english

defence

by Daniel King

EVERYMAN CHESS
Published by Everyman Publishers plc, London
First publishe d 1999 by Everyman Publishers plc, formerly Cado gan Books
p lc, G louces t er M ansi o ns , 140A Shaftesbury Avenue, London WC2H 8HD

g
C opy ri ht © 1999 Daniel King

The right of D aniel King to be identified as the author of this work has been
asserted in accordance wit h t h e C opy rights, Designs and Patents Act 198 8 .

All rights reserved. No part of this publication may be reprod uced, stored in a
retrieval system or transmitted in any f o rm or by any means, electr onic ,
elect r ostatic, magn etic tape , ph ot ocopying, rec ording or otherwise, wit hout
prior permission of the publisher.

British Library Cataloguing-in-Publication Data

A catalog ue record for this book is availab le from the British Library.

ISBN 1 85744 295 4

Distributed in North America by The G lo be Peq uot Press, 6 B us iness P ark

R oad, P 0 Box 833, Old Saybrook, Connectic ut 06475-0833.
. .

T ele ph one 1-800-243 0495 {toll free)

All other sales enquiries sh o u l d be directed to Everyman Chess, G lo ucest e r

M ansi ons, 140A S h aftes bury Avenue, London WC2H 8HD
tel: 0171 5 39 7600 fax: 0171 379 4060
em ail: dan@e veryman uk. co m
.

website: www .everyman. uk. com

The Everyman Chess O pe ning Guides were designed and deve lope d by First
Rank Publishing.

EVERYMAN CHESS SERJES (fo rmerly Cadogan Chess)

Chief Advisor: Garry Kasparov
Adviso ry Panel: Andrew Kinsman an d Byron ] acobs
Typeset and edited by First Rank P ublishing, Brig h ton .

r
P o duct i on by Book Production S e rvices
Printed and bound in Great Britain by The Cromwell Press Ltd., Trowbridge, Wiltshire.
 CONTENTS I
1 c4 b6

Bibliography 4
Introduction 5

Main Line with 3 e4 .ib7 4 it.d3: The new 4 .. )tJc6

' ' 10
Main Line with 3 e4 .lri.b7 4 it.d3: Other Fourth Moves for Black 29
. Main Line with 3 e4 J.b7 4 d5, 4 �c3 �b4 5 d5 and
4 �c3 �.b4 5 �d3 48
�
Main Line with 3 e4 .!b7 4 tLlc3 �b4 5 f3 and 4 f3 68
:Main Line with 3 e4 .ib7 4 4Jc3 il.b4 5 Wc2 and 4 'i'c2 81
;;. White Plays an early a2-a3 93
The Two-Pawn Attack (1 c4 b6 2 4Jc3 i.b7 3 e4} 113
Systems with .. . .ixf3 130

Index of Complete Games 143

 BIBLIOGRAPHY I

Books
Encyclopaedia of Chess Openings volume A, second edition (Sahovski
lnformator, 1996)
The English Defence, Ke ene Plaskett and Tisdall (Batsford, 1987)
,

]on Speelman 's Best Games, Speelman (Batsford, 1997)

Nunn s Chess Openings, Nunn, Burgess, Emms and Gallagher (Everyman
Chess, 1999)

Periodicals
Informator
New in Chess Yearbook
New in Chess Magazine
ChessBase Megabase CD-ROM
Chess Monthly
B1itish Chess Magazine
 INTRODUCTION I

