Você está na página 1de 385

(Grandmaster Repertoire)

Michael
Roiz

Tired of bad positions? Try the main lines!

QUALITY CHESS
Grandmaster Repertoire

The Nimzo-Indian Defence


By

Michael Roiz

Quality Chess
www.qualitychess.co. uk
First edition 20 1 7 by Quality Chess UK Ltd

Copyright 20 1 7 Michael Roiz

Grandmaster Repertoire-The Nimzo-lndian Defence


All rights reserved. No part of this publication may be reproduced, stored
in a retrieval system or transmitted in any form or by any means, electronic,
electrostatic, magnetic tape, photocopying, recording or otherwise,
without prior permission of the publisher.

Paperback ISBN 978- 1 -78483-027-4


Hardcover ISBN 978- 1 -78483-028- 1

All sales or enquiries should be directed to Quality Chess UK Ltd,


Suite 247, Central Chambers, 1 1 Bothwell Street,
Glasgow G2 6LY, United Kingdom
Phone +44 1 4 1 204 2073
e-mail: info@qualitychess.co. uk
website: www.qualitychess.co. uk

Distributed in North America by National Book Network

Distributed in Rest of the World by Quality Chess UK Ltd through


Sunrise Handicrafts, ul. Poligonowa 3 5A, 20 -8 1 7 Lublin, Poland

Typeset by Jacob Aagaard


Proofreading by Colin McNab
Edited by Ian Kingston, John Shaw & Andrew Greet
Cover design by adamsondesign.com
Cover photo by capture365 .com
Printed in Estonia by Tallinna Raamatutriikikoja LLC
Contents
Key to symbols used 4
Preface 5
Bibliography 6

Various 4th Moves


1 Rare Options 7
2 4.'1Wb3 17
3 4.i.d2 32
4 4.i.g5 48
5 4.f3 66
6 4.a3 82
7 4.g3 101
8 4.ltlf3 109
9 4.ltlf3- Main Line 120

4.e3
10 Rare 5th Moves 139
11 5.a3 153
12 5.ltlge2 163
13 5.i.d3 185
14 6.a3 202
15 6.ltlf3 220
16 10.i.g5- Main Line 233

4.'i'c2
17 Various 5th Moves 255
18 5.a3 264
19 7.ltlf3 279
20 7.i.g5 297
21 5.cxd5 310
22 6.e3 c5 7.i.d2 324
23 6.ltlf3 337
24 7.b3 352

Variation Index 375


Key to symbols used
;!; White is slightly better
i Black is slightly better
White is better
+ Black is better
+- White has a decisive advantage
-+ Black has a decisive advantage
equality
i with compensation
+t with counterplay
unclear
t with the initiative

? a weak move
?? a blunder
a good move
!! an excellent move
!? a move worth considering
?! a move of doubtful value
# mate
Preface

My madness for chess started in 1 989, when as a six-year-old kid I saw my father playing with
my uncle. Back then, I could see chess in almost everything, and I started to collect and explore
every chess book I could find. Those were tough times in the Soviet Union and it was not easy to
get good chess books, but my parents did their best to support my hobby. So in 1 990 I was lucky
enough to have plenty of books at my disposal, including David Bronstein's tournament book
about the Zurich 1 9 5 3 Candidates. There were many spectacular games in this book, but I was
especially impressed by the Geller - Euwe encounter, where the former World Champion played
the Nimzo-Indian and scored a memorable victory in counterattacking style, using the exciting
motif of a rook sacrifice. The influence of this game was so significant that for the next ten years
I avoided getting doubled c-pawns in my games!

When I look back on my childhood career, I can understand why I did not play 3 .tLlc3 with
White and allow the Nimzo-Indian - it is one of most complex openings from a strategic point
of view, and the arising positions are sometimes tough to handle, even for grandmasters, so it
would be impossible for a young child. Even after many years of playing the Nimzo-Indian with
both colours, and analysing various systems with top players (including preparing for the Anand
- Gelfand World Championship match in 20 1 2, where the Nimzo played an important role) I
still fail to evaluate some positions properly, and so does the engine!

So when Quality Chess asked me to write a book on this opening, focusing on Black's side, I
found this project very challenging and this appealed to me. Indeed, White has a large choice of
possibilities even on the 4th move - therefore, a thorough evaluation of all the possible responses
for Black is difficult to say the least.

The concept of this book is to enable players to feel knowledgeable enough in any system they
may encounter when playing the Nimzo-Indian. So I offer a complete repertoire for Black after
3 . . ..ib4.

I feel I have succeeded in improving my own understanding of the Nimzo-Indian, and I hope to
share this knowledge with the reader. Best of luck in your journey with the Nimzo-Indian.

Michael Roiz
Beer Sheva, December 20 1 6
Bibliography
Cox: Starting Out: J.d4!, Everyman Chess 2006
Dearing: Play the Nimzo-Indian, Everyman Chess 2005
Emms, Ward & Palliser: Dangerous Weapom: The Nimzo-Indian, Everyman Chess 2006
Hansen: The Nimzo-Indian: 4 e3, Gambit 2002
Kaufman: The Kaufman Repertoirefor Black and White, New in Chess 20 1 2
Kornev: A Practical White Repertoire with I.d4 and 2.c4: Volume 2, Chess Stars 20 1 4
Schandorff: Playing I.d4- The Indian Defences, Quality Chess 20 1 2
Sielecki: Opening Repertoire: Nimzo and Bogo Indian, Everyman Chess 2015
Sokolov: The Strategic Nimzo-Indian: Volume I, New in Chess 20 1 2
Vigorito: Challenging the Nimzo-Indian, Quality Chess 2007
Watson: A Strategic Chess Opening Repertoire for White, Gambit 20 1 2
Yakovich: Play the 4j3 Nimzo-Indian, Gambit 2004

Periodicals
New in Chess Yearbooks

Electronicllntemet Resources
ChessPublishing
ChessBase Magazine
Gustafsson: Grandmaster repertoire: 4.Qc2 agaimt the Nimzo-Indian, Chess24 20 1 4
Various 4th Moves
a b c d e f g h

Rare Options
Variation Index
l.d4 tLlf6 2.c4 e6 3.tLlc3
3....tb4
A) 4.e4? 8
B) 4.f;C/d3?! c5! 8
B 1) 5.dxc5 9
B2) 5.d5?! 0-0 9
B2 1) 6.d6N 10
B22) 6.i.g5N 11
C) 4.i.f4 0-0 5.e3 d5 6.tLl f3 c5 13
C 1) 7.a3 14
C2) 7.dxc5 15

A) after 1 0 . ti:l f3 B l ) after 8.'xc3 C l ) after l l .i.e2

a b c d e f g h a b c d e f g h

1 0 . . d6!N
. 8 . d6N
. .
8 Various 4th Moves

l .d4 f6 2.c4 e6 3. c3 .th4 8 ...5


We start our Nimzo-Indian journey with Killing any hopes of a kingside attack.
three rare and unpromising options: A) 4.e4,
B) 4.Yid3! and C) 4..tf4. 9.Yie2 .t6 10.6
In Roeder - Volpert, Unterfranken 1 987,
A) 4.e4? the most convincing continuation would have
been:
This move is over-ambitious: White has
absolutely no justification for sacrificing the
central pawn. 8

a b c d e f g h
IO...d6!N
a b c d e f g h White cannot stop ...e6-e5, so Black will
have a positional advantage on top of his extra
4 ... xe4 s.Yig4
pawn.
White is obviously pinning his hopes on this
double attack.
1 1.0-0 e5 12.dxe5 dxe5+
White's position should collapse soon.
5.Yic2 is hardly an improvement: after 5 ...ll:lxc3
6.bxc3 ie7 7.ll:lf3 b6 8.ie2 ib7 9.0 -0 0 -0
l O.ge l c5+ White had no compensation for
the pawn in T. Carlsen - Freydl, email 200 7.

s... xc3 6.a3


Even worse is 6.id2?! ll:ld5 7.cxd5 ixd2t
8.tJixd2 0 -0 -+ as in Schoengart - Tonndorf,
Hamburg 200 5.

6....te7!
The most natural and effective.

7.bxc3 0-0 s ..td3


The other attacking attempt, 8.ih6 if6
9.id3, runs into 9... d5 1 O.Yig3 e5! and Black
is winning. a b c d e f g h
Chapter 1 - Rare Options 9

This move was employed by the famous s ... ttla6! 6.a3 ttlxc5 7.Yffc2 .bc3t s.Y!fxc3
Lithuanian player Vladas Mikenas, and rhus We have reached a well-known theoretical
is usually called the Mikenas System. More position from the Classical System, but usually
recenrly, Richard Palliser devoted a chapter it is White's turn to move here! I think Black
to ir in Everyman's Dangerous Weapons: 7he should proceed wirh:
Nimzo-Indian. However, my analysis indicates
that purring the queen in the centre like this is
dangerous only to Whire, and it seems to be
one of his worst 4th move options.

4...c5!
Challenging the centre seems the most
principled reply.

Instead, both 4... d5 and 4...0-0 give White


a chance to transpose to the Classical System
with 5.a3 ixc3t 6.Wfxc3; and in the latter case,
he could go for a modified version of the 5.e4
system wirh the queen on d3 instead of c2.
8... d6N 9.f3
White may react to the text move wirh 9.b4 is hardly an improvement: 9...l2Ja4
Bl) 5.dxc5 or B2) 5.d5?!. 10 .Wlb3 i.d7 1 1 .l2Jf3 :i:l:c8 1 2.ig5 h6 1 3.i.h4
g5 1 4.i.g3 l2Je4 and Black has the initiative.
5.a3 i.xc3t 6.Wfxc3 is covered via the 4.Wlb3 c5
move order - see the note on 5.a3 on page 18. 9...e5 IO.e4 .te6 I I ..te3 gcs 12Jdl bS!:j:
Making full use of the extra tempo. White's
Bl) 5.dxc5 lack of development makes his position
unpleasant.

B2) 5.d5?!

a b c d e f g h
10 Various 4th Moves

Palliser thinks this is White's best move, but


8
entering into a Benoni type of position with
such a misplaced queen invites more trouble. 7

6
5 ...0-0
There have not been many games from this 5
position, so I will focus on Palliser's two main 4
suggestions ofB21) 6.d6N and B22) 6 ..tg5N .
3
6.e4?! i.xc3t 7.bxc3 exd5 8.exd5 d6 9.tLlf3 2
WaS 1 0.i.e2 if5FF highlights the awkward
1
placement ofWhite's queen.
a b c d e f g h
6.i.d2
6 ... b5!
This move is rather passive, and Black easily
I especially like this aggressive option.
builds an initiative by targeting the queen.
A good alternative is:
6 . . . exd5 7.cxd5 d6 8.g3
A desperate attempt to complete the
6 . . . tLl c6 7.tLlf3 b6
development of the kingside pieces. Palliser briefly suggests that 7 . . . e5 deserves
8 . . . b6 9.i.g2 i.a6 1 0.Wfc2 attention, but I prefer the text move.
We have been following the famous game 8.i.g5 h6 9.i.h4 i.b7 1 0.a3
Mikenas - Keres, Moscow 1 949. A simple 1 0.e4 e5 1 l .a3 ixc3t 1 2.bxc3 tLl a5+ leaves
and strong continuation would have been: White with no compensation for his pawn
weaknesses, which can be exploited by
. . . e8-e6 and . . . i.a6.
1 0 . . . ixc3t 1 l .Wfxc3

a b c d e f g h

1 0 . . J:e8N 1 l .a3 i.xc3 1 2.i.xc3 We7 1 3.e3


a b c d f g h
lLle4+ e
White has serious problems.
l l . . .g5! 1 2.lLlxg5
B2 1) 6.d6N 1 2.i.g3 lLle4 1 3 .Wfc2 f5 1 4.e3 Wf6 and the
d6-pawn will soon fall.
Palliser mentions this in passing as an 1 2 . . . hxg5 1 3 .i.xg5 tLle4 1 4.i.xd8 tLlxc3 1 5 .ie7
interesting idea. To me it looks dubious for lLl e4 1 6.f3 fe8 1 7. fxe4 lLlxe7 1 8.dxe7 i.xe4
White, as Black has a significant development Black's superior pawn structure gives him
advantage with many promising continuations. the upper hand in the endgame.
Chapter I - Rare Options ll

7.cxb5 a6

a b c d e f g h
10 ...c4! l l.Vxc4 xc3 12.bxc3 hd6
a b c d e f g h
White has no compensation for the
8.b6 damaged pawn structure, and he is also behind
This seems like the best of a bad bunch for in development.
White.
B22) 6..tg5N
8.bxa6?! i.xa6 9.'1Wc2 ll:l c6 gives Black an
overwhelming initiative, and White may
8
already be objectively lost.
7
8.e4 This advance is the most logical 6
continuation of White's previous ambitious
play, but it simply doesn't work: 8 . . . axb5 9.e5 5
!:_Jd5 4
3
2
1
a b c d e f g h
Palliser gives this move as White's best, but I
am not impressed by it.

6...exd5
a b c d e f g h 6 . . . ixc3t? 7.'1Wxc3 ll:lxd5 does not really
work, as 8.ixd8 ll:lxc3 9.ie7 :ge8 l O.ixc5
l O.ll:lf3
c4 1 1 .'1Wc2 ll:l c6+ Followed by . . .f6
ll:le4 I I .ia3 leaves White with the two
and White's position is going to collapse.
bishops, as Palliser points out.
8...f;Yxb6 9..tg5 d5 10.e4
7.cxd5 d6 8.f3 bd7
12 Various 4th Moves

White has a bad version of the Leningrad 10...ti'b6!


System, as the misplaced queen offers Black Highlighting White's lack of coordination.
lots of attractive options.
l l.gb l
9.a3 .tas 1 1 .0-0-0 is the only way to keep the
9 . . . ixc3t might be even simpler: 1 0.bxc3 material balance for a while, but now the king
( I O.Wfxc3 h6 1 I .ih4 g5 1 2.ig3 ll:l e4 1 3.Wfc l is in danger. 1 1 . . . l::1 e 8! It makes sense to restrict
ll:l df6+) 1 0 . . . h6 l l .ih4 l::1 e 8 1 2.e3 Wla5 the mobility of White's knight before starting
an attack. ( l l . . .l::1 b 8 1 2.ltle4 is not so clear)
1 2.Wfc2

a b c d e f g h

1 3 .ig3 ( 1 3.ltld2 simply loses a pawn after


1 3 . . . ltle5 1 4.Wfc2 ll:lxd5) 1 3 . . . ll:le4 1 4.ltld2 a b c d e f g h
ll:lxg3 1 5 .hxg3 b5 White is clearly in trouble: 1 2 . . . ixc3 1 3 .Wfxc3 ll:le4 1 4.Wfc2 ll:lxg5
Black has a simple plan of . . . c4 and . . . ltlc5, and 1 5 .ltlxg5 ll:l f6+
the pawns on c3 and d5 are weak.
l l ... xd5
This leads to a tactical sequence where Black's
lead in development makes the difference.

12.b4 xc3 13.ti'xc3

a b c d e f g h
10.e3
Palliser ends his analysis here, calling it a
"tough and roughly balanced struggle." I have
already pointed out the favourable comparison
with the Leningrad System, and Black can a b c d e f g h
cause serious problems with:
Chapter 1 - Rare Options 13

13 ... e5! 14.xe5 5.e3


1 4.ll:ld2 cxb4 1 5 .axb4 Wfc7! is a nice tactical 5 . lLl f3 d5 6.e3 transposes.
resource, enabling Black to keep a healthy
extra pawn. s ... ds

14 ...dxe5 1 5.J.e7 ge8 16.hc5 ti'g6 17,gdl


J.c7+
Material is level but White's coordination
remains poor.

C) 4.J.f4

8
7
6
5 a b c d e f g h
4 6.6
3 The alternatives hardly promise more:

2 6.a3 ixc3t 7.bxc3 c5 8.id3 WaS 9.ll:le2 cxd4


1 1 0.exd4 dxc4 l l .i.xc4 ll:l c6 ( l l . . .b6 1 2.ie5
ll:l bd7 1 3.id6 :i:l:e8=) 1 2.0-0 e5 1 3 .ig3 if5
a b c d e f g h
yields Black an excellent position.
This looks like a natural way to develop
- leaving the bishop on its initial square 6.Wfc2 c5 7.a3 ixc3t 8.bxc3 cxd4 9.cxd4
and playing 4.e3 is not to everyone's taste. WfaSt 1 0.Wfd2 Wxd2t l l .'it>xd2 ll:l c6 1 2.cxd5
However, this plan of development has a clear lLlxd5 1 3 .ig3 id7+ White has some trouble
drawback: the bishop turns out to be quite completing his development.
mlnerable on f4 in many lines, and cannot
take part in protecting the queenside pawns. 6 ... c5
In fact, I didn't find a single game with a top Preparing . . . Wfa5 , putting pressure on the
player on the white side. queenside and reminding White that his dark
squared bishop is no longer able to defend that
4... 0-0 part of the board!
I don't see any reason to delay castling.
The main options to consider are Cl) 7.a3 and
After the immediate 4 . . . d5 White might C2) 7.dxc5 .
consider 5.cxd5 ll:lxd5 6.i.d2 0-0 7.ll:lf3 c5
8.ll:lxd5 i.xd2t 9.Wfxd2 Wfxd5 1 0.dxc5 Wfxc5, Mter 7.i.d3 cxd4 8.exd4 dxc4 9.ixc4 ll:ld5
which leads to an equal and somewhat boring 1 0.id2 ll:l c6 1 1 .0-0 ll:l b6 1 2.id3 ll:lxd4
position. 1 3 .ll:lxd4 Wfxd4 1 4.Wfc2 Wfh4 White doesn't
get much for the pawn.
14 Various 4th Moves

7Jk 1 cxd4 8.exd4 was played in Daenen - 9.b3


Potemri, email 20 1 0. (Dubious is: 8.ll:lxd4?! After 9.'1Wc2 cxd4 1 0.exd4 id7 1 l .Ae2
'1We7! 9.ll:lf3 gdg 1 0.Wc2 ll:l c6 The lack of dxc4 1 2.ll:le5 ll:l d5 it's obvious that White can
development causes White definite problems.) hardly profit from having the bishop on f4:
Now the simple: 1 3.id2 '1Wa4=

9 ... cxd4 IO.exd4


After 1 0.id6 dxe3 l l .ixf8 exf2t 1 2.'xf2
ll:l e4t 1 3.'it>e l xf8 1 4.'1Wb4t '1Wxb4 1 5 .cxb4
a5 Black should be better, as he has two clear
pawns for the exchange.

10 ... c6 I I ..te2
l l .cxd5 gives up the file too early: 1 l . . .exd5
1 2.id3 gest 1 3 .ie3 ig4+
a b c d e f g h
l l .i.d3 as in Jennen - Tonceri, Hastings 1 982,
8 . . . b6N 9.i.d3 dxc4 1 0.ixc4 i.b7 would lead allows Black to secure a stable advantage with
to a typical position from the 4.e3 system some neat tactics: 1 1 . . .lLlxd4!N 1 2.ll:lxd4 e5
where Black has gained a couple of tempos. 1 3 .'1Wb4 Wxb4 1 4.axb4 exd4 (less convincing
would be 1 4 . . . exf4 1 5 .c5, with an unclear
Cl) 7.a3 position) 1 5 .cxd4 gest 1 6.'it>d2 dxc4 1 7.i.xc4
ie6+

a b c d e f g h
Spending an important tempo, but at least
White grabs the bishop.

7 ... .bc3t 8.bxc3 a5


Not only attacking the c3-pawn, but also
pinning it, so White cannot improve his pawn l l ... b6N 12.cxd5
chain. Inferior is: 1 2.0-0?! ia6 1 3.ll:ld2 gac8+
Chapter 1 - Rare Options 15

12 ... tvxd5 13.tvxd5 xd5 14.J.d2 J.b7 15.c4 1 0 . . . Wfxd5 1 1 .0-0 i.xc3 1 2.bxc3 i.d7 1 3.c4
ct6 Wff5 1 4.llJd4 Wff6 1 5 .if3 Wfg6
I prefer Black in this endgame, since the
hanging pawns are under pressure.

C2) 7.dxc5

8
7
6
5
a b c d e f g h
4
Black has the better pawn structure and
3
excellent piece play, which fully compensates
2 for White's bishop pair.
1
a b c d e f g h 8
7 ... e4 8Jk l 7
The other way to support c3 is: 6
8.tvc2
This was played in Meyer - Mach, Germany 5
1 99 1 , and can also be well met by: 4
8 ... llJ a6N 9.cxd5
3
9.i.e2 llJ axc5 1 0.0-0?! (the better 1 0.cxd5
tvxd5 would transpose to the line below) 2
1 0 . . . ixc3 1 l .bxc3 f6! 1 2.l::1 fd 1 e5 1 3.E1xd5 1
tves 1 4.i.g3 i.e6+
9 . . . llJ axc5 1 0.i.e2 a b c d e f g h
1 0.dxe6 Axe6 gives Black a serious initiative 8 ... a6!N
for the pawn, with ... if5 and/or . . . Wla5 This way of regaining the pawn seems
coming soon. more attractive - it aims to complete the
development of the minor pieces first.

Less precise is 8 . . . Wla5 9.i.d3 llJxc3 ?! 1 0.bxc3


Axc3t 1 l .'.t>fl and White had a serious
initiative in Jose Abril - Shatko, Sant Boi
20 1 6.

9.cxd5 axeS lO.J.e2


Too greedy is 1 0.dxe6? 'Wxd 1 t 1 l .E1xd 1
i.xe6+, with an enormous lead in development.
16 Various 4th Moves

10 ...Yixd5 Conclusion

This first chapter offers Black an easy start


8
to Nimw-Indian life, with three rare and
7 unchallenging White options. 4.e4? sacrifices
6 a vital central pawn in return for attacking
chances that will never appear. 4.Wfd3?!
5 misplaces the queen on a vulnerable square
4 while blocking White's natural development.
4.if4 is the closest in this chapter to a
3 reasonable line, but the bishop is not ideal on
2 f4, being slightly vulnerable and putting little
pressure on Black.
1
a b c d e f g h
l l .Yixd5
1 1 .0-0 llJxc3 1 2.bxc3 ia3 1 3.Wfxd5 exd5
also reaches an approximately equal position
with mutual pawn weaknesses.

l l ... exd5 12.i.e5


1 2.0-0 llJxc3 1 3.bxc3 ia3 transposes to the
preceding note.

12 ... ttla4 13. axc3 14..lxc3 hc3


1 5.bxc3 .le6=
The arising endgame is equal but far from
dead.
8
7 vz_,.,,-__zz
6
-----J"'m''"'"
5
r'""' zz,z,zz/'''
4

Various 4th Moves 3


2

a b c d e f g h

Variation Index
l.d4 f6 2.c4 e6 3.c3 i.b4 4.Y;Yb3
4...c5
A) s..tgs 19
B) 5.dxc5 20
C) 5.8 c6 24
Cl) 6.a3 24
C2) 6.e3 26
C3) 6.dxc5 27

note to move 5 B) after 1 3 .ie2 C l ) after I O . f3

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2
I
a b c d e f g h a b c d e f g h a b c d e f g h

6 . . .b5!N I L.ia6!?N l O . . . tLl d6!N


18 Various 4th Moves

I .d4 6 2.c4 e6 3.c3 i.b4 4.Yib3 5 . . . 0-0


This natural move is quite tempting. Indeed, 5 . . . b5!? also looks annoying for White.
in comparison to the Classical System, White 6.ltlf3
not only avoids doubled pawns, but also
attacks the opponent's bishop! However, as
grandmaster practice proves, the queen is
much more useful on c2 in the fight over the
central squares than on b3. Still, this system
was successfully employed by many great
players of the past, such as Alekhine, Euwe and
Bogolj ubow. In recent years, it has been tried
from time to time by strong grandmasters such
as Van Wely and Epishin.
a b c d e f g h
8 This position occurred in Mecking -
7 Goncalves, Campinas 20 1 1 . In my opinion,
the most effective way to refute White's risky
6
strategy is:
5 6 . . . b5!N 7.cxb5
4 The greedy 7.dxe6 invites even more trouble:
7 . . . bxc4 8.exf7t :gxf7 9.'1Wxc4 d5 1 0.WI'b3
3 ll:l c6, and Black is almost winning due to the
2 enormous lead in development.
7 . . . exd5 8.a3 ia5 9.e3 ib7 l O.ie2 d6 1 1 .0-0
1
ll:l bd7+
a b c d e f g h Black has full control over the centre.
4 ... c5
A multi-tasking move. Apart from protecting 5.a3 ixc3t 6.WI'xc3
the bishop, the c5-pawn also challenges This position might also arise after 4.WI'c2 c5
White's centre. 5.a3 ixc3t 6.Yixc3 .
6 . . . cxd4 7.Yixd4 ltl c6
White's main continuations are A) S.i.gS, Alas, White must move the queen once
B) 5.dxc5 and C) 5.6.
more. Practice shows that the bishop pair
does not always compensate for such a lack
5 .e3 ll:l c6 6.ll:lf3 is a transposition to variation of development!
C2. 8 .Wfd l 0-0 9.ltlf3 d5 1 0.cxd5
1 O.e3 e5! l l .cxd5, as played in Kotronias -
5 .d5?! Pandavos, Peristeri 1 993, seems even worse.
White has only developed one minor piece, Now Black should have kept the queen on
while the queen might become a target on the board in order to develop the initiative:
b3 . It is not surprising that Black is able 1 1 . . .ll:lxd5!N 1 2.b4 e4 1 3 .ltl d4 ltle5 1 4.ib2
to put strong pressure on White's central ig4 1 5 .Yib3 Wg5
pawns: 1 0 . . . exd5 1 l .e3 ltle4 1 2.ie2
Chapter 2 - 4.Wfb3 19

So far we have been following the game This may be the lesser evil from White's
Golichenko - Laznicka, Pardubice 2009. perspective. Still, giving up the dark-squared
bishop is a clear positional concession, and
Black has excellent prospects here too.
7.e3 b6 8.lLlf3 ib7 9.ie2 tLl c6

a b c d e f g h

Now Black can choose between several


attractive ways of handling the position, but
a b c d e f g h
I prefer:
1 2 . . . if5N 1 3 .0-0 :ge8 1 4.tLld4 tLlxd4 1 0.a3?!
1 5 .WI'xd4 :gc8 Better was 1 0.0-0, but after 10 ... cxd4
The activity of Black's pieces forces White to 1 l .exd4 ixc3 1 2.Wxc3 0-0 Black still gets
be careful. an excellent position with chances to exert
strong pressure on the c4-pawn.
A> s.J.8s We have been following the game Silva -
Galego, Portugal 1 998. Now Black missed
a nice opportunity to exploit the lack of
harmony in the opponent's camp:
1 0 . . . tLl a5!N 1 l .Wfc2 ixc3t 1 2.WI'xc3

a b c d e f g h

a b c d f g h
White's mixture of the Leningrad System e
with the queen on b3 makes a weird impression.
1 2 . . . ixf3! 1 3.ixf3 :gc8 1 4.ie2 cxd4 1 5 .exd4
s ... h6 6 ..th4 tLlxc4 1 6.ixc4 b5 1 7.b3 0-0 1 8 .WI'd3
I also checked: 1 8.0-0 d5 leaves White a pawn down in a
6.ixf6 Wl'xf6 lost position.
1 8 . . . bxc4 1 9.bxc4 d5 20.cxd5 :gfd8+
20 Various 4th Moves

Black brought his queen out to a5, which was


8
not really necessary. I believe Black should
7 choose the following strategy:
6
5
4
3
2
1
a b c d e f g h
This position has occurred in a few games,
but nobody has tried:
a b c d e
6 ... g5!N 7.i.g3 c6
Now the absence of the queen from the d-file 10 ... xc5N l l .c2 e5 12.6 d6;
forces White to lose control over the centre: White is suffering; the passive g3-bishop and
vulnerability of the doubled pawns makes his
8.dxc5 position quite difficult.
8.d5? allows Black to benefit from his
enormous lead in development: 8 . . . exd5 B) 5.dxc5
9.cxd5 ltld4 IO.Wfd l Wa5
8
7
6
5
4
3
2
a b c d e f g h 1
I I.:i:l:c l Wxa2-+ a b c d e f g h
5 ... c6
The text move transposes to a number of I like this natural developing move - the
games. The next few moves are natural and knight eyes the exposed d4-square, and the
obvious: c5-pawn can be collected later.
8 ... e4 9.e3 hc3t 10.bxc3 6 ..tg5
In Agdestein - Hjartarson, Reykjavik 1 996,
Chapter 2 - 4.V;Vb3 21

A popular continuation - White hardly has


any other way of developing the dark-squared
bishop.

6.ll:l f3 ll:l e4 transposes to variation C3.

Weak is: 6.g3?! ll:l e4 7.lLl f3 (7.ig2? ll:l d4!


and White loses material) 7 . . . ixc3t 8.bxc3
lLlxc5 9.V;Vc2 b6 I O.ig2 ib7+ White has
no compensation for his ruined queenside
structure.

6.a3 ll:l d4
a b c d e f g h
This tactical resource enables Black to treble
White's pawns along the c-file. 6 ... h6 7 ..bf6
6 . . . ixc5 7.ll:lf3 0-0 has been played in Leaving the bishop on the board leads to
almost every game, but I would prefer to unfavourable consequences: 7.ih4?! g5 8 .ig3
avoid the quiet character of play that arises. and we have transposed to variation A above;
7.Wa4 this was the actual move order of the Agdestein
7.Wd l ixc3t 8.bxc3 ll:lc6 9.ll:lf3 W!a5 also - Hjartarson game referred to there.
offers Black rich counterplay.
7 . . . ixc3t 8.bxc3 ll:l c6 7 ..ti'xf6
.

I was quite surprised to discover that this


position has arisen in about fifty games! White
has no advantage and it seems to me that even
finding equality will be a challenge for him.

8. f3
The most common.

8.e3 has no real independent value, because


after 8 . . . ixc3t 9.Wxc3 Wxc3t 1 0.bxc3 b6
l l .cxb6 axb6 White has nothing better than
a b c d e f g h 1 2.lLl8, which would transpose to our main
9.f3N line.
Weaker is 9.if4?! ll:l e4 1 0.Wc2, as played in
Goregliad - Salman, Long Island 1 995, and 8J:k l
now the simple 1 0 . . . ll:lxc5N l l .id6 b6+ This move avoids any damage to White's
leaves White with no compensation for the queenside structure, but delaying the
weak queenside pawns. development of the kingside pieces is a high
9 ... b6 l O.ie3 bxc5 l l .ixc5 WaS price to pay.
l l . . .Wc7!? is another interesting option. 8 . . . ixc5 9.e3 b6 l O.llJB
1 2.Wxa5 ll:lxa5 Now Black should put his finger on the
Black has easy play against White's weak above-mentioned drawback of White's 8th
pawns. move by means of:
2.2 Various 4th Moves

a b c d e f g h

1 o . . . lLle5!
A novelty when I first analysed it, but it has a b c d e f g h
since been played. Now Black has a couple of reasonable ways
1 l .i.e2 i.b7 1 2.Wfd 1 of handling the position, but I definitely prefer:
1 2.tLlxe5 Wfxe5 1 3.0-0 Wfg5+ is also excellent
for Black. 8 ....bc3t 9.Wfxc3
1 2 . . . lLlxf3t 1 3.i.xf3 ixf3 1 4.Wfxf3 Wfxf3 9.bxc3 ?! is inferior. My analysis continues:
1 5 .gxf3 9 . . . Wfe7N 1 0.g3 ( l O.Wfb5 is an awkward
attempt to keep the extra pawn, which can be
strongly met by: 1 0 . . . b6 1 l .cxb6 Wfa3! 1 2.Wfb3
axb6+) 1 o . . . Wfxc5 1 1 .ig2 b6

a b c d e f g h

1 5 . . Jk8!N
In Zhou Jianchao - Wei Yi, China 20 1 6,
Black instead forced a drawish double-rook a b c d e f g h

ending with 1 5 . . . ib4 1 6.c;i;>e2 ixc3 1 7J:xc3 This position resembles the Romanishin
c8. The text move is more ambitious. Variation, but White's dynamic play is
1 6. 'it>e2 c;i;>e?+ significantly limited, so after 1 2.tLld2 lLl a5
White will have to work hard to draw this 1 3 .Wfb4 b8 1 4. 0-0 d6 Black can claim a clear
endgame. advantage due to his better pawn structure.

9 ...Wfxc3t 10.bxc3
White is still a pawn up, but his queenside
pawn structure is hideous. Black's best way
forward is:
Chapter 2 - 4.'?9b3 23

the balance, but it is psychologically difficult


for human players to make such a move)
1 5 . . . :i:l:a3 1 6.tLlb l :i:l:a5 and White is in a
difficult position.

a b c d e f g h
10 . b6!
. .

Black allows White to swap off one of his a b c d e f g h


tripled pawns, but look at the benefits: White's
1 4 . . . i.xc4 1 5 . .txc4 :i:l:xc4 1 6.'it>d3 :i:l:a4 1 7.:i:l:hb l
a-pawn becomes a target, and the aS-rook and
:i:l:a6 1 8.tLld2
c8-bishop spring to life.
We have been following the game Gofshtein
l l .cxb6 axb6 - Har Zvi, Israel 1 998. Black could have
Despite his extra pawn, White is undoubtedly maintained a small but clear advantage by
the defending side in the endgame. means of:

8
7
6
5
4
3
2 a b c d e f g h
1 1 8 . . . 'it>d8!N 1 9.lLlc4 c;i;>c?+
a b c d e f g h Avoiding any unnecessary pawn exchanges
on the queenside. White faces a thankless
12.e3
defensive task.
In one game White tried:
1 2.e4
12 ... ci!?e7 13.i.e2
This doesn't change the pleasant (for Black!)
This position arose in the game Z. Varga -
character of the position.
Z. Almasi, Kazincbarcika 2005, when Black
1 2 . . . :i:l:a4 1 3 . .td3 i.a6 1 4.c;i;>d2
could have set his opponent definite problems
I also examined 1 4.tLld2 lLle5 1 5 .i.c2?! (the
by means of:
computer points out that 1 5 .c;i;>e2 maintains
24 Various 4th Moves

This move forces White to clarify the


situation in the centre. We will consider the
minor alternatives Cl) 6.a3 and C2) 6.e3,
followed by the more popular C3) 6.dxc5 .
6.d5?! has been played a few times, but
White is not ready for such ambitious moves,
especially with his queen on a poor square.
6 . . . ltl a5 7.Vflc2 This position occurred in Silva
- Viterbo Ferreira, Matosinhos Lentas 20 I 4,
when Black had no reason to reject 7 . . . ll:lxc4!N
8.dxe6 dxe6 9.e4 ll:l b6.

a b c d e f g h
13 ... i.a6!?N 14.c:bd2
I4.0-0 ltl a5 I 5 .ltld2 :i:l:hc8 I 6.:i:l:fbi ll:lxc4
I 7 .llJ xc4 .ixc4 I 8 . .ixc4 :i:l:xc4 I 9 .:i:l:xb6 :i:l:xc3
20.g3 :i:l:ca3+ results in a rook ending where
White will have to work hard in order to save
half a point.

14 ... a5 1 5 Jhb l gabS!?


I 5 . . . ll:lxc4t I6 . .ixc4 .ixc4 enables White to a b c d e f g h
equalize with I 7.a4.
There are no weaknesses in Black's camp, so
White doesn't get sufficient compensation for
16,gb4 ghc8i
the pawn after I O.a3 .ixc3t I I .Vflxc3 ll:l bd7
Intending . . . .ixc4, with a lasting advantage
I 2.id3 b6+.
due to the superior pawn structure.
CI) 6.a3 Lc3t 7.ti'xc3

It makes no sense to recapture with the pawn:


7.bxc3 0-0 8.ig5 h6 9 . .ixf6 (9 . .ih4?! cxd4
IO.cxd4 WaSH) 9 . . . V!Jxf6 I O.e3 b6+
Chapter 2 - 4.Wfb3 25

7 e4!
... 1 2.dxc5 0-0 1 3.b4 ( 1 3 .WI'c3?! Wl'd8! 1 4.b4
The queen is forced to leave the c3-square, so d4 1 5 .WI'c2 i.e6 1 6.i.b2 lDd5 is too risky for
Black gets the opportunity to disturb White's White) 1 3 . . . lDxb4 1 4.Wfc3 l2J c6 1 5 .WI'xa5
king. lDxa5 Black has at least equal chances in the
queenless position.
8.d3
The following alternatives lead to the loss of
a pawn: 8
7
8.WI'c2?! Wl'a5 t 9.ltJd2 ltJxd2 1 0.i.xd2 l2Jxd4+
6
8.WI'e3?! Wl'a5t 9.l2Jd2 l2Jxd2 1 O.i.xd2 cxd4
1 l .WI'g3 Wl'e5+ 5
4
3
7 2
6 1
5 a b c d e f g h
4 We have been following the game
Ruckschloss - Pushkov, Cappelle-la-Grande
1 995, in which Black retreated the knight to
2 f6. A stronger continuation would have been:
1
IO ... d6!N l l .dxc5
a b c d e f g h 1 1 .cxd5 c4! 1 2.WI'c3 exd5+ is one of the ideas
8 ...a5t 9.d2 d5 behind the previous move.
White still has the bishop pair, but Black's
active piece play fully compensates for that. l l ... xc4 12.e3 0-0
Black's lead in development gives him fine
IO.f3 prospects, for instance:
A better choice for White would be
1 0.cxd5N exd5 1 l .f3 , bur 1 1 . . .l2J f6! still gives
Black fine prospects, for instance:

a b c d e f g h
26 Various 4th Moves

13.<1t>fl 4e5 14.c2 b6 15.b3 a4 In the event of 8 . . . 0-0 9.0-0 a6 White gets
With a useful initiative. the additional opportunity to keep the tension
by means of 1 0.:i:l:d 1 , when 1 0 . . .dxc4 1 1 .Wixc4
C2) 6.e3 Wfe7 1 2.a3 leads to a symmetrical position in
which White's chances are slightly preferable.

a b c d e f g h
a b c d e f g h
White supports his centre, but such a
modest-looking move cannot pose Black any 9.c:x:d5
problems. I also examined an interesting pawn
sacrifice: 9.0-0!? lD aS 1 0.Wic2 dxc4 (weaker
6 ... d5 is 1 0 . . . l2Jxc4?! 1 l .e4 0-0 1 2.gS with the
There are some decent alternatives, but I initiative)
see no reason for deviating from this natural
advance. Now we will consider two ways of
releasing the pressure in the centre.

7.dxc5
7.cxdS exdS opens the path for the c8-
bishop: 8.dxcS 0-0 9.e2 (9.a3 xeS leads
to a normal IQP position except that White's
queen is clearly misplaced on b3, so I prefer
Black) 9 . . . e6!? 1 0.l2Jd4 xeS 1 1 .l2Jxe6 fxe6
1 2.0-0 Wfe7 Black had the more pleasant game
a b c d e f g h
in Karpov - Kramnik, Nice (blindfold rapid)
1 997. 1 l .:i:l:d 1 N ( l l .e4 occurred in Sava - Bondoc,
Bucharest 2002, when 1 1 . . . bSN would have
7 ... hc5 8.e2 a6! called White's compensation into question)
In my opinion this move is the most precise. 1 l . . .Wfc7 It's hard to believe that White's
This prophylactic move is always useful in compensation offers more than equality.
positions with an isolated dS-pawn. At the Play might continue: 1 2.l2Je4 e7 1 3.d2
same time, 9 . . . ltJ aS is now a clear positional bS 1 4.xaS WixaS 1 S .a4 b7 1 6.axbS WixbS
threat. 1 7.xc4 Wib6 1 8.Wia4t c6=
Chapter 2 - 4.Wfb3 27

9 ... exd5 10.0-0 0-0 l l Jdl .te6 7 ..td2


We can see another benefit of the prophylactic We should also consider some minor
. . . a6 move: the b7-pawn is poisoned. alternatives:

7.ie3 Wfa5 8.l::k 1 lLlxc3 9.bxc3 ixc5 1 0.ixc5


8 Wfxc5 was better for Black in Vezzosi - Gast,
7 Switzerland 2002.
6
Also after 7.e3 ixc3t 8.bxc3 tLlxc5 9.Wfc2 0-0
5 1 0.ie2 b6+ White had no compensation for
4 the ugly queenside structure in Markovic -
Kosic, Vojvodina 20 1 0.
3
2 7.Wfc2 looks unnatural - White has no reason
to touch the queen again and again in the
1 opening. Indeed, after 7 . . . ixc3t 8.bxc3 tLlxc5
a b c d e f g h 9.g3 b6 1 0.ig2 ib7 White once again lacks
the activity needed to make up for his rotten
1 2.c d4 pawn structure.
1 2.Wfxb7?? tLl a5-+ explains the above
comment.

12 ...ti'e7 13.tihc6 bxc6 14.ti'a4 .td7


Black was already better in Sandalakis - 8
Rychagov, Paleochora 20 1 5 . 7

C3) 6.dxc5 e4 6
5
4
3
2
1
a b c d e f g h
8 ... 0-0
In most games Black has preferred 8 . . . ixc5 ,
but I see no reason to hurry to recapture the
pawn - the text move seems somewhat more
a b c d e f g h flexible.

I like this aggressive move, which threatens 9.e3


to damage White's structure. It practically Since Black's dark-squared bishop has no
forces 7.id2, but many will dislike giving up opponent, it makes sense for White to keep an
this bishop. eye on the d4-square.
28 Various 4th Moves

The other way to develop is: 9 .g3 ixc5 1 o.ig2 1 0.0-0-0?! is a poor choice: 1 0 . . . b6 1 l .i.e2
a6 ( l O . . . b6!? 1 1 .0-0 i.b7 is also perfectly ( l l .l2J de4 i.e7 transposes to the next note on
playable) 1 1 .0-0 :gb8. Black has an excellent 1 o.l2J de4) 1 1 . . .i.b7+
version of the Hedgehog set-up. Now after
1 2.Wfc2, as was played in Welling - B. Martin, 8
Oakham 1 994, Black should have played: 7

a b c d e f g h

Black was better due his bishop pair and


safer king in Hanauer - Seidman, New York
a b c d e f g h
1 940.
1 2 . . . i.e7N 1 3 .:gfd 1 lDe5 Objectively, the
position is about equal, but Black's play looks Also pointless is: 1 0.l2Jde4 i.e7 1 1 .0-0-0?!
easier from a practical point of view: White ( l l .:gd 1 can be met by 1 l . . .Wfc7, and if
will have to watch out for the . . . b5 break, and 1 2.lDb5 then 12 . . . Wfa5t 1 3.lDec3 d5 1 4.cxd5
the bishop pair is a long-term asset. exd5 offers Black a promising initiative)
1 l . . .b6 1 2.l2Jd6 This position occurred in
Giffard - V. Gurevich, Le Touquet 2002,
when Black could have developed a queenside
initiative by means of:

2
a b c d e f g h
9 . hc5 IO.i.e2
.. a b c d e f g h
The most consistent. White's intentions are 1 2 . . . a6!N 1 3.Wfc2 b5 1 4.c5 Wfc7 1 5.c;i;>b 1 i.b7
clear: he will castle and plonk one or possibly Sooner or later, Black will swap off the strong
both rooks on the d-file, hoping to apply some knight on d6 and press ahead on the queenside.
pressure there. Here are some other possible
continuations: 1 0.:gd 1 has also been played, but I don't see the
sense for White in delaying the development
Chapter 2 - 4.%Vb3 29

of his kingside pieces. A good continuation is in Verlinsky - Romanovsky, Leningrad 1 925,


1 0 . . . f5!? 1 l .g3 b6 1 2.ig2 ib7 1 3 .0-0 as in and now I suggest a natural improvement over
Epishin - Psakhis, Internet (blitz) 2004, when Black's play:
Black should have played:

a b c d e f g h
a b c d e f g h
1 3 . . . c8N 1 4.%Vc2 a6 1 5 .%Vd2 c7+ White
1 3 . . . %Vc8!N 1 4.a3 ltl e5+ With a fine position. suffers from the lack of a constructive plan,
while the pressure along the c-file is rather
annoying for him.
8
7
6
5
4
3
2
1
a b c d e f g h
IO ... f5!
I like this aggressive approach - the e4-square a b c d e f g h
is no longer available to White's knights, so the l l ... b6 12.adl
c5-bishop is secured for a while. Moreover, Black An attempt to cover the long diagonal by
puts White's monarch under some pressure, as means of 1 2.g3 ib7 1 3.if3 led White to an
the . . . f5-f4 advance is potentially dangerous. inferior position after 1 3 . . . %Vc8 1 4.ac l ltle5
in Donner - Bohm, Leeuwarden 1 98 1 .
1 1 .0-0
Just as on the previous move, 1 1 .0-0-0?! 1 2.a3 ib7 1 3.%Vc2 c8 1 4.if3?! (better was
is more dangerous for White than for Black. 1 4.l':iad 1 , but still after 1 4 . . . ie7 1 5 .ltlf3 ltla5
1 l . . .b6 1 2.ltlf3 ib7 1 3.'b 1 (hardly better is Black's position would be preferable) 14 .. .f4!
1 3.d2 %Ve7 1 4.hd 1 ad8+, as was played in gave Black a powerful initiative in Pakleza -
Alster - Piskov, Sofia 1 949) This position arose Wojtaszek, Warsaw (rapid) 2007.
30 Various 4th Moves

12 ....tb7 13.f3
Obviously, there was no better spot for the
d2-knight. This position has occurred in about
twenty games, but Black has only found the
strongest continuation in a couple of them.

a b c d e f g h
1 5 ... g4 16.d4 xd4 17.exd4 gbs! 18.cxb6
axb6 19.c3
1 9.d5 ixd5 20.1:'!:xd5 exd5 2 l .Wfxd5t h8
22 .id3 l':!:a8 doesn't yield White sufficient
compensation for the material losses.
a b c d e f g h
13 ... g5! 19 ....td6!
This aggressive measure is fully justified by Black's powerful bishops and kingside
positional factors: both of Black's bishops put space advantage make his position easier, at
pressure on White's king, while the queen is least from a human point of view. It is quite
offside on b3. important that White has no time to chase
The most popular choice has been 1 3 ... Wfe7, away the bishop:
but it allows White to reduce Black's attacking
potential with 1 4.ll:la4.
8
14.a4N 7
This innovation is definitely the best 6
reaction.

1 4.ll:ld4 Wf6 1 5 .ll:l db5 occurred in Sherwood


- Pijl, email 20 1 3 , when 1 5 . . J:!:ad8!?N 1 6.ll:ld6
ia8 1 7.ll:la4 ixd6 1 8.1:'!:xd6 f4 would have
given Black some initiative.

14 ....te7
In comparison to 1 3 . . . We7, the e7-square is a b c d e f g h
now vacant for this retreat. 20. b5? .tds 2 I ..tc4 hh2t! 22.xh2
ti'h4t 23.<1t>gl g6
1 5.c5 With a decisive attack.
It looks like White will regain his harmony,
but Black retains a fine position after:
Chapter 2 - 4.Y!fb3 31

Conclusion

If you are unfamiliar with the relevant opening


theory, then 4.Y!fb3 is a tempting move: the
queen supports the pinned c3-knight while
attacking the offending bishop. The drawbacks
become clear after we play a couple of moves.
Black replies 4 . . . c5 and generally follows up
with . . . lL! c6, so the b4-bishop is never troubled.
In contrast, White is frequently bothered by
. . . lL!f6-e4 ideas, which is a major reason the
c2-square is a far more popular destination
for the white queen. As we saw in several
variations in this chapter, if White plays .igS
then he often needs to answer . . . h6 with J.x6,
surrendering the bishop pair, as instead .ih4
would allow . . . gS followed by . . . lL!e4, with the
usual problems on c3.
Overall, 4.Y!fb3 cannot offer White more
than equality, and often allows Black the
chance to play for the advantage in interesting
positions.
8
7
6
bm/"'"'///////
5
f""" ,,,_,,J"C"" "'"'""'"" Nmu"""" mud
4

Various 4th Moves


Lon/' ,":;07./-/ n//J/W//
3
f"C'"'"'""'u-
2

a b c d e f g h

4.id2
Variation Index
l.d4 f6 2.c4 e6 3.c3 i.b4 4.i.d2
4...0-0
A) S.'i'c2 33
B) S.a3 34
C) S.e3 36
D) 5.8 c5 38
Dl) 6.a3 38
02) 6.e3 40
03) 6.dxc5 hc5 7.g3 dS 43
03 1) 8.i.g2!?N 44
032) 8.cxd5 45

B) after 1 O.c5 C) after 1 2.'1Wc2 03 1 ) after 9 . 0-0!?N

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2

a b c d e f g h a b c d e f g h a b c d e f g h

1 0 ... e5!N 1 2 ... g6!N 9 ... d4!


Chapter 3 - 4.i.d2 33

l.d4 f6 2.c4 e6 3.c3 i.b4 4.i.d2 8 . . . b5! 9.a4 i.b7! 1 0.axb5 a6 1 1 .bxa6 ltlxa6
This is quite a harmless line. Still, it has 1 2.:1k 1 c5 1 3 .ltlf3 cxd4 1 4.ltlxd4 c8+ With a
been tried by such great players as Petrosian, useful lead in development for Black.
Korchnoi and lvanchuk, and is generally
a sensible option if White wishes to avoid Also too passive would be:
mainstream theory. 6.cxd5?! exd5 7.ltlf3
7.e3 e8 8.ltlf3 was played in Dolezal -
4... 0-0 Weinzetd, Prague 2007, when the simple
Castling immediately is the most flexible 8 . . . c6N 9.id3 i.d6 1 0.0-0 Vf!e7 would have
choice. White has four main options: yielded Black the better position. White's
A) 5.c2, B) 5.a3 , C) 5.e3 and D) 5.f3. bishop is not at all well placed on d2.
And here Black has several good possibilities,
A) 5.c2
but I favour the most active and aggressive:
7 . . . c5!

a b c d e f g h
a b c d e f g h
8.a3N
This mixed set-up looks rather passive, and 8.dxc5 i.xc5 9.e3 ltl c6+ Aleksandrov - Low,
Black gets a nice position by natural play in AI Ain 20 1 3 .
the centre. 8 . . . ixc3 9.i.xc3 c4 1 0.b3
Black is simply better after a slow
s ... d5 6.e3 continuation such as 1 0.e3 ltl c6 1 1 .i.e2
Harmless is: 6.a3 i.xc3 7.ixc3N (7.bxc3 ?! ltle4 1 2.0-0 i.f5+.
b6+ makes even less sense for White, Nikolov The text move {intending to meet 10 ... b5
- Kostov, Sofia 20 1 2) 7 . . . dxc4 8 .e3 with 1 1 .a4) is a reasonable attempt to
undermine our pawn chain, but we can
emphasize White's slow development with:
1 0 . . . ltl e4!
With the following idea:
1 1 .bxc4 i.f5 1 2.Vfib3 dxc4 1 3 .WI'xc4 ltl c6
1 4.e3 c8
Black gets a dangerous initiative for the
sacrificed pawn.

a b c d e f g h
34 Various 4th Moves

This resource allows Black to avoid the


isolated pawn and complete his development.

After 1 2 .. .'1We7 1 3 .cxd5 exd5 1 4.xd5 ie6


1 5 .d 1 ac8 1 6. llJ f3 Black's compensation is
not obvious.

B) 5.a3

8 ... cxd4 9.hd4 c6 IO ..tc3


We are following Forintos - Renman, Eksjo
1 98 1 . Now I like:
a b c d e f g h
Black doesn't mind swapping some minor
pieces, so losing a tempo with this move to
force simplifications is unlikely to yield much
for White.

5 ....txc3 6.i.xc3 e4
Black prepares to eliminate the bishop, and
will set up a comfortable position with . . . d6
and . . . e5 at some point.

7.Yffc2
a b c d e f g h There are a couple of other options to
IO . e4N
..
consider:
I also considered 1 0 . . . d4, but there is no
need to allow 1 1 .0-0-0! e5 1 2.exd4 llJxd4 7.c l d6 8.g3 e5 9.i.g2 llJ xc3 1 0.xc3 gives
1 3 .i.xd4 exd4 1 4.llJf3, when Black must give Black a choice:
up a pawn for uncertain compensation.

ll.dl xc3 12.xc3 e7!


Chapter 3 - 4 .id2 35

a b c d e f g h

1 o . . . exd4 This is the simplest equalizer. (Also


good enough is 1 0 . . . We7!? 1 1 .ltlf3 e4 1 2.ll:ld2 a b c d e f g h
f5 1 3 .0-0 ltld7, with a more complex but 7 ... xc3 8.xc3 d6 9.6 d7 10.c5
also roughly equal position.) 1 1 .Wfxd4 ll:l c6 I O.e3 Wfe?N I l .id3 e5= is harmless.
1 2.Wfd2 a5 1 3 .ltlf3 :ge8 1 4.0-0 if5 = Black
had solved all his problems in Kveinys - Mter 1 0.g3 We7N 1 1 .ig2 e5 1 2.0-0
Kengis, Lubniewice 1 998. (or 1 2.dxe5 dxe5 1 3.0-0 e4 1 4.ll:ld4 ltl f6+)
1 2 . . . e4 1 3.ltld2 f5 Black has a space advantage
7.ib4 is an attempt by White to preserve the and the g2-bishop is blocked, so I do not like
bishop pair, but it costs time and is unlikely to White's position.
succeed anyway. For example: 7 . . . d6 8.Wc2 f5
9.ltlf3 b6 1 0.e3 ( 1 0.g3 ib7 1 1 .ig2 occurred The text move was played in Avshalumov -
in Llopis - Roca Galanza, Spain 1 997, when Huzman, Baku 1 988. White is trying to play
1 1 . . .c5N 1 2.dxc5 bxc5 1 3 .ic3 ltlxc3 1 4.Wxc3 actively, but Black could have highlighted the
a5 1 5 .0-0 a4 1 6.:gfd l :ga6+ would have been uncastled king with the energetic continuation:
excellent for Black.)

a b c d e f g h

I O . . . c5!N ( 1 0 . . . ib7 was played in Hernandez a b c d e f g h


Delgado - Nemutlu, corr. 2006. It is better to IO ... e5!N l l .cxd6
drive the bishop back and get ready to eliminate 1 1 .dxe5 dxc5 ( l l . . .ll:lxc5 1 2.e3 ltl e4 1 3 .Wfc2
it.) I l .ic3 ib7 1 2.ie2 ltld7 1 3.0-0 ll:lxc3 ltlg5 1 4.ll:lxg5 Wxg5 1 5 .exd6 cxd6= is also
I 4.Wxc3 We7= Black has no reason to worry. decent) 1 2.e3 We? 1 3 .:gd l :ge8 gives Black
36 Various 4th Moves

comfortable play due to the weakness of the 5 ... c5


e5-pawn. This advance seems especially well timed: in
the event of a further . . . cxd4 and exd4, Black
l l ... cxd6 12.dxe5 xe5 13.d4 will not have to worry about White's bishop
1 3.llJxe5 dxe5 1 4.WI'xe5 :ge8 1 5 .WI'c3 if5 emerging on an active square like f4 or g5 , as
gives Black a dangerous initiative for the pawn. this will entail the loss of a tempo.

The text move looks like a sensible try to keep 6.d5?!


the centre closed, but Black can exploit his This advance definitely earns White an 'Pl.
development advantage with: for Ambitiousness. It doesn't combine well
with his previous moves though; not only
is the bishop a bit passive on d2, but it also
8
blocks White's defence of the d5-pawn.
7
6 6.llJf3 is the usual move, and it will be covered
under the 5.llJf3 c5 6.e3 move order in
5 variation 02.
4
A minor, harmless alternative is:
3 6.a3 ixc3 7.ixc3 llJe4 s.:gcl d6 9.llJf3 Wl'e7
2 1 0.b4
1 0.id3 ?! is inaccurate in view of 1 0 . . . llJxc3
1
1 l .:gxc3 e5 1 2.dxe5 dxe5 1 3 .ie4 llJ d7 and
a b c d e f g h Black is already slightly better.
13 ...i.e6 14.e3 c8 1 5.ti'd2 ti'f6 16.i.e2
ti'g6! 17.g3 .th3
White has some problems connected with
his inability to castle.

C) 5.e3

a b c d e f g h

1 0 . . . llJxc3N
It's necessary to eliminate the strong bishop.
l O . . . llJ c6?! is dubious: l l .bxc5 dxc5
1 2.ib2i Gurgenidze - Karner, Tbilisi 1 983.
l l .:gxc3 b6 1 2.ie2 ib7 1 3.0-0 llJ d7=
Black has a solid position and may consider
some kingside activity by means of .. .f5 and
a b c d e f g h . . . e5.
Chapter 3 - 4.id2 37

8 ... xd5 9.cxd5 ti'g5 IO.e2


8
White is more or less forced to sacrifice a
7 pawn or two and hope for the best.
6
IO ...ti'xg2 l l .gl ti'xh2 12.ti'c2
5 After deep analysis, I managed to find a
4 significant improvement for Black.
3
2
1
a b c d e f g h
6 ... d6
6 . . . exd5 could certainly be considered as
well. This takes away the option mentioned in
the note to White's next move; on the other
hand, there is something to be said for keeping
the opponent guessing.
a b c d e f g h
7.id3
7.dxe6 ixe6 shows a complete lack of 12 ... g6!N
ambition from White, and after 8.ll:lf3N ll:l c6 1 2 . . . ll:l d7 1 3.0-0-0 ixd2t 1 4.Wfxd2 h6?!
9.ie2 d5 Black is at least equal. Still, at least 1 5 .ll:lf4 Wlh4? 1 6J::! h 1 Wfe7 1 7.l:!dg 1 gave
this would avoid the problems experienced by White a venomous attack in Ulko - Tunik,
White in our main line below. . . Korolev 1 999.

7 ... exd5 8.xd5


I also considered 8.cxd5N, when 8 . . . ll:l bd7
9.lLlge2 lLle5 1 0.ic2 a6 1 1 .a3 ia5 1 2.0-0 b5+
gives Black everything he could have wished
for from the Benoni structure.

a b c d e f g h
Black is now three pawns up. It looks like
White has some attacking chances, but Black

a b c d e f g h
38 Various 4th Moves

has full control over the dark squares, especially The most popular and flexible move. White
the e5-spot for the knight. For instance: delays the development of his bishop, keeping
the fianchetto option in mind.
1 5 Jhl ti'f6 16,gh6 .tg4 17.gdh l d7
18,gxh7 e5 19.g3 .tf3! 5 ... c5
Restricting the mobility of the g3-knight. Although this is only Black's third most
popular option according to the statistics,
20.e4 in my opinion it gives Black more dynamic
No better is: 20J:Bh6 1:'Ub8 2 1 .ie4 f8 chances than 5 . . . b6 or 5 . . . d5.
22.Wff2 tJie7+

a b c d e f g h
20 ... xd3t 2 1 .ti'xd3 he4 22.ti'xe4 gfe8
23.ti'g2 ge5 24.ti'h2 gh5 25.gxh5 gxh5
26.ti'xh5 ges+ Dl) 6.a3 hc3 7.hc3 e4
White's kingside initiative is under control,
and Black has good chances to convert his Once again, Black should liquidate the
extra pawn. powerful dark-squared bishop.
Chapter 3 - 4.id2 39

8.c2 10.e3
The other way to avoid doubled c-pawns is: ib7 l l .dxcS bxcS
In the event of l O .'a.dlN
8 .:S.c l b6 9.g3 ib7 1 0.ig2 d6 1 1 .0--0 lDd7 White is unable to exert meaningful pressure
1 2.Wfc2 along the d-file. An illustrative line is: 1 2.e3
1 2.b3 lDxc3 1 3.:S.xc3 Wfe7 was fine for Black aS 1 3.ie2
in Schaufelberger - Gyimesi, Kerner 2007.
1 2.Wid3 lDxc3 1 3.:S.xc3 has occurred a couple
of times; I suggest 1 3 . . . We7N= with similar
play to the main example below.
1 2 . . . :S.c8 1 3.:S.fd 1 lDxc3 1 4.Wxc3 Wfe7 1 5 .Wfe3
lD f6

a b c d e f g h

1 3 . . . :S.a6! The rook defends d6 laterally and


may eventually be used to exert pressure on the
queenside. 1 4. 0-0 d6 1 5 .:S.d2 tLl d7 1 6.:S.fd l
Wfe7+ White lacks any active ideas.
a b c d e f g h

Having failed to obtain any advantage from 10 ...ib7 l l .ie2 d6 12.0-0 d7 13JUdl
the opening, White played too optimistically e7
in the following example:
1 6.d5?! :S.ce8! 1 7.dxe6 fxe6+ 8
Black managed to seize the initiative in
Dj uric - Tiviakov, Formia 1 995. 7
6
8 ... lthc3 9.xc3 b6
5
Black aims for a familiar set-up with the
bishop on b 7. 4
3
2
1
a b c d e f g h
14J::M2?!
White should prefer something like 1 4.h3N,
or any other neutral move, with equality.
Doubling rooks along the d-file is obviously
something White would like to do, but here
it runs into:
a b c d e f g h
40 Various 4th Moves

14 ... 6! D2) 6.e3


Threatening a fork on e4.
8
1 4 . . . d5 was equal in Ajrapetjan - Harutjunyan,
Alushta 2007, but it would be a shame to miss 7
out on the opportunity that has j ust been 6
presented.
5
4
3
2
1
liL_:::::....<o==
a b c d e f g

6 .. cxd4
.

6 . . . d5 has been played lots of times, but


after 7.a3 i.xc3 8 .i.xc3 Black is j ust playing
for equality without posing any real problems
for his opponent.
a b c d e f g h
7.exd4
I SJ:Mdl
We should also consider the knight recapture:
1 5 .'1Mfc2 e4 1 6Jdd 1 is the same thing.
7.xd4 c6 8.i.e2
8.i.d3 is harmless: 8 . . . d5 9.cxd5 xd4
1 5 . e4 16.ti'c2
1 0.exd4 xd5 1 1 .0-0 f6= Zuberoski -
..

From here Black continued with the overly


Rusomanov, Skopje 1 998.
aggressive 1 6 . . . g5?! in Musialkiewicz -
8.a3 i.e? 9.i.e2 d5 1 0.cxd5 xd4 1 1 .exd4
Zmarzly, Wroclaw 2007, when both players
xd5 gives Black a nice position against the
missed the idea of 1 7.xg5 Wfxg5 1 8.d5!,
IQP; after the further 1 2.i.f3 b6 1 3 .i.e3
when White has no problems after shutting
c4 1 4.0-0 xe3 1 5 .fxe3 i.d7 the pawn
Black's bishop out of the game. (The trick is
structure had changed, but Black still had
that 1 8 . . . exd5 ? runs into 1 9.i.f3!.)
the better chances thanks to the bishop pair
in Lenic - Ponomariov, Khanty-Mansiysk
A better continuation would be:
(ol) 20 1 0.
16 . f5N
..

Keeping . . . g5 and other attacking options


available. The position is close to equal but
clearly more comfortable for Black.

a b c d e
Chapter 3 - 4.id2 41

8 . . .d 5 9.cxd5 lLlxd4 1 0.exd4 lLlxd5 9.0-0 d5 l O.tLlxd5 lLlxd5 1 l .cxd5 occurred


10 ... exd5 has occurred in several games but I in Gasztonyi - Barczay, Budapest 1 965,
see no reason to enter a boring position with when l l . . .ixd2!N 1 2.ixa6 lLlxa6 1 3 .WI'xd2
a symmetrical pawn structure. Wl'xd5+ would have exemplified Black's
1 1 .0-0 lLl f6 1 2.ig5 h6 1 3 .ih4 ie7 1 4.if3 strategy.
All this occurred in Goess - Trockmann, 9 . . . h6 1 0.ih4
Mittelfranken 2007. Now the simple This was Erenberg - Goczo, Budapest 20 1 4 .
approach works well: Now I suggest the natural improvement:

a b c d e f g h a b c d e f g h

1 4 . . . id7N 1 0 . . . d5!N l l .cxd5 ixc3t 1 2.bxc3 ixd3


The b7-pawn is not really en prise, as Black 1 3.Wfxd3 exd5
would be happy to activate his rook and take Black is not worse, for example:
on b2. 1 4.0-0 lLl bd7 1 5 .tLle5 Wl'c7
1 5 .WI'e2 ic6 The pressure along the c-file fully
Black has comfortable play. After the likely compensates for White's centralized knight.
exchange on c6, his isolated pawn on c6 will
not be any weaker than the one on d4.
8
7 . b6
..
7
This move has two ideas. Preparing to 6
develop the bishop on b 7 or a6 is the obvious
one, but I also want to prepare . . . d5 without 5
allowing White to get a strong pawn chain by 4
advancing c4-c5.
3
8.a3 2
The most natural alternative is:
1
8.id3 ia6!?
This move prepares a future exchange of a e g
the light-squared bishops in order to limit 8 .. ie7
.

White's attacking potential and secure I definitely prefer to keep the bishop on the
control over the d5-square in the future. board.
9.ig5
42 Various 4th Moves

9.b4
8
9 ..id3N can be well met by: 9 . . . .ia6! (9 . . . d5?
would be premature in view of 1 0.cxd5 ll:lxd5 7
l l .Wfc2! h6 1 2.ltlxd5 exd5 1 3.0-0;!;) 1 0.Wfe2 6
( I O.b4 d5 1 1 .b5 dxc4 1 2 ..ixc4 .ib?+) 1 0 . . . d5
l l .cxd5 .ixd3 1 2.Wfxd3 ltlxd5+ 5
4
The text move is the only realistic way to
avoid an IQP, as White is now ready to meet 3
. . . d5 with c4-c5 . However, there is an obvious 2
drawback as White loses more time. In Atia -
1
Hussein, AI Ain 2008, a logical continuation
would have been: a b c d e f g h

14.-tbS
14 . .id3 d6+ is pleasant for Black.

I also analysed 1 4.ie3 ?! d5! 1 5 . .ib5 ll:l xb4!


