥Jic2 d5!? 29

tt:Je6 14 �d6 f6 I S CDf3 �f7 16 e3 l:td8 1 7 �g3 and now, instead of 1 7 ... b6?! (Van Wely-Gol­oshchapov, Bundesliga 200617), Black can ob­tain good play with 1 7 . . . dxe3 ! 1 8 fxe3 aS ! 19 bS tLle7 20 �c4 CDfS.

ll ... gxf6!? 12 CDf3 If White prevents the check on aS with 1 2

b4!? then Black can use the idea of penetrating with the queen on b3: 1 2 . . . �e6 1 3 CDf3 'iYdS ! 14 �xf6 �b3 ! .

12 ... �a5+!? Here 12 .. . .Si..e6! deserves consideration, wait­

ing for White to determine how he is going to continue his development. If then 13 b4, Black can respond with 1 3 . . . 'iYdS! , while if 1 3 g3 then 13 . . . �aS+ followed by the exchange of queens and .. . 0-0-0.

13 �d2 'iYxd2+ 14 �xd2 (D)

We have reached a strange position where Black has a catastrophic pawn-structure (iso­lated pawn on d4, doubled pawns on f7 and f6, four pawn-islands), but in return has dynamic development which provides sufficient coun­terplay.

The value of the whole variation depends on the assessment of this position. White possesses the structural advantages indicated above, but Black's forces are almost fully mobilized, while White still needs several tempi to bring all his pieces into play.

14 ... �f5 The bishop supports a possible . . . d3 advance

and leaves the e-file free for the king's rook. Here 14 . . . �e6!? was also worth considering,

because on e6 the bishop would be secure and would also control the b3-square.

15 b4 (D)

With this move, White cuts out once and for all the annoying manoeuvre . . . tt:laS-b3, which would give White a lot of headaches. From now on, Black's play is very similar in all variations: after castling, he brings the rooks to the centre, creating problems for the white king with the threat of . . . d3.

Another plan is to initiate skirmishes on the queenside with . . . aS, taking advantage of the fact that a large part of the white force is located on the kingside.

15 •.• 0-0-0 16 licl �b8 17 g3 aS!? (D) Another plan was 1 7 . . . lihe8 1 8 ..ig2 d3 !? 1 9

e3 ..ie4 with the idea of . . . lDeS or . . . lidS-fS, with good counterplay.

18 ..ig2!?

30 CHESS EXPLAINED: THE NIMZO-INDIAN

If 1 8 b5, then 1 8 . . . lt:la7 followed by . . . lt:lc8-d6/b6, or l 8 . . . lt:le5 !? and . . . lt:lg4!?.

18 ••• ..te4 It is essential to restrict the action of the g2-

bishop. If Black allows the exchange of the white bishop for his knight, his structural prob­lems will start to be significant: 1 8 . . . axb4?! 1 9 lt:lh4! .ite6 ( 1 9 . . . ..td7 20 axb4 lt:lxb4 2 1 .itxb7 !? 'iti>xb7 22 �bl and all the black pawns are weak) 20 ..ltxc6 bxc6 2 1 axb4 and White is in com­mand.

19 lt:lh4 White sacrifices the b-pawn in order to com­

plete the mobilization of his forces, obtaining considerable pressure on Black's weaknesses and on the b-file but, as the game itself will demonstrate, this compensation is sufficient only for a draw. The struggle would remain alive after 1 9 b5 ! ? lt:la7 ! 20 b6 !? lt:lb5 !? 2 l l:tc5 d3 ! with great complications.

19 .•• ..txg2 20 lt:lxg2 axb4 21 axb4 lt:lxb4 22 .l:.c4 (D)

B

Black has won a pawn but now it is White who takes the lead and mobilizes his rooks with gain of time. If White manages to regain the pawn he will have a won game owing to his better structure and his centralized king, but this cannot be forced.

22 ••• lt:lc6 23 �bl lt:leS! 24 .l:.i.cb4 .l:Id7 25 lt:Jf4 llc8

Black retains his extra pawn and has acti­vated all his pieces, but his weaknesses give White sufficient chances to equalize.

26 I!a4 Threatening ll l b4. 26 ••• lt:Jc4+ 1/z-•!z The players considered that the best course

was to repeat moves with 27 'ifi>d3 lt:le5+.

Conclusions

The solid Capablanca line can be met with the aggressive 4 . . . d5 !?, which aims above all to open the game as soon as possible and take ad­vantage of White's lag in development.

If White plays the immediate 5 a3 Black can play for . . . lt:le4 and . . . c5, as demonstrated by Game 4 and its notes. The position becomes very complicated if White decides to win the pawn with 8 dxc5 and 9 cxd5, but Ivanchuk's idea 1 0 . . . 'ii'f6 !? seems to give Black enough counterplay for the pawn. Black can also em­ploy the quiet line 6 . . . dxc4 7 llVxc4 b6!? to ex­change the light-squared bishops, equalizing without many problems.

When White plays 5 cxd5 exd5 6 ..ltg5, Black has 6 . . . c5 ! ?, a line which is giving good results, although the last word on it has not yet been spoken. However, the psychological advantage is with Black, since White generally aspires to a quiet position when he employs this variation but here he has to enter fully into the complica­tions by playing 7 dxc5 if he wants to obtain any advantage, since the alternatives do not set Black many problems. Remember that in this case Black must exert pressure as soon as possi­ble with . . . lt:le4 and . . . 'ii'a5 and not fear for his king, or worry about a bad pawn-structure; the priority is to develop the greatest possible activ­ity by mobilizing the queenside and creating threats against White's king and queen.

3 Capablanca Variation: 4 �c2 0-0

Introduction

In this chapter we shall see a quieter way to meet the Capablanca Variation, in which Black plays in a logical manner but without trying to exploit White's lag in development immediately as in the previous chapter.

Here Black castles and plans a scheme of development with . . . b6, . . . ..tb7 and . . . d6, . . .'!iJbd7 and .. . c5, with the idea of exerting pressure on the c-file and controlling the e4-square. Obviously, White has many ways to play and we shall examine each of them.

The Games

In Game 7 (Radjabov-Alekseev) White tries to seize the centre right away with the aggressive 5 e4 !?, but Black reacts immediately with 5 . . . d5, followed by putting his knight on e4 and breaking with .. . c5, which makes way for the queen to go to a5. A recurrent theme in this variation is that Black keeps his knight on e4, even when White plays f3, with the powerful . . . f5 ! , so that if White captures the knight, Black takes on e4 with the f-pawn, attacking the bishop on d3 and opening lines against the white king, which generally has problems escaping from the centre. In this game White gave priority to development with .i.d3, lLle2 and 0-0, but did not achieve anything special.

Games 8- 1 1 all feature 5 a3 .i.xc3+ 6 �xc3 b6.

In Bocharov-Lysy, Game 8, White employed a very clever system based on prioritizing the de­velopment of his kingside (lLlf3, .i.e2 and 0-0) after gaining the bishop-pair. But Black reacted cor­rectly, creating an offensive on the kingside based on a Dutch-type set-up with . . . ltJe4, .. .f5 and the deployment of the queen and the rook to that sector, generating a strong attack. Only mistakes in the conduct of the attack sent Black to his defeat.

In Game 9 (Van Wely-Wells) White played the logical .i.g5 followed by e3, with the modem plan of bringing the g ! -knight to c3 and the queen to c2, controlling the e4-square. In the game, Wells placed unfounded hopes on the attack on White's c4 point, but in so doing he left his d6-pawn very weak, which was exploited by Van Wely to gain a clear advantage. Black's attempts to attack the white king did not yield the results for which he hoped and with some precise moves White managed to castle, retaining both a material and positional advantage. In the notes to the game the best line for Black in this difficult variation is analysed.

The Gustafsson-Postny clash (Game 10), is an example of another knight manoeuvre (lLlf3 and ctJd2) before completing the development of the kingside. In general in these variations, when Black attacks the centre with . . . c5, White captures with dxc5, producing a particular pawn-structure where Black has central superiority and a useful b-file, while White possesses the bishop-pair and pressure on Black's d-pawn. This game is a good example of coordination on the part of Black, who manages with precise moves to bring each piece to a good square, in accordance with the general

32 CHESS EXPLAINED: THE NJMZO-INDJAN

plan of exerting pressure along the half-open b-file. The strongest plans for each side are explained in the notes. In the middlegame White had an opportunity to exploit an inaccuracy by his opponent but he missed it and Black was victorious in the endgame.

Game 11 (M.Gurevich-Leko) is a confrontation between two of the greatest experts in this vari­ation. White employed here the most ambitious system after 7 .1Lg5, which consists of playing 8 f3, supporting the e4 advance. In the game Black played ... d5 and White chose to take play into an end­game which has been debated at the highest level for some years. We shall also analyse other possi­bilities, especially the typical structure following the exchange of pawns by cxd5 and . . . exd5 with White remaining with a backward pawn on the e-file.

Leko's play in this game was very accurate, since he combined the occupation of the open c- and d-files with the creation of some weaknesses in the white centre, which was sufficient to equalize the chances. Overestimating his position led Gurevich to commit some mistakes which finally left him into a hopeless endgame.

Game 7 [ E32]

Teimour Radjabov - Evgeny Alekseev Sochi 200 7

1 d4 lLlf6 2 c4 e6 3 tLlc3 .1Lb4 4 �c2 0-0 (D)

One of the most important variations against 4 '1i'c2. Black waits to see White's intentions and will react accordingly.

5 e4!? This ambitious move became popular in the

rnid- 1 980s and reached its climax when Karpov adopted it in some games from 2000 onwards. It is necessary to know that 5 lLlf3 should be an­swered with 5 . . . c5 ! (not 5 . . . b6? in view of 6 e4! d5 7 e5 lLle4 8 .1Ld3 with a white initiative) 6 dxc5 tLla6 ! and Black will seize control of the e4-square, taking advantage of the white knight

being on f3, which guarantees him equality, even when he loses the bishop-pair. An effec­tive plan is to follow . . . lLlxc5 with .. . b6, . . . ..lk.b7 and . . . lLlce4! or . . . J:k8 with pressure on the c­file.

5 a3 ! is the main line, and will be seen in Games 8-1 1 .

5 ... d5! Black strikes at the white centre. 6 e5 Gaining space. Naturally White cannot cap­

ture twice on d5, since he would end up with an isolated pawn on d4 without any compensation.

6 ••• l2Je4! From now on, many of the variations are

based on the theme of the insecurity of this knight, because it is usual for Black to try to maintain it in this position as long as possible, since the defence of his king side depends on it. Note that if White gives up his light-squared bishop for this knight then Black will always obtain excellent play on the light squares.

7 .1Ld3 Giving priority to development. Remember

that if White tries to dislodge the knight with 7 f3? he runs into 7 . . . �h4+ 8 g3 lLlxg3 9 '1i'f2 lLlf5 and Black saves his piece, remaining a pawn up.

CAPABLANCA VARIATION: 4 ifc2 0-0 33

The main alternative is to strengthen the cen­tre at the expense of development with 7 a3 .1xc3+ 8 bxc3. After 8 . . . c5 !, the attack on the pawn-chain begins, and a path is opened for the queen to aS . White then has two lines:

a) The strange 9 �b2!? cxd4 I 0 cxd4 pro­tects the centre, but posts the bishop rather passively and delays the development of the kingside. Now IO . . :�Va5+? ! fails to I I We2, when the threat of f3 forces Black to make po­sitional concessions. One of the best responses is I 0 . . . ..1d7 ! (keeping the queen on d8, which prevents I I f3 for the moment in view of l l . . .'iVh4+) I I ..id3 'iVaS+ I 2 We2 .llc8! (ex­erting pressure along the file) I 3 f3 (D).

Black should now play 1 3 . . . f5 ! . Remember this idea, which comes up in many positions. Black maintains his knight on e4 or secures a good retreat for it to f6, since if White captures the knight he will be unable save his bishop on d3. 14 exf6 (Black was threatening I4 . . . dxc4 15 .1xc4? .llxc4! and . . . ..ibS) 14 . . . tt'lxf6 (admit­tedly Black remains with a weakness on e6, but for the moment White is not coordinated enough to attack it and is more worried about the secu­rity of his king and the pressure on the c-file) 15 cS .ibS ! and Black exchanges his passive bishop and subsequently prepares to break with . . . e5 or . . . b6.

b) The main variation is 9 ..1d3 'iVaS ! (D). This queen move leaves White with the few­

est options. Black puts pressure not only on c3 but along the whole a5-e I diagonal. I 0 tt'le2 cxd4 I I cxd5 (now I I f3?! is not advisable:

l l . . . dxc4! 12 �xe4 d3 with advantage to Black, T.Rej-Timoshenko, Sydney 2007) I I . . . exd5 1 2 f3 tt'lxc3 1 3 tt'lxd4. Here, Black has an extra pawn and several possible knight discoveries. For his part, White threatens ..ixh7+ and �d2, winning the knight. Usually Black continues 1 3 . . . tt'lb5+ 14 �d2 tt'lxd4 but after 1 5 �xh7+ Wh8 16 ..1xa5 tt'lxc2+ 17 ..ixc2 he stands worse because White has the bishop-pair, the more compact pawn-structure and the more central­ized king. However, Leko' s idea I 3 . . . tt'le4+ ! 14 We2 f5 ! changed the assessment of this posi­tion. We have already mentioned this idea. The knight is maintained on e4 by means of tactics. The disadvantageous position of the king on e2 gives Black sufficient resources. White is un­able to capture the knight since, as Leko points out, Black wins after 1 5 fxe4 ?? fxe4 1 6 ..ibS ..ig4+ 1 7 We3 'i¥b6! and the white king will not survive. Instead, 1 5 exf6 tt'lxf6 leaves Black with an extra pawn, and is no good either. Vallejo-Leko, Morelia/Linares 2006 continued 15 ..ie3 tt'lc6! 16 tt'lxc6 bxc6 17 .llhc l .llb8 !, when the game was very complex but satisfac­tory for Black. Even after the best move, 1 5 e6!, Black has good play following 15 .. . tt'lc6! ac­cording to Leko.

7 ... c5! It is essential to react in the centre, before

White consolidates with tt'le2 and 0-0 . 8 tt'le2 If 8 dxc5?! tt'ld7 ! Black regains the pawn

with the better development. Inserting 8 cxdS exdS before 9 tt'le2 just

gives Black additional possibilities, although it

34 CHESS EXPLAINED: THE NIMZO-INDIAN

generally leads to the same position as in the game after 9 .. . lt:Jc6 10 0-0 cxd4. However, Black can deviate with 9 . . . cxd4 1 0 lt:Jxd4 lt:Jd7 ! ? or af­ter 9 . . . lt:Jc6 l 0 0-0 �f5 !?, with good prospects in both cases.

8 ••• lt:Jc6 9 cxdS exdS 10 0-0 (D)

B

Finally White manages to remove his king from danger, but now his pawn-centre vanishes.

10 .•. cxd4! Preferable to I O . . . lt:Jxd4 I I lt:Jxd4 cxd4 1 2

lt:Jxe4 dxe4 1 3 �xe4, when White has slight pressure on both flanks.

11 lt:JxdS! Worse is 1 1 lt:Jxe4?! dxe4 1 2 �xe4 lt:Jxe5 1 3

�xh7+ 'ii>h8 and White has problems with the h7-bishop and dealing with the passed pawn.

l l ... 'iVxdS 12 �xe4 'i!VxeS 13 a3 (D) White wants to determine the position of the

black bishop as soon as possible and then post his own pieces accordingly. His hopes are based on combining an attack on the kingside (by de­ploying the queen and the rooks towards that sector) with a degree of pressure on Black's queenside and the siege of the queen's pawn. However, Black has enough resources to main­tain dynamic equality. His centralized queen is a major obstacle to White's attack and his passed pawn on d4 does not allow White to take too many liberties.

A more common approach is the direct 1 3 �xh7+ 'it>h8 but apart from a slight initiative for a few moves White has gained nothing con­crete. For example, 1 4 �e4 (if 14 �f4, then l4 . . . 'iVf6 or l 4 . . . �c5 !?, while 14 f4 is met by

1 4 . . . 'iVc5 ! ?, which is still OK) 14 . . . �e8 15 ctJg3 �e6 16 �d2 �d5 and Black stands well, A.Ramirez-Vesco vi, Buenos Aires 2003.

B

13 ... �a5 It is also possible to bring the bishop to the

kingside with 1 3 . . . i:!e8 !? 14 �xh7+ '<t>h8 1 5 �d3 �f8, but from a5 it controls the e 1 -square and later from b6 it supports the queen's pawn.

14 �xh7+ '<t>h8 15 �d3 �g4 A typical idea; by putting pressure on e2

Black provokes a weakness on the a7-gl diago­nal.

16 f3 �d7 If l 6 . . . �e6 then 1 7 f4 !? and 1 8 f5, winning

tempi to initiate the attack on the kingside. 17 �f4 'iVf6 18 'iVcS! (D)

One of the ideas of 1 3 a3 : the queen gains the c5-square to transfer to the kingside.

18 .•• '<t>g8 19 'iVhS g6

CAPABLANCA VARIATION: 4 'ilfc2 0-0 35

White has managed to weaken the royal for­tress further, but Black's queen is very well placed on f6 and his minor pieces are active enough to repel the white attack.

20 i.g5 �e5! 21 'ii'h6 i.d8! It is essential for Black to eliminate White's

annoying dark-squared bishop, or else transfer his own bishop to the king side, before the white attack gains intensity.

22 i.xd8 After 22 i.f4 'ii'g7 White has to exchange

the queens. 22 .• J::taxd8 23 .l:i.ad1 White combines the attack on the kingside

with pressure on the passed pawn. 23 .•• 1H5 Continuing with the policy of simplifying,

which will enhance the value of the passed pawn and weaken White's attack, but with these ex­changes White will gain tempi for the attack at the black queen's expense. Instead, 23 .. . .l:tfe8 !? was worth considering.

24 i.xf5 �xf5 25 f4 A typical plan, to involve the f-pawn and the

fl-rook in the attack. It was possible to bring the knight straight into play with 25 lLlg3 �f6 26 lLle4 but after 26 . . . 'ii'h8 ! 27 'ii'g5 �e5 the at­tack is unsuccessful.

25 ••• nres 26 lLlg3 �f6 27 f5 The future weakening of Black's castled po­

sition guarantees White a draw at least and of­fers some practical chances to create a surprise mating attack.

27 •• .'ilig7 This was the moment to advance the passed

pawn with 27 . . . d3 !? 28 fxg6 'ii'xg6 29 �h4 d2 !, tying the white rooks to the first rank.

28 'ilic1

Other retreats of the queen do not change the general assessment: White's attack is sufficient only to maintain the balance.

w

28 ••• d3! 29 fxg6 After 29 f6 'ii'f8 the attack grinds to a halt. 29 ••• �xg6 (D)

White has managed to weaken the oppo­nent's kingside even more, but the passed pawn and the activity of the centralized black rooks do not allow White's attack to generate serious threats.

30 'it>h1 If 30 lif5, then 30 . . . M.e5 ! defending g5. 30 • • • l::te5 31 'ilif 4 !':tdd5 Controlling the fifth rank, especially the f5-

square. 32 .:d2 �e6 33 h3 �e1 34 !':txe1 'ii'xe1 + 35

Wh2 'ilie6 35 . . . 'ii'e5 !?. 36 ltJe4 Wf8! 37 ltJg3 'ilie5 If2.1f2 Black could have played on, although a draw

is the most probable result, owing to the insecu­rity of his king.

Game 8 [E32]

Dmitri Bocharov - Igor Lysy Moscow 2007

1 d4 lLlf6 2 c4 e6 3 lLlc3 i.b4 4 �c2 0-0 5 a3 The continuation most in tune with 4 'ilic2;

White obtains the bishop-pair without wreck­ing his pawn-structure.

5 .•• i.xc3+ 6 'ii'xc3 b6 The most logical; Black develops his queen­

side and exerts pressure on the centre. The bishop will go to b7 (or a6 in some variations)

36 CHESS EXPLAINED: THE NIMZO-INDIAN

controlling the long diagonal and especially the e4-square. It also supports the central counter­attack with . . . c5.

7 lLlf3 �b7 8 e3 (D)

B

A dangerous system, despite its modest ap­pearance. It has been used with some success by GMs I.Sokolov, Karpov and Van Wely. White speeds up the deployment of all his forces by �e2 and 0-0, followed by b4 (or b3) and �b2, in some move-order or other. Once his develop­ment is complete, White will seek to open the position to favour his bishop-pair. Other sys­tems, based on �g5, will be seen in Games 1 0 and 1 1 .

Black has several schemes for combating this system:

1 ) Complete his development, placing his pawns on dark squares while maintaining con­trol of e4, which he achieves by playing . . . d6, followed by . . . lbbd7 and subsequently . . . c5. To my way of thinking, this formation is rather passive and does not cause White many wor­ries; he just completes his development with �e2, 0-0, b4 and �b2 and subsequently opens the long diagonal with dxc5, obtaining the ad­vantage in all cases.

2) The plan of playing . . . d5, . . . lLlbd7 and then . . . c5, by which he seeks to open the game as soon as possible to exploit White's slight lag in development and the exposed position of the white queen on the c-file. However, with this method Black runs a great positional risk, in that he has to achieve something concrete with his temporary initiative (weakening of the white

pawns, penetrating the enemy camp with the rooks, winning material, etc.), because other­wise he will end up suffering in an open posi­tion against very powerful bishops.

3) The third method of play is the one used in this game. It consists of using the e4-square as a launch-pad for an attack against White's kingside castled position. Black plays ... d6 and .. . lbbd7 and subsequently .. . lbe4 and .. .f5 (remi­niscent of the Dutch Defence) and sends the queen ( . . . �h4 or . . . �f6) or the rook via the third rank ( .. JH6-g6-h6) towards the kingside. In the first games played with this system, the pioneers Larsen and Smyslov obtained brilliant results, but White strengthened his play subse­quently by means of temporary pawn sacrifices (d5 ! ) to activate his bishops. Nowadays the struggle in this variation is wide open.

8 ... d6 (D) It is hardly ever advisable to play . . . �xf3?!

because as well as gaining the bishop-pair White would strengthen his centre and have the g-file available for attacking purposes.

9 �e2 If White plays 9 i..d3, Black can pursue the

same plan of attack recommended against i..e2 ( . . . lbe4 and . . .f5) or else try to exploit the posi­tion of the bishop on d3 with 9 . . . lbbd7 !? 1 0 0-0 c5 1 1 b4 cxd4! . Then if White recaptures with 1 2 exd4 Black replies 1 2 . . . �xf3, when White's structure will be seriously damaged, while after 1 2 lLlxd4 lbe5 ! (hitting the bishop) 1 3 �e2 .l:tc8 the pressure on the c-file gives Black the initia­tive.

CAPABLANCA VARIATION: 4 'i¥c2 0-0 3 7

The move-order 9 b4 !? followed by .ib2 is more dangerous, delaying the deployment of the king's bishop until Black shows his hand. It also retains the possibility of queenside cas­tling. For example, 9 . . . CLJbd7 1 0 .ib2 CLJe4 I I '1Jiic2 f5 12 .id3 ! . Now:

a) 12 . . . CLJg5?! isn't sensible in view of 1 3 d5 ! and the white bishop is hitting f5: 1 3 . . . CLJxf3+ 14 gxf3 lDe5 15 0-0-0 !, with a strong initiative on the g-file.

b) Bringing the queen to the kingside with 12 .. .''1Jie8 and . . . "i¥h5 is not very advisable either since after 1 3 0-0 "i¥h5 14 d5 ! Black is in diffi­culties, because if he takes on d5 the c7-pawn drops.

c) I suggest 12 . . . c5 !?, a temporizing move, waiting for White to determine the position of his king. 1 3 0-0 CLJg5 ! 14 Ct:Jxg5 (if White re­treats the knight, Black takes on d4 and if White recaptures with the pawn then comes . . . CLJh3+ and .. . CLJf4, attacking both the bishop and the g2-square, with an initiative) 14 . . . "i¥xg5 15 f4 '1Jiig6 gives Black the e4-square for his pieces and the b7-bishop has an excellent diagonal.

9 ... CLJbd7 10 0-0 Ct:Je4 1 1 "i¥c2 fS! (D)

We have reached the type of Dutch position that we discussed above. Black is planning to attack on the kingside, while White tries to eject the knight from e4 and bury the b7-bishop with d5.

12 b4 This is the plan chosen by the majority of

experts with White, although the retreats of the f3-knight to e l or d2 in order to dislodge

the black knight from e4 with f3 are also rele­vant.

If 1 2 CLJe l it makes little sense to wait pas­sively for White to play f3, followed by b4, .ib2 and CLJd3, because Black would be left without a plan. It is best to send the queen into the attack with 1 2 . . . "i¥h4 ! , and after 1 3 f3 CLJg5, the threat of . . JH6-g6/h6 gives Black good play.

In the event of I2 CLJd2, 1 2 . . . "i¥g5! is best, taking advantage of the fact that the knight on d2 obstructs the c ! -bishop. Now the typical pawn sacrifice 1 3 d5 !? is indicated, to restrict Black's initiative. Black must either capture on d5 or maintain the tension in the centre, since it would be a serious positional mistake to play 1 3 . . . e5 because then the b7-bishop would be­come a dead piece. After 1 3 . . . exd5 ! 14 cxd5 .ixd5 15 .ic4! CLJef6 1 6 CLJf3 "i¥g4 White had some compensation but not more in M.Gure­vich-Van der Wiel, Hoogeveen 200 I .

12 .•. l:tf6! (D) The start of the thematic plan of attack. The

rook goes to g6 or h6 and in combination with the b7-bishop creates unpleasant threats against the white king. Moving the queen ( 1 2 . . . "i¥f6) has also given Black good practical results. However, trying for counterplay on the queen­side with 1 2 . . . a5?! favours White, since he can tum the open a-file to his advantage. Even if all four rooks are exchanged, White can harass Black with his bishop-pair, as well as threaten to penetrate with his queen at a7.

w

13 dS!

38 CHESS EXPLAINED: THE N!MZO-INDIAN

An important resource in this position, dis­covered by GM I van Sokolov. White tempo­rarily sacrifices the pawn but speeds up his development and limits the opponent's aggres­sive intentions. In these positions, restricting the action of the b7-bishop with d5, or f3 and e4, is White's objective.

Delaying d5 would be met by an instructive stratagem: 1 3 .ib2 l:i.g6 1 4 .l:!.ad l VJJie7 1 5 d5 c5 ! (D).

w

Remember this positional point. Black wants to capture on d5 but without leaving the c7-pawn backward. If White takes on c6 or e6, the long diagonal stays open and the b7-bishop re­mains in the game. For example: 16 dxe6 'iVxe6 and all Black's pieces are active, Rubinetti­Smyslov, World Team Ch, Lucerne 1 985.

13 ••• l:i.g6! Better than 1 3 . . . exd5, because after 14 i.b2

.l:Ig6 1 5 cxd5 i.xd5 16 i.c4 ! (another theme that keeps cropping up: the exchange of Black's active bishop) 1 6 ... i.xc4 17 VJJixc4+ �h8 1 8 VJJid5 ! , the f5-pawn drops and White keeps some initiative.

14 ltJd4! Neither side wants to make any positional

concessions. The knight is in contact with the critical points (e6 and f5) and at the same time makes way for the f-pawn to expel the black knight. If 1 4 i.b2, then 14 . . . c5 ! , while 1 4 dxe6 helps Black to bring his pieces into the attack after 14 . . . ltJf8 followed by . . . ltJxe6.

14 ••• 'iVg5! 15 g3 (D)

15 ••• ltJe5!

B

This is the only continuation that keeps the initiative and does not resign Black to having the worse position. That is what would happen after 1 5 . . . exd5 16 cxd5 .ixd5 1 7 .tc4! (with this simplification an important attacking piece is eliminated and the queenside light squares are weakened, to be penetrated by the white queen) 1 7 . . . .ixc4 1 8 'iVxc4+ 'it>h8 1 9 'ikc6! �d8 20 'iVxc7 ltJe5 2 1 .l:Ia2! and White defended successfully on the kingside and subsequently gained material on the queenside in Kramnik­J.Polgar, Sofia 2005 .

16 ltJxe6 Accepting the challenge. More conservative

is 16 ltJf3 (to reduce the number of attacking pieces) 1 6 .. . 'i¥e7 17 ltJxe5 dxe5 1 8 �d1 , de­fending d5 and keeping the black bishop out of play, but after 1 8 . . . exd5 1 9 cxd5 c6! Black suc­ceeds in opening the diagonal, or else White has to sacrifice a pawn, as in Dobrov-Yakovich, Moscow 2007. After 20 i.b2 cxd5 2 1 �ac l .l:Ic6 22 'i¥a4 l:tac8 White's compensation was only sufficient for a draw at best.

16 .• Jixe6! A forced sacrifice of the exchange which

opens the long diagonal for the b7-bishop; this, combined with the coordinated action of Black's knights and queen, places the white king in se­rious danger.

17 dxe6 VJilg6! Leaving the g5-square free for the knight. 18 l:.d1!? Preparing to return the exchange and vacat­

ing fl to allow the king to escape. In Drozdov­sky-Grishchuk, Odessa 2007, 1 8 'i¥b3 ltJg5 !?

CAPABIANCA VARIATION: 4 'iVc2 0-0 39

( 1 8 . . . h5 !? is interesting, followed by .. . h4) 1 9 e4! f4 ! was played, with great complications.

18 ... ctJg5 19 !:i.d5! (D)

White returns the exchange to block the b7-bishop. If Black accepts, the e6-pawn will be passed and protected.

19 ... ctJgf3+ 19 . . . 'i¥xe6!? was also possible. 20 i.xf3 l2lxf3+ 21 'it>fl l2lxh2+ After 2 l . . .�g4 !?, keeping the white king

confined, Black has at least a draw. 22 'it>e2 'i¥g4+ 23 'it>d2 Now the king takes refuge on the queenside. 23 .•. ctJf3+ 24 'it>c3 b5? (D)

Losing the thread of the game. It was essen­tial to play 24 . . . ..txd5 25 cxd5 ctJel or 25 . . . c5, and the complications continue.

w

25 l:.xb5 ..te4 26 'i¥e2 �g6 27 ..tb2 'i¥xe6 28 'it>b3

The king has finally slipped out of reach of the enemy pieces and White remains the ex­change up.

28 .•. a5?! This hastens his defeat, but after 28 ... d5 29

..tc3 ! dxc4+ 30 '1t>b2 White would safeguard his king, after which his extra material and pressure on the long diagonal should prove decisive.

29 .l:Ixa5 .l:txa5 30 bxa5 d5 31 'it>a2! dxc4 32 ..tc3

The struggle is over. The active black pieces have vanished into thin air, and now all White has to do is mobilize his rook and go over to the counter-attack.