Whenever I am faced with an opponent 1 c4 Move Order

who opens 1 d4 or 1 c4, I always get the
feeling that I am up against a different 1 c4 b6
breed of player to someone who starts
with the e-pawn. 1 e4 players tend to want
to 'have a go' at you in an open fight,
sleeves already rolled up, quite prepared to
give as good as they get. My impression of
1 d4 and flank opening players is different.
They prefer to defer the struggle until
they have brought out their pieces and
rather like to stay on well-worn tramlines
of development, a bit like carrying a com­
fort blanket around. (I won't be taking
this cod chess psychology too far - it is
too dangerous for me: depending on the
day of the week I open 1 c4, or 1 d4, 1 This is the pure starting position of the
llJf3, 1 g3 and 1 e4. At the very least I am English Defence. Black invites his oppo­
quite confused!) nent to occupy the centre with his pawns,
That said, assuming there is a grain of so that he can then shoot them down in
truth in my view, then the English De­ flames.
fence is the perfect weapon against these 2 d4 e6
kind of players. From the outset White 2 ...i-b7 is also possible, but I want to
cannot stick to his usual moves; he actu­ show how Black can reach the English
ally has to think for himself as standard Defence via different move orders. This is
responses are often not good enough. Be­ crucial if White plays 1 d4.
fore we get into looking at some of the
typical strategies that Black employs, the 1 d4 Move Order
question must be askeq, 'What exactly is Supposing your opponent opens with.. .

the English Defence?' 1 d4

5
English Defence
... and you would still like to play the
English Defence. I would recommend that
you play...
1 . . . e6!
It is true that this gives White the op­
portunity to transpose into a French De­
fence with 2 e4, but in my experience few
players who open with 1 d4 like to do this
(though you should have something up
your sleeve just in case).
2 c4
2 li:Jf3 is also common. After the text
move Black can safely play...
2 ... b6!
and we are back to the English Defence.
If you check out the move orders in the
main games of this book you will begin to
understand the best way to get into the
opening according to your own overall
repertoire.
Not Owen's Defence
By the way, I would not recommend that
you answer ...
1 d4
...with...
1...b6
In that case White plays . .
. this is probably the best course, it is very
2 e4!
...and if...
2 . . . .il.b7 3 iLd3 e6
...White should not move the c-pawn,
but just develop a piece, for instance.
6
4 ll'lf3
Here White's centre is far more secure
than in many of the games we will be con­
sidering, where the c-pawn is already on
c4. I won't be examining this so-called
Owen's Defence.
The English Defence in Practice
Since the 19th century various English­
men had dabbled with 1...b6 from time to
time, to little effect. However, the English
Defence really only began to 'earn' its
name in the early 1970s, when Michacl
Basman and Tony Miles began experi­
menting with the fianchetto of the queen's
bishop. Miles in particular was able to
pioneer the system at the highest levels of
international chess - with great success.
Over the years he has contributed so
many remarkable ideas to the opening
that he deserves the chief credit for its
current cult status in the repertoire of
some of the more imaginative players on
the tournament circuit.
To help explain some of the main con­
cepts of the English Defence, we could do
worse than look at one of Miles's early
games with the opening. It features three
crucial concepts which have recurred in
many games.
F arago-Miles
Hastings 1976/77
1 c4 b6 2 d4 e6
Note Miles's move order. Instead of
preventing e2-e4 with 2 . ...ib7, he posi­
.
tively invites White to play it, and you
would be amazed at how many players
like to take up the challenge. Although
tricky - Grandmaster Farago doesn't
make it past move 20.
3 e4
Game on. Black's task 1s to knock
down the centre.