1 6.xb4 d4 1 7.ltlxd4 AxeS 1 8.ltlc6 when
Black has a beautiful idea:

3
a b c d e f g h
9 ... a5!N IOJbl 2
1 0.b5?! is inferior in view of 10 ... d5 1 1 .cxd5
ll:lxd5 1 2 . .id3 .ib7 1 3.0-0 ltl d7+, when
a b c d e f g h
everything is in order for Black.
1 8 . . . .ixb4! 1 9.ltlxb4 ( 1 9.ltlxd8 .ixc3t 20.'e2
10 ... axb4 l l .axb4 c6 12.c5 [20.tJifl .ig4+] 20 . . . a2t 2 1 .tJifl a 1 +)
1 2.b5?! ltl a5 1 3 . .id3 .ib7 1 4.0-0 We?+ 1 9 . . . Wa5 20.Wfa4 Wfxa4 2 1 ..ixa4 .ib7 White
leaves White under pressure on the queenside. is under pressure, since he lacks harmony
between his pieces.
12 ... bxc5 1 3.dxc5 e5!
I like this aggressive idea - Black should be 14 ... ttl d4
aiming to exploit his lead in development. Black could also consider 1 4 . . . e4!?, when
Now White has to play precisely in order to my analysis continues 1 5 .ltl g5 ll:l d4 1 6 . .ic4
avoid trouble. d5 1 7 . .ie3 dxc4 1 8.Wfxd4 Wfxd4 1 9 ..ixd4 .if5
20.0-0 fd8 2 1 . .ie3 d3 22.1':1fc 1 , reaching an
extremely complex endgame.
Chapter 3 - 4.id2 43

The text move is a simple way to ensure at least 7.g3


equal chances. A logical continuation is: Too passive is:
7.e3 d5 8.l:k 1
White is not ideally placed to fight against
the isolated pawn: 8.cxd5 exd5 9.ie2 ll:lc6
1 0.0-0 a6 1 1 ,:gc 1 ia7 and Black was doing
well in Grigoriadis - Sumets, Kavala 20 1 4.
8 . . .'1We7
Over-protecting the c5-bishop, though there
was also nothing wrong with 8 . . . ll:l c6!?.
9.cxd5 exd5 1 0.ie2 ll:l c6 1 1 .0-0 ig4 1 2.ll:la4
id6 1 3 .ic3 :gadS

a b c d e f g h
1 5.xd4 exd4 16.e2 d6 17.xd4 dxc5
18.c6 Y!lc7 19.0-0 id6
Black obviously has nothing to worry about.

03) 6.dxc5

a b c d e f g h

1 4.id4?!
1 4.ll:ld4 id7 with approximate equality.
1 4 . . . ll:le4
Black had a dangerous initiative in the classic
game Petrosian - Portisch, Palma de Mallorca
1 974.

White has also tried:


7.a3 d5 8.cxd5
8.e3 makes a poor impression - White's set
a b c d e f g h up looks rather passive. Black can choose
between several attractive possibilities, but
6 ...hc5 8 . . . ll:lc6 is the most ambitious: 9.b4 id6
6 . . . ll:l a6 is another typical way of regaining
1 0.ll:lb5 ib8 1 1 .cxd5 exd5 1 2.ic3 ll:l e4 and
the pawn, but after 7.e3 lLlxc5 8.ie2 Black
Black was better in Lajthajm - lvanisevic,
will most likely have to swap off his dark
Ulcinj 20 1 4.
squared bishop in an unfavourable situation.
8 . . . exd5 9.b4
For instance, 8 ... a5 9.0-0 b6 1 0.a3 ixc3
Mter 9.ig5N ll:lc6 1 0.e3 d4 1 1 .ixf6 gxf6
1 l .ixc3 ll:l fe4 1 2.ie 1 a4 1 3 .ll:l d4 ia6 1 4.f3
1 2.exd4 ll:lxd4 1 3.ll:lxd4 Wxd4 1 4.Wxd4
ll:l d6 1 5 .l:k 1 was better for White in Mamedov
ixd4 White also doesn't get full equality.
- Goganov, Plovdiv 20 1 2.
44 Various 4th Moves

9 . . . ib6 1 0.ig5 ie6 1 l .e3 h6 1 2.ih4 lt'l c6


8
.i. -
7 --- - %--,Y.rr
6 .:afr
- :1{/ , ;
l{
5
.t.% %..
. . . .

4 /i
-r----%
- - - iir0 =
w-:m
r
- - %--0
3

2 ----%
a b c d e f g h
a b c d e f g h
1 3 .lt'la4?
9.0-0!?N
White is playing with fire!
White plays in the spirit of the Catalan,
1 3 .ib5N a5 1 4.ixc6 bxc6 1 5 .0-0= was
offering a pawn sacrifice in order to free a path
safer.
for the light-squared bishop.
1 3 . . . g5! 1 4.ig3 d4!
The lack of development put White in a
9 ... d4!
critical situation in Varga - Babula, Plovdiv
In the event of 9 . . . dxc4 1 0.a4 e5 1 l .ig5
2003.
( l l .xc4 e7 1 2.ig5 ie6 1 3 .h4 h6
1 4.ixf6 xf6 1 5 .xf6 gxf6 is about equal)
7 ... d5
1 l . . .ie6 1 2.1"lad 1 c8 1 3 .ixf6 gxf6 1 4.lt'lh4
From this, the final branching point of
White gets promising compensation for the
the chapter, I examined the more ambitious
sacrificed pawn.
D3 1) 8.i.g2!?N as well as D32) 8.cxd5 .
10.ttla4 e7 l l .e3
D3 1) 8.g2!?N

This has not been tried here, although it does 8


transpose briefly to another game. 7
8 ... ttlc6 6
8 . . . dxc4 9.0-0 lt'l bd7 1 0.c2 offers White 5
enough compensation.
4
The text move was played in Wallinger - 3
Boehlig, Germany 1 992, and it also transposes
2
to a couple of other games. Each game
continued 9.cxd5 exd5, reaching a position
examined under variation 032 below. a b c d e f g h
However, White could also seriously consider
maintaining the tension with: l l ... dxe3
Chapter 3 - 4.i.d2 45

Too risky would be l l ... d3?! 1 2.i.c3 and it's D32) 8.cxd5 exd5
not easy to protect the advanced pawn. For
instance: 1 2 . . . ll:l e4 1 3 .'\Wb l e5 1 4.:i:l:d l i.f5
1 5 . ltl e l;!;

l l . . .e5!? 1 2.exd4 exd4 1 3 .:i:l:e l Ae6 1 4.:i:l:c l


( 1 4. '1Wb3 'IW d7 gives Black enough counterplay)
1 4 . . . :i:l:c8 1 5 .a3 reaches a highly complex
position. However, it seems to me that White
has the slightly easier game, as the d4-pawn is
liable to become weak soon.

12 ..be3 e5
Black has a solid position without
weaknesses. The game may continue: a b c d e f g h
This allows Black to develop all the pieces
13.c2 c7 14.a3
quite easily.
1 4.ltlc3 Ae6 1 5 .ltlb5 '1Wc8= seems fine for
Black. 9.i.g2 c6 10.0-0 :i:l:e8
If you put this rook back on f8 and White's
dark-squared bishop on c 1 , we would arrive
in one of the main variations of the Tarrasch
Defence. Here Black is essentially a tempo up
on that scenario, as his rook invariably goes to
e8 in that line anyway, whereas White's bishop
accomplishes nothing on d2.

l l .:i:l:cl
l l .ltla4 if8 1 2 .a3 h6 1 3 .:i:l:c l d4 1 4.e3 ig4
gave Black an excellent position in Goy -
Langer, email 20 1 3.

a b c d e f g h l l ... d4 12.a4 i.ffi


14 ... h6
Securing a nice spot for the bishop. 8
7
14 . . . g6!? is also a decent option.
6
1 5.b4 i.e6 16.c5 .bc5 17 ..bc5 :i:l:fd8= 5
White's bishop pair is balanced out by Black's
central control and well-coordinated pieces. 4
3
2
1
a b c d e f g h
46 Various 4th Moves

13.b4 Also level is: 1 3 . . . d3 1 4.exd3 Wfxd3N


The following mmtature demonstrates a (or 14 . . . ltJxb4 1 5 .d4, Stern - Keymer,
typical tactical motif for such a position: Saarbruecken 20 1 5 , 1 5 . . . ltJ c6N 1 6.Ae3 ie6=)
1 3 .i.g5 h6 1 4.Axf6 Wfxf6 1 5 .ltk5 Axc5 1 5 .ltJc5 i.xc5 1 6.bxc5 Ag4=
1 6Jhc5 i.e6 1 7.a3 :gadS 1 S.b4 Ag4

a b c d e f g h 1
"""=--==....

1 9.:gc2?? d3! 20.:gxc6 dxe2 0- 1 Wallinger - a b c d e f g h


Boehlig, Germany 1 992. 14.c5 i.xc5
Now White should aim for equality with:
I also considered 1 3 .e3N dxe3 1 4.ixe3 i.e6
1 5 .ltJ c5 Axc5 1 6.:gxc5 Wfxd 1 1 7.:gxd 1 :gadS 1 5.bxc5N
when White must tread carefully to remain The more ambitious 1 5 .:gxc5?! ltJ e4 1 6.:gh5
equal. The danger is illustrated after: was tried in Stern - Svane, Saarbruecken 20 1 3.
Here Black missed a chance to exploit the
vulnerability of the rook by means of:

a b c d e f g h 2

1 S.ltJd4? ( l S,:gcc l ! is better, when 1 S . . . :gxd 1 t


1 9.:gxd 1 ixa2 20.ltJ d4= enables White to a b c d e f g h
win back the pawn and equalize) 1 S . . . ltJxd4 1 6 . . . h6!N 1 7.a4 Ag4 1 S.:gh4 if5 1 9.b5 axb5
1 9.:gxd4 ( 1 9.ixd4 :ge7 leaves White stuck in 20.axb5 ltJ e7+
an unpleasant pin) 1 9 . . . :gxd4 20.Axd4 :gds
and White has problems on the first rank. The text move doesn't look too inspiring for
White, but it should enable him to keep the
1 3 ... a6 balance. For example:
Chapter 3 - 4 .id2 47

Conclusion

4.id2 is no threat to the Nimzo-Indian, as it


is too passive a development. I recommend
castling in reply, quite often followed by plans
involving . . . c7-c5 . One general idea to note is
that if White plays a2-a3, we usually take on
c3, White recaptures with the bishop, then it
is vital to immediately play . . . ll:\ e4 and take
the bishop. If instead we allow White to keep
the bishop pair then the feeble 4.id2 could
be transformed into a promising try for an
a b c d e f g h advantage.
1 5 ... e4 16.Yfc2 if5 17.h4 x:d2
1 8.Yfxd2 ig4 19JUel Yfd7=
Black is by no means worse.
Various 4th Moves
a b c d e f g h

4.ig5
Variation Index
l.d4 tLlf6 2.c4 e6 3.tLlc3 i.b4 4.i.g5
4...c5
A) 5J3cl 49
B) 5.d5 d6 51
Bl) 6.tLl 3 52
B2) 6.3 54
B3) 6.e3 exd5 7.cxd5 tLlbd7 56
B3 1) 8.i.b5 57
B32) 8.i.d3 'i'a5 9.tLlge2 tLlxd5 10.0-0 hc3 ll.bxc3 c4! 59
B32 1) 12.i.5?! 60
B322) 12.i.c2 0-0 61
B322 1) 13.tLlg3?! 62
B3222) 13.i.h4 63

A) note to 7.ixf6 B I ) after 9 . e4 8322 1 ) after 1 5 .ie7.

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2

9 . . . '1Wb6!N 9 . . . '\WeS!N 1 5 . . Elf7!N


.
Chapter 4 - 4 .ig5 49

l .d4 c!Ll6 2.c4 e6 3.c3 ib4 4 ..tg5 bishop on g5 , which can be hit by either . . . lfl e4
or . . . cxd4. Clearly, the g5-bishop is misplaced
here.

5.lflf3 isn't in the spirit of the Leningrad


System at all. 5 . . . h6 6.ih4? (6.ixf6 is the
lesser evil, although 6 . . . Wxf6 7.e3 cxd4 8.exd4
ixc3t 9.bxc3 b6 1 0.ie2 0-0 was comfortable
for Black in Ye Rongguang - Nisipeanu,
Groningen 1 997) 6 . . . g5 7.ig3 g4 8.lfle5 In
Elstner - Karabalis, Germany 2008, Black
should have continued:

a b c d e f g h
Known as the Leningrad System, this is one
of White's most aggressive ways to tackle the
Nimzo-Indian. White pins the opponent's
knight in order to gain control over the
important e4-square. Obviously, breaking
the pin along the h4-d8 diagonal by means
of . . . ib4-e7 would be a loss of tempo. Even
a b c d e f g h
though 4.ig5 isn't popular any more at the
highest level, some aggressive players such as
Mamedyarov, Korobov and Moiseenko still 8 . . . lfle4!N 9.Wfd3 lflxg3 1 0.hxg3 d6! 1 l .lflxg4
use this weapon from time to time. e5 White has serious problems, for instance:
1 2.dxe5 ixg4 1 3.0-0-0 ixc3 1 4.Wxc3 lfl d7
4 ... c5 1 5 .exd6 Wg5t 1 6.E1d2 0-0-0+
This is the most natural and direct way to
exploit the white bishop's departure from the
queenside.
Black often starts with 4 . . . h6 5 .ih4 before 8
playing 5 . . . c5. However, it is crucial for our
7
repertoire that we refrain from . . . h6 any time
soon! When we get to the main lines, we will 6
see that some of Black's ideas only work because 5
of the exposed bishop on g5 , and would be
completely ineffective if this piece was on h4. 4
3
We will analyse the sideline A) SJcl before
moving on to the normal B) 5.d5 . 2
1
Mter 5 .e3? Wa5 White is already in trouble
a b c d e f g h
due to the unpleasant pin and the vulnerable
50 Various 4th Moves

This is not really in the spirit of the chosen 9 . . .'1Wb6!N


system, but it has been tried by some strong Provoking White's next move.
players, most notably Viktor Korchnoi. 1 0.1:'k2
And now Black can get an improved version
s ... cxd4 6.ti'xd4 c6 of the aforementioned game with:
1 0 . . .'1Wa5! l l .a3
l l .id3? is now strongly met by l l . . . lLl b4!,
8
so White has to waste a tempo on the
7 queenside.
6 1 1 . . . 0-0

5
4
3
2
1
a b c d e f g h
7.i.xf6
This is White's safest approach. a b c d e f g h

1 2.ixf6
7.'1Wh4?! With the bishop still on fl , the piece sacrifice
This is playing with fire, since White's set-up is useless: 1 2./ixh6? gxh6 1 3 .WI'xh6 Wl'h5-+
lacks the necessary coordination. 1 2 . . .ixf6 1 3 .WI'f4 d8 1 4.ie2 d5+
7 . . . 1ie7 8.lLlf3 h6 9.e3 Black has the more comfortable position due
We have been following Gonzalez Vassallo to the bishop pair.
- Contreras, Santiago 2007. In the game,
Black brought his queen to a5 and obtained 7 ...ti'xf6 8.ti'xf6 gxf6
a good position.
However, the following finesse would have 8
been even stronger:
7
6
5
4
3
2
1
a b c d e f g h
a b c d e f g h
9.6
Chapter 4 - 4.ig5 51

9.a3 ie7 1 0.e3N is slightly more precise, of exchanging on c3, followed by setting up
but Black still has excellent play after 1 0 . . . b6. a dark-squared pawn wall with . . . h6, . . . e5
and . . . g5, I will instead be proposing a more
9 ... b6 IO.a3 dynamic approach involving . . . exd5, leading
This position was seen in Ionescu - Benjamin, to a Benoni structure. One way to think of it is
Moscow 1 987. In the game Black took on c3 that we will be playing a so-called Snake Benoni
and made a draw, but I would strongly prefer where, instead of taking the rather convoluted
to keep the dark-squared bishop on the board: route from f8-d6-c7-a5, our bishop has saved
time by going to the queenside using one
tempo instead of three. And as I mentioned
8
previously, we will also look to target the
7 bishop on g5 in some lines.
6
5
4
3
2
1
a b c d e f g h
IO ...ie7!N l l .e3 ib7 12.d2
1 2.ie2 :gg8+ is also good for Black.
a b c d e f g h
12 ... f5 13.ie2 e5
White finds himself in a passive, unpleasant We will analyse the important options of
position. Bl) 6.6. B2) 6.6 and B3) 6.e3, after first
checking a couple of minor alternatives:
B) 5.d5
6.e4?! makes little sense, as after 6 . . . ixc3t
This is White's usual choice, and the only way 7.bxc3 h6 8.ixf6 '1Wxf6+ White has no
to fight for an advantage. compensation for his damaged pawn structure.

5 ... d6 6.'1Wc2
Let me repeat: do not be tempted to touch This is a slightly more interesting sideline,
your h-pawn! The bishop on g5 may be and has been used a few times by GM Nana
White's greatest asset, as it yields an annoying Dzagnidze.
pin, but it's also his greatest liability, as it can 6 . . . exd5 7.cxd5 h6
be tactically vulnerable. On this occasion you are allowed to prod the
bishop.
Before we start analysing any variations, let 8.ih4 ll:l bd7 9.ltlf3 0-0 1 0.ltld2
me say something about the plan I have in This position was seen in Dzagnidze -
store. Rather than the well-known scheme N. Kosintseva, Hangzhou 20 1 1 . In my
52 Various 4th Moves

optmon, the best way to meet White's 1 5 .0-0 ixc3 1 6.bxc3 llJxd5+
somewhat slow method of development is:
Bl) 6. 6

a b c d e f g h

1 0 . . . l::1 e 8N
A natural move. Black is fighting against the
e2-e4 advance and will put strong pressure a b c d e f g h
on the d5-pawn. This has been tried by several strong players
l l .e3 Wia5! - and the number rises again if we count the
A concrete approach. White's next move is games where White has gone for a llJ f3 set
practically forced. up with the moves . . . h6 and ih4 included.
1 2.llJc4 We? 1 3.ie2 Chris Ward advocated this set-up for White in
White has no time to secure the c4-knight: a chapter of Dangerous Weapons: 1he Nimzo
1 3 .a4 llJxd5 1 4.l::1 d l llJ 7b6+ !ndian. The English GM mainly focuses on the
version with . . . h6 and ih4 included; he does
briefly discuss the possibility of Black avoiding
those moves, but concludes that in most cases it
does not make much of a difference. However,
I managed to find a precise sequence where
Black can benefit from the exposed bishop on
g5 - see the main line below for full details.

6 .. exd5 7.cxd5 bd7 8.d2


.

The d2-knight has a couple of useful


a b c d e f g h functions - it keeps an eye on the key
e4-square and also unpins the second knight.
1 3 . . . b5!
This means the knight on c3 now protects d5,
Instead, 13 ... llJ xd5 1 4.E1d l llJ7b6 1 5 .ig3
so the d5-pawn is protected. On the other
would yield White definite compensation
hand, the knight on d2 blocks the queen from
for the pawn.
defending that pawn.
1 4.llJd2 a6
Black has obviously made significant progress
8 .e4 looks more ambitious, but it forces
on the queenside. Now there is more reason
White to exchange the powerful bishop: 8 . . . h6
to grab the d5-pawn after:
9 .ixf6 Wxf6 (9 . . . llJxf6!? also looks good:
Chapter 4 - 4 .ig5 53

1 0.ib5t id7 1 l .ixd7t Wxd7 1 2.Wfc2 0-0 capture the c3-pawn but White will have a lot
1 3.0-0 l::1 fe8+) 1 0.Wfc2 0-0 1 l .ie2 tLle5 of compensation.
1 2. 0-0 tLl g6 Black was better in V. Georgiev
Eames, Hastings 2008. 10.f3
The somewhat more natural 1 O.id3 is
8 ... 0-0 9.e4 strongly met by: 1 o . . . ixc3 1 1 .bxc3
After the modest 9.e3 h6 1 0.ih4 lLl b6
1 l .id3 ixc3 1 2.bxc3 tLl bxd5+ White does
not get sufficient compensation for the pawn.

9.a3, as was played in Petkevich - Gerchikov, Sr


Petersburg 1 997, should be met by 9 . . . ixc3N
1 0.bxc3 Wa5 1 l .c4 b5!, with a powerful
initiative due to the enormous development
advantage.

This position occurred in Kristiansen -


a b c d e f g h
Schandorff, Denmark 2008. In my opinion,
the best way to make use of Black's lead in 1 l . . .Wfe5! By exploiting the loose bishop on
development was: g5 , Black gets a clear positional advantage.
( l l . . .lLlxd5 is also possible, but White is very
much in the game after 1 2.0-0) 1 2.ixf6
lLlxf6+

10 ... ixc3 l l .bxc3

a b c d e f g h
The aforementioned game continued 9 . . . h6
1 0.ih4, reaching a position which is also l l ... xd5!
covered by Ward via a different move order 1 1 . . .We5 is less effective here than in the
with an earlier . . . h6. Both the game and note above, as after 1 2.ih4 Wxc3 White does
Ward's analysis continued 1 O . . . l::1 e 8 1 1 .ie2 not have a bishop hanging on d3, and the
Wa5 1 2.f3 ixc3 1 3.bxc3, when Black can f3-pawn is useful for supporting his centre.
54 Various 4th Moves

On the other hand, capturing the d5-pawn 7 . . . exd5 8.cxd5 0-0 9.e4 is another possible
works better here than in the note above. move order.
After the text move the game goes on and 8.e4 exd5 9.cxd5
the position is still complex, but White will Now Black can immediately benefit from
struggle to j ustify the loss of a pawn and I the lack of a pin on the f6-knight.
strongly prefer Black. Play might continue: 9 . . . ll:lh5!
Threatening a check on h4 while preparing
1 Vc4 xc3 13.ti'xd6 b5 14.e3 gbsi to put pressure on White's centre with . . . f5 .

B2) 6.f3

a b c d e f g h

1 0.g3
1 0 .ll:lge2 f5 1 1 .WI'c2 fxe4 1 2.fxe4, as played
in Parker - Laurier, Mondariz 2000, can be
a b c d e f g h strongly met by 1 2 . . . ig4!N 1 3.0-0-0 lLld7+.
This move strikes me as over-ambitious. 1 0 . . . f5
Mixing the 4.f3 and 4.i.g5 systems is a risky Black's play is very natural and consistent.
approach, but it has been tried by many 1 1 .ig2
strong grandmasters, including Epishin, 1 1 .i.d3 ll:l d7+ was even worse for White in
Azmaiparashvili, Bauer and others. Berhhorst - King, Hamburg 1 98 5 .
1 1 . ..ll:lf6 1 2.exf5
6 ... h6! Hardly better is 1 2.ll:lge2 fxe4 1 3 .fxe4 ll:l bd7
In this variation there is not much to be 1 4.0-0 ll:le5. Having occupied an important
gained from leaving the bishop on g5 . On the central square, Black keeps a stable edge.
other hand, as we will see, it is worth forcing the 1 2 . . . i.xf5 1 3.ll:lge2
bishop to declare its intentions, as each possible
retreat has some sort of drawback for White.

7 . .th4
Occasionally White has tried:
7.id2
This retreat makes a weird impression and
doesn't suit White's aggressive intentions in
this line.
7 . . . 0-0
a b c d e f g h
Chapter 4 - 4.ig5 55

We have been following the game Pedersen


- Ostergaard, Aarhus 1 99 1 . Black had no
reason to deviate from the most natural way
to develop:
1 3 . . . ll:l bd7N 1 4.0-0 ll:le5 1 5 .a3 ia5
Black has the initiative.

7 ... exd5 8.cxd5 0-0 9.e4 bd7


In this position the h4-bishop is clearly
misplaced: the g1 -a7 diagonal is exposed, so
White would prefer to have the bishop on e3 .

IO.a3 a b c d e f g h
White can attempt to complete his IO ...J.xc3t!?N
development in a couple of other ways: 1 O . . . i.a5 is also a decent option. For instance,
1 1 .i.d3N a6 1 2.ll:lge2 b5 1 3 .0-0 ll:le5 with a
1 0.ll:lh3 ll:le5 1 1 .lLlf2 allows 1 1 . . . ll:lg6 1 2.i.g3 good game for Black.
ll:l h5+ when Black eliminates the important
dark-squared bishop and gains the upper l l .bxc3 e8 12.ie2 a6!
hand, as in V. Toth - Najdorf, Mar del Plata This is not so much intended to prepare . . . b5
1 956. (although that might be a useful option at
some stage) ; but rather to prevent White from
1 O.i.d3 lLle5 1 1 .i.c2 (After 1 1 .ll:lge2 Black was utilizing the b5-square.
able to exploit the above-mentioned drawback
of putting the bishop on h4: 1 1 . . . c4! 1 2.i.c2 The immediate 1 2 . . . ll:le5 allows 1 2.i.b5
i.c5 and White got into trouble in Mohandesi followed by ll:l e2, when White manages to get
- Barsov, Leuven 2002.) 1 1 . . . c4 1 2.i.f2 his kingside in order.

8
7
6
5
4
3

a b c d e f g h
2

We have been following the game Happel - 1


Chabanon, Saint Affrique 1 995. Now Black a b c d e f g h
missed a nice chance to develop an initiative
13.h3
on the kingside: 1 2 . . . lLl h5!N 1 3 .ll:lge2 '1Wg5
After 1 3.a4?! lLle5 White has serious
1 4. lLlg3 lLl f4 1 5 .0-0 i.xc3 1 6.bxc3 h5+
problems with completing his development.
56 Various 4th Moves

13 ... e5 14.fl g6 1 5 ..tg3 h5 16.0-0


f5
Black can eliminate the g3-bishop whenever
he feels like it, and he has excellent prospects
in the centre and on the kingside.

B3) 6.e3

8
7 a b c d e f g h

6 1 0 . . . 0-0 1 l .f3 E1e8


Now White faces serious problems with
5
completing development.
4 1 2.'1Wd2 c4 1 3 .i.xf6
3 The immediate 1 3.e4? is nicely refuted by
1 3 . . . llJxe4! 1 4. fxe4 llJ c5 1 5 .0-0-0 ( 1 5.llJg3
2 llJxe4 1 6.llJxe4 E1xe4t-+) 1 5 . . . E1xe4 1 6.ig3
1 if5-+, with a decisive attack.
1 3 . . . llJxf6 1 4.e4
a b c d e f g h
We have been following the game Milov -
This is the most common and flexible Pelletier, Switzerland 20 1 4. Black should
continuation. White has a few reasonable ways have played:
to development his kingside pieces from here.

6 ... exd5 7 .cxd5 bd7


A flexible continuation. The f6-knight is
protected now, so Black's queen is free to
move. Now White is at a crossroads, with
B3 1) s .tb5 being the main alternative to
.

B32) s . .td3.

8.llJf3?! '1Wa5+ immediately puts White in


nasty pin trouble; Black has scored a perfect
a b c d e f g h
616 from this position.
1 4 . . . i.d7N 1 5 .llJf4 '1Wc5 1 6.ie2 b5+
White can hardly spend another tempo for a
prophylactic move like: 8.llJge2?!
8 .i.h4?! i.xc3t! 9.bxc3 Wa5 1 0 .llJe2 This looks awkward, as the fl -bishop is
Other moves lead to the loss of a pawn: blocked now.
1 O.'c2 llJxd5+ 8 . . . 0-0 9.a3 ia5 1 0.E1b 1
Or 1 0.'1Wb3 llJxd5 1 1 .l::1 c 1 llJ 5f6+. I was quite surprised to discover that this
position has occurred in no fewer than seven
games! White is trying to get rid of the
Chapter 4 - 4 . .ig5 57

unpleasant pin by means of b2-b4, but it B3 1) 8 .ib5


takes a long time.


8
7
6
5
4
3
2
a b c d e f g h 1
1 0 . . . h6 1 1 .ih4 llJe5 1 2.b4 ib6
The c3-knight is unpinned, but White's This is a challenging line which demands
problems with development are not solved. attention. In comparison to the main line
1 3.c l with the bishop on d3, the dS-pawn remains
1 3. llJ c l ifS 1 4.b3 c8+ Black was much protected. On the other hand, the b5-bishop
better in Axelrod - Nakar, Acre 20 1 3. is quite vulnerable.
We have been following Conquest - Emms,
Oakham 1 994, when Black could have 8 h6 9 .ih4 .ixc3t
. . .

obtained a huge advantage with: Also perfectly playable is: 9 . . . a6! ? l O.ixd7t
ixd7 1 l .llJge2 0-0 1 2.0-0 e8 1 3.'1Wc2 Vfle7
1 4.';t>h 1 White has scored surprisingly well
from this position, but this in no way reflects
the outcome of the opening, especially after:

a b c d e f g h

1 3 . . . cxb4!N 1 4.axb4 aS! 1 5 .b5 a4


Gaining access to the aS-e 1 diagonal. White's
position already seems lost, for instance:
1 6.llJxa4 xa4! 1 7.'1Wxa4 llJd3t 1 8.tJid2 if5-+ a b c d e f g h

1 4 . . . g5! 1 5 .ig3 llJhS and Black's posmon


was much more pleasant in L.C. Schmidt -
Schatzberg, Germany 1 994.

10.bxc3
58 Various 4th Moves

a b c d e f g h a b c d e f g h
10 ... 0-0! l l ... b6!
A natural improvement over Kramnik's play. It turns out that there is no way to protect
This move has only been played in a couple the d5-pawn.
of email games, never over the board. Black's
development advantage should tell in all cases 12.0-0
now. Even giving up the dark-squared bishop
doesn't save the pawn: 1 2.ixf6 Wl'xf6 1 3 .WI'd2
1 0 . . . WI'a5 1 1 .ixd7t lDxd7 was also fine for Wl'e5 1 4.gd 1 a6 1 5 .id3 l2Jxd5+
Black in Korobov - Kramnik, Tromso 20 1 3,
but the text move seems more ambitious. 1 2 ... bxd5 13 . .td3
We have been following the game Vodicka
1 1 .e2 - Larwinski, email 20 1 2. Now I suggest the
Wasting one more tempo on the light natural innovation:
squared bishop with 1 1 .id3 looks dubious.
Mter 1 1 . . .Wa5 (or 1 1 . . .WI'c7!? 1 2.c4 b5! 8
1 3 .cxb5 a6 with the initiative) 1 2.lDe2 lDxd5+
White doesn't get any compensation for the 7
pawn. 6
5
I also examined 1 Uk 1 l2J b6 1 2.c4, but it
leaves White undeveloped, so Black develops a 4
powerful initiative after 1 2 . . . a6 1 3 .ia4 lDxc4! 3
1 4Jhc4 b5.
2
The text move is the most natural continuation,
and was played in both games. However,
Black can immediately exploit the awkward a b c d e f g h
placement of the b5-bishop by means of: 13 ... .tg4!N
Indeed, why not develop a piece with tempo?

14Jcl
Chapter 4 - 4.ig5 59

1 4.f3 leads to an inferior endgame: The only drawback of this set-up is that the
14 . . . lLlxe3 1 5 .ixf6 Wfxf6 1 6. fxg4 tLlxfl d5-pawn remains unprotected.
1 7.Wfxfl Wfxfl t 1 8.'it>xfl d5+
8 ..a5
.

14 .. Je8 1 5.h3 The most popular and principled reply, with


No better is: 1 5 .ib5 id7 1 6.ixd7 Wfxd7 a double attack on c3 and d5.
1 7.c4 l::1 e4!+ 8 ... tLle5 is another perfectly playable option,
but it leads to a long strategic battle a Ia the
Benoni, and is not to everyone's taste.
8
7 9.ge2 xd5 10.0-0
6 This position is quite important for the
whole line with 4.ig5 . The d5-pawn is gone,
5 so what does White have in return? Well, the
4 answer is: quite a lot! The d-file is open now,
so the d6-pawn is under pressure. Apart from
3
this, Black's queenside pieces are undeveloped
2 and the king is still in the centre. Luckily for
1 Black, there is a nice tactical resource:

a b c d e f g h
1 5 ... id7 16.c4 b6 17.c3 ic6
Black has successfully completed his
development and doesn't have any weaknesses,
so White's compensation is questionable.

B32) s.id3

a b c d e f g h
IO .ixc3 l l .bxc3 c4!
..

The point! The vulnerability of the g5-bishop


enables Black to regain coordination. Now we
see how it important it is not to hurry to attack
the bishop with . . . h6 earlier in the opening.

Instead, l l . .. tLl 5f6?! leaves Black with the


a b c d e f g h above-mentioned problems. In the following
This is by far the most popular move. There game White was able to develop a powerful
is no better spot for the light-squared bishop, initiative: 1 2.c4 h6 1 3.ih4 0-0 1 4.E1c l Wfc7
while the e2-square is now vacant for the knight. 1 5 .tLlc3 Baron - Riazantsev, Eilat 20 1 2.
60 Various 4th Moves

We will analyse B32 1) 12.i.f5?! and compensation for the pawn in Ovetchkin -
B322) 12.i.c2, after first considering a couple Lysyj , St Petersburg 20 1 2.
of minor lines:
B321) 12.i.f5?!
1 2.i.xc4? drops a piece after 1 2 . . . ltJ 5b6, but
it's worth continuing the analysis j ust a little
8
further: 1 3.f4!?N ( 1 3.i.xf7t c;i;>xf7 gave White
no real hope in the two existing games) 7
1 3 . . . l2Jxc4 6
5
4
3
2
1
a b c d e f g h
On a5 the queen keeps an eye on both
a b c d e f g h
bishops, so this is definitely amongst White's
worst possibilities, even though it was once
1 4.W/d4 The g7-pawn hangs, but Black can successfully employed by Jan Timman against
still secure a big advantage by returning some Anatoly Karpov.
of his extra material with 1 4 . . . l2J de5! 1 5 .fxe5
i.e6+. 12 .. 0-0!
.

1 2 . . . f6?! was played in the above-mentioned


1 2.i.e4 allows Black to liquidate one of game, but I do not see any reason to weaken
White's strong bishops: 1 2 . . . ltJ 5f6 1 3 .i.xf6 our light squares while driving White's bishop
lDxf6 1 4.i.f3 d5 to a safer square.

13.e4
The following line nicely illustrates the
vulnerability ofWhite's bishops: 1 3.Wfc2 ltJ 7f6
1 4.i.xc8 axc8 1 5 .ad l

a b c d e f g h

1 5 .Wd4 ( 1 5 . l2J f4N may be better; still, after


1 5 . . . i.e6 1 6.Wfd4 0-0 1 7.a4 b6 1 8.fb 1
Wc5 Black is well out of danger in the
endgame) 1 5 . . . 0-0+ White had insufficient
a b c d e f g h
Chapter 4 - 4.ig5 61

1 5 . . . ll:lxc3! 1 6.ixf6 ll:lxe2t 1 7.Wixe2 gxf6 8322) 12.ic2


1 8.Wif3 WleS+

Even worse is 1 3.Wid2 ll:lxc3 1 4.ll:lg3 , as was


played in Yuneev - Dautov, Daugavpils 1 989.

a b c d e f g h

a b c d e f g h
This is the best retreat.
Now the best way to limit White's attacking 12 ... 0-0
potential was: 1 4 . . . ll:l c5!N 1 5 .ixc8 ( 1 5 .ic2 Having inserted the . . . c4 move to protect the
l::1 e8 1 6.if4 d5-+) 1 5 . . . E1axc8 1 6.E1ac l lLl 5a4 knight on d5, Black gets time to complete his
1 7.e4 E1fe8 1 8.E1fe 1 b6-+ development.
13 ... xc3 14.xc3 xc3 Clearly inferior is 1 2 . . . ll:lxc3 ?! 1 3 .ll:lxc3 Wlxc3
(after 1 3 . . . Wixg5 ? 1 4.ll:le4 White would restore
the material balance while keeping the black
king in the centre) 1 4.Wixd6 Wle5 1 5 .if4
Wlxd6 1 6.ixd6 ll:lf6 1 7.l::1 ab l ! b6 1 8 .l::1 fd 1 ie6
1 9.ia3 White had a powerful initiative for a
mere pawn in Ulko - Lukjanenko, Voronezh
2005.

White has tried several moves here, but by far


the most popular have been 8322 1) 13.g3?!
and 83222) 13.ih4.

We should also check the following alternatives:


a b c d e f g h
1 5 .E1cl Dubious is: 1 3 .ll:ld4? lLlxc3 1 4.Wih5 ll:l f6
Hardly better was 1 5 . WIxd6 WI e5 1 6.if4 (possibly even stronger was 1 4 . . . f5!? 1 5 .ie7
Wlxd6 1 7.ixd6 l::1 e 8+. ll:l f6 1 6.Wih4 E1f7 1 7.ixd6 id7+)

15 ...e5 16Jxc4 b6+


Jacob - Luther, Austria 2005. Black has
simplified matters and keeps a solid extra pawn.
62 Various 4th Moves

a b c d e f g h a b c d e f g h

1 5 .'1Wh4 ll:l ce4! 1 6 . .ixe4 ll:lxe4 1 7.ie7 :ge8 1 5 . . . ltl c5!N 1 6.:gfd 1
1 8.'1Wxe4 Wfe5! Black won the piece back 1 6.ltle4 is well met by 1 6 . . . Wfe5 1 7.:gfd 1 b5!
and kept an extra pawn in Bareev - Dautov, with the idea 1 8.Wfxb5 if5+.
Moscow 1 990. 1 6 . . . .ie6 1 7.Wfe2
1 7.Wld4 b6 1 8.ll:le4 Wfg6+
In the more recent game Ratkovic - Perunovic, 1 7 . . . :gfd8 1 8.:gd4 Wfe7 1 9.:gad 1 h6+
Serbia 20 1 5 , White tried the fresh idea Despite the activity of White's pieces, he
1 3 .:gb 1 ?!, but it is unlikely to attract many does not have full compensation for the
followers. Black should have continued: pawn.

B322 1) 13.g3?!

8
7

a b c d e f g h 3
1 3 . . . ltl7b6!N 1 4.W/d4 ltlxc3 1 5 .ltlxc3 Wfxg5+ 2
1
1 3 .Wfd4 ltlxc3!
Mter 1 3 . . . ltl 7b6 1 4.i.h4 i.d7 1 5 .a4 White a b c d e f g h
has no more than sufficient compensation This has been played by many good players,
for the pawn, but Black's pieces lack but it leads White down a forcing path to a
breathing room. clearly inferior position after:
1 4.ll:lxc3 Wfxg5 1 5 .Wfxc4
We have been following the game Dekker 13 ... xc3 14.Yih5 5! 1 5 ..ie7
- Van der Werf, Dieren 1 99 1 . I suggest the Even worse is 1 5 .e4? ll:lxe4 1 6.:gad 1 ll:l df6-+
following way of handling the position: as in V. Georgiev - Leko, Istanbul (ol) 20 1 2.
The text move was played in A. Mastrovasilis
Chapter 4 - 4.ig5 63

- Wojtaszek, Gibraltar 20 1 3 , and could be B3222) 1 3.ih4


strongly met by:
This has been the most popular choice, though
it's a pity to waste a tempo in such a dynamic
position. However, the lines above clearly
illustrate the need for White to remove the
bishop from the black queen's deadly gaze.

a b c d e f g h
1 5 .. JH7!N 16.xf5
1 6.ixd6 lLle5 1 7.ixf5 ixf5 1 8.lLlxf5 lLl f3t
1 9.Wxf3 Wxf5+ leads to the same.
a b c d e f g h
16 ... e5 17.hd6
No better is: 1 7.lLlxg7 lLlf3t 1 8.'hl Wxh5 13 ... xc3 14.xc3
1 9.tLlxh5 :gxe7 20.gxf3 b5+ The following little tactic leads White to a
clearly inferior position: 1 4.ixh7t ?! xh7
17 ...hfs lS.ix5 1 5 .Wc2t tJig8 1 6.lLlxc3 tLle5 1 7.:gab l (hardly
better is 1 7.ie7 :ge8 1 8.ixd6 tLl d3+, with
a similar assessment to the line below) This
position occurred in Chirila - Hernandez
Carmenates, Houston 20 1 3 . A natural
improvement over Black's play would be:

a b c d e f g h
1 8 ... f3t! 19.ti'xf3 ti'x5 20.ti'xf5 e2t
2 I .thl gx5+ a b c d e f g h
Despite the material balance, White finds
himself in a very difficult position: Black's 1 7 . . . a6!N 1 8.ie7 :ge8 1 9.ixd6 lLl d3+ Even
queenside passers are too strong. though there is material balance on the
64 Various 4th Moves

board, Black's position is much better due to with no real compensation for the pawn. For
a strong d3-knight, which is supported by the instance: 1 8.e4 ( 1 8 .E1e 1 also fails to impress
queenside passers. after 1 8 . . . b5 1 9.Wff3 lLl b6 20.l::1 cd 1 Wlg4+)

14 Y;Yxc3

a b c d e f g h

1 8 . . . lLle5 (also good enough is 1 8 . . . b5!? 1 9.f4


i.b7 20.e5 Wfd5 2 1 .Wlg4 lLl c5+) 1 9.f4 lLl d3
20.i.xd3 cxd3 2 1 .Wfxd3 Wlxe4+ Despite the
a b c d e f g h opposite-coloured bishops, White was unable
I SJcl to save the game in Hujbert - Andreev,
Temporarily Black is two pawns up, but Budapest 20 1 3.
the activity of White's pieces shouldn't be
underestimated. 8
7
After 1 5 .i.e7 I like 1 5 . . . d5! ( 1 5 . . . l::1 e 8 1 6.i.xd6
lLl f6 is also OK for Black) 1 6.E1cl Wfe5 1 7.i.xf8 6
lLlxfB+. Black has two pawns for the exchange, 5
while White's heavy pieces are quite passive.
4
1 5 ...Ve5 3
There were several alternatives, but I believe
keeping the queen in the centre is the most 2
natural idea. 1
a b c d e f g h
16 ..tg3
1 6.e4 is an attempt to extend the kingside 17 ...Vxd6 1 8.bd6 ges 19Jfdl
initiative by advancing the pawns, but it fails White could have won the pawn back by
to achieve the goal after: 1 6 . . . Wfe6 1 7. f4 b5 means of 1 9.i.d 1 N, but after 1 9 . . . lLl e5 20.i.xe5
1 8.f5 Wfh6 1 9.i.e7 l::1 e 8 20.i.xd6 i.b7 2 1 .Wld4 E1xe5 2 1 .E1xc4 i.e6 Black is completely fine.
Wff6+
We have been following the game Martinovic
16 ...f;Ye6 17.Vxd6 - Stevie, Croatia 20 1 3 . Now I suggest the
Liquidating into an endgame looks like a following natural improvement over Black's
concession, but 1 7.i.xd6 l::1 e 8 leaves White play:
Chapter 4 - 4 .ig5 65

Conclusion
8
7 4.ig5 is an ambitious and aggressive approach
from White, but the bishop's abandonment
6
of the queenside gives Black ample counter
5 chances. The logical reply is 4 . . . c5 when s .:gcl
4 is rather harmless, so 5 .d5 is the critical line.
After 5 . . . d6 there are various options but one
3 tough test is 6.e3 exd5 7.cxd5 l2J bd7 8.id3 .
2 Black does indeed have a good antidote in
8 . . . WI'a5 9.l2Jge2 l2Jxd5 1 0.0-0 ixc3 l l .bxc3
1
c4! but Black should pay close attention to this
a b c d e f g h line and all the details I give about it. Black has
19 .. f6!N 20.ia4 d8 2 1 .ib5
.
no theoretical problems but if this variation
After 2 l .:gxc4 White's lack of harmony leads caught you unawares over the board, then you
to the loss of a pawn again: 2 l . . .ie6 22.:gcd4 might have a difficult job finding the answers.
(22.:gc2?? l2Je4-+) 22 . . . ixa2+ Doing your homework in advance will be
rewarded against 4.ig5 .
2 1 ... e4!
The most ambitious way of handling the
endgame.

Instead, 2 l . . .ie6 22.ixc4 :gac8 23.ixe6 fxe6


leads to equality.

a b c d e f g h
22.ie7 xdl t 23.xdl ie6 24,gd4 6
25.hc4 c8 26.ifl d5
Black's pieces are much more active, though
White should be able to hold with careful
defence.
Various 4th Moves
a b c d e f g h

4.f3
Variation Index
l.d4 6 2.c4 e6 3.c3 J\b4 4.3
4...c5
A) 5.a3 hc3t 6.bxc3 c6 7.e4 d6 67
Al) 8.e2 67
A2) 8.J\e3 68
B) 5.d5 d6 6.e4 b5 69
Bl) 7.cxb5 70
B2) 7.J\d2 70
B3) 7.a3 71
B4) 7.J\g5 72
B5) 7.ge2 74
B6) 7.J\d3!? 75
B7) 7.dxe6 he6 8.J\f4 0-0! 76
B7 1) 9Jfxd6 77
B72) 9.hd6 78

82) after 9 . a3 83) after 9 . tLlge2 87) after 8 .if4

8 8 8
1 1 1
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2

a b c d e f g h a b c d e f g h a b c d e f g h

9 tt:\e5!N
. . . 9 . . exd5N
. 8 0-0!N
. . .
Chapter 5 - 4.f3 67

l .d4 6 2.c4 e6 3.c3 .tb4 4.6 A) 5.a3 hc3t 6.bxc3


This move is one of White's most aggressive
4th move options. It was introduced into This takes the game along Samisch paths
high-level practice by Grandmaster Efim (see Chapter 6) , but the version that arises is
Bogoljubow in 1 93 1 , but it gained popularity comfortable for Black.
in the 1 950s mainly due to the efforts of
Mikhail Tal and Viktor Korchnoi. Black has 6 ... c6 7.e4 d6
a wide choice of possibilities, many of which Compared with the Classical Samisch, Black
lead to complex and irrational positions. puts strong pressure on the d4-pawn, so White
Among the top modern players who frequently has no time for his preferred set-up with i.d3
employ this move are Anand, Nakamura, and l2J e2.
Mamedyarov and Shirov.

4 ... c5
4 . . . d5 has been the most popular choice, and
4 . . . 0-0 is another big move; both of these have
certain points in their favour, but I found the
text move the most attractive of all.

My recommended move is arguably the most


ambitious reply to 4.f3 . Black doesn't mind
entering a Benoni-type position which offers
White a pleasant space advantage. In return,
Black aims to challenge the opponent's
centre by means of . . . b7-b5 or . . . f7-f5 , taking a b c d e f g h
advantage of his lead in development. The two main options are AI) 8.e2 or
A2) s . .te3.

After the more ambitious 8.i.g5 h6 9.i.h4


cxd4 1 0.cxd4 Wfa5 t 1 1 .f2 Wfd8!+ White
would suffer from a lack of harmony. We will
see a similar approach in the 4.e3 0-0 5.a3
line!

AI) 8.e2 b6 9 ..tg5

After 9.lDg3 0-0 1 0.d5 lDa5 1 1 .i.d3 i.a6


1 2.We2 ltJ d?+ White was doomed to passive
defence in Zakharov - Karpov, Moscow 1 976.
a b c d e f g h
We will analyse two options, with A) 5.a3 9 ... h6 IO .th4

being the only adequate alternative to the After 1 O.ie3 , as was played in Gutman -
standard B) 5.d5. Spassky, Germany 1 986, Black had no reason
to deviate from the most natural 1 0 . . . lDa5N
68 Various 4th Moves

1 1 .llJg3 ia6 1 2.id3 cxd4 1 3.cxd4 l::1 c 8 A sort of concession - instead of pinning
1 4.l::1 c 1 0-0 1 5 .We2 We?, forcing 1 6.c5 ixd3 the opponent's knight, the bishop is tied to
1 7.Wxd3 bxc5 1 8.dxc5 d5+. defensive functions.

This position was reached in Merry - Jedynak, 8 . b6


..

Isle of Man 20 1 4. Now it makes sense to Since the pin on the f6-knight doesn't cause
clarify the situation in the centre by means of: Black any concrete problems, it makes sense to
attack the weak c4-pawn as soon as possible.
8
9 .id3
.

7 9.Wa4 id7 1 0.Wc2 llJa5 1 l .e5 dxe5 1 2.dxe5


6 llJ g8 leaves White with no compensation for
an ugly pawn structure.
5
4 9 . . a5 10.h3
.

1 0.llJe2 ia6 1 1 .0-0 llJ d7! A typical


3 prophylactic manoeuvre - avoiding an
2 unpleasant pin on the f6-knight significantly
limits White's activity on the kingside. 1 2.f4
1
ixc4 1 3 .llJg3 We? 1 4.f5 ixd3 1 5 .Wxd3, as
a b c d e f g h was played in Danielsson - Koenig, Warsaw
1 O ... eSN l l ..tfl 1 93 5 , could be met strongly by:
Black would benefit from having a closed
type of position after 1 l .d5 llJ a5 1 2. llJ c l g5
1 3 .if2 llJh5+.

1 1 . 0-0 12.cg3 cxd4 13.cxd4 xd4


.

14.bd4 exd4 1 S.f;Yxd4 .te6 16J:M1 gc8=


Black has excellent play.

A2) s ..ie3

8 a b c d e f g h

7 1 5 . . . 0-0-0!N 1 6.Wb5 e5+ - Black's king feels


quite safe on the queenside, when most of the
6 opponent's pieces are blocked by the pawns.
5
10 ....ta6 l l .f;Ye2
4
So far we have been following the top-level
3 game Spassky - Hubner, Bugojno 1 982. Now
2 Black should have neutralized the coming
kingside pawn storm by means of:
1
a b c d e f g h
Chapter 5 - 4 . f3 69

The immediate 5 ... b5 is also possible. The


usual reply is 6.e4, when 6 . . . d6 transposes to
our repertoire. (6 . . . 0-0 transposes to the note
below; 6 . . . bxc4 is the main line, but I am not
so keen on it.)

5 . . . 0-0 6.e4 b5!? is another variation on the


. . . b5 theme; this way Black keeps the option
of playing . . . d5 in one move. The advantage of
the early . . . d6 is that we prevent e4-e5 while
preparing to meet dxe6 by recapturing with
the bishop.

6.e4 b5
According to the database, this move was
introduced by GM Leonid Yudasin in 1 990.
B) 5.d5 It has still not been played in many games but
I like the concept: Black is ready to give up a
pawn in order to weaken White's pawn centre
and exploit his lead in development.

a b c d e f g h
No doubt this advance is the most principled
reply - White seizes a lot of space and hopes to a b c d e f g h
keep the massive pawn centre.
The most principled response to this gambit
5 ... d6 idea is 7.dxe6, but I will also mention a range
Surprisingly, this popular move was not even of other lines. So the main lines we will see
mentioned by Yakovich in Play the 4j3 Nimzo are Bl) 7.cxb5, B2) 7.i.d2, B3) 7.a3,
Indian. In most of the older games Black went B4) 7.i.g5, B5) 7.ge2, B6) 7.i.d3!?N and
on to either block the centre with . . . e5 or go B7) 7.dxe6.
for a Benoni structure with . . . exd5. Instead, I
want to make life more difficult for White by 7.i.e3?!
undermining his pawn structure with . . . b5. Considering that the centre is about to blow
up, this way of developing seems too slow.
70 Various 4th Moves

7 . . . 0-0 As in the Benko Gambit, White is ready


The most accurate move, postponing the to grab a pawn, hoping to neutralize Black's
capture on c4 to allow White to waste a queenside activity. However, Black's lead in
tempo with his bishop. development and the open centre offer us
8.id3 bxc4 9.ixc4 much more dynamic play than in the above
mentioned opening.

7 ... exd5 8.exd5 0-0


The open e-file obviously benefits Black.

9 .tc4?
.

9.id3N a6 1 0.lLlge2 Wfb6 1 1 .ie3 was the


lesser evil, giving back a pawn. Still, after
1 1 . . .:i:l:e8 1 2.if2 axb5 1 3.0-0 ixc3 1 4.bxc3
:i:l:a3 Black gets a comfortable position.

a b c d e f g h 9 .. lLlh5!
.

A powerful resource, which enables Black to


9 . . . lLlxd5!?
extend his initiative. Obviously, White has no
A standard tactical resource in this variation.
time for 1 0.lLlge2? Wfh4t, winning the bishop.
Although the text move is fine, it is worth
mentioning that the calmer 9 . . . ia6!?N also IO.g3 ti'f6 l l .ti'c2 eSt I2.Cbfl .t5
offers Black an excellent game. Black had a winning attack in Mann -
1 0.ixd5 Ploch!, Stockerau 1 992.
1 0.exd5? Wfh4t-+ is the key point of course.
1 0 . . . exd5 1 1 .Wfxd5 Wfb6 1 2.lLlge2 B2) 7 .td2

1 2.Wfxa8? ib7-+
1 2 . . . ia6 1 3.<;i;>f2 lLl c6+
Lindqvist - Haapasalo, Jyvaskyla 1 99 1 .

Bl) 7.cxb5

a b c d e f g h
Breaking the pin.

7 ... a6!
a b c d e f g h I do not like supporting White's development
with 7 . . . bxc4 8.ixc4 e5 9.lLlge2 lLl bd7
Chapter 5 - 4.f3 71

1 0.0-0 0-0 1 1 .a3;!; as i n Krizsany - Lengyel,


Hajduboszormeny 1 996.

8 .ld3

Or 8.a3 ia5 9.b4 (9.dxe6 ixe6 1 0.cxb5 0-0


is too risky for White) 9 . . . ib6 1 0.cxb5 exd5
1 l .ig5 d4 1 2.lLld5 ie6, with counterplay.
White experiences definite problems
developing the kingside pieces.

8 ... bd7 9.a3


Accepting the challenge now doesn't make
any sense, since White would have to lose time a b c d e f g h
retreating the bishop: 9.dxe6 fxe6 1 0.cxb5
lLle5 1 1 .ie2 0-0, with the initiative. 14 ... xe4!
This enables Black to fan the flames of his
Now in Gerard - De Sousa, France 1 996, initiative.
Black missed a clear way to seize the initiative:
1 5.fxe4 h4t 16.'it>d2 g5t 17.'it>c2 :i:l:fb8
Black will regain the piece while keeping an
overwhelming position:

1 8.a4 xg2t 19.e2 xe4t 20.'it>cl gxb5-+

B3) 7.a3

8
7
6
5
4
3
2
12.dxe6 be6 13.cxb5
After 1 3 .ixe5 dxe5 1 4.Wfxd8 :i:l:fxd8 Black is
clearly better in the endgame. a b c d e f g h
This move has only been played one game,
13 ... axb5 14 .hb5

but it's a reasonable option which forces us to
At first glance, the powerful c3-bishop gives make an important decision.
White reasonable chances to consolidate.
However, Black has a beautiful tactical 7 ....la5
resource:
72 Various 4th Moves

In general, Black prefers to keep the 1 0.ixd6?! Wlb6 1 1 .l2Jge2 l2J c6 gives Black
dark-squared bishop on the board. some initiative.
A double-edged endgame would arise after:
On the other hand, the value of a tempo in 1 0.Wfxd6 Wlb6 1 1 .Wfxb6 axb6 1 2.lDge2 l2J c6
such sharp positions is exceptionally high, 1 3.0-0-0 0-0?
so 7 . . . ixc3t!?N 8.bxc3 0-0 deserves serious 1 0 . . . l2J bd7 1 1 .ixd6 ic7 1 2.ixc5 ie5
consideration as well. Play may continue: 1 3 .l2Jge2 Wfc8
Black gets interesting compensation for the
sacrificed pawn.

8 0-0 9.ge2
..

This position was seen in Frugah - Busch,


Germany 1 993. Now Black should have opted
for:

a b c d e f g h

9.dxe6 (after 9.ig5 h6 1 0.ih4 ge8 1 1 .ie2


a6+ White experiences the usual problems
with developing the kingside pieces) 9 . . . ixe6
1 O.cxb5 a6 1 1 .a4 ge8 Black has excellent
compensation for the pawn due to his serious
development advantage.

s .td3
a b c d e f g h
.

I also examined:
8.dxe6!?N ixe6 9.if4 9 exd5N IO.cxd5 a6 1 1 .0-0 bd7
. . .

Black has comfortable play.

B4) 7.-tgs

a b c d e f g h

9 . . . bxc4
9 . . . l2J c6!? 1 0.Wfxd6 l2J d4 is also interesting.
1 0.Wla4t
Chapter 5 - 4.f3 73

Pinning the f6-knight temporarily helps IO ... g5


White to hold the centre, but it doesn't support Black is not afraid of ghosts! I believe
the development of the rest of his minor pieces. White's lack of development should prevent
him from exploiting the weaknesses created by
7 . . h6 s .th4 0-0
. .
this advance.
Now White has to release the pressure in the
centre in order to complete the development 1 0 . . . a6!? 1 l .bxa6 tLlxa6 also seems perfectly
of his kingside pieces. playable, but the text move is more forcing and
definitely in the spirit of position.
9.dxe6
9.i.d3 exd5 1 0.exd5 (after 1 0.cxd5 c4 1 1 .-tn ds IVge2
1 1 .i.c2 i.c5 ! White's king is stuck in the centre, White can hardly afford any activity on the
so Black gets a powerful initiative) This was kingside in such a situation: 1 2.h4?! g4
played in Salmela - Luukkonen, Laukaa 1 998.
Now I suggest the following improvement:

a b c d e f g h

a b c d e f g h
1 3 .a3 (The following line illustrates well Black's
attacking potential: 1 3 .exd5? tLlxd5 1 4.tLlge2
1 0 . . . bxc4N 1 l .i.xc4 i.xc3t 1 2.bxc3 tLl bd7 Wf6, with a decisive attack.) 1 3 . . . i.xc3t
1 3.lLle2 lLle5 1 4.i.d3 gbs 1 5 .0-0 g5 1 6.i.f2 Exchanging the dark-squared bishop isn't
lLlxd5+ White has no compensation for the really a concession - Black wins another tempo
missing central pawn. for developing the initiative! 1 4.bxc3 tLl bd7
1 5 .Wd2 dxe4 1 6.Wxh6 1he queen is the most
9 ...be6 IO.cxb5
powerful piece, but it doesn't bother Black on
its own:
8
8
7
7
6
6
5 5
4 4

3 3

2
2
1
a b c d e f g h
a b c d e f g h
74 Various 4th Moves

1 6 . . . exf3 1 7.gxf3 e8 1 8.i.e2 ic4 Black has In such dynamic positions, time is worth
the more dangerous attack. more than material!

l S .bfB x8
8
The lack of development and the vulnerability
7 of the dark squares around the king put White
6 under strong pressure.
5

4
3

a b c d e f g h
12 ... dxe4
Also possible is 1 2 . . . d4 1 3 .a3 ia5 1 4.b4
i.e? 1 5 .llJa4 d3 1 6.llJxc5 dxe2 1 7.ixe2, but I
do not like the character of the play that arises
- White gets full compensation for the piece!
a b c d e f g h
13.a3
The endgame arising after 1 3 .'1Wxd8 xd8 This is a rare move, but significant in that
1 4.a3 ia5 1 5 .ixc5 llJ bd7 1 6.id4 ac8 still it has been played by 4.f3 specialist Sergey
offers Black full compensation for the pawn. Volkov.

13 ... .ta5 14 ..bc5 7 ... bxc4 8.f4


This is White's idea: he is willing to sacrifice
a couple of tempos with his knight to provoke
. . . e6-e5 and thus secure his pawn centre.

8 ....bc3t
A solid alternative is 8 . . . e5 9.llJfe2 llJ bd7
1 O.llJ g3 llJ b6 1 I .ie2 0-000 as in Baron -
Ushenina, Jerusalem 20 1 6.

9.bxc3 e5 10.e2
Now I suggest a simple developing move:

a b c d e f g h
14 ... bd7!
Chapter 5 - 4.8 75

This move has not yet been tried over the


board. Visually, it looks like a loss of tempo if
Black exchanges on c4, but it allows White to
arrange his pieces most harmoniously.

7 ... 0-0N
7 . . exd5 8.cxd5 a6 was seen in Galiano
.

Martinez - Borst, email 2002, but I have


something else in mind.

8.ge2
After 8.dxe6 .ixe6 9.cxb5 a6 1 0.bxa6 c4
a b c d e f g h 1 1 . .ic2 .ic5 White's king is stuck in the centre,
IO ... bd7N so Black is fine.
1 0 . . . Wla5 was played in Volkov - Bartel,
Dubai 20 1 5 , but the queen is ineffective on 8 ... bxc4 9 ..hc4 exd5 lO.i.xdS xd5
aS, as shown after 1 1 .tLlg3!N. The tactical l l .'!Wxd5
justification is: 1 1 . . . Wfxc3t?! 1 2 . .id2 Wla3 The aS-rook is trapped, but it leads to
1 3 . .ixc4 0-0 1 4.Wfb3! Wfxb3 1 5 .axb3;!; Black's interesting complications:
extra pawn will not last due to the plan of f2
and doubling rooks on the a-file.

l l .g3 b6CD
This is similar to the Baron - Ushenina game
mentioned above. Black's extra pawn is not so
significant yet, but if White wants to win it
back at once then he will have to give up the
bishop pair.

B6) 7.i.d3!?

a b c d e f g h

l l ...i.e6 12.'!Wxa8 '!Wb6


Now the queen is trapped!

13.0-0
The following alternatives also offer Black
sufficient counterplay:

1 3 . .ie3 lLl c6 1 4.Wfxffit c;i;>xffi 1 5 .0-0 .ixc3


1 6.bxc3 Wlb2+ Black's active queen is certainly
not worse than the two rooks.
a b c d e f g h
76 Various 4th Moves

1 3.i.d2 lD c6 1 4.'1Wxf8t tJixf8 1 5 .0-0 l2Je5 16 ... .tc4 17Jdl e5 1 8 . .td2 b2


1 6.l2Jf4 c4t l ?.c;i;>h l Wfd4 also gives Black The penetration of Black's queen offers full
plenty of activity, for instance: equality.

B7) 7.dxe6 he6

a b c d e f g h

1 8Jhd l liJ d3 1 9.l2Jxd3 cxd3 20.a3 i.c5 2 1 .b4


i.b6?

a b c d e f g h
Black's development advantage is obvious
now, but the point behind White's last move
is revealed after:

s . .t4
The important d6-pawn falls.

8.cxb5N
This is the other natural move to consider,
although it has not yet been tested in
practice.
a b c d e f g h
8 . . . d5
13 ... c6 14.xf8t cj;>x8 This seems perfectly playable for Black:
White has a small material advantage, but 9.exd5 lDxd5 l O.ltJge2
it's not easy to neutralize the activity of all
Black's pieces.

15.4
1 5 .d l c4t 1 6.c;i;>h l l2Je5 also offers Black
enough play.

1 5 ...hc3 16.bxc3
White should avoid 1 6.l2Jxe6t?! fxe6
1 7.bxc3 c4t 1 8.c;i;>h l lDe5 when Black has the
initiative.
a b c d e f g h
Chapter 5 - 4.f3 77

1 0 . . . ltl d7! This natural novelty is an improvement


It would be a serious mistake to delay the over 8 . . . Wa5 ?!, which can be met strongly
development of the queenside pieces: by: 9.ltlge2!N (9.c;i;>f2 i.xc3 1 0.bxc3 0-0 was
1 0 . . . 0-0? 1 l .c;i;>f2! c4 1 2.lt:lxd5 i.xdS 1 3 .i.e3 unclear in Dreev - Yudasin, Manila 1 990)
ge8 1 4.ltlc3 White stabilizes the position 9 . . . 0-0 (9 . . . i.xc4?! 1 0.Wfxd6) 1 0.a3 i.xc3t
and keeps a safe extra pawn. 1 1 .lt:lxc3 bxc4 1 2.Wfxd6;!; Black has managed to
1 1 . c;i;>f2 lt:lxc3 1 2.bxc3 regain the pawn, but the positional advantage
1 2.lt:lxc3 0-0 1 3 .ie3 c4 offers Black is clearly on White's side.
excellent compensation for a pawn as well.
1 2 . . . ia5

a b c d e f g h
a b c d e f g h
1 3 .Wfc2 Now we have another split, as White may
1 3 .i.e3 0-0 1 4 . lt:l f4 Wff6 1 5 .ltlxe6 Wfxe6 choose B71) 9.Yfxd6 or B72) 9 ..bd6.
1 6.Wfb3 Wff6 leaves White in danger.
1 3 . . . 0-0 1 4.ie3 Wff6 B71) 9.Yfxd6 Yfa5 10.ge2 hc4
Black has an active position and White's
king is far from safe, while his extra pawn has This may seem scary for White, but his position
little value. is not as bad as it looks.

8 l l .a3 a6
7
6
5
4
3
2
1
a b c d e f g h
8 ... 0-0!N
a b c d e f g h
78 Various 4th Moves

12.i.g5! Let's see what happens ifWhite tries to prolong


Not an easy move to find. The main idea the game: 1 4.ixe2 ixc3t 1 5 .bxc3 Wxc3t
behind it is to vacate the g3-spot for the queen. 1 6.f2 l::1 ad8 1 7.Wxa6 Wfd4t 1 8.'it>g3 l::1 d6
1 9.Wfxb5 ( 1 9.Wfxa7 f5 20.l::1 ad l f4t 2 1 .xf4
1 2.Wfd l l::1 fd8 1 3.Wfc l i.xc3t 1 4.ll:lxc3 i.xfl E1f6t 22.g3 E1g6t=)
1 5 .E1xfl c4

a b c d e f g h
a b c d e f g h
1 9 . . .f5! Black's attack is sufficient to secure
This puts White under some pressure, since a draw at least. For instance, 20.h4?! E1g6t
the king is still in the centre. 2 1 .h3 Wf2 22.g4 fxg4t 23.fxg4 Wfe3t
24.g2 Wxe4t 25 .'it>h3 E1b6+ and by now
12 ...be2! White would be wishing he had taken
1 2 . . . l::1 ad8 1 3 .Wfg3 h8 1 4.f2 i.xe2 a draw.
1 5 .ixe2 ixc3 1 6.bxc3 seems less precise - the
potential of White's bishops might tell in the I4 ... .txfl t s .Y;Ygst 'it>hs I6.Y;Yf6t 'it>gs=
long run.
B72) 9 . .bd6 ges
13 ..bf6 gxf6

8
7
6
5
4
3
2
1
a b c d e f g h
a b c d e f g h
IO.cxb5
14.f;Yxf6 This seems like the most consistent choice.
This leads to a draw by perpetual. Obviously White is going to suffer from
Chapter 5 - 4 . f3 79

undeveloped pieces anyway, so why not grab A little tactic which helps Black to make use
a second pawn? of his development advantage and the unstable
placement of the bishop on d6.
Black has nothing to worry about after:
1 0.a3 i.xc3t l l .bxc3 'WaS 1 2.AxcS
(dubious is 1 2.'1Wd2? :gd8! 1 3.eS ll:l eS+)
1 2 . . . '1Wxc3t 1 3 .c;i;>f2 bxc4 1 4.ll:le2 'WaS
1 S .ie3 ll:l c6 The strong c4-pawn and active
pieces fully compensate for White's pair of
bishops.