32 ••• �d5 33 .l:tdl ..td3 34 'i¥b2 'it>f7 35 a6 ctJg5 36 �b7 1-0

Game 9 [E32]

Loek van Wely - Peter Wells London 2007

1 d4 ctJf6 2 c4 e6 3 ctJc3 ..tb4 4 'i¥c2 0-0 5 a3 .bc3+ 6 �xc3 b6 7 ..tg5 (D)

The most logical and popular continuation. The bishop goes outside the pawn-chain and pins the knight, taking advantage of the ab­sence of its black counterpart.

7 ... i.b7 This move, followed by . . . d6, . . . ctJbd7 and

the subsequent . . . c5, introduces Black's most popular and logical set-up. If White takes on

c5, Black can recapture either with the b-pawn, obtaining a pawn-majority in the centre and the half-open b-file, or with the knight, if it is im­portant to operate on the c-file and allow the knight to gain the e4-square quickly. If White does not exchange in the centre, Black can cap­ture on d4 himself at some point and then exert pressure on the c-file or play . . . d5 .

The move 7 .. . ..ta6!? also makes a lot of sense, in order to exert pressure on c4 with . . . d6,

40 CHESS EXPLAINED: THE N!MZO-INDIAN

B

. . . tZ'lbd7, . . . c5 and . . . l:tc8. But recently Black has encountered some difficulties after 8 �f3 !?, which forces him to play 8 ... tZ'lc6, blocking the c-pawn.

It is hard to determine whether the inclusion of the moves 7 ... h6 8 ..ih4 makes a significant difference to the position. Normally these moves can be included at any point in the following variations, but in general Black prefers to delay this move until White has committed himself to a kingside development scheme. It is possible to attack the centre immediately with 7 . . . c5, but in that case Black would be showing his hand too early and his opponent could take advan­tage of this by quickly opening the d-file and castling queenside at an opportune moment.

The simplifying idea . . . tZ'le4!? is very com­mon, attacking the white queen on c3 while ex­posing Black's own queen to the attack of the g5-bishop. For this to work, Black has to make sure that he will not lose material in the result­ing exchanges. It usually works if, after captur­ing the queen on c3, the knight is attacking another enemy piece.

8 e3 (D) White has several possible kings ide develop­

ment systems. In this game, he employs a rather modem one which consists of the following piece regrouping: the knight goes to c3 via e2, the queen makes room for the knight and goes to c2 (or d3) controlling e4, while the rook goes to d l to support the d5 advance, to gain space and block the pressure exerted by the black bishop against g2, so that White can finally de­velop his king's bishop to e2 and castle.

Despite the time that all this takes, Black does not have many active plans available and, although Black can easily break in the centre with . . . d5, or . . . cxd4 and .. . d5, White always has slight pressure on the dark squares owing to the absence of Black's king's bishop.

8 ... d6 9 lZ'le2 lZ'lbd7 (D) The immediate 9 . . . c5 is answered by 10

dxc5 ! dxc5 (if IO . . . bxc5? ! , then 1 1 0-0-0! fol­lowed by ..ixf6, when the weakness at d6 forces Black to recapture on f6 with the pawn and, since the queens are still on the board, White can develop a strong attack with lZ'lg3 and tZ'lh5) 1 1 �c2 h6 1 2 ..ih4 �e7 1 3 0-0-0 .l:i.d8 14 l::!.xd8+ �xd8 15 lZ'lc3 and White is a bit better, Kramnik-Bologan, Dortmund 2003. In these stable positions, without any pawn-breaks, the bishop-pair can operate at long distance, sup­porting advances on both wings.

10 �c2!?

CAPABLANCA VARIATION: 4 'iVc2 0-0 41

The main alternative is I 0 'iVd3, which also vacates the c3-square for the knight and con­trols e4, as well as supporting a future d5 ad­vance, but Black can take advantage of the queen's position to create pressure on the c­pawn with the thematic I 0 . . . .ta6!, the strategic objective of which is to exchange one of White's bishops. Now, with or without the inclusion of ... h6 and .th4, many games have been played with I I CLJc3, but after l l . . . d5 12 'iVc2 .txc4 1 3 i.xc4 dxc4 14 'iVa4 c5! the most probable result has been a draw, since White can obtain very little from such a symmetrical position; for ex­ample, 1 5 �dl !? 'iVc8 1 6 'iVxc4 cxd4 17 �xd4 (17 "iVxd4 'iVa6 !) 1 7 . . . 'iVxc4 1 8 �xc4 �fc8 main­taining equality, M.Gurevich-Cheparinov, Metz 2005.

IO ••• cS (D)

The struggle for the centre begins. Note that it is almost never appropriate, in this type of po­sition, to advance in the centre with . . . e5, since White plays d5 and the b7-bishop is left very badly blocked.

11 �dl ! Before continuing with his plan of CLJc3,

White defends d4 with the rook, so he can an­swer .. . cxd4 by recapturing with the rook, main­taining pressure on d6. If White plays 1 1 CLJc3 then Black captures with I I . . . cxd4, closing the d-file, and then focuses on attacking the c4-pawn by . . . 'iVc7, . . . �ac8 and . . . .ta6. It is also possible to follow up with . . . d5, which would leave White with an isolated pawn.

ll. .. cxd4

This move is generally linked with the plan of opening the c-file (after a timely . . . d5) with the hope of achieving counterplay on that file. However, it is better to maintain the flexibility of the position with l l . . .'iVc7 !? 12 CLJc3 a6 ! (D).

With this Black waits for at least one more move to see how White is going to solve the problem of his kingside development, which he has three ways of doing:

a) Sacrificing the g-pawn, which proved in­adequate in the game Lafuente-S.Garcia, Ha­vana 2004 after 1 3 .te2 h6 ! 14 .th4 .txg2 1 5 �gl .tb7 1 6 d5 �fe8 and, with the king on e l , it is difficult to launch an attack on the g-file.

b) Carrying out his scheduled d5 advance, which in this particular position allows Black some queenside activity which influences the centre and exposes the main defect of White's position: his lag in development. After 1 3 d5 exd5 1 4 cxd5 b5 ! White does not have time to stabilize his centre and simultaneously com­plete development: 1 5 .td3 b4 16 axb4 cxb4 1 7 CLJe2 'iVxc2 1 8 .txc2 .txd5 1 9 0-0 .l:Ifc8 20 CLJd4 a5 and the missing pawn counts for more than White's bishop-pair and slight initiative, I.Sok­olov-Ponomariov, Wijk aan Zee 2005 .

c) 1 3 f3 might be the best option, but after l 3 . . . cxd4! 1 4 �xd4 ( 14 exd4?! is answered by l4 . . . d5 ! ) l 4 . . . d5 ! 15 .te2 (there is no time to evade the pin with 1 5 'iVdl due to l 5 . . . e5 ! , while 1 5 cxd5 is answered by 1 5 . . . CLJxd5 ! , threatening . . . CLJxc3 and . . . 'iVe5 ! ) l 5 . . . e5 ! l 6 llh4!? h6 ! and White loses the bishop-pair, after which the game is equal.

42 CHESS EXPLAINED: THE NJMZO-INDIAN

12 llxd4 h6 The swift counter-attack on c4 continues to

be attractive, but with a series of precise moves Van Wely highlights the defects of Black's po­sition.

A more modest plan is to get rid of the weakness at d6 with a well-timed . . . d5 ad­vance, but then White completes his develop­ment and the bishop-pair gives him a slight advantage: 12 . . .'tiVc7 l 3 lLlc3 a6 14 'ifdl ! d5 1 5 cxd5 lLlxd5 1 6 lLlxd5 �xd5 1 7 e4, I.Sokolov­Stefansson, Cal via Olympiad 2004.

13 i.h4 lLle5 14 lLlc3 11c8 15 �g3! (D)

B

The lack of the dark-squared bishop makes itself immediately felt.

15 •• J!fe7 16 lLlb5! Increasing the pressure, regardless of the

lag in development. Now Black loses a pawn and seeks compensation against the white king which is still on e l .

16 ••• lLlc6

Better than 16 . . . 11fd8 17 lLlxa7 and the knight returns later to b5 .

17 .:dl ! e5 18 lLlxd6 ltJd4 Trying to open the centre to get at the white

king.

B

19 exd4! iVxd6 Here 1 9 . . . exd4+? loses a piece to 20 'iVe2. 20 dxe5 iVe7 21 i.e2! (D)

Finally the bishop emerges and the threat of exf6 prevents Black from capturing on g2.

21 • • • lLle4 22 ..itf4 White will lose one of his extra pawns but

will retain the bishop-pair which, combined with the extra pawn, suffices for victory.

22 .•. g5 23 �e3 'iVxe5 24 0-0 'iVe6 25 .Ufel The game is decided: extra pawn, bishop­

pair and safer king. 25 • • • Ir.fe8 26 b3 g4 27 �n 'iWg6 28 l:td7 ..ltc6

29 .l:i.xa7 h5 30 11a6 h4 31 l:!.xb6 .U.e6 32 b4 g3 33 b5 i.a8 34 l:!.xe6 fxe6 35 c5 gxf2+ 36 ..itxf2 .l::tf8 37 c6 1-0

Game 1 0 [E32]

Jan Gustafsson - Evgeny Postny Greek Team Ch� Ermioni 2006

1 d4 lLlf6 2 c4 e6 3 lLlc3 ..itb4 4 'i!Vc2 0-0 5 a3 �xc3+ 6 'iWxc3 b6 7 ..ltg5 i.b7 (D)

8 lLlf3 A modem plan, which consists of first put­

ting the knight on d2 and then playing f3 and e4 (or f3 and e3) followed by i.e2 and 0-0. In this

way the knight does not get in the way of the light-squared bishop and, in addition, on d2 it fulfils the mission of overprotecting c4.

At one time 8 e3 d6 9 f3 lLlbd7 was very pop­ular and White tried two models of develop­ment:

CAPABLANCA VARIATION: 4 "VJilc2 0-0 43

I ) bringing the knight into play via h3, put­ting his bishop on e2 and castling;

2) simple development with .i.d3 followed by lL'le2.

In the first case (lLih3 and ..ie2) Black ob­tains good play by breaking with . . . c5 and an­swering the capture dxc5 by retaking with the b-pawn, which gives rise to a typical pawn­structure in which Black has a strong centre and the half-open b-file to compensate for White's bishop-pair.

In the second case, when White plays ..id3 and lL'le2, Black also plays . . . c5, but if White takes on c5, Black recaptures with the knight, attacking the bishop on d3 and subsequently exerting pressure on the c-file. If White keeps the pawn on d4, Black takes on d4 at some point and creates play against the c4-pawn and the queen on c3 ( . . . .l::!.c8, . . . ..ia6 and . . . d5).

Let us see a few examples of both positions: a) I 0 lt'Jh3 h6 I I ..ih4 c5 !? 1 2 dxc5 bxc5 1 3

�e2 (if 13 0-0-0, then 1 3 . . . 'iVb6! followed by ... d5, . . . ..ic6 and . . . ki.ab8 with an attack on the white king; here 1 3 . . . lt'Je4? fails to 14 'iVe l ! ­remember this idea! - by retreating to e l the queen defends the bishop) 1 3 . . . a5 14 0-0 a4 1 5 �ad l �a6 and Black will have counterplay on the b-file and can also play in the centre at some point with . . . d5 or, if necessary, with . . . e5. This position is quite similar to the one in the main game, but on h3 the knight is worse than on d2.

b) I 0 .i.d3 h6 I I .i.h4 c5! 1 2 lt'Je2 (after 12 dxc5 Black replies 12 . . . lt'Jxc5! 13 ..ic2 .l::!.c8, when he is ready to exert pressure on c4 with ... lL'lcd7 and ... d5; if then I 4 ..ixf6 'iVxf6 1 5

'iVxf6 gxf6, the ending is good for Black, be­cause he breaks with . . . d5 and his development advantage makes itself felt, or he puts the king on e7, defending d6 comfortably) 1 2 . . . .l::!.c8 and he starts counterplay on the c-file by . . . cxd4 and . . . d5 ! or . . . .i.a6.

8 ••• d6 9 lt'Jd2 (D) Against the natural development 9 e3 lt'Jbd7

I 0 ..ie2 Black has achieved excellent results with I O . . . 'iVe8 ! followed by . . . lt'Je4 and . . . f5. Here the bishop on g5 only helps Black's game and if White exchanges on f6 Black does not have any problem at all.

9 ... c5!? I consider that this i s the most precise move

to reach the typical structure of this variation in which White takes on c5 and Black recaptures with the b-pawn. However, the most common line is 9 . . . lt'Jbd7 I 0 f3 (if White plays I 0 e3, Black takes advantage of the omission of ... h6 and ..ih4 by playing I O . . . .l::!.c8 I I f3 c5 1 2 dxc5 lhc5 ! 1 3 ..ih4 b5 ! with excellent counterplay on the c-file, Krush-Shirov, Edmonton 2005). Now:

a) Black can oppose the e4 advance with I O . . . d5 I I cxd5 ! exd5 (Black will exert pres­sure on the weakened e3-pawn with . . . .l::!.e8 and . . . 'iVe7, but even then he will fail to equalize) 12 e3 .l::!.e8 1 3 .i.e2 ! .l::!.c8 14 0-0 'iVe7 1 5 .i.b5 ! c6 16 .i.a4 h6 1 7 .i.xf6 lt'Jxf6 1 8 l:i.fe l , Topalov-Leko, Morelia/Linares 2006, and according to Topa­lov the position is more pleasant for White to play, with moves such as b4, .l:!.ac I and the prep­aration of e4.

44 CHESS EXPLAINED: THE NIMZO-INDIAN

b) It is preferable to play 1 0 . . . c5 !? 1 1 e4 ! ? ( 1 1 dxc5 bxc5 brings about the pawn-structure of the main game) 1 1. . .Ite8! 1 2 �d3 ( 1 2 �e2 lt:Jd5! leads to equality) 1 2 .. . cxd4 1 3 'ii'xd4 lt:Jc5 14 �c2 e5 ! (a novel strategy, seeking control of the dark squares) 1 5 �f2 ( 1 5 �e3? lt:Je6 16 �h4 lt:Jg4) 1 5 .. . lt:Jg4 ! and White loses his dark­squared bishop, after which the position be­comes equal.

10 dxc5 10 f3 h6 1 1 �h4 cxd4 12 'ii'xd4 lt:Jc6 1 3 'iVc3

gives Black a choice between 1 3 . . . lt:Je5 fol­lowed by . . . l:Ic8, and 1 3 . . . e5 followed by . . . l:.c8 and . . . d5, when Black has good prospects ow­ing to his better development.

1 0 e3 lt:Jbd7 1 1 f3 .l:.c8 transposes to Krush­Shirov, mentioned above.

10 ... bxc5 11 e3 The endings after 1 1 �xf6 'ii'xf6 1 2 'ib'xf6

gxf6 are comfortable for Black, who has better development and can bring the king to e7 to de­fend d6 if necessary.

ll ... lt:Jbd7 12 �d3 (D)

12 ... a5 Black will double rooks on the b-file in the

most elegant way ( . . . .l':la6-b6 and . . . l:.b8); addi­tionally the a-pawn has the function of pre­venting b4. Accepting the pawn by 1 2 . . . �xg2 would give White good compensation on the g-file.

13 0-0 l:.a6 It is more accurate to play 1 3 . . . a4 !? (fixing the

queenside pawns and preventing the plan of b3 and a4, followed by lt:Jbl -a3-b5, neutralizing

Black's pressure on the b-file) 14 l:.adl .Ma6! (D).

Seeking the most harmonious set-up for his pieces: the rook defends d6 and will go to b6, the black queen will gain a tempo with .. . �a8, attacking g2 and defending a4, while the king's rook heads for b8. Then the bishop can go to c6 or a6 and the queen to b7 or b8, exerting maxi­mum pressure on b2. I.Sokolov-Beliavsky, Bel­grade 1 99 1 continued 15 'iVc2 't!Va8 16 f3 h6?! (according to Sokolov it was better to play 16 .. . .l:.b8 and .. . �b6) 17 �h4 llb6 18 lt:Jbl .Mb8 19 �g3 ! but now instead of the passive 19 . . . e5?!, Beliavsky should have complicated the game with 19 .. . lt:Je5 ! .

14 �c2?! The plan of blocking the b-file by putting a

knight on b5 is the best option for White here, although I have no doubt that subsequently Black can gain counterplay by the central mo­bilization with . . . d5 and . . . e5 . Hence White should play 14 b3 ! and a4 as soon as possible and only then bring the knight to b5 . In the game B.Esen-Aiexandrov, Saint Vincent 2005, White delayed this manoeuvre and was out­played after 14 .l:tadl .U.b6 ( 1 4 . . . a4 ! ) 1 5 �c2?! ( 1 5 b3 !) 15 . . . �a8 16 f3 h6 17 �h4 l:tb8 1 8 lLlbl lt:Je5 1 9 �xf6 gxf6 20 lt:Jd2 �c6 21 .l:tbl a4 22 �e4 f5, when Black was better.

14 ... h6 (D) The advantage of including . . . h6 is, natu­

rally, that the king has luft, but from h4 the bishop can go to g3, attacking d6. Perhaps the immediate 14 .. .'�Va8 is better.

CAPABLANCA VARIATION: 4 'iYc2 0-0 45

15 ..lth4 �a8 16 f3 t!.b8 He should prevent b3 with 1 6 . . . a4!?. 17 lt:Jb1 t!.b6 18 lt:Jc3 ..lta6 19 t!.ab1 Here 19 lt::lbS !? was worth considering, al­

though Black can exchange on b5 and then try to win the b-pawn: 1 9 . . . ..\txbS 20 cxb5 a4 !?.

19 ... lt:Je5! (D)

To prevent White from plugging the b-file with lLlbS, although the kingside is now weak­ened.

20 ..ltxf6 gxf6 21 lt:Je4 f5!? 21 . . .�g7 ! is more solid. 22 lLlf6+ ®g7 23 lLlh5+ ®f8 The black king abandons the weakened king­

side and takes refuge on e7. 24 e4 t!.b3?! Trying to exert pressure on the white pieces,

so that they take up passive positions and can­not go on to the counter-attack, but this move leaves d6 unprotected. After White's next move,

the queen will come to d2 with simultaneous threats against d6 and h6. It was better to play 24 . . . 'iib7 !?.

25 .te2! fxe4 26 fxe4? (D) Losing a precious tempo. It was necessary to

go after the black king as soon as possible with 26 �d2 !. Black then has difficulty defending, as the following variations demonstrate: 26 ... ®e7 (26 . . . �c6 27 �xh6+ ®e7 28 �f6+ ®d7 29 f4 !) 27 t!.bd l �c6 28 �xh6! exf3 (28 . . . e3 29 �f6+ ®f8 30 lt::lg7 !) 29 �f6+ ®f8 30 i.xf3 ! lt:Jxf3+ 3 1 t!.xf3 t!.xf3 32 gxf3 with a fierce attack on the dark squares by the queen and the knight.

26 ... ®e7! Once his king is under cover, Black's posi­

tional advantages will tip the balance in his fa-vour.

27 t!.fd1 It was better to choose 27 lLlf6! with counter­

play. 27 ... t!.8b6 Preparing to triple on the b-file and at the

same time making room for the queen to come to the defence of the kingside with . . . �h8 or . .. �g8-g5, as appropriate.

28 lt:Jg3?! Passivity leads directly to defeat. 28 i.fl

was better. 28 ... �b8 Now either the b2-pawn or the a3-pawn

drops. 29 �d2 Trying to launch a belated attack on the black

king. If 29 t!.d2 Black has 29 . . . t!.xa3 or the

46 CHESS EXPLAINED: THE NIMZO-!NDIAN

positional 29 . . . iVg8, followed by . . . iVg5 with total command.

29 •• J::txb2 30 �xb2 �xb2 31 iVxh6 �bl ! (D)

With the disappearance of the rooks, all pos­sibility of an attack is eliminated, because the queen and the knight are unable to hunt down the black king.

32 iVg5+ 'it>d7 33 iVd2 33 iVxe5? fails to 33 . . . �xdl + and . . . dxe5 . 33 •.• Si.xc4 The white position is collapsing. The passed

c-pawn will soon decide the struggle, whereas the h-pawn has a long way to go.

34 h4 Si.xe2 35 iVxe2 c4! 36 'it>h2

Trying to get the king under cover. If 36 h5, then 36 ... iVb6+ 37 'it>h2 (37 'it>h I c3 !) 37...l:ib2' 38 iVfl lLlg4+ 39 'it>h I lLlf2+ and the rook drops.

B

36 ... �xdl Here 36 . . . c3 ! was better. 37 iVxdl 'iVb3 38 iVd4 (D)

The last attempt to seek perpetual check, but it is well-known that with a knight in close at­tendance the king is very well protected from queen checks.

38 ... c3 39 iVa7+ 'it>e8 40 iVa8+ cJ:ie7 41 iVa7+ lLJd7 42 iVxa5 c2 43 iVg5+ f6 44 'i'd2 'iVb2 45 lLle2 cl iV 0-1

Game 1 1 [E32]

Mikhail Gurevich - Peter Leko Elista 2007

I d4 lLlf6 2 c4 e6 3 l2Jc3 Si.b4 4 'iVc2 0-0 5 a3 Si.xc3+ 6 iVxc3 b6 7 Si.g5 Si.b7 8 f3 h6!

In this line, where Black will play . . . d5 and in some variations recapture with . . . lLlxd5, at­tacking the queen, it is appropriate to force the bishop to h4, so that White does not have iVd2, defending the bishop.

9 Si.h4 d5 (D) The natural response, in order to prevent the

central advance e4. It is essential to keep one point in mind: after playing f3, if White takes on d5 then Black will have the option of recap­turing with the pawn, to exert pressure on the

CAPABLANCA VARIATION: 4 'iYc2 0-0 47

backward pawn on the e-file. As the reader will notice, it is also quite usual to recapture on d5 with the knight; it all depends on small details W which we shall try to explain in the following annotations.

10 e3 Exchanging immediately with 1 0 cxd5 forces

IO . . . exd5 (because I O .. .<!iJxd5? would lose a pawn after 1 1 �xd8 lLlxc3 1 2 �xc7) 1 1 e3 but now Black has available the accurate 1 1 . . .l:re8! (D), which attacks e3 and prevents White's ideal development with 12 �d3 and 1 3 lLle2.

White can complete his development in vari­ous ways, but Black always obtains counter­play:

a) I 2 lLlh3 �e7 1 3 �f2 c5 ! 14 �b5 �c6! 1 5 �e2 ( 1 5 .txc6 ltJxc6 16 0-0 .:.ac8 and the white pieces are very passive) 1 5 . . . �d7 ! . It should be noted that the knight is not very well placed on h3.

b) 12 �f2 !? (defending e3, to reach the op­timal position with �d3 and lLle2) 12 . . . c5 ! (the immediate break in the centre is best, so that the knight can come to c6) 1 3 �b5 cxd4 14 �xd4 and now the safest is 14 . . . ltJc6! 1 5 �xc6 �xc6 16 lZ'le2 �b5 !, making it difficult for White to castle.

c) 12 .tb5 c6 1 3 �a4 ltJe4 !? (taking advan­tage of the position of the bishop on a4) 14 �xd8 'Llxc3 1 5 bxc3 z:i.xd8 with approximate equality, since White has lost the bishop-pair, Van Wely-Grishchuk, Biel 2007.

10 ... <i:Jbd7 (D) 1 1 cxd5!?

If White plays 1 1 �d3 before capturing on d5, Black seizes the initiative by attacking the centre immediately with l l . . . c5 !? 12 cxd5 cxd4! 1 3 �xd4 (if 1 3 exd4 l::!.c8 14 �d2 �xd5 the iso­lated pawn on d4 means that Black is at least equal) 1 3 . . . e5 ! 14 �xf6 lLlxf6! 1 5 �xe5 ltJxd5 16 �d4 lLlxe3 ! 1 7 �xe3 .l:te8 1 8 �e4 f5 regain­ing the piece with an advantage in develop­ment, Candela-Xu Jun, Linares 1 998.

1 1 ••• ltJxd5! 1 1 . . .exd5?! allows White to set up his ideal

position with 1 2 �d3 and 1 3 lLle2. 12 �xd8 ltJxc3 13 �h4 Black would regain the pawn immediately

after 1 3 �xc7?! ltJd5 ! . 13 ... ltJd5 14 �f2 (D)

B

The classic position, which has been reached in innumerable contests. White retains the bishop-pair and does not have any permanent weakness; furthermore he threatens to push the

48 CHESS EXPLAINED: THE NIMZO-INDIAN

black pieces back with e4. His only problem is his lag in development, which will be exploited by Black to try to penetrate with his rooks or create some permanent weakness to compen­sate for the advantage of the bishop-pair. If Black achieves any of these, or manages to ex­change one of White 's bishops, he succeeds in equalizing completely.

14 ... c5!? Consistent with the above-mentioned ideas

of opening files. Black can also restrain White's central expansion with 14 . . .f5, and subsequently break in the centre with . . . c5 or . . . e5 .

15 e4 For a time it seemed that 1 5 i.b5 placed

Black in difficulties, but Kasparov showed that 15 . . . .llfd8 ! keeps everything in order. White achieves nothing by taking with 16 i.xd7 .llxd7 17 dxc5 because after 1 7 . . . bxc5 1 8 CDe2 CDb6! Black has excellent counterplay on the light squares and commands the d-file. 16 e4 is stron­ger, but then comes l6 . . . CDc7 ! (a resolute pawn sacrifice to open the centre and exploit the posi­tion of the white king on e 1 ) 1 7 i.xd7 .llxd7 1 8 dxc5 f5 ! . This is the natural follow-up to the preceding moves; if White now takes on f5 he opens the e-file onto his king, while allowing Black to capture on e4 after 19 cxb6 axb6 20 CDe2 fxe4 2 1 fxe4 i.xe4 leaves White strug­gling for equality, Kramnik-Kasparov, World Ch (8), London 2000.

15 • . . ltJe7 The knight retreats to a square where it can­

not be attacked and is ready to support the . . .f5 break. 1 5 . . . CDf4 16 ..lte3 CDg6 is somewhat infe­rior as the knight is worse on g6 than on e 7.

16 ltJe2 (D) Development can be completed rapidly with

16 ..ltb5, but then Black exchanges one of the bishops with 1 6 . . . i.c6, which is sufficient to equalize.

Here 16 0-0-0?! is very risky because it com­mits the king prematurely to the queenside, giv­ing Black the chance to seize the initiative by combining an attack on the king with weakening White's centre: l6 .. . Mac8 1 7 'it>bl c4 ! (blocking the development of the fl -bishop and preparing to attack the white centre) 1 8 CDe2?! ( 1 8 .lle l was better) l 8 . . .f5 ! 1 9 exf5 CDxf5 20 CDc3 CDf6

2 1 i.e2 CDd5 22 CDxd5 i.xd5 and, in M.Gure­vich-Leko, Elista 2007, Black had managed to damage White 's pawn-structure and secure an outpost in front of the isolated pawn. His plan consists of advancing the queenside pawn­majority without allowing White's bishops to become active.

B

16 ... f5! Immediately attacking the white centre in

order to inflict some permanent weakness, or at least hinder White's development. Trying to penetrate along the open files allows White to gain a small advantage: 16 . . . .llac8 17 CDc3 cxd4 1 8 i.xd4 CDc5 19 .U.d l l:i.fd8 20 ..te3 ! .U.xd l + 21 'it>xd l e5 22 b4 ! CDe6 23 'it>c2 CDc6 24 'it>b2 and White sheltered his king on the queenside and subsequently prevailed in Kramnik-Leko, Dort­mund 2006.

17 ltJg3 Worse is 17 CDf4 ?! fxe4 18 CDxe6 .llf6 and

White's poor development becomes obvious. The only way to avoid being left with a weak pawn on e4 is 1 7 exf5 ! CDxf5 1 8 dxc5 CDxc5 but then the activity of the black pieces allows him to equalize, even though his pawn-structure is inferior.

17 ... fxe4 18 fxe4 cxd4 19 ..ltxd4 .U.ac8 (D) The position is equal: the bishop-pair is bal­

anced by the better development of the black pieces.

20 1:i.d1 Preventing . . . CDc5 as well as . . . CDf6 because

White would exchange, followed by . . . .lld7. 20 •.• 1:i.fd8 21 ..tc3 ltJg6 22 ..ltbS ltJcS 23 0-0

CAPABIANCA VARIATION: 4 "VJic2 0-0 49

Gurevich is hoping for an exchange of e­pawns, so that the bishops will give him a small advantage in an open position where he has no weaknesses.

23 ... �a6! (D)

As usual in this variation, when Black ex­changes one of White's bishops he is com­pletely out of danger.

24 �xa6 tt:Jxa6 25 I!.xd8+?! Surrendering the d-file in order to try to keep

the a6-knight out of play. It was better to play 25 tt'lh5 e5 26 'it>f2, bringing the king to e3, with equality.

25 ... .llxd8 26 b4?! Now the bishop loses its natural support and

the queenside pawns become weak. 26 tt'lh5 was indicated.

26 ... tt'lc7 27 a4 Continuing with the idea of keeping the

knight out of play.

27 ... b5! 28 aS tt:Je8 (D)

w

The knight is best placed on e8 for the time being. lt defends the g7 -pawn and prepares to in­vade enemy territory with ... tt'ld6-c4 or . . . l2lf6.

29 l2le2?! Misplacing the knight. 29 .llf3 !? is better. 29 •.• e5 30 l:!.cl tt:Jf6 31 l2lg3 h5! Black has the initiative. 32 h3? It was necessary to defend with 32 h4!? but

after 32 . . . a6 ! ? White has an arduous defence. 32 . . • a6! Leko slightly improves his position and elim­

inates the possibility of his opponent compli­cating the game with the advance a6.

33 c,i{h2? (D) It was essential to seek immediate counter­

play on the c-file with 33 ii.el ! .

33 . .. �c8!

50 CHESS EXPLAINED: THE NIMZO-INDIAN

Now he wins a pawn after the exchange of rooks and White no longer has counterplay with a6.

34 i.d2 l:txcl 35 i.xcl h4 36 ltJf5 ltJxe4 37 g3 hxg3+ 38 ltJxg3 ltJxg3 39 'it>xg3 'it>f7 40 'it>g4 ltJe7 0-1

White resigned because he is unable to put up much resistance to the advance of the e-pawn (supported by the knight and the black king) and at the same time defend his b- and h-pawns.