3 ...�b7
A good start.
4 tZlc3
White protects the e-pawn, but.. . that if White ever gets fed up and wants to
4...if..b4
...Black pins the knight and renews the
attack on the e-pawn. This is Concept
No.J. The pin is an extremely useful
weapon for Black: not only does he in­
crease the pressure on the centre, but he
also has the positional threat of doubling
the c-pawns, intending to attack them
later on, very much as in the Nimzo­
lndian Defence.
5 'i'c2
White defends the e-pawn .. .
5 ...'i'h4
...and Black attacks it again. Concept
No.2: the early development of the queen
to h4 to attack the pawn on e4 is an
Introduction
gral part of Black's system. It also creates
some nasty threats on White's kingside.
One of the bonuses of the queen move is
boot it out with g2-g3, then serious weak­
nesses have been created on the long di­
agonal - remember that bishop sitting
snugly and smugly in safety on b7.
At the time this game was played this
kind of early queen placement would have
come as quite a shock to someone like
Farago. Nowadays it is commonplace,
though no less difficult to handle.
6 1l.d3
White protects the pawn again ...
6 ... f5!
...Black attacks. Concept No.J: the ad­
vance of the f pawn. This is a key part of
-
Black's strategy in so many games. It is all
very well attacking Black's centre with
pieces, but that usually won't be enough
to do the damage. A pawn break must
come, and this fits the description per­
fectly. It is all about increasing the scope
of that bishop on b7. � once reached this
position and my opponent captured on f5,
so I took on g2 and that was a whole rook
in the bag.)
7g3�5
White relieves the pressure on his cen­
tre for a moment, but the move g2-g3 cre­
ates serious long-term weaknesses on the
ime- kingside and on the long diagonal.
7
English Defence

a �e2 'i!H7 9 f3 fxe4 1 0 fxe4lllf6 11 d5

The centre lurches forward, but this is
just desperation. It isn't going anywhere.

A few moves before I had decided to

11 ... 0-0 12 �f3 �g6 13 Ad3 'Cii'hs 14
0-0 llla6! 15 a3 �xc3 16 bxc3
It is interesting to see how Miles waited
to be pushed by a2-a3 before he captured
on c3: the knight wasn't running away
and it was possible that the bishop could
have been used elsewhere. 16 'ii'xc3 liJcS
17 e5 ctJg4 would also have been very
good for Black.
16 ..<�3c5 17 j:_e3 llJxd3 1 8 �·xd3 exd5
. healthier pawn st:ructure than his oppo­
19 exd5 lt:lxd5! 20 cxd5 �xf3 0-1
Here White resigned, clearly demoral­
ised by his disastrous start. He could have
struggled on with 21 ttxf3 1!Vxf3 22 Il.dl
c6, though clearly the situation is grim. �f
you would like a more detailed analysis of
this variation, then turn to Game 49 )
castle on the queenside. There my king is
well protected by four pawns - safer than
White's king. This is an extremely com­
mon scenario in the English Defence.
Chances at this point are about equal. Al­
though Black is a pawn down I was rea­
sonably optimistic in view of White's
shattered kingside pawns; and that brings
me on to the other theme I want to men­
tion. After the confusion of the opening
10-15 moves, Black often emerges with a
nent. Let me show you the position we
reached a few moves further on:
.

Webster-King
British Championship, Eastbourne 1990

As well as the three main attacking ideas I

mentioned in the Farago-Miles game, there
are a couple of other important themes
that you should be aware of. Often, when
Black plays ... f7 -f5 the kingside can look a After a little good fortune and a little
bit of a mess. Although this is not a criti­ persuasion, I managed to reach this end­
cal problem, as in return White's centre game. As you can see, several of those
has collapsed, it can mean that castling on loose pawns have dropped off (well, they
the kingside is inadvisable. tend to) leaving me with my mass of four