I O.if4 'WaS
Black has some reasonable alternatives, but
keeping the queens on the board looks like
an attractive idea.
1 1 .'\Wc l
l l .cxbS a6 1 2.bxa6 c4 1 3.'1Wc2 ll:lxa6 a b c d e f g h
offers Black rich play which more than l l ..te5
compensates for two pawns. l l .exdS ifSt 1 2.tJif2 '1Wxd6 looks extremely
l l . . . bxc4 1 2.a3 Axc3t 1 3.Wxc3 '1Wa4 dangerous, since Black has full control over the
dark squares.

Too risky is 1 1 .'1Wd2?! '1Wxd6 1 2.exdS . Now


Black can choose between several good
looking possibilities, but most natural seems
1 2 . . . Af5t 1 3 .ltlge2 ltl d7 1 4.0-0-0 ig6.
White's extra pawns have little value here, for
instance:

a b c d e f g h

1 4.:gc l
After 1 4.ll:le2 ll:l c6 1 S .lLlg3 lLl d4 Black is not
worse, at least.
1 4 . . . ll:l c6 1 S .lLle2
1 S .i.xc4? ll:lxe4! 1 6.fxe4 ixc4-+
1 S . . . ltl d7 1 6.ig3 fS 1 7.ltlf4 if7
The pressure along the e-file forces White to
liquidate into an equal endgame: a b c d e f g h
1 8 .ixc4 Axc4 1 9.Wxc4t Wxc4 20.:gxc4 fxe4 1 S .h4 h6 1 6.ltlf4 ixc3 1 7.bxc3 ih7 1 8.id3
2 l .fxe4 ll:l d4= ixd3 1 9.ltlxd3 a6 With an attack.

10 ... d5!
80 Various 4th Moves

The vulnerable placement of White's bishops 13 ... xe5


helps Black to complete his development. The text move seems most convincing.

Another tempting try, l l . . . WI'g5 ?!, can be Also quite playable is: 1 3 .. .f6 1 4.id3 ( 1 4.d6
strongly met by 1 2.f4! lDxf4 1 3. l2J f3 Wl'g4 fxe5 1 5 .ic4t tJih8 1 6.l2Jge2 Wfh4t 1 7.g3 Wfh3
1 4.tJif2!, and White's set-up is becoming with counterplay) 14 . . . ixc3t 1 5 .bxc3 ixd3
harmonious. 1 6.WI'xd3 fxe5 1 7.lDe2

8
7
6
5
4
3
2
a b c d e f g h
1
1 7 . . . c4! 1 8.Wxc4 :gc8 Black's counterplay is
a b c d e f g h sufficient for equality: 1 9.Wfd3 lD c5 20.Wfd l
12.exd5 .tf5! Wl'b6 2 l .d6 exf4 22.0-0 l2J e4t 23.'h l Wxd6
The materialistic approach should be 24.WI'xd6 l2Jxd6 25.l2Jd4 :gxc3=
rejected: 1 2 . . . ixd5?! 1 3 .'f2 :gxe5 1 4.l2Jxd5
ie l t 1 5 .Wxe l :gxe l 1 6.:gxe l Wfh4t 1 7.g3 14.fxe5 h4t 1 5 .g3 e4t 16.'it>fl xh1
Wl'd4t 1 8.l2Je3;!; And the unbalanced position 17.3
seems better for White.
8
13.4
7
6
5
4
3
2
1
a b c d e f g h
17 ... c4!
This practically forces both players to find
a b c d e f g h the only moves.
Chapter 5 - 4.f3 81

My first intention was to liquidate into an 22 g4


.

endgame by means of 1 7 . . ..ig4 1 8 . .ig2 VNxd l Now White must force a draw by perpetual:
1 9.xd l .ixf3 20 . .ixf3 xeS, but after 2 1 .d6
d8 22.ic6;!; the passed d-pawn is a source of 23.'\W4
concern. After 23.ltlh4?! VNh2t 24.ig2 .ixg2
25.ltlxg2 e6 White is in trouble.

It looks like White has consolidated and will 23 .. gd3 24.'1Wg5t 8 25.'1Wh6t g8=
.

benefit from the powerful central pawns and


the trapped queen, but Black has an amazing Conclusion
resource:
4.f3 is a bold and ambitious attempt to build
a big centre. We will meet it forcefully with
4 . . . c5 when 5.a3 is harmless after s . . .ixc3t
6.bxc3 ltl c6. So the real test is 5.d5 d6 6.e4,
when I recommend fighting for the initiative
with the sacrificial 6 . . . b5. We saw no fewer
than seven main options, but the most critical
is 7.dxe6 he6 followed by hitting the exposed
d6-pawn with 8 . .if4. After my suggested
improvement 8 . . . 0-0!N White can take on
d6 with either piece, but I am satisfied with
Black's play in both cases.

a b c d e f g h Overall 4.f3 is a forceful try and my reply to


19 .. h5!
.
it sharpens the struggle, so this is a chapter
Creating a strong threat of 20 . . . .ig4, and that should be studied carefully. With much
thus keeping Black's counterplay alive. of the chapter consisting of original analysis,
you should have excellent chances to cause
20.h3 .bh3 2 1 .MI problems for your opponents across the board.
White gives up too much material after
2 I ..ixh3 VNxa l 22.d6 cd8+.

21 ... g5!
The same concrete approach - White doesn't
have time to consolidate and attack the queen.

22.d6
After 22 . .ixh3 VNxh3 23.d6 g4 24.d7 gxf3
25 .xf3 cd8 26.dxe8='VNt xe8 White's
king is still in danger, so the position remains
dynamically balanced.
Various 4th Moves
4.a3
Variation Index
l.d4 f6 2.c4 e6 3.c3 i.b4 4.a3
4...i.xc3t 5.bxc3 c6!?
A) 6.f;C/c2 83
B) 6.e4!? 84
C) 6.i.g5 88
D) 6.e3 90
E) 6.f3 b6 7.e4 i.a6 92
El) 8.h3 92
E2) 8.e5 93
E3) 8.i.d3 a5 94
E3 1) 9.e5 94
E32) 9.f;C/e2 95
E4) 8.i.g5 97

A) after 9 .e3 E2) after 1 2 .g5 E3 1 ) after 1 4.g5

8 8
7 7
6 6
5 5
4 4
3 3
2 2 2
I
a b
9 . . . e8!N 1 2 . . d6!N
. I 4 . . .li:l g6!N
Chapter 6 - 4.a3 83

l .d4 6 2.c4 e6 3.c3 i.b4 4.a3 6 ... d6 7.e4 e5


This system is named after Fritz Samisch,
one of the great players of the past. Samisch
had a lot of success with this aggressive
concept, using it to defeat Capablanca and
Reti, amongst others. White doesn't mind
wasting a tempo in order to clarify the central
structure and establish the bishop pair.

4 ...i.xc3t 5 .bxc3 c6!?

a b c d e f g h
8.6
The somewhat awkward 8.tlJe2 0-0 9.f3 is
well met by 9 ... tlJh5! 1 0.g4 Wfh4t l l .tJid l tlJ f6.
White's position lacks development, so there
is no way to exploit the slight vulnerability of
the queen on h4. 1 2.E1gl Wff2 1 3 .Wid3 tlJa5+
White was in trouble in Mikenas - Jahner,
Prague 1 93 1 .
a b c d e f g h
A relatively rare continuation, but I like this 8 ... 0-0 9.i.e3
concept: Black is ready to exert strong pressure 9.ig5 ?! h6 l O.ih4 exd4 l l .cxd4 g5 1 2.ig3
on the c4-pawn as soon as possible. tlJxe4 sees Black pick up a pawn for not much
compensation.
6 . . . b6 can be played with similar ideas in
mind, and it may easily transpose within a few The text move has occurred in a couple of
moves. That said, I find it slightly more flexible games. I recommend:
to start by developing the knight.

The main lines to consider are A) 6.'!Wc2,


B) 6.e4!?, C) 6.i.g5, D) 6.e3 and E) 6.6.

6.tlJf3 is sometimes played, but after 6 . . . b6


I don't see anything better for White than
7.ig5 , when 7 . . . h6 8.ih4 transposes to
variation C.

A) 6.'1Wc2

Supporting the e2-e4 advance in this way is


not without drawbacks.
a b c d e f g h
84 Various 4th Moves

9 ...ti'e8!N A temporary pawn sacrifice that completely


Attacking the e4-pawn and removing the changes the nature of the game.
queen from a vulnerable spot.
6 ... xe4
IO . .td3 g4 l l ..tg5 h6 12 ..th4 f5! Accepting the challenge is the most
This move forces favourable complications. principled reply.

1 3.h3 7.ti'g4
White manages to win the pawn back, but
in doing so he allows us to force a queen
exchange, which clearly favours Black.

7 ... f5 8.ti'xg7
8 .h5t? g6 9.h6 f6 l O.ltlf3 b6 l l .id3
ia6+ leaves White with no compensation for
the pawn.

8 ... 6
Obviously there is no way for White to avoid
the exchange.

a b c d e f g h
8
1 3 ... exd4 14.hxg4 fxe4 1 5.he4 !U4 16.d2
dxc3 17.ti'xc3 gxe4t 1 8.xe4 ti'xe4t
7
19.fl ti'xg4 6
Black has more than enough for the
5
exchange.
4
B) 6.e4!?

8
7
a b c d e f g h
6
9.ti'xf6
5 9.h6?! is inferior, as long as Black responds
4 with: 9 . . . xh6 l O.ixh6 gg8! (avoiding the
simple trap: 1 0 . . . ltlxc3?? l l .d5 exd5 1 2.ig7+-)
3 l l .ltle2 b6 1 2.f3 ltl d6 1 3 .ltlg3 ia6+
2
9 ... xf6 10.f3
1
It makes little sense to start with 1 O.if4 d6,
a b c d e f g h when White has nothing better than l l .ltl f3 .
Chaprer 6 - 4.a3 85

1 0.g3 makes Black's rask even easier: 10 . . . b6 A quierer way of handling rhe posirion is ro
1 l .ig2 ib7 1 2.lLlh3 0-0-0+ complere developmenr:
1 l .id3 ia6 1 2.0-0 lLl a5 1 3.E1e 1
IO b6!
... Now Black has a choice of decem oprions,
This enables Black ro keep a flexible pawn bur I especially like:
srrucrure (for comparison, 1 0 . . . d6 would
weaken rhe e6-pawn) and pur pressure on rhe
weak c4-pawn.

8
7
6
5
4
a b c d e f g h
3 1 3 . . . ll:le4!?
2 Now Whire is forced ro give up rhe lighr
squared bishop in order ro keep rhe marerial
1
balance:
a b c d e f g h 1 4.ixe4 fxe4 1 5 .E1xe4 ib7!
l l .d5!? 1 5 . . .ixc4 allows 1 6.ll:le5 idS 1 7.E1h4
A sraric characrer of rhe barrie can hardly when rhe pressure on rhe h7-pawn is rarher
suir Whire, so rhis advance is a narural arrempr annoying, rhough Black should srill be okay.
ro make rhe play more concrere. 1 6.d5 0-0-0 1 7 .ig5 E1df8 1 8.E1d 1 E1hg8
Arracking rhe c7 -pawn is illogical, since The b7-bishop is blocked ar rhe momenr,
Black would benefir from opening up rhe bur ir srill exerrs srrong pressure. The
c-file: 1 l .if4 ia6! conrinuarion mighr be:

a b c d e f g h
a b c d e f g h

1 2.ixc7?! l::1 c 8 1 3.id6 lLl a5 1 4.ie5 e7 1 9.h4 h6 20.if4 b5!


Black wins rhe pawn back and obrains a dear Black grabs rhe pawn back and gers a berrer
posirional advanrage. posirion.
86 Various 4th Moves

White's initiative seems sufficient only to


8
maintain the balance. For instance:
7 1 6.lLlh4!
6 Less effective is: 1 6.ig5 :gde8 17 .i.b3 h6
1 8.i.d2 lLl f6+
5

4
3

a b c d e f g h
ll a5
. . .

Declining the pawn seems safest.


Taking on d5 is playable too: a b c d e f g h
1 1 . . .exd5N 1 2.cxd5 lLlxd5 1 3 .ic4 1 6 . . . :ghg8! 1 7.f3!
This must have been White's idea. 1 7.i.d3 allows 1 7 . . . lLlg6! 1 8.lLlxf5 lLl df4
Weaker is: 1 3 .i.d3 i.b7! 1 4.0-0 (after 1 9.ixf4 lLlxf4 20.lLle7t mb8 2 1 .lLlxg8
1 4.i.xf5 0-0-0 White is in serious trouble :gxg8+ and Black is on top.
- Black's pieces are much better mobilized) 1 7 . . .f4 1 8.i.xf4
1 4 . . . 0-0-0 1 5 .c4 lLl de7 1 6.i.b2 :ghg8+ Or 1 8.i.d3 lLl e3 1 9.i.xe3 fxe3 20.c4
d5 2 1 .:gxe3 dxc4 22.i.xc4 lLld5 with
counterplay.
1 8 . . . lLlxf4 1 9.:gxe7 d5 20.i.b3 :gde8 2 1 .:gae 1
lLl h3t=
Forcing a draw, as moving the king to f1
would invite a nasty check on a6.

12.o d4
Here I found a useful improvement over a
top-level game.
a b c d e f g h
8
The text move prepares to put Black's king
under fire. However, the position is still 7
perfectly playable for Black after: 6
1 3 . . . lLl ce7! 1 4.0-0 i.b7 1 5 .:ge 1
Black is also doing reasonably well after 5
1 5 .ig5 h6 1 6.i.h4 lLl f4 1 7.:gfe 1 :gh7 4
1 8.i.xe7 :gxe7 1 9.lLlh4 c;i;>f8 20,:gxe7
3
mxe7 2 1 .lLlxf5t c;i;>f6 22.lLle3 :ges, with
counterplay. 2
1 5 . . . 0-0-0
1

a b c d e f g h
Chapter 6 - 4 .a3 87

12 ... 0-0!N 1 4 . . . b3! It is important to exchange the


1 2 . . . tJif7 1 3 .dxe6t dxe6 14 . .if4 .ia6 1 5.f3 powerful dark-squared bishop. 1 5 .:i:l:b l xc l
led to unclear play in Jobava - Carlsen, 1 6.exd7 .ixd7 1 7.:i:l:xc l White has won a pawn,
Khanty-Mansiysk (ol) 20 1 0 . but after 1 7 . . . c5! 1 8 . .ie2 .ic6+ White finds
himself in a difficult situation - his forces are
Th e text move places Black's king o n a safer uncoordinated and the b5-knight is in trouble.
square, so White's tactical resources are limited.
13 ....ta6
1 3.3 1 3 . . . c5! ? 1 4.b5 b3 1 5 .:i:l:b l xc l
White's king is not so safe either, so it makes 1 6.:i:l:xc 1 e 5 also offers Black comfortable play,
sense to prepare a home for it on f2. but the text move is more ambitious.

14.dxe6
Now, in comparison to the above-mentioned
After 1 4 .ih6 :i:l:fe8 White has nothing better
game, 1 3 .dxe6 dxe6 14 . .if4 can be met
than 1 5 .dxe6 dxe6, with similar play to the
strongly by:
main line.

14 ... dxe6 1 5 .'itf2


There is not much sense in 1 5 .b5 :i:l:ad8
1 6.ih6 :i:l:f7+, when the b5-knight can be
pushed away.

1 5 ... <itf7 16.b5


After 16 . .ih6 :i:l:fe8 1 7.b5 :i:l:e7 1 8.if4 e5
1 9.ixe5 :i:l:xe5 20.xc7 .ib7 2 l .xa8 ixas+
White is in trouble - all Black's pieces are
a b c d e f g h
active and ready to attack the weak pawns.
1 4 . . . .ia6! 1 5 .b5 :i:l:ad8 and Black takes over
the initiative: 1 6 . .ixc7 :i:l:d7 1 7 . .ie5 e4 1 8.f3
d2+

1 3 .b5 e8 also works out well for Black


after: 1 4.dxe6 ( 1 4 . .ih6 :i:l:f6 1 5 . .ig5 :i:l:g6
1 6 . .if4 d6+ doesn't help White)

a b c d e f g h
16 ...hb5!
It is worth straightening out White's pawn
structure in order to gain time and liquidate
White's most active piece.
a b c d e f g h
88 Various 4th Moves

17.cxb5 b3 ISJibl xc1 19Jxcl d7;


Black is better due to his superior minor
piece and control over the dark squares.

C) 6 ..tg5

a b c d e f g h

8 ....tb7!
This is an important moment, as the direct
attack on the c4-pawn seems less effective:
8 . . . i.a6?! This position has been seen in a
couple of games, including at GM level.
a b c d e f g h
However, I managed to find a new way of
The pin on the f6-knight is quite unpleasant, developing White's kingside initiative: 9.e4!N
so White is trying to provoke some weakening g5 1 0.ll:lxg5! hxg5 l l .i.xg5 :gg8 1 2.h4 e5
reaction, like . . . h6 and . . . g5 .
8

7
6 ... h6 7.Ah4
7.i.xf6?! is a serious positional concession:
6
7 . . . Wfxf6 8.e4 d6 9.ll:lf3 e5+
5

7 ... b6 4

Since 8.e4 will now drop a pawn, Black feels 3


free to develop the queenside pieces. 2

8.f3
a b c d e f g h
The most consistent move - the knight
is heading to d2, where it will protect the 1 3 .f4! Wl'e7 1 4.Wff3 White has a venomous
c4-pawn and support the thematic e2-e4 attack, and the pin of the f6-knight puts
advance. Black under unpleasant pressure. For instance,
1 4 . . . exd4 1 5 .e5 :gxg5 1 6.hxg5 ll:lg8 1 7.cxd4
8.f3 i.a6 9.e4 transposes to variation E4. yields White a decisive advantage.

8.e3 is met by 8 . . . ib7, when the natural 9.d2


9.i.d3 ? runs into 9 . . . ll:lxd4! 1 0.cxd4 ixg2 9.e3 d6 1 0.id3 ( 1 0.ll:ld2 g5 l l .i.g3 Wl'e7
with heavy material gains for Black. White transposes to the Bacrot - Carlsen game
should therefore prefer 9.ll:lf3, transposing to referred to under 1 O.e3?! in the notes to the
9.e3 in the notes below. main line below) This position was reached
Chapter 6 - 4.a3 89

in Smailovic - Blaeser, Luxembourg 1 998, Better is 1 4.fxe5, but then Black gives up
when Black missed a chance to eliminate the extra piece and gets a definite positional
the dark-squared bishop in a favourable advantage: 1 4 . . . l2Jxe5 1 5 .dxe5 Wfxe5+
situation: 14 . . . exd4 1 5 .e5 l2Jxe5 1 6.Wfxb7 l2Jf3t-+
The white king can choose which side of the
board to be mated on.

a b c d e f g h

1 0 . . . g5N 1 l .i.g3 lD a5 1 2.0-0 l2Je4 1 3.Wfc2


lDxg3 1 4.fxg3 Wfe?'? Perhaps it's dynamically
balanced, but it seems to me that White has
the more difficult task to prove that he has a b c d e f g h
adequate compensation for the damaged pawn 9 ...ffe7 IO.e4
structure. White ought to handle the position
aggressively.
Compared with the note to Black's previous 1 O.e3?! is too timid. Having taken a
move, the following central expansion and significant positional risk, White can hardly
piece sacrifice doesn't work: afford such slow play. 1 0 . . . g5 It is no problem
9.e4?!N g5 1 0.lDxg5 hxg5 1 l .i.xg5 gg8 1 2.h4 to break the pin in this way, as Black will soon
1 2.i.h4 e5 1 3 .f4 gh8 1 4.ig5 Wfe7 1 5.fxe5 be able to castle on the queenside. 1 1 .i.g3 d6
l2Jxe5 1 6.dxe5 Wfxe5 1 7.i.xf6 Wfxf6+ also 1 2.h4 0-0-0 Black was already slightly better
leaves White with an ugly pawn structure. in Bacrot - Carlsen, Baku 2008.
1 2 . . . e5 1 3.f4 Wfe7
IO ... g5 l l .i.g3

a b c d e f g h

a b c d e f g h
90 Various 4th Moves

l l ... eS!?N 1 7 . . . c6 1 8.dxc6 i.xc6 1 9.ixc6 ltlxc6 20.ltl c4


A typical strategy - the pressure on the c7 White has some compensation for the
d4-pawn forces White to close the position, so pawn, but Black's chances are not worse.
the bishops become much less effective.
The text move is an attempt to change the
1 1 . . .d6!? also deserves attention: 1 2.i.d3 character of the position, but it can be well
0-0-0 1 3 .ltlb3 h5 1 4.h4 ltl d7 1 5 .Wd2 f5 met by:
1 6.hxg5 fxe4 1 7.i.xe4 d5 1 8.i.d3 ltl de5 With
interesting complications in Arreaga Orozco - 13 ...xc5! 14.he5 e7 I S .h6
Grams tad, email 2009. And not 1 5 .i.d4? ltlxe4 1 6.We2 ltlxd2
1 7.i.xh8 ltlxfl 1 8.'xfl i.xd5+.
12.d5 aS
1 5 ...xf6 16.c4 0-0-0 17Jcl d6
Black has comfortable play due to his control
8
over the dark squares.
7
D) 6.e3
6
5 This move may appear slow, but it actually
comes with aggressive intent, as explained in
4
the next note.
3
2
1
a b c d e f g h
13.c5!?
After 1 3 .id3 d6 1 4.ltlfl 0-0-0 1 5 .ltl e3
ltld7 1 6.ltlf5 Wf6+ White suffers from a lack
of constructive ideas, while Black has a clear
plan of regrouping the kingside pieces.

I also considered: 1 3 .h4 0-0-0 1 4.Wf3 d6


1 5 .c5!? dxc5 1 6.Wf5 t ltl d7 1 7.i.b5 a b c d e f g h
6 b6 7.i.d3 aS!
. .

It makes sense to start with this move - the


knight is heading to a5 anyway, whereas the
light-squared bishop has other options too.

7 . . . i.a6 looks obvious and has been by far the


most popular choice, but it has a drawback:
8.e4 ltla5 9.e5 ltlg8 1 0.Wg4! White may have
lost a tempo with e3-e4, but he has actually
benefitted from omitting 2-8. Now Black must
a b c d e f g h
Chapter 6 - 4.a3 91

make some sort of concession with 1 0 . . 'it>f8 or . 8 ... .tb7!


1 0 . . . g6, which I would prefer to avoid. The main idea behind this move is to
provoke the f2-f3 advance, which will block
8.e4N the d 1 -h5 diagonal.
This is a novelty, but it's the most obvious 8 . . . i.a6 leads back to the 7 . . . i.a6 8.e4 ltla5
move and it may easily still transpose back into line as mentioned above.
one of the main lines.
9.e2
The only preceding game saw: 8.l::1 b 1 ?! White White should prefer 9.f3 , when 9 . . . i.a6
hardly can afford the luxury of spending a reaches a position covered via the move order
tempo on prophylactic measures in such a 6.3 b6 7.e4 i.a6 8.i.d3 ltl a5 in variation E3.
position. 8 . . . i.a6 9.Wfe2 This position was
reached in M.V. Santos - Delgado Ramirez, The text move is an independent alternative
Sao Paulo 2004, and here I like the thematic but it has the significant drawback of leaving
approach: the b3-spot vacant for Black's knight.

9 ... b3 10Jb 1 xc1 1 Uxcl 0-0


Now that the dark-squared bishop has been
liquidated, White's attacking possibilities are
heavily limited.

1 2.c f3 d6 13.0-0

a b c d e f g h

9 . . . c5!N (9 . . . d5!?N is a good positional


alternative) 1 O.e4 cxd4 1 1 .e5 ll:lg8 1 2.cxd4
E1c8 1 3 .Wfg4 'it>f8+ The c4-pawn will fall,
and White doesn't have enough play on the
opposite side.

a b c d e f g h
13 ... e5!
Black reaches a comfortable position, as
grabbing the e-pawn only leads to hardship for
White:

14.dxe5?! dxe5 1 5.xe5 d6 16.f4 xa3i


White has a difficult position with many
weaknesses.
a b c d e f g h
92 Various 4th Moves

E) 6.6 EI) 8.h3

This has been played twice by Russian


8
Grandmaster Sergey Volkov, who is a real expert
7 in such positions, so we should definitely pay
6 attention to it.

5
4
3
2
1
a b c d e f g h
Definitely the most ambitious continuation
- White wants to seize space in the centre
before bringing his pieces into play.

By the way, the majority of games from this


position have been reached via the move order
a b c d e f g h
4.f3 llJ c6 5.a3 ixc3t 6.bxc3, which is why I
have referred to Yakovich's book in some of the 9.c5!?
following variations. So why do we not aim for 9.Wla4 was tried in the later game Volkov
this position against 4.f3 as well? There are two Ovetchkin, Internet (blitz) 2006. Now it was
reasons. Firstly, I really like the 4 . . . c5 plan as possible to expose the drawbacks ofWhite's last
covered in the previous chapter. And secondly, move by simple means: 9 . . . 0-0N l O.ig5 h6
after 4.f3 llJ c6, 5 .e4! is a serious option which l l .ih4 g5 1 2.if1 d6, followed by . . . Wid? or
has been doing well for White. . . . Wfe8. White will not get any compensation
after losing the c4-pawn.
6 ... b6 7.e4 .ta6
This popular position can also arise via a The text move sees the pawn get out of harm's
different move order - 5 . . . b6 6.f3 ia6 7.e4 way, but the price is high.
llJ c6.
9 ....txfl IOJhfl d6
Now White is at a crossroads, with four main White's king is stuck in the centre, while the
options: EI) 8.h3, E2) 8.e5 , E3) 8 ..td3 and c4-square still belongs to Black's knight.
E4) s ..tg5 .
I I ..tg5
Quite pointless is 8.ie3 ?! llJ a5 9.e5 llJ g8 In Volkov - Shaposhnikov, Russia 2004,
1 0.Wla4 llJ e7+ as in Gross - Kabatianski, Black could have secured some advantage with:
Germany 1 998.
Chapter 6 - 4.a3 93

Seizing a lot of space and forcing the


knight's retreat looks very tempting. However,
this advance is connected with an additional
strategic risk.

8 ... g8 9.h3
9.i.d3 llJ a5 is covered via the 8.i.d3 move
order in variation E3 1 below.

9 ... a5 IO.f;Ya4
Alas, there is no other way for White to
protect the weak c4-pawn and keep the light
squared bishop on the board.

IO ... e7
Black re-develops the knight and prepares to
castle. This position has been well known since
a spectacular victory by Kotov over Keres at
the 1 950 Candidates Tournament!

I I .Ad3
12.e5 h6 13.exf6
The knight transfer 1 l .llJg5 h6 1 2.llJe4
1 3 .i.h4 g5 14.llJxg5 ?? is impossible, as doesn't really help White: 1 2 . . . 0-0 Now
1 4 . . . llJd5 wins a piece. the d7-pawn is unpinned, so White's next
prophylactic move is almost forced: 1 3.i.f4
I3 ... hxgs I4.fxg7 ggsi Now Black can obtain a clear positional
advantage by means of:
Black will capture on g7, with an extra pawn
and a clear advantage.

E2) 8.e5

a b c d e f g h

1 3 . . .'1We8!N 1 4.f2 d5 Exchanging the queens


is an indisputable achievement for Black.
1 5 .Wxe8 fxe8 1 6.cxd5 i.xfl 1 7.hxfl In
Kuna - Almarza Mato, email 2009, the players
a b c d e f g h
agreed a draw, but Black could have played on
with 1 7 . . . llJxd5+.
94 Various 4th Moves

1 1 ... 0-0 13 ... dxe5 14.dxe5 ti'e8!+


Yakovich recommended l l . . .h6 to prevent This strong unpinning manoeuvre is an
White's next, but there is actually no need. additional point behind Black's 1 2th move.
White has no attack and no compensation for
12.!g5 his terrible pawn structure.
1 2.0-0 has been played a few times but after
1 2 . . . d6!N Black is in excellent shape. E3) 8.id3

The text move has been played in several


games, but I found a new idea which seems to
refute White's aggressive strategy:

a b c d e f g h
This more consistent mode of development
is usually connected with the loss of the
c4-pawn. As tournament practice shows, it is
a b c d e f g h
not easy to prove that White has long-term
1 2 ... d6!N compensation.
Avoiding the weakening 1 2 . . . h6 1 3 .ih4,
as was played in the above-mentioned classic 8 ... a5
game. Let me show you how Alexander White may proceed with E3 1) 9.e5 or
Kotov managed to develop a powerful attack: E32) 9.YlYe2.
1 3 . . . d5 1 4.i.b 1 !! g5 1 5 .W1 c2 llJg6? (correct was
1 5 . . .f5 1 6.exf6 llJ f5 with unclear consequences) E3 1) 9.e5 g8
1 6.llJf4! White went on to win in fine style in
Kotov - Keres, Budapest 1 950. Even though Black is forced to take a step back
in development with this move, it doesn't help
Similarly, 12 ... d5?! 1 3.i.b l ! also gives White White to develop an initiative on the kingside.
a dangerous attack. The text move is much Indeed, White's pawn centre is becoming even
better, as Black avoids blocking the centre and more vulnerable and can be easily attacked by
makes the e5-pawn into a second target (along Black's d-pawn.
with c4) .
IO.ti'e2
13.fl The other way of protecting the pawn
I don't see anything better for White. is too awkward: 1 0.V!ff a4? c5 l l .i.e3 :i:l:c8+
Kravchenko - Shaposhnikov, Yalta 1 996.
Chapter 6 - 4.a3 95

IO ... d5!
I like this concept, as the ensuing bishop
exchange will guarantee Black full control over
the light squares.

Opening-up the c-file by means of 1 0 . . . c5!? is


another decent way of handling the position.

l l .cxd5 .bd3 12.f;Yxd3 f;Yx:d5

I S .gdl
Liquidating into an endgame with 1 5 .%Ve4
%Vxe4 1 6.fxe4 c5+ would leave White with
vulnerable central pawns and a passive knight
a b c d e f g h
on e2.
13.ttle2 e7
Finally Black manages to complete the 15 ... h6 16.i.cl gd8 17.f;Yc2 c5+
development of the kingside pieces. Black is firmly in control.

14.i.g5 E32) 9.f;Ye2


The alternatives also lead White to an
inferior position:
8
1 4.0-0?! %Vc4 1 5 .%Ve4 ltl d5+ was great for 7
Black in Posedaru - Ernst, Belgrade 20 1 1 . 6
1 4.ll:lf4 %Vc4 1 5 .%Vxc4 lLlxc4+ gave Black the 5
better endgame in De Ia Rocha Prieto - Rivas 4
Pastor, Linares 1 99 1 .
3
The text move is the most aggressive, and was 2
seen in Can - Saric, Sarajevo 20 1 1 . White's
1
plan involves castling followed by pushing the
f-pawn as soon as possible. I like the following a b c d e f g h
prophylactic approach: 9 ... c5!
96 Various 4th Moves

A standard way of developing queenside We have been following the game Kelires
counterplay - opening-up the c-file is - Stamatoupoulos, Greece 20 1 2. Here
extremely annoying for White here. Black could have consolidated his positional
advantage with 1 3 . . . llJ d7!N 1 4.0-0 0-0.
Black has opted for 9 . . . llJ b3 1 0J::! b 1 llJxc l White has no constructive ideas apart from
1 1 J::hc 1 i n most games. However, my opinion 1 5 .f4, but then 1 5 . . . e5!+ would significantly
is that releasing the pressure on the c4-pawn restrict the activity of all White's pieces.
makes White's play much easier, despite the
liquidation of the powerful dark-squared
8
bishop.
7
10.h3!?N 6
This move has never been seen in practice,
but it makes sense to complete development 5
as soon as possible, even if the c4-pawn is lost. 4

The following alternatives have been tested in 3


tournament practice: 2
1
1 0.d5 was played in N. Adams - R. Burnett,
Philadelphia 1 996. With the centre now closed, a b c d e f g h
it makes more sense to exchange the poor 10 ... cxd4 1 l .e5!?
a5-knight: 1 0 . . . llJb3N 1 1 .:i:l:b1 llJxc l 1 2.:i:l:xc 1 The natural recapture 1 1 .cxd4 ixc4! enables
0-0 1 3. llJ h3 ( 1 3.e5? exd5 14.exf6 :i:l:e8-+) Black to win a pawn using a little tactical trick:
1 3 . . . e5 1 4.0-0 d6 In this complex position I 1 2.ixc4 WeB The c4-bishop cannot move
prefer Black due to his better pawn structure. (otherwise 1 3 . . . Wc3t would win the a 1 -rook) ,
so White is forced to fight for a draw in the
1 0.ie3 Avoiding the . . . llJ b3 fork and freeing endgame: 1 3 .e5 llJg8 1 4.id2 Wxc4 1 5 .ixa5
the c l -square for the rook makes sense, but Wxe2t 1 6.'it>xe2 bxa5+
Black can force White on to the defensive:
1 0 . . . cxd4 1 1 .cxd4 d6 (there is no reason to let l l ... g8 12.fl
White get rid of the weak c4-pawn: 1 1 . . . :i:l:c8?!
1 2.c5! ixd3 1 3 .Wfxd3;!;) 1 2.llJh3 :i:l:c8 1 3 .:i:l:cl

a b c d e f g h
Chapter 6 - 4.a3 97

12 ... 5! The text move is much better than 1 0.i.g3 ?!


Ensuring that the knight will not reach d6. lLlh5 1 1 .tLl h3 as was played in Padevsky -
Platz, Halle 1 954. Here Black should have
13.cxd4 gcs 14.0-0 .ixc4 15 ..td2 .ixd3 played 1 1 . . .tLla5!N 1 2.Wfa4 0-0 1 3 .c5 i.xfl
16.ti'xd3 e7i 1 4.<;hfl f5 , developing a powerful initiative.
White is still in the game, but he does not
have full compensation for the pawn.

E4) s . .tgs

a b c d e f g h
10 ... h5!
I really like this prophylactic concept. Black's
priority for now should be reducing White's
a b c d e f g h
active possibilities on the kingside, as White's
This is the most ambitious and challenging queenside weaknesses will not run away.
approach. Since it's hard to protect the c4-pawn
anyhow, White concentrates on creating some 1 0 . . . tLla5?! 1 1 .h4 lLl h7 1 2.hxg5 Wfxg5 1 3 .lLlh3
threats on the other side of board, pinning the gave White some initiative in Panov - Sidorov,
knight and forcing Black to think about e4-e5 . Yalta 1 99 5 .

I n the majority o f games Black has preferred 1 0 . . . Wfe7?! 1 1 .h4 0-0-0


. . . tLla5, either immediately or after chasing the
bishop back to h4. Another option is to step
out of the pin with . . . Wfc8. However, I would
like to suggest a third idea:

8 ... h6 9 ..th4 g5!?


Provoking this weakening move is a definite
achievement for White, as now he has a clear
target for an attack on the kingside. However,
he has still only developed a single piece, and
Black is only a couple of moves away from
a b c d e f g h
long castling.
This appears more reasonable, but it also
10.-tfl leads to some problems for Black after:
98 Various 4th Moves

1 2.d5 tLla5?!
The lesser evil was 1 2 . . . tLle5 1 3.c5 xfl
1 4.d6 cxd6 1 5 .cxd6 Wfffi 1 6.'xfl tLle8
1 7.hxg5 lLlxd6 1 8.gxh6;t, with a slight edge
for White in this complex position.
1 3.c5 xfl 1 4.d6 cxd6

a b c d e f g h

1 2 . . . xfl 1 3 .tJixfl d5
13 ... bxc5 is playable, but after 1 4.d5 tLld8
1 5 .h4 the black king is much less secure.
1 4.cxd6 Wxd6
Followed by . . . 0-0-0, leading to a long
a b c d e f g h
strategic battle where Black's chances are
certainly not worse.
Now in Carlsen - Leko, Monte Carlo
(blindfold) 2007, White could have obtained
a clear advantage with: 8
1 5 .cxb6!N axb6 7
Even worse is: 1 5 . . . xg2? 1 6.bxa7 tJib7
l ?Jb l t tJia8 1 8.Wfd4 tLl c6 1 9.Wfb6+-
6
1 6.c;i;>xfl d5 1 7.xb6 5
4
l l .h4N
This is a novelty, but it's surely the most 3
logical move. 2
l l .g3 1
This was played in Akvist - Nyberg, Sweden a b c d e f g h
20 1 0, but I don't see much sense in spending
l l . Wf6 12.g4
a tempo restricting the h5-knight. Black gets
. .

This move looks like a concession, as the


a good position after:
awkward h5-knight gets a great outpost.
l l . . . Wfe7!N 1 2.c5
However, it seems like White's best option to
Getting rid of the weak c4-pawn is White's
force the play on the kingside.
only reasonable idea.
After 1 2.h4?! 0-0-0 1 3 .c5 xfl 1 4.c;i;>xfl d5!
1 2.e5 Wig? works out well for Black: 1 3.g4
Black's development advantage is becoming
( 1 3.lLlh3 is well met by 1 3 . . .f5!, exploiting the
threatening.
awkward placement of the h3-knight. Mter
1 4.g4 fxg4 1 5 . fxg4 tLl f4 1 6.lLlxf4 gxf4 1 7Jgl
0-0-0 Black completes his development and
takes over the initiative.) 1 3 . . . tLl f4
Chapter 6 - 4.a3 99

8 8

7 7

6 6

5 5

4 4

3 3

2 2

a b c d e f g h a b c d e f g h

1 4.l2Je2 l2Jxe2 1 5 .ixe2 lD a5 1 6.Wa4 0-0-0 1 7.Wd l


1 7.0-0-0 f5+ It will be very difficult to activate 1 7.l::1 xh5 E1xh5 1 8.gxh5 i.b3! 1 9.Wfb4 l2J c4
White's pair of bishops, while the pressure on 20.i.g3 l2J d2 2 l .c4 lDxfl sees Black regain
the c4-pawn is very annoying for him. the piece with a big advantage.
1 7 . . . Wxf3 1 8 .l::1 x h5 0-0-0
I also examined: Black has more than sufficient compensation
1 2.Wa4 lDa5 1 3 .lDe2!? for the piece, as White's king is stuck in the
This pawn sacrifice isn't toothless and centre.
requires accurate handling by Black:
1 3 . . .i.xc4 1 4.hxg5 hxg5 1 5 .e5

a b c d e f g h

1 5 . . . Wg6!
Leaving the h8-rook unprotected is a b c d e f g h
concretely justified. 13.h3 xh3
The natural-looking 1 5 . . . Wg7?! allows Exchanging the knights is part of Black's
1 6.E1xh5! E1xh5 1 7.l2Jg3 and after 1 7 . . . E1h4 restricting strategy; now White's dynamic play
1 8.i.xc4 l2Jxc4 1 9.Wfxc4 Wxe5t 20.l2Je4 is significantly limited.
E1xe4t 2 l .fxe4 Wxe4t 22.'fl ;!; Black's pawns I have also examined 1 3 . . . l2Jg6 1 4.e5 Wg7
would not fully compensate for the bishop. 1 5 .h5 lDge7, but White has a powerful knight
1 6.g4 Wd3! transfer: 1 6.i.g3! lD a5 1 7.l2Jf2 i.xc4 1 8.i.xc4
Exploiting the awkward placement of lDxc4 1 9.l2Je4 With a promising initiative for
White's queen. a pawn.
1 00 Various 4th Moves

14Jxh3 ffg7 1 5 .hxg5 hxg5 1 6.xh8t Wfxh8 1 7 . .ig3 0-0-0


A flexible continuation - freeing the way 1 8. tJif2 d6 reaches a complex position where I
for the f-pawn makes a lot of sense, as the prefer Black - there are clear targets in White's
g5-pawn is somewhat vulnerable. camp.

At first I rejected 1 4 . . . 0-0-0!? in view of 1 5 ... h5!


1 5 .c5 .ixfl 1 6.'xfl and it looks like White Less convincing seems 1 5 . . . 0-0-0 1 6.c5!
has made some progress on the queenside. .ixf1 17 .'xfl h5 1 8.d5 and White gets some
However, Black can strike back with: attacking chances.

16.g2
1 6.gxh5?! xh5 1 7.g2 Wfh7 1 8.hxg5 h 1
1 9 . .ig 1 ltl a5+

16 ... hxg4
1 6 . . . ltl a5!? is also possible.

17.gxg4 f6 18.hxg5 fxg5 19.ffd2 a5


20.ffxg5 ffxg5 2l .gxg5 .bc4
Black has no weaknesses and cannot be
a b c d e f g h
worse here.
1 6 . . . h5! 1 7.hxg5 Wfg6! (of course, not
1 7 . . . Wlxg5 ? 1 8 . .ih4) 1 8.gxh5 xh5 1 9.xh5 Conclusion
Wfxh5 Black is completely fine here as well.
4.a3 is a direct, almost crude, answer to the
8 Nimzo-Indian. After 4 . . . .ixc3t 5.bxc3 White
has the bishop pair, but he has paid a price
7 in his doubled c-pawns and the tempo spent
6 on a2-a3. I like the relatively rare 5 . . . ltl c6!?
when we looked at a few options, but the
5 most significant is 6.f3 b6 7.e4 .ia6, which
4 transposes to a more common position. Once
again White has various options, but I would
3
highlight 8 . .ig5 as particularly worthy of
2 attention, as it provokes Black into weakening
1 his kingside with . . . h6 and . . . g5.
As is typical of sharp attempts to crush the
a b c d e f g h Nimzo, Black has effective answers against
1 5.gh2 4.a3 , especially if he is well prepared.
This seems a reasonable attempt to play
flexibly.

After 1 5 . .ig3 0--0-0 1 6.Wla4 .ib?+ White is


running out of useful ideas.
Various 4th Moves
a b c d e f g h

4. g3
Variation Index
l.d4 6 2.c4 e6 3.c3 i.b4
4.g3
A) 4...hc3t!? 5.bxc3 d6 6.i.g2 0-0 7.f3 c6 8.0-0 e5! 102
A1) 9J::b 1 104
A2) 9Jfc2 104
A3) 9.c5 105
B) 4 c5 106
B 1) s..tg2 106
B2) 5.d5 107

A I ) after 9.B:b l A3) after I O.li) g5 82) after l l .li) f3

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2

a b c d e f g h a b c d e f g h a b c d e f g h

9 . h6!N
. . 1 0 B:e8!N
. . . l l . f5N
. .
1 02 Various 4th Moves

I .d4 f6 2.c4 e6 3.c3 i.b4 4.g3 A) 4 ...i.xc3t!?

Black is aiming to reach a complex,


double-edged position where the weakness of
White's doubled pawns might tell in the long
run. It is important to note that the power of the
g2-bishop is less effective when Black doesn't
play . . . c5 .

5.bxc3 d6 6.i.g2 0-0

8
7
a c e f g h
6
This move often leads to the main lines of
5
the so-called Romanishin System, which will
be covered in the next two chapters via the 4
move order 4.tLlf3 c5 5 .g3 . However, nowadays 3
4.g3 is considered less flexible than 4.tLlf3, as
it gives Black a couple of good options apart 2
from the main 4 . . . c5.
The fianchetto was first tried in the
a b c d e f g h
early 1 920s, but was deeply explored and
successfully employed much later by Ukrainian 7.f3
Grandmaster Oleg Romanishin, starting in the Dubious is: 7.:gb 1 tLl c6 8.c5?! (8.tLlf3 e5
1 970s. White is aiming to put strong pressure 9.0-0 transposes to variation A1 below)
along the h 1 -a8 diagonal and prevent Black 8 . . . dxc5 9 . .ia3 as in Miana - Giardelli, Buenos
from developing the light-squared bishop. In Aires 1 98 5 , when 9 . . . :ge8N 1 0 . .ixc5 e5 gives
most cases, White should be ready to play very Black the initiative.
energetically, without being worried about
sacrificing a pawn or two. This concept became The ambitious 7.e4?! e5 8.lLle2, as tried in
especially popular after Kasparov's successful Makarichev - Razuvaev, Moscow 1 982, leaves
experience with 4.tLlf3 in his second World the g2-bishop locked in for a long time.
Championship match against Karpov.
As just mentioned, 4.g3 allows Black to
choose between several ways of avoiding the
usual lines of the Romanishin Variation. So in
this case I will offer a choice of replies: I believe
A) 4 ...i.xc3t!? is one of the most attractive
alternatives to the more usual B) 4 ... c5 .

a b c d e f g h
Chapter 7 - 4.g3 1 03

Now Black could have exploited the


permanent drawback ofWhite's pawn structure
by means of: 8 . . . b6!N 9.f4 ia6 1 0.fxe5 dxe5
1 1 .0-0 ixc4 1 2.ig5 llJ bd7+ White does not
get adequate play for the pawn.

7.llJh3!?
This has been seen only in some online
engine games, but definitely deserves some
practical tests.
7 . . . l2k6!
Instead, 7 . . . e5 8.f4 llJ c6 9.0-0 leads to a
highly double-edged struggle. a b c d e f g h
8.0-0
Now 8.f4 can be met by 8 . . . d5!, getting 8 ... e5!
control over the e4-square. This move makes Black's play much more
8 . . J::! e 8! active and prevents White from seizing more
An excellent prophylactic move. space in the centre. Moreover, the further
Once again, 8 . . . e5 seems less flexible. advance of the e-pawn might block the
g2-bishop and yield Black some attacking
possibilities.

White's main options are AI) 9Jb l ,


A2 ) 9.ti'c2 and A3 ) 9.c5.

The preparatory move 9.ge 1 was seen in


I. Sokolov - Kuraj ica, Sarajevo 1 987, and
is again well met by 9 . . . e4!N 1 0.llJd2 ge8
1 1 .llJf1 h6+.

a b c d e f g h The following knight transfer to e3 seems


9.e4 somewhat slow: 9.llJe1 ge8 l O.llJc2 llJ a5
9.f4 is well met by 9 . . . d5! 1 0.llJf2 llJ a5 1 l .c5 1 l .llJe3 This was Van Laatum - Lee, Dieren
b6 1 2.e4 dxe4 1 3 .llJxe4 ib7 1 4.llJxf6t 1 989, when Black should have continued:
Wfxf6+.
9 . . . e5
Now the e4-pawn is under attack, so White
has no time to pursue the initiative with fL-f4.
1 0.ge 1 llJ a5
Black has promising counterplay; the
permanent weakness of the doubled pawns
starts to tell.

7 ... c6 8.0-0
a b c d e f g h
1 04 Various 4th Moves

l l . ..e4!N 1 2.'1W a4 b6 1 3.c5 d5 1 4.c4 ll:lxc4 excellent counterplay due to the opponent's
1 5.ltlxc4 dxc4 1 6.WI'xc4 i.e6 1 7.Wa4 Wl'd5+ weaknesses.
With full control over the light squares.
The text move seems like a reasonable attempt
AI) 9J:bl to swap off one of the weak pawns and open
the position for White's bishops, but Black has
This posmon has been reached in several a good answer.
games, but so far nobody has played the strong
prophylactic move: IO ... e4 l l .d2

8
7
6
5
4
3
2
1
a b c d e f g h a b c d e f g h
9 ... h6!N 1 1 ..Je8!
Black is not in a hurry; White has no direct 1 1 . . .d5 is playable but the text move is much
threats, so it makes sense to restrict the minor simpler.
pieces.
12.cxd6 cxd6 13.c4 d5 14.e3 b6 1 5.cxd5
9 . . J:e8 1 0.i.g5!? h6 l l .i.xf6 Wl'xf6 1 2.e3 aS
gave White easier play in Epishin - Barlocco, Black is firmly in control.
Di Roseto 20 1 0. Unlike some other Nimzo
variations, White's dark-squared bishop does A2) 9.WI'c2
not have many prospects in this structure, so
it makes sense to restrict it.

Even though the text move has never been


played, the resulting position has been reached
a few times from the Four Knights variation
of the English, where White has misplayed the
opening; in that case, it is even Black to move!

10.c5!?
I also checked 1 0.Wfc2 :ge8 l l .dxe5 ll:lxe5
1 2 .ltlxe5 :gxe5 1 3 .if4 i.f5 when Black has
Chapter 7 - 4.g3 1 05

9 .. Je8 10.dxe5 A3) 9.c5


Dubious is: 1 0.l':id 1 ?! e4 1 1 .tLlg5? (better
was 1 1 .tLl d2 i.f5 1 2.e3 tLla5+, but also then
White's position doesn't look attractive)
1 l . . .if5 1 2.Wfa4 WeB-+ White's stranded
knight and general lack of harmony meant
that her position was already losing in Botsari
- Kiriakov, Halkida 1 996.

The text move should be met by:

10 ... xe5!N
1 0 . . . dxe5 1 1 .tLl d2 gave White reasonable
prospects in Potapov - Spirin, Pardubice
2007. The text move makes it harder for him a b c d e f g h
to justify his ugly pawn structure, for instance: Among White's possibilities, this looks most
ambitious: White is trying to get rid of the
weak pawns and open up the position so that
his bishops become more powerful. However,
Black can interfere with those plans by means
of:

9 ... e4 10.g5
We have been following the game
V. Georgiev - Mancini, La Fere 20 1 2. Now
Black should choose:

IO ... e8!N
Echoing the earlier variation A l .
a b c d e f g h

l l .xe5
1 1 .c5 tLlxf3t 1 2.i.xf3 dxc5 1 3 .E1d 1 We?+
doesn't give White full compensation for the
pawn.

l l ... E1xe5 12 ..tf4 e8 13.adl We7 14.fel


h6i
White's pair of bishops cannot compensate
for the damaged pawn structure.
1 06 Various 4th Moves

l l .cx:d6
l l .d5 looks ambitious, but it merely yields
Black the c5- and e5-squares for the knight:
l l . . . llJ b8 1 2.cxd6 cxd6 1 3 .c4 h6 1 4.llJh3 if5
1 5 .ib2 llJ bd7+ With . . . llJe5 to follow.

l l ... cx:d6 12.3 exf3 13.exf3 dS:j:


Stabilizing the position. Black has good
control over the light squares, especially the
c4-outpost, while White's bishops are not
really working. a b c d e f g h

8 . . . d6N 9.llJf3 e5 1 0. 0-0 h6 1 1 .d5 llJe7 In


B) 4 ... c5 this complex position Black's chances seem
preferable - it will be difficult for White to
make the bishops work.

BI) s.i.g2

This speeds up Black's development and thus


cannot be good.

a b c d e f g h
The two main options we will consider in
this chapter are Bl) 5.i.g2 and B2) 5.d5.

Most significant of all is 5.llJf3, which will be


discussed under the 4.llJf3 c5 5 .g3 move order.

5.dxc5 is harmless at best; after 5 . . . ixc3t (but a b c d e f g h


not 5 . . . llJ e4, as 6.'1Wd4! is annoying) 6.bxc3 s ... cx:d4 6.Wfxd4 llJc6 7.Wfd3
Wla5 Black must be at least equal. White hopes to put pressure along the d-file,
but wasting time by moving the queen again
5 .a3 seems slow - the extra tempo helps Black affords Black a significant lead in development.
to put strong pressure on White's doubled Dubious is 7.Wfe3, as was played in Delitzsch
pawns. 5 . . . ixc3t 6.bxc3 llJ c6 7.ig2 (7.llJf3 - Hammes, Fuerth 200 1 , in view of 7 . . . d5!N
transposes to variation E 1 in the next chapter) 8.cxd5 llJxd5 9.ixd5 Wfxd5 1 0. llJ f3 0-0
7 . . . Wla5 8.id2 This position was reached in 1 1 .0-0 W/f5+. Black is clearly better due to
Foerster - Daum, Berlin 1 999. Now I suggest the bishop pair and the potential weakness of
a natural way to handle the position: White's king.
Chapter 7 - 4.g3 1 07

This position occurred in the game Adamis Black's active piece play fully compensates
Fidriliakis, Greece 20 1 5 . Now Black shou l d for White's bishop pair.
have played:
B2) 5.d5
8

7
6
5

- a b c d e f g h
7... d5!N 8.cxd5 exd5
I nstea d 8 . . . "l..l
.r.-- xd5 9.i.d2 Axc3 1 0.i.xc3 0-0

s ... e4
leads to an equal position with a symmetnca 1 Black has some other attractive possibT
I mes,

pawn structure. but I like this aggressive move.

6.
9.a3 bc3t IO.xc3 0-0 I I .llJf3 d4 12.d3
d5 move looks like a concession, but there
is no other way to keep the material balance.

B c2, Wif6! 7.llJf3 (even worse for


Mter 6 . \lb
. .

r.-- r.-- -
Wh Ite Is 7 ."l..l h3 llJxc3 8.i.d2 "l..l xd5+ as was
.
I a d in Vaganian - Karpov, Lenmgrad
r ;
9 ) 7 . . . llJxc3 8.i.d2 llJxd5 9.cxd5 ixd2t
1 0.llJxd2 d6+ White does not get adequate
compensation for the pawn.
1 08 Various 4th Moves

6 ....bc3 7.hc3
Hardly better is:
7.bxc3 f6 8.f3!?N
8.tLlf3 lLlxc3 9.c l lLl e4 10 . .if4 d6+ didn't
give White much for the missing pawn in
Cativelli - Adla, Buenos Aires 1 994.

a b c d e f g h
l l ... f5N 12 ex:f5 L5 13.0-0 d7 14.h4
.

g6
White suffers from a lack of active play,
a b c d e f g h while the weakness of the doubled pawns
might soon tell.
8 . . . tLlxd2!
I do not like the unclear position after Conclusion
8 . . . lLlxc3 9.c l tLl a4 1 0Jb l . In my opinion,
the poor placement of the a4-knight might This chapter dealt with 4.g3 , and the first
tell in the long run. thing to note is that closely related lines can be
9.xd2 d6 1 0. lLl h3 0-0 1 1 . tLl f2 e5 1 2.ig2 reached after 4.lLlf3 c5 5 .g3, as will be covered
e7 1 3.0-0 f5 shortly. By starting with 4.g3, White offers his
Black's position seems preferable due to the opponent a little more freedom, and I decided
better pawn structure. to exploit that by giving two possible replies.
Firstly, 4 . . . ixc3t!? avoids reaching the usual
7 ... x:c3 8.bx:c3 d6 9.Ag2 e5 IO.e4 0-0 lines, and offers Black comfortable play after
1 1 .6 5.bxc3 d6 6.ig2 0-0 7.lLlf3 lLl c6 8.0-0 e5!.
We have been following the game Kluger Instead the main line is 4 . . . c5 when the critical
- Schneider, Eksjo 1 977. Now I suggest the test is 5.tLlf3. In this chapter we looked at
natural move: 5 . .ig2 and 5.d5, neither of which causes Black
much trouble.
Various 4th Moves
a b c d e f g h

Variation Index
l.d4 f6 2.c4 e6 3.c3 i.b4 4.8
4...c5
A) 5.c2 1 10
B) 5.dxc5 111
c) 5.d5 1 12
D) 5.a3 1 12
E) 5.g3 c6 1 15
El) 6.a3?! 1 15
E2) 6.d5 1 16
E3) 6.dxc5 1 17

D) note to 1 o . lLl e 1 D) after 1 4. lLl a3 E3) after 7.Vf!c2

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2

a b c d e f g h a b c d e f g h a b c d e f g h

1 3 . . . lLl c6!N 1 4 . . . Vf!e7!N 7 . . . xc3t!N


1 10 Various 4th Moves

l .d4 f6 2.c4 e6 3.c3 i.b4 4.6 Chapter 2. 5 .ig5 is a harmless sideline of the
This move was first seen back in 1 887(!), Leningrad System - see the note on 5 . lLl f3 at
but it was mainly explored by the great the start of Chapter 4. And finally, 5.e3 0-0 is
players of the 1 920s and 1 930s: Alekhine, variation B of Chapter 1 0.
Euwe, Rubinstein and others. Developing the
A) 5.Yic2
knight in this way keeps White's position quite
flexible, and the dark-squared bishop can still This leads to a harmless line of the Classical
be placed on g5 in the future. Nowadays this System with 4.Wfc2 c5, where White responds
can be considered as an invitation to debate with 5.lLlf3 instead of the more critical 5.dxc5 .
the Romanishin System - most White players
prefer to enter it via this move order rather s ... c:x:d4 6.xd4 c6
than with 4.g3. White has to take care of the d4-knight, so
it's obvious that the queen is misplaced on c2.
4 ... c5
4 . . . b6 is also highly topical, with a Nimzo/
Queen's Indian hybrid, and moves such as
4 . . . 0-0 and 4 . . . d5 are of course possible, the
latter being a Ragozin. But I will recommend
the text move, directly challenging the
d4-pawn and keeping the game in pure
Nimzo-Indian territory.

8
7
6
a b c d e f g h
5
7.xc6
4 The modest 7.e3 0-0 8 .ie2 d5 9.lLlxc6 bxc6
3 1 0.0-0 Wfe7 1 l .b3 e5 didn't pose Black any
problems in Yakimenko - Popilski, Golden
2
Sands 20 1 4.
1
7 ...dxc6 8.a3 i.e7 9.i.f4 VIaS!
a b c d e f g h
The . . . e6-e5 advance will solve the problem
The options we will cover in this chapter of the c8-bishop. Black is already completely
are A) 5.Yic2, B) 5 .dxc5, C) 5.d5, D) 5.a3 fine, and in the following game he was able to
and E) 5.g3. The last move is by far the most take over the initiative.
important, and the analysis of it will continue
into the next chapter as well. 10.e3?! h5!
Gaining the advantage of the bishop pair.
There are three other significant moves, but
each of them transposes to a separate variation. l l .i.d3 e5 12.i.g3 g6 13.i.e2 xg3 14.hxg3
5 .Wfb3 has been covered in variation C of i.e6
Chapter 8 - 4.tLlf3 111

Black was better in Fedoseev - Narayanan, 7 ... a6!


Pune 20 1 4. It is too early for 7 . . . l2Jxc3?!, as 8.id2! gives
Black some problems to solve.
B) 5.dxc5
8.Yfxe4N
Inferior is 8.id2 ixc3 9.ixc3 lDxc3 1 0.bxc3
l2Jxc5+, and Black was obviously better in
Ulanov - Molchanov, Togliatti 20 1 4.

8 ....bc3t 9. c:bdl

7
6
1 5
a b c d e f g h 4
This offers comfortable play after: 3

5 ... e4! 6.Yid4 2


Also harmless is 6.id2, as played in Marwitz 1
""""'"'----'"""""<o.....;;;;..=--""'""'-=..1
- Kolessov, Germany 2003: 6 . . . ixc3N 7.ixc3
a b c d e f g h
l2Jxc3 8.bxc3 l2J a6 9.g3 0-0 1 0.ig2 l2Jxc5
1 1 .l2Jd4 :i:l:b8 White has to take care to equalize. 9 ....bb2!
Less convincing is 9 . . . l2Jxc5 1 0.Wfc2 ie5
6 ...Yif6 7.e3 1 1 .l2Jxe5 Wfxe5 1 2.id2, when White's bishops
7.'1Wxf6 gxf6 8.id2 ixc3 9.bxc3 lD a6 may cause Black significant problems in the
gave Black comfortable play in Medvedev - long run.
Pantykin, Novokuznetsk 2009.
10.bb2 Ylxb2 l l .Yfd4 Yfxd4t 12.exd4 b6!
Creating some breathing room for the
bishop, while forcing the following exchange
to the benefit of the rook on a8.

13.cxb6 axb6 14.c:bd2 i.b7 1 5.i.e2 i.e4=


Preventing :i:l:hb l . Both sides have a weak
pawn in this endgame, and overall the chances
are equal.

a b c d e f g h
1 12 Various 4th Moves

C) 5.d5 9.WI'c2 i.f5 1 o.llJh4 llJxc3 1 1 .llJxf5 llJxa2t


1 2.i.d2 ixd2t 1 3 .WI'xd2 llJ b4 1 4.WI'c3 f6+
doesn't offer White adequate compensation
for the pawn.

9 ...hc3 10.bxc3 d7
Black had excellent play in Fritz - Soelter,
Lieme 2004, due to his control of the
e4-outpost.

D) 5.a3 hc3t 6.bxc3

8
a b c d e f g h 7
Gaining space does not seem to be effective 6
in this situation - the d5-pawn becomes
5
vulnerable when White cannot support it by
e2-e4. 4
3
s ... exd5 6.cxd5 d6 7.g3
7 .i.g5 transposes to a line of the Leningrad 2
System which was covered in variation B 1 of 1
Chapter 4.
a b c d e f g h
7.e3 0-0 8.i.d3 will be covered via the 4.e3 In comparison to the usual Samisch System,
move order - see variation B 1 of Chapter 1 0. White's active possibilities are limited - it's
difficult for him to gain control over e4.
7... 0-o s ..tg2
6 ... 0-0
Since pinning the f6-knight isn't effective in
this situation, there is no reason to reject this
natural move.

7.e3
It is amazing how one line can transpose to
another in chess. Here is one more example:
7.WI'c2 d5 8.e3
8.ig5 is completely harmless after 8 . . . cxd4
9.cxd4 dxc4 1 0.Wxc4 b6. This way of
handling the position resembles the Classical
System as covered later in the book. 1 1 .e3
a b c d e f g h ia6 1 2.Wfa4 ixfl 1 3 .l::1 xfl llJ bd7=
8 ... e4! 9 ..td2 8 . . . b6 9.cxd5
Chapter 8 - 4 . tLl f3 1 13

9 . . . Wfxc3t 1 0.tLld2 gxf6 1 l .d5 d6 1 2.g3 exd5


1 3 .i.g2

a b c d e f g h

9 . . . Wfxd5
a b c d e f g h
This suddenly takes the game into Classical
paths - see variation B2 of Chapter 2 1 , where
this position arises after 4.Wfc2 d5 5.cxd5 This interesting position was reached in the
Wfxd5 6.e3 c5 7.a3 i.xc3t 8.bxc3 0-0 9.tLlf3. game Ivanisevic - Kravtsiv, Jerusalem 20 1 5 .
Incidentally, 9 . . . exd5!? 1 0.c4 cxd4 1 1 .tLlxd4 White was trying to exploit the opponent's
i.b? is also perfectly playable for Black. exposed kingside structure, but Black
actually has no reason to deviate from the
Let's see why pinning the knight on f6 achieves 'greedy' approach:
nothing for White: 1 3 . . . d4N 1 4.0-0 f5 1 5 .l:!b 1 tLl c6 1 6.l:!b3 Wla5
7.i.g5 h6 8.i.h4 Wla5! 1 7.e3 l:!es+
Exploiting the lack of harmony in White's The reduced material leaves White with
camp. insufficient attacking potential.

a b c d e f g h

9.i.xf6?!
This pawn sacrifice is dubious, but it's the a b c d e f g h
only way to fight for the initiative. 7 b6
...

The passive 9.Wfc2 is not in the spirit of the There is also nothing wrong with 7 . . . d5,
position: 9 . . . tLl e4 1 0J::! c l d5 1 l .e3 cxd4N but I like the text move - it allows Black to
Black grabbed the a3-pawn in one game, but keep control over the e4-square without letting
the text move is much easier: 1 2.lLlxd4 tLl c6 White get rid of the weak c4-pawn.
1 3 .cxd5 exd5 1 4.i.d3 l:!e8 1 5 .0-0 i.d7=
1 14 Various 4th Moves

s ..td3 .tb7 9.0-0 e4 1 2.lLld2?!


This theoretical position can be reached via 1 2.ixe4N is better, but after 12 ... fxe4
various move orders. Practice proves that it is 1 3 .ll:l d2 d5 Black has at least equal chances.
difficult for White to make the bishops work 1 2 . . . ll:lxd2 1 3 .ixd2 lLl a5 14.:gfe 1 Wf6+
effectively. Black had a better structure and the more
harmonious position in Matinian - Bocharov,
lO.el Voronezh 20 1 5 .
Also possible is 1 0.ll:ld2, but the immediate
exchange of knights also doesn't bother Black:
1 0 . . . ll:lxd2 l l .i.xd2 f5 1 2.f3 d6 1 3.Wfc2
( 1 3.e4 fxe4 [ 1 3 .. .f4!?] 1 4.fxe4 :gxfl t
1 5 .Wfxfl ll:l c6 1 6.Wf2 Wf6 offers Black a very
comfortable endgame)

a b c d e f g h

This was played in Orr - Joyce, Armagh


1 994, and could be well met by: 1 3 . . . ll:l c6!N l l .f3 d6 12.a4
1 4.e4 f4 1 5 .e5 h6 1 6.exd6 Wfxd6 With After 1 2.ie2 We? 1 3.dxc5 bxc5 1 4.:gb l i.c6
excellent play for Black. Black had a clear advantage due to his better
pawn structure in Yurtaev - Timman, Yerevan
1 0.Wfc2 f5 l l .a4 (ol) 1 996.
After 1 1 .lLl d2 ll:lxd2 1 2.i.xd2 ll:l c6 Black's
chances were already preferable in Gevorgyan 12 ... c6
- Papin, Samara 20 1 5 .
1 1 . . .ll:lc6

a b c d e f g h
a b c d e f g h
Chapter 8 - 4.lt:l f3 1 15

13.tiJc2!
8
White chooses the wrong way to handle
the position - the c4-pawn isn't worth such 7
measures. 6
Better was 1 3 .dxc5N bxc5 1 4.a3 llJ e5 5
1 5 .xc5 Vfic7 1 6.xd6 Vfixd6 1 7.e2 We?= 4
when Black gets full compensation for the
pawn, but not more. 3
2
13 .. .tt:l a5 14.tlJa3
This position arose in Laurier - Gelfand, 1
Biel 1 997, when Black's strongest continuation a b c d e f g h
would have been:
The lines we will consider in depth in
this chapter are El) 6.a3?! , E2) 6.d5 and
E3) 6.dxc5 . The main line is 6.g2 and we
will cover it in the next chapter.

6.Wd3 ?! runs into 6 . . . cxd4 7.llJxd4 llJe5,


and after 8.Wc2 llJxc4 9.g2 (9.Wb3 xc3t
1 0.Wxc3 d5 l l .g2 0-0+) 9 . . . 0-0 1 0.0-0 d5
White did not have much for the missing pawn
in Plastowez - Wiechert, Mannheim 1 994.

El) 6.a3!

a b c d e f g h This is too slow.


14 ... We7!N 1 5 .We2 e5!i 6 ... .bc3t 7.bxc3
Securing a definite advantage.

E) 5.g3

Finally we arrive at the main line, which can


also be reached via 4.g3 c5 5.llJf3.