Conclusions

The line with 5 e4 !? (featured in Game 7) has two basic interpretations: White either gives priority to his development with 7 i.d3 and ltJe2 or else strengthens his centre with the im­mediate 7 a3. In the first case, after the simplifi­cations in the centre, Black is left with a passed pawn on d4 and White with a slight initiative which with correct play by both sides tends to disappear. In the event of 7 a3 the struggle be­comes sharper but with accurate play ( . . . c5 ! and .. .'iVa5 ! ) the chances level out. Remember that it is important to maintain the knight on e4 to defend the kingside and harass White, even when there is a white pawn on f3, as occurred in Vallejo-Leko (note 'b' to White's 7th move in Game 7), since the resource .. .f5 ! is almost al­ways available.

The system with 8 e3 (seen in Game 8), keeping the queen's bishop inside the pawn­chain, is very dangerous, despite its quiet ap­pearance. Black does well to initiate play on the light squares with . . . ltJe4 and .. .f5. Here it is es­sential for Black to keep in mind the possible retreats of the white knight, in order to eject the black knight with f3, and at that moment con­sider sending the queen into the attack (for in­stance, meeting ltJel with ... �h4 and ltJd2 by . . .'ifg5).

White's d5 advance, even as a pawn sacrifice, is one of the basic ideas of this position. If White manages to activate his queen's bishop on b2, exchange the light-squared bishops and pene­trate on the c-file, he will gain the advantage, so it is essential for Black to speed up his attack on the kingside and even sacrifice material, as in

Game 8, Bocharov-Lysy ( l6 .. J:he6). Another important idea is to answer d5 with a well-timed . . . c5, to be able to capture .. . exd5 without leav­ing the c7-pawn exposed.

The line with 7 i.g5 followed by e3 and the manoeuvre �c2 and ltJe2-c3 (Game 9) is quite strong. White brings his pieces to secure squares and with .l:!.d I overprotects his centre and pre­pares the advance d5 to restrict the action of Black's b7-bishop. The best course for Black seems not to hurry with the capture on d4 and to adopt a waiting policy with l l . . .�c7 !? 1 2 l'Llc3 a6 ! , to be prepared against the d5 advance or else force White to play f3 in order to complete his development.

When White chooses ltJf3-d2 and i.e2, or ltJh3 and i.e2 (lines examined in Game l 0), Black should play ... c5 and after the compul­sory capture dxc5 (otherwise at some point Black plays . . . cxd4, exerting pressure along the c-file and later creating permanent weaknesses in White's centre with . . . d5) Black should re­capture with . . . bxc5, seeking counterplay on the b-file and advancing his central majority at some point, which allows him to gain sufficient counterplay. White has several plans, such as exerting pressure on d6 and trying to bring a knight to the b5-square, blocking the b-file.

If White chooses the formation ii.d3 and lLle2, Black should first develop the queenside ( . . . d6 and . . . ltJbd7) and then play . . . c5, so that if White captures by dxc5 Black can recapture with the knight, attacking the bishop on d3. Otherwise, Black creates pressure on c4 with . . . .:tc8, . . . i.a6 and . . . d5 .

If White plays i.g5 followed by the ambi­tious f3 (Game I I ), Black should reply . . . d5 and bear in mind that if White captures on d5 there are two ways to recapture: with the knight, exchanging the queens, and with the pawn, in order to exert pressure on the backward pawn on e3. Black should choose the recapture with the pawn if White is unable to achieve his ideal set-up with ii.d3, lLle2 and i.f2.

In the variations in which White captures with the queen on c3 and then plays li.g5, both sides must constantly keep an eye on the unpin­ning move . . . ltJe4!?.

4 Rubinstein System: 4 e3 0-0 : Variations with ttJe2

Introduction

4 e3, known as the Rubinstein System, is the most popular move against the Nimzo-lndian De­fence. White simply fortifies the centre and continues his kingside development, for which he has three logical patterns of development available:

I) .i.d3 followed by tZ'lf3 and 0-0. 2) .i.d3 followed by lZ'le2. 3) lZ'le2 followed by a3, to gain the bishop-pair without doubling the pawns. White's most problematic piece to bring into play is the c 1 -bishop. The most logical continuations for meeting the Rubinstein System are 4 . . . b6, 4 . . . d5, 4 . . . c5 and

4.. .0-0. The move-order that I suggest is the most natural and flexible: 4 . . . 0-0, followed by . . . d5 . With

4 ... 0-0, Black allows White to show his cards, i.e. he waits for White to choose the type of set-up he will use and then selects the most suitable development in response.

One advantage of 4 . . . 0-0 over 4 . . . c5 is that in the event of White playing 5 lZ'le2 and a3, the b4-bishop can retreat along the a3-f8 diagonal, leaving the white pieces rather awkwardly placed.

4 . . . 0-0 is generally associated with the central advance . . . d5 and . . . c5 but, as we have already noted, Black first wants to see what moves White will make.

In this chapter we shall see some of White's main set-ups in response to 4 . . . 0-0, before reaching the position that occurs after I d4 lZ'lf6 2 c4 e6 3 lZ'lc3 .i.b4 4 e3 0-0 5 .i.d3 d5 6 tZ'lf3 c5 7 0-0, which will be the theme of the next chapter.

It should be pointed out that at any moment White can insert the move a3 (without having de­fended the c3-knight with another piece) to gain the bishop-pair and reach positions characteristic of the Samisch Variation, in which Black has already played the move . . . d5 or . . . 0-0, so that (after ... d5) White can easily get rid of the doubled pawns on the c-file whilst maintaining the advantage of the bishop-pair, or, in the event of the location of the black king having been determined by ... 0-0, organize his offensive more easily.

However, it should be noted that in these variations with 4 e3 the c 1 -bishop cannot be activated as easi ly as in the Samisch or Capablanca Variations, in which it could go to g5 at any moment.

The Games

In Nakamura-Volokitin (Game 12) we see White's attempt to exploit Black's early commitment of his king by transposing into a Samisch Variation. Black blockaded the doubled pawns with . . . c5 and subsequently went after the c4-pawn with the sequence . . . lZ'lc6, . . . l2la5, . . . b6 and . . . .i.a6. We shall see how Black neutralized White's attack on the kingside but how several slack moves on his part allowed White to obtain a dominating position which the American GM rounded off with an elegant manoeuvre.

52 CHESS EXPLAINED: THE NIMZO-INDIAN

In Game 13, I.Sokolov-J.Polgar, we see once again the advance a3 by White, but only after Black has carried out the advance . . . d5, so that after the exchange of bishop for knight White can quickly get rid of the doubled pawns. We shall see how Polgar chooses the most favourable pawn­structure to combat the bishop-pair and the white centre, avoiding positions of the Botvinnik­Capablanca type, which we shall discuss during this chapter and the next. This game is very in­structive, because Black achieved a blockade of the centre and of the enemy bishops and concluded the game with a strong attack on the king.

In the game Alexandrov-Le Quang (Game 14), White opts for the set-up with ..id3 and lt:Je2, but after capturing on d5. In this very fluid position, we shall see how Black should retain his dark­squared bishop and when and how he should react to White's plans of expansion in the centre and on the kingside. The game is a good example of well-timed counterplay by Black in the centre, but he let his experienced opponent off the hook on at least two occasions.

In Milov-Sharavdorj (Game 15), the Swiss GM also arranges his minor pieces as in the previous example (..id3 and lt:Je2) but without the exchange cxd5. In these positions, White is left with an isolated pawn on d4, but with the unusual feature that the king's knight is placed on e2. We shall see some of Black's schemes for fighting against the isolated pawn in these conditions, but we shall de­vote particular attention to Tal's popular retreat ( I O . . . ..id6) and Ivanchuk's method ( 1 2 .. .'iYc7 !?) used in the game. We shall also look at Karpov's plan of 1 2 . . . .l:le8.

Y.Gonzalez-A.Aivarez (Game 16) looks at the famous Reshevsky Variation (5 lt:Je2) in which White gives priority to placing the knight on e2 before developing the king's bishop, immediately seeking the advantage of the bishop-pair while retaining a perfect pawn-structure.

Black does well to retain his two bishops, but must be careful in choosing how to deploy his forces, because although the white position seems inoffensive and Black has many different ways to develop, it is essential to take care not to fall under an irresistible attack, as occurred in the game.

Game 1 2 [E29]

Hikaru Nakamura - Andrei Volokitin Cuernavaca 2006

1 d4 lt:Jf6 2 c4 e6 3 lt:Jc3 ..ib4 4 e3 0-0 5 a3!? In the first chapter, we studied the Samisch

Variation with 4 a3 or 4 f3, in which White's usual plan consists of gaining the bishop-pair and bolstering the d4-square, so as then to play e4. So it would seem that playing 4 e3 first is a waste of time if he continues 5 a3 !? . However, White wants to show that Black's quick king­side castling actually makes it easier for him to launch an attack. Moreover, there are lines where White regains the tempo by playing e4 and f4 (rather than f3-f4 ) , and a direct transpo­sition to the Samisch occurs - indeed, this i s the case in our main game here. Nevertheless,

practice has shown that Black has sufficient resources for active defence combined with a counter-attack on the c4-pawn.

5 ••• ..ixc3+ 6 bxc3 c5! (D) This move is easy to recommend; Black fixes

the doubled pawns and is ready to attack them with the typical sequence . . . b6, . . . ..ia6 and . . . lt:Jc6-a5, followed by . . . .l:lc8 if appropriate. It is possible to play this formation without in­cluding the move . . . c5, but in that case White can take advantage of not having played the move f3 to create direct threats on the castled king, because his queen has the open d l -h5 di­agonal at its disposal.

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH tjje2 53

7 i..d3 lbc6 8 lbe2 Best. The pawn sacrifice with 8 e4 !? can be

accepted with 8 . . . cxd4 9 cxd4 lbxd4 10 e5 l"Lle8 !? or declined with the central counter­attack 8 . . . d5 ! ? 9 cxd5 exd5 I 0 e5 lbe4 I I lbe2 Ji.g4 ! , reminding White that his king is still in the centre. 1 2 i..b2 ( 1 2 f3 i..xf3 ! ) 1 2 . . . cxd4 1 3 cxd4 �b6 14 i..xe4 dxe4 1 5 i..c3 Mfd8 ! gave Black strong pressure on White's centre in V.Georgiev-Efimenko, Port Erin 2007.

8 . . . b6 9 e4! Black must now take measures against the

threat of i..g5. 9 ... lbe8! (D)

A well-known manoeuvre devised by Capa­blanca. Black prevents the i..g5 pin and at the same time is ready to increase his pressure on the c4-pawn with . . . lbd6. In addition he frees his f-pawn, which will play a major role in the defence of his castled position.

10 0-0 This natural move has the idea of subse­

quently advancing the f-pawn, but other plans have been tried here, such as:

a) 10 e5 i..a6 I I �a4 lba5 1 2 dxc5 �c7 ! (a typical sacrifice that we shall see over and over again in the Nimzo-Indian when White has dou­bled pawns on the c-file) 1 3 i..e3 f6 !? ( 1 3 . . . bxc5) 14 Md I bxc5 15 exf6 C'bxf6 1 6 i..f4 e5 17 i..g3

d6 and White's bishops are not very active, Moskalenko-M.Roder, B arbera 2000.

b) 1 0 i..e3 i..a6 ! (without delay) I I lbg3 (remember that capturing the pawn with I I dxc5 always gives Black good play: l l . . .lbe5 ! ) i l . . .lba5 1 2 �e2 Mc8 and after 1 3 Me l or 1 3 d5 Black should play the prophylactic 1 3 . . . �h4! followed by ... lbd6. From h4 the queen has in­fluence along the fourth rank and restricts White's actions on the kingside.

1 0 . . . i..a6! (D)

1 1 f4!? With this move, we have now transposed to a

position that can also arise from the Samisch Variation.

The advance of the f-pawn is White's best plan of attack. Against other moves Black sim­ply carries on with his plan of . . . lba5 and . . . Mc8 or . . . d5 and defends the kingside with . . . f5 or . . . f6. The only alternative worth con­sidering is the immediate advance I I e5, with the idea of �c2, in order to force Black to weaken his kingside. l l . . .d6 ( l l . . .lba5 !? 1 2 �c2 f5 i s worth considering) 1 2 �c2 h6 1 3 Md l dxe5 1 4 dxe5 �c7 1 5 i..f4 lbxe5 1 6 i..e4

54 CHESS EXPLAINED: THE NIMZO-INDIAN

and now instead of 1 6 . . . J::tc8, Stefanova-Lahno, Krasnoturinsk (women) 2007, Black should play 1 6 . . . .l:tb8 ! in order, after 1 7 'i!Va4 lt:lf6 1 8 i.xe5 'iVxe5 1 9 i.f3, to have 1 9 . . . i.b7 ! sup­ported by the rook.

l l ... f5 (D) It is also interesting to wait for White to ad­

vance f5 and then reply with . . . f6, which also leads to a lively struggle: I I .. .lt:la5 1 2 f5 f6! (it is vital to prevent 1 3 f6! lt:lxf6 14 i.g5 with a strong attack) 1 3 lt:lf4 i.xc4 14 i.xc4 lt:lxc4 1 5 fxe6 lt:lc7 ! with complex play, Stefanova-Sar­gissian, Zafra 2007.

w

12 exf5 It is best to exchange on f5 and then close the

c-file. If White keeps the tension with 1 2 lt:lg3 g6 1 3 i.e3, then the plan devised by Karpov works peifectly: 1 3 . . . cxd4 1 4 cxd4 d5 ! (an excellent move, designed to block the e3-bishop com­pletely and exchange the light-squared bish­ops) 1 5 cxd5 .ixd3 1 6 'iVxd3 fxe4! 17 �xe4 'iVxd5 1 8 'iVxd5 exd5 and, owing to his control of the light squares and his pressure on d4, Black does not have any great problems. This has occurred in a number of games, starting with Yusupov-Karpov, Linares 1993.

The move-order 1 2 dxc5 is less accurate in view of 1 2 ... fxe4!? 1 3 i.xe4 i.xc4 14 i.e3 lt:lf6 and the control of the light squares gives Black good play.

It had appeared that White could gain better chances with the advance 1 2 d5 ! ?, because after the thematic !2 .. . lt:la5?! 1 3 e5 ! i.xc4 14 i.xc4 lt:lxc4 1 5 d6 the strong chain of white pawns

hems in Black' s forces. However, in Radjabov­Leko, Monte Carlo (rapid) 2007 Black played the simple 1 2 .. . exd5 ! 1 3 cxd5 i.xd3 14 'ii'xd3 fxe4 15 'i!Vxe4 lt:ld6 16 'ii'd3 lt:la5. Here the ac­tive black knights have outposts on c4, b3 and d6, whereas White does not have anywhere to attack.

12 •.• exf5 13 dxc5 bxc5 14 i.e3 d6! Better than moving the queen away with

1 4 . . . �a5?! . 15 lt:lg3 g6 (D)

This interesting position has been debated in several games recently. In the long term Black has the advantage, because White's bishops and knight do not appear to be very active, and it is easier to attack White's weakness on c4 than Black's backward pawn on d6. However, the black king is more vulnerable than the op­posing king and White can benefit from the present lack of coordination among Black's forces.

16 i.xf5!? White has also tried to gain the advantage by

regrouping the rook to e I , the knight to d5 via fl -e3 and then the e3-bishop to h4, but Black can become active on the e-file and bring his knight to e4, with good play. Let us see a recent example: 16 .l:te I lt:la5 17 lt:lfl lt:lf6 1 8 i.f2 'ii'd7 19 lt:le3 .l::!.ae8 20 i.h4 .ib7 !? 2 1 i.fl lt:le4 22 lt:ld5 i.xd5 23 cxd5 c4 and the outposted black knights are better than the clumsy white bishops, I.Sokolov-Grishchuk, Canada de Cala­trava (rapid) 2007.

16 ••• gxf5 17 'ii'd5+ :r7

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH ClJe2 55

Better than 17 . . . �h8 !? 1 8 'i¥ xc6 .i.xc4 19 �fd 1 'i¥c8 20 .l::i.xd6!? lt:Jxd6 21 'iVxd6 'iVe6 22 'i¥xc5 with compensation for the exchange, Hillarp Persson-K.Ong, Stockholm 2007.

18 'i¥xc6 .i.b7!? 19 'i¥a4 (D)

An important moment. At the cost of a pawn, Black has managed to divert the white queen from the kingside and increase the activity of the b7-bishop.

19 .• .'�f6 Activating the queen, but the older move

19 .. . h5 !? is possibly better, followed by the im­mediate further advance of the h-pawn, creat­ing strong counterplay against g2.

20 �ab1 lt:Jg7 lf 20 . . . 'i¥xc3 2 l .l::i.fe 1 lt:Jg7 22 .i.f2 White has

a small advantage based on his command of the open files and his safer king.

21 'i¥d1 �e8! (D) Activating the rook. 2 l . . .'i¥xc3?! is inferior

due to 22 .i.xc5! dxc5 23 .l::i.xb7 ! ; after these simplifications the weakness of the black king becomes clearer.

22 �e1 After 22 .i.d2 'i¥g6 and . . . h5, Black would

continue with his counterplay on the kingside, by advancing his h-pawn.

22 ... 'i¥g6?! Very ambitious. After 22 . . . 'iVxc3 ! the pres­

sure of the black pieces would keep the bal­ance.

23 .i.f2! lixe1+ 24 'i¥xe1 hS Black was relying on this attack, but Naka­

mura's unexpected reply blocks the h-pawn and

at the same time gives his king an escape­square.

25 h4! .l::i.d7 26 'iVd1 Tying Black to the defence of h5 and creat­

ing threats on the light squares. 26 ••• �h7 27 'i¥a4 'i¥f7 28 lld1 lt:Je6 29 .i.e3

'i¥g7?! Occupying the knight's ideal square; it was

better to play 29 . . . �g6 ! right away. 30 �f2 �g6 (D)

w

The black pieces are defending each other. Now White carries out a tremendously effec­tive regrouping.

31 l:ld2! Suddenly there is a threat of 'i¥d 1 hitting d6

and h5. 31...'i¥e7 32 'i¥d1! 'iVxh4 33 llxd6 lle7 Passive defence also would not have helped:

33 . . . l:txd6 34 'iVxd6 'iVf6 35 'i¥d7 .i.e4 36 'i¥e8+ 'iVf7 37 �h8 and White wins material.

56 CHESS EXPLAINED: THE NIMZO-INDIAN

34 l:!.d7! 'ifh2!? Staking everything on the complications. 35 !Ixe7 'ii'xg2+ 36 'it>el �xg3+ 37 �d2?! More accurate was 37 i..f2 ! 'ifxc3+ 38 �d2

and White wins.

w

37 ••• 'it>f6! 38 !th7 tt:Jxf4 39 �n Here 39 'iig 1 ! would allow less counterplay. 39 ... lLlg2 40 i..xc5 i..e4! (D)

Maintaining threats and allowing Black to fight on, although objectively the game is lost.

41 i..d4+ We6 42 c5! With the deadly threat of 'ifc4+. 42 ... �g5+ 43 'it>dl �g4+ 44 'lt>cl 'ifg5+ 45

Wdl 'ifg4+ 46 �e2! lLlf4 It is essential to exchange queens; otherwise

the black king will perish immediately.

47 !Ih6+ 'it>d7 48 '*iVxg4 hxg4 49 l:.h7+ 'tt>c6 50 �xa7 g3 51 :lg7 g2 52 '1t>d2

52 c4 !? is better, making it more difficult for the black king to become active.

52 ... tt:Je6 53 1lg8 (D)

53 ... f4 53 . . . 'it>d5 !? 54 i..f2 'lt>c4 would offer more

resistance. 54 i..f2 f3 55 a4 tt:Jf4 56 We3 With the pawns blockaded, Black does not

have anything he can do. The rest is simple. 56 ... i..d5 57 l:tg4 lLle2 58 c4 i..f7 59 a5 lLlc3

60 !If4 i..h5 61 '1t>d2 lLldl 62 1:1f6+ ctlc7 63 'lt>el tt:Jb2 64 i..g3+ Wd7 65 'it>f2 lLldl + 66 'lt>gl tt:Jc3 67 i..f2 tt:Je4 68 c6+ '1t>d8 69 l:,e6 tt:Jxf2 70 'it>xf2 i..f7 71 a6 ..ixe6 72 a7 1-0

Game 1 3 [E49]

Ivan Sokolov - Judit Polgar Hoogeveen 2006

1 d4 lLlf6 2 c4 e6 3 lLlc3 ..ib4 4 e3 0-0 5 i..d3 d5 6 a3

This move has the same objectives as in the previous game, but it is better now, since White will not have any problems getting rid of his doubled pawns.

6 ... i..xc3+ 6 . . . i..e7 is possible, leading to positions of

the Queen's Gambit in which White has his bishop on c 1 , behind the pawn-chain. However, Black's wasted tempo with the bishop favours

White, because the move a3 can also prove use­ful.

7 bxc3 dxc4! If Black delayed this capture, White would

take on d5 himself. At first sight it seems to come to the same thing, but this is not the case. After 7 . . . c5 8 cxd5 exd5 9 lLle2 (D) we would reach a fixed structure in the centre that be­came very famous after Botvinnik's victory over Capablanca at the AVRO tournament in 1 938.

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH fDe2 57

B

Black enjoys easy development and does not have any weaknesses; he even has a plan for eliminating one of White's bishops ( . . . b6 and .. . .ia6), which in other cases guarantees equal­ity. However, White's central majority is very strong and his plan is clear: carry out the ad­vance e4 and e5 and attack on the kingside. This is achieved with 0-0, f3, etJg3 and in some cases .ib2 to defend d4. There are also other manoeuvres such as �a2-e2 and even the ad­vance g4. In this type of contest, White aban­dons the queenside, where he loses material in many cases, but the attack generated on the kingside is often decisive.

8 .ixc4 cS (D) Attacking the centre and giving the queen

access to aS or c7 .

White has the bishop-pair and a pawn-ma­jority in the centre, but he is slightly behind in development and the position is more open

and unstable than in the Botvinnik-Capablanca structure. His long-term plans are to advance with e4 or c4, to activate his dark-squared bishop and generate threats on the kingside, or else create a passed pawn on the d-file.

Black's slight development advantage can be used to disturb White's development and ex­change one of his bishops (keep in mind the variations with . . . b6 and . . . �a6), or else Black can strike at d4 with the advance . . . e5, which is not possible if Black allows the exchange on d5. It should be pointed out that in the end­game, after the exchange on d4, Black retains a queenside pawn-majority which, combined with control of the c-file, can yield an advantage.

9 etJe2 Best. White develops his knight and defends

his c-pawn while leaving the f-pawn free to ad­vance.

9 etJf3 is answered with 9 .. .'iVa5 !? and Black answers 10 �b2 or I 0 Vi'c2 with I O . . . cxd4, leaving White with an isolated pawn-couple on c3 and d4. After 10 �d2 Black can eliminate the bishop-pair immediately with I O . . . etJe4 ! ?, but even better is 1 O . . . Vi'c7 ! in order to continue with the plan of . . . b6 and . . . �a6. For example, I I �d3 b6 1 2 e4 ( 1 2 Vi'e2! ? prevents . . . �a6, ex­changing the bishops, but after 1 2 . . . �b7 1 3 e4 Vi'c6 ! White is compelled to enter complications in order to defend his central pawns) 1 2 . . . �a6 1 3 �xa6 etJxa6 14 Vi'e2 'iVb7 15 e5 etJe4! (D).

This is the type of position that Black is aim­ing for, since he has the better pawn-structure and control of the light squares, especially d5.

58 CHESS EXPLAINED: THE NIMZO-INDIAN

His queenside pawn-majority constitutes an ad­vantage in the endgame and he only has to take precautions so that White cannot bring a knight to d6 or develop an attack on the kingside, where he has more space.

9 .. :fic7 Vacating the d-file and putting the question

to the bishop, forcing it to decide which diago­nal it will occupy.

10 i..a2 Against 1 0 �d3 Black can continue with the

plan of 1 0 . . . b6 followed by . . . �a6 although, bearing in mind that the bishop has abandoned the a2-g8 diagonal, it is also logical to play . . . e5 and . . . tt:Jc6.

10 ... b6! Polgar takes control of the a6-fl diagonal

and c4 in particular, so that subsequently she can exchange bishops on that square.

11 0-0 �a6 12 l:re1 (D) The other way to move the rook away from

the pressure of the bishop is 1 2 f3 tt::lc6 1 3 l:rf2, but on f2 the rook does not make a good impres­sion. Black would bring his rooks to the centre and exchange bishops on c4 ( . . . tt::la5 and . . . �c4 ).

B

12 •.• tt:Jc6 13 tt::lg3 From g3 the knight supports e4 and can jump

to h5, or f5, if Black plays . . . e5 at some point. If 1 3 �b2 Black is fine if he follows the

well-known recipe of bringing the bishop to c4: 1 3 . . . tt::la5 ! 1 4 tt::lg3 �c4.

13 .. Jlad8 The idea of restraining the e4 advance will

be reinforced with ... e5 and ... .l:1d7 followed by

... l':.fd8. To me it seems more logical to play 1 3 . . . .l:1fd8 ! in order to place the other rook on c8 and continue quickly with the plan of ... CDa5 and ... �c4, so that White has no time to bring his queen' s rook to d l and answer . . . i..c4 with �b l . If Black exchanges the light-squared bishops, he does not mind much if White ad­vances with e4, since Black would then obtain counterplay on the light squares and the c-file.

14 �b2 14 tt::lh5 eliminates the defender of Black's

castled position, but after 14 .. . tt::lxh5 1 5 'i¥'xh5 �d3 ! White will have difficulty mobilizing his central pawns to activate his bishops.

14 ... e5 15 'ikc2! (D) In this way White is able to deploy his rook

harmoniously on d l , instead of the less logical 1 5 .i:i.c l played in the game Vallejo-Illescas, Spanish Team Ch, Sanxenxo 2004.

Polgar's recommendation 1 5 d5 !? was played in Haimovich-E.Berg, European Clubs Cup, Kerner 2007; after 1 5 . . . tt:Ja5 1 6 e4 tt::le8 !? (here Polgar recommended 16 . . . �c4 but 17 i.c l ! gives White the better game) 1 7 'ikg4 lt:'ld6 1 8 .l:.ad 1 'iVd7 1 9 "fih5 Black embarked upon the conquest of the queenside with 19 . . . 'i.Va4, giv­ing White chances of counterplay on the king­side. It was more prudent to keep the queen in the centre and penetrate with the minor pieces at c4.

B

1S . . J:td7 1 5 . . . tt::la5 is answered with 16 dxe5 ! and 17

c4! sacrificing White's structure but opening the a l -h8 diagonal for the b2-bishop.

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH tiJe2 59

16 .l::!.ad1 .l::!.fd8 17 h3 A useful move which does not spoil any-

thing. According to Polgar, it was preferable to W exchange the knights on e4 to weaken the king-side and then create threats in that sector with .1bl .

17 ... c4!? (D) Note that Black is unable to increase the

pressure on d4, which is why she decides to complete the total blockade of the position so that the bishops do not have any diagonals. 1 7 . . . tt'la5? fails to the thematic 1 8 dxeS 'i¥xe5 1 9 c4 'i¥e6 20 ..lii.bl !? 'i¥xc4 2 1 .l::!.xd7 .l::!.xd7 22 'i'fS and Black is defenceless.

18 a4? Weakening the b4-square prematurely. It was

better to play 1 8 ..lii.b I ! g6 and only then 1 9 a4, answering 1 9 . . . .l::!.e8 with 20 e4 ! .

18 ... .l::!.e8! 19 �a3 Sokolov should have prevented the advance

.. . e4 with 1 9 lbe4, but not 19 e4? because with his last move he weakened the b4-square and 1 9 . . . exd4 20 cxd4 lbb4 ! would follow, with ini­tiative to Black.

19 ... e4! (D) Completing the blockade of the position,

which leaves Black with more space and the dS-square for her pieces. This allows her to take the offensive on the kingside. Meanwhile, White is confined to only three ranks and lacks pawn­breaks.

20 ..lii.b1 .l::!.d5 The concentration of force on the kingside

will create problems for White.

21 lDe2! .l::!.gS 22 lDf4 �c8?! This leaves the c4-pawn undefended, allow­

ing some counterplay. It was better to play 22 . . . 'i¥d7 !? with the idea of . . . lbd5 exchanging the defending knight.

23 'i¥e2 As Polgar points out, the correct 23 ..lii.a2 !

would have compelled Black to acknowledge her error with 23 . . . ..lii.a6 or retreat the knight from c6.

23 ••• 'i¥d7 24 'i¥xc4 lba5 25 'i¥b4?! This puts the queen in range of the knight. 25

'i¥e2 was preferable. 25 .•. lbd5! (D)

26 lbxd5? White should have kept his defensive bastion

f4 at all costs by means of 26 'i¥d6! ? lbxc3 27 .l::i.c l lbxbl 28 .l::!.xbl 'i¥xa4 29 �b4 with strong counterplay for the pawn, because the black pieces have become disorganized.

60 CHESS EXPLAINED: THE NIMZO-INDIAN

26 ••• 'i'xd5!? 27 h4?! The best defence was 27 'i'd6 ! 'i'f5 28 'i'f4 !

l2lc4 29 ..ltc I iV g6 30 'i'h2 ..ltd7 3 1 ..ltc2 and White resists, although obviously Black keeps the initiative.

27 .. ..l::i.g4 28 'i'b5 'i'd8! 29 g3 a6! Dislodging the queen from b5, where it at­

tacks the rook on e8. 30 'i'b4?! It was more prudent bring pieces to the de­

fence of the king with 30 �fl 'i'xh4 3 1 ..ltd6!?. 30 ... 'i'xh4 31 ..ltxe4 lixg3+! (D) After this exchange sacrifice, the white king

remains at the mercy of the black pieces, be­cause there is no way to defend the light squares.

32 fxg3 'i'xe4 33 l::i.d2 33 d5 l2lc4 34 l::i.d4 'i'f3 ! also loses for White.

33 ... 'i'f3 34 'it>h2 'i'h5+ 35 'it>gl 'i'f3 36 'it>h2 .ll.f5! 37 e4 .ll.xe4 38 lixe4 'i'xe4 39 dS l2lc4 0-1

Game 14 [E48]

Alexei Alexandrov - Kim Le Quang Abu Dhabi 200 7

1 d4 l2lf6 2 c4 e6 3 l2lc3 .ll.b4 4 e3 0-0 5 .ll.d3 d5 6 cxd5

Completely ruling out the possibility of . . . dxc4.

6 ... exd5 7 l2le2 (D)

B

This line became popular following the young Kasparov's victories over Yurtaev in 1 98 1 and later against Tal in 1 987. At the pres­ent time GMs Alexandrov and Milov are its strongest exponents. White stabilizes the centre

and develops the kingside harmoniously. As in other variations where the knight goes to e2, the f-pawn is left free to support his main plan based on the central expansion by f3 and e4 or on the flank attack with f3 and g4. Also typical is the transfer of the queen to the king side by 'i'e l -h4.