8

pawns supported and protected by my
king. The ending is technicall y winning.
Rahman-Speelman
Calcuua 1998
Here is another example of Black's su­
perior pawn structure making the differ­
ence b etwe en the two p os itions. Black has
challenged the centre in the standard way,
w i th .. .f7-f5. White should pr obab ly try to
hold the structure with 7 f3, but the ten­
sion was too great, and he cap tured:
7 exf5?! exd5!
Creating an imbalance in the structure
rather similar to the last example.
a liJgf3 li:Je7 9 g4
If White could maintain t he pawns like
this then the position would be rather
unclear. However. ..
9 . . . h5!
... wrecks the m.
10 ::g1 dxc4 11 .it.xc4 hxg4 12 J::!xg4
!!Jbc6 13 .i.d3 0-0-0
see following d;agram
Having created chaos on the kingside,
the king evacuates swiftly behind the four
pawns and is perfectly safe. Now Black s ' the openings that I play, the English De­
task is to mop up the broken pawns.
14 b4 Il:df8 15 b5 <Zld8 16 li:Jh4 !iJf7 17
f4 .'L!d6 18 0-0-0 li:Jdxf5
And Black was on the way to capturing
the whole kingside. I'll be pointing out
Introduction
more themes as we go along .
How this Book is Organised
The first five chapters of the book deal
with positions where White accepts the
challe nge and occupie s the cen tre with
pawns. Th e sixth chapter cont ains games
wh ere White p lays an early a2-a3, prev ent­
ing ..ii.b4, and there b y gives himself more
.
control over the centre. Chapters 7 and 8
include games where White attempts to
divert the game into the more familiar
patterns of the E nglis h Opening and the
Reti. We have ways of avoiding that tedi­
ous outcome - that's the reason we play
l...b6 in the first p lace !
As ever when learning an opening, I
would recommend that you pick one of
its greatest exponents and study what the y
do. Tony Mi le s still pl ays the s ystem regu­
larly and continues t o produce highly
i m aginative ideas; and currently the other
practitioner who has enj oye d success with
l...b6 is the Latvian Grandmaster Edvins
Kengis. You'll find p le n ty of games by
both these pl ayers in the b ook .
And finally, on a personal note, of all
fence is the one which has given me per­
h aps the greatest sense of fun, and even
some reasonable results, when I have tried
it. One thi ng I can guarantee: play it in the
right sp irit and you'll ne ver be bored.
9
 CHAPTER ONE I
M ai n Li ne w ith 3 e4 iLb 7
4 i.d3: The 'new' 4 tt:Jc6 . . .

Let us begin with what has evolved in ment of the b' O? !.5 � :-!?
... as the g2-
recent years as the main line of the English pawn has been le:IT -e:1ded, but while
Defence. this is great Xe:! own to be
1 d4 e6 2 c4 b6 3 e4 risky. I supfJOse :.. is a ...,.-.er of taste. If
White places three pawns in the centre you are prepa.-ed � :o� i: en I would
and in so doing displays a metaphorical direct you to C�- .., &-.::::nes 12-19. Of
two-fingered salute to Black's opening the other alte.� es. .. ib4+ is a bit
strategy. The battle is on. If Black is un­
able to counter White's central domina­
tion then he could be driven off the edge
of the board.
3 . . . .i.b7 '
- s :nost popu-
The counterattack begins. The bishop lar move in rece::.: ;ec...-s.
snipes at the centre from the edge of the
board, reminding White that when he G:.
assumes such an imposing central stance Ruzele-Kengis
he also takes on great responsibility. After Bad GodesO:rro 996
White's centre had been chopped down in
various ways in the early years of the 1 d4 e6 2 c4 b6 3 e4 ..ib7 4l.d3
opening, it gradually became clear that the When the F-,�;--._ ::x:� �ail its re­
most solid and reliable method of defend­ surgence in e -<ee se.-� = eighties,
ing the e-pawn was ... 4 .. .f5 was the move 3� �ec o When
4 .i.d3 experience ro...OC :..::.;;: :.. -- � the ,

The alternatives, 4 tt:lc3, 4 dS, 4 f3, 4 search was 0:1 i"o:- ·-��. cs, = that is
'ilfc2 and 4 tt:ld2, all allow Black more when...
chances of attacking the centre (see Chap­ 4...l2Jc6
ters 3-5) . After 4 i..d3 Black faces some­ ... became po. U.:Z. -=:-o =y Q"';4_edge it
thing of a dilemma. How should he go was first tried a: :o _c-� :_ Sosonko­
about undermining White's centre? The Miles, Tunis b:cw� - ough
natural move to meet the early develop- that nine-move dra_,.- ··' eh,