5 ... tlJc6
This move is somewhat provocative - it
looks like White is being invited to seize a lot
of space with gain of tempo by pushing d4-d5.
However, the pin on the c3-knight offers
Black various tactical resources, so this idea is
justified. Two more common moves are 5 . . . cxd4
and 5 . . . 0-0, but after much analysis, I like a b c d e f g h
what is happening after the knight move.
1 16 Various 4th Moves

7... b6 E2) 6.d5


It makes sense to neutralize the pressure
along the long diagonal as soon as possible.

s ..tg2 .tb7 9.0-0


9.llJe5 can even be met by: 9 . . . llJxe5!?N (the
simple 9 . . . llJ a5 is also fine) 1 o.ixb7 llJxc4
1 l .i.xa8 Wxa8 1 2.0-0 Wc6 Black's position
seems preferable from the human point of
view, since White's rooks are useless in the
closed position that arises.

a b c d e f g h
Seizing space with gain of tempo is amongst
White's most natural replies. However, closing
the long diagonal helps Black to develop the
queenside pieces and attack the c4-pawn.

6 ...i.xc3t 7.bxc3 a5 8.d2 0-0 9.i.g2 d6


10.0-0
After a series of obvious moves, Black now
has to decide how to finish his development.

a b c d e f g h 8
10.i.g5!?N 7
This may be White's best attempt to j ustify 6
his opening play, although it still doesn't
inspire confidence in his set-up. 5
4
After 1 O.dxc5 bxc5 1 1 .i.f4 llJxc4+ Black was
3
obviously better in Stare - Morovic Fernandez,
Pula 2000. 2
1
10 ... xc4 l l . e5 bg2 1 2.<xg2 xe5
13.dxe5 h6 14 ..txf6 gxf6 1 5.f;Yd6 a b c d e f g h
White has some compensation for the 10 ... b6!?
sacrificed pawn, but Black is the only one who I like this concrete approach - White will
can realistically fight for the advantage. not be given time to protect the c4-pawn.
1 0 . . . l::1 e 8 1 l .e4 b6 1 2.l::1 e 1 i.a6 1 3 .i.f1 led
to a long, strategical battle in Miladinovic -
Short, Istanbul (ol) 2000.
Chapter 8 - 4 . lLl f3 1 17

l l .dxe6 Releasing the pressure in the centre should


1 l .e4?! .ia6 1 2.dxe6 fxe6 1 3.e5 dxe5 be met with:
1 4 . .ixa8 Wfxa8+ leads White to an inferior
position. 6 ... e4
I like this aggressive move. Since 7 . .ig2
l l ... .be6 12.ha8 f;Yxas would simply drop material, White is obliged
to waste a tempo to protect the knight.

s "if -
-
.... v.-
7.f;Yc2
Clearly dubious is 7.Wfd3?! as in Name -
7
-
- - l
,

Jacoba de Oliveira Reis, Dois Irmaos 2008, in


6

5
.. . .. ..... % '------
%


z
x% A ,gj:
.JL .:

view of 7 . . . .ixc3tN 8 .bxc3 lLlxc5 9.Wfe3 b6
10 . .ia3 d6 1 1 . .ig2 .ib7+.
4 , zr
-
. 0 x - White's only other plausible continuation is:
7 . .id2 lLlxc3 8 . .ixc3

8u- - - 8n- - -
3
8.bxc3 .ixc5 9 . .ig2 0-0 1 0.0-0 d6+ simply
2
>w\ili/if--- leaves White with an ugly pawn structure.
- 8 . . . .ixc3t 9.bxc3 Wfa5 1 0 . .ig2
a b c d e f g h
13.6 xc4 14.lihc4 hc4i
Black had an extra pawn plus long-term
positional compensation for the exchange
in Gulko - Kuzmin, Tashkent 1 984. Black's
minor pieces coordinate nicely, while it is not
so clear what White should do with his rooks
and bishop. The loss of the g2-bishop also
means that White's king could be vulnerable
in the long term.
a b c d e f g h

E3) 6.dxc5 1 0 . . . Wfxc5!


The other capture would be a mistake:
1 0 . . . Wfxc3t?! 1 1 .tLl d2 0-0 1 2.0-0 b6
1 3 .e3 bxc5 1 4.tLle4 Wfxc4 1 5 .Wfd6 White
has a powerful initiative, which more than
compensates for the pawn.
1 1 .tLl d2 0-0 1 2.0-0 b6 1 3.Wfa4 .ib7=
Black had successfully neutralized the
pressure along the h 1 -a8 diagonal in Giorgadze
- Novikov, Lvov 1 986. Although the position
is objectively equal, in a practical game it is
White who will face the greater challenge not
to end up in a bad endgame with a rotten
queenside structure.
a b c d e f g h
118 Various 4th Moves

This posmon has been seen five times in 0-0 to follow shortly. Note how strong
practice. In all those games, the knights were White's tripled pawns are!
exchanged on c3, seemingly automatically. I 8 . . . ixc5 is safer, but 9.Wfxc3 0-0 1 0.b4 ie7
would like to suggest something better: l l .ib2 if6 1 2.Wfd2;!; is pleasant for White.
9.axb4 lt:lxc l l O.Wfxc l lt:lxb4 l l .Wfc3 Wff6
1 2.Wfxf6 gxf6 1 3 .'it>d2 lt:l a6 1 4.lt:ld4 lt:lxc5
1 5 .ig2
White has at least enough compensation for
two pawns; the poor bishop on c8 is going to
have no moves for a long time.

8.bxc3 xeS
Even though the knight is somewhat less
effective in fighting for the dark squares,
Black's position still looks quite attractive due
to having stable squares for both knights and
potential play along the c-file.
a b c d e f g h
7 ...J.xc3t!N 8
To understand the necessity for this
7
improvement, we must consider the alternative.
6
7 . . . lt:lxc3 5
In Farago - Dely, Budapest 1 978, the
obvious 8.bxc3 ixc5 led to a fine position 4
for Black. However, I discovered a great new 3
idea for White:
8.a3!!N 2

a b c d e f g h
9 ..ig2
9.ie3 b6 1 0.ixc5 bxc5 l l .ig2 ib7 1 2.gb l
lt:l a5 1 3 .0-0 would transpose to the same
position.

Black has better chances after: 9.lt:ld4 lt:le5


1 0.ia3 d6 l l .gd l id7 1 2.lt:lb5 ixb5 1 3 .cxb5
gcs+
a b c d e f g h

8 . . . lt:lxe2t 9 b6 10.0-0 .tb7 u ..ta3 et as 12.bc5


...

8 . . . Wla5 ?! runs into 9.axb4! Wfxa l 1 0.bxc3 bxc5 13Jabl Yffc7 14Jfdl h6
and White is clearly better, with ig2 and
Chapter 8 - 4.tlJf3 1 19

Conclusion

4.tlJf3 is one of the most ambitious ways of


meeting the Nimzo. White keeps a flexible
position and avoids blocking the dark-squared
bishop, thus retaining the option of the
annoying ig5 pin. I recommend the direct
4 . . . c5, when the ambitious 5.d5 exd5 6.cxd5
illustrates the main drawback of having the
knight on f3: it will be difficult for White to
play e2-e4, which means that the d5-pawn will
be vulnerable.
a b c d e f g h
By limiting White's active possibilities Black 5.g3 is the most significant option, when
gets a comfortable position. I should mention I suggest the provocative 5 . . . tlJ c6, putting
that Black is not obliged to castle, as the king pressure on the centre. Once again White has
may feel safe in the centre, as in the following a choice, but in this chapter I looked at the
line: relative sidelines, saving the main line for the
next chapter. Black has a mostly comfortable
1 5.d2 .ixg2 16.c:bxg2 c:be7!? ride in the variations examined here, although
Followed by . . . ab8, intending to swap the it's worth familiarizing yourself with the
rooks and put pressure on White's doubled novelty on move 7 of variation E3, as the
pawns. alternative could lead to problems if your
opponent happens to be armed with the big
improvement I found for White.
8
7
6
5
4

Various 4th Moves


----u/""=._-;;;;J ='-
3
'""'""____--/'""..
2

a b c d e f g h

4. f3 - Main Line
Variation Index
l.d4 tLlf6 2.c4 e6 3.tLlc3 i.b4 4.tLl a c5 5.g3 c6 6.i.g2
6... tLle4
A) 7.d5 12 1
B) 7.flc2 cxd4 122
B 1) 8.a3!? 122
B2) 8.tLlxd4 124
C) 7.fld3 cxd4 8.tLlxd4 tLlxc3 9.bxc3 tLle5! 10.flc2 i.e? 125
C 1) 1 1.f!le4?! 126
C2) 1 1.fla4?! 126
C3) 1 1.flb3 127
D) 7.i.d2 tLlxd2 8.f!lxd2 cxd4 9.tLlxd4 0-0 129
D 1) 10.tLlc2 130
D2) 10.0-0 tLle5 ll.b3 a6 13 1
D2 1) 12.tLl8 133
D22) 12.tLlc2 134
D23) 12JUd1 135
D24) 12.a3 137
B 1 ) after 9 . bxc3 82) after I O.'Wd2 022) after 1 5 . lLl e4?!

8
7
6
5
4
3
2

a b c d e f g h a b c d e f g h a b c d e f g h

9 . . . \Wa5!N 1 0 . . . \Wc?!?N 1 5 . . . d5!N


Chapter 9 - 4 . lLl f3 - Main Line 121

l .d4 f6 2.c4 e6 3.c3 .tb4 4.6 c5 5.g3 9 ...hal !


c6 6 ..tg2 O n this occasion, the greedy approach i s the
As mentioned in the previous chapter, this best.
is the critical test of our provocative knight
development. 9 . . . i.xd2t was played in Vargyas - Kovacs,
Hungary 1 994, when 1 0.lLlxd2!?N tLle5 1 l .f4
6 e4
..
tLlg4 1 2.i.f3 would have given White full
This knight hop is by far the most common compensation for the pawn due to his big
and consistent choice. space advantage.

8 lO.xal d4 l l .xd4 cxd4 12.xd4 b6!


7
This accurate move forces the queen to leave
the perfect d4-square.
6

5 Less precise is 1 2 . . . 0-0 1 3 .i.c3 f6 1 4.d6! b6


1 5 .c5 b 1 t 1 6.c;i;>d2 xa2t 1 7.c;i;>e3, when
4 Black's queenside pieces are paralysed.
3
13.d3
2
White is unable to capture on g7 because
1 . . . b 1 t leads to mate.
a b c d e f g h
13 ... d6 14.0-0 0-0
The options for White we will look at are
A) 7.d5 , B) 7.c2, C) 7.d3 and D) 7..td2.
A) 7.d5

This ambitious sacrifice has only ever been


played in a couple of games, but it should not
be ignored.

7... xc3 8.bxc3 hc3t 9 ..id2

6 a b c d e f g h

5
l SJ:: b lN
The over-optimistic 1 5 .h4? e5 1 6.h5
4 h6 led White to a lost position in Zude -
3 T. Kosintseva, Moscow 2005. The text move
is a better try, but Black can retain the better
2 chances by giving back the exchange:
1 22 Various 4th Moves

1 5 ...f;Ya6! 16.dxe6 he6 17.bb7 Wfxa2 9.b3? was played in Moebus - Miller,
1 8.ba8 .bc4:j: Augsburg 1 997, when 9 . . . d5N 1 0.cxd5
White still has some problems to solve. Wfxd5 would have left White a pawn down
in a bad position.
B) ?.f;Yc2 9 . . . 0--0 l O.:i:l:d l cxb2
If the drawing line below is not fully
acceptable, I can also suggest 1 0 . . . f5!? l l .Wfc2
cxb2 1 2.ixb2 Wfe7 with a complex battle.
l l .ixb2 ic5 1 2.lLle5 Wfc7

a b c d e f g h
This method of protecting the c3-knight is a b c d e f g h
not without merits - the queen is comfortably
Black's position is very solid, so White
placed on c2, and it attacks the knight on
has nothing better than forcing a draw by
e4. However, the lack of control over the
perpetual:
d4-square is an obvious drawback.
1 3 .Wff4 d6 1 4.ll:lxc6 bxc6 1 5 .ixg7 c;!.?xg7
1 6.Wfg5t c;!.?hs 1 7.Wff6t mg8=
7 ... c:x:d4
Now White has two main options: Bl) 8.a3!
Bl) 8.a3! and B2) 8.xd4.
This original idea has been tried only once in
A harmless sideline is: practice, but it deserves attention.
8 .'1Wxe4 dxc3
8 ....bc3t 9.bxc3

3
a b c d e f g h 2
9.0-0N 1
a b c d e f g h
Chapter 9 - 4.tLlf3 - Main Line 1 23

9 ...a5!N White is a pawn down and will have to fight


This tactical resource is an important novelty. for a draw.

9 . . . d5?! proved to be a mistake after 1 0.ll:lxd4


ll:l d6 1 1 .cxd5 ll:lxd4 1 2.cxd4 exd5 1 3.0-0,
when White was clearly better due to his
preferable pawn structure and pair of bishops
in Stocek - Vavrak, Slovakia 2008.

10.0-0 xc3 l l .e3


Black is two pawns up at the moment but
the c3-knight is in danger, so it is necessary to
release the pressure at the cost of material.

l l ... a4!
White is at a crossroads now. a b c d e f g h
1 2 ... d5 13.cxd5 xd5 14 ..tb2 0-0
12.d3 Black has two extra pawns, so it makes sense
The alternative is: to get castled and force White to spend a
1 2.Wxa4 ll:lxa4 1 3 .ll:lxd4 tempo capturing on d4.
Mter 1 3.exd4 b6 1 4.d5 lLla5 1 5 .ll:l d4 i.b7
1 6.lLlb5 ll:lxc4 1 7.lLlc7t me7 1 8.ll:lxa8 ixa8 1 5.xd4
1 9 J::! e 1 f6 Black's chances are preferable in I also analysed: 1 5 .ac l b6 1 6.ll:lxd4
this complex endgame.
1 3 . . . ll:lb6! 1 4.c5 ll:la4 1 5 .lLlb5 cJle7 1 6.d 1

a b c d e f g h

a b c d e f g h
1 6 . . . ia6! (inaccurate is 1 6 . . . ll:lxd4?! 1 7 .c4!
We8 1 8.Wxd4 with a promising initiative)
1 6 . . . b6! 1 7.Wfd 1 Wxd 1 1 8.fxd 1 ll:lxd4 1 9.xd4 ac8
Activating the light-squared bishop is Black's The endgame is absolutely safe for Black after
main task in the Romanishin System! 20.xc8 ixc8! - but not 20 . . . xc8?!, when
Instead, the greedy 1 6 . . . ll:lxc5 1 7 .a4!, 2 1 .a4! wins back the pawn while giving
followed by 1 8.ia3 , would offer White White a chance to press with the two bishops.
excellent play for two pawns.
1 7.cxb6 axb6 1 8.i.d2 ib7+ 1 5 ... xd4 16.hd4 b6
1 24 Various 4th Moves

Black is ready to complete his development, 1 2.a4 e5 would offer Black a slight advantage
so White should force a draw by perpetual due to his better pawn structure.
while he has the chance.
9 ... xd4 IO.Yfd2
Here I would like to bring a new idea to your
8
attention:
7

17 ..lxd5 1
White should avoid: 1 7.:1l:fd 1 i.b7 1 8.e4
a b c d e f g h
llJ f6+
IO ...Yfc7!N
17 ... exd5 18.hg7 xg7 1 9.Yfxd5 i.e6 An interesting novelty, which has a tactical
20.Yig5t h8 2 I .Yif6t g8= basis.

B2) 8.xd4 1 0 . . . llJ c2t 1 1 .'1Wxc2 i.c5 1 2.0-0 0-0 occurred


in Ubilava - Suba, La Roda 20 1 3, when
1 3.'1Wd3!?N 'We? 1 4.:1l:d 1 :i:l:d8 1 5 .i.e3 i.xe3
1 6.'1Wxe3 Wxc4 1 7.:1l:d6 would have offered
White interesting compensation for the pawn.

l l .cxb4
1 1 .'1Wxd4 e5 1 2.'1Wd3 i.c5 1 3.0-0 d6 offers
Black comfortable equality.

l l ...Yixc4 12Jbl 0-0


The pressure against the e2-pawn makes it
hard for White to arrange castling.

a e f g h 13.i.b2
8 ... xc3 9.bxc3 I doubt that White has anything better.
Dubious is 9.llJxc6, as was played in Garcia
Roman - Jedlicka, Pardubice 20 1 5 . Now the 1 3.b5 gives Black a choice, with 1 3 . . .f6!?
simple 9 . . . dxc6N 1 0.bxc3 ic5 1 1 .0-0 0-0 being the ambitious option. (If a draw is an
acceptable result, then 1 3 . . . d5 1 4.i.b2 llJxb5
Chapter 9 - 4 . ltl f3 - Main Line 1 25

virtually forces White to take a perpetual: C) 7.ti'd3


1 5 .ixg7 c;i;>xg7 1 6.Wg5 t c;i;>h8 1 7.Wf6t tJig8=)
1 4.e3 tLlxb5 1 5 .l::1 b4 Wc5 1 6.a4 tLl c7
8

1
a b c d e f g h
a b c d e f g h
1 7.ia3 l::1 e8 1 8.0-0 lLld5 1 9.E1cl Wa5 The advantage of putting the queen here
20.E1d4 Wxd2 2 1 .E1xd2 a5 White has some rather than on c2 is that the d4-square is
compensation for the two pawns, but Black is adequately defended. The drawback is that the
well and truly out of danger. queen can be easily attacked by the opponent's
minor pieces, so White's dynamic play is
slowed down.

7 ... cxd4 8.xd4


8.Wxe4 has already been covered in the notes
to variation B above, via the 7.Wc2 move order.

8 ... xc3 9.bxc3 e5!


As mentioned previously, Black now gains
an additional tempo for attacking White's
doubled pawns.

a b c d e f g h 8

13 ...Wxb4! 7
A nice trick to force the queens off. 6

14.ti'xb4 ltlc2t 15.c;i;>d2 xb4 16.bg7 5


<tt>xg7 17J:xb4 d5 1 8Jcl b6 4
The activity of White's rooks provides
sufficient compensation for the pawn, but 3
Black is by no means worse. 2

1
a b c d e f g h
1 26 Various 4th Moves

10.Yic2 i.e7
8
I consider this the most logical retreat.
Another popular continuation here is 7
1 0 . . . ic5 , but I prefer not to block the c-file 6
without reason.
5
The main lines we will look at are 4
C1) l l .Yfe4?!, C2) l l .Yfa4?! and C3) l l .Yfb3.
3
White's development advantage is rather 2
symbolic here, so the following pawn sacrifice is
1
not justified: 1 1 .0-0?! lLlxc4 1 2JM 1 ( 1 2.Yid3
Ylc7 1 3 .:i:l:b 1 a6+) 1 2 . . . 0-0+ Battaglini - a b c d e f g h
Brunner, Mulhouse 20 1 1 . 13.c5
I also examined: 1 3 .:i:l:d 1 N ll:lxc4 1 4.ltlf5
C1) l l .Yie4?! exf5 1 5 .Wxc4 i.e6 1 6.Wb4 :i:l:b8+

White is trying to get rid of the weak pawn,


but it runs into:

13 ... 5! 14.Yif4 g6 1 5.Yie3 f4! 16.Yie4 d5


17.Yic2 i.xcs+
White had no compensation for the lost
pawn in Vasilev - Suba, Collado Villalba 2003 .

C2) 1 1 .Yia4?!

a b c d e f g h
The queen looks awkwardly placed here.

l l ... d6!
1 1 . . .ll:lxc4 1 2.ll:lxe6 d5 lets White off the
hook by giving him two routes to equality:
1 3 .ltlxg7t (there is also 1 3 .WI'xc4 i.xe6
1 4.Yib5t Wfd7 1 5 .Wxd7t c;i;>xd7 1 6.0-0 with
an equal endgame) 1 3 . . . c;i;>f8 1 4.Wxd5 Ylxd5
1 5 .ixd5 ltlb6 1 6.i.h6 ltlxd5 1 7.0-0-0 ll:lxc3
1 8.ltlf5t c;i;>e8 1 9. ltlg7t= With a perpetual. a b c d e f g h
This is also unsuccessful.
12.0-0 0-0
1 1 ...0-0 1 2.i.f4
Chapter 9 - 4.ltJ f3 - Main Line 1 27

Hardly better is 1 2J:b 1 d6 1 3.i.xb7 l::1 b 8


14.i.e4 E1xb 1 1 5.i.xb 1 'We?+ as seen in Anton
- Kolev, Albena 20 1 1 .

12 ...ti'c7 13 . .be5
This is the only way to relieve the pressure on
the c4-pawn, but giving up the dark-squared
bishop is a high price.

Even worse is 1 3.l2Jb5 '1Wc5 1 4.i.e3 ?! '1Wxc4


1 5 .'1Wxc4 lDxc4 1 6.i.xa7 d5+, when White
suffered from many weaknesses in Tikkanen -
Zakhartsov, Olomouc 2007. a b c d e f g h
1 1 ... 0-0
13 ... ti'xe5 14.gb 1 The somewhat slow 1 1 . . . a6?! allows White
to apply pressure along the b-file: 1 2.i.f4 'We?
( 1 2 . . . d6 1 3 .c5 'We? 1 4.cxd6 i.xd6 1 5 .0-0;!;)
1 3.l2Jf3 f6 1 4.l2Jxe5 fxe5 1 5 .i.e3 i.c5 1 6.i.xc5
'1Wxc5 1 7.'1Wb4;!; Sachdev - Georgiadis, Rijeka
2008.

12 ..tf4
Chasing the knight away with 1 2.f4?! isn't
in the spirit of White's system, as it closes
the path for his dark-squared bishop. In the
following game Black got the upper hand
quite soon: 1 2 . . . l2J c6 1 3 .i.e3 'We? 1 4.l2Jb5
Wb8 1 5 .c5 b6 1 6.cxb6 axb6+ Stamenkovic
a b c d e f g h
Mareco, Campinas 20 1 1 .
This was Guseva - Kashlinskaya, Skopje There is no independent value in 1 2.0-0
20 1 5 . Now I suggest a natural novelty: '1Wc7, when White has nothing better than
1 3 .i.f4.
14 ... a6N 1 5.0-0 gbs 16JUd1 b6;
Black consolidates a small but long-lasting 12 ...ti'c7!
positional advantage. Stepping into a pin looks somewhat
provocative, but Black can break it at any
C3) l l .ti'b3 moment.
1 2 . . . d6?! would allow White to get rid of
This is clearly the most harmonious way to one of his weak pawns and open up the d-file:
protect the pawn, and it has been the most 1 3 .c5! dxc5 1 4.i.xe5 cxd4 1 5 .0-0;!;
popular choice by far.
13.0-0
A natural-looking alternative is:
1 3 .l2Jb5
1 28 Various 4th Moves

This was played in Camarena Gimenez - 1 6.h3 Wfs


Gonzalez Garcia, Benidorm 2008. I suggest A decent alternative is: 1 6 . . . WI'h5!? 1 7.l2Jxc8
a natural improvement: E1axc8 1 8.ixb7 E1c7 1 9.ig2 l::1 fc8 20.l::1 c l
1 3 . . . WI'c5!N d5+
Unpinning the knight, so White is obliged 1 7.lDxc8 E1axc8 1 8.ixb7 E1c7 1 9.ia6 ic5
to force matters: 20.i.xc5 E1xc5
1 4.i.e3 Wl'xc4 White has an extra pawn but his king is
stuck in the centre, while most of his pieces
are uncoordinated. I definitely prefer Black's
chances.

13 d6
. . .

It is inadvisable to keep the e5-knight pinned


for longer than necessary: 1 3 . . . a6?! 1 4.l2Jf3!
l2Jxf3t (preferable was 1 4 . . . d6 1 5 .l2Jxe5 dxe5
1 6.i.e3;!; but White's pressure is annoying here
as well) 1 5 .ixf3 e5 1 6.ie3 i.c5 1 7.ixc5 Wl'xc5
1 8.E1fd l Wl'c7 1 9.Wfb4 White was clearly better
a b c d e f g h
in Gulko - Balashov, Tallinn 1 983.
1 5 .l2Jxa7
The alternative is: 1 5 .i.d4 Wl'xb3 1 6.axb3
lD c6 1 7.ib6 d5! (I do not like the
following materialistic approach: 1 7 . . . a6
1 8.0-0 d5 1 9.e4, and White gets sufficient
compensation) 1 8.0-0 i.d7 1 9 .l2Jxa7 White
regains the pawn, but it is Black who gets a
minimal edge after 1 9 . . . l2Jxa7 20.ixa7 if6
2 1 .i.d4 i.xd4 22.cxd4 ib5 23 .i.f3 E1fc8+.
The text move appears tempting, but Black
has a nice way to keep the dynamic character
of play:
1 5 . . . WI'g4!
Instead after 1 5 . . . d5 the prosaic 1 6.f4 Wl'xb3 a b c d e f g h
1 7.axb3 l2J g4 1 8.ib6 id7 1 9.0-0 offers
White a safe position with chances for a I4.b5 Y;Ycs
small edge. A drawish endgame would arise after
1 4 . . . Wxc4 1 5 .ixe5 Wl'xb3 1 6.axb3 dxe5
1 7.l2Jxa7 f5 1 8.l2Jc6 E1xa l 1 9.l2Jxe7t f7
20.E1xa 1 xe7, but why should we exchange
White's main weakness?

1 5 .i.e3
1 5 .l2Ja3?! would obviously be an awkward
way to protect the pawn. 1 5 . . . i.d7 1 6.l::1 ab l
ic6 1 7.Wb4 occurred in Jianu - Macak,
Chapter 9 - 4.lLl f3 - Main Line 1 29

Plovdiv 2008. Now the best way to take


advantage of the poor placement of the
a3-knight is:

a b c d e f g h

1 9.Wxb7?! ( 1 9.:i:l:fe l lLl f3 t 20.i.xf3 Wxf3


2 1 .Wxb7 i.f6 leaves White with obvious
a b c d e f g h
weaknesses too, but it was the lesser evil)
1 9 . . . lLl c6+ White's position was on the verge
1 7 . . .'1Wxb4!N 1 8.cxb4 a5 1 9.b5 ixg2 20.'kt>xg2 of collapse in Moiseenko V. Gaprindashvili,
-

:i:l:fc8+ White will have trouble holding his Kocaeli 2002.


position together in this endgame.
16 ... a6 17.lLlxd6 '!Wxb3 1 8.axb3 bd6
1 5 ...'1Wxc4 19.fxe5 .be5
White has some compensation for the pawn
8 due to the undeveloped bishop on c8. Still,
Black's position is completely safe.
7

6 D) 7.i.d2
5

1
a b c d e f g h
16.f4!N
This is the best chance to make White's
position work.

The greedy 1 6.lLlxa7?! only invites fresh a b c d e f g h


trouble. 1 6 . . . d5 1 7.i.d4 Wfxe2 1 8.lLlxc8 :i:l:fxc8 The most common choice, which can
deservedly be considered the main line.

7 ... xd2
1 30 Various 4th Moves

Liquidating the powerful dark-squared queen on c7 without fear of harassment from


bishop is an indisputable achievement for the enemy knight.
Black in the Romanishin System!
12.:1Ud1
The alternative line 7 . . ..ixc3 8.bxc3 0-0 The over-ambitious 1 2.tLle4 '1Wc7 1 3 .c5?!
offers White a lot of dynamic play, which doesn't really prevent Black from completing
fully compensates for the queenside pawn his development: 1 3 . . . b6 1 4.b4 b8 1 5 .fc l
weaknesses. bxc5 1 6.bxc5 tLl e5+

8.ti'xd2 cxd4 9.xd4 0-0 White might also consider:


We have reached the key position of the 1 2.ad 1 N We?
5 . . . tLl c6 variation. Black has the bishop pair Although White's last move was a novelty, this
and is aiming to solve the problem of the position has been reached via transposition
passive c8-bishop. In turn, White will try to in a few games.
apply pressure along the h 1 -a8 diagonal and 1 3.tLle4 tLle5 1 4.b3
the d-file in order to disrupt the opponent's Considering that White has committed
plans. his queen's rook to d 1 rather than c l , it
We will consider 01) 10.c2 followed by makes sense for Black to play actively on the
the more critical 02) 10.0-0. queenside, as in the following game:

Avoiding the pin with 1 0.a3 i.e? 1 1 .0-0 does 8


not have independent value: 1 1 . . .tLle5 1 2.b3 7
( 1 2.'1Wf4 tLlxc4 1 3.tLlxe6 fxe6 1 4.'1Wxc4 d5= is 6
toothless) 12 . . . a6 and we have transposed to
5
variation 024 below.
4

D1) 10.c2 i.e7 1 1 .0-0 a6 3

a b c d e f g h

1 4 . . . b8! 1 5 .f4 tLlg4 1 6.h3


16 . .if3 tLl f6 is similar, but Black might also
provoke complications with 1 6 .. .f5!?.
1 6 ... tlJ f6 1 7.tLld6 b5! 1 8.cxb5 axb5 1 9.e4
This position was reached in Rashkovsky
Khalifman, Minsk 1 985. Black has several
decent moves, but perhaps the clearest way
to secure at least equal chances is:

a b c d e f g h
White's set-up is rather harmless, but
Black still needs to find the right way to get
organized. The last move prepares to put the
Chapter 9 - 4 . ttJ f3 - Main Line 131

8
I do not see any reason why Black should be
worse here.
7

6
02) 10.0-0
5

4 This is the main line, keeping different options


3 open for the knight on d4. After careful
consideration, I believe Black's most accurate
2
continuation is:

a b c d e f g h

1 9 . . . tLld5!?N 20.lLlxb5
20.exd5 Wfxd6 is no problem, and 20.tLlxc8
:gfxc8 also gives Black the most comfortable
side of equality.
20 . . . :gxb5 2 l .exd5 Wla7t 22.tLld4 .ic5 23.'h2
.ixd4 24.Wfxd4 Wfxa2
The position is close to equal, but Black
has some chances to press against the weak
b-pawn.

12 ...'!Wc7 13.:gacl
1 3.tLle4 lLle5 1 4.b3 d6 1 5 .:gac l :gd8 was a b c d e f g h
equal in Potapov - Platonov, Orel 1 996. IO e5
..

Black is aiming to remove all the pieces from


This position was reached in Farago - Suba, the long diagonal, so it makes sense to start
Baile Herculane 1 982. At this point it looks with the knight and force White to take care
good for Black to play: of the hanging c4-pawn.

8 My first intention was:


1 0 . . . a6
7 Preparing to put the queen on c7, as in
6 variation D 1 above. However, the fact that
White has not yet retreated his knight to c2
5
yields him additional tactical resources.
4 l l .:gac l !
1 1 .tLl c2 .ie7 transposes to variation 0 1
3
above, and l l .:gad 1 N Wfc7 1 2.tLlc2 .ie7
2 converts to 1 2.:gad 1 N Wfc7 in the notes to it.
1 The text move is more problematic, as the
following lines demonstrate.
a b c d e f g h
13 d6N 14.e3 ltJeS 1 5 .f4 ltJd7 16.e4
. .

:gds
1 32 Various 4th Moves

8
I have also examined I I .Wf4!?N, which can be
compared with the note above, but the absence
7
of the moves . . . a6 and E1ac l helps Black. The
6
most convincing way to equalize is to enter an
5 endgame: l l . . .Wf6!? 1 2.Wxf6 gxf6 Doubling
4 the opponent's f-pawns isn't a real achievement
3
for White, and the c4-pawn cannot be
defended. Play might continue: 1 3 .l::1 fc l l2Jxc4
2

a b c d e f g h

l l . . . l2J e5!?N
Mter l l . . .l::1 b 8 1 2.l::1 fd U Black had trouble
completing development in Moiseenko -
Shkapenko, Warsaw 2006.
1 1 . . .'1Wc7N transposes to a few games;
after 1 2.a3 Ae7 1 3 .E1fd l !N Black is under
pressure, since 1 3 . . . l2Je5 can be strongly met
a b c d e f g h
by 1 4.c5!.
The text move is a typical device to provoke
b2-b3, but it can be strongly met with: 1 4.l2Jdb5 (after 1 4.l2Je4 d5 1 5 .l2Jxf6t c;i;>g7
1 2.'1Wf4! l2Jxc4 1 3 .l2Jxe6 fxe6 1 4.Wxc4 Ae7 1 6.b3 l2J d6 1 7.lDh5t c;i;>h6 only Black can
1 5 .'1Wb3 l::1 b 8 1 6 .l::1 fd l b5 1 7.e3;!; be better) 1 4 . . . l2Jb6 1 5 .a4 The activity of his
Black is under unpleasant pressure. pieces gives White reasonable compensation
for the pawn, but after 1 5 . . . d5 1 6.e4 dxe4
1 7.lDxe4 ltJd5 1 8.l2Jed6 a6 Black is absolutely
fine.

1 1 ... a6
Taking control over the b5-square is usually
necessary when playing a Hedgehog structure.

I like the text move more than l l . . .Wa5


1 2.l::1 fc l ia3 , as was played in Ivanchuk -
Bruzon Batista, Havana 20 1 4, in view of
1 3.E1d l !N a6 14.l2Ja4 l2Jxc4 1 5 .Wxa5 l2Jxa5
1 6.l2Jc2 Ae7 1 7.l2Jb6 l::1 b 8 1 8.l2Jxd7 Axd7
a b c d e f g h 1 9.E1xd7, reaching an endgame where White
can exert pressure without any risk.
l l .b3
A consistent way to protect the pawn, but it White has a wide choice of continuations; we
has an obvious drawback - it exposes the dark will consider 02 1) 12.f3, 022) 12.c2,
squares. However, it is still the best choice. 023) 12J:Ud1 and 024) 12.a3 .
Chapter 9 - 4.ltJf3 - Main Line 1 33

02 1) 12.ltJf3 14.a3 !J.e7 15.Yif4


1 5 .lt:le4 d6 1 6.gfd 1 'Wc7= is comfortable for
Black.

a b c d e f g h
This has been seen only once in practice, but
a b c d e f g h
it's a genuine attempt to fight for an opening
advantage. I suggest the simple: 1 5 ... d6 16,gfdl Ylc7 17J:: acl b6 1 8.e4
After 1 8.b4 fib? 1 9./ixb? gxb7 20.ltJe4
12 ... xf3tN gdg 2 1 .gd3, trying to put some pressure on
1 2 . . . 'Wa5 1 3 .gfc l lt:lc6 was seen in Grischuk the d6-pawn, Black can simplify matters by
- Ivanchuk, Sochi 2008, when White could means of:
have exploited the vulnerable placement of
Black's pieces by means of:

a b c d e f g h

2 l . . .a5! 22.gcd 1 axb4 23.axb4 d5 24.'Wxc7


a b c d e f g h
gxc7 25.cxd5 gxd5 26.gxd5 exd5 27.gxd5
1 4.lt:le l !N fie? ( 1 4 . . . d5 1 5 .cxd5 exd5 1 6.a3! gc4 28.ge5 c;i;>fg 29.b5 gb4= The activity
fixa3 1 7. 'Wc2 gives White a dangerous of Black's pieces fully compensates for the
initiative) 1 5 .tLld3 d6 ( 1 5 . . . 'Wc7 1 6.c5;!;) missing pawn.
1 6.gab 1 lid? 1 7.b4;!;
1 8 .. Jd8 19.b4 /J.b7
13.ixf3 gbs If the b8-rook was on c8, we could end
As usual, solving the problem of the the line here and conclude that Black has
c8-bishop is Black's main task. no problems. As things stand, we need to
1 34 Various 4th Moves

check the following attempt to exploit the reaches an endgame where White has some
undefended queen: practical winning chances.

26.bxc5 ks 27.J.xh7t xh7 28.Yixc5 gd2=


The activity of Black's rooks makes his
position safe.

022) 12.c2

a b c d e f g h
20.c5! he4 21 .be4 bxc5 22Jxc5
22.bxc5 ? Wfa5! leads to the loss of a pawn for
White, since 23.cxd6? blunders the exchange:
23 . . . ig5-+
a b c d e f g h
22 ... dxc5 23.Yfxc7 gxdl t 24.g2 i.d6
25.Yic6 White can also push away the opponent's
Despite having two rooks for a queen, bishop in this way, but the c2-knight is placed
Black's lack of coordination means that he still quite passively, so Black shouldn't face serious
has to be slightly careful. problems.

12 ... i.e7 13.gfdl


White can also try playing on the kingside
with: 1 3 .:i:l:ad l Wfc7 1 4.f4!?N ( 1 4.llJe4
transposes to Rashkovsky - Khalifman, as
referenced in the note to White's 1 2th move in
variation D l ) 1 4 . . . llJ g4 1 5 .if3 llJ f6

4
a b c d e f g h
3
25 ... gd4!
2
25 . . . cxb4 26.ixh7t f8 27.id3 :i:l:d8
28.axb4 ixb4 29.Wfa4! :i:l: l xd3 30.exd3 a5
a b c d e f g h
Chapter 9 - 4 . lLl f3 - Main Line 135

1 6.g4!?N ( 1 6.e4 d 6 was roughly equal in


Vorwerk - Wassilieff, email 20 1 3; in general,
I don't think Black has much to fear when
White blocks the long diagonal for his bishop)
1 6 . . . l::1 b 8 1 7.g5 ll:l e800 We have reached a
double-edged position. White has gained some
space and Black is temporarily passive; on the
other hand, Black is solid on the kingside and
he can look for ways to open the position
and eventually exploit the holes in White's
position, especially on the dark squares.

a b c d e f g h
1 5 ... d5!N 16.cxd5 exd5 17.c3 .teSt
18.d4
Liquidating into an endgame after 1 8.e3
ll:lxe3 1 9.ltlxe3 d4 20.ll:led5 dxc3t 2 1 .c;i;> h l
cxd2 22.ltlxc7 l::1 a7 23.E1xd2 b 6 doesn't fully
solve White's problems either.

18 ... 6 19.c;i;>h1 gds 20.gacl Yld6;


White suffers from the exposed dark squares
around the king.
a b c d e f g h
023) 12,gfd1
13 ...Yic7 14.4
It looks odd to play this after moving the
8
rook away from the f-file, but Mamedyarov
must have felt it would be useful to keep the 7
second rook on the queenside to keep Black's 6
counterplay in check.
1 4.l::1 ac l has been played a few times; after 5
1 4 . . . l::1 b 8 we reach a position covered in the 4
note to White's 1 4th move in variation 023
below. 3

2
14 ... g4 1 5.e4?!
1
More to the point was 1 5 .if3N ltl f6 1 6.e4
d6, but Black is by no means worse here. a b c d e f g h
White can also ignore the pin for a while,
We have been following the game Mamedyarov but it doesn't look challenging.
- Dominguez Perez, Huai' an (rapid) 20 1 6.
Now Black could have refuted his opponent's 12 .. ,gb8
artificial strategy by means of:
1 36 Various 4th Moves

Black has preferred 1 2 .. .'1Wa5 in a few games, The text move differs from the note above:
but I don't fully trust the early development of rather than bringing his last piece into play,
the queen, as a2-a3 tactics will be in the air. White is moving an already-developed piece in
the hope of invading on d6. This demands a
The text move seems safer: Black gets on with more energetic response from Black:
his main strategic plan of neutralizing the
pressure along the h 1 -a8 diagonal.
8
13.c2 !J.e7 14.e4 7
1 4Jhc l is a logical alternative. Black's
6
soundest continuation is:
5
8
4
7
3
6

5 2
4 1
3 a b c d e f g h
2
14 ... b5!
Inaccurate is 14 . . .Wc7 1 5 .tLld6 b5 1 6.cxb5
a b c d e f g h axb5 1 7.E1ac l i.
1 4 . . . Wc7!N ( 1 5 . . . b5 is slightly premature, and
1 5 .Wfe3! left Black with some coordination 15.c5 ffc7 16.ffc3 /J.b7 17.d6
problems in Sandipan - Bindrich, Gibraltar We have been following Raupach -
20 1 0) 1 5 .tLl e4 d6! Black is ready to complete Firsching, email 20 1 3 . Now Black should have
played:
development with . . . l::1 d 8 and . . . b6, so
the following tactical sequence is critical:
1 6.tLlxd6!? l::1 d 8 1 7.c5 Wxc5 1 8 .tLle3 8
7
8

7
6
6 5
5 4
4
3
3
2
2
1
a b c d e f g h a b c d e f g h

1 8 . . . E1xd6! (but not 1 8 . . . Wxd6? 1 9.Wfc3! when 17 .../J.xg2N 18.'it>xg2 !J.f6 19.e4 IJ.e7
Black is in trouble) 1 9.E1xc5 E1xd2 20.l::1 xc8t 20J:: acl
E1xc8 2 1 .E1xd2 E1c 1 t With an equal endgame. 20.tLld6= would lead to a repetition.
Chapter 9 - 4 . lLl f3 - Main Line 1 37

20 .. Jfc8 2 1 . b4 f6 22.3 a7 23.d6


c7
Black has enough breathing room for
his pieces, while the knight on d6, though
powerful, is not entirely stable.

D24) 12.a3

8
7
6
a b c d e f g h
5
Here I found a natural way to improve
4 Black's play from a high-level game.
3
14 ... d6!N
2 The more aggressive 1 4 . . . 5 1 5 .lLl c3 :gbs
1 1 6.e4 fxe4 1 7.lLlxe4 b6 occurred in Vitiugov
- lvanchuk, Reggio Emilia 20 1 2. Now White
a b c d e f g h
could have secured a solid space advantage by
Chasing the bishop seems White's most means of 1 8.f4!N tLl g6 1 9.b4 .ib7 20.:gac l ;!;.
natural way of breaking the pin. However,
the presence of the pawn on a3 renders his I S.acl d8
queenside less stable. The last preparation before solving the
problem of the c8-bishop.
12 ....te7 13.fdl
Since the a 1 -rook has some defensive 16.e3 b8
functions now, White puts the other rook Black has a solid position and will prepare
on d 1 in order to cause Black some concrete the thematic . . . b5 break, which will make both
problems. of his bishops more effective.

White doesn't really benefit from taking ultra Conclusion


aggressive measures here. For example: 1 3.f4
lLlc6 1 4.e3 b8 1 5 .:gfd 1 Wff c7, and Black After 6.ig2 tLl e4, the sacrificial 7.d5 is
has no reason for complaint. The standard interesting but ultimately favourable to Black,
counterplay based on the . . . b7-b5 advance is while 7.Wff c2 and 7.Wid3 both have certain
still possible. drawbacks: the former weakens the d4-square
while the latter leaves the queen exposed, giving
13 ...Wffc7 14.e4 Black enough time to consolidate his position
The quiet 1 4.e3 isn't challenging: 1 4 . . . :gb8 and put pressure on the weak doubled pawns.
1 5 .Wib2 b6 1 6.:gac l i.b7= and Black was fine No doubt, 7.i.d2 is the best way of protecting
in Thorsteins - Stefansson, Reykjavik 20 1 3 . the knight; it generally leads to a Hedgehog
structure where Black has the bishop pair but
1 38 Various 4th Moves

must work to neutralize the pressure along


the h l -a8 diagonal. Even though White has a
couple of decent plans, such as manoeuvring
the knight to d6 or advancing the queenside
pawns to a3, b4 and c5, Black has every
reason to feel happy: he has a solid, flexible
position with two bishops, with potential for
counterplay based on the . . . b5 break.
4.e3
a b c d e f g h

Rare 5 th Moves
Variation Index
l.d4 ttlf6 2.c4 e6 3.ttlc3 i.b4 4.e3
4...0-0
A) 5JNc2 c5! 140
A1) 6.i.d3 14 1
A2) 6.a3 142
A3) 6.tB a 142
B) 5.ttl8 c5 145
B 1) 6.d5 145
B2) 6.i.d2 146
B3) 6.i.e2 d5 148
B3 1) 7.0-0 149
B32) 7.a3 i.xc3t 8.bxc3 'i'c7 9.cxd5 exd5 150
B32 1) 10.0-0 15 1
B322) 10.dxc5N 15 1

A I ) after 1 3 .d3 B l ) after 1 5 .f4 832 1 ) after J J .ttle5

8
7
6
5
4
3
2

a b c d e f g h a b c d e f g h a b c d e f g h

1 3 . . . ttl b8!?N 1 5 . . .e8!N l l . . . ttl c6!N


1 40 4.e3

I .d4 6 2.c4 e6 3.c3 i.b4 4.e3 5.i.d3 will be considered in Chapters


This is one of the 'Big Two' moves against 1 3- 1 6.
the Nimw-Indian (the other being 4.WI'c2 of
course) . Even though the dark-squared bishop 5.3?! Logically, this move should be connected
gets locked in for a while, White benefits from with an e2-e4 advance, so it makes little sense
great flexibility in how he can develop his here: 5 . . . c5 6.a3 (6.i.d3 ll:lc6 7.ll:lge2 cxd4
pieces. The Rubinstein Variation combines 8.exd4 d5 9.i.g5 dxc4 1 0.i.xc4 b6+ Nguyen
both solidity and ambition, which has helped - De Sousa, Bagneux 2002) 6 . . . i.xc3t 7.bxc3
make it a popular choice at all levels. We will ll:l c6 8.i.d3 b6 9.ll:le2 i.a6 1 0.e4
spend this and the following six chapters
analysing it.

4... 0-0
4 ... b6, 4 ... c5 and 4 ... d5 all have their
supporters, but the text move is the most flexible
of all, as well as the most popular. This is a huge
branching point for the Nimw-Indian.

8
7 a b c d e f g h

6 1 0 . . . ll:l e8! White's kingside play has slowed


down, so Black gets an improved version of the
5
Samisch System. The last move is important to
4 prevent the i.g5 pin. 1 1 .0-0 lLla5 1 2.f4 f5+
3 Black was better in Radjabov - P.H. Nielsen,
Tripoli 2004.
2
1
a b c d e f g h This mixing of the Rubinstein and Classical
variations makes a strange impression.
In this chapter we will make a start on
the Rubinstein by considering the rare lines
s ... c5!
A) S.Wfc2 and B) 5.f3. The latter has the
There is nothing wrong with 5 . . . d5, but
possibility to transpose to a huge main line, but
attacking the d4-pawn highlights the drawback
for now we will only look at the independent
ofWhite's last move.
options. Of course these two moves are not the
most important lines:
The three main replies we shall consider are
AI) 6.i.d3, A2) 6.a3 and A3) 6.f3.
The unusual 5 .i.d2 has been covered in
6.dxc5 is mentioned by the analysis engines
Chapter 3.
and is how White typically responds in the
5.a3 can be found in the next chapter. Wfc2 system when faced with . . . c5, but the
huge difference is that in those Classical lines
5.ll:lge2 will be discussed in Chapter 1 2. White would not volunteer e2-e3 with the
Chapter 1 0 - Rare 5th Moves 141

bishop stuck on c l . Specifically, after 4.WI'c2 c5 Any normal developing move is fine, but
5.dxc5 0-0, White's best continuation is 6.a3 this one particularly emphasizes the misplaced
ixc5 7.lLlf3, intending to develop the bishop queen on c2.
actively on f4 or gS . Instead, the weird 6.e3
would transpose to our position after 6.dxc5 . 9.8
If you do face this position, then the simple 9.lLle2, as was played in Michenka - Rigo,
6 . . . lLl a6 already gives Black a comfortable Slovakia 2009, leaves the d3-bishop no squares
game. for retreat, and after 9 . . . b6!N 1 0.cxd5 ( 1 0. 0-0?
runs into 1 0 . . . dxc4 I I .ixc4 cxd4 1 2.cxd4
AI) 6 ..td3 d5 ia6-+) 1 0 . . . c4 1 I .ie4 exd5 1 2.if3 lLlc6
1 3 . 0-0 lLl a5+ Black gets the better position.

9 ... dxc4 IO.bc4 b6 l l .td3.

1 1 .0-0? loses material after 1 1 . . . cxd4


1 2.cxd4 ia6, so White must spend valuable
time moving the bishop again.

l l ... .ta6 12.ba6 xa6 13.d3

8
7
6
5
7.a3
The poor placement of White's queen 4
becomes obvious after 7.cxd5 exdS 8.dxc5 3
lLlc6 9 .lLl f3 ixc5 1 O.a3 ig4, as in Kraidman -
Vadasz, Skara 1 980. 2
1
7 ....txc3t 8.bxc3 c7!?
a b c d e f g h
We have been following the game Luther -
Grandelius, Cappelle-la-Grande 2008. Now I
like the following way of regrouping the pieces:

13 ... b8!?N 14.0-0 c6 1 5.e4 h6


As a result of White's slow play, Black has
managed to put strong pressure on White's
central pawns.

a b c d e f g h
1 42 4.e3

A2) 6.a3 .bc3t

a b c d e f g h

9 . . . ll:l e4!N 1 O.Wfc2 e5 1 1 .ll:l f3 ( l l .dxe5 ll:l c6


1 2.ll:lf3 leads to the same) 1 1 . . . ll:l c6 1 2.dxe5
WlaSt 1 3 .id2 ll:lxd2 1 4.Wfxd2 Wfxc5 1 5 .1:' k 1
a b c d e f g h Wlb6 1 6.ie2 ges+
7.f;Yxc3
This position resembles the Classical line 9 ... dxc4 10 ..bc4 b6 1 1 .0-0 .tb7 12.e5
with 4.Wfc2 0-0 5 .a3 ixc3t 6.Wfxc3, but here c6i
it's easier for Black to take advantage of his lead White had no compensation for the
in development. vulnerable isolated pawn in Kraidman -
Switching to Samisch paths by means of David, Zurich 20 1 1 .
7.bxc3 d6 8.id3 e5 9.ll:le2 makes little sense.
This position arose in Bank Friis - 0. Vovk, A3) 6.f3 cxd4 7.exd4 d5
Aarhus 1 999.

a b c d e f g h

Now Black should have continued 9 . . . e4!N a b c d e f g h


1 0.ixe4 ll:lxe4 1 1 .Wfxe4 ll:l c6 1 2.0-0 lLla5, In comparison with normal positions with
regaining the missing pawn with excellent play. an isolated pawn, here the pressure on the
d4-pawn is much more significant.
7 ... cxd4 8.exd4 d5 9.f3
Black's development advantage should tell s ..tgs
after 9.c5 as in Dollahite - Tears, Fort Worth There are also a few alternatives to consider:
1 95 1 . The correct response was: 8.cxd5 can be met in a few ways, but I prefer
Chapter 1 0 - Rare 5 th Moves 143

8 . . . exd5! when Black is at least equal. be met by: 1 2 ... b6N 1 3 .0-0 i.a6 1 4.:i:l:fc l
( 1 4.tLle5 i.xd3 1 5 .Wfxd3 lLlc6+) 1 4 . . . i.xd3
8.c5 b6 9.a3 This seems consistent. (Instead 1 5 .Wfxd3 lLl c6 1 6.c4 lLl a5 1 7.cxd5 Wid?+
after 9.cxb6 Wfxb6 1 0.i.d3 lLl c6 l l .i.e3 i.a6 Black regains the pawn and keeps some
Black was clearly better in Kracunov - Orlov, positional advantage.
Sombor 2008.) 9 . . . ixc3t I O.Wfxc3 bxc5 1 2 . . . b6 1 3.0-0 i.a6 1 4.i.xa6 lLlxa6 1 5 .i.f4
1 l .dxc5 Now 1 1 . . .aS gave Black a playable :i:l:ac8 1 6.:i:l:fc l Wlb7
position in Simmons - Maggiora, email 2007, White had no compensation for the exposed
but much stronger would have been: queenside structure in Braga - Fernandez
Romero, Albacete 200 1 .

8 . h6 9 .th4 c6
. .

This position, which i s similar to the Panov


Attack in the Caro-Kann, is very comfortable
for Black, mainly due to the poor placement of
the queen on c2. The following practical tests
fully prove this assessment.

a b c d e f g h

1 l . . . e5!N 1 2.b4 d4 1 3.Wfb2 Wfd5 1 4.i.e2 lLlc6


1 5 .0-0 e4 Black's powerful central pawns put
White in a dangerous situation.

8.a3
White is taking a risk in playing such a move
while already behind in development.
8 . . . ixc3t 9.bxc3 Wfc7 1 0.cxd5 exd5 1 I .i.d3
:i:l:e8t
a b c d e f g h
IO.a3
1 0.:i:l:d 1 was well met by 10 ... g5!? 1 I .i.g3 tLle4
1 2.i.d3 f5+ in Stremavicius - Sheykhhasani,
Maribor 20 1 2.

Castling long does not help: 1 0.0-0-0 i.e?


1 l .bl b6 1 2.i.xf6 i.xf6 1 3.cxd5 exd5+
Roussel-Roozmon - Maze, Montreal 2009.

a b c d e f g h IO .bc3t l l .bxc3
..

1 2.tLle5 I found eight games that reached this


The more modest 1 2.i.e3, as in Rozkovec position, but Black only found the strongest
- Vojta, Czech Republic 1 998, should also continuation in one of them.
1 44 4.e3

a b c d e f g h

1 4.ie2 ( 1 4.dxe5?! tLlxe5 1 5 .tLld4 :i:l:e8 1 6.0-0-0


a b c d e f g h id7+ is even worse) 1 4 . . . exd4 1 5 .0-0 '1Wc5+
l l ... e5!
An excellent move to exploit White's lag in 12 ... :i:l:e8 13.0-0-0 xe5
development. If White exchanges knights on e5, we will
return to the correspondence game noted
12.dxe5N above, so we should check to see if he can
This is a logical attempt to improve, but it benefit from avoiding the trade.
doesn't solve White's problems.

The game continued: 1 2.lLlxe5 :i:l:e8 1 3 .0-0-0


tLlxe5 1 4.dxe5 :i:l:xe5 1 5 .cxd5

a b c d e f g h
14.:i:l:xd5
a b c d e f g h
1 4.cxd5 allows Black to make use of the
1 5 . . . if5 1 6.id3 ixd3 1 7 .:i:l:xd3 g5 1 8.ig3 knight: 1 4 . . . tLl g6! 1 5 .ig3 id7 1 6.tJib2 'WaS
:i:l:xd5+ White was under pressure in With excellent attacking chances.
Kameneckas - Kunzelmann, corr. 2008.
14 ... ed7 1 5.i.d3 Ylb6 16.:i:l:b5 Ylc6
1 2.cxd5N '1Wxd5 1 3 .ixf6 gxf6 does not really White's extra pawn is relatively meaningless,
help White, as the doubled f-pawns are not but his shattered structure and unsafe king will
enough to make up for the deficiencies in his be relevant for a long time to come.
position. For example:
Chapter 1 0 - Rare 5th Moves 145

B ) s .a 6 d6 7.i.d3
. .

Or 7.id2 exd5 8.ll:lxd5 ll:lxd5 9.cxd5 ixd2t

m
1 O.Wfxd2 ig4+ and Black had an edge in
8 E -.i.B
-:--,/. ,Y.
Rosenberg - Kacheishvili, New York 2008.

" ! -
7
7 . exd5 8.cxd5
.

6 --
. .

5 -- This position has been tested a few times at

-
GM level, and was also advocated for White by
4 w,w
w o o

IM Palliser in Chapter 9 of Dangerous Weapons:
The Nimzo-Indian (via a slightly different

-----%m .... /.Dm


3 move order involving 5 .id3 c5 6.d5 ) . The
2 8D. ... :_ ... . /. 8r! continuation I like most is:

- - - %ffD:
. .

1
a b c d e f g h
This position has been tested in a huge
number of games, but it usually converts to
the main lines after a subsequent 6.i.d3 . Most
strong players tend to put the bishop on d3
first, as that way White keeps his opponent
guessing as to whether the knight will go to
f3 or e2.

s c5
. . .

The lines to consider in this chapter are a b c d e f g h


Bl) 6.d5 , B2) 6.i.d2 and B3) 6.i.e2. Of
course, these moves do not tell the full story 8 ... tflxd5!
about this position . . . Palliser calls this "misguided" in view of an
old game ofKorchnoi, but Black's real mistakes
6.'1Wc2 leads back to variation A3 above. came later.

6.a3 i.xc3t 7.bxc3 has been covered via the 9.hh7t 'it>xh7 IO.Wfxd5 'it>gs 1 1 .0-0 hc3
4.ll:lf3 c5 5.a3 move order - see variation D 12.bxc3 c6
of Chapter 8.
8
6.i.d3 d5 leads to the absolute main line of 7
the Rubinstein Variation, as discussed from
the start of Chapter 1 5 . 6
5
Bl) 6.d5
4
The presence of the pawn on e3 makes this 3
advance less effective, as the d5-pawn doesn't
2
receive enough support.
1
a b c d e f g h
1 46 4.e3

Interestingly, a couple more games reached B2) 6.i.d2


this position via a slightly different move order.
We will follow the example of a future World White can hardly hope to fight for the
Champion from almost 60 years ago! advantage with such a move.

13.e4 i.e6! 8
1 3 . . . Wf6 1 4.i.g5 Wg6 1 5 .:gfd 1 i.g4 1 6.lLlh4
Wfh5 1 7.f3 i.e6 ( 1 7 . . . tLle5!) 1 8.g4 i.xd5
7
1 9.gxh5 Ae6 20.:gxd6;!; was Korchnoi - Parma, 6
Soviet Union 1 965, as quoted by Palliser. Even
5
this would have been quite reasonable for
Black after the improvement noted at move 4
1 7, but the text move is better still. 3
14.f;Yh5 f6 15 ..tf4 2
Black also has good compensation after: 1
1 5 .:gd l We8!?N ( 1 5 . . . Wa5 was played in
a b c d e f g h
Knaak - Adamski, Polanica Zdroj 1 979, but
it feels strange to place the queen so far from 6 ... cxd4 7 .exd4 d5
the kingside) 1 6.Wxe8 :gfxe8 1 7.:gxd6 :gads;; White's active possibilities are limited by the
timid placement of his bishop.

8.c5
This is the only ambitious try - White is
aiming to build a solid pawn chain, seizing a
lot of space on the queenside. There are a few
alternatives:

After 8.a3 Axc3 9.ixc3 dxc4 1 0.Axc4 b6


1 1 .0-0 ib7= Black had full control over d5 in
Makoli - Doettling, Kerner 2007.

s.:gc l b6 9.cxd5 was played in Hort - Ribli,


a b c d e f g h
Manila 1 976, when the natural 9 . . . tLlxd5N
would have given Black easy play, for instance:
We have been following the game Antoshin
- Spassky, Leningrad 1 957. Now correct is: 8

7
15 ... Ve8!N 16.f;Yxe8 fxe8 17.i.xd6 b6i
6
Black has superb compensation for a pawn,
thanks to the vulnerability of White's pawns 5

and Black's control over the light squares, 4


especially c4. 3

a b c d e f g h
Chapter 1 0 - Rare 5 th Moves 1 47

I O.l2Jxd5 i.xd2t 1 1 .Wfxd2 Wxd5 1 2.i.c4 Wfe4t - Grund, Viernheim 1 99 5 . Simple and strong
1 3 .Wfe3 i.b7 1 4.0-0 l2J d7= would have been:

8 .i.d3 dxc4 9.ixc4 b6 1 0.0-0 ib7 Since


the d2-bishop is clearly misplaced, Black is
almost a tempo up compared with the main
line - see Chapter 1 6, where the bishop goes
to the much more sensible g5-square. A game
continued:

a b c d e f g h

1 0 . . . Wfd6N 1 1 .i.d3 i.xd2t 1 2.Wfxd2 l2Je4


1 3 .Wfc2 ia6 1 4.0-0 :i:l:c8 1 5 .We2 ixd3
1 6.Wxd3 l2J d7 White must fight for equality
due to his inferior pawn structure.

9 . . ..bc3 10 ..bc3 e4 n .ti'c2


a b c d e f g h
Another way to protect the bishop, 1 1 .:i:l:c l ,
1 1 .:i:l:e 1 l2J bd7 1 2.a3 ixc3 1 3 .ixc3 l2Jd5= is also well met by 1 1 . . .a5! 1 2.i.b5. We have
Potapov - Kasimdzhanov, Tashkent 2009. been following the game Cossin - Godart,
Once again, control over the d5-square assures Saint-Quentin 20 1 6, when Black should have
Black of a comfortable game. played:

a b c d e f g h

1 2 . . . bxc5N 1 3 .dxc5 id7! 1 4.a4 We? 1 5 .0-0


a b c d e f g h :i:l:c8+ With strong pressure on White's
8 ... b6 9.a3
queenside pawns.
The most consistent.
9.cxb6 Wxb6 has been played; but if White We have been following the game Sandipan
has to resort to this, it suggests that his whole - Eljanov, Doha 20 1 4. Now I suggest the
strategy is flawed. 1 O.l2Ja4 This was A. Sokolov following way of handling the position:
1 48 4.e3

16 ...d7 17.0-0 b5
8
Black has been able to set up a stable
7 blockade, so there is no reason to worry.
6
B3) 6 . .te2
5
4
3
2
1
a b c d e f g h
l l ... aSN 12.b4 axb4 13.axb4 gxal t
14.bal .ta6!
It makes sense to get rid of the passive light
squared bishop.
a b c d e f g h
1 5 ..ha6
1 5 .b5?! runs into 1 5 . . . bxc5! 1 6.Wla4 ib7 This modest-looking move has been tried
1 7.dxc5 lDxc5 1 8.Wfd4 Wff6 1 9.Wfxf6 gxf6 by many strong players, including Viktor
20.ixf6 ltJ bd7+ and the lack of development Korchnoi.
causes White serious problems.
6 d5
. .

The two main lines we will look at are


B3 1) 7.0-0 and B32) 7.a3 .
8
7.cxd5 i s now well met by 7 . . . cxd4! ( 7 . . . exd5
7 is possible, though after 8.dxc5 White has
6 reasonable chances to put pressure on the
isolated d-pawn) 8.exd4 lDxd5 9.id2 l2J c6
5 1 0.0-0 ie7 Black was fine in I. Ivanov - Suba,
4 Hastings 1 983, as the combination of White's
IQP with the passive position of the e2-bishop
3 looks really awkward.
2
7.dxc5 is harmless and rather dull after
1
7 . . . dxc4. Alternatively, to keep more life in the
a b c d e f g h position, Black could instead try 7 . . . l2J bd7! ?N.
16 . .tc3
Once again, dubious is 1 6.b5?! bxc5 1 7.0-0
l2J b4 1 8.Wfa4 l2J d3 1 9.dxc5 l2J dxc5 20.Wlb4
f6+.
Chapter 1 0 - Rare 5th Moves 1 49

B3 1) 7.0-0 dxc4 1 1 ... c5!?N


If a draw is not an acceptable result, I can also
suggest 1 L. J:b8 1 2.tLla4 ( 1 2.ig5 h6 1 3.ih4)
1 2 . . . id6 1 3.b3 tLld5 with a balanced position
- the pawns on d4 and c6 are equally weak.

12.'1!1!16 gbs 13.'!Wg3 gb7 14.dxc5 ixc5

a b c d e f g h
8.e5!?
I have to admit that this move is not without
merit: it enables White to change the pawn
structure so that both sides have a weakness.

8.ixc4 cxd4 9.exd4 transposes to the main


tabiya, as analysed in Chapters 1 5 and 1 6.

8 ... cxd4 9.exd4


So far this is Savic - Pavlovic, Valjevo 20 1 2.
Now I suggest:
1 5 ... h5 16.'1Wg4 '!Wd4 17.'1Wxh5 '!Wxc4
9 ... c6N 10.xc6 bxc6 l l ..bc4 And now White has nothing better than
We have transposed to a game, Mantilla forcing a draw by means of:
Reyes - Guo Qi, Tromso (ol) 20 1 4, where
1 1 . . . 'IW aS was played.

8
7
6

a b c d e f g h
1
1 8 ..bg7 <lt>xg7 19.'\WgSt <lt>hs 20.'!Wf6t=
a b c d e f g h
1 50 4.e3

B32) 7.a3 tough task for White after 1 0.a4 b6 l l .i.a3


i.a6 1 2.WI'c2 d8 1 3.lDe5 cxd4 1 4.cxd4 c3+
in Basson - Wellen, email 20 1 0.
8
1 o . . . cxd4 l l .cxd4
7 White is trying to regain the pawn and
6 activate the bishops, but after:

5
4
3
2
1
a b c d e f g h
This aims to avoid the isolated pawn.
a b c d e f g h
7 ...hc3t 8.bxc3 ffc7
8 . . . dxc4 might transpose to one of the most l l . . .b5! 1 2.a4 b4 1 3.ixc4 l2J fd7! 1 4.Wfe2
explored positions in the Nimw-Indian after lDxe5 1 5 .dxe5 lD c6 1 6.f4 i.b7+
9.i.xc4 l2J c6 1 0.0-0, but this is not part of our Black is better due to the strong passed
repertoire. b-pawn.

9.cxd5 9 ... exd5


I also examined:
9.0-0 dxc4 8
7
6
5
4
3
2
1
a b c d e f g h a b c d e f g h
1 O.l2Je5!?N So far, B32 1) 10.0-0 is the only move
1 0.Wa4 b6 l l .Wfxc4 i.a6 1 2.WI'a2 i.xe2 to have been tested from this position, but
1 3.Wxe2 Wl'b7 1 4.l2Je5 lD c6 was fine for B322) 10.dxc5N also deserves attention.
Black in Troncoso Flores - Iniguez, email
2008.
Regaining the pawn turned out to be a
Chapter 1 0 - Rare 5 th Moves 151

B32 1) 10.0-0

This seems less comfortable for White in view


of:

10 ... c4!
This ambitious move makes both of White's
bishops passive and yields Black a clear
superiority on the queenside.

1 1 . e5
1 l .Wfc2, restncnng the mobility of the
c8-bishop, is not effective: 1 1 . . . lLl c6 1 2.lLld2 a b c d e f g h
E1e8 1 3 .if3 ig4+ Black was clearly better in 1 I . .. tLl c6!N
Hrescak - Brkic, Split 20 1 5 . A natural novelty.
1 1 .lLld2 was played in Petschar - Exler, Austria Also possible is 1 1 . . . lLle4, as was played in
20 1 2. Since White's main goal is to advance B.A. Toth - Neagu, Calimanesti 20 1 3 . However,
the e-pawn, I suggest 1 1 . . .if5N 1 2.f3 lLl bd7 in that case White could have tried 1 2.Wfc2N
1 3.a4 ig6 1 4.l::1 e 1 E1fe8 1 5 .ifl E1e6 1 6.ia3 l::1 e8 1 3.f3 lLld6 1 4.h3 ifS 1 5 .e4! dxe4 1 6.Wa2,
l::1 ae8, mobilizing all Black's forces. Play might with definite compensation for the pawn.
continue:
12.xc6 f;Yxc6 13.Yfc2 i.g4 14.6 i.h5
Black's play seems much easier, as White
lacks a clear plan to make his bishops work.