However, White has difficulties in mobiliz­ing the c 1 -bishop and as soon as he plays f3, Black replies with . . . c5, putting pressure on d4 and hindering e4.

7 .•. lie8 The rook holds up the e4 advance. In addi­

tion, the f8-square becomes available for the bishop.

The immediate 7 . . . c5 has the drawback of shutting off the retreat of the b4-bishop. White's reply would be 8 0-0 l2lc6 9 a3 cxd4 (after 9 . . . .ll.xc3?! 1 0 bxc3 White reaches the favour­able Botvinnik-Capablanca structure that we mentioned earlier) 10 axb4 ! dxc3 I I b5 ! l2Je5 (after l l . ..cxb2?! 12 ..ltxb2 White's bishops are worth more than the pawn) 12 l2lxc3 and White has control of the dark squares (in particular the d4-square, which is useful as a base for the queen or the rook after lia4-d4) together with

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH t'De2 61

pressure on the a-file and on Black's isolated queen' s pawn.

8 .id2 This flexible developing move (8 'i¥c2 is sim­

ilar) has the aim of delaying kingside castling so that Black cannot create any immediate threats against White's king. It also delays the move f3, so that if Black rushes with the advance . . . c5, White captures on c5 and plays against the iso­lated pawn without having a weakness on e3. After the natural 8 0-0, Black can retreat the bishop to a more active square with 8 . . . .id6! (D), creating threats against White's castled po­sition.

Let us see the main ideas in this position: a) White 's first attempt to gain an advan­

tage here was with 9 f3. Black reacted in an ex­emplary manner with 9 . . . c5! and after 10 �el tt:lc6 I I �h4 Black defends successfully with I l . . ..ie7 12 'i¥f2 a6 1 3 g4 b5 followed by . . . b4, . . . aS and . . . it..a6 with excellent counterplay or with i l . . .lZ:le7 ! followed by . . . lZ:lg6.

b) White has also played 9 a3, which rather neglects the safety of the king. Kasparov-Junior, New York (5) 2003 featured 9 . . . c6 and after 10 "ti'c2 the computer sacrificed the bishop unex­pectedly with I O . . . .ixh2+!? I I 'it>xh2 lZ:lg4+ 12 Wg3 �g5, obtaining a strong attack which ended in a draw, while Acs destroyed Van Wely at Hoogeveen 2002, with the spectacular 9.. .l2lg4!? 1 0 h3? ( 1 0 lZ:lf4! is better) I O . . . lZ:lh2 ! I I .l:tei lZ:Jf3+! (D).

12 gxf3 �g5+ 1 3 Wh l �h4 I4 lZ:Jf4? .ixh3 15 l2lcxd5 �e6 ! ! ! 6 lZ:lxe6 .if5+ I7 Wg l 'iVh2+

1 8 'it>fl .ig3 ! ! and White resigned because the mate is unstoppable.

c) Epishin has regularly used the prophylac­tic 9 h3 and afterwards he continues with 'i¥c2 and �d I hindering the . . . c5 break and intending to follow up with the minority attack with �bl and b4. In that case, Black should look for counterplay on the kingside by bringing the knight from b8 to h4 ( . . . lZ:lbd7-f8-g6-h4).

We now return to 8 .id2 (D):

8 . . . a6 An unpretentious but useful move. Black

covers the b5-square and prepares to play ... .id6 against any other white waiting move.

The direct 8 . . . c5?! is not accurate owing to 9 a3 ! , when White gains the bishop-pair.

Once again 8 . . . .id6! ? is possible, to which White can respond with 9 �c2 c6 I 0 f3 ( I 0 h3!? is interesting, followed by g4 !? and 0-0-0) I O . . . c5! I I dxc5 .ixc5 1 2 e4 dxe4 1 3 lZ:lxe4

62 CHESS EXPLAINED: THE NJMZO-INDIAN

tt'lxe4 14 i.xe4 �6! 1 5 .i.xh7+ <Ji>h8 1 6 .i.e4 i.e6 and the insecure position of the white king gave sufficient compensation for the pawn in Milov-Ehlvest, Merida 2006.

9 tt'lg3 Alexandrov has tried plans of containment

which include this move, 0-0, f3 and the transfer of the c3-knight to e2. Objectively this brings nothing special, but in practice Black has to de­cide when i t i s prudent to advance with . . . c5 and how to arrange his pieces in the most effective way. These positions are reached basically after the moves 9 0-0 i.d6 1 0 l:tc 1 . If Black plays . . . c5 immediately, without waiting for White to play f3, then he remains with an isolated pawn without the compensation of a white weakness on e3. For example, 1 0 . . . c5 1 1 dxc5 i.xc5 1 2 tt'lb5 tt'lbd7 1 3 tt'lbd4 and White has rather the more pleasant position, Delchev-Graf, Spanish Team Ch, Torrelavega 2007. So it is better to play 1 0 . . . tt'lbd7, to meet 1 1 f3 with 1 1 . . .c5 ! .

9 ••• tt'lc6!? (D)

An interesting idea which forces White to castle right away in view ofthe threat of . . . tt'lxd4. The knight temporarily obstructs the c-pawn, the advance of which is essential as soon as White plays f3, but Black can recycle it to e7 or b4, as occurs in the game.

10 0-0 i.d6 11 f3 1 1 a3 prevents the following manoeuvre, but

Black has various interesting options, such as 1 1 . . .tt'le7 !? controlling f5 and preparing . . . h5 or . . . c5 as appropriate.

ll ... tt'lb4! (D)

Reaffirming that as soon as White plays f3, Black should react immediately with .. . c5.

12 .i.b1 c5! 13 dxc5 With the following sequence of moves, White

tries to take advantage of the loose position of the black pieces, but this will be counterpro­ductive. 1 3 tt'lce2!? was more prudent.

13 ... .i.xc5 14 tt'la4 i.f8 15 �b3 d4! This unexpected tactical shot takes advan­

tage of the momentary lack of coordination among the white forces.

16 .i.xb4 .l:!.xe3! This is the key: the queen cannot maintain

the defence of both bishops. 17 .i.d3 .i.e6! More accurate than 1 7 . . . .i.xb4. 18 �a3 .i.xb4 19 .i.xh7+ 'it>xh7 20 �xb4 bS

21 tt'lc5 .i.c4! (D)

This shows the advantage of having placed the bishop on e6 on move 17 . Black's bishop

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH fbe2 63

supports the advance of the d-pawn and blocks the path of the queen.

22 l':!.fdl d3 23 'it>hl Removing the king from the g 1 -a7 diagonal. 23 .. .'iVb6 24 b3 CLJd5?! Stronger was 24 . . . a5 ! 25 'i¥a3 (25 CLJa4?!

fails to 25 ... axb4 26 CLJxb6 i..xb3 !, winning) 25 . . . b4 26 CLJa4 ! 'i¥e6 27 'iVc I i..b5 28 CLJc5 'i¥e5 29 ctJce4 CLlxe4 30 CLlxe4 (30 fxe4 loses to 30 . . . l':!.xg3 ! 3 1 hxg3 'i¥d4 !) 30 ... l':!.e2 with a dom­inant position.

25 l2Ja4! (D)

An unexpected defence in an extremely awk­ward position .

25 ... CLJxb4 Against 25 . . . 'i¥c7 or 25 . . . 'iVh6, 26 'i¥c5 ! de­

fends. 26 lLlxb6 l':!.b8 27 l2Jxc4 bxc4 28 bxc4 l':!.c8 Black maintains the initiative, but the posi­

tion has simplified and the passed pawn on d3 is

not very secure, because the pieces guarding it can be dislodged.

29 a3?! More accurate was 29 l':!.ab1 ! CLlxa2 30 CLJfl

l':!.e2 3 1 l':!.xd3 l':!.xc4 32 l':!.d2 with a minimal ad­vantage to Black.

29 ••• l2Jc2 30 l':!.acl l':!.xc4 31 CLJfl (D)

3l...l':!.e2?! Overlooking a golden opportunity to main­

tain the passed pawn with 3 1 . . .l':!.c3! , when 32 CLlxe3? is not possible because after 32 . . . CLlxe3 33 l':!.xc3 CLJxd 1 34 'Jd..c7 CLJf2+ 35 'it>g 1 d2 the rook would drop. White must settle for 32 '>t>g l l':!.e2, with a great advantage for Black.

32 l':!.xd3 With the elimination of the passed pawn, the

dangers are over. 32 ••• l2Jxa3! 33 l':!.xc4 CLJxc4 34 \t>gl The king comes to support the defence. 34 ••• a5 35 .l:!c3 CLJb2 1/z- 1/z

Game 1 5 [E4 1 ]

Vadim Milov - Dashzeveg Sharavdorj Las Vegas 2006

1 d4 CLJf6 2 c4 e6 3 tLlc3 i..b4 4 e3 0-0 5 i..d3 d5 6 ctJe2 (D)

In this variation the knight supports its col­league on c3 but without defining the central position (with 6 cxd5) as in the previous game. Black must strike at once in the centre, since otherwise White plays a3 and is a bit better.

6 ••• c5 The thematic response. Another interesting

variation is 6 . . . dxc4 7 i..xc4 e5 ! ?, liberating the c8-bishop and attacking the centre.

7 cxd5 If White calmly castles here, he will end up

with an isolated pawn on d4 which Black can

64 CHESS EXPlAINED: THE NIMZO-INDIAN

B

easily blockade: 7 0-0 cxd4 8 exd4 dxc4 9 �xc4 and now 9 . . . tt:'lbd7 ! followed by . . . tt:'lb6, . . . .i.d7 and . . . �c6, or 9 . . . tt:'lc6 followed by . . . b6, . . . �b7 and . . . tt:'lce7-d5. In both cases Black establishes a cast-iron blockade on d5, which serves as a base of operations for Black. White would base his hopes on attacking on the kingside with .i.b3-c2 and 'iVd3 to provoke weaknesses and then bring the rooks to the centre (�fe 1 and .l:i.ad 1 ) and his queen to the kingside (via h3 and g3), avoiding piece ex­changes as far as possible. The struggle is bal­anced.

After the ever-present 7 a3 Black can choose from two options:

a) 7 . . . cxd4! 8 exd4 (after 8 axb4 dxc3 9 tt:'lxc3 tt:'lc6! 1 0 b5 tt:'le5 ! 1 1 cxd5 tt:'lxd3+ 1 2 'ikxd3 tt:'lxd5 the position becomes completely equal) 8 . . . dxc4 9 �xc4 �e7 and Black prepares to fight against the isolated pawn, safe in the knowledge that the knight is worse placed on e2 than on f3.

b) 7 . . . �xc3+! ? 8 tt:'lxc3 (8 bxc3 transposes after 8 . . . dxc4 9 �xc4 'iVc7 ! to Game 1 3, I.Soko-lov-J.Polgar) 8 . . . cxd4 9 exd4 dxc4 1 0 .i.xc4 tt:'lc6 I I �e3 and Black plays . . . b6 followed by . . . �b7 and . . . tt:'le7 or . . . tt:'la5, whereas White's bishops are rather passive.

7 ... cxd4! 8 exd4 'Llxd5! Leaving White with an isolated d-pawn.

8 ... exd5 is worse, because rigid central struc­tures favour the side with the initiative, in this case White. Furthermore, with this structure the knight on e2 is more useful than the one on f6, since it can go to f4, putting pressure on Black's

queen's pawn, and it allows f3, protecting the squares e4 and g4.

9 0-0 tt:'lc6 (D)

The main position. White has an isolated pawn and tries to organize the typical kingside attack, but on e2 the knight is more passive than on f3, since it does not control e5 or g5. The most usual attacking formation is i.c2 and �d3, provoking a slight weakening of Black's castled position, although other ideas have been employed which we shall see later.

10 �c2 1 0 �c2 does not achieve anything; Black

can secure easy play with I O . . . h6 (Gelfand­Topalov, Monte Carlo (blindfold) 2006), intend-ing . . . i.d6, . . . tt:'lce7 and . . . �d7. Also possible is I O . . . g6 !? followed by . . . .l:i.e8 and . . . �f8-g7.

The quiet move I 0 i.e4 deploys the bishop on the long diagonal, puts pressure on d5 and prepares �d3 gaining time, but Black does not have any problems completing his develop-ment with I O . . . tt:'lf6 ! I I .i.f3 h6!, followed by . . . tt:'le7-d5 or . . . i.d7 and . . . .l::tc8 with good play, since White does not have any active plans and d5 breaks can be answered with . . . tt:'le5, attack­ing the bishop on f3.

10 ... .i.d6! (D) One of Tal's ideas. Black deals with White's

attack on h7 by posting his queen actively on h4. In some cases, after the exchange .. .':tJxc3 the bishop supports the . . . e5 break.

Karpov popularized another plan for Black in which prophylaxis and harmony are com­bined: 1 0 . . . I:te8 ! ?. Black will forestall White's

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH tDe2 65

intentions with . . . g6, so that the bishop can re­turn immediately to f8 and subsequently go to g7 to defend the castled position.

1 1 ct:le4 It appears that White is gaining time with the

attack on the bishop, but the knight on e4 ob­structs the attacking bl -h7 diagonal and loses control of the d5-square.

The idea of I O . . . �d6! becomes clear after I I 'i'd3 'iYh4! and then after 1 2 g3 'iYh5 the queen defends h7 and d5, restricting the white attack from this comfortable square.

ll ... i.e7 12 a3 This is essential to be able to play 'iYd3 with­

out allowing ... lbb4. 12 ... 'ii'c7!? (D) An idea of Ivanchuk's that has several objec­

tives: from c7 the queen controls the f4-square, and makes way for the rook to go to d8 to put pressure on the d-pawn. In the future this pres­sure can be increased by retreating the knight from d5 to f6 and bringing the queen to b6.

It is possible to begin Karpov's plan with 1 2 . . . �e8, although in this case the black bishop will take two tempi to reach f8.

If Black is satisfied with a draw, I 2 . . . e5 !? equalizes completely, since the central pawns disappear.

13 'ii'd3 The move-order actually used in the main

game was 4 . . . c5 5 �d3 lbc6 6 lbe2 cxd4 7 exd4 dS 8 cxd5 lDxd5 9 a3 �d6 I 0 lbe4 �e7 1 1 �c2 0-0 I 2 'ii'd3 'ii'c7 I 3 0-0, but I have adjusted it for ease of reference.

13 .•. .l:.d8 14 lbgS (D) Evaluated as dubious by many analysts, but

probably White's best option. The knight puts pressure on the kingside and forces . . . g6. Then White will begin manoeuvres to penetrate on the dark squares or force other weaknesses such as the advance of the h-pawn. After 14 lb4c3 g6 15 �h6 lbxc3 1 6 bxc3 e5 ! we see another of the points of placing the rook on d8.

B

14 ... g6 15 'iYf3 After provoking the advance . . . g6, it is logi­

cal to bring the bishop to the a2-g8 diagonal: I 5 �b3 lbf6!? I 6 l:i.d I �f8 1 7 �f4 'iJle7 and here, instead of I 8 'li'e3?, which loses the bishop-pair after I 8 . . . lbd5 !, Kamsky-Karpov, FIDE World Ch ( I 4 ), Elista I 996, 1 8 �ac I should have been played, with equal chances.

15 ... �f8 1 5 . . . �f6 is more accurate, immediately hit-

ting d4.

66 CHESS EXPLAINED: THE NIMZO-INDIAN

16 lld1 .i.g7 Black completes his strategic plan of bring­

ing his bishop to g7, where it puts pressure on d4 and defends the kingside.

17 h4!? (D) In the stem game of this variation, Rogers­

Ivanchuk, Moscow Olympiad 1994, White pre­ferred 17 .i.b3, but after the thematic manoeuvre 17 . . . h6 1 8 tt'le4 tt'lce7 ! the knight overprotected d5 and solidified the kingside, after which it only remained to bring his queenside pieces into the game by . . . .i.d7 and . . . .l:Iac8, with excellent play.

B

17 .•. h6 After this move Black will be forced to open

the b1 -h7 diagonal, along which the white queen will penetrate. Another option was to complete development of the queenside immediately with 1 7 . . . b6 ! ? 1 8 h5 .i.a6 ! exerting pressure in the centre and, although White can open the h-file and penetrate on h7, his attack is not decisive.

18 tt'le4 tt'lce7 19 tt'l4c3 .i.d7? 19 . . . b6 !? is more precise. If White plays as in

the game with 20 h5 g5 2 1 tt'lxd5 tt'lxd5 22 �d3, it transpires that after 22 . . . .i.a6! 23 �h7+ �f8 it is not possible to play 24 tt'lc3? on ac­count of 24 .. .'�Jf6, trapping the queen.

20 h5 g5 21 tt'lxd5 tt'lxd5 22 �d3 .i.b5 23 �h7+ �f8 24 tt'lc3! (D)

Perhaps Black overlooked this strong move, which attacks the bishop and the knight.

24 •.• .i.c6! The c3-pawn could not be captured on ac­

count of 24 . . . tt'lxc3 25 bxc3 'iVxc3? 26 .i.d2 fol­lowed by .i.b4+.

B

25 tt'lxd5 :xd5?! 25 ... .i.xd5 ! was necessary, keeping the bishop

active and not exposing the rook. 26 a4! (D)

Preparing the way for the bishop to come powerfully into play on a3.

26 • . J:te8?! Here 26 .. .f5 !? is better, although the attack

would continue powerfully after 27 b4 ! , with the idea of b5 and .i.a3+.

27 b3! rs 28 .i.a3+ �f7 The king is in a very dangerous situation. 29 .i.d3! 'iVd7 30 .l::f.acl ! g4 If 30 . . . Itxd4, then 3 1 .i.b2 ! . 31 .i.c4 'iVd8 32 .i.xd5 'iVxdS 33 l::txc6! -qwxc6 33 . . . bxc6? loses on the spot to 34 -qwg6+. 34 d5! �d7 1-0 Black resigned without waiting for 35 dxe6+

�xe6 36 .i.b2 l:l:g8 37 llc 1 ! �d7 38 l:l:e 1 ! with unstoppable threats.

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH fDe2 6 7

Game 16 [E46]

Yuri Gonzalez - A. Alvarez Havana 2007

1 d4 CLJf6 2 c4 e6 3 4'lc3 .ltb4 4 e3 0-0 5 4'le2 (D)

This move of Rubinstein's, later popularized by Reshevsky, has the objective of bolstering the position of the knight on c3, to be able to recap­ture on c3 with the knight, keeping the pawn­structure intact. However, the knight on e2 does not have as much influence on the centre as it would on f3 and furthermore it obstructs the natural development of the king's bishop, which slows down White's development.

5 ... d5 Black occupies the centre and leaves open

the option of retreating the bishop from b4. 6 a3 .lte7 6 . . . .ltxc3+? 7 CLJxc3 gives White what he

wants, but 6 .. . .ltd6!? is interesting, because if White takes on d5 the bishop would be well placed on d6, and it is not clear whether White obtains any advantage with the expansion 7 c5 fi..e7 since with the plan of . . . b6, . . . a5 and . . . .lta6, Black exchanges his worse bishop and simpli­fies the position on the queenside, which reduces the importance of White's space advantage.

7 cxdS Maintaining the central tension with 7 CLJf4

is answered by reinforcing the centre with 7 ... c6 and . . . ctJbd7 so Black is ready to break in

the centre with . . . dxc4 followed by . . . e5, hit­ting both d4 and the knight on f4 . 1f White de­fends c4 with 8 b3, Black follows the same plan: 8 . . . ctJbd7, when 9 .ltb2?! dxc4 1 0 bxc4 e5 ! I I dxe5 ctJxe5 gave Black excellent play in Milos-Lafuente, Santos 2006. It is more logi­cal to play 9 .ltd3 !? to respond to 9 ... dxc4 1 0 bxc4 e5 with I I ctJfe2, but Black should play 9 ... e5 !? 10 dxe5 4'lxe5 I I cxd5 g5 ! ? 1 2 ctJfe2 ctJxd5 and the initiative in the centre is more important than the weakening of the kingside.

7 . . . exd5 (D) A useful alternative is 7 . . . ctJxd5 !?, seeking

greater dynamism in the centre. After 8 ctJxd5 exd5 the identical pawn-structure to the game is reached, but the exchange of knights eases Black's game. There is no advantage in 8 e4 lLlxc3 9 lLlxc3 c5 or 8 ctJg3 c5 either, while the ending that appears after 8 g3 c5 ! 9 dxc5 lLlxc3 1 0 lLlxc3 �xd I + I I �xd I .ltxc5 should be drawn. The flexible 8 .ltd2 and 8 �c2 are more common, but Black has obtained satisfactory results with 8 . . . ctJd7 followed by . . . c5 or . . . b6, . . . .ltb7 and . . . c5.

We have now reached the basic position of this variation.

This position is very fluid and both sides have tried various different development patterns.

68 CHESS EXPLAINED: THE NIMZO-INDIAN

The basic plans for White are the break with f3 and e4, the attack on the kingside with f3 and g4 and the minority attack on the queenside with b4 and b5.

Black can complete the mobilization of his queenside in various ways, but must be particu­larly attentive to White's central break with f3 and e4 - replying with . . . c5 is indicated in most cases.

8 g3 White has the following alternative schemes

available: a) Reshevsky's favourite system of 8 lLlf4,

.ltd3 and 0-0. Here Black equalizes with 8 . . . a5, followed by . . . tt::la6, . . . c6 and . . . lt:'lc7-e6 exchang­ing the white knight.

b) 8 lt:'lg3 and .ltd3. In this case the immedi­ate 8 . . . c5 ! exploits the fact that the knight on g3 does not have any influence on the centre.

c) The most topical line with 8 b4 (control­ling c5) intending lt:'lg3 and .ltd3. Michael Ad­ams has shown that Black has good play by first seizing the b 1 -h7 diagonal with 8 . . . .ltf5 !? (D).

w

Although the bishop will remain slightly ex­posed on g6, White gains no advantage. Let us examine a few examples:

c 1 ) 9 lt:'lg3 .ltg6 I 0 .lte2 .l::i.e8 1 1 0-0 .ltd6 1 2 f4 h6 1 3 f5 jt_h7 14 .ltd3 lt:'lbd7 1 5 'iff3 c6 and Black does not have any great problems, Milov­Adams, European Clubs Cup, Kerner 2007.

c2) 9 f3 h6 (9 . . . h5 !?) 1 0 tt::lg3 .ltg6 I I .ltd3 .ltd6 1 2 lLlf5 lt:'lc6 ! with the idea of . . . lt:'le7 to contest White's control of the f5-square, Aro­nian-Adams, Wijk aan Zee 2006.

8 ••• c6 It is possible to defend the d-pawn with

8 . . . lt:'lbd7 and . . . lt:'lb6, leaving open the possibil­ity of breaking with . . . c5 in one move. Further­more, from b6 the knight can jump to c4. An example: 9 .ltg2 lt:'lb6 10 0-0 l::i.e8 I I b3 i.f5!? 12 f3 c5 ! 13 g4 ..tg6 1 4 tt::lf4 cxd4 15 exd4 I:tc8 with harmonious development and pressure along the open files, Vera-Chandler, Novi Sad Olympiad 1990.

Another successful plan is to restrain the queenside with . . . a5 followed by . . . tt::la6, which practically limits both sides' play to the centre and the kingside.

9 .ltg2 (D)

B

9 .•. -tfS?! This facilitates two of White's main ideas: the

central advance e4 and the wing expansion g4. The experts prefer the scheme with . . .lt:la6

followed by . . . lt:'lc7, supporting the d-pawn, and as soon as White plays f3 Black replies with . . . c5. The move . . . a5 !? is also employed (before . . . tt::la6 or following the manoeuvre . . . tt::la6-c7) preventing any ideas of b4 altogether. In some positions, Black can even create counterplay with the advance of his queenside pawns ( . . . a5 and . . . b5-b4) to dislodge the c3-knight and find work for the queen's bishop on the a6-fl diago­nal.

A basic idea is to respond to f3 with . . . c5! without fearing the creation of an isolated pawn on d5, for White's weakness on e3 would pro­vide compensation.

10 f3

RUBINSTEIN SYSTEM: 4 e3 0-0: VARIATIONS WITH liJe2 69

Preparing both e4 and g4. 10 ... :i.e8?! 10 . . . c5 !? is indicated, since once White has

played 1 0 f3 the pressure on the d5-pawn is re­duced and e3 is weakened.

11 0-0 c5? One move later and this advance is no longer

any good, since White has castled and Black, with his previous move, has weakened the f7-square. After the provocative 1 l . . . .itd6!? 1 2 e4 dxe4 1 3 fxe4 .itxe4 1 4 :i.xf6 .itxg2 1 5 l:i.xd6 'i'xd6 16 'it>xg2 tLld7, Gunnarsson-J.Rodriguez, Leipzig Olympiad 1 960, Black has a rook and a pawn for two pieces, and the white king is somewhat exposed, but White can retain a slight advantage by means of the manoeuvre tLlgl -f3, 'i'b3, iH4 and :!.fl .

12 g4! (D)

Implementing the second plan in this posi­tion: expansion on the kingside.

12 ... jg_e6 Jf 12 . . . .itg6, White takes advantage of the

underdefended queen's pawn and the bishop's exposed position by 1 3 g5 t2Jfd7 14 f4 ! tLlb6 1 5 dxc5 jg_xc5 1 6 tLlg3 ! and the bishop on g6 is in trouble.

13 g5 t2Jfd7 Equally after 1 3 . . . tLlh5 14 f4 White gains the

advantage. 14 f4! Threatening f5 and putting pressure on d5. 14 •.• cxd4?! This move allows the knight to reach d4 with­

out loss of time. It was better to play 14 . . . t2Jb6.

15 t2Jxd4 tLlb6 16 f5 (D)

B

16 ••• ii.c8 17 g6!? The break 1 7 f6 !? gxf6 1 8 �h5 ! is even

stronger. 17 .•. t2Jc6?! It is necessary to eliminate the attacking

pawns by 1 7 . . . hxg6! 1 8 fxg6 fxg6 19 tLlxd5 ! tLlxd5 20 �b3 <;.t>h7 2 1 jg_xd5 Sg_f6! , with good survival chances.

18 gxf7+ <;.t>xf7 19 tLlxdS! A pretty combination based on Black's weak

king and lack of coordination. 19 .•. t2Jxd5 20 �h5+! <;.t>r6 Worse was 20 . . . <;.t>f8 2 1 .itxd5 �xd5 22 tLle6+

.itxe6 23 fxe6+ and the queen drops. 21 t2Je6 It is natural that there should be more than

one way to win. For example, 2 1 b4 ! tLle5 22 jg_b2 and the pin is fatal.

2I. • • .itxe6 22 fxe6+ <;.t>xe6 23 :i.d1 (D)

B

70 CHESS EXPLAINED: THE N!MZO-[NDIAN

White regains the piece without abandoning the hunt of the black king. The game is decided.

23 ••• ..td6 24 �xd5+ <i;e7 25 ..th3 ..te5 26 �c5+ ..td6 27 �d5 ..te5 28 "i¥e6+ <t;rs 29 1!xd8 l:i.axd8 30 "i¥f5+ ..tf6 31 'ifc2 ll'le5 32 ..tg2 .l:i.c8 33 "i¥xh7 l:!.cd8 34 �c2 llc8 35 �b3 lled8 36 ..td2 1-0

Conclusions

The move-order 4 . . . 0-0! against the Rubinstein seems to me to be the most flexible and practi­cal.

If White plays 5 a3 immediately (as in Game 1 2), Black does well to choose the formation with . . . c5 and after the typical development of the queenside, bringing the knight to a5 and the bishop to a6, put immediate pressure on the c4-pawn. It is very important to remember Capa­blanca's manoeuvre . . . ll'le8 ! , which not only avoids the deadly ..tg5 pin but also leaves the f­pawn free to neutralize the advance of the white pawns. Both . . .f5 and .. .f6 are giving good prac­tical results. The struggle here is quite equal.

After 5 ..td3 d5, White can once again em­ploy 6 a3 (we saw this in Game 1 3), but after 6 . . . ..txc3+ 7 bxc3 dxc4 ! Black ensures that the centre will not be completely closed and then proceeds to lay siege there with . . . c5 and in some cases . . . e5 . 1t is essential to remember that the manoeuvre . . . b6 followed by . . . ..ta6 to ex­change the light-squared bishops greatly re­lieves the black position and limits White's attacking possibilities. If White hides his bishop on a2, Black obtains good play by bringing his bishop to c4 with help of the knight on a5. In this variation Black also has sufficient resources to maintain the balance.

The line in which White, with his knight on e2, allows the creation of an isolated pawn on d4 (covered in Game 15) is no longer so popu­lar, although it continues to be dangerous. Keep

in mind Tal ' s idea of retreating his bishop to d6 to neutralize the attack with . . . �4. Remember how Black puts pressure on the isolated queen's pawn with . . . �c7 and . . . .l:!.d8 and how the block­ade can be strengthened with . . . ll'lce7. Impor­tant: do not underestimate the white attack! In this variation the white queen quickly reaches the kingside and can create serious threats with the support of the other pieces.

The apparently passive line where White takes on d5 and adopts a set-up with ..ltd3 and ll'le2 (see Game 14) needs careful handling. White's plans are very varied and Black must pay attention to White's e4 break and kingside attack. The direct transfer of the queen towards the kingside with �el -h4 can be neutralized in two ways: with . . . ..te7 or with the defensive manoeuvre . . . ll'le7-g6, both designed to force the white queen to retreat. Recently White has preferred waiting plans, in which he keeps his pawn-structure intact (without f3) and even de­lays castling, but here too Black has sufficient resources.

When White plays 5 ll'le2 (we examined this in Game 1 6), the best response is 5 . . . d5 ! and af­ter 6 a3 ..te7 7 cxd5 exd5 White has various op­tions leading to different types of positions, which are difficult to classify. As in other varia­tions, Black must wait for the correct moment to break with . . . c5. If White fianchettoes his bishop, Black has several plans available, among which I would highlight the plan of bringing the queen's knight to b6, with the idea of break­ing with . . . c5 at an appropriate moment, and the plan of developing the knight on a6, with or without the restraining move . . . a5. Black should be cautious about bringing his queen's bishop into play, so that it does not just help White's expansion in the centre or on the kingside. This line, on account of its fluid and manoeuvring character, continues to be one of White's main weapons against the Nimzo-Indian.

5 Ru binstein System: 4 e3 0-0 with �d3 and ttJf3

Introduction

In this chapter we shall study White's most natural development in the Rubinstein System. White brings his pieces to their natural squares with ..id3 and CLJf3, followed by castling, without clarify­ing the situation in the centre. A typical move-order is I d4 CLJf6 2 c4 e6 3 CLJc3 .ib4 4 e3 0-0 5 ..id3 d5 6 CLJf3 c5 7 0-0. Against this, Black can choose from various set-ups, but I have concentrated on the flexible Parma Variation (7 . . . dxc4 and . . . CLJbd7 ! ?), a favourite of Nimzo-Indian experts such as Anand and Adams. We shall also see the variation in which Black maintains his centre and plays in a natural manner with 7 . . . CLJc6!?.