10
 Main Line with 3 e4 i.b7 4 i.d3: The 'new' 4 . . .tiJc6

4. . . ll'lc6 has nevertheless pr oved since then 7.Ji1e7

to be a dur able and reliable method of I favour this method of development,
coping with White's system. Black does even though it is fairly slow. The straigh t­
not seek to destroy the pawn centre forward development of the knight to f6
straight away, as in so many other lines of would be asking a little too much of the
this ope nin g, but rather seeks to grab the positio n: it would kicked around by the
bishop on d3 with his k night. Surpris­ white pawn advancing to e5. However, it
ingly, this takes the sting out of White's is possible to develo p the knight to f6 with
entire set-up. First White must think how a c ouple of preparatory moves (see Game
he is to deal with the threat to his d-paw n. 4 ).
5 !iJf3 8 ti:lc3'Llg6
This and 5 t"Lle2 (Games
6-10) are the The knight arrives at a square wh e r e it
most popular moves, whereas 5 d5 is the is not going to be hassled by an advancing
move which Black would really like to pawn.
provoke (Game 11). 9d5
5 . . .!iJb4 This is a common move for Wh ite in
This is the idea of bringing the knight thes e kinds of position. He feel s he should
out so early, to force an exch ange for the be using his lead in development to 'do
bishop. The bishop cannot retreat as thee­ something', though it isn't clear to me
paw n would be en prise. what that something is. At least the cl­
6 0-0 lt:\xd3 p aw n is now blocking the bishop on b7,
Take the bishop quickly! Otherwise the though that piece is still the pride of
knight might have had a wasted journey. Black's position - it has no o pponent, so if
7 'i'xd3 it ever finds the right diagonal it will be a
great asset. White's alternatives here are
considered in Games 2 and 3.
9 ... i.e7 10 i.e3 0-0 11 �ad1 e5
It might not seem to make sense to
close the position when Black has the two
bishops , but they will emerge later!
12 tile2 d6
The blocked p aw n structure is reminis­
cent of a King's Indian. Black would love
to get in the pawn break ... f7-f5 to attack
on the kingside. Normally White counters
this with pl ay on the queenside (via b 2-b4
and c 4-c 5 ) but that would be far too slow
Black is a long way behind in develop­ in this case (it is quite handy for Bl ack that
ment and is quite seriously cr amped. the pawn is already on b6, covering the c5-
However, he is able to survive (more than square). Instead White attempts to thwart
survive, I believe that Bl ac k ' s chances are Black's kingside play:
no worse than his opponent's) as the posi­ 13 <2lg3 i.cS
tion is still closed. Since there are no open Bringing the bish op back to life.
files Black is able to shuffle his pieces onto 14 h3
the right s qu ares while White is only an I don't quite see the point of this move.
onlooker. If White had played l"Of5 we would have