B322) 10.dxc5N

a b c d e f g h

1 7.e4 dxe4 1 8.ixc4 e3! 1 9.ixe6 exd2


20.ixf7t ixf7 2 1 .l::1 xe8t ixe8 22.Wfxd2 if7
Even though White has a rook and two pawns
for the minor pieces, I still prefer Black due to
his full control over the light squares.

a b c d e f g h
Even though this move has yet to be tested, I
believe it is the best choice. White is trying to
open lines for his bishops.
1 52 4.e3

IO ...Yixc5 Conclusion
l O ... ll:l bd7!? l l .c4 dxc4 1 2.xc4 ll:lxc5= also
looks perfectly playable. In the position after 4.e3 0-0 White can
employ various set-ups, most of which will
l l .c4 be covered in subsequent chapters. Amongst
l l .a4 Ylxc3t 1 2.d2 Yfc7 offers White White's minor possibilities, 5.ll:\f3 c5 6.d5
some compensation for the pawn, but Black is is the most ambitious try, and it leads to
certainly not worse. interesting complications after 6 . . . d6 7.d3
exd5 8.cxd5 ll:lxd5, when the subsequent
Also harmless is l l .b2 ll:l c6 1 2.0-0 g4, novelty 1 5 ... Yie8!N makes Black's position
when Black's active piece play fully compensates quite attractive. Another tricky set-up is 5.ll:\f3
for White's bishop pair. c5 6.e2 d5, but the drawbacks of White's
rather passive placement of both bishops are
l l ...dxc4 12.Yid4 Ylxd4 13.xd4 illustrated after both 7.cxd5 cxd4! 8.exd4
White intends to restore material equality lLlxd5 and 7.a3 xc3t 8.bxc3 Yfc7 9.cxd5
and enjoy his bishop-pair advantage, but he will exd5, followed by . . . c5-c4 if the opportunity
have to sacrifice some time and coordination presents itself.
while regaining the pawn after:

a b c d e f g h
13 c3! 14.b5 c6 1 5.xc3 i.e6 16.i.b2
..

gac8 17.0-0 aS
Black is fine; the exposed light squares along
the c-file prevent White from putting his
bishop pair to good use.
8
>-=""''wC.oJ'
7
.-=-'-'=/.._;:,:;,pm
6
5
4

4.e3
lm_o/ ' " ' " ,i<"77/' ' Nmd
3
e ....." m..
2

a b c d e f g h

5.a3
Variation Index
l.d4 ttlf6 2.c4 e6 3.ttlc3 i.b4 4.e3 0-0 5.a3
5...hc3t 6.bxc3 d6
A) 7.3 154
B) 7.ttl f3 156
C) 7.ttle2 157
D) 7.i.d3 e5 8.ttle2 e4! 159
D 1) 9.i.c2 160
D2) 9.i.b 1 16 1

A) after l l .e4 B) after 7 . ltl f3 02) after I O .b3

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2

a b c d e f g h a b c d e f g h a b c d e f g h

1 J . h6!N
. . 7 . . . e5!?N I O . . . c5!N
1 54 4.e3

I .d4 6 2.c4 e6 3.c3 .tb4 4.e3 0-0 5.a3 White's play is slowed down by the early e2-e3
bc3t 6.bxc3 advance.

7 ... c5 s ..td3
The over-ambitious 8.e4 tLl c6N 9.i.g5 h6
1 O.i.h4 leaves White with undeveloped pieces.
His only real achievement is the unpleasant
pin of the f6-knight, which can be addressed
as follows:

a b c d e f g h
This way of handling the position obviously
resembles the Samisch, as White spends a
tempo to practically force the exchange on
a b c d e f g h
c3 , putting his faith in the bishop pair and
pawn centre. However, in my opinion, mixing
the 4.e3 and 4.a3 systems makes little sense 1 0 . . . cxd4! 1 l .cxd4 '1Wa5t 1 2.'it>f2 '1Wd8! The
- sooner or later White will have to push threat of . . . tLlxe4t puts White in a difficult
e3-e4 anyway. Still, it has been employed by situation.
such strong players as Carlsen, Eljanov and
Alekseev, among others. 8
7
6 ... d6
It would be a mistake to try and follow the 6
plan from Chapter 6 against the pure Samisch, 5
as we have already committed to short castling.
4
There are reasonable alternatives, but I like 3
the text move the most. Black is preparing
. . . e5 followed by activating the light-squared 2
bishop as soon as possible, all of which fits in 1
well with the early castling.
a b c d e f g h
We will look at four options: A) 7.f3, 8 ... c6 9.e2
B) 7.f3, C) 7.e2 and D) 7 ..td3. Developing the knight in a different way
doesn't bother Black: 9.tLlh3N b6 1 0.0-0 ia6
A) 7.f3 1 1 .'1We2 lLl a5 1 2.lLlf2 :gcs 1 3.d5 lLl d7!+ and
the weakness ofWhite's queenside pawns starts
In comparison with the usual Samisch, to tell.
Chapter 1 1 - 5.a3 1 55

9 ... b6 10.0-0 .ta6 1 1 .e4 1 1 ... h6!N


I also considered: Avoiding the unpleasant pin on the
1 1 .lDg3 lDa5 1 2.dxc5 f6-knight.
Hardly better is 1 2.WI'e2, as played in
Heranval - Yacob, France 2009. After 12.4 aS 13.e5
1 2 . . . cxd4N 1 3 .cxd4 l::1 c8 White loses a pawn 1 3.Wfa4 cxd4 1 4.cxd4 l::1 c 8+ also leads to the
for nothing: 1 4.e4 i.xc4 1 5 .ig5 h6+ loss of a pawn.
1 2 . . . dxc5 1 3 .WI'e2 Wfd7 1 4.E1d 1 Wa4
13 ... h7!
This allows Black to win the pawn and
consolidate.

1 3 ... lD d7 looks natural but is actually less


convincing after: 1 4.exd6 cxd4 ( 1 4 . . . i.xc4
1 5 .f5! is dangerous for Black) 1 5 .cxd4 i.xc4
1 6.f5 ltJ f6 1 7 .ixc4 lDxc4 1 8.fxe6 fxe6 1 9. WI d3
l2Jxd6 20.i.xh6 gxh6 2 1 .Wfg6t <;i;>h8 22.l2J f4
White's attack is sufficient to maintain the
balance.
a b c d e f g h

1 5 .e4
8
1 5 .ic2 WeB 1 6.i.d3 E1d8+
1 5 . . . i.xc4 1 6.i.g5 i.xd3 1 7.E1xd3 l2J d7+ 7
White had no compensation for the pawn in 6
Kunicki - Jakubowski, Wisla 2000.
5
This position has been seen in a couple of 4
games. In my opinion, the best way to prove
that losing a tempo is critical for a system as 3
sharp as the Samisch is: 2
1
a b c d e f g h
14.ti'c2
The following alternatives do not solve
White's problems:

1 4.l2Jg3 i.xc4 1 5 .ixc4 lDxc4 1 6.We2 b5+


leaves Black a pawn up with a fine knight.

1 4.f5 dxe5 collapses White's centre while


stopping his hoped-for attack, particularly as
1 5 .dxe5 i.xc4 wins a pawn while making an
a b c d e f g h
exchange of queens likely.
1 56 4.e3

14... f5 1 5 .g4 correspondence game, but it doesn't pose


White's structure is crumbling so he might Black any problems either:
as well try this lunge. 8 . . . ig4!?N 9.h3
After 9.dxe5 dxe5 1 0.l2Jxe5 ixe2 1 1 .W.xe2
1 5 ....bc4 16 ..bc4 xc4i W.d6 1 2. l2J f3 E1e8 1 3 .0-0 c5 1 4.E1d 1 W.e6
White does not have enough for the pawn. Black obtains at least sufficient compensation
for the pawn.
9 . . . ih5 1 0.dxe5
In the event of 1 0.0-0 ltJ c6 1 1 .ib2 l::1 e 8+
White's set-up seems too passive.
8
7
6
5
4
3
2

a b c d e f g h
a b c d e f g h
This move does not seem to be in the spirit 1 0 . . . l2J e4!
of the position, but it was tried by Akiba Admitting that the bishop pair is just
Rubinstein so it deserves some attention. I as significant as the weakness of White's
like the following new way of handling the doubled pawns.
position: 1 o . . . dxe5 is less convincing due to 1 l .W.xd8
E1xd8 1 2.g4 e4 1 3 . gxh 5 exf3 1 4.ixf3 when
7 .. e5!?N
.
White's bishops may become powerful.
The most dynamic, although I should l l .exd6
mention that there is nothing wrong with a I l .ib2 dxe5 1 2.W.xd8 l::1 xd8 1 3 .g4 ig6
normal continuation such as 7 . . . W.e7 followed 1 4.l2Jxe5 l2J c6 1 5 .l2Jxg6 hxg6 looks too
by . . . e5 . passive for White.
l l . ..l2Jxc3 1 2.W.d3 lDxe2 1 3 .'it>xe2 cxd6
8.d2 1 4.E1d l l2J c6 1 5 .ib2 f6
Mter 8.dxe5 dxe5 9.W.xd8 E1xd8 1 0.l2Jxe5
l2Je4 Black regains the pawn and gets excellent
play, since 1 l .ib2?! l2J a6 1 2.f3 l2J ec5 1 3.E1d 1
ie6 1 4.ie2 l2J a4 1 5 .ia 1 f6 puts White in a
difficult position.

8.ie2
This is another natural developing
move which briefly transposes to an old

a b c d e f g h
Chapter 1 1 - 5 .a3 1 57

1 6.xd6 b6 12 ...b6 13.b3 ti'xb3 14.xb3 d5 1 5 .c5


Black has good compensation in this unclear e8
position. White's bishops are ineffective, so Black is by
no means worse.
8 ... c5 9 . .te2
C) 7.e2 c5
8
7
6
5
4
3
2
1
a b c d e f g h
a b c d e f g h
9 .. exd4!?
.

An interesting concept: Black is trading the With White's knight heading for g3, it makes
better pawn structure for active piece play, sense to attack the c4-pawn as soon as possible.
while getting rid ofWhite's space advantage.
8.g3
10.cxd4 cxd4 l l .exd4 c6 12 ..tb2 The ambitious 8.d5N isn't supported by
Mter 1 2.d5 lLl d4 1 3 .0-0 if5 1 4.i.b2 b6 White's undeveloped pieces, so after: 8 . . . tLl bd7!
Black forces an exchange of the opponent's 9.dxe6 tLle5 1 0.exf7t E1xf7
dark-squared bishop and gains some positional
advantage.

a b c d e f g h

1 1 .tLlg3 ( l l .tLlf4 lLl e4! 1 2.i.d3 tLlxd3t


1 3 .xd3 i.f5+) 1 l . . .i.e6 1 2 .i.e2 Axc4 1 3 .f4
i.xe2 1 4.xe2 tLl c6 1 5 .0-0 d? Black regains
a b c d e f g h the pawn and gets the superior position due to
his better pawn structure.
1 58 4.e3

8 ... c6 9 ..td3 1 6 . . . llJe8!


The lack of development prevents White That's the point! The e8-knight is the key
from developing any initiative on the kingside defensive piece now.
after: 1 7.llJh5 ixd3 1 8.Wfxd3 Wid?+
9.e4 b6 1 0.ie3
1 0.ig5 h6 1 1 .h4 cxd4 1 2.cxd4 was seen
in Oestreich - Zeitler, Buschhuetten 1 967.
Now the correct 12 . . . hxg5!N 1 3 .hxg5 g6!
1 4.gxf6 Wfxf6 1 5.llJe2 ia6 1 6Jk 1 tJig?+
would have put White in a difficult position.
1 0 . . . ia6 1 l .id3

a b c d e f g h
9 ... b6 10.0-0 .ta6 l l .Wfe2 c8!
Since the d4-d5 advance isn't possible,
it's better to keep the tension for as long as
a b c d e f g h
possible.
This was played in Averbakh - Matanovic, 1 1 . . .cxd4 1 2.cxd4 c8 1 3 .ib2 llJ a5 1 4.ac l
Rijeka 1 963, and should be met by: would allow White to consolidate and protect
1 1 . . .llJa5N 1 2.Wfe2 cxd4 1 3 .cxd4 Wfc7! the c4-pawn; 1 4 . . . d5 1 5 .cxd5 ixd3 1 6.Wfxd3
A precise move, preparing to put the king's Wfxd5 is still okay for Black, but the text move
rook on c8. poses more problems to White.
Instead, 13 . . . c8 1 4.cl Wfc7 1 5 .0-0
ixc4 1 6.ig5 llJ d7 1 7.llJh5 offers White 12 ..tb2
interesting play for a pawn. Obviously the dark-squared bishop is
1 4.c l feB 1 5 .0-0 ixc4 1 6.ig5 misplaced now, but White has to get ready
to defend the c4-pawn, and 1 2.d5? llJe5 is
horrible for him.

12 ... a5 13.d5
1 3 .ac l d5 1 4.cxd5 ixd3 1 5 .Wfxd3 exd5+
led White to a passive position in Golz -
Radovici, Ploiesti 1 957.

13 ... exd5 14.c:x:d5


This position was reached in Ribeiro -
a b c d e f g h Belem, Fortaleza 1 95 1 . Now Black should
have exchanged White's light-squared bishop:
Chapter 1 1 - 5 . a3 1 59

secure the c5-spot for the knight. 1 1 .d5 ic8


1 2.0-0 llJ bd7 1 3 .llJ g3 tD c5 White's position is
extremely risky from a strategic point of view.)
9 . . . llJxe4 1 o.tlJf3

a b c d e f g h
14 ... .ixd3N 1 5.Yixd3
And then blocked the other one with: a b c d e f g h

15 ... c4! 16.Yif5 g6 17.Yif3 ge8 1 8.e4 b3 1 o . . .f5 1 l .Yfc2 Wle7 1 2.0-0 tlJd7 1 3 .a4
19,gadl c5+ c5 1 4.tlJd2 tlJ df6 Black's play in the game
Black has strong pressure against the Eidelson - Minogina, Vitebsk 1 985, was
e4-pawn, and the c5-knight threatens to j ump highly instructive. White was doomed to
to d3 at any moment. passive defence and eventually lost.

D) 7 ..td3 Mter 8.e4 Black may take advantage of his


development advantage by means of: 8 . . . exd4
9.cxd4 llJxe4! 1 0.llJe2 ( l O.ixe4 :i:l:e8+ Tissari
- Veingold, Jyvaskyla 1 999) 1 0 . . . d5 1 1 .0-0
Now in Ghyselen - Bomans, Westerlo 20 1 0,
Black should have played:

a b c d e f g h

1 l . . .:i:l:e8!N 1 2.Vflc2 if5 1 3 .cxd5 Wlxd5 1 4. tlJ f4


7 ... e5 8.e2
Wld7 Intending to exchange the bishops by
8.f3 doesn't stop the e-pawn: 8 . . . e4! 9.fxe4
means of 1 5 . . . tlJd6. White has insufficient
(I also examined: 9.ic2N :i:l:e8 1 0.llJe2 ie6!
compensation for the pawn.
It's no loss to expend a tempo in order to
1 60 4.e3

1 4.E1f2 ltlh5 1 5 .Wfd4 was seen in Perez Perez


- Gragger, Marianske Lazne 1 96 1 , and now
Black should have played:

a b c d e f g h
8 ... e4! a b c d e f g h
This advance is part of Black's restricting
1 5 . . . ltle5!N 1 6.'it>h 1 ( 1 6.f4 ll:l g4 1 7.E1f3 Wh4
strategy - it makes both White's bishops look
1 8.id2 ig2-+) 1 6 . . . c5 1 7.dxc6 ll:lxc6 1 8.Wd5
poor.
g6+ Securing a big positional advantage.
The two lines to look at are of course
The attempt to protect the c4-pawn with
D l ) 9.i.c2 and 02) 9 ..lb l .
1 0.ib3N fails after: 1 0 . . . ltl c6! 1 1 .ltlf4 (The
tactical justification for Black's last move is
Dl) 9 ..lc2 .le6
1 1 .d5 ltle5 1 2 .dxe6 ltl d3t 1 3 .'fl ll:lg4+ and
White is in trouble; the text move is not much
I like this natural developing move. Protecting
of an improvement though . . . )
the c4-pawn would slow down White's
counterattack.

a b c d e f g h

1 1 . . .ic8 1 2.f3 ltl a5 1 3.ia2 ( 1 3.0-0 ll:lxb3


1 4.Wxb3 b6+) 1 3 . . . b6 1 4.0-0 ia6+ White's
a b c d e f g h
opening strategy has clearly failed.

10.g3 10 ... bc4 1 1 .xe4 xe4 12.be4


1 O.d5 is, once again, a positional concession. White has been able to get rid of the
1 0 . . . ig4 1 1 .0-0 ll:l bd7 1 2.f3 exf3 1 3.gxf3 ih3 e4-pawn, but it is still not easy to activate the
Chapter 1 1 - 5 .a3 161

dark-squared bishop. Moreover, the king is 02) 9.-tb l


stuck in the centre, so there is no way to avoid
the exchange of light-squared bishops, which
8
will favour Black.
7
6
5
4
3
2
1
a b c d e f g h
This is the most common retreat. Most of
White's pieces are located on the first rank now,
a b c d e f g h
but it's easier to protect the main weakness: the
12 ... d5 1 3 . .td3 .bd3 14.f;Yxd3 d7 1 5.c4 c4-pawn.
I also considered 1 5 .0-0N ltl b6 1 6.e4 dxe4
1 7.'1Mfxe4 c6, when Black gets some positional 9 ... .te6
advantage due to his control over the light The same concept as in the preceding line.
squares.
9 . . . b6!? is another decent way to handle the
.. i s position.

&(t-ilB .iV-Wi
8
1 %
.....% ....%
IO.Vb3

T
. . . . . .

6 1 0.ll:lg3 .ixc4 1 1 .ltlxe4 ll:lxe4 1 2 . .ixe4 leads


back to variation 0 1 above, where 1 2 . . . d5
5 >'//.'' gives Black the better game.
4 ---- -
8
m

". 'l.-
r'm%'m"
The ugly-looking 1 O . .ia2, as was played in
3 - %. Einarsson - Vidarsson, Hafnarfjordur 1 996,

w[l!j
2 can be met strongly by: 10 . . . ll:l c6!N We saw
...
exactly the same idea in the note to move
1 . I] : 1 0 in the previous variation. The critical line
a b c d e f g h continues: 1 1 .d5 ltle5 1 2 .dxe6 tDd3t 1 3 .tJifl
We have been following the game Podolny ll:l g4
Furman, Vilnius 1 949. Now Black could have
put White under some pressure by means of:

1 5 ... b6!N 16.cxd5 Vxd5 17.0-0 c5i


1 62 4.e3

Mter 1 l .Wfxb7 tLl bd7 1 2.d5 i.g4 1 3.tLlg3


tLl b6 1 4.Wa6 e8+ Black's pieces are
dominating and White's extra pawn is of no
value.

1 l .d5 is also ineffective after 1 l . . .i.c8 1 2.tLlg3


e8+.

l l ... c6 12.d5 aS 13.ti'a2 .td7

a b c d e f g h

1 4.tLlf4 tLl gxf2 1 5 .e7! White needs to keep


the f-file closed. 1 5 . . .'1Wxe7 1 6.Wfd2 lLlxh 1
1 7.tLlxd3 exd3 1 8.'kt>g 1 tLl g3 1 9.hxg3 Wg5+
Despite the approximate material balance,
Black's position is preferable due to White's
hideous pawn structure.

Here I found a useful improvement over some


existing games.

Conclusion

The character of play in this chapter is similar


to the Samisch Variation, but the inclusion of
e2-e3 and . . . 0-0 has merits for both sides. Black
has committed his king so ideas such as the ig5
a b c d e f g h pin have the potential to be more dangerous,
IO ... cS!N but White's early e2-e3 may represent a loss of
Black has usually preferred 1 O . . . b6, dating all a tempo if the pawn subsequently advances to
the way back to the game Kotov - Pachman, e4. It seems to me that 6 . . . d6 followed by . . . e5
Moscow 1 946. However, I discovered that is the best way of neutralizing White's bishops.
there is no need to spend a tempo safeguarding Once this pawn reaches the e4-square it will be
the b-pawn. hard for White to develop any activity on the
kingside, whereas Black's positional advantages
l l .g3 are long-lasting.
4.e3
a b c d e f g h

Variation Index
l.d4 tLlf6 2.c4 e6 3.tLlc3 i.b4 4.e3 0-0 5.tLlge2
5.. Je8!?
A) 6.g3 164
B) 6.a3 i.8 165
B 1) 7.tLl f4 165
B2) 7.g3 d5 8.cxd5 exd5 9.i.g2 a5 10.0-0 tLla6 168
B2 1) 1l.i.d2 169
B22) 1l.'i'c2 170
B3) 7.e4 d5! 8.e5 tLl fd7 170
B3 1) 9.c5?! 17 1
B32) 9.cxd5 172
B4) 7.tLlg3 d5 175
B4 1) 8.cxd5 175
B42) 8.i.e2 177
B5) 7.d5 a5! 180
B5 1) 8.g3 1 80
B52) 8.tLlg3 tLla6 181
B52 1) 9.i.e2 181
B522) 9.i.d3 183
1 64 4.e3

l .d4 f6 2.c4 e6 3.c3 .th4 4.e3 0-0 placement of the d2-bishop by means of:
s.ge2 8 . . . lLl bd7N 9.ig2 lLl b6 1 0.0-0 c6 1 l .b3
This move was introduced and deeply if5 1 2.f3 ia3+
explored by one of the greatest players of the
past: Akiba Rubinstein. White avoids doubled
c-pawns and keeps the pawn chain flexible.
The drawback is that the development of the
light-squared bishop is somewhat delayed,

s .. Je8!?
This move was introduced back in 1 937,
but it has become fashionable again in recent
years. The main continuation is 5 ... d5 6.a3
ie7, whereas the text move secures a more
a b c d e f g h
comfortable square on f8 for the bishop's
retreat. Of course, the rook move costs a 8 . . . c5!
tempo and does not immediately occupy the Since the d-file is blocked, this advance is
centre, so we have to consider various attempts more effective than usual.
by White to gain space in that area. 9.a3 ixc3 1 0.ixc3 lLl c6 1 1 ./ib5
Too passive is 1 1 ./ie2?! c4 1 2.0-0 b5+.
1 1 . . ./ig4 1 2.'1Wd2 cxd4 1 3 .ixd4 lLlxd4
8
1 4.'1Wxd4 '1Wa5t 1 5 .'1Wb4 '1Wxb4t 1 6.axb4 l::1 e 6=
7 The weakness of the b4-pawn yields Black
6 sufficient counterplay.

5 A) 6.g3 d5
4
3
2
1
a b c d e f g h
We will start by considering A) 6.g3, which
is an attempt to deviate from the main line of
B) 6.a3 .

6./id2
For some reason, this natural-looking move
a b c d e f g h
has been seen in only one tournament game.
6 . . . d5 7.cxd5 exd5 8.tLlg3!?N 7. .tg2
8.g3 was played in Aleksandrov - Lopez White should probably return to the usual
Martinez, Lugo 2009. Now Black could paths with 7.a3 /if8, as covered in variation
have exploited the somewhat awkward B2.
Chapter 1 2 - 5 . tLl ge2 1 65

After 7.cxd5?! xd5! 8.gg1 h5 9.h3 e5


1 0.g4 h4+ White was suffering from a lack
of development in Bluvshtein - Onischuk, White's set-up resembles the well-known
Montreal 2009. theoretical line 5 . . . d5 6.a3 i.e? 7.cxd5 exd5
8.lLlf4, where the pressure on the d5-pawn
7 ... dxc4 8.0-0 might be annoying sometimes, especially
I also examined 8.a4N a5 9.a3 id7 if White manages to advance the g-pawn.
1 0.c2 i.c6 1 1 .i.xc6 i.xc3t 1 2.xc3 tLlxc6 Therefore, I suggest:
1 3 .xc4 e5+, when White manages to regain
the pawn, but faces serious new problems. 7... d6!
Rather than fixing the central structure with
8 ... c6 9.c2 bd7 10.e4 b6 l l .xf6t . . . d5, Black sets up the possibility of . . . e5 . Now
xf6 the f4-knight looks rather awkwardly placed.

8
7
6
5
4
3
2
1
a b c d e f g h a b c d e f g h
12.e4 s ..td3
No better is: 1 2.f4 e7 1 3 .a3 ia5 1 4.e4 c5 8.lLlh5, as was played in Shirazi -
1 5 .i.e3 cxd4 1 6.lLlxd4 i.d7+ Shchekachev, Paris 20 1 4 , looks artificial, and
the simple 8 . . . lLlxh5N 9.xh5 e5 1 0.d5 e4
12 ... e5 13 ..te3 exd4 14 ..ixd4 e7i offers Black a better position.
Black remained a pawn up in Giorgadze -
Bacrot, Groningen 1 997. White has also tried: 8.i.e2 e5 9.tLlfd5
(releasing the tension in the centre is a definite
B) 6.a3 concession: 9.dxe5 dxe5 1 0.xd8 gxd8
1 1 .lLlfd5 lLl e8! 1 2.0-0 lLla6 1 3.gd 1 i.e6+
This is the most natural and consistent choice. Black was better in Schandorff - P.H. Nielsen,
Aalborg 2006) In Lutsko - A. Zhigalko, Minsk
6 ...-tm 2008, a good continuation would have been:
Now we have a major branching point; we
will look at Bl) 7.4, B2) 7.g3, B3) 7.e4,
B4) 7.g3 and BS) 7.d5 .
1 66 4.e3

9 . . . cxd5 1 0.cxd5 e5 1 1 .lLlh5 lLlxh5 1 2.Wfxh5


g6 1 3 .Wfd 1 ll:ld7 1 4.e4 offers White some
space advantage.
1 0.0-0 cxd5 1 1 .cxd5 e5

a b c d e f g h

9 . . . ll:lxd5N 1 0.cxd5 exd4 1 1 .Wfxd4 g6 1 2.0-0


ig7 1 3 .Wfd 1 c5! It makes sense to get rid of
the weak c7-pawn or neutralize White's space
advantage. 1 4.dxc6 Otherwise Black gets a a b c d e f g h
pleasant version of a Benoni. 1 4 . . . ll:lxc6 The
activity of Black's pieces fully compensates for 1 2.lLlh5 e4 1 3 .lLlxf6t ll:lxf6 1 4.i.e2 if5
the isolated pawn. I prefer Black, who has definite attacking
prospects.
8.d5!?N
This hasn't been played yet, but it's a typical 8 ... e5
idea and it seems like White's most ambitious Now the most challenging move is:
try. I think the best response is:
8
7
6
5
4
3
2

a b c d e f g h
1

8 . . . c6 9.i.d3 a b c d e f g h
Too passive is: 9.ie2 cxd5 1 0.cxd5 e5 9.fe2N
1 1 .lLlh5 ll:lxh5 1 2.ixh5 e4 1 3.0-0 g6 Previously 9.dxe5?! dxe5 1 O.ll:lfd5 was
1 4.i.e2 ig7+ played in Moor - Mantovani, Switzerland
Black is also doing well after 9.dxe6 fxe6 2005 . Now Black could have exploited the
1 0.e4 e5 1 1 .lLlh5 ll:lxh5 1 2.Wfxh5 ie6. lack of harmony in the opponent's camp by
9 . . . ll:l bd7! means of:
The most precise.
Chapter 1 2 - 5 . lLl ge2 1 67

10 ... exd4 1 1 .xd4 xd5

a b c d e f g h

1 0 . . . tlJfd7!N 1 1 .0-0 c6 1 2.lLlb4 aS 1 3 .lLlba2


tlJ c5+ With a big positional advantage.

9 ... d5! a b c d e f g h
I don't like releasing the tension in the centre
12.hl7t
with 9 . . . exd4 1 0.exd4 d5, since after 1 l .c5 b6
The simple 1 2.tlJxd5 Wfxd5 1 3 .0-0 is
1 2.b4 a5 1 3 .E1b 1 the space advantage is on
unthreatening, as Black has easy development
White's side.
after, for example, 1 3 . . . lLl c6.

The vulnerability of Black's centralized knight


enables White to grab a pawn with the text
move, but Black gets full compensation.

12 ... <.txh7 13.ti'h5t <.tgs 14.ti'xd5 c5

8
7
6
5
a b c d e f g h
10.cxd5 4
This leads to some tricky tactics. 3

1 0.dxe5 allows an exchange of queens after 2


1 0 . . . dxc4 1 l ..ixc4 Wfxd l t 1 2.tlJxd 1 E1xe5= 1
when Black is fine.
a b c d e f g h
1 0.tlJxd5 lLlxd5 1 1 .cxd5 can hardly pose any 1 5.lLlde2 c6 16.ti'xd8 gxd8 17.e4 .te6
problems after: 1 l . . .exd4 1 2.e4 ( 1 2.lLlxd4 Black has excellent play for a pawn, due
Wfxd5 1 3. 0-0 lLl c6 1 4.lLlxc6 Wfxc6 1 5 .b4 .ie6 to the powerful bishops and control over the
1 6 . .ib2 l::1 ad8=) 1 2 . . . tlJ d7 1 3 .0-0 lLl c5 Black b3-square.
has plenty of activity.
1 68 4.e3

B2) 7.g3 9 ... a5 10.0-0


The ultra-aggressive 1 O.h3 c6 1 1 .g4 is
well met by 1 1 . . .h5! 1 2.g5 llJ e4 1 3 .h4. We
have been following the game Vyzmanavin -
Mochalov, Belarus 1 983, where Black missed
a strong idea:

a b c d e f g h
7... d5 8.cxd5
a b c d e f g h
Instead 8 .ig2?! dxc4 9.WI'a4 seems dubious
White manages to regain the pawn, but it takes
a long time: 9 . . . llJ bd7 1 0.WI'xc4 e5 1 1 .0-0 e4 1 3 . . . llJ d6!N 1 4.llJg3 g6+ Black is firmly in
1 2 .b3 llJ b6 1 3 .Wfc2 if5+ As a result, Black control.
was able to seize a lot of space in I. Sokolov -

Kryvoruchko, Antalya 20 1 3 . IO ... a6


A typical manoeuvre: Black restricts White's
s ... exd5 9.ig2 activity on the queenside and prepares to meet
Once again, the posmon resembles the the thematic f2-f3 advance with . . . c7-c5.
well-known theoretical line with 5 . . . d5 6.a3
ie7 7.cxd5 exd5 8.g3, where White's main plan
is to prepare the e3-e4 advance. Obviously, in
our case Black benefits from applying pressure
along the e-file.

a b c d e f g h
White may proceed with B2 1) l l .id2 or
B22) I I .f;Yc2.
A minor alternative is 1 1 .llJf4 c6. Provoking
the . . . c6 advance in this way is hardly a
a b c d e f g h significant achievement for White, as we
Chapter 1 2 - 5 . lLl ge2 1 69

can see after the further: 1 2.f3 0. c7 1 3.e4?! White's play in the following game seems too
A tempting but incorrect advance. 1 3 . . . c5! slow: 1 2.h3 0.c7 1 3.Wfc2 0. e6 1 4.<kt>h2 g6
White has lost control of the centre and found 1 5 .:gad 1 Aleksandrov - Lopez Martinez, Lugo
himself in an inferior position after: 2009. Now Black could afford some aggressive
measures:

3
2 r=ru
h

a b c d e f g h
a b c d e f g h
1 4.e5 ( 1 4.ie3 dxe4 1 5.0.xe4 c4 1 6.if2 b5+)
14 . . . cxd4 1 5.Wfxd4 tt::l d7 1 6.tt::l d3 :gxe5 1 7.f4 1 5 . . . b5!N 1 6.0.a2 Wfc7 1 7.0.ac l b4 1 8.axb4
:ges 1 8.tt::l xd5 0.xd5 1 9.Wfxd5 Wfb6t 20.<it> h 1 axb4 1 9.0.d3 ia6+
0.f6+ S . Bekker Jensen - Hammer, Borup 2008.
12 ... c7 13.a4
B2 1) l l .i.d2 c6 In the event of 1 3 .f3 0. e6 1 4 .ie 1 , intending
to regroup the pieces and push e3-e4, Black is
able to seize the initiative with energetic play:
14 . . . b5! 1 5 .if2 b4 1 6.axb4 axb4 1 7.0.a4 ia6
1 8.:ge 1 ib5+

8
7
6
5
4

a b c d e f g h 3
12Jcl 2
As was mentioned earlier, 1 2.f3N should 1
be met by 1 2 . . . c5 , and then 1 3.dxc5 0.xc5
a b c d e f g h
1 4.0.d4 id7 offers Black comfortable play.
13 ... e6 14.b4 e4 1 5 .i.el axb4 16.axb4
Black did not face any problems after 1 2.0.f4 d6i
if5 1 3.f3 h5 1 4,:gf2 b5 1 5 .ifl id6 in Milov Black had valuable control over the light
- Alekseev, Santo Domingo 2003. squares in Volkov - Morozevich, Moscow 2005.
1 70 4.e3

B22) I l .'i;Yc2 c6 A typical confrontation in the centre had


occurred in Reilly - Makarov, Yerevan 2004.
Black should have kept the tension by means
8
of:
7
6
5
4
3
2
1
a b c d e f g h
12.f3
Instead 1 2.id2 I:J c7 1 3.l::1 ad 1 I:J e6 1 4.ic l
a b c d e f g h
seems too slow. It may look as if White had
reached an optimal set-up in Pogosian - 13 ... .td7N 14.g4
Moiseenko, Olginka 20 1 1 , but Black could Or 1 4.1:Jf4 E1c8 1 5 .Wff2 ic6, with double
have seized the initiative with a familiar plan: edged play.

14... h6 1 5.h4?!
I will show this move just to illustrate what
happens if White is over-ambitious.

15 ... h7! 16.1:Jxd5 'i;Yxh4 17.b6 gad8i


White is being punished for exposing his
king.

B3) 7.e4
a b c d e f g h

1 4 . . . b5!N 1 5 .b3 ia6 1 6Jfe 1 b4 1 7.axb4 axb4


1 8 .1:Ja4 ib5+ White's position turns out to be
quite passive.

12 ... c5
Once the g2-bishop has been blocked, Black
is better placed to accept the creation of an
IQP. Moreover, the e3-pawn provides a target
for counterplay.

13Jdl
Chapter 1 2 - 5 . tLl ge2 171

There hardly can be a more tempting move


- White's pawn centre looks optimal, and the
dark-squared bishop is now freed and ready
to pin the opponent's knight. However, this
advance provokes the following confrontation
in the centre.

7 ... d5! 8.e5


White's set-up looks rather ugly after 8.i.g5
dxe4 9.lLlxe4 i.e7 l O.i.xf6 i.xf6+ - it will be
difficult to maintain the pawn centre in the a b c d e f g h

long run. 1 3 . . . tLl c6!N (a clear improvement over


1 3 . . . bxc5 1 4.bxc5 i.b7, which led to unclear
8.cxd5 exd5 9.e5 lLl fd7 transposes to variation play in Khismatullin - Zherebukh, Voronezh
B32 below. 2008) 1 4.Wa4 i.b7 1 5 .b5 lLl cxeS! 1 6.dxe5
bxc5 White's position will collapse soon.
8 ... tLlfd7
Now White can attempt to squeeze Black on l l ...axb4 12.axb4 bxc5 13.bxc5 i.a6!
either side of the board with B3 1) 9.c5?! or Now it's obvious that White's king will be
B32) 9.cxd5 . stuck in the centre for a long time.

B3 1) 9.c5?! b6 10.b4 aS
8
7
6
5
4 I. . . . . . . /W//,r.. .

3
WN//r.. , . . . .

2
1
a b c d e f g h

14.g3?!
a b c d e f g h
1 4.f4 would be the lesser evil, but even then
This only invites serious trouble for White, after 1 4 . . . tLl c6 1 5 .i.e3 f6 1 6.i.f2 ic4 White
as we can see from tournament practice. would be under strong pressure.

l l Jbl 14 ...hfl 1 5.xfl


Even worse is: l l .i.e3?! axb4 1 2.axb4 xa 1 1 5 . tJixfl f6+ does not help White.
1 3 .'1Wxa 1
15 ... 6 16.f4 c6 17.g3 fxe5 1 8.fxe5 Yfh4
19.i.e3
1 72 4.e3

We have been following the game l O.lLlxd5?! c5 only leads to a loss of time, and
Konstantinov - Arsic, Paracin 20 1 5. Black's White will not remain a pawn up for long.
easiest and most effective way to break through Play might continue: l l .lLlec3 cxd4 1 2.Wixd4
would have been: ll:lc6 1 3.Wid l lLldxe5 1 4.ie2

8
7
6
5
4

3
2
a b c d e f g h
1
1 4 . . . ge6! 1 5 .0-0 gd6+ and the pin on the
a b c d e f g h d5-knight causes White a lot of problems.
l9 ... tldxe5!N 20.dxe5 d4
With a winning position at an early stage of The attempt to limit Black's active play by
the game. means of I O.b4?! is not effective either: I O . . . a5
l l .b5 c5 1 2.bxc6 bxc6 1 3.g3 c5 1 4.ig2 cxd4
B32) 9.ad5 exd5
1 5.ll:lxd4 ll:lxe5 1 6.0-0

8
7
6
5
4
3
2 a b c d e f g h

1 1 6 . . . ig4! 1 7.Wib3 ic5 1 8.ie3 ixd4


a b c d e f g h
1 9.ixd4 lLlf3t 20.ixf3 ixf3+ White had
no compensation for the pawn in Le Roux -
IO.f4 Oleksienko, Guingamp 2007.
Over-protecting the key e5-pawn seems like
White's best idea. IO ... c5 l l ie3
l l .g3 ll:lc6 1 2.ie3 transposes to the note to
l O.ie3 c5 1 1 .f4 is another possible move White's next move.
order.
Chapter 1 2 - 5 . ttJ ge2 1 73

1 1 .dxc5!? of the kingside pieces by means of 1 2.g3 as in


This should be met by: Baginskaite - V. Ni, Saint Louis 20 1 3 , fails to:
1 1 . . . l2J c6!N 1 2 . . . l2J b6!N 1 3 .i.g2 l2J c4 1 4.if2
Less precise is 1 1 . ..l2Jxc5 1 2.b4 l2J e6
1 3.l2Jxd5 a5 1 4.b5 i.c5 1 5.Ae3, when Black
still had to prove his compensation for the
pawn in Sipila - B. Socko, Lisbon 20 1 4.

a b c d e f g h

1 4 . . . i.g4! 1 5 .ixd5 l2Jxd4 and Black manages


to develop a powerful initiative, since 1 6.ixc4
is impossible: 1 6 . . . l2J f3t 1 7.tJifl ih3#
a b c d e f g h

1 2.b4?
White should prefer 1 2.i.e3 , transposing to
our main line.
The greedy 1 2.Wfxd5? l2Jxc5 1 3 .Wfxd8 :gxd8
1 4.lDg3 lD b3 1 5 .:gb 1 i.c5+ also invites a lot
of trouble.
1 2 . . . a5 1 3.:gb 1 axb4 1 4.axb4 l2J dxe5! 1 5 .fxe5
if5 1 6.:gb3 d4
Black regains the piece with great effect!

8
7 a b c d e f g h

6 12 ... xc5
Weaker is 1 2 . . . i.xc5?! 1 3 .i.xc5 lDxc5 1 4.b4
5
d4 1 5 .bxc5 dxc3 1 6.Wfxd8 :gxd8 1 7.lDxc3 ,
4 when claiming full compensation for the pawn
3 turns out to be a tough task.

2 13.xd5
1 Apparently, 1 3.b4 is a safer way to maintain
the balance. Play continues 1 3 . . . d4! 1 4.l2Jxd4
a b c d e f g h
l2Jxd4 1 5 .Wfxd4 Wfxd4 1 6.ixd4 l2J b3 1 7.:gd 1
l l ... c6 12.dxc5 a5! , when White's lack of development
An attempt to complete the development prevents him from keeping the extra pawn:
1 74 4.e3

1 8.ll:ld5 axb4 1 9.axb4 (bad is 1 9.ll:lc7? as in 14 ... e4


Pantzar - Haug, Stavanger 20 1 6, in view of There is one more interesting tactical
19 . . J:'d8!N 20.ll:lxa8 ll:lxd4 2 1 .axb4 ll:l c2t resource: 1 4 . . . ll:lxe5!?N 1 5 .ixc5 :i:l:c8!
22.'e2 ig4t-+) 19 ... ie6 20.ie3 :i:l:ed8
2 l .ic4 ixb4t 22.tJif2 lLla5 23.ia2 if8
24.ll:lc3= With an equal endgame.

a b c d e f g h

1 6. fxe5 (after 1 6.ie3 ?! ll:l c4 1 7.:i:l:c3 b5! White


would be almost paralysed) 1 6 . . . ixc5 1 7.b4
:i:l:xe5 1 8.:i:l:xc5 :i:l:xc5 1 9.bxc5 :i:l:xd5 20.Wfc l
We7 The lack of harmony in White's camp
a b c d e f g h offers Black full compensation for the piece.
13 .tg4! 14Jcl
15.dc3
...

The following game vividly illustrates


the potential danger of White's position: A desperate attempt to simplify matters.
1 4.h3? ie6 1 5 .:i:l:c l ll:l e4 1 6.ll:ldc3 We have
been following the game Shulman - Milov, 8
Oak Brook 2007. Now Black missed a great
7
opportunity to develop a crushing attack:
6
5
4
3
2
1
a b c d e f g h
1 5 Yih4t
a b c d e f g h
.

I like this way of handling the position, as it


1 6 . . .'1Wh4 t!N 1 7 .g3 ll:lxg3 1 8.if2 :i:l:ad8! and allows Black to keep the position more tense.
there is no space for White's pieces: 1 9.Wfc2 Instead, 1 5 . . . ic5 1 6.Wfxd8 :i:l:axd8 1 7.ixc5
( 1 9.Wfa4 ll:lxe5 20.fxe5 Wxa4 2 1 .ll:lxa4 ll:lxh 1 -+) ll:lxc5 1 8.ll:lg3 f6 would regain the pawn
19 ... ll:l d4! 20.ll:lxd4 Wxf4-+ and liquidate into an approximately equal
endgame.
Chapter 1 2 - 5 . lLl ge2 1 75

16.g3 h6! 17.lthe4 7 ... d5


1 7.i.g2 gadS 1 8.'1Wc2 lLlxc3 1 9.Wfxc3 Wg6 The two main lines to consider are
maintains Black's initiative. B41) 8.cxd5 and B42) 8 ..te2.
8.b4 a5 9.b5 c5 1 0.dxc5 was tried in Graf
- Barsov, Tashkent 1 987. Now Black should
have solved his development problems with:

a b c d e f g h

a b c d e f g h 1 O . . . lLl bd7!N 1 l .i.b2 (inferior is 1 l .cxd5?!


20.fxe5 xe3 2 1 .d4 xd4 22.cxd4 gxe5t lLlxc5 1 2.i.b2 exd5 1 3.i.e2 a4 1 4.0-0 ie6+)
23 . .te2 gae8 1 1 . . . ltlxc5 And White's lack of development
In this unbalanced position Black did not forces him to fight for equality.
face any problems in Gonzalez Vidal - Ortega,
Santa Clara 20 1 4. B41) 8.cxd5 exd5 9 ..td3

B4) 7.g3 9.ie2 b6 transposes to variation B42 below.

a b c d e f g h
White is not trying to refute Black's set-up
with this move, but instead aims to complete
development and build upon his slight space
advantage.
1 76 4.e3

9 ... c5 IO.dxc5
1 0.0-0 ltl c6 1 1 .ic2, as was played in
Kuruppu - Karas, Gyor 20 1 4, should be
met by: 1 1 .. .ig4N 1 2.f3 ie6 1 3 .ltlge2 :gc8
1 4.id2 cxd4 1 5 .exd4 ltl a5+

1 o ... .bc5 1 1 .0-0


1 1 .b4 id6 1 2.ib2 ie5 1 3.0-0 ltl c6
transposes to the main line below.

a b c d e f g h

1 5 . . . d4!?N 1 6.b5
1 6.e4 a5 1 7.b5 ltle5 leads to double-edged
play.
1 6 . . . ltl a5 1 7.exd4 b6
The passive placement of his pieces prevents
White from benefiting from his extra pawn:
1 8.ltlc4
Also level is: 1 8.Wa4 ie6 1 9.:gac l Wfd6
20.:gfe 1 :gad8=
18 ... ig4! 1 9.lLle2 lLlxc4 20.ixc4 :gc8 2 1 .id3
ltld5 22.Wfd2 ixe2 23.ixe2 ltl c3 24.ig4
a b c d e f g h :gc7=
l l ... c6 12.b4
It doesn't make much sense to postpone
development with 1 2 .ltlh5 ie6 1 3.ltlf4 -
after 1 3 . . . d4! 1 4.ll:lxe6 :gxe6 1 5 .ltl a4 if8
1 6.exd4 Wfxd4 1 7.ie3 Wfh4+ Black was
better in Lilienthal - Sokolsky, Kuibyshev
1 942.

12 ....td6 13 ..tb2 .te5


The fight over control of the blockading
square (in this case, the d4-square) is one
of the prevailing themes in positions with
an isolated pawn. As practical tests prove,
White cannot claim any advantage from this a b c d e f g h
one. 14 ... h5!
Exploiting the awkward placement of the
14.d2 g3-knight and developing a dangerous
Instead a classic game continued: initiative on the kingside.
1 4.ll:la4 ixb2 1 5 .ltlxb2 In my opinion, the lack of development
Gligoric - Bronstein, Zurich 1 953. Now I of the queenside pieces is a good reason for
recommend the following pawn sac: avoiding 1 4 . . . d4 1 5 .exd4 ixd4 1 6.:gad l .
Chapter 1 2 - S . lLl ge2 1 77

For instance, after 1 6 . . . Ag4 1 7.lLlce2 Axb2 Better was 1 6.c;i;>h 1 h4 1 7.ltJfS AxfS 1 8.Axf5
1 8.WI'xb2 Wl'e7 1 9.h3 Ad7 20.ltlf4:t White d4 1 9.lLlbS, maintaining the balance.
maintains a small but stable advantage.
16 ...d4! 17.exd4 h4 18.e2 h3 19.g3
15.4 1 9 .g3 is a better try, although 1 9 . . . Ag4
1 S .:i:l:ad 1 !?N is less drastic way to handle the 20.:i:l:ae l WfdS 2 1 .1':1f2 :i:l:ad8 still gives Black a
position. Still, Black's position seems perfectly promising initiative.
playable after: 1 S . . . h4 1 6.ltlge2 h3 1 7.g3 Ae6
19 ... xd4
White was in trouble in Michalik - Swiercz,
Czech Republic 20 1 2.

B42) 8 .te2

8
7
6
a b c d e f g h 5
1 8.lLlf4 (or 1 8.Ab l :i:l:c8 1 9.f4 Axc3 20.Axc3 4
Ag4, with plenty of counterplay) 1 8 . . . d4
1 9.lLlxe6 :i:l:xe6 20.ltle4 dxe3 2 1 .ltlxf6t Axf6 3
22.fxe3 Wl'b6 The position remains dynamically 2
balanced.
1
1 5 ....tc7 a b c d e f g h
This was Watson's recommendation in A
Strategic Chess Opening Repertoire for White,
but he only mentions 8 . . . cS for Black. I don't
like that move at all, since 9.dxcS AxeS 1 0.b4:t
followed by Ab2 gives White easy play, while
Black has wasted too much with his bishop,
causing a delay in the development of the
queenside pieces. Instead I would like to
recommend:

8 ... b6!
This move not only aims for a fianchetto, but
also makes the counterattacking . . . cS advance
a c e g
more effective.
16.a4?! 8 . . . dxc4 9.0-0 cS 1 0.dxcS AxeS 1 l .Axc4
A dubious decision - White cannot afford to Wl'xd l 1 2.:i:l:xd l ltl bd7, as played in Fier
move the knight away from the centre. - Harikrishna, Eppingen 20 1 S , seems
1 78 4.e3

insufficient for equality in view of: 1 3 .ie2!N 1 2.b5 cxb5 1 3.llJxb5 llJ c6, when the weak
a6 1 4.llJge4 i.e? 1 5 .llJd6;!; c4-spot in White's camp yields Black excellent
counterplay.
9.cxd5
9.0-0 c5 1 O.dxc5 bxc5 1 1 .e4?! ( l l .cxd5 exd5 IO ... c5
would transpose to our main line below) was With White's kingside knight on g3 instead
played in Manolache - Cheparinov, Pontevedra of the normal 3, Black has more reason to
2004. Black should have continued: create hanging pawns.

a b c d e f g h

1 1 . . .dxe4!N 1 2.llJgxe4 llJxe4 1 3.llJxe4 i.b7


1 4.llJc3 llJ c6+ Black's split queenside pawns a b c d e f g h
are not weak at all, and his knight will be l l .dxc5
excellent on d4. The attempt to keep the tension fails to pose
Black any problems: 1 1 .i.b5N i.d7 1 2.a4
9 ... exd5 cxd4 1 3 .exd4 a6 1 4.i.xd7 Wfxd7 1 5 .ig5 llJ e4
1 6.llJgxe4 dxe4 1 7.l::1 e 1 llJ c6 1 8.d5 llJ b4=

Another possible attempt to put pressure on


the hanging pawns is:
1 1 .i.f3 i.b7 1 2.dxc5 bxc5 1 3.b3
1 3.e4 is harmless: 13 . . . dxe4 1 4.llJcxe4 llJxe4
1 5 .Wfxd8 E1xd8 1 6.llJxe4 llJ c6 1 7.llJc3 E1d7=

a b c d e f g h
10.0-0
1 0.b4 c6 1 1 .0-0, as was played in
Schepetkova - Charochkina, Satka 20 1 5 ,
can be comfortably met by 1 1 . . . i.a6N
a b c d e f g h
Chapter 1 2 - 5 . lLl ge2 1 79

1 3 . . . d7! 13 ...i.a6!N
Protecting the bishop and getting ready to A natural way of handling the position.
break open the centre.
1 4.i.b2 The previously played 1 3 . . . i.b7?! 1 4.i.f3;!; was
1 4.a2 is also well met by 1 4 . . . d4! 1 5 .i.xb7 too passive for Black in Berovski - Dobrev,
xb7 1 6.exd4 cxd4 1 7.lLlce2 lLl c6 1 8.d2 Shumen 1 995.
ab8 1 9.b4 a5= and White cannot avoid an
exchange of all the queenside pawns. 14J:el
14 . . . d4! 1 5 .i.xb7 xb7 1 6.exd4 cxd4 1 7.lLlce2 Over-optimistic is: 1 4.i.f3?! i.xfl 1 5 .xfl
1 7.xd4 xb3 1 8.lLlce4 lLl bd7 is slightly tLl d7! (instead 1 5 . . . tLl c6 1 6.i.xd5 c8 1 7.i.d2
better for Black. offers White more compensation) 1 6.i.xd5
b8+

14 ... c6!
This pawn sacrifice allows Black to complete
development and exploit the light-square
weaknesses in the opponent's camp.

a b c d e f g h

1 7 . . . lLl c6
The passed d-pawn is strong, so White has
nothing better than:
1 8.lLlxd4 lLlxd4 1 9.i.xd4 adS 20.a2 e6
2 l .d2 i.xa3=

l l ... bxc5 1 V h5 xh5 1 3 ..bh5


a b c d e f g h

8 1 5 .xd5
1 5 .xd5 lLle5 1 6.i.e2 i.xe2 1 7.xe2 b8
7
leads to a double-edged battle.
6
5 1 5 .i.f3 leads to a similar character of play:
1 5 . . . lLle5! 1 6.i.xd5 b8 1 7.e4 h4! The
4 c l -bishop is tied to protecting the b2-pawn,
3 whereas Black has a few dangerous ideas, such
as transferring the rook to h6.
2
1 1 5 ...i.c4 16.c3 b6 17.a4 i.b3i
White experiences serious problems with the
a b c d e f g h
development of his queenside pieces.
1 80 4.e3

BS) 7.d5 1 1 .cxd5 b6! 1 2.0-0 i.a6


White suffers from the unpleasant pin on
8 the e2-knight and the vulnerability of the
d3-square.
7
6
5
4
3
2
1
a b c d e f g h
a b c d e f g h
7 ... a5!
Securing the excellent square c5 for the 1 3 .ga2
knight, where it will also put unpleasant Hardly better is: 1 3.a4 tLl fe4 1 4.lLlxe4 lLlxe4
pressure on the weakened b3-spot. 1 5 .Wfc2 lLl c5+ Steingrimsson - Grandelius,
Reykjavik 20 1 4.
The two main options are BS I) 8.g3 and 1 3 . . . lLl d3
B52) 8.g3. I like this concept - after liquidating the
c l -bishop, Black will have control over
8.lLld4 This natural-looking move has never many dark squares.
been seen in practice. The vulnerability of the 1 4.gd2 lLlxc l 1 5 .Wfxc l b5
d5-pawn should tell after: 8 . . . exd5 9.cxd5 tLl a6
1 0.i.e2 c6 1 1 .0-0 lLl c7 1 2.dxc6 dxc6+

BS I) 8.g3

This means of development looks most


consistent - the d5-pawn will be well
protected, and the fianchetto will allow White
to castle quite soon.

8 ... a6 9.i.g2 c5 10.0-0


White can instead attempt to keep the pawn a b c d e f g h
structure more flexible with: 1 6.tLla2?!
1 0.b3 1 6.gfd 1 N b4 1 7.axb4 axb4 1 8.tLla4 would
This was recommended by Lars Schandorff have maintained the balance.
in Playing J . d4 - 1he Indian Defences, but it 1 6 . . . b4+
can be strongly met by: Black had the edge in Hammes - Brunner,
1 0 . . . exd5 Skopje 20 1 5 .
This move was not considered by the Danish
GM.
Chapter 1 2 - 5 . lLl ge2 181

Black excellent play in Dubov - Jakovenko,


Ekaterinburg 20 1 3 .

B52) 8.g3

a b c d e f g h
IO ... a4 l l .Yic2 d6 12.i.d2 i.d7
Black should also be okay after 1 2 . . . lLl b3
1 3 .gad l exdS 1 4.cxd5 id7.

13.dxe6
Releasing the tension in the centre looks This is the most popular way to handle the
like a concession, but I do not see a better position. The knight is freeing the way for the
alternative. bishop and taking control of the important
e4-square.
The natural 1 3.gad 1 ?! c6! 1 4.e4 exd5 1 5 .exd5
cxd5 1 6.cxd5 tLl g4 leads White to a passive 8 ... a6
and inferior position. Now the two main lines are B52 1) 9.i.e2
and B522) 9.i.d3.
13 ... .be6
White cannot afford to delay development
8 even more with 9.gb 1 ?! in view of 9 . . . exd5
1. -=C7. /''Hm. , . . . . . .
1 0.cxd5 c6! 1 l .ie2 b5 1 2.dxc6 dxc6, and
7
White was already in trouble in Berezjuk -
6 Markos, Czech Republic 2002.
5
I also examined 9.b3 exd5 1 0.tLlxd5 ( l O.cxd5
4 c6 1 1 .ie2 b5!+ would force White to release
3 the tension in a worse situation) 1 0 . . . lLlxd5
1 l .cxd5 c6 1 2.ie2 tLl c7 1 3 .dxc6 dxc6 1 4.Wfc2
2 ie6, when Black has excellent play.
1
B52 1) 9.i.e2 c5 10.0-0
a b c d e f g h
14.d5 c6 1 5.xf6t Ylxf6 16.d4 i.d7 There are a couple of serious alternatives:
The vulnerability of the c4-pawn and
ideal placement of the c5-knight afforded
1 82 4 .e3

1 0.b3
This prophylactic move can be met by:
1 0 . . . a4
In my opinion, liquidating the dark-squared
bishop is the most challenging reply.
1 l .b4 ltl b3 1 2.1:h2
Or 1 2.:i:l:b 1 lLlxc l 1 3 .:i:l:xc l exd5 1 4.cxd5
c5! 1 5 .bxc5 ( 1 5 .0-0 cxb4 1 6.axb4 ixb4
1 7.ltlxa4 Wla5 1 8.Wfd4 Wlxa4 1 9.:i:l:c4
Wla5 20.:i:l:xb4 Wfxd5= leads to major
simplifications) 1 5 . . . ixc5 1 6.0-0 Wla5 with a b c d e f g h

promising counterplay. 1 2 . . . g6
1 2 . . . ll:lxc l 1 3 .Wfxc l Another attractive idea is 1 2 . . . :i:l:xe3!?N
We have been following the game Thingstad 1 3 .fxe3 Wfe7, when Black has excellent
- Predojevic, Norway 20 1 6. In my opinion, compensation due to his superb control over
Black should have activated the dark-squared the dark squares.
bishop by means of: 1 3 .0-0 ig7 1 4.:i:l:b 1 id7 1 5 .b4 axb4 1 6.axb4
ll:l fe4 1 7.ltlgxe4 ll:lxe4 1 8.ll:lxe4 :i:l:xe4+
The activity of both Black's rooks put
White in an unpleasant situation in Zueger -
Greenfeld, Munich 1 987.

a b c d e f g h

1 3 . . . exd5N 1 4.cxd5 c5! 1 5 .0-0


1 5 .bxc5 ixc5 1 6.0-0 d6 is also okay for
Black.
1 5 . . . cxb4 1 6.axb4 ixb4 1 7.:i:l:xa4 :i:l:xa4
1 8 .ll:lxa4 Wfa5
Black has counterplay. a b c d e f g h

Another natural continuation is: 10 ... a4


1 O.e4 d6 1 1 .ie3 Black is ready to exchange the important
Mter 1 1 .0-0 exd5 1 2.exd5 g6 1 3 .Wfc2 ig7 dark-squared bishop, so White's next move is
1 4.id2 a4+ Black's position seems preferable almost forced.
due to the awkward placement of the
l l .:i:l:bl
g3-knight.
In the event of 1 l .e4 ltlb3 1 2.:i:l:b 1 ll:lxc l
1 1 . . .exd5 1 2.exd5
1 3.:i:l:xc l exd5 1 4.exd5 g6 1 5 .ltlxa4 h 5 ! the
strong dark-squared bishop offers Black
Chapter 1 2 - 5 .ltJ ge2 1 83

sufficient compensation for the pawn, at the White will have to expend another tempo after
very least. . . . ll:la6-c5 .

This position was reached in Kuzubov - 9 ... c5 IO.i.c2 exd5 l l .cx:d5


Pavlov, Nakhchivan 20 1 6. Now I suggest the White is ready to complete development
following plan of activating all Black's minor and push the central pawns, but . . .
pieces:
l l ... b5! 12.xb5
Grabbing the pawn is the most principled
8
reply.
7
6 After 1 2.0-0?! ib7 1 3 .WI'd4 b4 1 4.axb4 axb4
1 5 Jl:xa8 Wl'xa8 1 6.WI'xb4 ll:lxd5 1 7.WI'h4 h6
5 Black was better in Ipatov - Eljanov, Dubai
4 20 1 4.
3
2
1
a b c d e f g h
l l ...i.d6!?N 12.e4 exd5 13.exd5 i.e5
14.i.e3 d6
The poor placement of the g3-knight and
Black's control over the e-file puts White in an
uncomfortable situation.

B522) 9 ..td3

a b c d e f g h
13 ...'\WbS!?N
I like this ambitious new approach.

The previously played 1 3 . . . c6 1 4.b4 ltl b7


1 5 .dxc6 dxc6 1 6.WI'xd8 l::1 exd8 1 7.l::1 b 1 axb4
1 8 .axb4 ll:l d6 1 9.f3 ltl b5 20.ll:lge2 ll:lxc3
2 1 .ltlxc3 E1db8 led to an equal endgame in
Aleksandrov - Granda Zuniga, Pavlodar 20 1 5 .

14Jbl
a b c d e f g h 1 4.ll:lge2 i.d6 1 5 .0-0 c6 1 6.g3 ie5 puts
A double-edged continuation. The bishop White under strong pressure - both of Black's
is placed more actively here than on e2, but bishops are extremely powerful.
1 84 4.e3

14 ... a4! 16 ... cx:d5 17J:el i.c4 18.xa4


It's vital to secure the c5-knight. After 1 8.llJd4?! llJ fe4 1 9.llJxa4? E1xa4!
20.ixa4 .id6 the massive attack by all Black's
15.ge2 forces decides the game on the spot:
And certainly not 1 5 .e4? as White cannot
afford such an 'active' move: 1 5 . . . c6! 1 6.dxc6
d5 and Black's attack is almost decisive.

a b c d e f g h

2 l .g3 llJxf1! 22.'xf1 .ixg3t! 23.hxg3 llJ e4t-+

18 ... x:a4 19 ..ixa4 .txa3


The activity of Black's pieces fully
a b c d e f g h compensates for the slightly inferior pawn
1 5 ... c6 16.0-0 structure.
It's better not to be too greedy.
Conclusion
The following line illustrates well White's
troubles with development: 1 6.dxc6? dxc6 The Rubinstein System is a perfect weapon for
1 7.0-0 l::1 d 8 1 8 . .id2 '1Wd6 1 9 . .ie 1 positional players and leads to a long strategic
battle, where both players have different plans
at their disposal. In comparison to the main
line with 5 . . . d5 6.a3 i.e?, my recommendation
of 5 . . . E1e8 6.a3 if8 gives Black a more flexible
set-up, but delaying the . . . d5 advance yields
White some aggressive options. However, since
7.e4 is strongly met by 7 . . . d5!, and 7.d5 offers
Black promising play on the dark squares, it
may be that White has nothing better than
playing patiently with 7.g3 or 7.llJg3 . The
arising positions are strategically complex, but
a b c d e f g h if Black chooses the correct moment to initiate
1 9 . . . '\Wc? 20.'1Wc l llJ d3 2 l ..ixd3 .ixd3 22.l::1 a 1 counterplay with . . . c5 (or, in some variations,
llJg4! White's awful development makes i t . . . b5), he will have no cause for complaints.
impossible for him t o defend his kingside,
for instance: 23.f4 ic5 24.E1f3 Wb6 25 .id2
l::1 e 8-+
4.e3
s.id3
Variation Index
l.d4 t[} 6 2.c4 e6 3.tLlc3 i.b4 4.e3 0-0 5.i.d3
5...d5
A) 6.tLlge2 dxc4 7.hc4 c5 1 86
A1) 8.a3 186
A2) 8.0-0 cxd4 9.exd4 tLlc6 187
A2 1) 10.a3 189
A22) 10.i.g5 190
B) 6.cxd5 exd5 7.tLlge2 e8 192
B 1) 8.i.d2 193
B2) 8.0-0 i.d6 195
B2 1) 9.tLl f4 195
B22) 9.i.d2 196
B23) 9.a3 198
B24) 9.3 198

A2) note to 1 Oth move o p tions A22) after 1 5 .e4 B l ) after 1 3 .b3

8
1
6
5
4
3 Wlf1'"""f%';m
///A/' ' "" ///C'/A
2

a b c d e f g h a b c d e f g h

' 'Ld6!N 1 5 . . . f5!N I 3 . . . ti:l a5!N


1 86 4.e3

l .d4 f6 2.c4 e6 3.c3 i.b4 4.e3 0-0 s.i.d3


This is White's most popular continuation.
The bishop comes to its most active square,
while the knight may go to e2 or f3 .

s ... ds
Taking space in the centre is Black's most
popular reply.

5 . . . c5 is a major alternative which sometimes


leads to the same positions after a subsequent
. . . d5, while wholly independent options
include 5 . . . b6 and 5 . . . d6. a b c d e f g h
White must choose between AI) 8.a3 and
A2) 8.0-0.

AI) 8.a3 i.xc3t 9.xc3

9. bxc3 transposes to variation C of the next


chapter.

9 ... cxd4 IO.exd4 c6


White has nothing better than protecting
the isolated pawn with:

a b c d e f g h 8
In this introductory chapter to the 5 .i.d3 7
set-up we will focus on A) 6.ge2 and 6
B) 6.cxd5 .
5
6.a3 will be discussed in the next chapter, 4
while the most important 6.tLlf3 is covered in
Chapters 1 5 and 1 6. 3
2
A) 6. ge2 dxc4
1
Another common variation is 6 . . . c5 7.cxd5 a b c d e f g h
cxd4 8.exd4 lLlxd5 9.0-0 tLl c6, but I don't I I .i.e3
like the position after 1 0.tLlxd5 exd5 1 1 .tLlf4, Of course, such a modest method of
when White can claim a risk-free edge. development doesn't suit White's attacking
ambitions, so Black gets comfortable play. By
7.i.xc4 c5 the way, the text move actually transposes to
Chapter 1 3 - 5 .id3 1 87

another well-known theoretical line of the


Nimzo where it is usually White to move after
4.e3 c5 5.l2Jge2 cxd4 6.exd4 d5 7.a3 ixc3t
8.l2Jxc3 dxc4 9.ixc4 l2J c6 1 0.ie3 0-0. That
particular line is slightly favourable to White,
but in our version the extra tempo means that
Black has no problems whatsoever.

l l ... b6 12.0-0 ib7 13.ia2


Mter 1 3.'1Wd3 h6 1 4J:ad 1 c8 1 5.ia2
lDe7 Black was already better in Gruenberg
Enders, Dresden 1 98 5 .
a b c d e f g h
1 3 .i.d3 lD e7 1 4.i.g5 , as played i n Christiansen 1 8 ...hg2! 19.xe6t xe6 20.be6t h7
- Hjartarson, Dubai (ol) 1 986, can be 2 l .xg2 f4t 22.g3 xe6
comfortably met by 1 4 . . . l2J ed5N with equal Black was doing well in Wiedenkeller -
chances. Carlsson, Borlange 20 1 4.

13 ... e7 A2) 8.0-0


Once again this move plays an important
role, preventing White from opening the
position with d4-d5 .

a b c d e f g h
8 ... cx:d4 9.exd4 c6
a b c d e f g h This position with the isolated d4-pawn
is similar to the main tabiya from Chapters
14 ..ig5 g6 15.d5 1 5 and 1 6. However, the somewhat modest
Finally it comes, but White has to exchange placing of the e2-knight severely limits White's
the powerful dark-squared bishop after: attacking possibilities.