The Games

In Game 17 (Kamsky-Elianov), White, using Bronstein's old idea 9 �b3, tries to obtain the bishop­pair without altering his pawn-structure, but Black's accurate play brought to light White's prob­lems in bringing the c ! -bishop into play. Seeing himself completely tied down, Kamsky sacrificed a pawn to change the character of the game and bring his bishops to life, but he failed to achieve suf­ficient counterplay and lost in the end. In this game we shall also see the positions that arise after White continues his development with natural moves such as 9 �e2 and I 0 .l:i.d I .

In Bernasek-Ragger (Game 18), White employed the most traditional move in the Rubinstein System, 9 a3, seeking an immediate clarification of the position of the black bishop. After 9 . . . cxd4 10 axb4 dxc3 I I bxc3 �c7 ! a peculiar pawn-structure arose, in which White chose the promising pawn sacrifice 1 2 .lie2 !? obtaining a lot of compensation, as is usual in this variation. Later inaccu­racies by Black allowed White to round off the game with a brilliant attack.

If Black is not satisfied with the positions that arise after 9 a3 cxd4 I 0 axb4 !? then a good alter­native is 9 . . . .ia5 !?, which was employed in Zilberman-Babula (Game 19). Black delays the cap­ture . . . cxd4 to limit White's options and waits for the right moment to capture with . . . cxd4 and ... ..ixc3, leaving White with an isolated pawn-couple on c3 and d4. In this game we shall see the typical struggle between White's bishop-pair on one side and Black's better pawn-structure on the other. After some instructive manoeuvres, Black embarked upon a risky central break which was objectively unfavourable to him, but in the complex tactical position that arose, White failed to cope with the turn of events and was decisively beaten.

In Game 20 (Korotylev-Zakhartsov), Black chooses another method of play, which consists of completing his development with 7 . . . CLJc6 without relieving the central tension. Here we shall look at the various ways for both sides to handle this position. We shall see how, after White obtains the two bishops, Black has good resources for keeping the position sufficiently closed to neutralize them.

72 CHESS EXPLAINED: THE NIMZO-INDIAN

White succeeded in opening the position with the e4 break, but in doing so he had to exchange his light-squared bishop for the annoying e4-knight and, while White controlled the dark squares, his op­ponent became the master of the light squares. A final tactical skirmish ended in a draw - a fair result.

Game 1 7 [E55]

Gata Kamsky - Pavel Elianov Montreal 200 7

1 d4 lbf6 2 c4 e6 3 lbc3 ..ltb4 4 e3 0-0 5 ..ltd3 d5 6 lbf3

The classical way of developing: the bishop goes to d3 and the knight occupies its natural square, followed by kingside castling.

6 ••• c5 7 0-0 dxc4 8 ..ltxc4 lbbd7!? (D)

w

This pattern of development is known as the Parma Variation and at the moment it is one of the most popular among the world elite. Black wants to complete his queenside development with . . . b6, . . . ..ltb7 and . . . .l:Ic8 and is ready, at the right moment, to play against an isolated pawn after a timely . . . cxd4 or against an isolated pawn-couple after . . . cxd4 and . . . ..ltxc3.

The flexible 8 ... lbbd7 !? keeps the central tension so as not to help the c 1 -bishop develop, which is a frequent strategic theme in the Rubin­stein System.

9 'tlVb3 This move of Bronstein's has experienced a

great boom in the last few years. The queen is activated and vacates the d 1 -square for the rook. It also aims to gain the bishop-pair without any great alteration to the pawn-structure.

The main alternatives are 9 'tlVe2 and 9 a3!?; the latter move will be analysed in Games 18 and 19.

With 9 iYe2 and 10 lid 1 , White mobilizes his major pieces and threatens to open the centre with d5 or dxc5. It is possible to play 9 .id3 fol­lowed by 10 'tlVe2 or vice versa, but Black con­tinues with the mobilization of his queenside ( . . . b6, . . . ..ltb7 and . . . .l:!.c8) in order to capture on d4 later and play against the isolated pawn. The exchange on d4 is postponed for as long as possible, to hinder the development of the c 1 -bishop, although White can create problems with the advance d5. Let us see some lines after 9 �e2 b6:

a) 1 0 d5 !? (giving rise to a forced variation in which White wins two pawns and in return Black develops a strong initiative on the king­side, based on pinning the f3-knight and later weakening White's castled position) 10 . . . i..xc3 (following 1 0 . . . exd5? l l lbxd5 the b4-bishop is left out of play) 1 1 dxe6 lbe5 ! 1 2 exf7 + (less ambitious is 12 bxc3?! lbxf3+ 1 3 �xf3 i..xe6 with equality) 1 2 . . . �h8 1 3 bxc3 1Lg4! 14 e4 �e7 ! (D).

w

RUBINSTEIN SYSTEM: 4 e3 0-0 WITH ii..d3 AND li:Jj3 73

Threatening . . . ltJxc4 and . . . �xe4. For the moment Black is two pawns down, but if he re­gains them, his queenside pawn-majority will grant him a slight advantage in the endgame. However, the threats against the white king are more serious once the pawns on the f-file are doubled. 1 5 .U.el (defending e4) 1 5 . . . b5 ! (de­flecting the bishop from the defence of f7) 1 6 ..txb5 ltJh5! with excellent counterplay.

b) It is more prudent to play 1 0 .l::td l cxd4 (with the rook on d l the d5 advance would be stronger) I I exd4 i.b7 (D).

Black has a satisfactory position. The stan­dard plan is to exchange the bishop for the knight on c3, making the d5 advance impossible, and play against White's isolated pawn-couple. The black queen goes to c7 and the rooks to c8, e8 or d8, while his bishop is very strong on the long diagonal. Later he can weaken White's structure even more with the advance ... e5 or blockade White's pawns using the d5- and c4-squares. When the queen is on c7 White will have to be alert to the threat of . . . ltJg4 ! .

If White begins the manoeuvre i.g5-h4-g3 to harass the queen on c7 along the h2-b8 diago­nal, this will be met by . . . ltJh5 ! when the bishop is on h4, which paralyses that plan and at the same time threatens to establish a strong knight on f4. Naturally, White also has his trumps, be­cause his greater central control gives him an advantage in space, and after Black takes with ... ..txc3 he will also have the bishop-pair and the plan of a4-a5 to break up Black's queenside structure.

9 ... cxd4! (D) This natural move had escaped the attention

of both the theoreticians and the grandmasters, who preferred 9 . . . �e7 or 9 . . . a6 1 0 a4 �e7.

w

At first sight, White will now attain the main objective of �b3, since he gains the bishop­pair with a symmetrical pawn-structure.

10 �xb4 The natural response, retaining a healthy

pawn-structure. However, the Ukrainian grand­master will show with some subtle moves that White's difficulties in bringing the c l -bishop into play give Black equal chances.

A less classical, but perhaps better, move is 10 exd4 !?. Now the most prudent course is to retreat the bishop and continue with the tradi­tional manoeuvre . . . ltJb6 followed by . . . i.d7 and . . . i.c6, regardless of the fact that White eliminates this bishop with ltJe5 and ltJxc6 -since Lasker-Capablanca, World Ch ( 1 2), Ha­vana 1 92 1 , it has been known that Black's chances in this type of structure, where a back­ward pawn on c6 confronts another isolated pawn on d4, are fully acceptable. For example: I O . . . i.e7 ! (worse is 1 0 . . . i.xc3? ! 1 1 bxc3 b6 1 2 lbg5 ! and the threat of ltJxf7 gives White a dan­gerous initiative, Topalov-Ponomariov, Vitoria Gasteiz 2007) 1 1 Itel lLlb6 1 2 i.e2 i.d7 1 3 lbe5 i.c6 ! 1 4 lLlxc6 bxc6 1 5 i.e3 ltJfd5 1 6 .U.adl ltJxe3 1 7 fxe3 l:tb8 and Black creates counterplay on the b-file and against White's central pawns, Liascovich-Vescovi, Buenos Ai­res 2005.

10 ... dxc3 11 �xc3 �c7! (D)

74 CHESS EXPLAINED: THE NIMZO-INDIAN

w

A universal move which pins the c4-bishop and generates threats against White's castled position, in combination with the emergence of the bishop on b7 and the jump of the knight to g4.

12 �b3 Naturally it is not good for White to ex­

change the queens and saddle himself with the worse pawn-structure by 1 2 .1i.e2?! �xc3 1 3 bxc3 ctJe4, while after 1 2 .1i.d2 ctJe4! Black equalizes completely.

12 •.• b6 13 .1Le2 White has retained the bishop-pair, but this

has been achieved at the expense of develop­ment. After 1 3 .1i.d2 ct:'lc5 1 4 �c2 .1i.b7 Black threatens . . .. �:Jg4 and . . . .1Le4 or . . . .1i.xf3.

13 ... .1i.b7 14 ctJd4?! Parrying the threat of ... ctJg4 and preparing

the central occupation by f3 and e4, which would leave the b7-bishop dead, but Black is going to forestall him in the centre, taking ad­vantage of the fact that White has not com­pleted his development. The modest 1 4 .1i.d2 ct:'lc5 1 5 �b4 ! ctJfe4 1 6 .U.fc I a5 1 7 �a3 ctJxd2 1 8 ctJxd2 .1i.d5 was preferable, with equality.

14 .•. ct:'lc5! 15 �a3 eS! (D) Black takes the initiative. This pawn will be­

come a spearhead which will completely re­strain White's play.

16 ctJb3 If 1 6 ctJb5, then 16 . . . �c6 1 7 f3 .1i.a6 1 8 �b4

e4! hindering e4, which leaves the c l -bishop shut in. After 16 ctJf5 l:i.fd8 1 7 f3 e4! once again White has difficulties completing his develop­ment.

16 ... ct:'lxb3 17 axb3 There will be problems developing the c ) ­

bishop after 1 7 �xb3 l:i.fd8 18 f3 e4 ! . 17 .•. l:i.fd8 18 l:i.e1 Vacating the fl -square for the bishop. If 1 8

b4, then 18 . . . �c2! would be quite awkward. 18 •.• a5! Leaving the white queen out of play and pre­

paring the exchange of the light-squared bish­ops with . . . .1i.a6. It is curious how with quite simple moves Black has managed to dominate the position, preventing the enemy bishops from becoming active. Kamsky reacts in the best way, sacrificing the e-pawn to bring his posi­tion to life.

19 e4 .1Lxe4 20 Si.gS (D)

B

20 • . . h6! Clarifying the position; White decides to give

up one of his valuable bishops in return for weakening Black's castled position a little. But

RUBINSTEIN SYSTEM: 4 e3 0-0 WITH .il.d3 AND tDj3 75

the white forces are too far away to worry the black king.

21 l:'!.acl �b7 22 .1Lxf6 Forced, since Black was attacking g2. 22 ... gxf6 23 .1Lfl l:'!.d4! (D)

Black not only has an extra pawn, but he also controls the dark squares and is preventing the white queen from getting back into play.

24 b4!? Kamsky sacrifices another pawn to activate

his queen and try to create some sort of threat against Black's weakened castled position.

24 •.• .l:'!.xb4! 25 .l:'!.c4 To activate his bishop but, two pawns down,

every exchange of pieces brings the end closer. 25 ••• l:'!.xc4 26 ..ixc4 cJ;;g7 27 h4 ..ig6 28 �g3

h5 29 l:'!.d1 White has not achieved anything for his sacri­

ficed pawns and the rest is a matter of technique. 29 ... �c7 30 ..ie2 c,i;;h7 31 �f3 .l:'!.d8 32 l:'!.e1 After 32 �xf6 .l:'!.xdl + 33 ..ixd l �c l 34 �f3

�xb2 the passed pawns decide. 32 ... l:'!.d6 33 b3 �c5 34 g3 cJ;;g7 35 ..ia6 b5

36 .1Lb7 l:'!.d3 0-1

Game 1 8 [E55]

Jan Bernasek - Markus Ragger Mitropa Cup, Szeged 2007

1 d4 CDf6 2 c4 e6 3 Cbc3 ..ib4 4 e3 0-0 5 ..id3 c5 6 ctJf3 d5 7 0-0 dxc4 8 Jl.xc4 CDbd7!? 9 a3!? (D)

This is Kramnik's favourite move in this po­sition and the one that seems to cause Black the most problems.

9 ... cxd4 The main move, though 9 . . . ..ia5 !? is also

good, as we shall see in the next game. The

capture 9 . . . ..ixc3?! is less sensible because it strengthens the white centre, while the black knight on d7 neither puts pressure on d4 nor hinders the advance a4-a5 . Then 1 0 bxc3 b6 I I ..id3 ..ib7 1 2 .l:'!.e I ! (D) is the most logical for­mation for White, who wishes to advance in the centre (e4), releasing his dark-squared bishop and generating ominous threats on the king­side.

B

76 CHESS EXPLAINED: THE NIMZO-!NDIAN

Now: a) I2 ... tt:'le4 has also been played but after

I 3 i.b2, followed by c4, White is a little better since he can aim for a4-a5 or simply improve his pieces with !lac I and exchange the black knight on e4 with tt:'ld2.

b) If I2 . . . e5 White plays 1 3 e4 ! and devel­ops his bishop to g5 with a slight advantage, be­cause with the disappearance of Black's central pawns the bishop-pair and central superiority give White the advantage.

c) I2 . . . i.e4, opposing White's planned ad­vance and offering the exchange of bishops, gives White a choice between I3 i.fl ! ? , retain­ing the bishops and preparing tt:'ld2 and f3, and I 3 ..i.xe4!? tt:'lxe4 I4 �d3 ! , followed by I 5 c4 and I 6 i.b2, and while White has prospects in­volving d5 or opening the queenside with a4-a5, Black is rather passive and lacks a plan.

10 axb4 After I 0 exd4?! i.xc3 !? I I bxc3 'W/c7 ! Black

reaches, with an extra tempo, the hanging pawns position of Game 1 9, and continues with the typical plan ( . . . b6, . . . i.b7, . . . l!ac8, etc.).

10 • . • dxc3 11 bxc3 (D)

B

White has obtained the bishop-pair with a relatively healthy structure (without hanging pawns in the centre or an isolated pawn), and if can complete his development untroubled he will have the advantage.

l l ... �c7! Essential. Black must exploit the momentary

lack of coordination among the white forces and exert pressure on the bishop and the c-pawn. If

he manages to prevent the c4 advance and place a minor piece on that square he will not have any problems. I I . . . b6 is weaker, since both 1 2 �d6!? and I 2 ..i.e2, followed by c4 and .ib2, give White the advantage.

12 ..i.e2!? This powerful pawn sacrifice, an idea of GM

Guillermo Garcia, is the most energetic. White organizes his pieces in the best way and for this he must make this small investment of material.

Instead, I2 �e2 tt:'lb6 ! 1 3 i.d3 e5, followed by l 4 . . . i.e6 and . . . i.c4, gives White nothing, but 1 2 'i¥b3 !? is very interesting and has been debated at the highest level; e.g ., I2 .. .<'2Jb6! 13 i.e2 e5 ! I4 l:.a5 ( 1 4 c4 fails to l4 . . . ..i.e6 15 .ib2 tt:'lxc4 I6 .l:r.fc l tt:'lxb2! with equality) I4 . . . .ie6 1 5 �c2 tt:'lbd7 1 6 c4 (D).

B

l 6 . . . b6 ! 1 7 :a6 (the most incisive; if l7 l'r.a3, then I7 . . . M.fc8 I 8 tt:'ld2 b5 ! 1 9 c5 a5 ! with good counterplay) I7 .. . 1:rfc8 I 8 '2ld2 and now the sim­plest is I 8 . . . i.d5 ! followed by . . . i.b7 and . . . a5 ! (Rogozenko-I.Farago, Hamburg 2002); Black conquers the c5-square for the knight and cre­ates a passed pawn on a5.

12 •.• tt:'ld5!? (D) The other way to take the pawn, I 2 .. ."iVxc3,

has fallen into disuse because White obtains fan­tastic compensation after I 3 i.a3, fol lowed by �bi , l!fc l and i.d3, as in Kramnik-Kasparov, World Ch ( I 2), London 2000. I3 l:tbi , followed by i.b2, has also yielded excellent results; e.g., 1 3 . . . �c7 I4 i.b2 a6 I 5 l:tc l �b6 (Potkin­Alexandrov, Abu Dhabi 2005) I 6 �a4 ! gives White strong pressure.

RUBINSTEIN SYSTEM: 4 e3 0-0 WITH i..d3 AND Ci'Jj3 77

If Black does not capture on c3, White sim­ply continues with his plan (c4 and .1l..b2) and enjoys a small advantage without risk.

13 .1i..b2 M.Gurevich preferred 1 3 c4 !? lL!c3 (certainly

not 1 3 . . . 4:Jxb4? 1 4 .ia3 !) 1 4 'iVc2 lDxe2+ 1 5 'lixe2. Although Black has eliminated one of the enemy bishops and can easily complete his development with 1 5 ... b6 1 6 .1l..b2 .ib7 17 �fd I �fd8, he must be careful, since after 1 8 lDh4! White threatens the lonely black king. How­ever, according to Gurevich, Black holds the balance with 1 8 . . . lDf8 19 b5 a6! and the queen­side pawns are exchanged.

Another, rather strange, way of sacrificing the c-pawn is 1 3 ..ll.d3 !? 'iVxc3 14 �a3 'iVf6 (14 . . . 'ihb4!?) 1 5 'iVc2 h6 16 b5 lDb4 17 .11..h7+ 'tt>h8 1 8 'it'bl aS ! with a position that is hard to assess, Aronian-Topalov, Morelia/Linares 2006.

13 .•. lL!xc3 (D)

14 'iVd2!? A new approach to the position. White re­

tains the strong dark-squared bishop, leaving his opponent far behind in development. These factors, combined with the pressure exerted by the a) -rook, give sufficient compensation for the pawn.

14 .11..xc3 'iVxc3 15 'it'd6 has been played on various occasions but the game is greatly sim­plified after 1 5 . . . 'iVc6 !.

14 • . . 4:Jxe2+ 15 'iVxe2 lL!b6 Removing this defensive knight from the

kingside proves fatal. It seems more natural to continue 1 5 . . . b6 !? , although then White can maintain the pressure with 1 6 l!ac I ! ? or 1 6 �fdl !?.

16 �fcl 'it'd8 17 �d1 'it'e7 After 1 7 . . . .id7 1 8 lDe5 or 1 8 .1i..d4 !? White's

pressure is intense. 18 ..ll.d4! 'iVxb4 19 l:Idb1 (D)

Black has won a second pawn, but he is walking a tightrope because White has threats on both wings.

19 •.• 'it'c4 20 'iVb2 Maintaining the threat of .ixb6 and attack­

ing g7 . 20 •.• lDa4 21 'iVd2 Stronger was 2 1 'iVa3 ! b5 22 �b4 'iVc6? 23

.ixg7 ! with advantage. 21...'iVc6?! It was essential to give the knight immediate

support with 2 1 . . . b5! although White has com­pensation after 22 e4 ! .

22 'iVa5!

78 CHESS EXPLAINED: THE NIMZO-IND/AN

The queen attacks a4 and generates hidden threats on the kingside.

22 ••• b5 23 lixb5 i.d7? This allows a beautiful finish. It was essential

to block the action of the b2-bishop with the in­genious 23 .. .f6! 24 �xa4 i.d7 25 l:i.b4 'iVxa4 26 libxa4 i.xa4 27 lixa4 lifc8 and White has the advantage in the endgame but Black is not with­out counterplay, based on White's momentary passivity and the presence of the passed pawn.

24 :g5 f6 (D) 25 .l:.xg7+! ! �h8 Instead, 25 . . . �xg7 1oses to 26 'iVg5+ �h8 (if

26 ... �f7, then 27 i.xf6! and there is no defence against lt:'le5+ or 'iVg7+) 27 i.xf6+ .l:!xf6 28 'iVxf6+ �g8 29 lL'le5 !.

26 �g5! 'i¥d5 There was no salvation in 26 . . . e5 27 lt:'lxe5 !

fxg5 28 lt:'lxc6, nor in 26 . . . fxg5 27 .l:tg6+ e5 28

.1xe5+. In both cases Black faces catastrophic loss of material.

27 .Mg8+! 1-0 And mate in two. A perfectly executed at­

tack.

Game 1 9 [E55]

Vaacov Zilberman - Vlastimil Babula Pardubice 2007

1 d4 lL'lf6 2 c4 e6 3 lL'lc3 i.b4 4 e3 0-0 5 i.d3 d5 6 'Llf3 c5 7 0-0 dxc4 8 i.xc4 lt:'lbd7!? 9 a3 i.a5!? (D)

This retreat, retaining the bishop for the time being, is Adams's favourite. Obviously I 0 dxc5? is not good for White because his structure would be seriously damaged after I O . . . i.xc3 ! 1 1 bxc3 'Llxc5 .

9 . . . i.a5 ! ? and also the previous 8 . . . ctJbd7 are flexible moves, designed to limit White's op­tions. In general, Black wants to play . . . cxd4 and then . . . i.xc3, to leave White with hanging pawns, but delaying these captures deprives White of the characteristic plans of that struc­ture, such as immediately bringing the bishop out to g5, followed by .l::!.c I ; neither does it make much sense to play 10 l:i.el while the e­file remains closed. Furthermore, the move­order 9 . . . .1a5 ! ? followed by . . . cxd4 avoids the variation 9 . . . cxd4 1 0 axb4! ? seen in the previ­ous game.

10 'iVe2 The most natural continuation. An alternative is I 0 .ii.d2, but then Black con­

tinues with the idea of giving White an isolated pawn-couple by I O . . . cxd4 I I exd4 (White gains no advantage with I I lt:'lxd4 lt:'le5 12 .ii.e2 i.d7 because his bishop is passive on d2 and Black has easy development) l l . . . .ii.xc3 !? 1 2 bxc3 (if he plays 1 2 .1xc3 the blockade of the isolated pawn with 1 2 . . . lt:'lb6, . . . lt:'lbd5 and . . . b6 !? or

RUBINSTEIN SYSTEM: 4 e3 0-0 WITH il..d3 AND ti:JJ3 79

12 .. . b6 !? followed by . . . �b7 ensures pleasant play for Black) 1 2 ... b6 1 3 �e2 �b7 14 �d3 �c7 15 l:tfe I and now with 1 5 . . . l:tac8 or 1 5 . . . l:tfe8 we reach the type of position that we have dis­cussed, where the play is balanced.

White can try to leave the bishop out of play on a5 with I 0 CLJb5, but after I O . . . cxd4 I I �xd4 (if I I CLJbxd4 Black takes the centre with l l . . . e5 ! 12 tt'lb5 e4 1 3 CLJd2 CL'Je5) I l . . .�b6 12 �h4 a6! 13 tt'lbd4 l:te8 14 b4 CLJf8 (threatening . . . e5) 1 5 tileS CLJ6d7 ! and the ensuing exchanges equal­ize the struggle, Wells-Adams, London 2007.

10 ... cxd4! Now that White cannot capture on b4 as in

the previous game, Black takes on d4 and c3, reaching the pawn-structure with an isolated pawn-couple on c3 and d4.

Note that the tempo lost with . . . �a5 has only helped White to place a pawn on a3, which is no help to him and could even be a weakness, since it is easier for Black to attack and prevents the possible development of the queen's bishop to a3.

11 exd4 �xc3 12 bxc3 �c7! (D)

This position can be reached by different move-orders. The most usual pattern of devel­opment is . . . b6, . . . �b7, . . . l:!.ac8 and . . . l:!.fe8, while White must bring his bishop to d3, his rooks to the centre and the queen's bishop to g5 or b2. Generally speaking, chances are equal.

13 i.d2 White defends the c3-pawn before retreat­

ing the bishop from c4. It is possible to post­pone the development of the queen' s bishop and

play 1 3 �d3 immediately, sacrificing the pawn, but Black can decline the offer and continue with his development plan ( 1 3 . . . b6 followed by ... �b7 and . . . l:tac8) because the capture of the pawn would give White many tempi.

White can also defend c3 with 13 �b2 and then play �d3 and c4, although in that case White loses control of the f4-square, which Black could use for his queen, threatening . . . �xf3, or a knight.

l3 ... b6 14 l:tfe1 i.b7 15 i.d3 (D)

B

15 •.• l:tac8 The eternal problem of where to place the

rooks. As I understand it, this is the correct square for this rook and the king's rook should go to e8. However, Carlsen preferred to place them differently: 1 5 . . . l:tfe8 16 c4 l:!.ad8 1 7 �g5 CL'Jf8 ! ? - an interesting idea; Black puts imme­diate pressure on d4 and threatens to bring the knight to f4 via g6. He will get doubled pawns on the kingside, but in return the bishop-pair disappears and White loses control of the f4-square: 1 8 i.xf6 gxf6 1 9 i.e4? ! (a sacrifice without foundation, since Black's kingside is well defended; 1 9 .l:Iad 1 ! ? <tlg7 !? was prefera­ble, with complex play) 1 9 . . . �xe4 20 �xe4 �xc4 and White has little compensation for the pawn, Yusupov-Carlsen, Amsterdam 2006.

16 c4 White must stay on the alert, because one of

Black's plans is to set up a blockade of the iso­lated pawn-couple using the light squares d5 and c4. For example, if I 6 lDe5, then 1 6 . . . CL'Jxe5 1 7 dxe5 CLJd7 1 8 l:!.ad I �d5 ! .

80 CHESS EXPLAINED: THE NIMZO-INDIAN

16 ••• :fd8 Inaccurate; this rook should go to e8, where

it can support a possible . . . e5 advance, after . . . SLxf3. Furthermore, from e8 the rook pre­vents the advance d5 .

17 .l:tacl "iVd6 One of Black's resources in this variation.

From d6, the queen attacks the a3- and d4-pawns, creating the threat of . . . SLxf3 . Black could also have made use of the rook's posi­tion on d8 by 17 . . . lt:Jf8 !?, attacking the queen's pawn.

18 SLb4 "iVf4! (D)

w

One of the key squares in Black's strategy. The threat is once again . . . SLxf3.

19 "iVe3 lt:Jh5 A recurrent idea in this variation. From h5

the knight comes into play on f4. Another ap­proach is 19 . . . "iYxe3 20 fxe3 (20 l1xe3 is met by 20 . . . lt:Jd5 !) 20 . . . lt:Jc5 2 1 SLfl lt:Jce4 with ap­proximate equality.

20 SLe7! White would not have had this possibility if

Black had placed the rook on e8. 20 . • Jle8 21 SLg5 The bishop returns to cover the f4-square

and force Black's queen to make a decision. 21. • .'ii'xe3 Here 2 1 . . .'iVd6! ? deserved consideration. 22 fxe3 White brings a pawn to the support of his

centre. Less ambitious was 22 SLxe3 lt:Jhf6 with balanced play.

22 ••• h6 23 iLh4 g6!?

An interesting idea, but positionally risky. Black wants to break on e5 and so prevents SLf5 . The move has the drawback of substan­tially weakening the dark squares on the king­side, although it is unlikely that White could launch a mating attack against the black king without the queens on the board. Nevertheless, weakening the dark squares, when you do not have a dark-squared bishop, can be important even in the endgame.

24 lt:Jd2 It is very logical to bring the knight to d6. 24 •.• e5!? Continuing with his plan. Now the threat is

25 . . . exd4 26 exd4 SLxg2 followed by . . . lt:Jf4+, winning a pawn.

25 lt:Je4!? Entering complications. A more positional

approach is 25 d5 ! ?, restricting the action of the b7-bishop.

25 ••• exd4!? (D) Both players are prepared for the tactical

battle. In view of what follows, it can be stated that the prudent 25 . . . SLxe4 was objectively the correct move. After 26 SLxe4 lt:Jhf6 27 .id3 �g7 28 a4 the two bishops give White a slight advantage, but the black position remains quite solid.

26 lt:Jd6! lt:Jc5! 27 SLb1? This was not the moment for slackening the

pace. It was essential to play 27 lt:Jxc8 ! lt:Jxd3 ! (now 27 . . . SLxc8? is impossible due to 28 exd4! ; 27 . . . .l:!xc8 is a little better but after 28 .i:tcd I dxe3 29 .l:txe3 lt:Jxd3 30 :dxd3 .l:txc4 3 1 .i.e I !

RUBINSTEIN SYSTEM: 4 e3 0-0 WITH j_dJ AND ti'Jj3 81

the white rooks would go on the offensive) 28 4Jd6 .l:!,b8 29 exd4 CLJxc I 30 .l:!,xc I g5 ! 3 1 .ie 1 4Jf4 32 .l:!,c2 .i.c6 and White has the better end­ing.

27 ... d3! On d3, the pawn restricts the activity of the

white pieces. 28 .l:!,cdl? It was better to play 28 .l:!,ed I ! to vacate the

el -square so that the bishop could reach b4, at­tacking the knight.

28 ... g5 29 CLlxc8 .l:!,xc8 30 .i.f2 ..ie4 (D)

The position is a nightmare for White. The bishops and the rooks can hardly move.

3I .l:!,fl Vacating the e ! -square for the bishop. 3l...CLJg7 The only inactive piece starts to move. 32 .i.el lLJfS 33 �f2?! It would be sad to play 33 .i.d2 h5, when

White could still hardly move, but the move in the game allows another little tactic.

33 ..• .i.xg2! 34 .l:!,gl .i.e4 35 .i.b4 White has lost another pawn but at least he

has activated his bishop. 35 ... .l:!,d8 36 .i.xc5 bxc5 37 .l:!,d2?! White could have simplified the position to

reach a pawn-down rook ending by playing 37 .l:!,g4!? CLJd6 3 8 ..ixd3 ..ixd3 39 .l:!,xd3 CLle4+ ! 40 .l:!,xe4 .l:!,xd3 4 1 h4 ! with some chances of salva­tion.

37 ••• CLJh4 38 .l:!,gdl CLJf3? (D) In strategically won positions, a tactical slip

can spoil all the previous work. Black could

have maintained a very advantageous position by 38 . . . .l:!,d6! 39 ..ixd3 .l:!,f6+ 40 �g3 .l:!,f3+ 41 �g4 ..ib7 ! , with serious threats against the white king.

39 .l:!,b2? Overlooking the shot 39 .l:!,xd3 ! .ixd3 40

.ixd3 , not only attacking the knight but also threatening .i.h7+, winning the exchange. The best then would be 40 . . . CLlxh2 4 1 .i.h7+ �xh7 42 .l:!,xd8 CLJg4+ 43 �e2 and Black should lose, because the a7-pawn drops and it is impossible to stop the passed a-pawn supported by the rook.