15 ... h6! 16.i.xf6 xf6 17.dxe6 fxe6 1 8.d7 The two main moves are A21) 10.a3 and
A prophylactic move such as 1 8.f3 can be A22) IO .ig5, but I will also mention a few
.

met by 1 8 . . . ad8, when Black's piece activity alternatives:


matters more than the isolated e-pawn.
1 88 4.e3

1 0.if4 id6 1 1 .Wfd2 ixf4 1 2.Wxf4 l:i:J e7 Positions with an isolated pawn always contain
Forget any notion about White benefitting strategic risk, so White cannot afford slow
from the exchange of his 'bad' bishop. Such play. For instance, 1 0.h3 b6 1 1 .ie3 ib7 1 2.a3
exchanges almost always benefit the side was seen in the high-level game Vaganian -
playing against the isolated pawn, and in the Tukmakov, Reggio Emilia 1 987. Now Black
following example Black soon took control: should have played:

a b c d e f g h a b c d e f g h

1 3.gad 1 b6 1 4.gd3 ib7 1 5 .gh3 l:i:Jg6 1 6.Wfe3 1 2 . . . id6!N 1 3.ia2 ( 1 3.d5? l:i:J a5 1 4.ia2
Wd6+ Hoensch - Knoedler, Germany 1 987. exd5+ leads to the loss of a pawn) 1 3 . . . gc8
1 4 .ig5 (once again, 1 4.d5 fails to solve
1 0.Wfd3 The queen's transfer to the kingside White's problems: 1 4 . . . exd5 1 5 .1:i:Jxd5 l:i:J a5+)
is one of White's main attacking ideas in such 14 . . .ie7 1 5 .gcl l:i:Jd5 1 6.ixe7 l:i:J cxe7+
positions. However, in this variation it has a
concrete drawback: 1 0 . . . ti:Ja5! That's the point! Another pretty harmless continuation is:
Black manages to liquidate the light-squared 1 0.ie3 id6 1 l .l:i:Jg3 b6
bishop and thus limit White's attacking There are only two moves left before Black
potential. can complete the perfect set-up with
1 2 . . . ib7 and 1 3 . . . ti:J e7.

a b c d e f g h
a b c d e f g h
1 1 .ib5 id7 1 2.ig5 ie7 1 3.gfd 1 ixb5
1 4.Wxb5 a6 1 5 .Wfd3 l:i:J c6 White had to 1 2.ti:Jh5
take care in order to equalize in Graf - Weaker is 1 2.Wfe2?! ixg3 1 3 .hxg3 l:i:Jxd4+
Mchedlishvili, Batumi 2002. and White had no compensation for the
pawn in Aarland - Ekeberg, Oslo 2008.
Chapter 1 3 - 5 .id3 1 89

I also checked 1 2.d5N exd5 1 3 .lLlxd5 lLl g4!? on e2. In particular, now 1 1 .ig5 ? doesn't work
1 4Jk 1 ie6, with a sharp position where in view of 1 1 . . . ixh2t!.
Black is not worse.
1 2 . . . ib7 l l .Yid3
Once again, slow play will not suffice: 1 1 .h3
h6 1 2.ie3 ll:l e7 1 3.Wfd3 id7 1 4.:ihd 1 ic6+
and Black was better in Alatortsev - Bukhtin,
Moscow 1 976.

l l . .. b6
This move is sound and reliable, but in my
main line it has the drawback of allowing a
forced draw.
To avoid that outcome, I also examined:
1 1 . . . e5!? This aggressive reply is connected
a b c d e f g h with an interesting pawn sacrifice. 1 2.d5 e4!
1 3.lLlxf6tN 1 3.lLlxe4 lLle5
1 3.a3 is rather slow, and 13 . . .ie7 1 4. ll:l f4
Wlb8!? (intending . . J:!:d8) was good for Black
in Christensen - Skorna, corr. 200 1 .
1 3 . . . Wfxf6 1 4.W/g4
And of course not: 1 4.ll:le4? ixh2t 1 5.tJixh2
Wfh4t 1 6.tJig1 Wfxe4+
1 4 . . . Wfg6=
Black doesn't face any problems.

A2 1 ) 10.a3
a b c d e f g h

8 1 4.ll:lxf6t Wfxf6 1 5.Wfb3 ll:lxc4 1 6.Wfxc4 id7


7 1 7.ll:lg3 l:'!:fe8 Black's powerful bishops provide
full compensation for the pawn.
6
5
4
3
2
1
a b c d e f g h
10 ...id6!
There is nothing wrong with 1 0 . . . ie7, but I
prefer to focus on the drawbacks of the knight
1 90 4.e3

12J:Ml .tb7 13.f;Yh3 h6


8
This is not the only move, but it's the simplest
way to prove that Black has no theoretical 7
problems. 6
14 . .txh6N 5
If White does not go for the draw then 4
he is more likely be worse than better. For
example, Rakhmanov - Georgiadis, Dubai 3
20 1 5 , continued 1 4.l2Jg3 and now the 2
simple 1 4 . . . l2J e7N would have given Black a
1
comfortable position.
a b c d e f g h

8 10 ...i.e7 l l .a3
This is a multi-tasking prophylactic move:
7
White secures a good spot for the retreat of
6 the c4-bishop and restricts the mobility of the
c6-knight.
5
4 As before, 1 1 . "W d3 can be well met by:
3 1 1 . . . lD a5! 1 2.i.b5 a6 1 3.i.a4 b5 1 4.i.c2
g6 1 5 J:l:ad 1 i.b7+ - White has lost time
2 preventing the bishop from being exchanged,
1 so Black has been able to make a lot of progress
on the queenside.
a b c d e f g h
14 ... gxh6 1 S.f;Yxh6 g4 The strategical danger of having an isolated
I don't see much point in risking 1 5 . . . l2J h7?! pawn is well illustrated in the following
1 6.l2Je4 lD e7 1 7.d5! with a strong attack. It's example: 1 l .:i:l:c l b6 1 2.a3 i.b7 1 3 ."Wd3 :i:l:c8
probably still a draw with best play, but it's 1 4.i.a2 In Fish - Dinstuhl, Germany 2007,
messier and more dangerous. If you wish to Black should have continued:
avoid the early draw, then go for the alternative
at move 1 1 rather than this.

16.Vh5 f6 17.Vh6
With a draw by repetition coming soon.

A22) lO .tgS

The most popular and logical move, developing


the last minor piece and pinning the knight.
a b c d e f g h

1 4 . . . l2Jd5! 1 5 .ixe7 lD cxe7+ Once again, Black


Chapter 1 3 - 5 .id3 191

benefits from the exchange of dark-squared 14 . .txd5


bishops and solid control over the blockading White relinquishes the bishop pair, banking
d5-square. on his more active pieces.

Also too slow is: 1 1 .Wfd2 b6 1 2J::! ad 1 ib7 As usual, the bishop trade with 1 4.ixe7?!
1 3 .a3 1:'k8 1 4.ia2 lLl cxe7+ favours Black.

Black also has nothing to worry about in the


event of: 1 4.ic l lLlxc3 1 5 .tLlxc3 (inferior is
1 5 .bxc3 ?! lLl a5 1 6.ia2 c8 1 7.fe 1 Wfc7+ and
White was suffering from a weakened pawn
structure in Nadanian - Khademi, Teheran
20 1 4) 1 5 . . . if6 1 6.tLle4 lLl a5 1 7.ia2 idS
1 8.ixd5 exd5= Khudiakov - Alexeev, Alushta
2007.

a b c d e f g h 8
1 4 . . . tLld5! 1 5 .ie3!? White tries to keep the 7
position complex by avoiding a straightforward 6
bishop exchange, but now Black can get
active on the queenside: 1 5 . . . tLla5! 1 6.Wfd3 5
Wfd7 1 7J'k 1 ?! fd8+ De Vita - Womacka, 4
Olomouc 2004.
3
2
1
a b c d e f g h
14 ...hg5 1 5.ie4
Clearly inferior is: 1 5 .ia2?! lLl e7 1 6.d5
What else? Now in Sarosi - Tompa, Hungary
2005, Black could have played:

a b c d e f g h
l l ... b6 12.Wfd3 ib7 13,gadl
Now, with the d4-pawn protected, White is
ready to target the opponent's king with Wfh3.

13 ... d5
This is the right moment to release the tension!
a b c d e f g h
1 92 4.e3

1 6 . . . tLlxd5N 1 7.lLlxd5 ixd5 1 8.ixd5 Wfxd5 B) 6.cxd5 exd5


1 9.Wfxd5 exd5 20Jhd5 l::1 fe8+ and Black is
better. This early fixing of the central structure leads
to a completely different type of struggle.
We have reached a critical position from the
game Torre - Petursson, Bid 1 98 5 . Black
8
should have continued:
7
6
5
4
3
2
1
a b c d e f g h
7.ttlge2
Even though this position was first seen back
a b c d e f g h
in 1 932(!) , this system gained some popularity
1 5 ... 5!N in the 1 980s, due to the efforts of Grandmaster
This weakening move allows Black to gain Mikhail Gurevich. At the moment, White's
an important tempo for consolidation. set-up looks somewhat passive, especially the
dark-squared bishop. However, the plan of
1 5 . . . h6?! 1 6.d5 exd5 17 .ixd5 leaves Black f2-f3 and e3-e4 could lead to serious attacking
under strong pressure. potential, so Black has to watch out and be
ready to counter with a timely . . . c5.
16 .tf3 f;Yd7

The e6-pawn is no weaker than the one on Chasing the bishop would be premature:
d4. 7.a3 id6 8.tLlb5 ie7 9.Wfc2 a6!
I like this concrete approach, but there is
17JUel gadS also nothing wrong with 9 . . . c6 1 O.tLlc3 l::1 e 8.
The queen on d3 is clearly misplaced, so
White has to spend more time moving it.

I 8.ti'h i gres I9.g3


1 9.d5? would be a serious mistake due to
1 9 . . . tLl e5+.

19 ... g6 ! 20.d5 exd5 2 1 .hd5t hs


It will not be easy for White to demonstrate
compensation for the opponent's powerful
bishops.
a b c d e f g h
Chapter 1 3 - 5 .id3 1 93

1 0.ll:lc3
After 1 0.lLlxc7?! a7 1 l .id2 id6 1 2.c l b6
the knight would be trapped. Even though
White can grab two more pawns with
1 3 .ll:lxd5 lLlxd5 1 4.ixh7t tJih8 1 5 .id3
c7+, Black is still better.
1 0 . . . c5 1 l .dxc5 ixc5 1 2.ll:lf3
Too artificial is 1 2.ll:lce2 ia7 1 3 .id2
ll:l c6 1 4.ic3 e8, and Black was better in
Rakhmanov - lturrizaga Bonelli, Linares
20 1 4. a b c d e f g h
1 2 . . . ll:l c6 1 3 .0-0 ig4 1 2.ixe4 ( 1 2.Wfc2 if5 1 3 .Wfd l ll:l c6 1 4.0-0
White's pieces were not well placed for Wfd7+) 1 2 . . . xe4 1 3 .b3 cxb3 1 4.Wfxb3 ll:l c6
fighting against the isolated pawn in Diekers 1 5 .id2 b6 1 6J'k 1 ib7 1 7.0-0 ia6 Black is
- Rother, Bavaria 2002. at least equal.

7... e8 Bl) 8.id2


This is the most common move for good
reason: it maintains flexibility by putting the
rook on obviously the correct square while
leaving it until later to define Black's queenside
pawn structure, in particular the destination of
the c-pawn.

a b c d e f g h
Delaying castling in this way can be well met
by:

8 c5!?
. .

a b c d e f g h
In general, I prefer not to exchange the
dark-squared bishop for the knight on c3
The two main moves to consider are in this central structure, but in this specific
Bl) 8.id2 and B2) 8.0-0. situation it is the best way to prove that the
In the event of 8.Wfc2 the immediate break in d2-bishop is misplaced.
the centre seems effective: 8 . . . c5!? 9.a3N (after
9.dxc5 lLl c6 1 0.0-0 ixc5 1 l .a3 ib6 Black had 9.0-0
excellent play in Shengelia - Berkes, Hungary Completely toothless is: 9.dxc5 ixc5
20 1 0) 9 . . . ixc3t 1 0.Wxc3 c4 1 l .ib 1 ll:l e4 1 0.lLlb5 ( 1 0.0-0 ll:l c6 1 l .cl a6=) 1 0 . . . ll:l c6
1 94 4.e3

1 U'k 1 i.b6 1 2.0-0 i.g4 1 3.ic3 a6 1 4.ll:lbd4 Reykjavik 1 992, can be strongly met by
ll:lxd4 1 5 .exd4 ll:l e4+ Milanovic - Ki. Georgiev, 1 0 . . . c4!?N l l .i.b l a6 1 2.f3 b5+ when Black's
Skopje 20 1 6. play is faster.

9.a3!? might transpose to our main line, but 1 o ..bc3 l l .bc3 c4


..

it also has some independent significance: This advance makes White's dark-squared
9 . . .i.xc3 1 0.i.xc3 c4 l l .i.c2 bishop passive again.

12 ..tc2 .tg4!
Once again Watson does not cover this
. . . i.g4 idea, giving instead 1 2 . . . a5, which
occupies a square I want to reserve for a piece.

13.b3
Or 1 3 .f3 ih5 1 4.Wfd2 i.g6? gives Black
pleasant play.

This interesting posmon was reached in


a b c d e f g h
Stocek - Navara, Havlickuv Brod 20 1 5 . In my
l l . . . i.g4! Since the f5-square is not free for the opinion, Black should have played:
bishop, it's important to find another way for
it to get into the game. John Watson does not
mention this move in A Strategic Chess Opening
Repertoire for White. 1 2.f3N ( 1 2.0-0 ll:l c6
transposes to the main line below) 1 2 . . . i.h5
1 3 .'it>f2 i.g6 1 4.g4 ll:l c6 1 5 .h4 b5 1 6.ll:lf4 a5
The queenside pawn majority starts to gather
momentum, so Black is by no means worse.

a b c d e f g h
13 ... a5!N 14.b4
1 4.i.xa5 Wfxa5 1 5 .bxc4 dxc4 1 6.f3 i.d7
1 7.e4 b5 leads to double-edged play with
roughly equal chances.

14 ... b3!
This brave move is connected with a
a b c d e f g h
positional pawn sacrifice.
9 . c6 IO.a3
..
Instead, 1 4 . . . ll:l c6?! 1 5 .f3 i.h5 1 6.b5 ll:l b8
1 0J'k 1 as in Conquest - Stefansson, 1 7.Wfd2 yields White a promising position.
Chapter 1 3 - 5 .id3 1 95

15 ..bb3 cxb3 16Jbl e4 17Jxb3 launch an attack like this, but it's good to
know that the ideas exist, and White will still
have to worry about them.

The most challenging options are: B2 1) 9.4,


B22) 9.id2, B23) 9.a3 and B24) 9.f3.
9.tlJg3 can hardly bother Black: 9 . . . c6
1 0.3 c5 1 1 .dxc5N ( l l .tlJce2?! tlJ c6 was just
worse for White in Shariyazdanov - Lukacs,
Budapest 1 996) 1 1 . . .ixc5 1 2.'it>h 1

a b c d e f g h
17 ...ti'g5 1 8.ti'd3 gac8
Black has full compensation for the pawn,
due to his control over the c-file and the light
squares.

B2) 8.0-0 a b c d e f g h

1 2 . . . ixe3 1 3 .ixe3 :gxe3 1 4.ie4 tlJ c6 1 5.tlJxd5


tlJxd5 1 6.Wfxd5 Wfxd5 1 7.ixd5 tlJ e7 1 8.ic4
ie6=

B2 1) 9.4 c6 IO.f3

Normally when the pawn arrives on f3 , Black


will be quick to counter with . . . c5. Here the
d5-pawn is under pressure, so another plan is
needed.

The most natural and common move.

8 ...id6
The bishop has little to do on b4, especially if
Black is planning to challenge the opponent's
centre with . . . c7-c5 . In addition, its presence
on d6 yields some attacking resources, such as
. . . tlJg4 or even . . . ixh2t followed by . . . tlJg4t.
In most cases we will not be in a hurry to
a b c d e f g h
1 96 4.e3

10 ... b6! B22) 9 ..td2


Solving the problem of the passive bishop
on c8. 8
l l .a3 7
This was played in the only game where 6
Black's last move was played, but it is rather
slow.
5
4
1 l .e4? simply loses a pawn: 1 1 . . .dxe4 1 2.fxe4
3
i.xf4 1 3 .i.xf4 '1Wxd4t 1 4.<;i;>h 1 ig4 1 5 .'1Wd2
gds+ 2
1
1 1 .'1Wc2N is a natural attempt to improve; still,
after 1 l . . .ia6 1 2.gd 1 ixd3 1 3 .'1Wxd3 'We? a b c d e f g h
1 4.id2 ttl bd7? Black has no reasons to worry. This flexible move develops a piece and
postpones the f2-f3 advance for a moment.
l l ... .ta6 1 2.ha6 xa6 1 3.Yid3 c7 Now I like:
Having gained no advantage from the
opening, White played too optimistically with: 9 ... b6!?
This move is rare, but I find it an attractive
plan for developing counterplay.
8
7 10.4
An alternative is: 1 0.gc1 N c5 1 1 .tLlb5 i.f8
6
1 2.dxc5 bxc5 1 3 .i.c3 lLl bd7 1 4.'1Wd2
5
8
4
7
3
6
2 5

1 4

a b c d e f g h 3

2
14.e4? L4 1 5.Lf4 e6 16 ..te5 dxe4
17 .fxe4 g4+
White was in trouble in Gavrish - Aroshidze, a b c d e f g h
Sitges 20 1 5 . 1 4 . . . a5! It's important to cover the aS-square.
1 5 .a4 i.b7 1 6.gfd l '1Wb6= White cannot put
especially strong pressure on the hanging
pawns, while Black's set-up looks harmonious.

After the text move Black went for 1 O . . . c6 in


R. Jacobs - Nyvlt, email 20 1 3 . I prefer:
Chapter 1 3 - S .id3 1 97

a b c d e f g h

1 3 ... d4! 1 4.l2Ja4 lDeS 1 5 .ib5 id7 1 6.ixd7


a b c d e f g h l2J fxd7 1 7.cxb6 axb6 1 8.b3 Wff6 with
10 ... Lf4!?N promising counterplay.
Since . . . c7 -c5 is coming, Black can feel free
to exchange the important bishop and destroy 12 ... bxc5 13Jel ia6 14Jxe8t Wfxe8 15.Wif3
the opponent's pawn chain. Black's pawn structure is much better, so
White hopes to take advantage of his lead in
l l .exf4 c5 12.dxc5 development.
Releasing the tension is a reasonable decision
- White has the bishop pair, so it makes 15 ...Wid7 16Jel ixd3 17.c!ihd5!
sense to open up the position. The following White should avoid 1 7.Wfxd3 ?! l2J c6+.
alternatives are hardly more promising:

1 2.ie3 c4 1 3 .ic2 lD c6 1 4.f5 l2J b4 l S .ib l

a b c d e f g h
a b c d e f g h
This tactical resource aims to favourably
1 5 . . . l2J e4 1 6.a3 l2J d3! 1 7.ixd3 cxd3 1 8.'1Mfxd3 change the character of the play. However,
ixf5= Black has no problems. Black can get two minor pieces for a rook after:

1 2.f5 l2J c6 1 3.dxc5 can be met by: 17 ... c6! 18.e7t! xe7 19.Wfxa8t c8
20.ic3 ib5 2 1 .hf6 ic6 22.Wib8 gxf6
In this unbalanced position Black should be
fine.
1 98 4.e3

B23) 9.a3 12.e4 dxe4 13 ..ba6


The natural 1 3.fxe4?! is now strongly met by
1 3 . . . ie5! and White loses an important pawn.
8
7 13 ... c!iha6
6
5 8
4 7
3 6
2 5
1 4

a b c d e f g h 3
9 ... c6 2
The aggressive 9 . . . tLlg4!? certainly contains 1
a drop of poison, but I have chosen a calmer
a b c d e f g h
approach.
14,c!ihe4
IO.f;Yc2 After 1 4.fxe4 c5 1 5 .ig5 ixh2t 1 6.tJih l
This was tried once by former World cxd4 1 7.ixf6 gxf6 1 8.tJixh2 dxc3 1 9.lLlxc3
Champion Garry Kasparov. lLlcS+ White doesn't get full compensation for
the pawn.
IO ... b6!?
1 0 . . . ixh2t was played in a memorable 14 .. Jc8 1 5 .i.g5 i.e7 16.hf6 hf6 17J:adl
encounter between Garry and a computer i.e7=
and ended in a quick but spectacular draw:
l l .tJixh2 tLl g4t 1 2.<;i;>g3 '1Wg5 1 3.f4 '1Wh5 B24) 9.f3
1 4.id2 '1Wh2t 1 5 .tJif3 '1Wh4 1 6.ixh7t <;i;>hs
1 7.lLlg3 lLl h2t 1 8.tJif1 lLlg4t 1 9.<;i;>f3 lLl h2t
8
Y2-Y2 Kasparov - Comp Deep Junior, New
York (5) 2003 . Even though the result wasn't 7
bad for Black, improvements for White have 6
been pointed out - and in any case, I do not
see the need for such drastic measures. 5
4
l l .f3 i.a6N
l l . . . c5!? also led to interesting play in 3
Jambrich - Kupec, Slovakia 2003 . The critical 2
continuation looks to be 1 2.dxc5 !?N bxc5
1
1 3 .lLlxd5 lLlxd5 1 4.ixh7t <;i;>hs 1 5 .ie4 :i:l:xe4
1 6.'1Wxe4 ie6 with double-edged play. a b c d e f g h
Chapter 1 3 - 5 .id3 1 99

9 ... c5! lO.f;Yei 1 2 . . . ixg4! 1 3 .fxg4 lDxg4 1 4.E1g2N The best


A more solid way of handling the position chance. ( 1 4.E1f3? Wfh4 1 5 .h3 lD h2 1 6.E1f1
seems toothless: 1 0.l2Jb5 if8 1 l .dxc5 ixc5 Wfxh3 led White to a hopeless position in
1 2.lDbd4 l2J c6 1 3 .id2 Wfb6 Black was by no Barsov - Hoerstmann, Zwolle 1 997) 14 . . . Wlh4
means worse in Gelfand - Carlsen, Moscow 1 5 .if4 ixf4 1 6.ltJxf4 ltJ e3
2007.

The prophylactic 1 0.a3 l2J c6 1 1 .l::1 f2 was tried


by one of the main experts in this system, but
it doesn't look very attractive. 1 l . . .a6 1 2.ic2
b5

a b c d e f g h

1 7.Wff3 l2Jxd4 1 8.Wfg3 lDxg2 1 9.'xg2 Wfxg3t


20.hxg3 E1ad8+ White will be able to regain
the d5-pawn, but Black's kingside pawns
secure him a clear advantage.
a b c d e f g h
8
1 3 .g4 b4 1 4.axb4 lDxb4 1 5 .ia4 id7 1 6.g5
ixa4 1 7.lDxa4 l2J d7+ The lack of harmony 7
and the exposed king caused White serious 6
problems in M. Gurevich - Hjartarson,
Akureyru 1 988. 5
4
1 0.l::1 f2 l2J c6 1 l .g4?! has been tried by some
grandmasters, but White can hardly afford 3
such drastic measures at this early stage of the 2
game. 1 l . . .cxd4 1 2.exd4
1
a b c d e f g h
10 ... c6 l l .f;Yh4
The queen transfer is aimed at creating
serious threats on the kingside, but most of
White's pieces cannot support the attack.

l l ... h6 12.id2
White gets nothing special with the
aggressive advance: 1 2.g4 a6 1 3 .Wff2 b5
a b c d e f g h 1 4.l2Jg3 This position was reached in
200 4.e3

Giorgadze - Hjartarson, Yerevan (ol) 1 996, This move exposes some light squares in
when Black should have continued: White's camp, but it's impossible to leave the
knight on b4 for a long time.

8
7
6
5
4
a b c d e f g h

1 4 . . . Wlb6!N 1 5 .l2Jce2 aS 1 6.id2 b4 With a 3


clear positional advantage. 2
1
8
a b c d e f g h
7
1 5 . .tc2
6 It was safer to cover the a6-fl diagonal
5 with 1 5 .id3N; still, after 1 5 . . . ib7 1 6.l':iad l
l':ic8 1 7.l':ife l ie7 1 8.Wfh3 if8 Black has no
4 problems.
3
15 ... .ta6 16J:Uel c8
2 All Black's pieces are perfectly mobilized
1 now.
a b c d e f g h
17.adl?
12 ... 1l'l b4 Much better was l ?.'it>h l N lD a5 1 8.dxc5
This move forces White to lock in the ixc5 1 9.l2Jd4, maintaining the balance.
a l -rook for a while.
8
12 . . . c4!? 1 3 .ib l a6 1 4 .'it>h l b5 1 5 .a3 ib7
also deserves attention, with a complex battle 7
in which Black's chances are not worse. 6
13.-tb l 5
1 3 .ib5? id7 1 4.ixd7 Wfxd7 would decrease 4
White's attacking potential and yield Black a
clear advantage. 3
2
13 ... b6 14.a3
1
a b c d e f g h
Chapter 1 3 - 5.id3 20 1

17...he2! 18.xe2 Conclusion


1 8.!he2 cxd4 1 9.exd4 c!Llxd4+ simply wins
a pawn. After 5.id3 d5 6.c!Llge2 White's knight is less
active than it would be on f3, but the idea
18 ... cxd4 19.exd4 gxe2! 20,gxe2 xd4 is not without merits, as it avoids doubled
2I .f;Yxd4 ic5+ c-pawns and may enable the knight to find
a meaningful role on the kingside via f4
or g3. Nevertheless, after 6 . . . dxc4 7.ixc4
c5 the placement of the knight on e2 limits
White's attacking possibilities and offers Black
comfortable play against the soon-to-be
isolated d-pawn.

A more challenging scheme is 6.cxd5 exd5


7.c!Llge2, switching to a Carlsbad structure,
where White's play is mainly based on
advancing the central pawns by means of
f2-f3 and e3-e4. Black will generally want
to be ready to meet f2-f3 with . . . c5, but the
a b c d e f g h bishop's presence on d6 also yields some
White had to fight for a draw until the end attacking chances on the kingside. In general,
of the game in Gelfand - Laurier, Cap d'Agde this variation leads to a tough positional fight,
1 996. but objectively Black should not have much
difficulty equalizing.
4.e3
6 .a3
Variation Index
l.d4 ttlf6 2.c4 e6 3.ttlc3 i.b4 4.e3 0-0 S.i.d3 dS 6.a3
6...hc3t 7.bxc3 dxc4 s.hc4 c5
A) 9.i.b2 203
B) 9.i.d3 204
C) 9.ttle2 'i'c7 206
C 1) 10.i.a2 206
C2) 10.i.d3 2 10
D) 9.ttl f3 'i'c7 2 12
D 1) 10.'i'c2 2 14
D2) 10.'i'd3 2 14
D3) 10.i.a2 2 15
D4) 10.i.e2 2 17

A) after 1 4 . g4 C2) note to 1 3 . e4 02) after 1 2 .e5

8 8
1 1
6 6
5 5
4 4 4
3 3 3
2 2 2
I
a b c d e f g h a b c d e f g h a b c d e f g h

1 4 . . . tLl xg4!N 1 5 . . . tLla5!N 1 2 . . . i.e4!N


Chapter 1 4 - 6.a3 203

I .d4 f6 2.c4 e6 3.c3 i.b4 4.e3 0-0 A) 9.i.b2 ffc7


s ..td3 dS 6.a3
A double-edged continuation. Compared
to the Samisch System, there is less strategic
danger in the doubled pawns, since the presence
of Black's pawn on d5 will enable White to
repair his structure. On the other hand, White
spends a valuable tempo and enables Black to
build a useful lead in development.

6 ...hc3t 7.bxc3

a b c d e f g h
IO.ffe2
1 O.id3 transposes to variation B below.

1 O.ie2 was once employed by the great


Botvinnik, but it has little independent value
- after 1 0 . . . ll:l c6N White has nothing better
than 1 1 . lLlf3 :i:l:d8 1 2.0-0, when we have
transposed to the 1 2.ib2 line mentioned in
the notes to variation 04.
a b c d e f g h
7 ... dxc4 IO ... c6 l l .f3 e5 12.0-0?!
Forcing White to move his bishop for a 1 2.h3 is more accurate; still, in Borsi -
second time. Tompa, Hungary 2002, Black could have
targeted the uncastled king with: 1 2 . . . exd4!N
7 . . . c5 8.cxd5 exd5 9.lLle2 leads to a well-known 1 3 .cxd4 'Wa5t 1 4.c;i;>fl
variation which may arise via a few different
move orders. Even though Black's recent
results have not been too bad, I still don't really
trust his position. Indeed, White's bishops
may support the creation of a powerful and
mobile pawn centre. Botvinnik's masterpiece
(Botvinnik - Capablanca, Holland 1 938) is
a perfect example highlighting the strategic
danger of Black's set-up.

s ..txc4 c5
a b c d e f g h
Black prepares . . . 'We?, developing the queen
while targeting the bishop. We will analyse 1 4 . . . ie6! 1 5 .ixe6 fxe6 1 6.'it>g1 :i:l:ad8 Black
A) 9.i.b2, B) 9.i.d3, C) 9.e2 and D) 9.f3. has excellent counterplay.
204 4. e3

The text move avoids the queen check I4 ... xg4!N 1 5.hxg4 ixg4 16 .td5 gadS

that occurred in the above line, but allows 17 ..te4 gd6


something even more unpleasant: With an enormous attack.

B) 9 ..td3 flc7

12 ....tg4! 13.h3
a b c d e f g h
1 3 .d5? e4! 1 4.dxc6 Wxc6 has given Black an
easily winning position in a couple of games. IO . .tb2
The threat of . . . cxd4 followed by . . . Wfc3t
13 ... .th5! meant that White's choices were limited. The
Much weaker is 1 3 . . . i.xf3 ?! 1 4.Wxf3 cxd4 only other logical option is 1 0.llJe2, which
1 5 .cxd4 exd4 1 6.exd4t, when White's bishops transposes to variation C2.
were powerful in Lund - De Verdier, Bugibba
20 1 1 . IO ... llJc6 l l .llJf3
1 1 .llJe2 e5 1 2.0-0 converts to variation C2
14.g4 again.
We have been following the game Cruz 1 1 .e4?! is premature. 1 1 . . .Wa5! 1 2 .Wfd2
Lledo - Cerrato Torrijos, Padrun 20 1 1 , where Now in Milov - Guido, Genova 2003, Black
Black missed a golden opportunity: should have played:

a b c d e f g h

1 2 . . . b6!N 1 3.llJe2 i.a6 1 4.i.xa6 Wxa6


a b c d e f g h
Chapter 1 4 - 6.a3 205

1 S .f3 llJ aS+ Exposing all the drawbacks of This type of position is typical of cases when
White's set-up. Black manages to free himself with the . . . e6-eS
advance.
8
7
6
5
4
3
2
1
a b c d e f g h
a b c d e f g h
l l ... e5!
A well-timed advance: Black shouldn't let 14.Wfc2
White complete his development and establish 1 4.Wfe2 gd8 1 S .c4 WigS 1 6.f3 i.f5! I like
his central superiority. this provocative move. 1 7.e4 ( 1 7.i.xf5 Wfxf5
1 8.0-0 Wfe6 1 9.gfd l gxd l t 20.gxd l llJ d7
12.dxe5 followed by . . . llJ b6 would also guarantee
The attempt to keep the tension in the centre Black comfortable equality) 1 7 . . . ie6 White's
by means of 1 2.Wfc2 turns out to be risky: position becomes less safe with the pawn on e4,
1 2 . . . cxd4 1 3 .cxd4 e4! 1 4.i.xe4 llJxe4 1 S .Wfxe4 as Black's knight gets some potential outposts.
WaSt 1 6.llJd2 i.f5 1 7.Wff3

a b c d e f g h
a b c d e f g h
1 8.gd l ( 1 8.0-0? llJhS! is nasty) 1 8 . . . gd7 1 9.g3
1 7 . . . WfbS! Forcing White's king to remain in gad8 20.0-0 h6. Despite the bishop pair, it is
the centre. 1 8.ic3 gfe8 1 9.dS llJ e7 20.a4 Wla6 White who suffers from a lack of activity.
2 l .d6 Wfxd6 Black was somewhat better in
T. E. Carlsen - Fenwick, corr. 2007. 14 ... c4!
A brilliant pawn sacrifice to delay White's
12 ... xe5 13.xe5 Wfxe5 development.
206 4.e3

1 5.hc4 i.f5 16.f;Ye2 gac8 17.0-0?! 9 ...Vc7


Better was 1 7 .b3 ltJ e4 1 8.0-0 ltJ xc3 There are two main options to consider:
1 9.'1We 1 Wfb5 20.xc3 Wfxb3 2 l .d4 with Cl) IO.i.a2 and C2) IO.i.d3. Each has good
equality. and bad points, as we will soon see.

1 0.Wfd3 ?! makes a poor impression, and Black


8
can choose between different ways of exploiting
7 the awkward placement of White's pieces. For
6 instance, 1 0 . . . e5!? 1 1 .0-0 e4 1 2.Wfc2 cxd4
1 3.cxd4 e6 1 4.b3 c8 1 5 .Wfxc7 xc7
5 1 6.xe6 fxe6, with excellent prospects in the
4 arising endgame.
3 Cl) IO.i.a2
2
Compared to the other retreat, this option
1
discourages the . . . e5 advance, as White's bishop
a b c d e f g h will then be perfectly placed. However, it also
17 ... g4! 18.g3 Vc5 19.i.d3 .bd3 20.Vxd3 has a drawback, which becomes apparent after
Wfh5 2 1 .h4 e5 Black's reply.
Black had a powerful initiative in Milov -
Lautier, Biel 1 997. 8
7
6
5
4
3
2
1
a b c d e f g h
IO ... b6! 1 1 .0-0 i.a6
It turns out that now Black has another
attractive way to activate the bishop. The
a b c d e f g h pressure along the a6-fl diagonal exposes the
White covers the c3-spot and keeps the somewhat passive placement of the knight on
f-pawn mobile. This important position might e2.
also arise after 6.l2Jge2 dxc4 7.xc4 c5 8 .a3
xc3t 9.bxc3. l2,gel
Sidestepping the pin is the most natural
continuation.
Chapter 1 4 - 6.a3 207

Harmless is: 1 2 .ib2 l2J c6 1 3 Jk 1 fd8 1 4.c4 been following the game Aleksandrov - ltkis,
cxd4 1 5 .exd4 This position was reached in Kishinev 1 998. Now Black should have
Tregubov - Efimenko, Muelheim 20 1 6, and played:
now the most logical continuation would have
been:

a b c d e f g h

1 6 . . . ic4!N 1 7.ixc4 lDxc4 1 8.e4 f6! 1 9.dxe6


a b c d e f g h
Wfe7 20.Wfh3 e8+ Black regains the pawn and
1 5 . . . b5!N 1 6.d5 ( 1 6.c5 b4 1 7.axb4 lDxb4 is secures a positional advantage due to his better
statically bad for White) 1 6 . . . bxc4 1 7.ixc4 pawn structure and superb knight.
ixc4 1 8.xc4 exd5 1 9.c2 Wfd6 White's
compensation for the pawn is uncertain.
8
7
6
5
4
3
2
1
a b c d e f g h
14 ... a5
a b c d e f g h
Black is aiming to occupy the exposed light
12 ... c6 13.g3 squares in White's camp.
White's attacking abilities are connected
solely with the e3-e4 advance. Therefore Black 1 5.e4
should apply strong pressure to the d4-pawn. The most consistent.

13 .. J:Ud8 14.ib2 1 5 .a4 makes little sense now. 1 5 . . . ic4 1 6.ixc4


1 4.l2Jh5?! is dubious; exchanging the knights lDxc4 1 7.Wfe2 Ki. Georgiev - Hjartarson,
only makes Black's position more comfortable. Linares 1 988. In my opinion, the most
14 . . . l2Jxh5 1 5 .Wfxh5 l2J a5 1 6.d5 We have practical decision would be:
208 4.e3

a b c d e f g h a b c d e f g h

1 7 . . . ll:lxb2!?N 1 8.'1Wxb2 E1ac8 White is under 1 8 . . . Wfd7N The pressure on the d4-pawn is too
pressure along the c-file. strong, so the next few moves are forced: 1 9.d5
exd5 20.e5 ll:l e4 2 1 .lLlxe4 dxe4 22.Wfxd7 E1xd7
1 5 .E1c l ic4 1 6.ib 1 is a natural regrouping. 23.ixe4 l::1 c 5 24.f4 ib7 The activity of Black's
(Exchanging the bishops seems a clear pieces fully compensates for White's bishop
concession: 1 6.ixc4 lLl xc4 1 7. Wb3 lLl a5 pair.
1 8.Wfd 1 l::1 ac8 1 9.Wfe2 h6 20.h3 Wc6+ Volke
Petursson, Radebeul 20 1 6.) Now in Horowitz
- Rossolimo, Havana 1 952, Black could have
exploited the awkward placement of White's
pieces by means of:

a b c d e f g h
1 5 ...i.c4 1 6.hc4
a b c d e f g h
1 6.e5, as played in Bhat- Aroshidze, Balaguer
2008, is connected with serious strategic risk. I
1 6 . . . ll:l b3!N 1 7.E1c2 e5 1 8.f3 E1d7 1 9.E1f1 E1ad8 suggest the following regrouping of the pieces:
20.Wfc2 g6+ White is under pressure in the 1 6 . . . ll:ld5N 1 7.ib 1 ll:l b3 1 8.l::1 a2 ll:le7! This
centre and he has problems organizing any accurate move severely limits White's attacking
active play. possibilities, so Black should be better.
1 5 .ib 1 l::1 ac8 1 6.e4 ll:l c4 1 7.ic l cxd4 1 8.cxd4 16 ... xc4 17.i.cl
occurred in Marin - Dokhoian, Budapest Since the pressure on the d4-pawn has been
1 988. Now I suggest a simple improvement released, White feels free to move the bishop
over Black's play: from such a passive spot.
Chapter 1 4 - 6.a3 209

17 ... e8! Black has good prospects on the queenside, so


This prophylactic approach is typical of I prefer Black's chances.
several lines in the Nimzo-Indian: avoiding
the unpleasant pin on the f6-knight is an 18 ... cxd4 19.cxd4 gd5 20.Wlg4?!
important measure. 20.Wff3N c8 2 I .ie3 was correct. In this
case Black has some positional advantages, but
White maintains the balance due to the passive
8
placement of the knight on e8.
7
6 20 ...Yid7 2 1 J:e4
It looks like most of White's pieces are ready
5 to join the attack, but . . .
4
3
2
1
a b c d e f g h
18.e5
Just as before earlier, this advance seems risky
from a positional point of view. Still, it makes
White's attacking potential more significant.

The quieter 1 8.i.g5 , provoking the 'weakening'


a b c d e f g h
1 8 . . .f6 advance, is less effective: 1 9.i.e3 ac8
20.Wfb3 cxd4 2 l .cxd4 This position arose in 2 1 ... cd6!
Rakhmanov - Agopov, Finland 20 1 6, and here This excellent tactical resource enables Black
it would have made sense to play: to force simplifications.

22.gf4
22.exd6 f5 is the idea. White can keep the
material balance with 23.lDxf5 exf5 24.Wfg5
(24.Wfh3 l2Jxd6+) but after 24 . . . h6 25 .Wfe7
xd6 26.Wfxd7 xd7 27.e5 xd4 28 .ie3
d3 29.xf5 ad8+ he still has to work to
secure a draw.

22 ... 5 23.xf5 exf5 24.Yixf5 Ylxf5


25.gxf5 gxd4 26 . .te3 gd5
a b c d e f g h
Black had the better endgame in Zajogin -
2 1 . . . WI d7!N Leaving the d6-spot vacant for the Brodsky, Minsk 1 997.
knight. 22.ad l l2J ed6 23 .i.c l l2Jb5 24.d5 e5
White's kingside activity is slowed down and
210 4.e3

C2) IO .td3 c6
. 1 4 . . . :i:l:ac8 1 5 .'1We2 :i:l:fd8 1 6.i.b2 cxd4 1 7.cxd4
idS 1 8.f5 i.c4! 1 9.:i:l:bc l i.xd3 20.Wxd3
ll:l c4 Black's control over the c4-square offers
excellent counterplay.

l l .. e5
.

This position has been tested many times at


a high level. Tournament practice shows that
the activity of Black's pieces prevents White
from benefiting from the mobile pawn centre,
while the bishops remain quite passive in most
cases.

a b c d e f g h
With White's bishop on the other diagonal
and the knight on the slightly passive
e2-square, it makes sense to challenge the
pawn centre by all possible measures!

1 1 .0-0
The presence of White's king in the centre
tells after 1 l .e4? cxd4 1 2.cxd4 ll:lxd4!+ with
the idea 1 3 .ltlxd4? '1Wc3t.
a b c d e f g h
The restricting 1 1 .f4 has the drawback of
exposing the light squares, giving Black good 12.i.b2
play after: 1 l . . .b6 1 2.0-0 ib7 1 3 .ltl g3 ltla5 White can also try to change the character
1 4.:i:l:b 1 !?N ( 1 4.We2?! ltl b3 and White felt of the play with an interesting pawn sacrifice:
compelled to give up the exchange in Huguet 1 2.e4!? cxd4 1 3 .cxd4 exd4 1 4.'1Wc2
- Casas, Buenos Aires 1 95 1 , since 1 5 .:i:l:b 1 This position was reached in Bagirov -
ll:lxc l 1 6.:i:l:bxc l :i:l:ac8 would put White under A. Ivanov, Frunze 1 979. White's attacking
strong pressure along the c-file) potential, which is based on f2-f4 and e4-e5,
shouldn't be underestimated, so I suggest:

a b c d e f g h
a b c d e f g h
Chapter 1 4 - 6.a3 21 1

1 4 ... tLl g4!?N 1 5 .f4


1 5 .e5 is playable but harmless: 1 5 . . . tLlgxe5
1 6.xh7t 'it>h8 1 7.e4 d3 1 8.ixd3 tLlxd3
1 9.'1Wxd3 e5 20.f4 f5=
15 . . . e6 1 6.h3 tLl e3 1 7.xe3 dxe3 1 8.c l
White manages to regain the pawn but Black
is doing well after:
1 8 .. J:fd8 1 9.xe3 b6=

1 2.c2 e6 1 3 .tLlg3
Releasing the tension with 1 3.dxe5?! is a
clear concession: 1 3 . . . tLlxe5 1 4.tlJf4 tLlxd3
1 5.xd3 ad8 1 6.tLlxe6 fxe6 1 7.c4 c6 a b c d e f g h
and Black was better in Aleksandrov - 13.e4
Polgar, Dresden (ol) 2008. This has only been played once but it seems
White also faces clear positional problems like the most natural move to consider.
after: 1 3.b 1 ac8 1 4.[5 xf5 1 5 .xf5
fd8 1 6.b2 d7+ Reshevsky - Piesina, Alternatives do not cause any problems. For
Vilnius 1 978. instance, 1 3.c l fd8 1 4.c2 ac8 1 5 .tLl g3 ,
as i n Ploehn - Scheipl, Bavaria 2000, can be
strongly met by:

a b c d e f g h

1 3 . . . fd8 1 4.b2 tLl a5N


Apparently this is a new move but it looks a b c d e f g h
absolutely normal. Play might continue: 1 5 . . . tLl a5!N 1 6.dxe5 c4! 1 7.e2 xe5+ and
1 5 .f4 White finds himself in a passive position.
A desperate attempt to develop counterplay.
1 5 . . . c4 1 6.e2 exd4 1 7.cxd4 ac8 1 8.f5 d5 13 ... b6 14.d5
1 9.e4 c3! 1 4.c l looks too awkward and after
White is in trouble. 1 4 . . . tLl a5 ! 1 5 .d5 g4 1 6.c4 xe2 1 7.xe2
tLl b3 1 8.c2 tLlxa 1 1 9.xa 1 d6+ White has
12 ...i.e6 insufficient compensation.
The bishop is perfectly placed here, where
it takes control of the weak light squares in 14 ...xb2
White's camp.
212 4.e3

a b c d e f g h
I S.gb lN
1 5 .dxe6 :i:l:ad8 1 6.exf7t :i:l:xf7 1 7.:i:l:b l Wfxa3
1 8.Wfc2 llJ b4! 1 9.cxb4 Wfxd3 20.Wfxd3 :i:l:xd3+
left Black a pawn up in Lees - Doudon, France
1 997.

15 ... ti'xa3 16,gal ti'b2 17.gb l ti'a2 1 8,gal It is important to recognise the difference between
this and another major variation in which the
moves 0-0 and . . . llJc6 have already been played.
The line in question usually arises after: 4.e3
0-0 5.i.d3 d5 6.llJf3 c5 7.0--0 llJc6 (I will
be recommending 7 . . . cxd4 in the next two
chapters) 8.a3 Axc3 9.bxc3 dxc4 I O.i.xc4 Wfc7

a b c d e f g h
1 8 ...bd5!?
It seems to me that Black has enough reasons
not to accept a draw too early! a b c d e f g h

This position has occurred in thousands of


19.gxa2 .hal
games; Black usually follows up with . . . e5 in
In this unbalanced position White's play
the near future. The version in our repertoire is
seems more difficult, at least from a practical
easier for Black to handle, for a few reasons. To
point of view.
begin with, the c4-bishop is in more immediate
danger, and the possibility of a queen check on
Chapter 1 4 - 6.a3 213

c3 reduces White's options. Moreover, Black 1 0.i.b3


enjoys some additional flexibility; depending I was surprised to discover how rarely this
on how White plays, Black may revert to the retreat has been tested in practice, as the
. . . l:i:J c6/ . . . e5 plan, but he may also develop the bishop on b3 is placed somewhat more
knight to d7 and bishop to b7 or a6. actively than after the more common
1 0.ia2.
White must decide how to safeguard the 1 0 . . . b5!N
hanging bishop; his main options are I see no reason to deviate from the plan seen
Dl) 10.c2. 02) IO.d3. D3) IO.i.a2 and in variation 03 below.
04) IO.i.e2.

1 0.i.d3? is positionally desirable, as the bishop


controls the e4-square and points towards
the kingside. Unfortunately for White, here
it is refuted by 1 0 . . . cxd4 1 l .cxd4 Wfc3t,
when he suffers huge material losses. It is
worth mentioning that the i.d3 plan is one
of White's most important options in the
analogous variation mentioned in the previous
note, so eliminating this plan already counts as
a b c d e f g h
a significant achievement for Black.
1 1 .0-0
1 0. ti:J d2 looks really artificial - the knight blocks I examined another principled reaction to
the dark-squared bishop and relinquishes Black's novelty: 1 1 .a4 cxd4 1 2.cxd4 b4 1 3.0-0
control over some key central squares. 1 O . . . e5 a5 14.ib2 ti:J bd7 1 5 .:gcl Wfd8 1 6.ti:Jd2
The most natural and tempting reaction. 1 1 .0-0 ( 1 6.ti:Je5 l:i:Jxe5 1 7.dxe5 ti:Jd5 offers White
:gd8 1 2.ib2 l:i:Jc6 1 3.:gc l cxd4 1 4.cxd4 exd4 no real attacking chances) 1 6 . . . ib7 1 7.f3
1 5 .exd4 id5 1 8.Ac2 ti:J b6 The strong, protected
b4-pawn yields Black sufficient counterplay
on the queenside.
1 l . . . ib7 1 2.ti:Je5
Mter 1 2.Wfe2 :gc8! 1 3 .i.d2 ti:J bd7 White's
active play has been delayed for a long time.
1 2 . . . ti:J bd7 1 3.1:i:Jxd7 Wfxd7 1 4.f3

a b c d e f g h

This position occurred in Kaloskambis


Sigalas, Athens 2004. Even though grabbing
the d4-pawn is playable, I prefer 1 5 . . . if5!N
1 6.ti:Jf3 Ae4, when Black is firmly in control.
a b c d e f g h
214 4.e3

1 4 . . . d5 ! 1 3 . . . Wfxc2N 1 4.xc2 llJ c6 1 5 .:i:l:cl :i:l:ac8+ With


Perfect timing, before White gets a chance to better chances for Black.
shut the bishop out of the game with e3-e4.
1 5 .c2 c4 1 6J:U1 :i:l:fd8 1 7.b2 a5 1 8.e4 l l ....ta6 12.ha6
:i:l:ac8+ 1 2.0-0?! cxd4 forces 1 3.exd4, when
White has seized some space in the centre, 1 3 . . . xd3 1 4.Wfxd3 :i:l:c8+ gives Black a
but most of his pieces are passive and in no clear positional advantage, with pressure on
position to support an attack on the kingside. the backward c-pawn and control over the
blockading c4- and d5-squares.
D l ) IO.f;Yc2 b6
1 2 ... xa6 13.f;Yd3
Now I like the following method of
regrouping the pieces:

a b c d e f g h
t t .Ad3
This involves a loss of time but there is
a b c d e f g h
nothing better:
13 ... b8N
1 1 .0-0? cxd4! 1 2.cxd4 a6 just drops material. 1 3 . . . Wfb7 1 4.0-0 :i:l:fd8 1 5 .Wfe2 :i:l:ac8 1 6.b2
llJ b8 also gave Black comfortable play in
1 1 .b2 cxd4 1 2.cxd4 a6 1 3 .d3 was tried Medina Garcia - Pomar Salamanca, Palma de
in the classic game Petrosian - Unzicker, Mallorca 1 965.
Gothenburg 1 95 5 . Simple and strong would
have been: 14.0-0 c6 1 5.e4 h6
As a result of White's slow play, Black has
managed to put strong pressure on White's
central pawns.

02) IO.f;Yd3

This time our bishop will not be able to go to


a6, bur it will find an excellent home on b7
after:

a b c d e f g h
Chapter 1 4 - 6.a3 215

the attack comes to an end, whereas White's


king is permanently exposed.

14 ... cxd4 1 5.0-0 xc4 16.exf6


1 6.WI'xa8? l2J c6 1 7.WI'b7 l::1 b 8 1 8.WI'c7 l2Jd5
1 9.WI'd7 dxc3-+ leaves Black with two pawns
for the exchange, plus dominant pieces and a
monstrous pawn on c3 .

8
7
6
5
4
3
2
1 1 ....tb7 1 2.e5 1
As a result of White's over-aggressive play,
a b c d e f g h
Black gained control over the light squares in
Li Shilong - Lupulescu, Golden Sands 20 1 2 . 16 ... d5 17.g3 g6 1 8J:M1 c6 19.4
However, the sharp character o f the position e5i
forces Black to play precisely. The correct Black succeeds in liquidating into a better
continuation is: endgame.

D3) 10 . .ia2

8
/=ij/'''!"--., .../'"'"'"""

7
6
5
4
3
2
a b c d e f g h
1
12 ... .ie4!N 13.e2 .bf3 14.xf3
a b c d e f g h
Even worse is 1 4.gxf3?! l2Jd5 1 5 .l::1 g 1 l::1 c 8
1 6 . .th6 g6 1 7 . .txd5 exd5 1 8.WI'e3 l2J c6+ and Just as in the earlier variation C 1 , White
hopes that the bishop will be well placed after
216 4 .e3

Black pushes with . . . e5. Although that plan is Removing the rook seems logical; now the
still playable, I find the following continuation b5-pawn really is hanging.
more logical:
1 3 .ib2 tLl bd7 1 4.gfc l lLl b6 1 5 .lLle5 lLl fd7
lO ... bS!? l l .O-O .th7 12.e2 gives Black good prospects.
White should not indulge in pseudo
aggressive measures like 1 2.tLlg5?! lLl bd7 1 3 .f3 I also considered: 1 3.ib l lLl c6 ( 1 3 . . . tLl e4 can
h6 1 4.lLlh3 idS 1 5 .ib l ic4 1 6J:'f2 e5+, be met by 1 4.ib2) 1 4.ib2 ( 1 4.Wfxb5 tLl a5
when Black was clearly better in Savchenko - gives Black a lot of compensation for the
Predke, Moscow 20 1 5 . pawn) 1 4 . . . tLl a5 1 5 .e4 tLl c4 1 6.ic l tLl d7

Also somewhat awkward is 1 2.ib l tLl bd7


1 3 .Wfe2 tLl e4 1 4.ib2 ic6 1 5 .id3 gab8 and
Black has managed to set up a perfect blockade
over the light squares.

The text move seems like the most natural


- White attacks the b5-pawn and starts to
fight for the extremely important e4-square.
Here I would like to deviate from a couple of
correspondence games by means of:
a b c d e f g h

1 7.id3 ( 1 7.a4 cxd4 1 8.cxd4 b4 1 9.id3 a5+)


8
17 . . . tLl db6 1 8.ge l h6 Here too, it will be hard
7 for White to build a meaningful attack as
6 Black is so active on the queenside.

5 13 ... bd7!
4 It turns out that, having such well-developed
pieces, Black does not need to spend time on
3 prophylactic moves!
2
1
a b c d e f g h
12 .. Jc8!?N
I like this flexible mode of development -
Black keeps the knight on b8 for a while, so
the b5-pawn remains indirectly protected.
At the same time, White has to deal with
the possibility of the c-file opening up at any
moment.

a b c d e f g h
Chapter 1 4 - 6.a3 217

14.ti'xb5 is justified against the timid ie2 set-up. Mter


Grabbing the pawn is practically necessary White's next move we will transpose to one
- White's pieces lack harmony, so there are no of the lines from the variation mentioned on
other useful ideas. page 2 1 2 in the note to Black's 9th move.

14 ... g4!
8
Black's threats are serious, so White's next
move is practically forced. 7
6
1 5 .d5 bd5 16 ..bd5 exd5 17,gxd5
White has won a pawn, but the undeveloped 5
a1 -rook and c l -bishop allow Black to extend 4
the initiative:
3
2
1
a b c d e f g h
1 1 .0-o gds
Black is perfectly prepared for the . . . e5
advance, so White will have to move the queen
sooner or later.

12.Vc2
1 2.ib2 e5 1 3.gcl is toothless. ( 1 3 .WI'c2 ig4
transposes to our main line) 1 3 . . . ig4 1 4.h3
a b c d e f g h ih5 1 5 .lLlxe5 ixe2 1 6.WI'xe2 tLlxe5 1 7.dxe5
17 ... df6 IS.gdl e4 19.ti'c4 gds 20.gfl Wxe5 1 8.c4 We6= D. Guseinov - Lagashin,
ti'e7 2 1 .h3 gf6 Moscow 2009.
Black has full, long-lasting compensation for
the pawn. 1 2.c4 b6 1 3 .ib2 This position was tested a
few times at a high level, including Pasman -
D4) 10.i.e2 Tal, Riga 1 9 54. In my opinion, it makes sense
to keep the tension in the centre by means of:
This has been played by many strong
grandmasters. In my opinion, however, it is
not in the spirit of the chosen system - the
bishop on e2 is placed rather passively, so it is
easier for Black to take the light squares under
control now.

10 ... c6
There is nothing wrong with 1 O . . . b6 1 1 .0-0
ib7, b r.opl that the more direct approach
a b c d e f g h
218 4.e3

1 3 . . . lLl a5N 1 4 .WI'c2 i.b7 1 5 .lLle5 ( 1 5 .:gfd 1 on b2 doesn't really bother Black, while the
cxd4 1 6.exd4 :gac8 1 7.lLld2 Wff4+) 1 5 . . . lLl d7 c4-pawn is a permanent cause of concern.
1 6.lLlxd7 :gxd7= The game Mozharov - Zhidkov, Moscow
2006, saw White trying to complicate
1 2 ... e5 13 ..tb2 .tg4 matters by means of:
Black's forces are perfectly mobilized, so
White cannot keep the tension any longer.

a b c d e f g h

25.f4!? Wxe4 26.i.xf6


But this wouldn't pose Black any problems
after:
a b c d e f g h 26 . . . :gxd l tN 27.:gxd l lLlg6 28.i.b2 Wxf4
14.dxe5 29.Wfxf4 lLlxf4 30.:gd7 lLlh5
White may also start by nudging the bishop: White's compensation for the pawn offers
1 4.h3 ih5 1 5 .dxe5 no more than a drawish rook endgame.
White's inaccurate play invited serious trouble
in the following high-level game: 1 5 .lLlxe5 14 ... xe5
lLlxe5 1 6.dxe5 i.g6 1 7.Wfb3?! lLle4! 1 8.c4
lLld2+ Laznicka - Karjakin, Dubai 20 1 4.
1 5 . . . lLlxe5 1 6.lLlxe5 ixe2 1 7.Wxe2 Wl'xe5

a b c d e f g h
a b c d e f g h
15.c4 xf3t 16.gxf3 .th3 17Jfdl
1 8 .c4 Wl'e6 1 9.:gfd l lLl e4 20.:gac l b6 2 l .f3 White is relying on the bishop pair, but
lLl g3 22.Wff2 lLl f5 23.e4 lLle7 24.WI'g3 f6 the exposed kingside structure offers Black
The position is about equal: the active bishop sufficient counterplay:
Chapter 1 4 - 6.a3 219

17...c6! IS.c;!?hi!N Conclusion


This is more precise than 1 8 .Wfc3, when
1 8 . . . tD e8! 1 9.c;i;>h l ie6 20J!g l f6 2 1 .g3 6.a3 is another way for White to aim for
d7 22.agl adS+ was better for Black in a strong pawn centre plus the bishop pair.
Khalifman - Kramnik, Linares 2000. Indeed, in comparison to the Samisch, the
presence of Black's pawn on d5 enables White
to repair the damage to his pawn structure.
8
However, White's slow development enables
7 Black to develop significant counterplay along
6 the c-file and solve the problem of the light
squared bishop. Throughout the chapter, we
5 have seen how Black can use these attributes to
4 prevent White from utilizing his strong pawn
centre to build an attack.
3
2
1

a b c d e f g h
1 8 g4 1 9Jixd8t gxd8 20.e4

The position is tricky for both sides, but the


correct outcome looks to be a draw after:

20 xflt 2 I .c;!?gi h6! 22.c;!?xfl gd2


.

23.c3 g5 24.c;!?ei d4 2S.c;!?tl gd2=


4.e3

Variation Index
l.d4 f6 2.c4 e6 3.c3 i.b4 4.e3 0-0 S.i.d3 dS 6.f3
6...c5
A) 7.cxd5 22 1
B) 7.0-0 cxd4 8.exd4 dxc4 9.hc4 b6 223
Bl) 10.a3 224
B2) lO.i.d3 225
B3) lO.eS 226
B4) 10.'i'b3 227
BS) 10.'i'e2 228
B6) lOJel 230

83) after 1 3 .ixe6!? 8 5 ) after 1 4 .ttl xd5 86) after 1 3 .d2

8
7
I''"'"':.::=.71:'7//
6
//N//,

5
t'"/////,wJ"/"///mh
4
3
2

a b c d e f g h a b c d e f g h a b c d e f g h

1 3 . . .c7!N 1 4 . . . ic5 !N 1 3 . . . ttl e7!N


Chapter 1 5 - 6.lt:lf3 22 1

l .d4 f6 2.c4 e6 3.c3 i.b4 4.e3 0-0


s ..td3 dS 6.a
This move leads into some of the absolute
main lines of the Nimzo-Indian.

6 ... c5
Virtually every sensible-looking move has
been tried by strong players, but the text move
is the most classical response, challenging
White in the centre.
a b c d e f g h
We will analyse A) 7.cx:d5 and B) 7.0-0.
1 0 . . . llJ e4! 1 l .ixe4N This seems like the only
7.a3 is the only other significant move, but way to challenge Black's last move. ( l l .ib2
in that case 7 . . . ixc3t 8.bxc3 dxc4 9.ixc4 ifS 1 2.0-0 Lindstrom - Tasic, email 2009,
transposes to variation D of the previous 1 2 . . . llJc6N 1 3 .llJ d2 llJxd2 1 4.'1Wxd2 ixc2
chapter. 1 5 .Wfxc2 bS+) l l . . .dxe4 1 2.llJd2 WigS! 1 3 .'it> fl
WfdS+ White has a passive position with a
A) 7 .cx:d5 exd5 misplaced king.

8.0-0 c4 9.ic2 ig4 l O.llJe2, as played in


Farid - Nguyen Anh Dung, Jakarta 20 1 5 ,
leaves most ofWhite's pieces passive. Black has
no reason to reject:

a b c d e f g h
8.dxc5
This approach is similar to another a b c d e f g h
fashionable line, 7.0-0 llJ c6 8.cxd5 exdS
9.dxc5 , where sometimes Black finds it hard to 1 0 . . . ixf3N l l .gxf3 :i:l:e8 1 2.b3 cxb3 1 3.axb3
prove that active piece play fully compensates Wid? 1 4.ia3 llJ c6 With excellent play.
for the isolated d-pawn. In comparison to that
line, Black now benefits from not having the 8 ... bd7!
knight on c6! This resource enables Black to save a tempo
rather than moving the dark-squared bishop
8.a3?! cannot be recommended for White. again. Moreover, the knight will be placed
8 . . . ixc3t 9.bxc3 c4 I O.ic2 Now simple and quite actively on c5, taking control of the
strong is: important e4-square.
222 4.e3

9 ..td2 12.b5 .tbs 13.0-0 a6 14. bd4 xd4


9.0-0 ltJ xc5 1 0 . .ie2 seems too passive after 1 5.xd4 e4 16 ..tb4
1 0 . . . ltJ ce4 1 l .liJb5 .id7, when Black was at 1 6.c l e8 1 7 . .ie 1 Wl'd6 1 8.f4 .id7 doesn't
least equal in Alatortsev - Levenfish, Moscow promise White any advantage.
1 940.
16 .. Je8 17.cl
9 ... xc5 10 .te2
Also harmless is: 1 7.Wfb3 a5 1 8 ..ic3 a4
White is keeping his queenside pawn chain 1 9.WI'b5 .ie5 20.ac l ltJxc3 2 l .bxc3 a5
flexible, but the pieces are placed rather 22.Wfb2 :8:c5=
passively and don't put strong pressure on the
isolated pawn.
8
10 ... e6 7
It's important to secure a safe retreat for the 6
b4-bishop.
5
l l .a3 4
After 1 1 .0-0 a6 1 2.WI'b3 .ie7 1 3.fd 1
b 5 1 4 . .ie 1 .ib7 1 5 .a3 c8 Black's excellent
3
piece play provided full compensation for the 2
isolated pawn in Krush - Woj taszek, Doha
1
20 1 5 .
a b c d e f g h
We have been following the recent top-level 17 ....te5 1 8.3!?
game Harikrishna - Topalov, Stavanger 20 1 6. The most challenging.
I think Black should have opted for the most
active: 1 8.Wfb3 a5 1 9 . .ie 1 Wfd6 20.liJf3 .if6 looks
comfortable for Black.
8
18 ... d6 19Jc5! f5!
7 It was possible to trap the rook by means of
6 1 9 . . . b6 20.xd5 .ib7, but after 2 l .xe5 xe5
22.e4 I prefer White.
5
4 20.xf5 J.x5
3 White can grab the d5-pawn, but Black gets
sufficient counterplay in all cases.
2
1 2 1 .xd5
After 2 l .Wfxd5 b6 22.Wfxd8 axd8 23.c6
a b c d e f g h
a5 24 . .ic3 .ixc3 25.xc3 d2 Black regains
1 1 .td6!?N
the pawn to reach an equal endgame.
Black is ready to play . . . a6 with a harmonious
position, so the following line appears critical: 2 1 ...ti'f6 22.ti'd2
Chapter 1 5 - 6 . ltl f3 223

22.f4 ixb2 23.id3 ixd3 24.Wxd3 We6 is posmon is also frequently reached via the
balanced. Panov Attack against the Caro-Kann. Black's
last move obviously prepares to post the
bishop on b7, while the b8-knight remains
8
flexible: it often goes to d7, but I also like the
7 idea of bringing it to c6 in some variations.
6 At the moment Black is playing against an
isolated queen's pawn, but it is not uncommon
5 to convert to a hanging pawns structure after
4 an exchange on c3 .
3 In this chapter we will study the following
2 options: Bl) 10.a3, B2) IO .td3, B3) IO.e5,
.

B4) IO.b3, BS) IO.e2 and B6) IOJie l .


1
a b c d e f g h 1 0.ig5 i s the main move, which i s covered
separately in the next chapter.
22 Jiad8 23.b3 h6 24.4
.

24.E1d l E1xd5 25 .Wxd5 Wh4 26.h3 Wg3 1 0.Wfd3 ib7 l l .l::1 d l tLl bd7 1 2.ig5 E1c8
27.';t>fl Wfh2 is equal claims the computer, but 1 3 .ib3 is a harmless continuation which
practically more dangerous for White. occurred in Peter - Schmitzer, Hessen 1 998.
A good plan for Black could have been:
24 ...Ac7 25 ..ic5 Ab6!
Black has full compensation; his pieces are
more active and White has numerous pawn
weaknesses.

B) 7.0-0

a b c d e f g h

1 3 . . . Wc7!N Stepping out of the pin. 1 4.tLlb5


( 1 4.l::1 ac l Wb8 is also preferable for Black)
1 4 . . . Wb8 1 5 .ih4 id6 1 6.tLlxd6 Wfxd6 White's
bishops do not provide full compensation for
the isolated pawn.

1 0.ie3 seems too passive. 1 0 . . . ib7 l l .l::1 c l


a b c d e f g h lLl c6 1 2.Wfe2 tLl e7 This typical move secures
7 ... cxd4 8.exd4 dxc4 9 ..hc4 b6 control over the d5-square. 1 3 .E1fd l This was
We have reached the main tabiya of what I. Sokolov - Woj taszek, Haguenau 20 1 3, and
is widely known as the Karpov System. This now simple and strong would have been:
224 4.e3

That's the point! The vulnerability of the


c3-pawn forces White to deviate from his
optimal set-up with id3 followed by c3-c4.

12.We2
White insisted on 1 2.id3 in L. Guliev -
Abasov, Baku 20 1 1 , but the blunt 1 2 . . . Wfxc3N
1 3.ig5 tLlbd7 1 4.l::1 c l Wla5+ would have left
White with no real compensation for the pawn.

a b c d e f g h 1 2.Wfd3 ib7 1 3.ia2 ( 1 3.ig5 l::1 c8 1 4 .lLld2


1 3 . . . tLl fd5N 1 4.ig5 ( 1 4.lLlxd5 lLlxd5 1 5 .a3 tLld5+) 1 3 . . . tLlbd7 1 4.h3 l::1 ac8 1 5 .c4 seems
ie?+) 1 4 . . . h6 1 5 .lLlxd5 ( 1 5 .ih4 lLl f4 1 6.'1We3 quite pointless. White's set-up looks artificial
l::1 c 8 1 7.ib3 lLl fg6+) 1 5 . . . ixd5 1 6.ixe7 ixe7 and Black can exploit it by means of 1 5 . . . e5!+ as
1 7.tLle5 E1c8 Black has the better position, in L. Guliev V. Gaprindashvili, Adana 2006.
-

with chances to press against the IQP.


12 ... .tb7 13.e5 c6 14 ..td3
Bl) 10.a3 White accepts a structural weakening in the
hope of progressing his attack.
This move looks principled, as White hopes
to profit from the bishop pair and the 14 ... c!the5
strengthening of his pawn centre. However, 1 4 . . . l::1 ad8?! gave White a chance to
losing a tempo at such an early stage leads to consolidate with 1 5 .f4! in lnkiov - Marciniak,
some inconveniences. France 20 1 1 .

1 5.dxe5

a b c d e f g h
1 5 ... d7!?N
1 5 . . . Wfc6 1 6.f4 tLl e4 is fine, but after
1 7 .ixe4 Wfxe4 1 8.Wfxe4 ixe4 1 9.ie3 Black
l l .bxc3 Wc7! has no more than a symbolic advantage, and
Chapter 1 5 - 6 . ltl f3 225

the players soon agreed a draw in Siebert -


Von Saleski, email 2000. The text move keeps
more winning chances alive.

16.i.f4 c5
Black has a clear positional advantage due to
the better pawn structure.

B2) 10 ..td3 i.b7

1 5.e4
Pinning the f6-knight seems to be White's
only dangerous idea.

1 5 .llJe5? drops a pawn now: 1 5 . . . llJxe5 1 6.dxe5


ixe5 1 7.%Yxe5 Vxd3 1 8.i.xf6 %Yg6!+
l l .a3
l l .:ge 1 converts to the later variation B6, 1 5 ...'ifc7!
and l l .ig5 is covered under 1 0.i.g5 ib7 Black should not fear a slight compromising
l l .i.d3 , in variation A of the next chapter. of his kingside structure, as White does not
have any real attacking potential.
The text move sees White aim for a modified
version of the plan from the previous variation. 16.g3
Here his bishop is safely on d3, so the plan of 1 6.llJxf6t llJxf6 1 7.ixf6 gxf6 1 8.ie4 ixe4
exchanging on c3 and putting the queen on c7 1 9.Vxe4 f5 20.%Ye2 :gfd8+ and Black has the
loses its bite. Instead it looks better to continue upper hand.
with:

l l ....td6!?N
I was quite surprised that this typical retreat
hasn't been tried yet. The bishop is more active
on d6 than on e7, and White's bishop would be
better on c4 than d3 in the ensuing position.

12.i.g5 ttlbd7 13.'ife2


1 3 .llJe4?! i.e?+ leads only to simplifications,
so the weakness of the d4-pawn becomes more
significant.

a b c d e f g h
226 4.e3

16 ...ti'c6!
Finally forcing favourable simplifications.

17 . .bf6 xf6 1 8.!Uel xe4 19 . .be4 ti'xe4


20.ti'xe4 .be4 2 1 .gxe4 .td6
White will have to work to hold the
endgame.

a b c d e f g h
8
1 5 . . . ltl a5! 1 6.i.xf7t Wfxf7 1 7.W/xf7t tJixf7
7 1 8.i.xf6 gxf6 Black's chances are better due to
6 his domination over the light squares.
5
1 3 .ltlxc6 i.xc6 is already at least equal for
Black, for instance: 1 4.i.e2 Wid? 1 5 .i.f4

1
a e
This looks active and aggressive, but White
is spending time moving an already developed
piece, and it is often Black who can look to
seize the initiative.
a b c d e f g h
IO ... .tb7 l l .ti'b3
l l .ig5 transposes to variation B of the next 1 5 . . . b5! 1 6.l':i fe l a6 1 7.Wfb2 ltl d5+ Black was
chapter. able to set up a solid blockade on the light
squares in Davidov - Dimitrov, corr. 20 1 0.
l l . .. .bc3 12.bxc3 c6
Challenging the active knight while creating
the annoying positional threat of . . . ltla5.

13 ..be6!?
This seems like the only challenging idea,
but Black has more than one decent reply.

1 3 .ltlxf7?! is a slightly inaccurate version of


the same sort of idea. 1 3 . . Jhf7 1 4.i.xe6 Wfe8
1 5 .ig5
Chapter 1 5 - 6.lLl f3 227

13 ...ti'c7!N 17 ... d5 1 8.g3


1 3 . . . lLlxe5 1 4.dxe5 fxe6 1 5 .WI'xe6t tJih8 1 8.ac l ? lLl f4-+ wins material for Black.
1 6.exf6 xf6 1 7.WI'e2 Wl'f8 1 8.3 e8 1 9.Wff1
fe6 gave Black decent compensation for the The text move looks ugly, but the f4-spot must
pawn in Malushko - Summers, corr. 20 1 3 . be guarded.
There i s nothing much wrong with following
that game, but I like the text move even more.
8
14.c!ihc6 7
Dubious is: 1 4.lLlxf7?! xf7 1 5 .if4 6
5
4
3
2
1
a b c d e f g h

1 8 ... fe8!?
a b c d e f g h
1 8 . . . lt:lxc3 1 9.ixc3 Wfxc3 20.WI'a4= leads to
simplifications.
1 5 . . . WI'e7! Black should eliminate the light
squared bishop rather than the dark-squared 19.gfel b5
one, so that his own bishop will be able Black's control over the light squares
to dominate the light squares. ( 1 5 . . . WI'xf4 and superior minor piece offer him full
1 6.i.xf7t tJih8 1 7. 3 lLla5 1 8.WI'c2 enables compensation for a pawn.
White to maintain the balance.) 1 6.i.g5 lLl a5
1 7.i.xf6 Wl'xf6 1 8.i.xf7t Wfxf7 1 9.Wfxf7t B4) IO.ti'b3
c;i;>xf7+ Even though White has a slight material
advantage, he is doomed to a passive defence.

14 ...ti'xc6 1 5 ..th3 .tcS!


The bishop has done its job on the long
diagonal, and now exchanges itself in order
to allow the rook to come to c8 with gain of
tempo.

16.hc8 gaxc8 17 ..tb2


1 7.i.d2 runs into 1 7 . . . lLle4 1 8.d5 Wc4
1 9.Wxc4 xc4 20.fd 1 d8, regaining the
pawn in a favourable situation. The arising
endgame is drawish, but White will still have a b c d e f g h
to be careful!
IO ....ixc3 l l .bxc3
228 4.e3

In comparison to variation B l , it looks as 1 4 . . . l2J a5N 1 5 .Wfd l :i:l:c8 1 6.:i:l:c l Wfd5 ! 1 7 . .txf6
though White has made an active developing gxf6 1 8.:i:l:e 1 f5 Here too, Black's play seems
move rather than a useless pawn move. However, somewhat easier - the pressure on White's
the queen proves to be misplaced on b3, so it queenside pawns is quite unpleasant, whereas
turns out that White actually loses time. Black's king still feels safe.

l l ... c6 12.dl 12 ... .tb7 1 3 ..tg5 gc8 14.i.d3


Admitting that White's l Oth move was
pointless.

1 2 . .te2 is too passive. 1 2 . . . .tb7 1 3 . .tg5


occurred in Aleksandrov - Grigoriants,
Warsaw 2005, and now Black could have
seized the initiative by means of:

a b c d e f g h
14 ... e7! I S . .tx6 gxf6 16.gcl g6 17.gel
:i:l:c7 18.d2 s;
Black had the better position in Sipila -
a b c d e f g h Alekseev, Jerusalem 20 1 5 .
1 3 . . . l2J a5N 1 4.WI'b2 :i:l:c8 1 5 .:i:l:ac l Wl'd5!
1 6.i.xf6 gxf6+ Once again, the doubled BS) IO.e2
f-pawns are of little consequence as the rest of
Black's pieces are so well placed. 8
Another game continued 1 2 . .td3 .tb7 1 3 . .tg5 7
h6 1 4 . .th4, Zilka - Kravtsiv, Yerevan 20 1 3 . 6
Now I prefer the following way of handling
the position: 5
4
3
2
1
a b c d e f g h

10 ... .tb7
This is the most natural, although 1 0 . . . l2J c6!?

a b c d e f g h
Chapter 1 5 - 6 . lLl f3 229

8
1 1 .d 1 lLl a5 1 2.d3 b7 1 3.g5 e7 seems a
reliable alternative.
7

6
l l J:MI
1 1 .g5 transposes to variation C of the next 5

chapter. 4

3
White's set-up became popular in roughly the
2
middle of the 20th century. The main idea
behind it is to prepare for the d4-d5 advance.
In particular, 1 1 . . . lLl bd7 can be met with a b c d e f g h

1 2.d5, so in most games Black has preferred 1 4 . . . c8!N 1 5 .a6 xa6 1 6.'1Wxa6 lLl ed5
to deviate from those complications with 17 .lLlxd5 ll:lxd5+
1 1 . . .xc3 1 2.bxc3 lLl bd7. However, I take
a different view. Since exchanging the dark White may try to force the exchange on c3 by
squared bishop is something of a concession, means of:
it makes sense to postpone the development 1 2.a3!?N xc3 1 3.bxc3
of the queen's knight for a while. Therefore I But in doing so, he loses an important
recommend: tempo. Black has a good position after:
1 3 . . . Wfc7 1 4.d3
1 4.lLle5 lLl bd7 1 5 .d3 lLlxe5 1 6.dxe5 lLl d7
is good for Black, for instance: 1 7 .f4 fd8
1 8.Wlg4 lLlxe5 1 9.Wfg3 f6 20.ixh6 'it>f8
2 1 .f4 Wfxc3+
1 4 . . . Wfxc3

a b c d e f g h
l l h6!
...

It is surprising that this natural prophylactic


move has only been tested in a few games.
a b c d e f g h
White's attacking abilities are severely limited
now. 1 5 .xh6
The over-aggressive 1 5 .ib2?! Wfc7 1 6.d5
12 .th3
.
lLlxd5 1 7.ac l Wfe7 1 8.c4 f5 1 9.lLle5 d8+
The more aggressive 1 2.lLle5 lLl c6 1 3 .e3 leaves White no real compensation for two
lLl e7 1 4.ac l , as played in Batchimeg - pawns.
Dzagnidze, Dilijan 20 1 3 , could have led to an 1 5 . . . gxh6 1 6.Wfe3
inferior position for White after: This more or less forces a draw by repetition:
230 4.e3

1 6 . . . i.xf3 1 7.'1Wxf3 tLld5 1 8.ac l Wa5 1 5.i.e3 tLlxd5 16.i.xc5


I 9.Wg3t cj;lhs 20.Wh4 cj{g7= Both 1 6.i.xd5 We7= and 1 6.xd5 We?
1 7.ad l ad8= are fine for Black.
8
16 ... bxc5
7
6
5
4
3
2
1
a b c d e f g h

12 ... c6
Black is well developed and ready to improve
a b c d e f g h
his position with . . . tlJa5 or . . . tLle7, so White
should not hesitate to get rid of the isolated 17.hd5
pawn. 1 7.xd5? turns out to be a blunder:
1 7 . . . ttJ d4!-+
13.d5 exd5 14.xd5
We have been following the game Agdestein 17 ...b6 1 8.h3 gadS=
- Kramnik, Stavanger 20 1 4. Now Black The activity of Black's pieces fully
should have played: compensates for the slight weakening of his
pawn structure.
14 ... i.c5!N
Instead, 1 4 . . . tLlxd5?! 1 5 .i.xd5 Wf6 1 6.i.e3;t B6) IO,gel
left Vladimir under some pressure.
8
7
6
5
4
3
2
1
a b c d e f g h
a b c d e f g h
Chapter 1 5 - 6 .lt:l f3 23 1

In most cases, this move leads by 12 ... c!iJc6 13.ti'd2


transposition to the main lines with 1 o.g5 . We have been following the game Salem -
In this section, we will deal with some lines Yu Yangyi, Incheon 20 1 3. With the dangerous
with independent value. xh6 motif in the air, it makes sense to bring
one more piece to the defence. Therefore I
IO .. .tb7 1 I ..td3
. suggest:
1 1 .g5 transposes to variation D of the next
chapter.
8
The text move looks natural - White removes 7
the bishop from the more exposed c4-square, 6
where it could be attacked by the opponent's
rook, and hopes to use its power for developing 5
some kingside initiative. The drawback is that 4
White reduces his control over the dS-square,
which means that there is no need to exchange 3
the b4-bishop. 2
1
l l . h6
..

Just as in variation B 5 above, this prophylactic a b c d e f g h


move allows Black to limit the opponent's 13 ... c!iJe7!N 14.c!lJe5
attacking abilities and secure control over the 1 4.xh6? doesn't work after 1 4 . . . xf3!
d5-square. 1 5 .gxf3 gxh6 1 6.Wxh6 ll:l g6, for instance:

12 ..tf4
1 2.a3 d6! 1 3.Wfe2, as played in Korobov
Lysyj , Legnica 20 1 3, can be well met by:

a b c d e f g h

1 7.tJih 1 xc3 1 8.bxc3 WfdS 1 9.e4 WhS


20.Wfxh5 lLlxh5 2 1 .xa8 1:ha8+
a b c d e f g h
14 ... 5 1 5.a3 .te7 16.Lf5 exf5
1 3 . . . ll:l c6!N 1 4.lLle4 lLl a5 1 5 .lLlxd6 Wfxd6+ and
White finds himself in a passive position.
232 4.e3

Conclusion

In this chapter we dealt with some of White's


less popular options after 6.tLlf3 c5. Some
players prefer to fight against an isolated pawn,
so 7.cxd5 exd5 8.dxc5 may appeal to them,
but 8 . . . tLl bd7! enables Black to place his pieces
actively; in particular, the knight coming to e4
is quite annoying for White, so the pressure on
the d5-pawn is not too great a concern.

7.0-0 is the normal move, and after 7 . . . cxd4


a b c d e f g h 8.exd4 dxc4 9 . .ixc4 b6 the shoe is on the other
17.Yid3 e4 18.6 xc3 19.bx:c3 gc8 foot, since it is now White who must play with
In this complex position, Black has an isolated pawn. Of the various possibilities
reasonable chances of taking over the initiative. we explored in this chapter, 1 0.%Ve2 .ib7
The unopposed light-squared bishop is an 1 l .d 1 is the most challenging, as it forces
especially useful asset. Black to watch out for the d4-d5 advance.
I find 1 1 . . .h6! to be the most convincing reply,
in conjunction with the novelty at move 1 4 to
improve on Kramnik's play. On a more general
note, the presence of an isolated d-pawn
puts White under some strategic pressure. If
Black manages to complete development and
establish control over the d5-square, White is
likely to find himself having to work hard for
a draw.
4.e3
a b c d e f g h

lO .igS - Main Line


Variation Index
l.d4 f6 2.c4 e6 3.c3 J\b4 4.e3 0-0 5.J\d3 d5 6.f3 c5
7.0-0 cx:d4 8.exd4 dxc4 9.hc4 b6 10.J\g5
10...J\b7
A) ll.J\d3 234
B) ll.e5 bd7! 235
Bl) 12.xf7!?N 236
B2) 12.xd7 238
C) ll.e2 bd7 239
Cl) 12.d5 240
C2) 12.e5 24 1
C3) 12Jacl 242
D) l l.el c6 244
Dl) 12.J\d3 245
D2) 12.a3 246
E) ll.cl c6 249
El) 12.d5 249
E2) 12.d3 250
E3) 12.J\d3 250
E4) 12.a3 252
E5) 12.el 252
234 4.e3

l .d4 f6 2.c4 e6 3.c3 i.b4 4.e3 0-0 5.i.d3 1 2 .i.c2 should be met by the typical 1 2 . . . i.e7,
d5 6.6 c5 7.0-0 cxd4 8.exd4 dxc4 9 ..bc4 breaking the pin. I don't see anything better
b6 lO.i.g5 for White than 1 3 .a3 , transposing to the main
This is White's most popular continuation, line below.
setting up an annoying pin. Depending
on circumstances, Black may counter it by A fairly toothless alternative is:
retreating his bishop to e7, but he can also 1 2.i.e4 i.e7 1 3 .i.xf6 i.xf6 1 4.Wfa4 a6!?
consider playing with doubled f-pawns in 1 4 . . . ll:la5 1 5 .ixb7 ll:lxb7 1 6.:1l:fd 1 ll:ld6 is
some variations. also fine for Black.

lO ... i.b7
First things first! Before addressing the pin
on the knight, Black brings the light-squared
bishop to the obvious square. White has five
main contenders: A) l l .i.d3, B) l l .e5,
C) 1 I .f;Ye2, D) l l Jel and E) l l Jcl .

A) l l .id3

White takes the bishop out of harm's way and


points it towards the kingside. a b c d e f g h

1 5 .d5N
1 5 .ixc6? b5 1 6.i.xb5 axb5 1 7.Wfd 1 b4
1 8.ll:le2 i.xf3 1 9.gxf3 :i:l:a5+ leaves White
with too many weaknesses.
The previously played 1 5 .:1l:fd 1 ?! is also
inaccurate: 1 5 . . . b5 1 6.Wfc2 g6 1 7.a3 :i:l:c8+
Black was better due to his strong bishops
in Portisch - Gheorghiu, Crans Montana
1 976.
1 5 ... exd5 1 6.ll:lxd5 :i:l:e8 1 7.:1l:ad 1 b5 1 8 .Wfc2
g6=
White's active knight makes up for Black's
bishop pair, but it's not enough for him to
a b c d e f g h
claim any advantage.
l l . .. c6
Often the knight goes to d7 in this structure. 12 ...i.e7 13.i.c2 h6
However, in the main theoretical lines of this Another decent option is 1 3 . . . lLld5!? 1 4.Wfd3
chapter, I favour a set-up with the knight g6 1 5 .i.h6 :i:l:e8 1 6.ll:le4 ll:l f6 and Black had
on c6. I suggest doing the same against this comfortable play in Krivoshey - Jakubek,
sideline, to allow for transpositions. Kosice 1 997.

12.a3 14.i.h4
1 2J:e 1 and 1 2 J:k 1 transpose to the later 1 4.i.f4 i.d6 1 5 .lLle5, as was played in
variations 0 1 and E3 , respectively. Karpatchev - Munkhgal, Moscow 20 1 2,
Chapter 1 6 - l O.ig5 - Main Line 235

should be met by 1 5 .. Jk8N 1 6.e 1 ib8+, B) l l . e5


when the pressure on the isolated pawn forces
White to exchange the powerful knight.

14 ... h5!
This is a typical simplifying mechanism in
such positions. We will encounter it again
in variation D 1 , where I will say a bit more
about it.

1 5.d5
1 5 .ig3N lLlxg3 1 6. hxg3 if6 1 7.Wfd3 g6
1 8.ad 1 lLle?+ leaves Black with the bishop
pair and a firm blockade of the isolated pawn.

1 5 ... exd5 16.Wfd3N


This move is connected with a positional This aggressive move aims to take advantage
pawn sacrifice. of the pin and prevent Black from completing
Clearly worse is 1 6.Wfxd5? Wfxd5 1 7.lLlxd5 his development.
ixh4 1 8.lLlxh4 tLl d4 1 9.ie4 fe8+ as in
Pourramezanali - Barsov, Baku 20 1 2. l l ... bd7!
Black is not afraid of ghosts! Indeed, White
16 ... g6 17.ixe7 xe7 18.d4 has insufficient attacking resources to exploit
the damaged pawn structure after taking on d7
and f6.

The merits of White's previous move are


illustrated in the following line: 1 1 . . . tLl c6
1 2.ixf6 Wfxf6 ( 1 2 . . . gxf6 1 3 .tLlxc6 ixc6
1 4.d5) 1 3 .tLld7 and White wins an exchange.
Black's active pieces yield some compensation,
but White should keep an edge with precise
play: 1 3 . . . Wfh4 ( 1 3 . . .Wff4N 1 4.lLlxf8 xf8
1 5 .tLl e2 Wfh4 1 6.Wfd3 id6 1 7.f4;!;) 1 4.tLl xf8
xf8 1 5 .a3! ie7 1 6.id3! lLlxd4 1 7.ie4 ixe4
1 8.Wfxd4;!; Jiminez - Blake, email 2003.

We will analyse the interesting Bl) 12.xf7!?N


followed by B2) 12.xd7.

1 2.Wfe2 has been the most common choice of


19.adl a6 20.fel fe8 all, and can be found in variation C2 under the
Black has managed to consolidate and keep 1 1 .Wfe2 tLl bd7 1 2.tLle5 move order.
the extra pawn, whereas White's temporary
compensation might disappear soon. 1 2.Wlb3?!
236 4.e3

This looks like a natural move but it runs I was surprised to discover that nobody has
into some tactics. tried this thematic sacrifice. Black should be at
1 2 . . . ixc3! 1 3.bxc3 least equal with precise play, but he will have
1 3.WI'xc3 l2J e4! forces a favourable endgame: to navigate some wild-looking positions, some
1 4.ixd8 lDxc3 1 5 .ie7 l2Jxe5 1 6.dxe5 :gfc8+ of which involve an 'active' king!
Now Black can exploit the lack of harmony
in White's camp by means of: 12 ... xf7!
1 3 . . . l2Jxe5 1 4.dxe5 Wl'c7!! 1 2 . . . :gxf7 1 3 .ixe6 is playable, but Black has
A simple but beautiful tactic. to be careful and White is not really behind in
material. The text move forces White to play
more accurately to justify his last move.

13.d5
1 3.Wfb3 ? ixc3 1 4.ixe6t 'it>g6 1 5 .Wfc2t ie4
1 6.WI'xc3 h6 leaves White with insufficient
compensation for the piece.

13 ... e5!
13 ... e5 1 4.d6t 'it>g6 1 5.f4! exf4 1 6.h4! is too
scary.
a b c d e f g h

1 5 .exf6 14.dxe6t g6
Also after 1 5 .ixf6 gxf6 1 6.exf6 h8 1 7.ie2
:ggs 1 8.g3 Wl'e5+ Black regains the pawn in a
favourable situation.
1 5 . . . WI'c6! 1 6.f3 Wc5t 1 7.h l Wxg5 1 8.fxg7
This was Maloberti - Petters, email 2000,
and now Black is better after the simple
recapture:
1 8 . . . xg7N+

Bl) 12.lthf7!?N

8
7
6
5
4
3
2
1
a b c d e f g h
Chapter 1 6 - l O.igS - Main Line 237

a b c d e f g h

1 8 . . . ll:l fg4! 1 9.'1Wd2t <j;lhs 20.e7 '1Wb8! 2 l .g3


'1Wxa8 22.exf8='1W '1Wxf8 With a massive storm a b c d e f g h
of all Black's pieces against the king. 1 8.gd4t 'it>g5 19.d2t 'it>g6
The king returns to something resembling a
1 5 ... 'it>xg5 16.gadl normal position.
A funny position might arise after 1 6.h4t
xh4 1 7.'\WfS ? (better is 1 7.gad l '1We8, 20.id3t xd3 2 1 .xd3t 'it>h6 22.h3t
transposing to the main line below) 1 7 . . . ll:l f3t
1 8.h l ( 1 8.gxf3 '1Wd2 1 9.f4 h6! is winning for
Black too; the last move covers the g5-square
in preparation for a knight move)

a b c d e f g h

a b c d e f g h 22 ... h5!?
This is an interesting winning attempt.
1 8 . . . ll:lg4! 1 9.g3t <j;lh3-+ A rare case in which
the king feels completely safe in the opponent's
22 . . . g6= is the safe option, when White has
camp!
nothing better than a perpetual.
16 ...e8 17.h4t 'it>xh4
23J:ixb4 '1We7 24JM4 gadS 25.gxd8 gxd8
26.g4 g6CD
Black is not worse in this sharp position; his
bishop is excellent and his king is no longer
any weaker than White's.
238 4.e3

B2) 1 2.xd7 f;Yxd7


8
7
6
5
4
3
2
1
a b c d e f g h
I6.Y;Yg4t <lt>hs I7.f;Yh4
a b c d e f g h
A quiet draw offer.
13.Lf6
Of course, White's previous move was Black also doesn't face any problems after
connected with this exchange. 1 7.WI'd4 Wl'd8 ( 1 7 . . . tJig7 1 8.ad l Wl'c6= is also
possible, since 1 9.d3?! runs into 1 9 . . . ie4!+)
13 ... gxf6 14.d5 1 8.i.xd5 exd5 1 9.fe l c8. Moreover, two
This temporary pawn sacrifice is the best way small inaccuracies led White to a difficult
of handling the position. position after:

1 4.WI'g4t tJih8 1 5 .WI'h4, as played in Arutinian


- Sherbakov, Moscow 2006, can be strongly
met by:

a b c d e f g h

20.e3 ?! c4 2 l .WI'd3 ?! xc3! 22.WI'xc3 d4


23 .Wfd3 dxe3 24.WI'xe3 e8+ Simonet Pons -
Schandorff, Bled (ol) 2002.
a b c d e f g h

1 5 . . . g8!N 1 6.d5 exd5 1 7.id3 g7 1 8 .ad l I7 ....hc4 Is.Y;Yxf6t <lt>gs I9.Vgst <lt>hs
Wl'g4 1 9.WI'xg4 xg4 and White would have to A draw was agreed in Shirov - Giri,
play precisely in order to keep the balance. Hoogeveen 20 1 4.

14 ....hc3 1 5 .bxc3 hd5


Chapter 1 6 - l O.i.gS - Main Line 239

C) l l .Yfe2

a b c d e f g h

1 4.ll:ld2 (after 1 4.:i:l:ac l i.xf3 1 5 .gxf3 lLlh5+


White was suffering from a couple of pawn
weaknesses in Arnold - Perez, Saint Louis
a b c d e f g h 20 1 2) 14 . . . lLld5 1 5 .Wfg4 ll:l 7f6 1 6.Wfh4 h6
This move has been known since 1 920, but 1 7.ixf6 lLlxf6+ Fedoseev - Movsesian, Dubai
it gained some popularity after the World 20 1 4.
Championship match in Moscow 20 1 2, where
Boris Gelfand used it to put Vishy Anand in all 1 2.:i:l:ad 1 can be met in much the same way:
sorts of trouble. 1 2 . . . ixc3 1 3 .bxc3 Wfc7 1 4.ll:ld2 Now in
Plaskan - Pasko, Kerner 2009, Black could
l l ... ttlbd7 have obtained some advantage with:
l l . . .ltk6 makes less sense when the queen
has already vacated the d 1 -square, and after 8
1 2.:i:l:ad 1 White already threatens d4-d5 . 7

6
1 2 . . . ll:l a 5 1 3 .i.d3 h 6 was seen i n Kashlinskaya
- Saduakassova, Skopje 20 1 5 , and now White
5
could have utilized the poor placement of the
4
knight on aS by means of 1 4.i.xf6!N Wfxf6
1 5 .ll:le4 Wfd8 1 6.a3 ie7 1 7.b4 ll:l c6 1 8.ll:lg3, 3

keeping some initiative. 2

Mter the text move White has tried several


a b c d e f g h
ideas, but we will focus on the most logical
options of Cl) 12.d5, C2) 12.ttle5 and 1 4 . . . h6!N 1 5 .ih4 lLld5 1 6.ig3 Wfc6+
C3) 12J:acl . White must either give up a pawn or make a
positionally unfavourable exchange on d5.
1 2.:i:l:fd 1 makes d4-d5 into a serious threat, so
the following exchange makes perfect sense: 1 2.id3 h6 1 3.ih4 occurred in Pyrich -
1 2 . . .ixc3 1 3 .bxc3 Wfc7 Farrell, Scotland 1 994, when Black could
have obtained a good position with a typical
plan: 1 3 . . . i.xc3N ( 1 3 . . . ie7!?N 1 4.lLle5 ll:ld5
also looks decent) 1 4.bxc3 Wfc7 1 5 .:i:l:ac 1 ll:lh5
1 6.ib 1 Wff4 1 7.Wfd3 ll:l df6 1 8.ig3
240 4.e3

a b c d e f g h

1 8 . . . ie4! ( 1 8 . . . lLlxg3 1 9.fxg3 Wle4 20.Wfd2


Wfd5 2 1 .tLle5 lLl e4 22.Wfd3 b5 23.gf4 f5 leads
to double-edged play) 1 9.ixf4 ixd3 20.ixd3
lLlxf4= Liquidating into a comfortable
endgame.

Cl) 12.d5
1 5Jadl
8 1 5 .bxc3 lLl c5 1 6.ixe6t lLlxe6 1 7.Wfxe6t
'it>h8 is similar to the main line, and the loss
7
of one of the bishops can hardly help White.
6
15 ...Wfc7 16.he6t cot>hs 17.bxc3
5
4
8
3
7
2
6
1
5
a b c d e f g h
4
This rare but important move leads to
interesting complications, but they are 3
acceptable for Black: 2
1
12 ... hc3
Much weaker is: 1 2 . . . exd5?! 1 3 .tLlxd5 a b c d e f g h
ge8N ( 1 3 ... ixd5?! 1 4.ixd5 was downright 17 .. Jae8 18.Wfc4 Wlb7
unpleasant for Black in Bindrich - Houriez, Despite his extra pawn and bishop pair,
Puerto Madryn 2009) 1 4.Wfc2 id6 1 5 .gad l White had too many weaknesses to hope
gc8 1 6.Wfd3 With some initiative. for an advantage in Arizmendi Martinez -
Gharamian, Cappelle-la-Grande 20 1 6.
13.dxe6 .b3 14.gxf3
Chapter 1 6 - l O.igS - Main Line 24 1

C2) 1 2.c e5 Another decent option is 14 . . . lLlxe5 1 5 .dxe5


:gc5, when the lesser evil for White would
be forcing a draw by means of: 1 6.id3!
8
( 1 6.f4? :gxc4 1 7.exf6 g6 gave Black a clear
7 positional advantage in Hawkins - Kramnik,
6 London [rapid] 20 1 4) 1 6 . . . WI'c7 1 7.ixf6 gxf6
A draw was agreed in Avotins - Chripko, email
5 2008, in view of: 1 8.WI'g4t c;i;>h8 1 9.WI'h4 f5
4 20.WI'f6t=
3 1 5.xd7 Wfxd7 16.ixf6 gxf6
2 The position is equal but far from dead. In
the following game, White went wrong by
1
trying to force a draw too quickly:
a b c d e f g h
This most aggressive option is well met by: I7.Wig4t c;i;>hs Is.Wih4
White hopes to give up his bishop to force a
12 ...ixc3 13.bxc3 c8 perpetual, but he has overlooked an important
Just as in variation B above, Black is not detail.
worried about the doubling of the f-pawns
that could result from exchanges on d7 and
f6, as White's remaining pieces would be in no
position to carry out an attack.

14.acl
This has been by far the most common
continuation, but Black has more than one
good reply.

8
7
a b c d e f g h
6
18 g8! 19.Wixf6t gg7-+
5
. .

White must either give up a piece or


4 succumb to a cute mating attack, as occurred
in the game:
3
2 20.ie2 Wfe7 2 I .Wixe7 gxg2t 22.<i!?hl ggl t!
23.ci!?xgl g8t 24.ig4 gxg4#
1
Szelag - Socko, Rostock 20 1 5 .
a b c d e f g h
14 ...Wfc7!
242 4.e3

C3) 12Jacl I3.!Udl


The black queen is keeping an eye on the key
e5-square, so 1 3 .lLle5?! only invites trouble:
1 3 . . . ll:lxe5 1 4.dxe5 Axc3 1 5 .bxc3 ll:le4 1 6.Af4
gc8 1 7.gfe 1

a b c d e f g h

a b c d e f g h
The most consistent. Since the isolated pawn
is less vulnerable with Black's knight on d7
instead of c6, White is free to develop his rook This position arose in Dambacher -
in a way that combines active and prophylactic Henrichs, Maastricht 20 1 5 . Had Black played
duties. In particular, the c3-knight is over 1 7 . . ,gc5!N 1 8.Ad3 ll:lxc3 1 9.VIig4 V!ifB, White
protected, and the c7 -square is no longer safe would not have had adequate compensation
for Black's queen. for the pawn.

12 ti'b8!
...
1 3.gfe l gc8 1 4.id3 h6! and now either
This elegant way of avoiding the pin and bishop retreat carries a drawback:
solving the problem of the dB-queen is one
more example of Vladimir Kramnik's deep
home preparation. Black takes control of the
important b8-h2 diagonal, while the blocked
a8-rook is only a temporary problem.

Here is a fragment of the World Championship


game which sparked a lot of the recent
attention on this variation:
12 .. Jk8 1 3 .Ad3 Axc3
1 3 . . . Ae7 and 1 3 . . . h6 deserve attention as f g h
well.
1 5 .id2 ( 1 5 .ih4 exposes the f4-square, so
1 4.bxc3 V!ff c7 1 5 .c4 Axf3? 1 6.V!ffxf3 gfe8
1 5 . . . lLlh5! 1 6.ib5 ll:l df6 1 7.lLle5 a6 1 8.Aa4
Anand must have intended 1 6 . . . e5, only
Ad6 offers Black comfortable play) 1 5 . . . V!id6
now realizing that 1 7.Af5! would lead to
Connecting the rooks and activating the
huge problems for him.
queen. 1 6.ll:lb5 V!ff e7 1 7.Axb4 V!ffxb4 1 8.a3.
1 7.gfd 1 h6 1 8.ih4
We have been following the game Peralta -
White was much better in Gelfand - Anand,
Riwuk, Barcelona 20 1 5 . Now Black could
Moscow (9) 20 1 2.
Chapter 1 6 - l O.igS - Main Line 243

have neutralized the temporary activity of with: 1 5 .d5 lDxdS 1 6.l2Jxd5 l::1xc l 1 7.ixc l
White's pieces by means of: ixd5 1 8.ixd5 exd5 1 9.l::1 xd5 lD f6=) 1 5.bxc3 h6

4
L . . ,J=''
3
-=------ ----
2

a b c d e f g h a b c d e f g h

1 8 . . . Wff8!N 1 9.E1xc8 Wfxc8 20.l2Jd6 Wfc6 1 6.ixf6 ( 1 6.ih4 Wff4 1 7.ig3 Wlg4 1 8.E1e l
2 1 .lDxb7 Wfxb7 22.l2Je5 WfdS The position ltJ hS =) 1 6 . . . l2Jxf6 1 7.l2Je5 E1c7 1 8.c4 lD d7
is still about equal, but the weakness of the 1 9.E1c3 lDxeS 20.Wfxe5 E1d7 The position is
d4-pawn might tell in the long run. equal, but plenty of play remains.

13 .. J:fc8! 14.id3
White safeguards the bishop and points 8
it towards the kingside. On the other hand, 7
Black obtains full control over the d5-square.
6
Black is well prepared for the thematic 5
advance: 1 4.d5 l2Jxd5 1 5 .l2Jxd5 exdS 1 6.ixd5 4
ixdS 1 7Jhc8t Wfxc8 1 8.l::1 xd5
3
2
1
a b c d e f g h
I4....td6
The bishop no longer has much of a role on
b4, so Black relocates it to a stable square in
the centre.

a b c d e f g h 1 5.g3
1 8 . . . Wfc6 1 9.Wfd l l2J f6 20.l::1 d 8t E1xd8 1 5 .h3 h6 1 6.ixf6 lDxf6 1 7.l2Je4 ie7 gives
2 1 .Wfxd8t ifB= Black easy play.

1 4.ib3 is also harmless. 1 4 ... ixc3!? This is the I also considered 1 5 .ia6 h6 ( 1 5 . . . ixa6
ambitious option. (If a draw is an acceptable 1 6.Wfxa6 if4=) 1 6.id2 E1c7 1 7.id3 lDdS
result, then 1 4 . . . id6 invites simplifications 1 8 .l2J e4 if4 and Black is doing well.
244 4.e3

D) l l .el

a b c d e f g h

1 5 ... a6
a b c d e f g h
1 5 . . . h6 1 6 . .ixf6 ltJ xf6 1 7.ltJe4 l::1 xc 1 1 8.l::1 xc l
'1Wd8 1 9.l2Jxf6t Wxf6 20 . .ie4 'We? is another l l ... c6
route to an equal position. l l . . .l2J bd7 is much more popular, but the
text move appeals to me. The knight puts
16 . .bf6 xf6 17.e4 xcl 1 8.xcl pressure on the d4-pawn and may later move
We have been following the top-level game towards the kingside via e7.
Gelfand - Kramnik, London 20 1 3 . The
position is roughly equal, but Black could We will look at two options: Dl) 12.i.d3 is a
have made things mildly unpleasant for his harmless alternative to D2) 12.a3 .
opponent with:
1 2.E1c l transposes to variation E5 .

1 2.'1Wd3 , as tried in lvanchuk - Van Wely,


Monte Carlo (rapid) 2002, can be well met by:
1 2 . . . l2J a5!N

a b c d e f g h
1 8 ....le7N 19.xf6t .bf6 20.i.e4 g6
It should be a draw, but White still has to a b c d e f g h
think about the potential weakness of the
1 3 .ib5 ixf3 1 4.gxf3 (in the event of
isolated pawn.
1 4.Wxf3 Wxd4 1 5 .E1ad l Wg4 1 6.'1Wxg4 l2Jxg4
Chapter 1 6 - l O .igS - Main Line 245

1 7J:e4 ixc3 1 8.bxc3 ltl f6 1 9.ixf6 gxf6)


White has insufficient compensation for the
pawn) 14 . . . ie7 1 5 .f4 ltl d5 1 6.WI'g3

a b c d e f g h

a b c d e 14 ... h5!
The exchange of dark-squared bishops
1 6 . . . ltlxc3 1 7.bxc3 ixg5 1 8.fxg5 c8+ White
would make Black's set-up more harmonious.
no longer has doubled pawns, but his position
This approach was also successfully employed
still contains numerous holes and weaknesses.
by former World Champion Anatoly Karpov
in such situations. I remember his impressive
Dl) 1 2 ..ld3 h6 13 ..lh4 i.e7 14J:kl
victory over Viktor Korchnoi in their 1 98 1
World Championship match in Merano.
This is the only move to have been tried.
1 5 ..lg3
1 4.a3N may be a touch more precise, although
1 5 .ixe7 ll:lxe7 1 6.ltle5 ll:l f4+ is unpleasant
Black seems fine after: 1 4 . . . ltlh5 1 5 .ig3 ltlxg3
for White.
1 6.hxg3 if6 1 7.ie4
1 5 .d5!?N is not quite correct, but it is worth
taking a quick look at the remarkable idea
connected to it: 1 5 . . . exd5 1 6.ll:lxd5 ixh4
1 7.ltlxh4 Wl'xh4 1 8.c4 White almost regains
the piece while keeping an active position, but
Black can counter with:

a b c d e

1 7 . . . WI'd6! This leads to an approximately equal


position. (I prefer to avoid 1 7 . . . b8 1 8.d5!
exdS 1 9.ltlxd5 ixb2 20.a2 if6 2 l .d2
with dangerous complications.) Now 1 8.d5 is
toothless: 1 8 . . . exd5 1 9.ltlxd5 fd8=
a b c d e f g h
246 4 .e3

1 8 . . . g5 ! 1 9.h4 xd5! 20.h7t xh7 lasting advantage due to the passive bishop on
2 1 .xd5 ltlf6 I believe Black's three pieces are b7) 1 5 .e4 f6 1 6.a4! ltl a5 1 7.ltle5 e7
worth more than a queen here. 1 8.ad l;!; White was better in Naiditsch -
Bluebaum, Deizisau 20 1 2.
1 5 ... xg3 16.hxg3
We have been following the game Barsov - 13.bxc3
Demianjuk, Moscow 20 1 5 . Now Black missed
a great opportunity to create problems for his
opponent:

a b c d e f g h
16 ....tg5!N 17.gal
1 7.ltlxg5 xg5 1 8.d5 exd5 1 9.ltlxd5 adS The previously played 1 3 . . . c8 14.d3 ltl e7
20.ltlf4 ltl b4+ is also problematic for White. seems weaker, as it doesn't force White to
exchange his bishop. Indeed, after 1 5 .b3
17 ... .tf6 IS .te4 gbs 19.d5 exd5 20.hd5
ltlg6 1 6.ad 1 c7 1 7.xf6 gxf6 1 8.c4;i; White
f;Yd7i was well prepared for the key d4-d5 advance in
White has swapped off his isolated pawn, Wojtaszek - Socko, Berlin (blitz) 20 1 5.
but Black's bishop pair gives him the upper
hand. 14 ..th4
This seems like the only principled reply.
02) 12.a3
1 4.f4 leaves Black at liberty to occupy the
This is a more challenging move, forcing Black light squares: 1 4 . . . c8 1 5.d3 ltl a5 1 6.c l
to decide what to do with the bishop. ltl c4 with excellent play.

12 ...bc3! 14 ... e7!


1 2 . . . e7 has been more popular. Then A typical manoeuvre - with the support of
1 3 .a2 ltld5 was played once by the great Black's queenside knight, Black's king will be
expert Anatoly Karpov, but I do not like it in considerably safer following the damaging of
view of: 1 4.xd5 xg5 ( 1 4 . . . exd5 1 5 .xe7 the pawn structure.
ltlxe7 1 6.cl also yields White a small but
Chapter 1 6 - I O.i.g5 - Main Line 247

I S.i.x6
Again, the most challenging move.

After 1 5 .i.d3 lt:l f5 1 6.ixf5 exf5 1 7.ixf6 Wxf6


1 8.lt:le5 ac8 1 9.c4 fe8 the pressure along
the a8-h 1 diagonal and the vulnerability of
the d4-pawn prevent White from enjoying his
better pawn structure.

1 5 ... gxf6
a b c d e f g h

1 8 . . .f5! 1 9.e3 f4 20.h3 lt:l g6! 2 1 .tLlf3


After 2 1 .lt:lxg6 fxg6 22.el f6 White
would suffer from the awkward placement
of the rook on h3.
21 ... ixf3 22.xf3 lt:lh4!
As usual, combining strategy and tactics!

a b c d e f g h
16.i.b3
After 1 6.Wfd3 We? 1 7 .i.b3 Wf4! 1 8.i.c2
f5 1 9.lt:le5 WigS (it makes sense to provoke
g2-g3 , so that White's rook will not have access
a b c d e f g h
to the kingside) 20.g3 ac8 2 1 .c4 fd8?
Black's pieces are perfectly mobilized for 23.h3
attacking the opponent's centre. 23.xf4?? WigS 24.g3 lt:lf3t!-+ is a nice
detail.
White may also try to develop a direct attack 23 . . . Wg5 24.ie4 e5 25.'it>h l f5 26.id5 fd8
with: 27.Wa2 exd4 28.cxd4 Wf6?
1 6.id3 c8 1 7.Wfd2 With such active pieces, Black should not be
After 1 7.c l Wd6 1 8.Wfd2 g7 1 9. lt:l h4 worse.
lt:l d5! 20.c4 Wf4 Black manages to liquidate
into a comfortable endgame. 16 tLlg6 17.ti'd3
. . .

1 7 . . . 'it>g7 1 8.lt:l h4!? After 1 7.e3 f5 1 8.c4 Wf6 1 9.d5 White


The idea of activating the rook along the manages to block the powerful bishop for
third rank is dangerous. However, Black is a while, but it doesn't offer any advantage:
well placed to trade punches on the kingside 1 9 . . . fd8 20.a2
with:
248 4.e3

a b c d e f g h

20 . . . b5! 2 1 .:gd2 bxc4 22 . .ixc4 e5 23.Wlal


:gac8 The position is dynamically balanced. a b c d e f g h
A sample continuation is: 24 . .ib3 Wig? 20.d5!?
25.lLlxe5 lLlxe5 26.Wfxe5 :gel t 27.:gd l :gxd l t 20.c4 is well met by the prophylactic
28 . .ixd l :gxd5= Leading to a drawn endgame. 20 . . . :gfe8! and if 2 1 .tLle5 then 2 1 . . .Wfg5 is
equal.

20 ... 4!
Interrupting the harmony of White's pieces.

2 1 .e2 hd5 22.hd5 exd5 23.gxd5 xc3


White's piece acuvtty gives definite
compensation for the pawn, but he has no
advantage. For example:

8
7
a b c d e f g h 6
17 ... 5 1 8.e3 5
In the event of 1 8 .:gad l Wff6 1 9.c4 Wig?
20.Wfe3 :gac8 White's play is slowed down, 4
since 2 1 .d5? exd5 22.cxd5 loses on the spot to: 3
22 . . . :gc3! 23.:gd3 :gxd3 24.Wfxd3 lLl f4 25.Wffl
.ia6-+ 2
1
18 . . <tt> g7 19Jad1
a b c d e f g h
.

The impulsive 1 9.d5?! .ixd5 20.:gad l f4


2 l .Wfc l :gc8+ leaves White with insufficient 24.d1 gfe8 25.gfl f6
compensation for the pawn. The greedy 25 . . . Wfxa3 ?! 26.tLld4 is not
recommended.
19 ...6
Chapter 1 6 - l O .ig5 - Main Line 249

E) l l .cl

This is the most popular continuation. Once


8
again, Black must decide how to develop the
queen's knight. 7
6
8 5
7 4
6 3
5 2
4 1
3 a b c d e f g h
2 16.a3 exd5 17.g3 g4 18.f6t! gxf6
1 19.axb4 c4 20.ie2
Also after 20.ib l ll:le5 2 1 .ll:lxe5 Wfxd l
a b c d e f g h
22.:gfxd l fxe5 23.:gc7 :gabS Black is out of
l l ... c6 danger.
There is nothing wrong with l l . . .ll:l bd7
1 2.a3 ixc3 1 3Jhc3 , but in this line the 20 ... fe8 2 1 .d4
mobile rook on c3 offers White some attacking It looks as though White is developing a
prospects. That's why I prefer the text move. powerful initiative, but Black can fight back
We will look at five options: El) 12.d5, with:
E2) 12.d3. E3) 12.id3, E4) 12.a3 and
E5) 1 2.e l .

El) 12.d5

For some reason, this tempting advance has


been tried only once in tournament practice.
It can lead to interesting complications.

12 ... a5 13.id3N
This is better than 1 3 .ll:le4? as played
in Yuksel - Baskara, Kayseri 20 1 0, when
1 3 ... ie7!N 1 4.d6 ixe4 1 5 .dxe7 Wfxe7+ would
have left White with no compensation for the a b c d e f g h
pawn. 2 1 ... x:e2! 22.xe2 e5
The weakness of the light squares gives Black
13 ... h6 plenty of counterplay.
1 3 . . . exd5 1 4.ib l ! offers White promising
compe : ' '1 .
250 4.e3

E2) 12,f;Yd3 c8 17.c4 f;Yh5 IS ..tx6 gxf6 19.d5 exd5 20.c5


c4i
White had insufficient compensation for the
8
pawn in Aronian - Caruana, Wijk aan Zee
7 20 1 5 .
6
E3) 1 2.i.d3
5
4
3
2
1
a b c d e f g h
13.fdl
1 3 .ltJ e4 i.e? leads to unfavourable exchanges,
after which White will have no compensation
for the isolated d4-pawn.
a b c d e f g h
13 ... a5
A ryp ical motif- attacking the bishop is often This move has been tried by several strong
effective when the d3-square is unavailable. players, but it often transposes to variation E5
after a subsequent :i:l:e l .
14.i.b5 h6 15.i.h4 .bc3 16.bxc3
1 2... h6 13.i.h4 i.e7
Black prepares the thematic bishop exchange,
which will simplify the position and establish
full control over the d5-square.

14.i.b l
1 4.:i:l:e 1 :i:l:c8 1 5 .ib 1 transposes to variation
E5.

1 4.a3 has been tried in a couple of games.


I suggest our usual approach: 1 4 . . . l2Jh5N
1 5 .ig3 ( 1 5 .ixe7 l2Jxe7 is simply better for
Black) 1 5 . . . l2Jxg3 1 6.hxg3 i.f6 1 7.ie4 :i:l:b8
a b c d e f g h
16 ...f;Yd5!
White cannot afford to lose control over the
c4-square, so the next move is forced.
Chapter 1 6 - l O.igS - Main Line 25 1

8
7
6
5
4
3
2
a b c d e f g h
1
The pressure on the isolated pawn practically
forces White to play 1 8.d5 exd5 1 9.ixd5, but a b c d e f g h
then after 1 9 . . . tlJa5 20.b4 ixc3 2 l .ixb7 ib2 20.ia2!N
22.gb 1 gxb7 23.gxb2 tlJ c4 Black's position is Supporting the powerful knight is the best
preferable due to his more active knight. way of handling the position.
20.tlJxf6t Wfxf6 2 l .Wfd2 tJig7 22.Wfc3 gfd8
8 23.Wfxf6t tJixf6+ led White to an inferior
endgame in Leenhouts - Ernst, Vlissingen
7 20 1 5 .
6
20 ... e5!
5 20 ... ixb2 2 l .gc2 ig7 22.gd2 would give
4 White full compensation for the sacrificed
pawn.
3 The text move is simpler - Black is inviting
2 favourable simplifications.
1
a b c d e f g h

14 .. Jc8 1 5.a3
Covering the b4-spot, so that the queen can
go to d3 without fearing harassment from the
enemy knight.
1 5 .ge 1 is another transposition to variation
ES .

1 5 ... h5 16.ig3
1 6.'1Wd3 ?! achieves nothing: 1 6 . . . g6 1 7.ig3
tlJxg3 1 8.hxg3 if6 1 9.gcd 1 ig7 20.gfe 1
a b c d e f g h
tlJe7+ Van der Stricht - Burg, Belgium 20 1 5 .
2 1 Jxc8 ixcS 22Jel xf3t 23.xf3 <lt>g7=
16 ... xg3 17.hxg3 if6 18.d5 exd5 19.xd5 White has nothing better than liquidating
g6 into an equal endgame.
252 4.e3

E4) 12.a3

This has been the most popular choice but


Black is well equipped to meet it.

a b c d e f g h
16 ... ha6!N 17.Yfxa6 d5 1 8Jc4
Mter 1 8.xc8 xc8 1 9.xa7 c2 Black will
regain the pawn with a positional advantage,
a b c d e f g h
since 20.b4?! c6 does not help White at all.

12 ...hc3 13Jxc3 18 .. J:xc4 19.Yfxc4 Yfd7 20Jcl e7


1 3.bxc3 has no logical connection with the Black has a small but lasting positional
rook on c l , and after 1 3 .. .1:'k8 1 4.e2 d6! advantage, as the IQP is a more significant
Black is at least equal. weakness than Black's damaged kingside.
13 .. Jc8
ES) 12.el c8
1 3 . . . tLle7 has been played a few times but it
makes more sense to delay it for another move.
Keeping the knight on c6 means that White
is tied to protecting the isolated pawn, and he
has no way to take advantage of pinning the
other knight.

14.Yid3
I 4.ia2 lLle7 1 5 .l'hc8 xc8 1 6.tLle5 lLl fd5+
was ideal for Black in Radjabov - Karpov,
Buenos Aires 200 I .

14 ... e7!
Now is the time to open a path for the
light-squared bishop, while solidifying Black's a b c d e f g h
control over the d5-square. Once again, Black develops the rook and
menaces the hanging bishop on c4, forcing
I S ..b6 gxf6 16 ..ia6 White to waste time on prophylaxis.
We have been following the game Potkin -
Alekseev, Moscow 20 1 2. Natural and strong 13.i.d3
would have been:
Chapter 1 6 - I O . .ig5 - Main Line 253

1 3 .Wfd3N isn't without drawbacks either: 1 9.E1e3 f5 20.E1h3 ig5+ With a healthy extra
1 3 ... h6 1 4.ih4 ie7 1 5 .a3 We have transposed pawn.
to a game, in which Black made use of a
familiar device: 14 ... h6 1 5 ..th4
1 5 .ixf6 ixf6 1 6. WI d3 g6+ leads nowhere for
White.

8
7
6
5
4
a b c d e f g h 3
1 5 . . . l2Jh5! 1 6.ixe7 ( 1 6.ig3N lDxg3 1 7.hxg3 2
if6 1 8.l:kd 1 l2Je7+) 1 6 . . . l2Jxe7 1 7.l2Je2 l2Jg6
1 8.g3 l2J f6+ Black had a comfortable position
playing again the isolated pawn in Pribyl - a b c d e f g h
Matejka, corr. 2005. 1 5 ... h5!
13 ....ie7 14 . .ibl The same approach as in some of the notes
I also considered: 1 4.a3 l2Jd5 As we often above, as well as the earlier variation D 1 - the
see in this variation, the exchange of dark exchange of bishops will benefit Black.
squared bishops makes Black's play easier. 16 . .tg3!
1 5 .h4!? This aggressive move is connected with This is the only way for White to maintain
attacking ambitions, but they are not realistic. the balance.
1 5 ... l2Jxc3 1 6.bxc3 h6 1 7 . .id2 So far we have
been following Henrichs - Walter, Germany 1 6.Wfd3 ?! g6 1 7.ig3 l2Jxg3 1 8.hxg3 if6+
20 1 2, and here Black should have played: was pleasant for Black in Bindrich - Meier,
Dresden 20 1 5 .

1 6.Wfc2?! g6

a b c d e f g h

1 7 ... l2J a5! Aiming at the c4-square and thus


provoking White's next move. 1 8.l2Je5 ixh4
a b c d e f g h
254 4.e3

1 7Jhe6 Conclusion
The quiet 1 7.i.xe7 CiJxe7 1 8.CiJe5 CiJ f4+ also
leads to a bad position for White. There is no doubt in my mind that 10.i.g5 is
1 7 . . . t'iJ f4! the best way of developing active play - with
This cold-blooded reply allows Black to grab the Nimw bishop far away on b4, the pin
the material in a comfortable situation. becomes quite annoying. After the natural
1 8J:he7 reply 1 0 . . . i.b7 there is a variety of possible set
1 8.e4 CiJ b4 1 9 .'1Wd2 i.xe4 20.i.xe4 i.xh4 ups, but White's main intentions are:
2 1 .%Vxf4 ig5-+ 1 ) getting control over the d5-square and/or
1 8.i.xe7 CiJxe7+ forces White to give up the pushing d4-d5 at a suitable moment;
exchange with 1 9.e4, since 1 9.e3? CiJ f5 2) creating threats on the kingside, utilizing
would be even worse. a powerful knight on e5 (which may sacrifice
1 8 . . . t'iJxe7 1 9.Wd2 g5+ itself on f7, supported by the bishop on c4)
Black was much better and eventually won and perhaps a transfer of heavy pieces along
in I. Sokolov - Leko, Wijk aan Zee 20 1 3 . the third rank.

In his turn, Black benefits from harmonious


development and pressure on the isolated
pawn, and it's useful to keep in mind that
straightforward bishop and knight exchanges
will almost always favour Black. There are
a few concrete lines to memorize, such as
1 1 .CiJ e5 CiJbd7 followed by 1 2.t'iJxf7!?N or
1 2.CiJxd7, but Black is doing well in the
arising complications. 1 1 .%Ve2 seems a good
try for White, as it makes our usual set-up
with . . . CiJ c6 less effective, but then Kramnik's
excellent 1 2 . . . %Vb8! enables Black to neutralize
a b c d e f g h the thematic d4-d5 break. After the most
16 ... xg3 17.hxg3 .tg5N common 1 1 .c l , Black gets a good game with
This is the simplest way to ensure an equal 1 1 . . . CiJ c6, which can generally be followed up
position. by . . . c8 and . . . CiJe7 in the near future.

17 . . . i.f6!? is more complex: 1 8.d5 Axc3


1 9.dxc6 he 1 20.cxb7 xc l 2 1 .Wxc l i.b4
22.Wc2 g6 23.CiJe5 The position looks messy,
but it soon resulted in a draw in Aberbach
Peltzman - Sukhorskij, email 20 1 4.

18.xg5 f;Yxg5 1 9.d5 exd5 20.Vxd5=


4.9c2
a b c d e f g h

Various 5 th Moves
Variation Index
l.d4 f6 2.c4 e6 3.c3 J\b4 4.f;C/c2
4...d5
A) 5.e3 256
B) 5.J\g5 257
C) 5.f3 dxc4 258
Cl) 6.J\g5 259
C2) 6.e4 260
C3) 6.a3 26 1

C) note to 6th move C 1 ) note to 7 .a4 C3) note to 1 0.xc4

a b c d e f g h a b c d e f g h

l l . . .e4!N 9 . lLl c6!N


. . l O . . . lLlaS!N
256 4.Yic2

l .d4 f6 2.c4 e6 3.c3 i.b4 4.Yic2 Rubinstein systems doesn't make much sense,
This is known as the Classical System, and but this is still White's third most popular
is arguably the most theoretically challenging choice! We won't have to spend much time
answer to the Nimzo-Indian. The main idea on it though, as my recommended solution is
of the queen move is to prepare a2-a3 without likely to transpose to a harmless line which has
suffering from doubled c-pawns. The queen already been analysed in another part of the
may also support the e2-e4 advance. book.

4 ... d5 5 ... c5 6.f3


Black has a number of other options, with I checked two other continuations:
4 . . . 0-0 and 4 . . . c5 deserving a mention as the
other two 'big moves'. I cannot claim that the 6.a3 i.xc3t 7.bxc3 (7.Wxc3 cxd4 8.exd4 dxc4
text move is objectively superior, but it's the 9.ixc4 0-0 1 0.tLlf3 transposes to variation
one I have chosen to recommend for this book. A2 of Chapter 1 0, where Black stands better
Black is fighting for the light squares in the as he has an improved version of a normal
centre, preventing e2-e4 and forcing White to IQP position) 7 . . . 0-0 This position has been
decide what to do about the hanging c4-pawn. covered on page 1 1 2 - see 7.Wfc2 d5 8.e3 in
the notes to variation D of Chapter 8; although
In this chapter we will deal with three that line itself may soon transpose to variation
comparatively rare lines: A) 5.e3, B) 5.i.g5 , B2 of Chapter 2 1 !
and C) 5.f3.
6.dxc5 0-0 should enable Black to equalize
5.a3 is the first of the major options, which quite easily. 7.a3 This position has been
will be discussed in Chapters 1 8-20. reached in two games. In my opinion, the
simplest solution is:
5.cxd5 is the most challenging move of all, and
will be covered in Chapters 2 1 -24.

A) 5.e3

a b c d e f g h

7 . . . i.xc5N 8.tLlf3 dxc4! (8 . . . tLl c6 transposes


to the note to Black's next move in the main
line below, where 9.b4 i.d6 1 0.i.b2 gives
White good prospects) 9.i.xc4 a6 1 0.b4 i.e?
1 l .i.b2 b5 1 2.i.d3 i.b7= In this position,
with symmetrical pawn structures, Black has
As often happens, mixing the Classical and no reason to worry.
Chapter 1 7 - Various 5th Moves 257

a b c d e f g h a b c d e f g h
6 ... cxd4! 7 ... 0-0
This is the most accurate move order. We can end the line here, as Black's last move
reaches a position analysed via the Rubinstein
6 . . . ll:lc6 move order - see variation A3 of Chapter 1 0.
This has been by far the most popular move,
and was even used by Nimzowitsch to score B) s ..t8s
1 V2/2 against Reti and Capablanca! However,
it gives White the annoying option of:
8
7.dxc5!
7.a3 i.xc3t 8.bxc3 0-0 is similar to the main 7
Rubinstein lines from Chapter 1 4, but the 6
presence of White's queen on c2 (instead of
the bishop on d3) makes Black's task easier. 5
The games Korchnoi - Yusupov, Switzerland 4
2008, and L. Horvath - Ragger, Mureck
2004, are good examples of Black's chances.
3
7 . . . 0-0 8.a3 i.xc5 9.b4 id6 1 0.ib2 2
This theoretical position can be reached via
1
different move orders, and first occurred as
early as 1 909! Solving the problem of the a b c d e f g h
c8-bishop is not an easy task here. For This looks like a natural move, but it is
instance: connected with a dubious pawn sacrifice.
1 0 . . . ll:le5 1 1 .0-0-0 Wff e7 1 2.lLlb5!
White obtained a dangerous initiative in the s .. dxc4
.

classic game Tal - Aronin, Moscow 1 957. Why not? It's surprisingly hard for White to
regain the pawn, so he will have to search for
7.exd4 compensation.
After 7.ll:lxd4 0-0 White is equal at best. In
practical terms, Black has an easy position with 6.e3
a lot of freedom, and has achieved a heavy plus 6.ll:lf3 converts to variation C l .
score.
25 8 4.'1Wc2

6 ... b5 the following natural way to complete


This is not merely a materialistic approach development:
- the extra c4-pawn greatly restricts White's
pieces, and the last move prepares to develop
8
the bishop on b7.
7
7.a4 6
White should hurry up and play this in order
to force . . . c6. IfWhite delays it, Black will play 5
. . . b7 and then be ready to meet a2-a4 with 4
... a6, when the light-squared bishop will have
a perfect view of the long diagonal. 3
2
7... c6
1
a b c d e f g h
10 ....tb7N 1 1 .ge2 d7 12.0-0 gb8
13.axb5 axb5 14.b3 cxb3 1 S.f;Yxb3 .te7
16.4 0-0
Even though White retains some
compensation, I still prefer Black - there are
no real weaknesses in his camp, while the
queenside passed pawn should tell in the long
run.

a b c d e f g h
This position might also be reached via the
s.L6 Ragozin move order of l .d4 d5 2.c4 e6 3.tLlf3
White is happy to destroy the opponent's lLlf6 4.tLlc3 b4, and now the somewhat
pawn structure at the cost of exchanging the premature 5.'1Wc2.
powerful bishop.
8
After 8.e2 b7 9.f3 a6 1 0.tLlge2 tLl bd7
1 1 .0-0 '1Wb6 Black was a sound pawn up in 7
Levitas - Sokolsky, Leningrad 1 938. 6
8 ... gxf6 5
Avoiding a thematic trap: 8 . . . '1Wxf6? 9.axb5 4
cxb5? 1 0.'1We4+- and the rook is trapped.
3
9 . .te2 a6 10 ..tf3 2
This position occurred in Wehmeier -
1
Kotronias, Munich 1 993. Now I suggest
a b c d e f g h
Chapter 1 7 - Various 5th Moves 259

5 dxc4
This was played almost a century ago in
Once again, Black should be happy to grab Vilardebo Picurena - Noteboom, Prague
an important pawn. 1 93 1 . Black could have taken full advantage
of the premature development of White's
White may proceed with Cl) 6 .tg5, C2) 6.e4
. queen with:
or C3) 6.a3.

I also considered the modest-looking:


6.e3 b5 7.a4 c6

a b c d e f g h

1 l . ..ie4!N 1 2.VIid 1 a5 1 3 .bxc4 b4 1 4.ib2


ltl bd7 1 5 .ie2 0-0 1 6.0-0 V!ff c7
The two strong queenside passed pawns
a b c d e f g h
allow Black to seize the initiative.
This transposes to a version of the Noteboom
variation, which occurs after the opening CI) 6 .tg5

moves l .d4 d5 2.c4 e6 3.ltlf3 c6 4.ltlc3 dxc4


5.e3 b5 6.a4 ib4. The main line continues This is a reasonable move, but once again the
7.id2 a5 8.axb5 ixc3 9.ixc3 cxb5 1 0.b3 set-up with an early V!ff c2 is rather slow.
ib7 1 l .bxc4 b4 1 2.ib2, with an extremely
double-edged game ahead. Going back
8
to move 7 of the Noteboom line, the less
than-optimal 7.V!ff c2 would transpose to our 7
position. 6
8.id2
White has nothing better, but he is almost a 5
tempo down on the Noteboom line, as there 4
is no real point in putting the queen on c2
at this stage. 3
8 . . . ib7 9.axb5 ixc3 1 0.ixc3 2
I also examined 1 0.V!ffxc3N cxb5 1 l .b3 a5
1
1 2.bxc4 b4 1 3 .VIic2 ltl bd7 1 4.id3, and now
1 4 . . . e5! leads to favourable complications. a b c d e f g h
For instance, 1 5 .dxe5? ixf3 1 6.exf6 V!ffxf6 6 . . b5 7.a4
.

1 7.0-0 V!ff g 5 1 8.g3 ltlc5 is already losing for 7.e4 h6 8.ixf6 V!ffxf6 9.a4, as played in
White. Esedov - P. Horvath, Chalkidiki 2000, can be
1 o . . . cxb5 1 1 .b3 strongly met by:
260 4.'%Vc2

C2) 6.e4

Once again, the position can be compared


with another opening. 1 .d4 d5 2.c4 e6 3.tLlf3
tLl f6 4.tLlc3 dxc4 5.e4 i.b4 is the well-respected
Vienna Variation, when White's main tries are
6.ig5 and the sacrificial 6.ixc4!?. Instead
of those moves, 6.Wc2 would be a slow and
relatively harmless move, which transposes to
our line - so once again Black should be doing
a b c d e f g h well.
9 . . . lLl c6!N 1 0.e5 '1Wf4 1 1 .'1Wd 1 i.a6+

7 ... c6 8.g3
White also failed to get anything concrete
after 8.axb5 cxb5 9.i.xf6 gxf6 1 0.We4 Wd5+
in Postl - Lerner, Graz 1 995.

8 ...i.b7 9.i.g2 bd7 10.0-0


1 0.tLle5, as played in Flores Alvarez -
Schwartzman, Mar del Plata 1 936, should be
met by 1 0 . . . '1Wc8 1 1 .0-0 0-0+.

8 a b c d e f g h
7 6 ... b5 7.a4
6 Once again, ifWhite is to cause problems, he
should play this move before the bishop arrives
5 on b7. Now Black has a couple of options, but
4 my preference is:
3
7 ...i.b7!
2 I like this counterattacking concept. Black is
1 ready to give up his extra pawn, but it forces
White to lose control over the d5-square.
a b c d e f g h 7 . . . c6 8.ie2 ib7 9.0-0 a6, as played in
10 ...ti'b6 l l .e4 a6 lots of games, leaves White with long-lasting
A complicated middlegame was reached in compensation due to the strong pawn centre.
Mamedyarov - Carlsen, Shamkir 20 1 4, but Still, Black's position is by no means worse
White did not really have enough for the pawn. here, so the choice is one of personal taste.

8.axb5!
The most dynamic move, giving up a central
pawn in order to open up the queenside.
Chapter 1 7 - Various 5 th Moves 26 1

The alternative is: 8 ... .he4 9.ti'a4


8.e5 i.e4 9.'1Wd2 tLld5 1 0.axb5 tLl b6 1 1 .ie2 This original position was reached in Rajlich
- Berczes, Budapest 2006. In my opinion, the
best way to protect the dark-squared bishop is:

8
7
6
5
4
a b c d e f g h 3
1 l . . .c6! 2
Dynamic factors are the most important
1
here. The b5-pawn was a potential weakness,
but securing the c6-square for the knight is a b c d e f g h
crucial. 9 ...ti'e7N IO.i.e2
1 2.'1Wf4 1 0.ixc4 i.xf3 1 1 .gxf3 lLl bd7+ leaves White
In the event of 1 2.0-0 i.g6 1 3.bxc6 tLlxc6 with no compensation for the damaged pawn
1 4.'\Wf4 0-0 Black's pieces are perfectly structure.
developed.
1 2 . . . ig6 1 3 .id2?! IO ... bd7 1 1 .0-0 b6 12.ti'dl i.b7
A dubious pawn sacrifice. 13 ..tg5 ti'd6
Better was 1 3 .bxc6 lLlxc6 1 4.0-0, Black has comfortable play due to his
transposing to the note above. excellent piece development and pressure on
the isolated pawn.