39 ... CLJe5 Attacking c4 and threatening . . . CLlg4+ and

. . . CLJxe3 winning a third pawn for the exchange. From now on everything is clear: the knight continues to capture pawns while attacking the white pieces.

40 ..ia2 CLlg4+ 41 �el CLJxe3 (D)

82 CHESS EXPLAINED: THE NIMZO-INDIAN

42 l:tcl White could not manage to exchange the

black rook with 42 k!dbl 'Llc2+ 43 �d2 'Llxa3, when 44 l:tb8 fails to 44 . . . 'Llxbl +. However, passive defence is also hopeless.

42 .•. 'Llf5 43 �d2 'Lld4 44 �bl fS

The advance of the black pawns decides the game.

45 l:tfl 'Llf3+ 46 �c3 �g7 47 I::tdl d2 48 �c2 �g6 49 a4 g4 50 aS hS 51 I::tbS h4 52 .l:!.xcS g3 53 hxg3 hxg3 54 �xe4 fxe4 55 .!':!.c6+ �fS 56 k!c7 g2 0-1

Game 20 [E58]

Alexei Korotylev - Viacheslav Zakhartsov Russian Ch, Krasnoiarsk 200 7

I d4 'Llf6 2 c4 e6 3 'Llc3 �b4 4 e3 0-0 5 �d3 dS 6 'Llf3 cS 7 0-0 'Llc6

This developing move leads to one of the most important positions in the Rubinstein Sys­tem. Both sides fight for the centre with the most natural moves, and in contrast to the pre­vious games, Black does not try to ease the cen­tral tension by exchanging pawns.

8 a3 (D)

B

This move is most in tune with White's in­tentions: to gain the bishop-pair.

8 ••• �xc3 In the aftermath of Spassky-Fischer, World

Ch ( I ), Reykjavik 1 972, the move 8 . . . �a5 gained a certain popularity, but today its status has declined in view of the variation 9 cxd5 exd5 I 0 dxc5, which forces Black to play I O . . . �xc3 I I bxc3. Then Black generally completes his development and creates play on the light squares and the kingside with l l . . . �g4, but af­ter I 2 c4 'Lle5 1 3 cxd5 �xf3 (if I 3 . . . tt'lxf3+? !,

then I 4 gxf3 �h3 15 e4 with a lot of compen­sation for the exchange in the shape of his bishop-pair and strong central pawns) 14 gxf3 'iVxd5 I 5 �e2 'iVxc5 16 i.b2, practice has shown that White's bishops are worth more than Black's better pawn-structure.

9 bxc3 'iVc7!? (D) If Black instead first takes on c4 by 9 . . . dxc4

10 �xc4 and then plays I O . . . 'iVc7, we reach a position very similar to that of Game 1 3, after the 8th move, but with one fundamental differ­ence: in that position Black had his knight on

b8, and thus he had available the excellent se­quence . . . b6 and . . . �a6, bringing about an ex­change of light-squared bishops or else gaining command of the active a6-fl diagonal.

At any rate, Black can play 9 . . . dxc4 I 0 i.xc4 'iVc7 and continue with the plan of putting pres­sure on d4 by means of . . . .l:!.d8 and . . . e5, which has been played on countless occasions with good results. The advantage of the move-order 9 . . . 'iVc7 is that it reduces White's possibilities.

RUBINSTEIN SYSTEM: 4 e3 0-0 WITH i.d3 AND ctJf3 83

10 cxdS White needs to open the centre to make way

for his bishops, although with this exchange he also frees the enemy bishop.

There are three alternatives. If 10 iLb2, then 10 . . . tt::la5 ! forces White to capture on d5, after which a fierce battle will rage for control of the light squares and around the advance e4. Black will try to prevent this and at the same time will manoeuvre with the aim of exchanging the light-squared bishops; for example, 1 1 cxd5 exd5 12 tt::le5 c4 1 3 iLc2 tt::le4! and the knight goes to d6 to support a future exchange of bish­ops on f5, while allowing the white knight on e5 to be dislodged with . . . f6.

Another line is 10 'ife2?!, but then Black takes advantage of the position of the queen to play I O ... dxc4 1 1 ..ixc4 e5 !, with threats of ... .\1Lg4 or ... e4, followed by ... tt::lg4, with excel­lent counterplay.

10 h3 !? is very interesting, because after 10 . . . dxc4 1 1 iLxc4 e5, the pawn move is quite useful. However, with IO . . . tt::le7 !? Black reaches an acceptable position, because in the event that White takes on d5, the knight on e7 will support .. .il..f5, exchanging the light-squared bishops.

10 ... exd5 (D)

Some readers will be wondering: isn't this the type of Botvinnik-Capablanca position (see page 56) that Black is supposed to avoid? De­spite the many similarities, there is one vital difference in Black's favour: the position of the knight on f3. With the knight on e2, White had a very clear plan of central expansion with f3

and e4 to unleash his bishops, while here this advance requires a lot of preparation by bring­ing the knight to d2 or h4, followed by f3.

11 a4 Seeking a diagonal for the c ! -bishop. White

would like to transfer the f3-knight to f5, but af­ter 1 1 tt::lh4, l l . . .tt::le7 ! prevents both the jump to f5 and the move f3, which would be answered with . . . g5.

Another plan for activating the c ! -bishop is to eliminate the central pawns with 1 1 dxc5 fol­lowed by c4 at a suitable moment, but after l l . ..i..g4 12 i..e2 tt::le4 1 3 i..b2 my recommen­dation is 1 3 . . . 1Ife8 !?, to have the knight pro­tected, and then 1 4 'ifxd5 is not sensible in view of 14 . . . l:Iad8 with excellent play; if White pre­pares c4 with 14 'ifc2, Black maintains the blockade with 14 . . . tt::la5 !?.

l l ... l:Ie8 This is rather more flexible than the immedi­

ate l l . . .c4 1 2 ..ic2 l:Ie8, which can give White extra possibilities such as the immediate 1 3 tt::ld2.

12 iLa3 c4 The capture on d4 is ruled out, since this

would leave White with a strong centre and a clear plan of occupying the c- and b-files. Black has to become strong on the light squares and keep the game closed, so that the enemy bish­ops, especially the one on a3, cannot increase their activity.

13 iLc2 (D)

B

This very interesting position has arisen in many master-level games. In compensation for

84 CHESS EXPLAINED: THE N!MZO-INDIAN

the bishop-pair, Black has more space and a de­gree of control over the light squares, as well as the queenside pawn-majority which could be valuable in the endgame.

The whole strategy of the position revolves round the e4-square. White tries to mobilize his central majority, which is looking rather sad right now, while Black opposes this by exerting pressure along the e-file and creating threats on the kingside with the pin . . . i.g4, or the transfer of a rook along the third rank. Ideally, White would like to break on e4 with the support of a pawn on f3, because in that way he would not be giving Black any squares, but this is made difficult by the presence of the knight on f3.

13 . . .'�a5!? (D) GM Tisdall's move, improving the position

of the queen by attacking c3, seems more pre­cise than the immediate l 3 . . .<�'\e4, because the inclusion of the moves l 3 . . . �a5 and 14 �el or 14 �c l favours Black.

l 3 . . . .ig4 has the idea of bringing the bishop to g6 ( . . . i.h5-g6), but after the accurate move 14 �el !, with the idea of l4 . . . i.h5 15 lLih4 ! followed by f3, White will end up breaking on e4 and exchanging his clumsy knight for Black's . bishop, while if l4 . . . i.xf3 15 gxf3 �d7 16 'it>g2, careful play by White, first making his king completely safe, will allow him in the long term to exert pressure along the g-file and break in the centre supported by the two bishops.

w

14 �cl White must defend c3 with the queen and

each of the two possible squares (el and c l ) are

worse positions than its present square dl . Af­ter 14 �e I lbe4 I 5 i.xe4 l:txe4 it is difficult to achieve the e4 break.

If, instead of capturing on e4, the bishop de­fends with 1 5 .ib2, Black does well to prevent White's expansion (lLih4 ! and f3) by means of the precise l5 . . . iVd8 ! , followed by . . . .if5 with good play.

14 ••• ltJe4 l4 . . . .ig4 is possible, but the text-move is

more forcing. 15 .ixe4 .l:Ixe4 16 lLid2 .U.e8 17 e4! .ie6 (D)

This is probably the critical position of the variation. Black maintains the blockade with­out allowing the white pieces to become active on the dark squares.

18 iVc2 Defending the a-pawn. Korotylev himself, in

a previous game against Grishchuk in Moscow 2004, sacrificed the pawn and obtained com­pensation after 1 8 e5 iVxa4 19 f4 ! ? .if5 20 .l:tf2! followed by lLifl-e3. However, if the queen heads for home immediately (20 . . . �a5 ! intend­ing . . . iVd8) the chances are equal.

18 •.• dxe4 19 ltJxe4 .if5 20 f3 �d5 One of the advantages of l 3 . . . �a5: the queen

can transfer easily to this comfortable square, after which the game is equalized, since all the black pieces are active and while White domi­nates the dark squares, his opponent is strong on the opposite colour.

21 I:tfe1 ltJa5 2 l .. . .l:i.e6!? and . . . !Iae8 is an alternative worth

considering.

RUBINSTEIN SYSTEM: 4 e3 0-0 WITH Ji.d3 AND 0Jj3 85

22 l:!.e3 l:!.e6 23 liael l:taeS 24 g4 i.g6 25 �f2 i.xe4 26 fxe4 tlxe4 (D)

It looks as if White is completely lost, but he manages to save himself by combining the pin and the weakness of Black's back rank.

27 "iYf3! .l:txe3 2S �xe3 �d7 28 . . . �d8 29 "iYd5 ! h6! is similar; White must

capture the piece. 29 �f5! �xa4 30 �xeS+ �xeS 31 �xa5 The draw is inevitable. 31..."�el + 32 �g2 �e2+ 33 �g3 'ife3+ 34

<;t>g2 lf2_lf2

Conclusions

The Parma Variation (7 . . . dxc4 8 i.xc4 ltJbd7) is a flexible system which restricts White's de­velopment options and leads to different posi­tions, depending on how White reacts.

If White develops normally with 'ife2 and �d I (examined in the notes to Game I 7), Black should capture with ... cxd4 and . . . i.xc3, creat­ing an isolated pawn-couple in the enemy camp. Remember that the best formation for Black in that case is to place the queen on c7, the bishop on b7 and the rooks in the centre. Black's plans include the use of the f4-square for the queen or the knight, as well as the blockade of the white pawns on the light squares c4 and d5. Also keep

in mind that at any moment Black can play . . . ltJg4, creating the annoying threat of . . . i.xf3.

Against Bronstein's idea 9 li'b3 (also Game I 7), Black should capture immediately with 9 . . . cxd4 and if White recaptures with the pawn, Black should retreat the bishop to e7 and subse­quently carry out the manoeuvre . . . ltJb6 and . . . i.d7-c6, establishing an iron grip on the d5-square. Remember that if 1 0 �xb4 dxc3 I I �xc3 Black obtains very good play with the subtle I l .. .�c7 !, since this takes advantage of White's difficulties in bringing his queen's bishop into play. This is the fundamental theme of Black's strategy.

The move 9 a3 can be answered by Black in two ways: 9 . . . cxd4 (Game 1 8) or 9 . . . i.a5 !? (Game I 9). The first brings about a strange po­sition in which White has the dangerous pawn sacrifice 1 2 i.e2 ! ?, which promises a lasting initiative for the bishop-pair and pressure along the open files, although Black has held out in the majority of cases. In the line with I 2 �b3 Black has equalized with the plan of ... ltJb6, followed by . . . e5 and . . . i.e6, trying to gain con­trol of the c4-square. Both positions are worth investigating.

9 . . . i.a5 ! ? is more prudent, generally leading to the positions with an isolated pawn-couple that we mentioned above, which are completely satisfactory for Black.

The line with 7 . . . 4Jc6 (Game 20), maintain­ing the centre with 9 . . . �c7, is perfectly play­able. After the exchange I 0 cxd5 exd5, and the necessary advance . . . c4, remember that the main thing is to prevent White from carrying out the advance e4 under favourable conditions and to be alert to the possible manoeuvres for activating the f3-knight. A well-timed . . ."�a5 !? can force the white queen to occupy a less fa­vourable position (c l or e i ) and prepares the transfer of the queen to the d5-square as soon as White breaks with e4. Keep in mind that White will try to support this break with moves such as ltJh4 or ltJd2, followed by f3.

6 Fianchetto Variation: 4 g3 or 4 ttJf3 c5 5 g3

Introduction

This variation, which gained great popularity through Romanishin's successes with it in the 1970s, is very different from the other ways of playing against the Nimzo-Indian Defence. The position of the bishop on g2 changes White's plans substantially and therefore also changes Black's plans.

Usually White plays for pressure on the long diagonal, making it very difficult for Black to de­velop his queenside, especially the c8-bishop. In many variations, White sacrifices a pawn (gener­ally the c-pawn) or ends up with an inferior pawn-structure but gains compensation in the form of the pressure that the white pieces exert on the centre and especially on the queenside. In contrast to other variations, where his light-squared bishop is on the bl -h7 diagonal, it is very unlikely that White will obtain any attack on the kingside.

There are various ways to play against this system, although I shall focus on the most topical and effective ones, in which Black tries to derive maximum benefit from the position of the white bishop on g2 by immediately attacking the c4-pawn or by permanently destroying White's pawn­structure on the queenside.

White has two move-orders leading to the Fianchetto: I d4 lL'lf6 2 c4 e6 3 tt:'lc3 �b4 4 g3 !? and 1 d4 lL'lf6 2 c4 e6 3 tt:'lc3 ..ib4 4 tt:'lf3 c5 !? 5 g3 !?.

If Black answers 4 tt:'lf3 with 4 . . . b6 !?, this transposes to the Queen's Indian Defence, given ex­cellent coverage in Chapter 6 of the book Chess Explained: The Queen 's Indian by Peter Wells.

The Games

In Game 21 (Vemay-Caruana) we shall see a very effective system which can be employed when White plays the immediate 4 g3 and which consists of the immediate 4 . . . d5 and the subsequent cap­ture on c4, taking advantage of the bishop's position on g2. This variation has much in common with the Catalan Opening, but the position of the white knight on c3 and the black bishop on b4 works in Black's favour. On reaching the critical position (after 7 . . . tt:'lc6) White has various ways to continue in addition to 8 a3 as played in the game, but there are two basic ideas: • occupy the centre with e4 and complete development, placing the rooks in the centre (d l , e l ) ; or • regain the pawn in some way.

For his part, Black should always consider defending his pawn by . . . .l:i.b8 and . . . b5, as well as the manoeuvres ... tt:'ld5-b4-d3 or ... tt:'la5-b3 if White plays a3 at some point. In this game the young Ital­ian/ American GM presents one of the strongest plans against the dangerous 8 a3, which consists of returning the bishop to e7, then the accurate manoeuvre . . . tt:'la5-b3, followed by the immediate break . . . c5 ! .

As White often plays the Fianchetto Variation beginning with the move-order I d4 lL'lf6 2 c4 e6 3 tt:'lc3 ..ib4 4 lL'lf3 c5 5 g3 !? it is important to know lines that include the move . . . c5 !?. In Game 22

FIANCHEITO VARIATION: 4 g3 OR 4 CDf3 c5 5 g3 87

(Wojtaszek-Akopian) we shall see a very fashionable system in which Black completely destroys White's queenside pawn-structure but in return has to withstand the early pressure of the bishop­pair and in many cases the presence of a knight on d6. But with patient and accurate play, Black can neutralize White ' s temporary advantages while retaining his own structural advantage. Akopian did precisely this.

In the final game of the chapter we shall see the most classical system for combating the fian­chetto, which has stood the test of time. After . . . c5 and . . . cxd4 Black simply breaks in the centre with . . . d5 ; White can capture on d5, followed either by 9 �d2 or the most topical move 9 �b3. White can also castle, sacrificing the c4-pawn, as is usual in this variation. In Game 23 (Wang Yue-Yakovenko) White played 8 cxd5 and 9 �b3 and after a more or less forced series of moves, he managed to keep the bishop-pair and the same pawn-structure as his opponent. However, Black showed that the bishop-pair can be neutralized, by taking advantage of his slight lead in develop­ment, which allows him to put pressure on key points in the white position.

Game 2 1 [E2Q]

Clovis Vernay - Fabiano Caruana Budapest 2007

1 d4 CLJf6 2 c4 e6 3 tLlc3 �b4 4 g3 Many players of the Fianchetto Variation use

the move-order 4 CLJf3 c5 5 g3 because the line that we shall see in this game, where Black plays . . . d5 and takes on c4, is considered quite risky for White, who has to sacrifice a pawn. However, thanks to the efforts of several grand­masters (the pioneer Romanishin, then Gulko and at present Carlsen and Moiseenko ), matters are not so clear and Black must play with great care. Practice has shown that White's initiative and firm grip on the centre compensates for the small material deficit. In reality, White obtains an unclear position with opportunities for cre­ative play.

4 • . . d5 (D) The most natural continuation against the

Fianchetto Variation. Black attacks the c4-pawn, which can no longer be protected by the fl­bishop which, to be consistent with 4 g3, will emerge on g2.

5 �g2 0-0 A necessary preparation for capturing on c4,

since 5 . . . dxc4?? loses to 6 �a4+ CLJc6 7 �xc6+ bxc6 8 �xb4, when White wins a piece.

6 CLJf3 There are not many valid alternatives to the

sacrifice of the pawn. Taking on d5 gives Black

w

very comfortable play, since this would solve the problem of the development of his queen' s bishop and give him a half-open file in the cen­tre. His typical development would be to bolster the centre with . . . c6 and play .. . lle8, . . . CLJbd7 (then . . . CLJb6 if White plays b4, or . . . CLJf8-g6) and retreat the bishop from b4 to d6, e7 or f8. In this pawn-structure, typical of the Exchange Variation of the Queen' s Gambit, White needs his light-squared bishop to be on d3, so that it can support the thematic advance b4-b5 and controls the e4- and c4-squares; furthermore, from d3 it can create threats against Black' s kingside. On g2 its action would be purely

88 CHESS EXPLAINED: THE NIMZO-INDIAN

defensive and it would be very restricted by the black pawn-chain b7-c6-d5; also, as we men­tioned above, if White carries out the natural advance b4, the c4-square falls into the hands of the black knights.

If 6 'ifb3, Black plays 6 . . . c5 ! and White has problems supporting his centre; for example, 7 a3 �aS! 8 e3 (8 lbf3? loses a piece after 8 . . . cxd4 9 lbxd4 e5 1 0 lbc2 d4, Fish-Morozevich, Al­ushta 1 993) 8 . . . cxd4! 9 exd4 lbc6 10 lbf3 dxc4 (M.Kopylov-J.Schulz, 2nd Bundesliga 2005/6) leaves White with an isolated pawn in an unfa­vourable position, since with the bishop on the h l -a8 diagonal, the possibilities of attack on the kingside are reduced; furthermore the queen on c4 will be an object of attack.

We now return to the position after 6 lbf3 (D):

6 ... dxc4 7 0-0 lbc6 The alternatives to this natural developing

move are somewhat inferior; e.g., 7 . . . �d7, with the idea of completing his queenside develop­ment and neutralizing the pressure on the long diagonal, is answered with 8 tt:Je5 !.

The other developing move 7 . . . lbbd7 gives White a comfortable position, since the knight does not put any pressure on d4 as it would from c6: 8 'ifc2 lbb6 (if 8 . . . c5 then with 9 l!d I ! White gains a lead in development, while Black has problems bringing his queenside into play) 9 l:td I , followed by liJe5 or the advance 9 e4 and �e3, leaves White with sufficient compen­sation for the pawn.

8 a3!?

White wants to clarify the position of the b4-bishop and, in the event that Black takes on c3, the d4-pawn will be strengthened, which al­lows White to concentrate on the advance e4-e5. Furthermore, if the dark-squared bishop disappears, the pin with �g5 will be very awk­ward, since Black has only two ways to get out of it: to make the risky advance of his kingside pawns ( . . . h6 and . . . g5) or remove the queen from the d8-h4 diagonal, allowing the destruc­tion of his kingside with �xf6.

In addition to 8 a3, there are the following options:

a) The slow 8 e3 (with the idea of 'i'e2, l:!.dl and e4) should be answered with the stan­dard plan of 8 . . . .i:!b8 ! 9 'ife2 b5 10 l:i.d l ctJdS ! (hindering White's play and anticipating the advance e4-e5) I I �d2 �e7 ! (vacating the b4-square and protecting g5) 1 2 e4 and now both 12 . . . lbxc3 1 3 bxc3 lba5 ! and 1 2 . . . CDdb4 !? offer Black good chances.

b) 8 l:!.e I supports the advance of the centre pawns to dislodge the black knights, obtaining play in the centre and on the kingside, but this runs into the sequence 8 .. . lbd5 ! (putting pres­sure on c3) 9 'ifc2 �e7! (D).

w

The student should remember this well­timed retreat. Black protects the g5-square and at the same time vacates the b4-square for his knights, so that if White advances in the centre with e4, Black plays . . . lbb4 and . . . lbd3 and the knight lodges at the heart of the white position. If White prevents this manoeuvre with a3, this weakens the b3-square, which allows another

FIANCHETTO VARIATION: 4 g3 OR 4 Ci'Jj3 c5 5 g3 89

typical manoeuvre in this variation: . . . lt:Ja5-b3, and on b3 the knight is also very annoying. Here two important lines should be mentioned:

b l ) 1 0 lt:Je4?! lt:Jdb4 ! 1 1 '¥Vc3 (after 1 1 'i'xc4? comes the surprising l l . . .'ii'd5 ! , which wins the exchange after the forced exchange of queens: 1 2 'ii'xd5 exd5 1 3 lt:Jc3 lt:Jc2) l l . . .b5 1 2 a3 CLJd5 1 3 'ii'c2 and now Black has a choice be­tween 13 . . . ..1tb7 and 1 3 ... ..1td7. Both moves put White in a position where it is hard to justify the m1ssmg pawn.

b2) 1 0 a3 lt:Ja5 ! (thematic; the knight comes to b3, putting pressure on d4, al and c l and freeing the c-pawn to advance) 1 1 e4 lt:Jxc3 1 2 bxc3 CLJb3 ! (it is important to force the rook to lose a tempo before it reaches d 1 ) 1 3 .l:.b 1 b5 14 .tf4 .ib7 and Black gradually completes his development whilst retaining the extra pawn, Vi lela-Yusupov, Cienfuegos 1979.

c) The line preferred by the specialist Moi­seenko is 8 .ig5 !? (D).

B

White completes his development and tries to regain his pawn as soon as possible. 8 . . . h6 (the other way to play is to hang on to the pawn with the typical 8 . . . .l:.b8 and . . . b5, but after 9 .l:!.c l !? b5 10 b3 ! White obtains good pressure on the c-file) 9 1i..xf6 'ii'xf6 1 0 .l:.c 1 ! (worse is 1 0 e3 ..l:!.b8 1 1 lt:Jd2 lt:Ja5 ! and White has difficulties regaining the pawn) 10 . . . l:td8 1 1 e3 and now Black has two paths to equality:

c l ) l l . . .�e7 12 'i¥e2 tt:Ja5 1 3 lt:Je5 .id7 14 CLJxc4 .ixc3 15 l:txc3 ..ltb5 16 b3 c5 ! 17 'i¥h5 .ixc4 18 bxc4 cxd4 19 exd4 'ii'b4 with a com­plex position in which White has to play with

great energy to compensate for the weakness of his central pawns, Moiseenko-Bologan, Rus­sian Team Ch, Ekaterinburg 2002.

c2) l l . . .e5 ! ? (the simplest) 1 2 lt:Jxe5 lt:Jxe5 13 dxe5 'i¥e7 14 �h5 ..ltxc3 1 5 .U.xc3 .U.d2 1 6 .l:.bl 'iVb4! and the counterplay against b2 and the first rank gave Black equality in J.Gonza­lez-Tiviakov, Cal via Olympiad 2004.

We now return to 8 a3 !? (D):

B

8 ••• .ie7! The most promising continuation; Black

prepares to defend the pawn with . . . lt:Ja5 and . . . tt:Jb3, intending subsequently to strike against the white centre with . . . c5.

If 8 . . . ..1td6, White has the move 9 'ii'a4! , which regains the pawn immediately. If 8 . . . .ia5, then 9 .ig5 is best and after 9 . . . h6 White can play 1 0 .ixf6 �xf6 1 1 �a4 1i..b6 1 2 e3 lt:Ja5?! 1 3 lt:Je5 �e7 14 lt:Je4! , regaining the c4-pawn.

The ambitious 8 . . . ..1txc3 gains time to com­plete development and consolidate the extra pawn, but it gives up the important dark-squared bishop and strengthens White's centre. After 9 bxc3 White threatens to regain the pawn with 1 0 'ii'a4, so Black plays 9 . . . .U.b8 but after 10 .ig5 b5 (the immediate 10 . . . h6 is also an­swered with 1 1 ..lth4!) 1 1 e4 h6? ! (Adams rec­ommended l l . . . lt:Je7 1 2 .ixf6 gxf6, but after 1 3 'ii'd2 or 1 3 a4 !? White has sufficient play for the pawn) 1 2 ..lth4 e5 13 lt:Jxe5 lt:Jxe5 14 dxe5 'iYxdl 1 5 llfxdl lt:Jd7 1 6 f4 Black did not man­age to equalize in Carlsen-Adams, Turin Olym­piad 2006.

9 e4

90 CHESS EXPLAINED: THE NIMZO-INDIAN

9 e3?! leaves the c l -bishop dead and Black would respond with the same plan as in the main game: 9 . . . tt:'la5 and . . . c5 when appropriate.

w

9 •.• tt:'la5! (D)

10 �f4 10 �e3 tt:'lb3 ! 1 1 l:tb 1 c5 ! is similar. 10 ••• tt:'lb3! More accurate than allowing White's ideal

set-up with 'iYe2 and .Mad ! , which gave him enough play to muddy the waters in Gulko­Popovic, Clichy 1 986.

1 1 .l::!.b1 c5! (D) The most annoying move for White to meet,

although i t is also interesting to defend the c4-pawn and develop the queenside with . . . a6 and . . . b5 .

12 tt:'lb5 The alternative would be the central advance

12 d5, but after 1 2 . . . exd5 1 3 e5 (if 1 3 exd5, then

1 3 . . . �d6 and the d5-pawn is blocked for the moment) 1 3 . . . tt:'lh5 14 'i¥xd5 tt:'lxf4 15 gxf4 iUS 1 6 ri.bd I 'i¥xd5 1 7 tt:'lxd5 Black keeps his extra pawn with 17 . . . �d8 or 17 . . . .l:Iae8.

12 ••• a6! Black does not fear the penetration by the

knight enemy. 13 tt:'lc7 .l:Ia7! 14 d5 tt:'lh5 Heading straight into favourable complica­

tions, but the clearest line was 14 . . . exd5 ! 15 tt:'lxd5 (15 exd5 �d6) 15 . . . tt:'lxd5 16 �xd5 b5 17 't'kh5 �b6 with a significant advantage, not only on account of the extra pawn, but because the knight can go to d4 and the rooks to the d­file, while the queen can defend the kingside with . . . 'iYg6 if need be.

15 �e5 �d6! 16 dxe6 More complicated was 16 tt:'lg5 !? .lixe5 17

'ikxh5 h6 18 tt:'lxf7 (after 18 tt:'lgxe6 �d6! 19 tt:'lxf8 Wxf8 the c7-knight remains trapped) 1 8 . . .Ihf7 19 tt:'lxe6, but after 1 9 . . :�f6 the extra piece must count.

16 •.• �xc7 17 �xc7 �xc7 18 tt:'lg5 f6! (D)

Forcing White to retreat or exchange his ac­tive knight.

19 tt:'lxh7 If 19 tt:'lh3, then 1 9 . . . g6, while 1 9 �d5 is met

by 19 . . . 'iYe7 20 tt:'lf3 g6, when the e6-pawn falls.

19 • . . Wxh7 20 'iVxh5+ Wg8 21 �h3 b5 22 .l:!.bd1

Or 22 �f5 g5 !, and Black's major pieces de­fend the second rank.

22 •.• lt:Jd4!

FIANCHETTO VARIATION: 4 g3 OR 4 fDJ3 c5 5 g3 91

The knight blocks the d-file, hindering any tactical shot by White.

23 .l::txd4 The last gasp before resigning. 23 ..• cxd4 24 'iVg6 'iVd6! 25 .l:i.dl

The checks come to an end after 25 ..ltf5 .ixe6! 26 'iVh7+ cJ;;;f7 27 .ig6+ c3;;e7 28 'iVxg7+ ..ltf7 0

25 •.• .ixe6 26 .l:i.xd4 'iVxd4 27 .ixe6+ .l:!.tT7 0-1

Game 22 [E20]

Radoslaw Wojtaszek - Vladimir Akopian Spanish Team Ch, Lugo 2007

1 d4 CLJf6 2 c4 e6 3 CL:lc3 .ib4 4 g3 c5 5 CL:lf3 cxd4 6 CL:lxd4 CL:le4 7 'iVd3

The most active move. White both defends c3 and attacks the knight on e4. 7 .id2?! makes no sense because Black can double the pawns on c3, exchanging White's bishop-pair, or even capture on d2 and keep his own bishop-pair. 7 't!Vc2?! is also weak because Black would play 7 . . . 'iVa5 8 CLJb3 (8 CLJb5 a6 9 'iVxe4 axb5 1 0 .id2 bxc4 1 1 'iVxc4 CL:lc6 1 2 .ig2 d5 1 3 'iVh4 0-0 and Black's strong centre gives him a very comfort­able game) 8 . . . 'iVf5 !, when White has serious problems because his queen is unprotected.

7 .•• ..ltxc3+ (D)

Black immediately doubles White's pawns and posts the knight on the outpost c5, gaining a tempo with the attack on the queen.

8 bxc3 CL:lc5 9 'iVd2 The advantage of 9 'iVd2, compared with 9

't!Ve3, is that it keeps an eye on the d6-square, so that after the manoeuvre CLJb5-d6 the knight will be supported and will therefore not be

forced to exchange itself immediately for the black bishop, but the positions resulting from both lines are in fact very similar. For example: 9 'iVe3 b6! 1 0 .ig2 (putting the knight on d6 im­mediately looks very menacing but with careful play Black can complete his development and subsequently exchange the annoying knight: 1 0 CLJb5 0-0 1 1 CLJd6 'iVc7 { attacking the unpro­tected knight } 1 2 CL:lxc8 .l:!.xc8 1 3 .ig2 CL:lc6 14 0-0 .l:!.ab8 and Black will patiently improve his position before finally attacking the weak white pawns) 1 0 . . . .ib7 l l .ixb7 CLJxb7 1 2 .ia3 !? (the immediate 1 2 'iVf3 can be met with the thematic 1 2 . . . CLJc6! , a typical pawn sacrifice to complete development and gain time: 1 3 CL:lxc6 dxc6 14 'iVxc6+ 'iVd7 1 5 ifxd7+ c3;;xd7 1 6 .ia3 .l:!.hc8 and with the queens off the board the ending is inferior for White, on account of the weakness of his pawns on the c-file) 1 2 . . . CLJa6 (the most logical plan; a knight is strong on c5, blocking the action of the bishop on the a3-f8 diagonal) 1 3 CLJf5 'iVf6 14 CLJd6+ CL:lxd6 1 5 ..ltxd6 .l:!.c8 1 6 .l:i.d 1 CL:lc5 1 7 0-0 (D).