C3) 6.a3

8
7
6
5
a b c d e f g h 4
1 3 . . . ixc3 1 4.bxc3 tLld5 1 5 .Wg3 cxb5 1 6.h4 3
h5+
2
Black had an extra pawn and full control
over the light squares in Bergez - Duhayon, 1
Charleroi 2003. a b c d e f g h
262 4.VNc2

6 ....txc3t 7.bxc3 However, the lack of development prevents


7.Wxc3 has no independent value, since him from fighting for an advantage.
7 . . . 0-0 would transpose to Chapter 1 9.
Moreover, 7 . . . b5!? 8.a4 c6 might even force 10.bc4
White to fight for equality. The sacrifice of a central pawn is a necessary
measure.
8
1 0.Ae3 ?! occurred in Hilverda - Borovikov,
7 Paderborn 20 1 3 , when Black should have
6 protected the pawn with:

5
4
3
2
1
a b c d e f g h
7 ... c5 8.e4
a b c d e f g h
The more modest 8.e3, as played in
I. Sokolov - Wells, London 2008, is well met
by 8 . . . cxd4!N 9.cxd4 b5. White can regain the 1 0 . . . tLl a5!N 1 l .e5 ( l l .ixc4 lLlxc4 1 2.Wfxc4
lost pawn with 1 O.a4, but this entails a serious tLlxe4+) 1 1 . . .tLld5 1 2.ixc4 tLlxc4 1 3 .Wfxc4
strategic concession: id?+ Black dominates the light squares.

10 ... xd4 l l .VNc3


White's bishops are very powerful, so Black
should be careful.

8
7
6
5
a b c d e f g h 4
1 0 . . . b4 1 l .i.xc4 0-0 1 2.0-0 ib7 The 3
strong passed pawn offers Black an excellent
position. 2
1
8 ... cxd4 9.cxd4 c6
a b c d e f g h
White's position looks promising due to his
powerful bishops and mobile pawn centre. l l ... c6!
Chapter 1 7 - Various 5th Moves 263

l l . . .l2Jxf3t?! 1 2.gx:f3 0-0 1 3.:i:l:gl offers 19.he6 :i:l:xe6 20.:i:l:xe6 fxe6 2 1 .xe5 d4!
White a promising initiative. 22.ti'd2 :i:l:d8
Black is by no means worse due to his active
12.0-0 0-0 13 ..tb2 piece play.
The pressure along the a l -h8 diagonal is quite
unpleasant, but Black can neutralize it with: Conclusion

13 ...ti'a5! 14.ti'e3 e5 1 5 .h3 This chapter has dealt with a few sidelines after
This position was reached in Vanheiden - 4.WI'c2 d5. White only has so many sensible
Meissen, email 20 1 3 . Now I recommend the moves, especially with the c4-pawn hanging.
following natural innovation: Indeed, after 5 .ig5 dxc4, followed by ... b5,
Black should be able to keep a healthy extra
pawn without compromising his position too
much. Of the lines where White gives up the
c4-pawn, his best option might be 5.l2Jf3 dxc4
6.a3 ixc3t 7.bxc3, but then the abandonment
of White's queen from the d-file makes the
centre unstable, so 7 . . . c5 offers Black excellent
play.

a b c d e f g h
1 5 h6N 16.:i:l:adl :i:l:e8 17.:i:l:d6 ti'b6!
. .

This defensive resource prevents White from


developing a powerful attack by means of:i:l:xf6.

1 8.ti'e2 .te6!
Black returns the extra pawn in order to
neutralize the activity of White's bishops.

a b c d e f g h
8
-=p..;;;,/-
7
6
5
4
lw--."'"""zz>
3
2 r-jr'uur--A"

a b c d e f g h

5 .a3
Variation Index
l.d4 tLlf6 2.c4 e6 3.tLlc3 i.b4 4.f;C/c2 d5 5.a3
5....txc3t 6.f;C/xc3 0-0
A) 7.cxd5 tLle4! 8.f;C/c2 exd5 265
Al) 9.e3 266
A2) 9.i.f4 tLlc6 10.e3 ge8 ll.tLlf3 g5! 12.i.g3 g4 13.tLle5 tLlxe5 267
A2 1) 14..ixe5 269
A22) 14.dxe5N 270
B) 7.e3 b6 272
Bl) 8.cxd5 273
B2) 8.tLl f3 i.a6 275
B2 1) 9.b3 275
B22) 9.b4 276

note to move 7 B) note to move 8 B22) after 1 4.li:lxe5?!

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2

a b c d e f g h a b c d e f g h a b c d e f g h

l l . . .e5!N I O c5!N
. . . 1 4 . . . li:lxe5!N
Chapter 1 8 - 5 .a3 265

l .d4 f6 2.c4 e6 3.c3 .tb4 4.c2 d5 5.a3 9.e3 ?


This is the first ofWhite's two major options. 9.b4 e 5 ! 1 0.i.xe5 d 4 1 l .i.xf6 Wfxf6 offers
I suggest the natural and popular continuation: Black a great initiative for the sacrificed
pawns.
5 ...1xc3t 6.xc3 0-0 9.llJf3!N llJ e4 1 0.Wfc2 Wfa5 t 1 l .i.d2 Wfxc5
6 ... llJ e4, 6 . . . dxc4 and 6 . . . c5!? all have their 1 2.e3 llJxd2 1 3 .llJxd2 is enough for White
supporters, but I like the flexibility of castling to hang on to equality.
immediately while maintaining the option of 9 . . . llJ e4 1 0.Wc2 WaSt 1 1 .e2
any of these moves. In Berkovich - Borisov, Omsk 20 1 0, Black
could have developed a crushing initiative
with:

a b c d e f g h
a b c d e f g h
In this chapter we will focus on two 1 1 .. .e5!N 1 2.cxd5
significant sidelines: A) 7.cxd5 and B) 7.e3. 1 2.i.g3 if5 is even worse for White.
1 2 . . . if5! 1 3.Wfc l exf4 1 4.dxc6 fxe3-+
7.llJf3 and 7.ig5 are the two main moves, White's king is fatally exposed.
analysed in Chapters 1 9 and 20 respectively.
A) 7.cxd5
7.i.f4
This rare move has achieved a heavy score 8
for White, but it allows Black to fight for the
7
initiative with:
7 ... c5! 8.dxc5 llJ c6 6
5
4
3
2
1
a c e g
7 ... e4!

a b c d e f g h
266 4 .Yic2

This intermediate move prevents White We will analyse AI) 9.e3 and A2) 9.-t4.
from pinning the knight and renders White's
development more problematic. 9.tLlf3 if5 1 0.Wfb3 tLl c6+ gives Black a lead in
development and good prospects on the light
8.Yic2 squares.
Definitely the best retreat.
AI) 9.e3
8.Wfb3?! misplaces the queen, as the following
game illustrates: 8 . . . exd5 9.tLlf3 c5 1 0.dxc5 This was played by Kramnik against Anand in
lLl a6 l l .e3? ( l l .Wid l lLl axc5 1 2.b4 Wf6 the stem game with 7 . . . tLl e4!. Playing this way
1 3.Wfd4 was the lesser evil, but even then is a significant concession, as White's dark
Black's lead in development forces White to squared bishop will now be locked in for a
fight for equality after 1 3 . . . tLl b3 1 4.Wxf6 tLlxf6 long time.
1 5 .l::1 b l tLlxc l 1 6.l::1 xc l id7 1 7.e3 l::1 fc8+)
9 ....t5 IO ..td3
This position was reached in Kramnik -
Anand, Nice (rapid) 2009, and a number of
subsequent games. For some reason, nobody
found the strongest continuation for Black:

a b c d e f g h

l l . . .tLl axc5 1 2.Wfd l id7 1 3.tLld4 ia4


1 4.Wf3 tLl b3 1 5 .lLlxb3 ixb3-+ Black's lead in
development decided the battle in Kiriakov -
Sjugirov, Sochi 20 1 5 .

8 ... exd5
a b c d e f g h

8 IO ... YigS!N l l .g3


l l .f3 Wh4t 1 2.<;i;>fl lLl g3t 1 3.hxg3 ixd3t
7
1 4.Wxd3 Wxh l + doesn't offer White any
6 compensation for the exchange.
5
l l .tJifl is hardly an improvement: l l . . .c5!
4 1 2.dxc5 tLld7
3
2
1
a b c d e f g h
Chapter 1 8 - 5 .a3 267

A2) 9.-t4

This is a more ambitious mode of development.


However, the f4-bishop isn't secure, so Black
gets some extra tactical resources.

8
7
6
a b c d e f g h
5
1 3.l2Jf3 (even worse is 1 3 .b4 l2Je5 1 4.l2Jf3 l2Jxf3
1 5.gxf3 ih3t 1 6.e l Wg2 l ?Jm l2Jg5-+) 4
1 3 . . . Wf6+ Black manages to regain the pawn 3
and develop a powerful initiative.
2
l l ...g6 12.3 1
Attempting a tactical solution. There is
nothing better, as Black's next move can hardly a b c d e f g h
be prevented anyway. 9 ... c6
9 . . . if5 was tried in some grandmaster
12 ... xg3 1 3.ggl hd3 14.xc7 games, but I don't like it because of: 1 0.Wfxc7
White will recapture on g3 to restore the Wxc7 l l .ixc7 :gcs 1 2.if4 l2J c6 1 3 .f3 l2J f6
material balance, but Black can establish a 1 4.g4 ig6
clear positional advantage with:

8
7
6
5
4
3
a b c d e f g h
2
1 5 .e3!N ( 1 5 .h4 l2Jxd4 1 6.:gd l l2Je6? gave
1
Black decent play in AI Sayed - Moradiabadi,
a b c d e f g h Ha Long City 2009) l S ... lD aS 1 6.h4 lD b3
14 ... .te2! 1 5,gxg3 a6+ 1 7.:gd l ic2 1 8.lDe2 White may not be able to
With full control over the light squares. claim an advantage, but he will certainly have
adequate compensation for the exchange.

IO.e3
268 4.Wc2

Statically, White's position is better due to 1 2 .Wfd 1 g5! 1 3.ig3 h5 1 4.h4 i.g4!+ and
the bishop pair and pressure along the c-file. White is in trouble.
However, there are also dynamic factors in the
position! 1 1 .0-0-0?!
This is too risky, as the king will not be safe
1o .. J:e8 on the queenside.
The more aggressive 1 0 . . . g5!? also seems 1 l . . .i.f5 1 2 .i.d3 c8!
playable. 1 1 .ig3 f5 1 2.0-0-0 This position The most consistent way to prove the
was reached in Karpov - Anand, Corsica significance of the above-mentioned factor
(rapid) 2009, and now I recommend: is to prepare a line-opening piece sacrifice.

a b c d e f g h a b c d e f g h

1 2 . . . Wfe7N 1 3 .id3 f4 1 4.f3 tLl d6 1 5 .exf4 gxf4 1 3.f3


1 6.i.f2 if5tt With a complex, double-edged The prophylactic 1 3 .tJib 1 still runs into:
battle. 1 3 . . . lLlxd4! 1 4.exd4 c5 1 5 .dxc5 xeS
1 6.Wa4 c3! 1 7.Wfd4 xd3 1 8.xd3 lLl d6!
The threat of ... tLlb5 prevents White from
develop her kingside pieces, so after 1 9.ixd6
i.xd3t 20.Wfxd3 e1 t 2 l .a2 Wfxd6 Black
gets a clear advantage.
1 3 . . . lLlxd4! 1 4.exd4 c5

a c e
1 1 .6
White doesn't have any adequate alternatives.
a b c d e f g h
1 1 .tLle2?, as tried in Kotanj ian - Zhao Xue,
Moscow 20 1 1 , should be met by 1 1 . . .if5N 1 5 .fxe4
Chapter 1 8 - 5 . a3 269

Hardly better is 1 5 .'b 1 c4 1 6.ixe4 dxe4, 14 ... c5 1 5 . .td3 .t5 16.ti'e2?!


with an enormous attack. This was Carlsen's choice, but delaying
1 5 . . . dxe4 1 6.ic4 cxd4 1 7.c;i;>b 1 ie6+ castling is a risky decision.
White was in trouble in Mkrtchian -
T. Kosintseva, Nalchik 20 1 0. 1 6.0-0 is safer, although Black is still at least
equal: 1 6 . . . c4 1 7.ixe4 ixe4 1 8.Wfd 1 (after
1 8.Wfd2 ge6 1 9.f3 gxf3 20.gxf3 gg6t 2 1 .c;i;>h 1
8
Wlh4 Black's king would feel safer than the
7 opponent's . . . )
6
5
4
3
2
1
a b c d e f g h

a b c d e f g h
l l ... g5!
Black has to play energetically in order to
make use of the development advantage. 1 8 . . . ge6 1 9.c;i;>h 1 gg6 20.f3 gxf3 2 1 .gxf3 g5
22.Wfd2 ifS 23.gg 1 Wfh5 24.Wlf2 f6 25.i.f4
12 ..tg3 g4 13.e5 xeS c;i;>f7 Black was absolutely fine in Bendana
So far A21) 14.he5 is the only move to Guerrero - Serner, corr. 20 1 1 .
have been tested, but it is also important to
examine A22) 14.dxe5N. Black is also doing well after 1 6.ixe4N ixe4
1 7.Wfd 1 Wlg5 1 8. 0-0 c4, but that would still
A2 1 ) 14 . .L:e5 have been an improvement for White over the
text move.

a b c d e f g h
a b c d e f g h
270 4.VNc2

16 ... 6 17 ..be4 .be4 1 8 ..tg3 14 ....tf5 1 5 . .td3 c5!


1 8.Wlxg4t?! tJih8 1 9.ig3 Wla5t 20.c;i;>fl cxd4 I like this counterattacking approach.
leaves White facing a strong attack.
The more modest 1 5 . . .ig6 1 6.ixe4 ixe4
1 8 ...ti'a5t 19.ti'd2 1 7.Wfe2 Wfg5 1 8.f3 gxf3 1 9.gxf3 ig6 20.if4
We have been following the top-level game Wfh4t 2 1 .Wfl-1 Wfxf2t 22.c;i;>x!-1 offers White a
Carlsen - Kramnik, Moscow 2009. Now Black slight advantage in the endgame.
should have played:
16.f3 ti'a5t 17.b4
Mter 1 7. tJifl lLlxg3 t 1 8.hxg3 ixd3 t
1 9.Wfxd3 Wla6 20.Wfxa6 bxa6 2 1 .f4 c4 Black
gets plenty of counterplay along the b-file.

8
7
6
5
4

a b c d e f g h 3
19 ...ti'a6!N 2
White's king is stuck in the centre. 1

20.3 gxf3 2 1 .gxf3 .txf3 22Jfl .th5+ a b c d e f g h


We could analyse further but it's obvious 17 ... c:x:b4 18.fxe4 dxe4 19 ..tc4
that White is under pressure, being a pawn 1 9.ie2? b3t 20.Wfd2 Wfxd2t 2 1 .c;i;>xd2
down with an exposed king. :i:l:ed8t gives Black a huge initiative for the
piece, as the following lines demonstrate:
A22) 14.dxe5N

a b c d e f g h

22.tJie l (22.c;i;>c3 :i:l:ac8t 23.tJixb3 ie6t


24.c;i;>a4 :i:l:d2-+ Despite the limited material,
a b c d e f g h
Chapter 1 8 - 5 . a3 27 1

Black has succeeded in developing a decisive


8
attack.) 22 . . . b2 23.:gb l :gac8 24.:gf1 :ge l t
25.tJif2 :gc2 26.tJig l 7
6
8

7
5
6 4
5 3
4
2
3
1
2
a b c d e

a b c d e f g 30.e6!
White has no time for 30.:gxa7?? g3 3 1 .hxg3
26 . . . :gdd2! 27.:gxf5 :ge l t 28.:gfl :gxb l
hxg3-+ when it's all over.
29.:gxb l :gxe2+ White will have to give up his
bishop for the b-pawn, leaving Black with an
30 ... fxe6 3 l ,gb5!
extra pawn and good winning chances in the
The only defence!
rook endgame.
3 I .. <It>f7
.

3 l . . .g3 would lead to the same outcome:


32.:gg5t @f7 33.hxg3 hxg3 34.:gh7t tJif6
3 5 .:gxg3 :gce l 36.:gh6t tJif5 37.:ggg6 :gxe l t
38.@2 :gn t 39.@g3 :gc3=

a b c d e f g
19 ... b3t 20.f;Yd2 b2 2 1 J:b l f;Yxd2t 22.<1t>xd2
gac8 23 .tb3 ged8t 24. <lt>e2

After a pretty forcing sequence, Black can


refuel his initiative with:
a b c d e f g h
24 ... .te6! 25.gxb2 hb3 26,gxb3 gc2t
32,:gg5 gxel t 33. <lt>xel gel t 34. <lt>d2 gxhl
27.<1t>fl Mit 28.-tel h5! 29,gxb7 h4
35.:gxg4 :gxh2=
The activity of Black's pieces forces White
Black's extra pawn is meaningless, and the
to be careful. The main threat is . . . g3 with a
endgame is a simple draw.
mating net.
272 4.'%Vc2

B) 7.e3

a b c d e f g h

I O ... c5!N l l .dxc5 llJ bd7! 1 2.llJe2 c8 White


is behind in development and his pieces are
vulnerable on the c-file.
a b c d e f g h
This move looks somewhat passive - the 8.b4
c l -bishop is blocked now. However, it allows This ambitious move has occurred in a couple
White to keep the tension in the centre, while of games, but it seems over-optimistic.
the bishop can be placed on b2 in the future. I like the following way of handling Black's
position:
7 . . b6
.
8 . . . a5!N 9.b5
Not only preparing a fianchetto, but also 9.ib2?! invites further trouble: 9 ... axb4
supporting the thematic . . . c5 advance. 1 0.axb4 llJ e4 I l .Wfc2 !ha l t 1 2.ixa l
We? 1 3 .c5 llJ c6 White is far behind in
We will examine two main options: B l ) 8.cx:d5 development and his queenside pawns are
is the principal alternative to B2) 8.6. under fire.
9.llJf3 axb4 1 0.axb4 llJ e4 l l .Wfb2 xa l
8.b3 has no independent value: after 8 ... i.a6 1 2.Wxa l ib7 1 3 .c5 llJd7 1 4.id3 Wa8 offers
9.i.b2 llJ bd7 White has nothing better than Black rich play along the a-file.
l O.llJf3, when 1 0 . . . c5 reaches a position
covered on page 275 - see 1 O.ib2 llJ bd7 in
the notes to variation B2 1 .

8.id3 dxc4 9.ixc4 ib7 1 0.f3 , as played in


M. Ernst - Berry, London 20 1 0, cannot be
recommended for White. White's development
is too slow, so Black can seize the initiative by
means of:

a b c d e f g h

9 . . . c5! 1 0.dxc5 llJ e4 l l .Wfc2 llJxc5 1 2.cxd5


ib7!
A promising pawn sacrifice
Chapter 1 8 - 5.a3 273

1 3 .dxe6 l2Jxe6 1 4.ltJf3 lD d7


With . . J:'k8 coming next, Black's lead in
development is becoming rather threatening.

Bl) 8.cxd5

a b c d e f g h

1 1 . . . e5!
Also possible is 1 1 ... ib7, but I prefer the
more aggressive concept.
1 2.lDf3 e4 1 3.l2Jd4 ib7 1 4.b3 l2J d7 1 5 .ib2
lDeS
The activity of Black's pieces fully
compensates for White's bishop pair.
a b c d e f g h
8 ... xd5!
I like this move, since it forces White to
spend another tempo moving his queen and
leaves the long diagonal open.

8 . . . exd5 has been more popular, but I think


the static character of the position should suit
White here.

9.f;Yc2
Another possible retreat is:
9.d2N
This hasn't been tested in practice yet. The a b c d e f g h
following line illustrates that the queen is 9 . .tb7!
..

placed rather awkwardly on d2: I was surprised to discover that this natural
9 . . . c5 developing move has hardly ever been seen in
I do not see a reason to delay this thematic practice.
advance.
1 0.l2Jf3 9 . . . ia6 1 0.ixa6 lDxa6 1 l .e4 l2Je7, as played
After l O.dxcS bxc5 1 1 .l2J f3 b6 1 2.ic4 ia6 in Carlsen - Kramnik, Moscow (rapid) 20 1 1 ,
1 3.e2 ixc4 1 4.xc4 l2J c6 the pressure doesn't look attractive: 1 2.c4!?N c8 1 3 .b4
along the b-file fully compensates for the c5 1 4.dxc5 bxc5 1 5 .b5 l2J b8 1 6.ib2;!; White is
weakness of the c5-pawn. better due to the powerful bishop and pressure
1 o . . . cxd4 1 1 .l2Jxd4 on the weak c5-pawn.
274 4 .VNc2

10.6
This natural developing move is the only
option to have been tested so far. I checked
three alternatives:

1 0.e4N ltl f6 1 1 .e5 ltld5 1 2.ltlf3 ltle7 1 3.id3


h6 offers Black excellent play thanks to his
control over the d5-square.

1 0.id3N h6 1 1 .ltl f3 c5 1 2.0-0 ( 1 2.dxc5 can


be met by 1 2 . . . ltl d7!, just as in some of the
lines below) 1 2 . . . ltl d7 1 3.e4 ltl e7
a b c d e f g h
IO ... cS!N l l .dxc5
The ambitious 1 1 .e4?! backfires after
1 1 . ..ltlf6 1 2.id3 c4! 1 3.xc4 ixe4+.

1 1 .id3 h6 transposes to the 1 0.id3N line in


the notes above.

l l ... d7 12.c6
The greedy 1 2.cxb6?! c8 1 3.d 1 xb6
a b c d e f g h leaves White under strong pressure - it will be
1 4.ie3 cxd4 1 5 .ixd4 c8 1 6.e2 ltlc5= not easy for him to complete development. For
White is forced to exchange the dark-squared instance: 1 4.b4 e5 1 5 .ib2 fd8 1 6.ie2
bishop, so Black gets comfortable play.

1 0.b4N ltl d7 1 1 .e4


1 1 .ll:lf3?! c5 1 2.bxc5 bxc5+ leaves White
with undeveloped pieces.
1 1 ...ltl 5f6 1 2.f3 c5 1 3.dxc5 bxc5
Black's development advantage fully
compensates for White's bishop pair. The
following line is hardly forced, but it nicely
illustrates Black's dynamic resources:
a b c d e f g h
1 4.ie3 cxb4 1 5 .axb4 a5 1 6.b5 c8 1 7.d2
ll:lxe4! 1 8.fxe4 h4t 1 9.if2
1 9.f2?! xe4t 20.ll:l f3 c3-+ 1 6 . . . ltlxb4! 1 7.axb4 ltl c5!+
1 9 . . .xe4t 20.e3 b4t 2 1 .d2 e4t=
With perpetual checks. I also checked 1 2.b4 bxc5 1 3 .b5, which
is a typical way to block the open file
We have been following Le Quang Liem - while establishing a powerful outpost on
Kaufman, Saint Louis 20 1 4. I managed to find c4. However, Black can thwart this plan
a clear improvement over Black's play: with:
Chapter 1 8 - 5.a3 275

B2 1) 9.b3 is a standard way of handling the


position, while B22) 9.b4 is more ambitious.

B2 1) 9.b3

This over-protects the c4-pawn.

a b c d e f g h

1 3 . . . c4! 1 4.'1Wxc4 1:'k8 1 5 .'1Wh4 '1Wxh4 1 6.lLlxh4


lLlc5 1 7.i.b2 lLl b3 1 8.:i:l:d l :i:l:c2 With an
obvious initiative for Black.

12 .. Jc8! 13.i.b5 e7
Black will regain the pawn while retaining a
lead in development.

B2) 8.f3 a b c d e f g h
9 ... c5!
For some reason, 9 . . . lLl bd7 has been a more
frequent choice. However, challenging the
centre at once seems a principled approach.

IO.dxc5
After 1 0.i.b2 lLlbd7 l l .:i:l:d l :i:l:c8 1 2.lLld2
'1We7 Black was already better in Mkrtchian
Xu Yuhua, Nanjing 2009.

In the event of l O.i.e2N dxc4 l l .bxc4 Black


may benefit from the decision on move 9 to
avoid committing the knight to d7: l l . . . lLl c6!
a b c d e f g h
This is the more solid continuation. White
prioritizes the development of the kingside
pieces, hoping to exploit the bishop pair in the
long run.

8 ...i.a6
8 . . . i.b7 is a decent choice too, but the text
move seems somewhat more active.

a b c d e f g h
276 4.Wfc2

1 2.ib2 ( 1 2.dxc5 may be safer, but after 13 ....ixc4 14.Yixc4 c6 15.i.b2 Yld3!=
1 2 . . . l2J e4 1 3.Wfc2 lD xc5 I still prefer Black due The activity of Black's pieces stops White
to the better pawn structure) 1 2 . . . cxd4 1 3 .exd4 from benefiting from having a strong dark
l::1 c8 1 4.E1cl l2J e4 1 5 .Wfe3 l2J d6+ Black exerts squared bishop.
strong pressure against the hanging pawns.
B22) 9.b4
8
7
6
5
4
3
2
1
a b c d e f g h
a b c d e f g h
10 ... dxc4!N
After 1 0 . . . bxc5 1 l .ib2 l2J bd7, as played in This is the most ambitious concept: White
Leyva - Lima, Dresden (ol) 2008, White could seizes space on the queenside and tries to make
have played 1 2.E1d l !N ib7 1 3.ie2 when the . . . c5 break more problematic.
Black would be doomed to passive defence.
9 ... bd7
l l . .ixc4 On this occasion I favour this natural
After 1 1 .b4 l2Jd5 1 2.Wfc2 bxc5 White drops developing move.
a pawn.
9 . . . ixc4 looks like a concession, as it supports
l l ... e4 12.Yfc2 xeS 13.0-0 White's development, so 1 0.ixc4 dxc4
1 1 .Wfxc4 Wfd5 1 2.Wfxd5 l2Jxd5 1 3 .id2 l2J d7
1 4.tJie2 c5 1 5 .dxc5 bxc5 1 6.l::1 h c l l::1 fc8 1 7.l::1 c2
gave White a small but stable advantage in
Nakamura - Giri, Tashkent 20 1 4.

If Black wishes to trade queens then 9 . . . dxc4!?


1 0.ixc4 l2Je4 1 1 .Wfc2 ixc4 1 2.Wfxc4 Wfd5
gives him an improved version of the above
line; the game Dubov - Rodshtein, Moscow
20 1 1 , is a good example of Black's chances.
This looks perfectly playable, but I will focus
on the more ambitious plan of preparing . . . c5.

a b c d e f g h lO.i.b2
Chapter 1 8 - 5.a3 277

1 0.b5N has not yet been tested but it's a the c-file is likely to cause problems with the
natural enough move. My analysis continues: queen on c3 .
l l . . .i.b7 l l .a4 ( l l .cxd5 i.xd5 1 2.a4 :i:l:e8
1 3 .i.b2 :i:l:c8 followed by . . . c5 also offers l l ... bxc5 12.b5 .tb7 13 . .te2
Black comfortable play) l l . . .c5 1 2.bxc6 i.xc6 White needs to catch up with the
1 3 .i.a3 :i:l:e8 1 4.cxd5 i.xd5= development of his kingside pieces.

1 0.i.d3 , as played in Selisek - Kunej , Ptuj The over-optimistic 1 3 .a4?! e5! 1 4.cxd5 llJxd5
2008, can be well met by: 1 O ... dxc4N l l .i.xc4 1 5 .Wfd2 e4 led White to an inferior position in
llJe4 1 2 .Wfc2 i.xc4 1 3 .Wfxc4 llJdf6 1 4.0-0 Laurent - Frayssinet, Paris 20 1 0.

8
7
6
5
4
3
2
a b c d e f g h
1
1 4 . . . Wfd5 ! The queen is perfectly placed in
the centre. 1 5 .Wfc2 (the greedy 1 5 .Wfxc7?! is a b c d e f g h
refuted by 1 5 ... :i:l:fc8 1 6.Wfe7 llJ c3!+ and Black 13 ... e5!
dominates the board) 1 5 . . . c5 1 6.dxc5 bxc5 The same idea works well here too.
1 7.i.b2 cxb4 1 8.axb4 :i:l:fc8=
14.lthe5?!
1 4.cxd5N llJxd5 1 5 .Wfc2 would be safer, but
even then Black can fight for the initiative with:

a b c d e f g h
a b c d e f g h
IO ... c5 l l .dxc5
White is virtually forced to release the 1 5 . . . c4! 1 6.i.xc4 ( 1 6.llJxe5 llJxe5 1 7.ixe5
tension in this way, otherwise the opening of :i:l:e8 also offers Black rich play for the pawn)
278 4.Wc2

1 6 . . . c8 1 7.Wfe2 Wla5t 1 8.Wfd2 Wfxd2t Conclusion


1 9.tlJxd2 tlJ 5b6 Black regains the pawn and
gets excellent play. This chapter has dealt with some rare but
significant lines after 5 .a3 .ixc3t 6.Wfxc3
We have been following the game Vera 0-0. First, 7.cxd5 should be met by 7 . . . tlJe4!
Gonzalez Quevedo - Barlov, La Laguna 2007. followed by ... exd5, preventing any .ig5 pin
Now Black should have continued: and making Black's development advantage
more significant. Since 9.e3 is too passive,
9 ..if4 seems White's best way of completing
development, but then Black gets some
extra tactical possibilities based on a timely
. . . g5 . Even though the arising positions are
somewhat unusual, it seems to me that White
faces the greater danger.

The quiet 7 .e3 avoids the problems experienced


by White in the aforementioned variations;
on the other hand, it leaves the dark-squared
bishop locked behind the pawn chain for a long
time. After 7 . . . b6 Black is ready to develop the
a b c d e f g h bishop on b7 or a6 according to circumstances,
14 ... xe5!N 1 5.Wfxe5 ges 16.Wff5 WfaSt and will choose a proper moment to launch
17.<.t>fl e4 a counterattack with . . . c5 . White may suffer
With a powerful initiative. from the vulnerable placement of the queen on
c3, especially in the main line with 8.tlJf3 .ia6
9.b4 tlJ bd7.
Variation Index
l.d4 ttlf6 2.c4 e6 3.ttlc3 i.b4 4.f;C/c2 dS 5.a3 i.xc3t 6.f;C/xc3 0-0 7.ttl f3
7 dxc4 8.f;C/xc4 b6
.

A) 9.g3 280
B) 9.i.f4 i.a6! 10.f;C/xc7 f;C/dS 281
B 1) 1 1.f;C/d6?! 281
B2) 1 1.f;C/c2 282
C) 9.i.g5 i.a6 283
C 1) 10.f;C/c2 283
C2) 10.f;C/c3 h6 286
C2 1) 1 1.i.h4 286
C22) 1 1.hf6 287
C3) 10.f;C/a4 cS 288
C3 1) 1 1.M1 288
C32) 1 1.dxc5 bxcS 289
C32 1) 12.h4!? 289
C322) 12J::M1 292
C323) 12Jc 1 294

C3 1 ) after 1 3 .ixf6 C32 1 ) after 1 5 .%hc5 C323) after 1 8 . ttl d2

a b c d e f g h a b c d e f g h

1 3 . . . d3!?N 1 5 . . . f6!N
280 4.c2

l .d4 f6 2.c4 e6 3.c3 .tb4 4.c2 d5 5.a3 IO.c2 bd7 u ..tg2 c5


bc3t 6.xc3 0-0 7.f3
This is the first of the two main options
8
we will consider from this position. White
develops flexibly, and will decide later whether 7
to place his bishop on f4 or g5 . 6
7 ... dxc4 5
This is the standard response. Opening up 4
the position may favour White's bishop pair
in the long run, but there are also dynamic 3
considerations: White is forced to recapture 2
with the queen, so Black will gain extra time
1
for development.
a b c d e f g h
8.xc4 b6 12.b4
After 1 2 . 0-0 1:'k8 1 3 .WI'd l Wl'e7+ Black had
the easier game in Fodor - Manca, Budapest
2009. It will not be easy for White to find an
adequate spot for the queen.

1 2 .. Jc8 13.a4 .tb7 14.dxc5 bxc5 1 5.0-0


We have been following the top-level
game Aronian - Topalov, Bilbao 2008. Now
I suggest the following natural improvement
over Veselin's play:

a b c d e f g h
Black prepares . . . ia6, developing with gain
of tempo. White may proced with A) 9.g3,
B) 9 ..tf4 or C) 9 . .tg5.

A) 9.g3

Putting the bishop on the long diagonal doesn't


bother Black here, since the problem of the
c8-bishop has been solved.
a b c d e f g h
9 ... .ta6 1 5 ....te4N 16 . .tb2 c4 17Jacl
There is nothing wrong with 9 . . . ib7, as has The greedy 1 7.ixf6 lLlxf6 1 8 .WI'xa7 c3
also been tried by some top players, but the 1 9.:i:l:fd l can be met by:
text move seems more active.
Chapter 1 9 - 7. tLl f3 28 1

IO.ti'xc7
Toothless is: 1 0.Wfc2 lLl bd7 1 1 .E1d 1 (the
over-ambitious 1 1 .e4?! xfl 1 2. xfl c5
1 3.d6 E1e8 1 4.e5 tLld5 led White to an inferior
position in Morozevich - Kramnik, Moscow
2009) 1 1 . . .Wfc8 1 2.g3 c5 1 3.g2 b7=
M. Gurevich - Adams, Khanty-Mansiysk
2007.

IO ...ti'd5
a b c d e f g h With the c-file open, White's king feels less
1 9 . . . c2! 20.l::1xd8 l::1 fxd8 2 1 .tLld4 c l =Wit comfortable in the centre.
22.E1xc l l::1 xc l t 23.fl h6 Black's position is
by no means worse. We will analyse Bl) l l .ti'd6?! and B2) l l .Wfc2.

17 .. Jc7! B I ) n .ti'd6?!
Covering the seventh rank and freeing a path
for the queen.

I S.!Udl ti'cS+!
Black has reached a harmonious set-up, and
White will have to keep a watchful eye on the
passed c-pawn.

B) 9 ..tf4

8
7
6
5
4
3
l l . ..ti'f5! 12.g3
2 No better is:
1 1 2.g4 Wc2 1 3 .l::1 c l Wfxb2 1 4.Wfb4
1 4.g5 tLld5 1 5 .d2 lLlc6 1 6.e3 xfl
a b c d e f g h 1 7.xfl tLl a5+ was unpleasant for White
9 ... .ta6! in Ki. Georgiev - Illescas Cordoba, Andorra
A typical concept: fast development is more 20 1 2.
important than the c7-pawn! 1 4 . . . Wfa2
282 4.c2

13 ... be2!N 14.'it>xe2 ti'b5t 1 5 .'it>el ti'xb2


16Jldl e4 17.gd2 ti'cl t 18.E1dl
Black has a draw in his pocket, but he can
aim for more by means of:

a b c d e f g h

1 5 .d2N
After 1 5 ..id6? l::1 d 8 the threat of . . . l2J d5
forced White into a humiliating bishop
retreat in Eames - G. Buckley, Brentwood
2008. The text move is better, but Black still
has fine prospects after:
1 5 . . . xa3 1 6 . .ig2 a b c d e f g h
The best White can do is to castle and hope 1 8 ... ti'c2! 19Jd2 xd2 20.xd2 gadS
for some long-term compensation for the 2 I .ti'b4 f6i
pawn. Nevertheless, after: Ordinarily two bishops would outgun a
1 6 . . . .ib7 1 7.0-0 .ie4 1 8.lDe5 .ixg2 1 9.'it>xg2 rook and two pawns, but Black's superior
l2J bd7 20.l2Jc6 l::1 fe8 coordination, king safety and pawn structure
I still prefer Black; he keeps an extra pawn, swing the balance in his favour.
and the queenside passed pawns could be huge
in a future endgame. B2) l l .ti'c2 gcs 12.ti'b l

12 ... bd7 13.i.g2 1 2.d l l2J c6 1 3 .e3 .ixfl 1 4.'it>xfl lDa5 gave
We have been following the game Black a great initiative in Flear - Franco
Gordievsky - Kachar, Moscow 20 1 2. In this Alonso, San Sebastian 20 1 1 .
critical position, Black overlooked a powerful
sacrificial resource:

8
7
6
5
4
3
2
1 a b c d e f g h

a b c d e f g h
Chapter 1 9 - 7. tLl f3 283

12 ... e4!N This is the most common and logical


After 1 2 . . . tLlbd7 1 3 .e4 %Vb7 Black had full continuation - White pins the opponent's
compensation in Likavsky - Bosiocic, Austria knight and starts to fight over the e4-square.
2009, but I prefer the more aggressive option.
9 ....ta6
13.g3 c6 14 ..tg2 f5 I S .ti'dl ti'bS I6Jbl Now White is at a crossroads, with
Cl) IO.ti'c2, C2) IO.ti'c3 and C3) IO.ti'a4 all
requiring attention.

Cl) IO.ti'c2

This is White's second choice in terms of


popularity, but it's the easiest move for Black
to meet.

a b c d e f g h
16 .. Jd8
Calmly shifting the rook to oppose White's
queen, while making room for the other rook
to go to c8 at some point. White has nothing
better than:

17.0-0 ti'xe2
Black is fine; he has restored material a b c d e f g h
equality, and his active pieces are at least as IO ... h6
important as White's bishop pair. There is nothing wrong with the immediate
1 0 . . . c5 l l .dxc5 bxc5, but I see no drawback in
C) 9 ..tg5 prodding the bishop first.

l l ..th4
Completely toothless is:
1 1 .i.xf6 Wxf6 1 2.e3
White's position lacks development, so
grabbing the pawn with 1 2.Wxc7? :i:l:c8 is
extremely dangerous: 1 3.We5 Wg6 1 4.g3
Wc2+ Uwahodo - Saitou, corr. 20 1 1 .
1 2 ... i.xfl 1 3 .'xfl c5 1 4.'it>e2 cxd4 1 5 .tLlxd4
lLld7
Black doesn't face any problems, and may
seize the initiative after:

a b c d e f g h
284 4.'1Wc2

1 6.l::1 ac l ?! l l . c5
. .

1 6.E1hc 1 N e5 1 7.tLlf3 e4 1 8.'1Wxe4 '1Wxb2t= Challenging the opponent's centre before


looks like White's best. he completes development is Black's main
objective in this variation. The d4-pawn is
under attack, and White can choose between
several ways of dealing with it.

8
7
6
5

a b c d e f g h
4

1 6 . . . e5! 3
16 ... '1Wg5 was only equal in lvanchuk - 2
Karjakin, Monte Carlo (rapid) 20 1 1 .
1
1 7.lLlb3
After 1 7.tLlf5 E1ac8 1 8.Wxc8 l::1 xc8 1 9.E1xc8t a b c d e f g h
@h7 20.tLlg3 tLlc5t White's king is in 12.dxc5
danger. The over-optimistic 1 2.e4?! i.xf1 1 3.E1xf1 ,
1 7 . . . Wg5 1 8.l::1 h d 1 as seen in Morozevich - Topalov, Nice (rapid)
In Eichner - Moreira, email 20 1 4, Black 2009, is strongly met by: 1 3 . . . tLl c6!N
could have safely played:

a b c d e f g h
a b c d e f g h
1 4.dxc5 ( 1 4.e5? lLlxd4 1 5 .lLlxd4 '1Wxd4-+
1 8 . . . '1Wxg2!N 1 9.Wf5 is one line that illustrates the benefit of
1 9.E1xd7?? '1Wg4t is the simple tactical point. including . . . h6 and ih4) 14 . . . bxc5 1 5 .'1Wxc5
1 9 . . . l::1 ad8! E1c8 The lack of development puts White in a
Black wins a pawn for insufficient critical situation: 1 6.l::1 d 1 lLlxe4 1 7 .E1xd8 lLlxc5
compensation, as 20.E1xd7? still loses the 1 8.l::1 xf8t @xf8+
exchange after 20 . . . g6.
Chapter 1 9 - 7. lLl f3 285

I also considered: 1 3.Wfxc5 lLl bd7 1 4.Wfc6 lLl b6 1 5 .E1d l E1c8


1 2.E1d l cxd4 1 3 .e4N 1 6.E1xd8 l::1 xc6 1 7.E1xf8t 'it>xf8+ Despite the
After 1 3 .lLlxd4 ll:l bd7 1 4.e3 l::1 c 8 1 5 .Wlb l simplifications, White was under strong
ixfl 1 6. 'it>xf1 WI c7+ White was suffering pressure in Zhu Chen - Pelletier, Cap d'Agde
from unconnected rooks in Socko - Azarov, 20 1 0.
Warsaw (rapid) 20 1 3 .
1 3 . . . ixfl 1 4.E1xfl After 1 3.E1d l %Va5t 1 4.Wfd2 lLlc6 1 5.ixf6 gxf6
1 6.Wlxa5 lLlxa5 Black's development advantage
matters more than his imperfect pawn structure:

a b c d e f g h

1 4 . . . Wfc8!
a b c d e f g h
1 4 . . . g5 ?! 1 5 .lLlxg5 ! hxg5 1 6.ixg5 lLl bd7
1 7.E1xd4 Wfc8 1 8.Wfd2 offers White 1 7.e3 ixf1 1 8 .l::1xf1 l::1 a b8+ Laznicka - Hracek,
promising compensation for the piece. Ostrava 2009.
1 5 .Wfxc8 E1xc8 1 6.ixf6 gxf6 1 7.lLlxd4 lLl c6=
13 ...J.xfl 14Jxfl
12 . bxc5
.. We have been following the game Ma
Zhonghan - Motylev, Ningbo 20 1 1 . Black
should have continued:

8
7
6
5
4
3
2
a b c d e f g h
1
13.e3
Black's development advantage tells also in a b c d e f g h
the following examples: 14 ... bd7!N
With the following idea:
286 4.Y!fc2

1 5.0-0-0 Y!fc7 16.bl gacs; 1 2.tLle5 tLlxe5! 1 3.dxe5 tLl e4 gives Black fine
The subsequent advance of the c-pawn will prospects after: 1 4.ixd8 tLlxc3 1 5 .i.xc7
put pressure on White's monarch.

C2) IO.Y!fc3

This comparatively rare retreat received some


attention after being successfully employed
by Hikaru Nakamura against one of the great
experts in this variation, Vladimir Kramnik.

10 ... h6
We will analyse C2 1) l l .i.h4 followed by
a b c d e f g h
the more popular C22) l l ..tx6.
1 5 . . . l::1 fc8! 1 6.id6 lLl e4 1 7.f3 tLlxd6 1 8.exd6
C2 1) l l .i.h4 E1c2 The activity of Black's rook forces White
to fight for equality.
8 12 .. ,gc8 13.gdl
7 Harmless is: 1 3 .i.xf6 lLlxf6 1 4.e3 i.xfl
1 5 .'it>xfl Wfd5 1 6.Wfxd5 exd5 l ?.c;i;>e2 c5=
6
5 13 ... b8 14.Y!Ia4
After 1 4.Wfc l c5 1 5 .dxc5 Wfe7 Black's lead in
4 development is becoming threatening.
3
2 8
1 7
a b c d e f g h 6
l l ... bd7N 5
A natural novelty. 4

The previously played l l . . .g5 1 2.i.g3 tLl e4, 3


from Maiorov - Lekic, Anogia 20 1 3, seems 2
a bit risky in view of 1 3.Wfe3!N Wfd5 1 4.h4
1
lLlxg3 1 5 . fxg3 g4 1 6.tLle5 tLl d7 1 7.E1d l , when
the king on g8 is becoming rather exposed. a b c d e f g h
14 ...Y!fd7
12.Y!fc6 A typical idea to break free from the
A somewhat artificial attempt to prevent the unpleasant pin .
. . . c5 break. However, it looks like the best way
of handling the position! 1 5.Y!fxd7 bxd7 16.e3 Lfl. I7.gxf1 c5=
Black has no problems at all.
Chapter 1 9 - 7. lLl f3 287

C22) l l ..b6 ti'xf6 This position was reached in Le Quang Liem


- Aronian, Tromso (ol) 20 1 4. I've managed to
find an interesting new way of handling the
8
position:
7
6 13 .. ,gad8!?N 14.c6
1 4.Wfxc7?! lDxe5 1 5 .Wfxe5 (the greedy
5 1 5 .dxe5 ? WigS 1 6.f4 Wlg6+ puts White
4 in a difficult situation - the extra pawn
has little value, as White's king is stuck in
3 the centre) 1 5 . . . ib7 1 6.Wfxf6 gxf6 1 7.f3
2 E1xd4 reaches a simplified position where
Black is in no danger at all, and it is White
1
who has to be a bit careful to maintain the
a b c d e f g h balance.
12.g3
This was Nakamura's choice. 14.. J:Me8 1 5.e4
1 5 .f4?! is an ambitious attempt by White
1 2.Wfxc7?! has already been covered under the to maintain his space advantage, but after
1 1 .ixf6 line in the notes to variation C 1 . 1 5 . . . ib7 1 6.ig2 l2J b8 1 7.0-0-0 E1d8! he is
unable to stabilize his position.
Also toothless is 1 2.e3 ixfl 1 3 Jhfl E1c8
1 4.'it>e2 c5= as seen in Loeschnauer - Flitsch, 1 5 ... Lfl. I6,gxfl b8!
corr. 20 1 0.

12 ... d7!
This improves over 1 2 . . . ib7, when 1 3.ig2
lD a6 1 4.0-0 c5 1 5 .E1ac l was better for White
in the aforementioned game Nakamura -
Kramnik, Antalya 20 1 3.

a b c d e f g h
17.xb8
1 7.lDxa7?! Wfd8 only invites trouble, as
White's knight is almost trapped.

1 7.lDe5 c5 1 8.f4 cxd4 1 9.Wfxd4 l::1 c 8= is fine


for Black.

a c e f g h
288 4.f;Yc2

17 .. Jxb8 1 8.0-0-0 !Ud8 19.f4 gd7 20,gd2 White was fighting for equality in Carow -
gbd8= Y. Vovk, Berlin 20 1 5 .
Black has strong pressure on the d4-pawn,
and White's space advantage does not matter 1 2.Wfb3 , as played in Bareev - Bu Xiangzhi,
much with no minor pieces left on the board. Moscow 20 1 0, can be met well by: 1 2 . . . tLl e4!N
1 3.dxc5 Wb7 1 4.cxb6 tLlxg5 1 5 .tLlxg5
C3) IO.f;Ya4

8
7
6
5
4
3
2 a b c d e f g h

1 1 5 . . . tLl d7 1 6.e3 i.xfl 1 7.'it>xfl gfd8 The lack


of harmony in White's camp offers Black, at
a b c d e f g h the very least, full compensation for the pawn.
This is the most common choice by far. The
queen is placed quite actively on a4, keeping 12 ... cxd4 13.Lf6
an eye on the a6-bishop and aiming to transfer After 1 3 .gxd4 Wc6 1 4.Wfxc6 tLlxc6 1 5 .ixf6
to the kingside via the fourth rank. The main gxf6 only White may experience problems
drawback of this move is that the b2-pawn is - there are still a few more moves needed to
now unprotected. complete his development.

10 ... c5 This position has occurred in several high-level


White's two main options are C3 1) u .gdl games, such as Khenkin - Bacrot, Geneva 20 1 0.
and C32) l l .dxc5. I've managed to find an interesting new idea:
Black was absolutely fine after 1 1 .e3 ixfl
1 2.'xfl lLl bd7 1 3 .'e2 cxd4 1 4.Wxd4 WeB in
8
Rychagov - Sargissian, Rethymnon 20 1 0.
7
C3 1) u .gdl f;Yd7 12.f;Yc2
6
The most ambitious try - the queen on d7 is 5
placed somewhat awkwardly, so White doesn't
4
mind wasting another tempo in order to avoid
the exchange and slow down the development 3
of Black's queenside pieces. 2
Harmless is 1 2.Wxd7 lLlbxd7 1 3 .e3 i.xfl 1
1 4Jl:xfl tLl e4 1 5 .dxc5 lLl dxc5 1 6.i.e7 gfc8 and a b c d e f g h
Chapter 1 9 - 7. lLl f3 289

13 ... d3!?N
Since the d4-pawn will fall anyway, it makes
sense to give it up in a better situation!

14.f;Yc3
1 4.exd3 gxf6 1 5 .i.e2 g7 1 6.0-0 :gc8
1 7.WI'd2 tLl c6 is at least equal for Black.

14 .. ,gc8 1 5 .f;Yd4 gxf6 16.WI'xf6

a b c d e f g h
25 .. ,gxdl t 26.xdl ga2 27.a4 f6 28. d2
d5
Black has full compensation for the pawn.

C32) l l .dxc5 bxc5

8
7
a b c d e f g h
6
16 ...f;Yd5! 17.exd3 L ; - .. . . F""'' u - .

1 7.h4 Wl'a5t 1 8.tLld2 tLl d7 1 9.Wfg5t Wxg5 5


20.hxg5 :gc2+ 4

17 ... d7 1 8.f;Yf4 3
1 8.WI'g5 t Wxg5 1 9.lLlxg5 :gc2 20.:gd2 :gac8 2
also holds no danger for Black.
1
18 .. ,gc2 19.f;Ya4 f;Ya5t 20.f;Yxa5 bxa5 a b c d e f g h
Black has at least enough compensation for Now C32 1) 12.h4!? is a tricky idea, but
the sacrificed pawn. the two main options are C322) 12,gdl and
C323) 12,gcl .
2 1 .gd2 gac8!?
If a draw is an acceptable result, then C32 1) 12.h4!?
2 1 . . .:gc l t 22.:gd l :gc2= leaves White with no
other option than to take it. This tricky move has only been played in a
few over-the-board encounters, but it has been
22.d4 gel t 23,gdl .ixfi 24,gxfl g8c2 more popular among correspondence players,
25.b3 which is quite telling. In any case, it requires
careful handling, and I recommend deviating
from most of the games by playing:
290 4.Wc2

12 ...f;Yd7! 1 7.e4 ( 1 7.f3 h6 1 8 . .if4 e5 1 9 . .ig3 l2Je3?)


This almost-new move stays true to our 1 7 . . . .ixfl 1 8.:gxf1 h6 1 9 . .if4 e5 20 . .ig3
general strategy in this variation: Black is ready l2Jgf6 2 1 .:gc3 l2Jh5 22 . .ih2 l2J f4 23 . .ixf4 exf4
to give up material in order to maximize his 24.c;i;>e2
lead in development.

1 2 . . . Wb6?! has been Black's usual response, but


then 1 3 . .ixf6 gxf6 1 4.:gb l ! shows why White
was happy to postpone moving his rook. He
has saved a tempo for protecting the b2-pawn,
and after 1 4 . . . c4 1 5 .l2J d2 .ib5 1 6.Wc2 l2J c6
1 7.Wc3 Black's king was under strong pressure
in Krysa - Tristan, Resistance/Saenz Pena
20 1 3 .
a b c d e f g h

24 . . . l2Jf8! 25 .:gfc l l2J e6= Black is fine, as 26.b4


:gd8! 27.bxc5 :gb2 may prove dangerous only
for White.

13 ... g4!
Exploiting the main drawback of 1 2.h4
- now it's not easy for White to push the
annoying knight away.

14,gdl
This brought White a victory in a
a b c d e f g h correspondence game. I considered two other
13.f;Yc2 moves:
Mter 1 3 .Wfxd7 l2J bxd7 1 4.l2Jd2 :gabS 1 5 .b3
:gfc8 1 6.:gc l l2Jg4! Black's lead in development 1 4.Wxc5N can be met by: 1 4 .. .f6 1 5 . .id2
offers good counterplay, for instance: ( 1 5 .:gd 1 Wb7 1 6.-ic l l2J c6 1 7. l2J d4 :gac8 is
unclear)

a b c d e f g h
Chapter 1 9 - 7 . ttl f3 29 1

1 5 . . . ttl c6 1 6.c2 tt:l ce5 1 7.i.c3 1"lac8 1 8.1"ld 1 full compensation for the pawn in Holroyd -
b7 White has parried the immediate threats, Sadzikowski, carr. 20 1 4 .
but Black maintains long-lasting compensation
due to his development advantage and pressure I briefly considered 1 5 . . . lt:l c6 but was not
along the a6-fl diagonal. satisfied with Black's chances after 1 6.lt:ld4!.

1 4.g3N b5 1 5 .i.g2 ( 1 5 .i.h3 f5 1 6.0-0 lt:l c6 16.i.c1


1 7.e4 1"lac8 offers Black excellent piece play) White might also try:
1 5 . . . lt:lc6 1 6.0-0 1"lac8 1 7.1"lfe 1 1 6.d4!? h5
The inclusion of these moves takes the play
in a different direction, but not in a bad way
for Black.
1 7.i.c l lt:l c6 1 8.d7 b6 1 9.xe6t
1 9.e3? tt:l ce5 20.tt:lxe5 1"lad8! 2 1 .i.xa6 fxe5
22.xd8 1"lxd8 23 .1"lxd8t xd8+ Material is
approximately even, but White is very much
on the defensive.
1 9 . . . mhs 2o.i.d
20.e3? is nicely refuted by: 20 . . . 1"lae8
2 1 .d5 lt:lxe3! 22.fxe3 1"lxe3t 23.md2 1"ld8
a b c d e f g h
24.xd8t lt:lxd8 25.i.xa6 1"lxf3 26.gxf3
1 7 . . .f5 ( 1 7 . . . e5!? 1 8.1"lac l h5 could also be xa6+ and White's king is still in trouble.
considered) 1 8.1"lac l h6 1 9 .i.d2 1"lfd8 The 20 . . . lt:lxe3 2 1 .xe3 xb2
activity of Black's pieces fully compensates for The computer calls it equal, but practically
the inferior pawn structure. the position is more dangerous for White, as
Black has a huge lead in development in return
14 ...b7 1 5.\':VxcS for a mere pawn.

a b c d e f g h a b c d e f g h
I S ... f6!N 16 ... ttlc6 17.ttld4
Mter 1 5 . . . 1"lc8 1 6.xc8t! xc8 1 7.1"ld8t Black's enormous lead in development forces
xd8 1 8 .i.xd8;!; Black was unable to find White to look for simplifications.
292 4.ti'c2

17 xd4!?
. C322) 12.dl ti'b6
Another good option is 17 . . . l::1 ac8 1 8.lLlxc6
E1xc6 1 9.Wfb4 Wfxb4t 20.axb4 E1c2, when best This is the right square for the queen; Black
play continues: should aim for counterplay rather than worry
too much about the ensuing damage to his
kingside structure.

a b c d e f g h

2 1 .l::1 h 3! l::1 fc8 22.l::1 c3 l::1 8xc3 23.bxc3 lLl h2!=


White's extra pawn is meaningless.

18.ti'xd4 fd8 19.ti'xg4 a b c d e f g h


1 9.Wfxd8t? l::1 xd8 20.E1xd8t c;i;>f7 2 1 .E1d l 13.Lf6 gxf6 14.e3!?
Wfb6 22.e3 xfl 23.E1xfl WI c7 reaches a White has also tried: 1 4.E1d2 ( 1 4.Wlg4t c;i;>h8
position where Black's initiative outweighs 1 5 .l::1 d2 tLl c6 leads to the same thing) 14 ... tLl c6
White's small material advantage - the queen 1 5 .Wlg4t h8
and knight make a perfect team!

a b c d e f g h

This position has arisen in more than 60


games, all of which ended in a draw! Indeed,
the lack of development leaves White with no
other choice than to repeat moves: 1 6.Wfh4
a b c d e f g h c;i;>g? 1 7.Wlg4t c;i;>h8 1 8.Wfh4=
22 .td2 gxd2 23.c;i;>xd2 ti'xb2t 24.c;i;>dl

ti'b lt= 14 ....bfl l5.<1t>xfl ti'xb2!


With perpetual checks.
Chapter 1 9 - 7 . tLl f3 293

1 5 . . . tLl c6 has been the usual choice, but I see


no reason to deviate from the more principled
option of grabbing the pawn.

a b c d e f g h
19 ... e7! 20.gbl
20.Wfxh7? backfires after 20 ... tLl f5 2 1 .1':1g8t
c;i;>e? 22.:i:l:xc8 l':ixc8 23.e4 Wfc2 24.1':1e l lLld4+
a b c d e f g h
when White's queen is completely out of play.
16.h4!?N
This seems to me to be the most natural 20 ... 5!?
attempt to develop the initiative. 20 ... Wfxa3 2 l .Wfxh7 Wla6t 22.tJigl <t f5
23.e4 tLlxg3 24.fxg3 also offers White no more
The quiet 1 6.g3 Wfb7 1 7.tJig2 :i:l:c8 offers no than equality.
advantage, for instance:
2 1 .gx:b2 fxe4 22.g5 d5

8
7
6
5
4
3
a b c d e f g h
2
1 8.1':1d6N ( 1 8.1':1d3 tLl c6= Pavlicek - Crapulli,
Internet 20 1 4) 1 8 . . . lLl c6 1 9.Wlg4t c;i;>fg 20.Wff4
1
tJig7= a b c d e f g h
23Jib7 gc7 24Jix:c7 xc7 25.xe4 c4
16 .. ,gc8 17.gh3 c6 1 8.gg3t t8 19.'1We4
26.c3 e8 27.gg5 5=
It looks like White has succeeded in bringing The strong passed c4-pawn should secure
all his pieces into the attack, but Black's Black an easy draw.
defensive resources prove fully adequate after:
294 4.V;Vc2

C323) 12Jc1 White has also tried:


1 8.b4 cxb4 1 9.axb4
White has swapped off the weak c-pawn in
an attempt to create concrete problems, but
Black is fine after:
1 9 . . . Wd6!
But not 1 9 . . . b7?? 20Jk6+- and the queen
is trapped.

a b c d e f g h
The main idea behind this move is not to
attack the isolated c5-pawn, but to prepare a
safe way of protecting the b2-pawn.
a b c d e f g h
12 ... YlYb6 I3 ..bf6 gxf6 I4J:c2 gds
Since the rook is not on d 1 , it makes sense to 20JM2
take control of the only open file. 20.1:'k3 llJ c4 2 Ud 1 Wxd 1 t 22.Wxd 1 gxd 1
23.c;i;>xd 1 llJ d6 24.llJd4 gb8= also leads
15.e3 Lfl. I6,gxfl c6 17.e2 nowhere for White.
Even though White's king is in the centre, 20 . . . Wc6! 2 l .Wfxc6 llJxc6 22.b5 llJ e7 23.gxd8t
Black still needs to take care to avoid falling gxd8 24.ga 1
Now in Oreev - Bacrot, Mulhouse 20 1 1
into a passive position where he is stuck
defending the c5-pawn. Black's most accurate continuation woul d
have been:

a b c d e f g h

a b c d e f g h 24 . . . gb8!N
24 . . . llJ c8!?N, followed by . . . gd5 and . . . llJ d6,
17 ... a5 18.d2 achieves the same purpose.
Chapter 1 9 - 7 . lLl f3 29 5

25Jha7
25.lLld4 lLl c8 26J:!:a5 ltl d6= does not change
anything; Black can follow up with . . J!b7
and . . . e5 if needed.
25 . . . ltld5=
Liquidating the last queenside pawn.

8
7
a b c d e f g h
6
2 1 . . .h6! Covering the g5-square, so the king
5
will feel much safer. 22.h3 l':iad8 23. fl Wfb8=
4
20 ... cx:b4 2 1 .axb4 b7 22.k7
3
22.1':1c8t E!:xc8 23.1':1xc8t g7 24.Wfa l + e5=
2 is no problem for Black.
1
22 ...f;Yb5t 23.f;Yx:b5 E!:xb5 24.E!:d7 gxb4
a b c d e f g h 25,gcc7
1 8 .. Jld5!N The last try - White manages to get full
This seems like the perfect place for the control over the seventh rank, but it doesn't
rook, as it covers both the c5-pawn and the bother Black:
kingside while vacating the dB-square for the
other rook.

After 1 8 . . . E!:ab8 1 9.E!:fc l E!:xd2t 20.E!:xd2


ltl b3 2 1 .E!:cd l lLlxd2 22.E!:xd2 White kept
a small edge in lvanchuk - Duda, Tromso
20 1 3 , although Black remains well within the
drawing margin and he held it without too
many problems.

19J:Ucl f5
Restricting the mobility of White's knight.

20.b4!? a b c d e f g h
White has to try this if he is to make any 25 ... d8 26.lLlc4 a5 27.x:a5 gx:a5
headway. 28Jxd8t g7 29Jdd7 ga2t 30. f3 gbb2
3 1 .gxf7 t g6=
Exchanging the knights cannot bother Black Black is obviously out of danger.
here: 20.ltl c4 ll:lxc4 2 1 .Wfxc4
296 4.c2

Conclusion

7.lL!f3 dxc4 8 .xc4 b6 is undoubtedly one


of the most topical and complex variations
in the Classical System. As usual, Black is
aiming to attack the opponent's centre as
soon as possible, making use of his lead in
development. The line which best illustrates
this advantage is 9.if4 .ia6! l O.YMxc7 YMd5 ,
when l l .YMc2 may be White's only way to
avoid being worse. 9.ig5 makes Black's task
tougher, but 9 . . ..ia6 followed by a quick . . . c5
offers Black good counterplay. There are a
few critical lines where White has the bishop
pair and/or a better pawn structure, but Black
always gets enough play with his active pieces
and pressure against the b2-pawn.
a b c d e f g h

7.ig5
Variation Index
l.d4 tLlf6 2.c4 e6 3.tLlc3 i.b4 4Jfc2 d5 5.a3 hc3t 6.'i'xc3 0-0 7.i.g5
7...h6
A) 8.i.h4 c5! 9.dxc5 d4 298
Al) 10.'i'g3 299
A2) 10.'i'c2 299
B) 8.hf6 'i'xf6 302
Bl) 9.tLl f3 302
B2) 9.cxd5 exd5 10.e3 i.5 305
B2 1) ll.tLlf3 305
B22) ll.tLle2 tLld7 306
B22 1) 12.tLlg3 306
B222) 12.tLlf4 308

Al) after 1 4 . tt:\ f3 B l ) note t o l l .c3! ? 8222) after 1 3 .e2

8 8 8
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
2 2 2
I
g h g h g h

1 4 d3!N
. . . 1 2 e5!N
. . . 1 3 a5!?N
. . .
298 4.V;Vc2

l .d4 f6 2.c4 e6 3.c3 i.b4 4.Vc2 d5 5.a3


i.xc3t 6.Vxc3 0-0 7.i.g5
What could be more natural than pinning
the f6-knight? It is certainly an annoying motif
for Black to deal with, especially when his own
dark-squared bishop has left the board.
Interestingly, in his 2007 book Challenging
the Nimzo-!ndian, IM Vigorito only considers
6 . . . 0-0 in an extremely brief note, mentioning
that 7.ig5 ll:l bd7 8.e3 gives White a good
version of a Queen's Gambit Declined. I a b c d e f g h

would rather put the question to the bishop 1 2 . . . Ad?! With a decisive attack - a pleasing
immediately with: finish would be 1 3.Vxe4? ia4t 1 4.'it>cl Ve l
mate! Quite an amusing final position, with
7 ... h6 most of White's pieces still on their original
I was quite surprised to discover that this squares!
natural move is only Black's fourth most
popular choice according to the database. 9 ... d4
White may react with A) 8.i.h4 or B) s . .t:x:6. Black has an improved version of two better
known theoretical lines:
A) s ..th4
7 . . . c5 8.dxc5 d4 has occurred in quite a lot
of games. In our version, the insertion of the
8 moves . . . h6 and i.h4 benefits Black, for two
7 reasons: the bishop on h4 is cut off from the
centre and queenside, and Black has the extra
6
option of throwing in . . . g5 if needed.
5
4 6 . . . c5 7.dxc5 d4 is a famous gambit. The
critical reply is 8.Vg3, with counter-chances
3 on the kingside. Here the queen move is less
2 troublesome as the g7-pawn is not en prise.
1
8
a b c d e f g h
7
Now Black has a choice between several
attractive possibilities, but I prefer the 6
following aggressive concept: 5
8 ... c5! 9.dxc5 4
9.cxd5?, as played in Kadimova - Pecorelli 3
Garcia, Port of Spain 2009, can lead to fatal
consequences after: 9 . . . g5!N l O.i.g3 ll:l e4 2
l l .Vc2 VaSt 1 2.'d l 1
a b c d e f g h
Chapter 20 - 7 .ig5 299

White is at a crossroads, with AI) IO.Yig3 I prefer Black in this wild-looking position:
and A2) I O.Yfc2 being the two most logical his king is reasonably safe, while White's pawns
options. might fall in the long run. For instance:

AI) IO.Yig3 c6N I9.e4


1 9 .:ghgl Wg4 20 .Wfh6 e5 2 1 .h3 Wg7+ also
1 O llJ bd7!? led to success for Black in Schwenk
favours Black.
- Kermer, corr. 2007. This could certainly be
investigated in more detail, but I prefer the
8
more active development of the knight on c6.
7
I I .0-0-0 6
There are no adequate alternatives:
5
l l .llJ f3 ? WaSt 1 2.llJd2 llJ e4+ puts White in 4
serious trouble.
3
l l .b4?! e5 1 2.llJf3 :ge8 is also excellent for 2
Black.
1
a b c d e f g h
8
I9 .. Jf6 20.ex5 ex5 21.h4 Ylg4 22.Yixg4
7 fxg4 23.ie4 ie6;
6 Black is out of danger, and the only real
question is whether or not White will be able
5
to salvage a draw in the endgame.
4
A2) IO.Yic2
3
2
1
a b c d e f g h
l l ... g5!
The threat of 1 2 . . . llJ e4 is hard to meet, so
White's next move is forced.

I2.hg5 hxg5 13.Yixg5t ths I4.f3 h7


I 5.Yih5 Yl6
White has three pawns for the piece, but
Black should be doing well as long as he takes
care over the next few moves. a b c d e f g h
This retreat looks more modest, but the
I6.e3 dxe3 I7.id3 Ylg7 I8.fxe3 5! queen can be used for defensive tasks.
300 4.Wff c2

10 ... e5 1 2.exd4 exd4t 1 3 .i.e2 occurred in Krush -


Black's compensation is base