92 CHESS EXPLAINED: THE NIMZO-INDIAN

One of the typical positions of the variation. It seems that White has a big advantage in view of the position of the bishop on d6 which pre­vents castling, but after 1 7 . . . tLlb7 ! the bishop has to retreat (which will be answered with . . . tLlc5 !) or allow itself to be exchanged, which would equalize the struggle, since White's pres­sure on d7 is balanced by his weak doubled pawns.

The main alternative is 9 iVf3 !? (D).

B

This hinders the development of Black's queenside but has the drawback that when a black knight reaches e5 it will attack both the c4-pawn and the queen: 9 . . . d6 10 ii.g2 tLlbd7 ! 1 1 0-0 (recently there have been some trials of 1 1 �g4! ? 0-0 1 2 ii.h6 �f6 1 3 ii.g5 �g6 1 4 �h4, diverting the black queen away from the queenside, but after 1 4 . . . tLlf6 1 5 0-0 i.d7 1 6 tLlb3 Itac8 ! 1 7 .l:tfd 1 i.c6! Black has little to fear) 1 1 . . .tLle5 1 2 �f4! (best, since 1 2 iVh5? ! 0-0 13 .l:i.d1 allows 13 . . . tLlxc4! 14 tLlb3 tLld7 ! ) 1 2 . . . ii.d7 (now 1 2 . . . tLlxc4? would be an error on account of 1 3 tLlb3 ! ; this retreat is the key to the position - White forces a capture on b3, which substantially improves his pawn-struc­ture and breaks the blockade on c5) and now:

a) 1 3 tLlb3 is not too good due to 1 3 .. J::i.c8 ! ? 1 4 �d4 f6 ! - this i s a key move in Black's de­fence. It supports the knight on e5 and provides a retreat-square on f7, from where the knight will protect the d6-pawn. It is true that at the moment the d6-pawn is being offered as a sacrifice, but this is a common idea, to gain tempi and sub­sequently attack White's doubled pawns: 15

�xd6 b6 ! 16 �d4 tLla4 17 tLld2 i.c6 and Black has excellent compensation, since the white pieces are tied to the defence of his doubled pawns, Kramnik-Serper, Gdynia 1 99 1 .

b) 1 3 tLlb5 i.xb5 1 4 cxb5 (D).

B

White has managed to get rid of his doubled pawns while retaining the bishop-pair, which makes his position slightly preferable. How­ever, he has a backward pawn on the c-file and the black knights enjoy strong and solid posi­tions. Black completes his development with 1 4 . . . 0-0 and . . . M.c8 and if the pressure increases on d6, he can play . . .f6 or . . . f5 and bring the knight from e5 to f7, comfortably defending the d6-square.

9 ... b6 10 tLlb5 0-0 11 tLld6 i.b7! Now White has to decide whether to take the

bishop immediately or keep his knight on d6 and play the awkward move 1 2 f3.

12 f3!? Preferring to maintain the knight on d6. If he

exchanges it by 1 2 tLlxb7 tLlxb7 1 3 ii.g2 tLlc6 1 4 i.a3 .l:i.e8 1 5 .l:i.d I �f6! 1 6 0-0 .U.ed8 17 f4 .l:i.ac8, then Black reaches his ideal formation and is ready to attack White's weaknesses with a timely . . . tLlca5, Parker-Adams, British League (4NCL) 2002/3.

12 .•• tLlba6 13 i.g2 i.c6 14 0-0 f5 Gaining space on the kingside. 15 i.a3 �f6! (D) In this type of position where Black is under

pressure on the d-file and has less space, one of his main problems is to find an acceptable posi­tion for his queen. The ideal square seems to be

FIANCHETTO VARIATION: 4 g3 OR 4 l:iJj3 c5 5 g3 93

f6, where it connects the rooks but is more ac­tive than on c7.

16 e4 fxe4 17 fxe4 �g6 18 �d4 �g5 19 �xf8+?!

It is not necessary to give up the file. 1 9 e5 !? is preferable.

19 ..• �xf8 20 .tel �g6 21 .tf4 Cba4! 22 �fl CD6c5 23 .te5?!

The exchange of rooks helps Black's game and leaves the white king rather weaker than his opponent's.

23 ... �xfl + 24 'itxfl ct:lb2 (D) Black activates his knights in exemplary

fashion. The pressure on the opponent's weak pawns, the greater security of his king and the passivity of White's bishops make Black's po­sition preferable.

25 Cbc8 �h5! Generating various threats: the h2-pawn,

. . . Cbxc4, the possible exchange of queens on d l

and . . . ct:lbd3 ! weaving a mating-net around the white king.

26 'itg1 After 26 Cbe7+ 'itf7 27 Cbxc6 dxc6 28 'itg l

Cbxc4 ! 29 .txg7 e5 30 �xc4+ 'itxg7 the queen and knight are better than the queen and bishop.

26 •.. �d1+! Going into an ending in which White's bish­

ops are rather clumsy. 27 �xd1 Cbxd1 28 Cbxa7 28 Cbe7 +? 'itf7 29 Cbxc6 dxc6 30 .tb8 CDd7

3 1 .txa7 Cbxc3 32 e5 c5 is pretty hopeless. 28 •.• .txe4 29 .txe4? Losing time. 29 .td4 ! should draw. 29 •.• Cbxe4 30 Cbc8 CDd2! 31 .ii.d4 Cbxc4 32

ct:lxb6 ct:lxb6 33 .txb6 Cbxc3 (D)

Black has ended up with an extra pawn, but must take care, because passed rook's pawns are dangerous enemies to a knight.

34 a3 'itf7 35 'itfl e5 36 .ta5 Cba4 It is essential to keep the passed pawn under

lock and key. The knight will be activated once Black has centralized his king.

37 'ite2 'ite6 38 '>td3 '>td5 39 .tc7 e4+ 40 'ite3 ct:lb2

The moment has arrived for bringing the knight to the support of his passed pawns.

41 'itf4 g5+!? 42 'itxg5 e3 43 g4 43 'itg4 is met by 43 . . . 'ite4. 43 •.• d6 44 .ia5 'ite4 Now the advance of the pawns decides the

game without any great problems. 45 'ith6 d5 46 g5 Cbc4 47 .tb4 d4 48 'itxh7

Cbe5 0-1

94 CHESS EXPLAINED: THE NIMZO-INDIAN

Game 23 [E201

Wang Vue - Dmitri Vakovenko Novgorod 200 7

1 d4 lt.Jf6 2 c4 e6 3 lt.Jc3 ii.b4 4 lt.Jf3 c5 5 g3 cxd4 6 lt.Jxd4 0-0 7 ii.g2 d5!? (D)

w

One of the most logical and popular lines against the Fianchetto Variation. Black castles and then immediately breaks in the centre with . . . d5 .

8 cxd5 It is possible to sacrifice the pawn temporarily

with 8 0-0 !? dxc4 9 Wia4 but after the precise 9 . .. lt.Ja6! 1 0 lt.Jdb5 lt.Jd5 ! (restricting the action of the g2-bishop) I I .i:tdi both l l . . .ii.xc3 !? I 2 lt.Jxc3 lt.Jxc3 1 3 bxc3 lt.Jc5 ! I 4 Wic2 'iVc7 and the pawn sacrifice I I . ..Wie8 !? I2 lt.Jxd5 exd5 I3 ii.xd5 ii.d7! give Black good play.

The ambitious 8 Wib3 takes care of the pawn­structure, but prematurely decentralizes the queen. Black has to choose either to give up the bishop-pair immediately in return for a strong and mobile centre with 8 . . . 3Lxc3+ 9 Wixc3 e5 I 0 lt.Jb3 lt.Jc6 I I ii.g5 dxc4 ( l l . ..d4! ?) I 2 Wixc4 ii.e6 I 3 Wih4 ii.xb3 ! I4 axb3 lt.Jd4, when the threats compensate for the enemy bishop-pair, or to defend the bishop and prepare .. . e5 with 8 . . . lt.Ja6 9 cxd5 lt.Jxd5 I 0 0-0 lt.Jxc3 I I bxc3 ii.e7 (D) (or I I . . .ii.c5 !?).

Black has succeeded in weakening the c3-and a2-pawns and the knight has an outpost on c5 from where it will protect b7 . On the other

hand, White has compensation in the pressure that his pieces exert on the centre and on b7.

8 ••• lt.Jxd5 9 'i¥b3 The main alternative is 9 ii.d2 !?. Here the

clearest way to equalize is to gain tempi for de­velopment with 9 . . . 3Lxc3 1 0 bxc3 e5 I I CDb3 lt.Jc6 I 2 0-0 lt.Jb6 I 3 ii.e3 Wic7. Black will have harmonious development and sufficient coun­terplay against the c4-square.

9 ••• Wia5 10 ii.d2 lt.Jc6 (D)

This allows White to weaken Black's pawn­formation with lt.Jxc6 but in that case the knight on d5 will be strengthened and at the same time a path will be opened for the bishop to come to a6.

FIANCHETTO VARIATION: 4 g3 OR 4 t:tJj3 c5 5 g3 95

1 1 tbxc6 White must try to take advantage of Black's

pawn-weaknesses and put his own bishop-pair W to use. Retreating the knight gives Black too many tempi: I I tbc2 .i.xc3 1 2 bxc3 Wile? with the familiar idea of bringing the knight to c4 via a5.

ll . . . bxc6 12 0-0 .i.xc3 White was threatening tbxd5 and tbe7+, win­

ning a pawn. 13 bxc3 (D)

B

It seems that White has managed to retain a healthy pawn-structure and the bishop-pair, but Black neutralizes these advantages with active play, using the a6-fl diagonal and his actively placed knight and queen.

13 ••. i.a6 14 .U.fd1 �c5 Defending against the threat of c4, which

even works after 14 . . . .U.ab8?! 1 5 c4 ! �c5 16 cxd5 ! M.xb3 17 axb3 i.xe2 1 8 l1el ! .i.b5 (if 1 8 . . . cxd5, then 1 9 .l:i.xe2 with an advantage in material) 1 9 dxc6 and the passed c-pawn gives White the better prospects.

15 e4 .i.c4! 16 'i!Va4 The queen must protect the a3-square since

after 16 'i!Vc2 tbb6, whether White plays 17 .i.f4 e5 1 8 i.e3 �a3, Carlsen-Leko, Moscow 2006, or 17 i.e3 'i!Va3, Rakhmanov-A.Ilyin, St Peters­burg 2007, the pressure on a2 and the well­placed black pieces guarantee equality.

16 ••• tt:Jb6 17 �b4 �h5! (D) Exchanging queens would improve White's

pawn-structure and deprive Black of one of his fundamental ideas in this position: the exchange

of the light-squared bishops with . . . .i.e2-f3. Now White will try to exploit his slight initia­tive to attack the black queenside pawns and dislodge the black knight from b6 with the ad­vance a4-a5 .

18 .i.f4 By placing the bishop on f4, White provokes

the advance of the black pawns. Another line played here is 1 8 .i.e3 .i.e2 ! 1 9 l:!.d2 (worse is 1 9 l:tel tbc4 20 'i¥c5 .l:i.fd8 and it is hard for White to improve his position, Bacrot-Gri­shchuk, Cap d' Agde (rapid) 2003) 1 9 . . . .l:!ab8! (it is curious that here 1 9 . . . tt:Jc4 ?? fails to the surprising 20 g4! 'iVxg4 2 1 f3 !, winning mate­rial) 20 .i.xb6 axb6 2 1 'iVd6 l:tfc8! 22 l:tb I h6 23 e5 .i.f3 and Black has everything defended, Bacrot-Leko, Moscow 2004.

18 .•. e5!? 18 . . . c5 ! ? is possible, although later this pawn

will be an object of attack for the white bishop and Black has to be creative to defend it or else sacrifice it in favourable circumstances. For ex­ample: 1 9 �b2 llad8 ! 20 l:i.el (20 f3 ! ? is better) 20 . . . .l:!d7 2 1 h3 h6 with equality, Yakovenko­Carlsen, Tal Memorial, Moscow 2007.

19 .i.e3 .i.e2 20 .l:!d2 After 20 l:te I tbc4 2 1 �c5 1:i.fd8 ! 22 l:!.ab I h6

23 h3 tbxe3 24 'i!Vxe3 .i.c4 25 a4 a5 ! 26 'i¥c5 i.e6, as well as control of the d-file Black has counterplay against h3 and a4, Nayer-Yako­venko, Moscow 2006.

20 •• Jiab8 21 'iVc5 i.f3 22 a4 The typical plan to destabilize Black's b6-

knight. 22 •.• tt:Jxa4! (D)

96 CHESS EXPLAINED: THE NIMZO-INDIAN

w

This fine tactical resource exploits the weak­ness of the first rank and solves all Black's problems.

23 'iVxc6 23 �xa4? fails to 23 . . . .l:!.bl + 24 .lin 'iVh3. 23 •.• tt'ib6 24 .Jtxb6 Resigning himself to the draw. After the ap­

parently logical 24 .l:!.xa7? White once again has back-rank problems: 24 . . . tt'ic4! 25 'iVxc4 .l:!.bl + 26 .lin ? 'iVh3! and mate in a few moves.

24 . . . axb6 1h-1h

Conclusions

The Fianchetto Variation is a dangerous weapon, because the position of the bishop on g2 grants White unusual pressure on the queenside. Black has various ways to combat it.

In Game 2 1 , Vernay-Caruana, the variation with . . . d5 and . . . dxc4 was employed (remember that Black has to castle before capturing on c4 ! ), and although White obtained a formidable cen­tre in return for the pawn, the young GM showed that the retreat of the bishop to e7, followed by a well-timed transfer of the knight to b3 and the break . . . c5 ! , puts White in trouble. This idea also shows up if White plays for 8 .l:!.el and e4, when the manoeuvre . . . tt'id5-b4-d3 usually forces the advance a3, weakening the b3-square.

Perhaps the safest line for White is the one with 8 ..ltg5, favoured by Moiseenko, but after

8 . . . h6 and the exchange of bishop for knight, Black has two ways to equalize. If White em­ploys the scheme with 8 e3, followed by 'iVe2 and .:di to support an advance in the centre, Black should defend the c4-pawn with the the­matic . . . .l:!.b8, followed by . . . b5, which leads to very lively play with equal chances.

The variation used in Game 22 (Wojtaszek­Akopian) is one of the most popular at all lev­els, because it requires White to play energeti­cally to make up for the permanent weakness of his doubled isolated pawns on c3 and c4. I have the impression that this type of position is eas­ier to play with Black, because his plan is clear: neutralize White's pressure and then take ad­vantage of White's weaknesses. Remember that in many cases Black has to sacrifice a pawn in order to complete his development, but this is compensated by the above-mentioned white weaknesses on the c-file. White frequently suc­ceeds in placing a piece on d6, but Black man­ages either to exchange it or render it fairly harmless, before concentrating on attacking White's doubled pawns.

Game 23, Wang Yue-Yakovenko, features the most logical variation for Black against the Fianchetto. After capturing on d4, Black breaks immediately in the centre with . . . d5. In this variation White is frequently left with an infe­rior structure (isolated pawns on a2 and c3 in the variation with 8 cxd5 and 9 ..ltd2 !?), but in return White exerts pressure on the queenside. Nevertheless, Black obtains counterplay against the c4-square and quite flexible development. In the game, after 8 cxd5 and 9 'iVb3, White seems to obtain the better game, because he re­tains the bishop-pair in a position where he has no inferiority in pawn-structure. However, the pressure of the black pieces in the centre and on the queenside (the a6-n diagonal, and the squares c4, c3 and a3 in some cases) promises Black equal chances. This variation is very popular and its constant use at the highest level means that its theory will continue to be devel­oped.

7 Leningrad Variation (4 ..tg5) and Other Lines

Introduction

In this final chapter we shall study the Leningrad Variation 4 .ig5 !? , 4 '!Wb3 (Spielmann) and other moves such as 4 .id2, 4 e4? and 4 '!Wd3.

The pinning move 4 .ig5 !? gives rise to the Leningrad Variation, so called because of the boost it received through its use by players such as Zak, Korchnoi and Spassky, natives of the city that is now known as St Petersburg. With 4 .ig5 the black knight is immobilized and a degree of control is exerted over the disputed central square e4. At the same time, the bishop is brought outside the pawn-chain, so that White can play e3 without this bishop being left inactive, as happens in other variations. On the other hand, 4 .ig5 temporari ly weakens the queenside and Black can create counterplay on the a5-e I diagonal.

We shall look at two lines to combat White's plan which have in common the idea of contesting White's occupation of the centre, which takes place after 4 .ig5 h6 5 .ih4 c5 6 d5 .

The Games

In Game 24 (Comas-Delchev) we shall see how the move-order with . . . d6 and . . . '!We7 ! , followed by putting pressure on the d5-pawn with . . . ctJbd7 and . . . ctJb6, is a plan that is just as viable as that of closing the centre with the advance . . . e5.

In Li Shilong-Harikrishna (Game 25), Black employed the popular gambit against this variation that starts with 6 . . . b5 !?. The Indian GM, a great connoisseur of the Leningrad Variation with both colours, employed the best plan, which consists of playing . . . a6 after White has captured on b5 . He reached a satisfactory position, but then he failed to notice the danger and was beaten by the Chi­nese GM with a strong attack on his king via the weakened light squares. As we shall see in the notes to the game, Black has other options, such as first placing the bishop on b7 to hinder the white knight from developing on e2.

The positional risk involved in allowing the capture . . . .ixc3, which on many occasions disrupts White's pawn-structure, has persuaded many players to try various other variations (4 e4?, 4 'lid3, 4 i..d2), in the hope of surprising the opponent. In Game 26 (Corral-Lautier), we shall look at these less common options for avoiding the main lines, with the main focus on the move 4 'lib3, which was used in the main game. Black chose a sound line against this variation (4 . . . c5 and . . . ctJc6) and soon reached a position in which he was the one who remained with the bishop-pair. White tried to alter the natural course of the game by castling on the opposite side, so as later to launch an offen­sive against his opponent's kingside, but some minor inaccuracies in the conduct ofthe attack facil­itated Black's defence and the game finished with Black executing the enemy king on the dark diagonals.

98 CHESS EXPLAINED: THE NIMZO-INDIAN

Game 24 [ E3 1 ]

Luis Comas Fabrego - Alexander Delchev Andorra 2006

1 d4 �f6 2 c4 e6 3 �c3 ..ib4 4 ..ig5 (D)

B

4 ••• h6 Black puts the question to the g5-bishop be­

fore initiating the usual counterplay with . . . c5. The advantage of the immediate 4 .. . c5 is that the feeble response 5 e3?! (5 d5 ! is normal) can be answered forcefully with 5 .. :iVa5 ! followed by . . . �e4, attacking the g5-bishop and increasing the pressure on c3. If then 6 ..ixf6 ..ixc3+ 7 bxc3 'ii'xc3+ 8 <;i;>e2 gxf6 Black has an extra pawn and the safer king. It i s worth pointing out that after 5 d5 ! (with or without the inclusion of . . . h6 and ..ih4) the tempting move 5 . . . �xd5?! fails to 6 ..ixd8 �xc3 7 'iVb3 �e4+ 8 <;i;>d l �xf2+ 9 <;i;>c2 <;i;>xd8 10 'iVg3 ! �xh l l l 'ii'xg7 l:te8 1 2 e3 and the knight on h l will drop; furthermore Black is undeveloped and has the weaker king.

5 ..ih4 c5 6 d5 ..ixc3+ 7 bxc3 d6 This is one of Black's basic plans: the block­

ade of the position with . . . d6 and . . . e5, reducing the scope of White's bishops. However, as we shall see, there are other, more modem methods based on immediate pressure on the d5-pawn, which force White either to sacrifice it or ex­change it for the one on e6, thus losing the char­acteristic space advantage of this variation. Furthermore, with this move-order Black avoids lines such as 7 . . . e5 8 d6!?.

8 e3 (D)

B

8 .. J�'e7! With this move-order, the e5-square is left

free for the time being. As we shall see, this can sometimes be useful in variations where White develops his bishop on d3, since Black has . . . �e5 available, attacking the bishop. Another of the advantages of this move-order is that White is prevented from employing lines with f3, since Black would then take on d5, attacking the e3-pawn with the queen. The basic idea for Black is to put pressure on the d-pawn in order to force White either to exchange it or to sacri­fice it for development.

The classical move 8 .. . e5 is less flexible, but more popular. It quickly defines the position in the centre and frees White from the need to de­fend his queen's pawn. The specialists with White are then showing a preference for 9 �c2 !?, to prevent . . . ..if5, and this leads to a struggle in which there are plenty of ideas for both sides. Black can strengthen his grip on the centre with the advance . . . e4, with or without the move . . . g5. Black can also keep the pawn on e5 and put into practice one of the basic ideas of these positions, which is to bring his king to c7 with the manoeuvre . . . <;i;>d8-c7. If Black includes the move . . . g5, White generally strikes on the

LENINGRAD VARIATION (4 .i..g5) AND OTHER LINES 99

kingside with h4 and tries to gain control of the h-file. If Black plays ... g5 and . . . e4, White still plays h4 and, after safeguarding his king with 0-0-0, followed by Wb2-al , he attacks on both wings using the h- and b-files. White 's position is slightly preferable in any case, because his plans are aggressive, whereas Black has to await events, which will be dictated by his opponent.

9 lt:Jf3 9 'i¥c2 is correct when Black has played . . . e5

and has no pressure on d5. In this position it is not appropriate, because Black can attack the queen's pawn: 9 . . . lt:Jbd7 I 0 lt:Jf3 lt:Jb6! 1 1 i.d3 ( I I dxe6 i.xe6) l l . . .exd5 12 cxd5 lt:Jbxd5 1 3 .lld l tt'lb6! 14 c4 i.d7 1 5 'iVb2 i.c6 16 0-0 'i¥e6 1 7 e4 0-0-0 and Black seems to be a pawn up for nothing, although the opposite-side castling gives White chances of counterplay, Gallardo Garcia-Leitao, Sants 2006.

If White develops with 9 i.d3, Black can also play 9 .. . lt:Jbd7 (D) followed by .. . lt:Jb6 or .. . lt:Je5. Then White has some options:

a) 10 dxe6 is not a problem for Black: 10 .. .'�xe6 I I tt'le2! g5 ! 12 i.g3 tt'le5 1 3 0-0 i.d7 ! preparing . . . i.c6 and . . . 0-0-0, followed by beginning operations on the kingside with . . . h5-h4 or breaking with . . . d5 at an appropriate moment.

b) I 0 lt:Je2 lt:Je5 ! shows one of the advantages of having held back the advance . . . e5 . From e5 the knight simultaneously attacks the c4-pawn and the bishop on d3, so that White loses the bishop-pair. I I 0-0 (after I I i.xf6 1Yxf6 1 2 0-0 1Vh4! Black is slightly better, Miles-Karpov,

Til burg 1 986) l l . . .g5 ! 1 2 i.g3 lt:Jxd3 1 3 "ifxd3 e5 1 4 f4 ! e4 and . . . lt:Jh5 gives Black good play.

We now return to 9 lt:Jf3 (D):

B

9 • • • lt:Jbd7 Continuing with the policy of not clarifying

the situation in the centre. 10 ..td3 This move also weakens d5 and now Black

can play against the pawn with . . . lt:Jb6! . More worrying i s 10 lt:Jd2 ! but, as well as

transposing to the traditional blockading lines with 1 0 . . . e5, Black can press on with the idea of exerting pressure on d5 by 10 . . . lt:Jb6! . White does not gain adequate compensation for the pawn after I I e4 ?! exd5 1 2 cxd5 g5 1 3 i.g3 lt:Jxe4, so it seems best to play I I dxe6 i.xe6 1 2 e4 0-0-0, when the bishop-pair is balanced by Black's better development. Note that 1 2 lt:Je4 fails to 12 . . . g5 ! and after 1 3 lt:Jxd6+ 'it>f8 !, fol-lowed by . . . l:i.d8, Black wins material.

10 ... lt:Jb6! (D) Black challenges White's approach by at­

tacking d5, which is just as viable a strategy as closing the centre with I O . . . e5 !?.

11 i.xf6 An important concession. White gives up the

bishop-pair and goes onto the defensive. The main alternatives are:

a) 1 1 0-0?! is not good because then Black closes the centre and begins an attack against the white king, which has prematurely taken up residence on the kingside, where he is ex­posed to the advance of the enemy pawns: l l . . .g5 ! 1 2 i.g3 e5 ! 1 3 tt'ld2 h5 14 f3 h4 with a

100 CHESS EXPLAINED: THE NIMZO-INDIAN

strong initiative, Dolmatov-Geller, Moscow 1 987.

b) I I dxe6 ..ixe6 12 0-0 is usually given as advantageous for White, but after 1 2 . . . g5 ! 1 3 i..g3 0-0-0 1 4 'iib3 ( 1 4 l!Ve2 tZ'lh5 !) 1 4 . . . 4Jh5 1 5 a4 tZ'lxg3 1 6 hxg3 d5 Black has the initiative.

11...'ti'xf6 12 'iVb3 exd5 13 cxd5 ..if5 Black does not have any obvious weaknesses

and his king can choose where to castle, while the d5-pawn can become a weakness.

14 ..ixf5 1 4 e4? fails to 1 4 . . . c4 !? 1 5 ..ixc4 ..ixe4 16

i..b5+ 'it>f8 and the white position collapses. 14 ••• 'ifxf5 15 c4? An error. White should have played 1 5 'iVb5+

'i:Vd7 ( 1 5 . . . 'it>f8 !?) 1 6 iVd3 with equality. 15 .• .'�g4! (D)

This double attack wins a pawn, for which White obtains very little compensation.

16 l:i.c1 'iVxg2 17 'it>e2 '*lVh3

Black has won a pawn and it only remains to limit White's counterplay along the half-open b- and g-files.

18 l:thg1 g5! (D) Defending the pawn and putting a stop to

White's activity on the g-file. However, the kingside is weakened and it will not be easy to find a safe refuge for the black king. Worse is 1 8 . . . 0-0? 19 l:tg3 followed by 20 llcg I with a strong attack.

19 'iVb2 White could have started an attack on the b­

file with 1 9 a4! ? but after the accurate reply 19 . . . tZ'ld7 !? 20 'iVxb7?! l:tb8 21 l!Vxa7 l::tb2+ 22 tZ'ld2 tZ'le5 Black has a decisive attack.

19 ••• 'it>e7?! (D) This risky decision to leave the king in the

centre of the board makes it easier for White to create counterplay. The correct way was to play 1 9 . . . 0-0-0! 20 a4 (20 �f6 l!Vd7) 20 .. . 'it>c7! (remi­niscent of one of the typical manoeuvres of this variation, . . . 'it>d8-c7) 2 1 l:tbl l:tb8 22 �f6 tZ'ld7 ! and Black consolidates his advantage.

20 .l:!g3 'i¥h5 21 'it>fl l::thg8 Black should have captured the h-pawn with

2 l . . .g4 22 tZ'ld2 'i¥xh2, not just to gain material but so that White's king is also in danger.

22 e4!? g4 23 tZ'ld2 .l'.:i.g5! Preventing the opening of the centre, where

the black king is stranded. 24 'it>g1 l:i.b8 Protecting b7 and preparing the centraliza­

tion of the knight by . . . tZ'ld7-e5. 25 l:i.b3?!

LENINGRAD VARIATION (4 Ji.g5) AND OTHER LINES 101

The Spanish GM could have caused his op­ponent more problems with 25 .l:!.a3 !, since after 25 . . . tt:lc8 the knight is badly placed and White can exploit this to break on e5 with 26 .l:tel 'it>d7 (if 26 .. .f6, then 27 tt:lfl followed by tt:lg3-f5 with good pressure) 27 t!.ae3 and Black is un­able to prevent the e5 break.

25 •.. Ii.e5 26 tt:lfl ! 'iVg5 27 tt:lg3 'it>f8 Finally the king heads for the kingside. 28 f4!? (D) White senses that his counterplay is running

out and rightly gives up a second pawn to open the f-file.

28 •.. 'iVxf4 29 .l:!.fl 'iVg5 30 'iff2 'iVg6 31 h4?! The natural 3 1 tt:lf5 !, threatening 32 tt:lxh6 and

32 tt:lxd6, would have complicated the game. 31...tt:lxc4 32 h5 'iVg7 33 'iVe2

B

After 33 tt:lf5 ! ? Black can play 33 . . . .l::i.xf5 ! 34 'iVxf5 'it>e7, followed by 35 . . . tt:le5, with a strate­gically won position.

33 ••• tt:lb6?! It was preferable to keep the knight on c4,

maintaining contact with the e5-square, by play­ing 33 .. . b5 !?.

34 tl.f4 In answer to 34 tt:lf5 the exchange sacrifice

was once again indicated: 34 .. . .l:!.xf5 ! 35 .l::i.xf5 tt:ld7 36 .l:lg3 'ifd4+ 37 'it>g2 tt:le5 achieving a fairly comfortable advantage.

34 •.• l:.g5 35 .l:!.xb6? Probably the result of time-pressure, although

White could now do little to complicate the game. 35 •.• axb6 36 'iVb5 'ife5 37 'iffl 'it>g8 38 'it>g2

.l:i.a8 39 .l:txf7 .l:.txa2+ 0-1

Game 2 5 [E3 1 ]

Li Shilong - Penteala Harikrishna Moscow 200 7

1 d4 tt:lf6 2 c4 e6 3 tt:lc3 .ib4 4 .ig5 h6 5 .ih4 c5! 6 d5

After 6 e3? ! 'iVa5 7 'ifc2 tt:le4 ! Black has the initiative.

6 •.. b5!? (D) The most aggressive way of playing against

the Leningrad Variation. This is the start of an attractive gambit, designed to blow up White's centre before he can finish his development and begin to make use of the space advantage pro­vided by the d-pawn.

7 dxe6 There are various ways to handle this gambit,

although often we are just talking about different move-orders leading to the same position. Some alternatives:

a) The counter-gambit 7 e4 !? usually leads to a draw: 7 . . . g5! 8 .ig3 tt:lxe4 9 .ie5 0-0! 1 0 'iVh5 (the white queen's proximity creates some dangers for the black king) I O . . . d6 1 1 .id3 tt:lxc3 1 2 'iVxh6 tt:le4+ 1 3 'it>fl dxe5 14 .ixe4 f5 1 5 'iVg6+ and the draw was soon agreed in

102 CHESS EXPlAINED: THE NIMZO-[NDIAN

w

Yermolinsky-Shabalov, USA Ch, Parsippany 1 996.

b) According to Kasparov, the most accu­rate move is 7 e3 (D).

B

b 1 ) However, this just constitutes a transpo­sition of moves if Black replies 7 .. . 0-0! , which has been traditionally rejected on account of the strange move 8 'i¥f3 !? with the threat of bring­ing about a favourable endgame after .i.xf6. However, if Black continues with Yusupov's ex­cellent recommendation of 8 . . . g5 !? (instead of 8 .. . .i.b7 9 .i.xf6!, Timman-Yusupov, Hilversum (6) 1 986) the ensuing complications are far from clear. For example: 9 dxe6 d5 ! 10 cxd5 fxe6 1 1 dxe6 .i.xe6 1 2 'i¥xa8 .i.xc3+ 1 3 bxc3 Vi'a5 ! 1 4 lLle2 .i.d5 and White loses his queen; 9 d6 lLlc6 1 0 cxb5 .i.b7 ! 1 1 bxc6 .i.xc6 1 2 'i¥h3 Vi'a5 1 3 lLle2 lLle4 1 4 !:te l fixa2 with a winning attack; or 9 .i.g3 !? exd5 1 0 cxd5 d6 and Black gains a lead in development with the threat of . . . .i.g4.

b2) 7 . . . .i.b7 ! ? is also interesting, because the bishop will be well placed here after White clarifies the situation in the centre: 8 dxe6 fxe6 9 cxb5 0-0 (now White is forced to develop the knight on f3) 1 0 lLlf3 Vi'a5 ! 1 1 .i.xf6 .l:txf6 12 Vi'c 1 ! ? . In some sources this position is evalu­ated as slightly better for White, but after 1 2 . . . a6 ! 1 3 bxa6 lt:'lxa6, Black obtains a clear advantage in development and his pieces are more active. After the coming exchange on c3, Black will have pressure on the white pawns at a2 and c3. For this, the plan is . . . c4 followed by . . . lLlc5 and . . . lLle4.

7 ... fxe6 8 cxb5 0-0 Tal once beat Spassky with 8 . . . d5, but at the

present time preference is given to developing and putting immediate pressure on c3 and b5, holding the pawn back on d7. This is because White has obtained good results by developing the bishop to d3 and the knight to e2 (protecting the c3-knight). For example, 9 e3 0-0 10 i..d3 ! and now the immediate advance 1 0 . . . d4 does not produce the desired results after 1 1 a3 ! i..a5 12 exd4 cxd4 1 3 b4 dxc3 14 bxa5 .i.b7 1 5 lt:'le2! i..xg2 16 l::i.g 1 .i.f3 1 7 �g3 ! . In this open posi­tion I prefer White on account of his bishop­pair, although Black has good chances of creat­ing complications since the white king is also weak.

9 e3 (D)

B

9 . . . a6! There are various alternatives, but the best

plans are all based on eliminating the b5-pawn with . . . a6:

LENINGRAD VARIATION (4 il.g5) AND OTHER LINES 103

a) It is inaccurate to occupy the centre im­mediately with 9 . . . d5? ! , because after 1 0 ..ltd3 ! jg_b7?! I I tt:Je2 tt:Jbd7 1 2 0-0 �e8 1 3 ..ltg3 e5 14 a3 ..lta5 1 5 �bl any advance of Black's central pawns gives White squares and diago­nals, whereas the white pawns can advance on the queenside, driving Black's forces back, Ward-Hinks Edwards, British League (4NCL) 1 999/00. This is the type of position that Black must avoid.

b) It is more logical to play 9 . . . 'i!Va5 !? 1 0 ..ltxf6 (necessary to prevent . . . tt:Je4 ! ) I O . . Jhf6 and subsequently the most appropriate plan is to play l l . . . a6 !? to open the a-file and liberate Black's minor pieces.

c) 9 . . . ..ltb7 !? is also playable, since this pre­vents White from comfortably developing the bishop to d3 and the knight to e2. Later Black follows up with . . . a6.

10 jg_d3 (D) Capturing with I 0 bxa6 allows Black to de­

velop fully with I O .. .'iVa5 !? (forcing �xf6), . . . tt:Jc6 and . . . d5, with the threat of . . . d4. A tac­tical theme that repeatedly crops up in these positions is the advance . . . d5-d4, after first eliminating the f3-knight with . . . .l::!.xf3. It might be necessary for Black to capture on a6 if White plays I I 'i!Vc I , defending c3 and preparing a3. This capture should preferably be made with the knight, to have available the manoeuvre . . . tt:Jc7-d5, increasing the pressure on c3.

B

10 •.. axb5 1 1 tt:Je2 1 1 jg_xb5 is met by I I . . . 'iVa5 ! 1 2 'iVd3 d5 1 3

a4 tLlbd7, followed by . . . tt:Je5 and . . . c4, and the

knight penetrates on d3 with strong counter­play.

l l ... .i.xc3+ (D) This immediate exchange does not seem

logical. In Harikrishna-J.Polgar, Bilbao 2007, Judit preferred to retain both her bishops and continue her development with I I . . . ..ltb7 1 2 0-0 c4 ! 1 3 jg_c2 �b6.

12 bxc3!? Forcing Black to defend the b5-pawn. Black

would not be worried by 1 2 tLlxc3 c4 1 3 �c2 'i!Vb6 14 0-0 ..ltb7, because after . . . d5 White's bishop-pair is balanced by Black's strong cen­tre.

B

12 .•. jg_a6 13 a4 (D)

Trying to seize the initiative at all costs. 13 • • • bxa4?! This hands over the initiative and stabilizes

the position, which is not wise when fighting

104 CHESS EXPLAINED: THE N!MZO-INDIAN

against the bishop-pair. The correct course was to keep the game strategically unbalanced with 1 3 . . . b4 !? 1 4 cxb4 cxb4 1 5 0-0 tt::lc6 and Black has a harmonious position.

14 Itxa4 ii.b7 15 l!xa8 ii.xa8 16 0-0 �c7 (D)

On his scale of static criteria for evaluating a position, GM Dorfman awards top ranking to king safety, and he is absolutely correct. In this game, White exploits the weaknesses of the black king to conjure up a strong attack from an apparently harmless position.

17 ii.xf6! Eliminating an important defender of the

light squares . 17 •. Jlxf6 18 tt::lf4 tt::lc6?! Black does not sense the danger. Otherwise

he would have taken urgent defensive measures such as 1 8 . . . c4 !? (preventing the manoeuvre 'ii'a4-e4) 19 ii.bl tt::la6 (to blockade on e4 or d3) 20 'ii'c2 l:tf5 2 1 l:Idl ! and White has only slight pressure.

19 �a4! Gaining time to reach h7. 19 • • . ii.b7? The decisive error. It was essential to orga­

nize his defence with 19 . . JH8 ! 20 �e4 CiJe7 covering the g6-square.

20 'i!Ve4 (D)

B

The penetratjon on the Jjght squares leads to material gains.

20 ••• 'it>f8 20 . . JH5? would fail to 2 1 tt::lxe6! dxe6 22

�xe6+ llf7 23 'ii'e8+ k!.f8 24 i.c4+ winning. 21 �h7 tt::le5 22 'ifh8+ 'it>e7 22 . . . 'it>f7 does not help either, in view of 23

tt::lh5, and g7 falls. 23 �xg7+ The position is collapsing. White has a mate­

rial advantage and Black's king lacks protec­tion.

23 •.• .i:!.f7 24 'i!Vxh6 tt::lxd3 25 tt::lxd3 .lia6 26 �g5+ 1-0

After 26 ... �f8 27 .i:!.al ! Black is defenceless.

Game 26 [E23]

Juan Corral Blanco - Joel Lautier Spanish Team Ch, Cala Galdana 1 999

1 d4 tt::lf6 2 c4 e6 3 tt::lc3 ii.b4 4 �b3 This is an obsolete line which shares some

similarities with 4 �c2. On b3 the queen at­tacks the bishop but is more exposed, does not control the e4-square and obstructs White's

b-pawn. Various famous GMs have used it, such as Piket, Malaniuk and Epishin. Black has several ways to equalize. But first, let us examine some other rarely played alternatives of lesser worth:

LENINGRAD VARIATION (4 iJ..g5) AND OTHER LINES 105

a) 4 e4? (D). attack by the black pieces. Now 4 . . . c5 ! is the best way to exploit the white queen's position, according to Keres:

b l ) 5 lt:Jf3 cxd4 6 lt:Jxd4 0-0 7 a3 .i.xc3+ 8 'i!Vxc3 d5 and Black's better development guar­antees him the initiative.

b2) With 5 d5 White seeks to stabilize the situation in the centre, but White's gain of space will be met by the activity of the black pieces. 5 . . . 0-0 6 .i.d2 exd5 7 cxd5 d6 8 g3 b6 ! (after this move and . . . .i.a6, the activity of the black bishop on the a6-fl diagonal will be decisive) 9 .i.g2 .i.a6 I 0 'i!Vc2 lt:Jbd7 I I lt:Jh3 l:!.e8 with tre­mendous pressure on e2, Mikenas-Keres, Mos­cow 1 949.

c) There is a more logical possibility in 4 This move has attracted the interest of a few .i.d2 (D).

adventurous souls who have employed it with disastrous results. After 4 . . . lt:Jxe4, 5 'i!Vg4!? is the move in which the partisans of this unsound gambit place their hopes. Now simple and natu­ral play by Black neutralizes any attempt to complicate the game: 5 . . . lt:Jxc3 6 a3 (6 'i!Vxg7 is not possible on account of 6 . . . 'i!Vf6! 7 'i!Vxf6 lt:Je4+) 6 . . . .i.e7 ! 7 bxc3 .i.f6 8 .i.d3 b6 and by continuing with the development of his queen­side ( . . . .i.b7, . . . d6 and . . . lt:Jd7), Black obtains a comfortable game with an extra pawn.

b) Rather better, but also inadvisable, is 4 'iHd3?! (D).

The d3-square is not the most appropriate square for the queen since, as well as getting in the way of the possible development of White's fl -bishop, the queen is exposed to immediate

With this modest move, White relieves the pin in a simple manner and avoids the doubling of his pawns. 4 . . . d5 (Black can also choose a method of play that is closer to the Indian sys­tems; for example, 4 . . . 0-0 5 lLlf3 b6 6 e3 .i.b7 7 .i.e2 d6 8 0-0 lt:Jbd7) 5 e3 (after this move we can better appreciate some of the defects of the white position: the bishop on d2 has limited scope, in comparison with other variations where it can pin Black's f6-knight or go to b2 to support the white centre) 5 . . . 0-0 6 lt:Jf3 c5 (it is also possible to complete the development of the queenside with 6 . . . b6, ... .i.b7 and ... lt:Jbd7, followed by . . . c5) 7 a3 .i.xc3 8 .i.xc3 cxd4 9 .i.xd4 lt:Jc6 I 0 cxd5 lt:Jxd5 I I .i.e2 l:!.e8 1 2 b4 (intending to retain his bishop-pair, but allow­ing Black to develop a dangerous initiative; 1 2

106 CHESS EXPLAINED: THE NJMZO-INDIAN

0-0 is better) 1 2 .. . e5 1 3 i.b2 e4 14 lt:\d4 lt:\e5 1 5 lt:\b5 (once again 1 5 0-0 was appropriate, al­though after 1 5 . . . �g5 ! 1 6 'i¥th 1 i.g4! Black has good prospects of installing his knights on the light squares, which have been left weak by the exchange of bishops) 1 5 . . . a6 l6 lt:\c7 lt:\d3+! 1 7 i.xd3 lt:\xc7 1 8 i.c4 �g5 and Black's initia­tive compensates for the bishop-pair, R.Stern­Almasi, Dresden 2007 .

Let us return to 4 �b3 (D):

B

4 ••• c5 Now that the white queen has abandoned the

defence of the d4-pawn, Black immediately at­tacks it. One major difference compared to the 4 �c2 lines is that here of course 4 . . . 0-0?? is impossible because the b4-bishop hangs.

5 dxcS White is no position to maintain his centre,

since after 5 lt:\f3 lt:\e4! or 5 d5 lt:\e4! Black has the initiative.

s ... lt:\c6 (D) 5 . . . lt:\a6!? is equally good. 6 lt:\f3 Here 6 i.g5 is an important alternative. White

prevents . . . lt:\e4, but after 6 .. . h6 must be willing to give up his dark-squared bishop right away:

a) After 7 i.h4 it is essential to play the en­ergetic 7 . . . g5 ! 8 i.g3 lt:\e4 9 e3 i.xc3+! (the most accurate move, before the white knight can go to e2, defending c3) 1 0 bxc3 �a5 and after eliminating the g3-bishop and recapturing on c5, Black is better on account of his healthy pawn-structure, Agdestein-Hjartarson, Reykja­vik 1 996.

b) 7 i.xf6 �xf6 8 lZ'lf3 (if 8 .l:.c I then 8 . . . a5 ! ? deserves consideration, with the idea of . . . a4 and . . . a3, weakening the dark squares) 8 . . . i.xc3+! (Stern-Landa, Bundesliga 2006/7 ended dramatically after 8 . . . b6 !? 9 cxb6 axb6 { 9 . . . i.xc3+! } I 0 e3 ?? l:i.a3 ! and White resigned, in view of the inevitable loss of material; White should have played I 0 l:i.c 1 ! keeping his pawn­structure intact, with advantage) 9 'ifxc3 (if 9 bxc3, then 9 . . . b6! is equally strong) 9 .. . 'ifxc3+ 10 bxc3 (D).

Now the typical positional pawn sacrifice 1 O . . . b6! guarantees Black sufficient counter­play. By opening files for his rooks, Black puts pressure on the weak white pawns, with good chances of restoring the material balance, after which his position would be preferable. 1 1 cxb6 axb6 1 2 e3 'i¥te7 and by continuing with the plan of . . . l:i.a4, . . . i.a6 and . . . lt:\e5 or .. . lZ'la5 Black develops strong pressure against White's

LENINGRAD VARIATION (4 ii..g5) AND OTHER LINES 107

weaknesses, which will finally lead to his re­gaining the pawn.

6 •• .lt'le4 7 .i.d2 lt'lxd2 8 lt'lxd2 0-0 9 e3 .i.xcS (D)

10 lt'lde4 The great Alekhine showed how to play if

White employs the passive 1 0 .i.e2: I O . . . b6 I I 0-0 (wasting the last opportunity of reaching equality with I I lt'lde4 !) l l . . .f5 ! (with this, the white knights are denied the use of the e4-square, and at the same time Black prepares a strong attack on the kingside) 1 2 llac I llb8 1 3 lt'lf3 't!Vf6 1 4 llfd I g5 ! with a strong attack, Nor­man-Aiekhine, Hastings 1 925.

10 ... b6 11 .ie2 Letting the black bishop escape is a strategi­

cally risky decision. White should have chosen the quiet I I lt'lxc5 bxc5 1 2 l:!.d I �b8 1 3 'ifc2 .ib7 14 't!Vd2 't!Va5 1 5 .i.e2 �fd8 16 0-0, when the game is equal.

ll ... .ie7 12 0-0-0 Risky, since on the queenside the white king

can be subject to unpleasant surprises. 12 ... a6 13 g4 (D) Launching the attack. It was more prudent to

play 1 3 �d2 or 1 3 'ifc2 and then double the rooks, exerting pressure on the d-file.

13 •.• 'ifc7?! 13 . . . .ib7 is more accurate, followed by .. . .M.c8

and .. . lt'la5, keeping the queen on d8 to protect the f6-square.

14 gS ltJaS 15 't!Vc2 After 1 5 't!Va4?, Black opens lines against the

white king with 15 . . . b5 ! .

B

15 •.• lt'lxc4 16 .i.xc4 'ifxc4 17 llhg1 '.io>hS The defence requires great care: if 1 7 . . . d5? ! ,

then 1 8 lt'lf6+ .ixf6 19 gxf6 g6 20 .l:.d4 'ifc7 21 llh4 with a dangerous attack.

18 llg4 'ifc7 19 llh4 g6 20 f4? Losing his bearings. It was better to play 20

tt'Jf6! .i.xf6 2 1 gxf6, reaching an extremely complicated position in which Black should re­tain his advantage with precise defence, start­ing with 2 I . . .'ife5 ! .

20 • • • .i.b7 21 'ifd3 White does not have the resources to con­

tinue his attack, since after 2 1 tt'Jf6 .ixf6 22 gxf6 'ifd8 23 'iff2 'ifxf6 24 llxd7 .ic6 25 l:td4 .l:.ad8 Black can concentrate on exploiting his material advantage.

21.. . .ixe4 22 'ifxe4 dS (D)

w

Black has consolidated his material advan­tage and the danger to his king is over.

23 'ifg2 fS!

108 CHESS EXPLAINED: THE NIMZO-!NDJAN

Leaving the h4-rook permanently out of play.

24 �bl z:i.ac8 25 �d2 �c4 26 �d4+ �g8! Giving back the pawn in order to coordinate

his pieces with gain of time and launch the final attack on the white king.

27 �xb6 l:Ic6 28 �b7 It was better to play 28 �a7 ! l:If7 29 l:Id4 and

White exchanges the queens, easing the de­fence.

28 •• Jif7! 29 l:td4 �cs 30 �b3 .tr8 3I l:Ih3 l:Ifc7 32 l:Id3 i..g7

Black completes his regrouping, after which the game is decided, because the white rook on h3 is not participating in the struggle.

33 �a4 l:Ib7 34 �c2 l:Ixb2+! (D)

The most elegant. The rest is a hunt ofthe white king.

35 �xb2 il.xc3+ 36 �c2 After 36 �bl i..g7 ! there is also no defence. 36 ••• i..b4+ 37 �dl �cl+ 38 �e2 �el+ 39

�f3 �fl+ 0-1 And mate on the next move.

Conclusions

The Leningrad Variation has seen a revival in master practice with the recent contributions of several outstanding grandmasters. The games

that we have examined demonstrate that Black does not have to limit himself to the central blockade systems (based on . . . e5), but can un­dermine White 's centre by putting pressure on d5 (keeping the pawn on e6 and bringing the knight to b6). Generally White ends up having to capture on e6; Black recaptures with either his queen or bishop, castles on the queenside and then either attacks on the kingside or breaks in the centre with . . . d5. In these positions, White will attack on the queenside by advancing with a4-a5 and using the b-file.

The gambit with . . . b5 !? is offering Black very good play at present. It is necessary to pay careful attention to the move-orders and the plans to follow. Black must combine pressure on the a5-e l diagonal with the freeing move . . . a6 !, which allows him to complete the active development of his queenside. After the cap­ture cxb5 Black cannot obtain good play by combining natural moves such as . . . d5, . . . lL'lbd7 and . . . i..b7, since White will have no problems containing the black centre and can mobilize his forces in ideal fashion with .id3 and lL'le2. Black should therefore play for .. . i..b7, or .. .'�a5 followed by . . . a6.

The secondary lines against the Nimzo­Indian do not offer any advantage to White. Re­cently 4 i..d2 has gained some support from players seeking a safe position, but Black can equalize in various ways. The simplest is 4 . . . d5, followed by . . . c5; note that in this variation White will often gain the bishop-pair by play­ing a3, but Black then gets it back with . . . lL'le4 or . . .'�Jd5 at some moment.

Of these rare fourth-move options, only 4 �b3 has any venom. Black should react by 4 .. . c5 and .. .'�Jc6, with the idea of . . . lL'le4, putting pressure on c3 and c5. Remember that if White keeps an extra pawn in the endgame, but has tri­pled pawns on the c-file, then the correct way for Black to treat the position is to open the a- and c-files with .. . b6 !, which guarantees sufficient counterplay against White's doubled pawns.

List of Games

Alexei Alexandrov - Kim Le Quang, Abu Dhabi 2007

Evgeny Bareev - Vasily lvanchuk, Havana 2006

Jan Bernasek - Markus Ragger, Mitropa Cup, Szeged 2007

Dmitri Bocharov - Igor Lysy, Moscow 2007

Miso Cebalo - Dimitri Komarov, Bratto open 2006

Luis Comas Fabrego - Alexander Delchev, Andorra 2006

Juan Corral Blanco - Joel Lautier, Spanish Team Ch, Cala Galdana 1 999

Yuri Gonzalez - A Alvarez, Havana 2007

Mikhail Gurevich - Peter Leko, Elista 2007

Jan Gustafsson - Evgeny Postny, Greek Team Ch, Ermioni 2006

Tiger Hillarp Persson - Peter Heine Nielsen, Danish League 2001/2

Rasul Ibrahimov - Shakhriyar Mamedyarov, President's Cup, Baku 2006

Gata Kamsky - Pavel Elianov, Montrea/ 2007

Alexei Korotylev - Viacheslav Zakhartsov, Russian Ch, Krasnoiarsk 2007

Li Shilong - Penteala Harikrishna, Moscow 2007

Vadim Milov -Dashzeveg Sharavdorj, Las Vegas 2006

Hikaru Nakamura - Andrei Volokitin, Cuernavaca 2006

Teimour Radjabov - Evgeny Alekseev, Soc hi 2007

Teimour Radjabov - Emir Dizdarevic, Calvia Olympiad 2004

Ivan Sokolov -Judit Polgar, Hoogeveen 2006

Loek van Wely - Veselin Topalov, Wijk aan Zee 2004

Loek van Wely - Peter Wells, London 2007

Clovis Vernay - Fabiano Caruana, Budapest 2007

Wang Yue - Dmitri Yakovenko, Novgorod 2007

Radoslaw Wojtaszek - Vladimir Akopian, Spanish Team Ch, Lugo 2007

Yaacov Zilberman - Vlastimil Babula, Pardubice 2007

60 2 1 75 35 1 5 98

1 04 67 46 42 12 24 72 82

1 0 1 63 52 32

8 56 27 39 87 94 9 1 78

Index of Variations

w

Now: A: 4 e3 B: 4 'tWc2

1 d4 2 c4 3 lt'lc3

C: Other lines

A)

w

4 e3

5 �d3

lt'lf6 e6 �b4

0-0 51

5 a3 !? 52 5 . . . �xc3+ 6 bxc3 c5 ! 7 .id3 lt'lc6 8 lt'le2 b6 9 e4! lt'le8 ! 53

5 lt'le2 67 5 . . . d5 6 a3 .lte7 7 cxd5 exd5 67 5 d5

6 lt'lf3 6 a3 56 6 . . . .ltxc3+ 7 bxc3 dxc4! 56 6 cxd5 60 6 . . . exd5 7 lt'le2 l:te8 8 .id2 a6 9 lt'lg3 62

6 lt'le2 63 6 . . . c5 7 cxd5 cxd4 ! 8 exd4 lt'lxd5 ! 64

6 7 0-0 71

c5

7 • • • dxc4 7 . . . lt'lc6 82 8 a3 i.xc3 9 bxc3 Wl/c7 !? 10 cxd5 exd5 83

9 Wlie2 72

8 i.xc4 lt'lbd7!? 72 9 a3!? 75

9 'iVb3 72 9 . . . cxd4 ! 1 0 WVxb4 73

Now (after 9 a3 ! ?) : 9 . . . .ltxc3?! 75

9 . . . cxd4 75 1 0 axb4 dxc3 1 1 bxc3 'tWc7! 76 9 . . . .lta5 ! ? 78 10 Wl/e2 cxd4! 79

B)

4 Wl/c2 20

INDEX OF VARIATIONS 111

B

4 0-0 31 4 . . . d5 21 and then: a) 5 a3 21 5 . . . ..ixc3+ 6 '*lVxc3 tLle4 7 '*lVc2 c5 ! 22

b) 5 cxd5 24 5 .. . exd5 6 ..ig5 24 6 . . . c5 ! 27:

b1) 7 0-0-0!? 27

b2) 7 tt:Jf3 27

b3) 7 dxc5 24 7 . . . h6 8 ..ih4 g5 ! 9 ..ig3 tLle4 10 e3 'iVa5 ! 25

b4) 7 a3 277 . . . ..ixc3+ 8 'iVxc3 ! cxd4 !? 9 'iVxd4 tt:Jc6 I 0 '*lVf4 28

5 tt:Jf3 32

5 a3 35

5 e4!? 32 5 .. . d5 ! 6 e5 tt:Je4! 32

5 ..ixc3+ 6 'iVxc3 b6

7 ..ig5 39

7 tLlf3 36 7 . . . ..ib7 8 e3 d6 9 ..ie2 36 7 ..ib7

Now:

8 e3 d6: 9 ctJe2 40; 9 f3 42

8 ctJf3 42 8 . . . d6 9 tt:Jd2 c5 !? 43 8 f3 46 8 . . . h6 ! 9 ..ih4 d5 1 0 e3 ctJbd7 1 1 cxd5 !? tt:Jxd5 ! 47

C)

4 tt:Jf3 a) 4 e4? 105 b) 4 'iVd3? ! 105 c) 4 ..id2 105 d) 4 'iVb3 104 e) 4 g3 87

e l ) 4 .. . c5 5 tt:Jf3 - 4 tLlf3 c5 5 g3 e2) 4 . . . d5 87 5 ..ig2 0-0 6 tt:Jf3 dxc4 7 0-0 tt:Jc6 88

f) 4 a3 8 4 . . . ..ixc3+ 5 bxc3 b6 6 f3 tt:Jc6 7 e4 ..ia6 9

g) 4 f3 12: g 1) 4 .. . tt:Jc6 5 e4 d5 ! 6 cxd5 exd5 7 e5 tt:Jg8 12 g2) 4 . . . 0-0 !? 15 5 a3 ..ixc3+ 6 bxc3 tt:Jh5 !? 16 h) 4 ..ig5 98 4 . . . h6 5 ..ih4 c5 6 d5: h 1) 6 .. . ..ixc3+ 7 bxc3 d6 98

h2) 6 . . . b5 !? 101 4 c5

4 . . . b6 - Queen's Indian

5 g3 5 e3 - see 4 e3 0-0 5 ..id3 d5 6 tLlf3 c5

Now:

5 cxd4 6 tt:Jxd4

6 . . . tt:Je4 91 7 �d3 91

6 . . . 0-0 7 ..ig2 d5 !? 94

Other Chess Openings Books from Gambit Publications

Chess Explained: The Queen's Indian Peter Wells 1 d4 li:Jf6 2 c4 e6 3 li:Jf3 b6 is a mainstream strategic opening, rich in nuances. Many of the modem main lines involve moves that look extravagant, but are backed up by a deep underlying logic. 128 pages, 248 x 172 mm; $19.95 I £12.99

Chess Explained: The c3 Sicilian Sam Collins

The c3 Sicilian is a popular way to sidestep the main lines of the Sicilian. White denies his opponent the chance to demonstrate his preparation in some heavily-analysed system; instead Black will have to fight carefully for equality in a more rational position of White's choosing. 112 pages, 248 x 172 mm; $19.95 I £12.99

Chess Explained: The Queen's Gambit

Declined James Rizzitano

Topical explanation of the Queen's Gambit Declined - one of the most important of chess openings. This solid and classical choice for Black has remained in vogue for more than a century. 128 pages, 248 x 1 72 mm; $23.95 I £12.99

Chess Explained: The Modern Benoni Zenon Franco This opening is a perennial weapon of aggressive players looking to create winning chances with Black. It is one of the few openings where White has no easy way to force drawish simplifications. 112 pages, 248 x 1 72 mm; $23.95 I £12.99

Chess Explained: The Meran Semi-Slav Reinaldo Vera The Semi-Slav is one of Black's most aggressive defences, and the Meran Variation is its traditional main line. Vera is superb at explaining the chaotic positions - meaning rich pickings for the player who is better prepared. 1 12 pages, 248 x 1 72 mm; $23.95 I £12.99

Chess Explained: The Classical Sicilian Alex Yermolinsky Former US Champion Yermolinsky explains, in masterful fashion, the key plans and ideas of one of his favourite openings. This dynamic Sicilian weapon has been a favourite of players such as Anand, Kramnik and Shirov. 1 12 pages, 248 x 172 mm; $19.95 I £12.99

Chess Explained: The French Viacheslav Eingorn & Valentin Bogdanov This important defence suits a wide range of chess temperaments - an opening that is solid yet uncompromising. In many critical variations White accepts major structural weaknesses in return for piece-play and dynamic chances. 1 12 pages, 248 x 172 mm; $23.95 I £12.99

Chess Explained: The Taimanov Sicilian James Rizzitano This opening system is one of the most flexible options for Black in the Open Sicilian. Black avoids committing himself to a rigid set-up, and tailors his response once White has shown his hand. 1 12 pages, 248 x 172 mm; $19.95 I £12. 99

About the Publisher: Gambit chess opening books are designed to be different. We recruit the fin­est authors, writing passionately on their specialist systems. Our editorial team is obsessive about double-checking variations and transpositions. Gambit believes trustworthy analysis is important, and that an openings book should remain an asset over many years.

www.gambitbooks.com

Chess Explained books provide an understarding of a n opening and the middlegames to which it leads, enabling you to find the right moves and plans in your own games . It is as if you were sitting at the board with a chess coach answering your questions about the plans for both sides, the ideas behind particular moves , and what specific knowtedge you need to have.

The Nimzo-lndian is one of the most i mportant of al l chess openings, and popular at al l levels of play. Together with the QueeA's I ndian (also covered in this series), the Nimzo has provided the backbone of the opening repertoire of a large portion of the world's top players over the last 80

years. It is considered a ful ly rel iable defence, and provides wi nning chances for both sides as it leads to structures of great strategic variety and complexity. Key battlegrounds in the N imzo include the blockade, IQP positions, the handling of unbalanced pawn-structures, and the struggle between bishop and knight. All these topics are covered in depth in this book, and an understanding of them wi l l prove valuable in a much broader context than just the Ni mzo-l ndian.

Reinaldo Vera comes from Cuba and g ained his grandmaster title in 1 988.

He has won the Cuban Championship on two occasions and was a member of the Cuban national team for more than two decades. Vera graduated from Havana University, and has written for many chess magazines, incl uding New in Chess, Jaque and Peon de Rey. He is a lso a FIDE Senior Trainer, and the coach of the Cuban team.

Mastering the Chess Openings

2 ........ ----..

--

£12 99 $23 95

I S B N - 1 3 : , 7 8 - 1 - 9 0 4 6 0 0 - 7 2 - 5

I S B N - 10 : 1 - 9 0 4 6 0 0 - 7 2 - 7

9 7 8 1 9 0 4 6 0 0 7 2 5 >