11
English Defence
arrived at a position similar to the next
game.
14...llJh4
To rel ie ve a little of the c o ngest ion in
his position Black seeks exchanges.
14 . . . tllf4 would have been more ambi­
tious, for instance:
a) 15 �xf4 exf4 16 ltJe2 (not 16 tt:Jhs
�gS and the knight on h5 is stranded)
16 . . . �£6 17 b3 gS 18 tbed4 �d7 with
chances for both sides.
b) 15 'i'd2 i.gS 16 ltJxgS 'i'xgS and
Black is on the attack.
15 !Dxh4 .i.xh4 16 '-'Llf5
16 ... g6
This is fine, but afterwards the game
simplifies very quickly. Black had a couple
of alternatives which would have kept a
bit more tension in the posit ion:
16 . .. �g5!? or 16. . .i..f6!?, threatening to
12
either capture on fS and break out with
... e5-e4; o r to kick the knight out wit h
. ..g7-g6, ret reat the bishop to g7 again and
pl ay for ...f7-f5.
17 lbxh4 'i'xh4 1 B b3 .i.d7 19 f3 f5 20
exf5 gxf5 21 f4 exf4 22 'i'd4 =.aeB 23
'*'xt4 "llkxf4 24 .i.xf4 Y..-%
Fair enough. There is not too much to
play for now, but bo t h sides could have
pl ayed more ambitiously e arlier - as the
next game illustrates.
Game2
Magerramov·Ehlvest
Moscow 1992
1 d4 e6 2 c4 b6 3 e4 .i.b7 4 .i.d3 lbc6 5
l0f3 lbb4 6 0-0 li:Je7 7 tt:lcJ lbxd3 B
'it'xd3 tt:lg6 9 b3 .1J..e7 10 d5 e5 11 lbe2
0-0 1 2 lLlgJ d6 13 tt:lf5 .lieS
We have virtually the same position as
the last game, t hough White has pl aye d
more purposefully. The knight looks very
fine on fS, but the question is, can he
make use of it?
14 i.d2 J:!.e8 15 J::tae 1 .i.d7
16 �h1 i.fS
Ehlvest has defended very coo lly, with­
drawing the bishop out of the k night ' s
range. The bishop is also well placed on d7
and White must alway s be careful tha t the
knight on fS doesn't get exch anged off at a
bad moment.
17 g3 �cB
 M a in L in e with 3 e4 il.. b 7 4 il..d3: Th e 'n e w ' 4
Now there is the possibility of captur­
ing on f5, playing ...e5-e4, and taking with
the queen on f5. So the knight retreats.
1 8 lLle3 h6 1 9 lLlg1 JJ.. e 7 20 f4
Dangerous? Well, yes, but Black has or­
ganised his forces efficiently so that he can
cope. In fact White's pawns look more
exposed than threatening.
20 . . . exf4 2 1 gxf4 il.. h4 22 .!;le2 JJ..t6 23
lLlg2
If 23 f5 then the knight gets a fantastic
square on e5. In other words, it is going to
be difficult to advance the pawns further,
and in the game they prove to be more of
a liability than an asset. But Black still
lacks space; so what does he do? Create
some:
23 . . . b5!
A smart pawn sacrifice ensuring that
the light-squared bishop (the one without
an opponent) will be able to get into the
game.
24 c5
If White had captured the pawn Black
would have had plenty of play: 24 cxb5
�b7 (or 24...a6 25 bxa6 'i'xa6 26 "i'xa6
�xa6 27 a4 �b8 and Black recovers the
pawn) 25 a4 a6 26 bxa6 llxa6 to be fol­
lowed by ...c7-c6; then just imagine that
bishop on the long diagonal ...
24 . . . dxc5 25 e5
The pawns appear impressive, but
Black has it all under control.
. . .li'Jc6
25 . . . il..f 5 26 'ii'x b5 c6! 27 dxc6
27 'ilfxc6 'ilfxc6 28 dxc6 it..d3 picks up
the exchange.
27 . . . JJ.. e 7
The point behind Black's pawn sacrifice
was to break up White's pawns...
28 J:!.e3 J:!.b8 29 'i!Va4 l:!.b6 30 c7 J:!.a6 3 1
il.. a 5 il..d 7 3 2 'i!Va3 iLc6
...and get that bishop to the long diago­
nal. White is in desperate trouble.
33 lLlt3 'i!Vg4 34 e6 fxe6 35 b4 cxb4 36
Wic 1 Ji.. b 7 0-1
The bishop on a5 is about to drop,
among other things, so White resigned.
For me, there was poetic justice in this
game: White was punished for over­
extending his centre and the black light­
squared bishop (which had reigned su­
preme since its opposite number was
nabbed at the start of the game) played a
significant part in its downfall.
I once had a game myself which also ar­
rived at the same pawn structure, though
in this case I was allowed to develop an
attack on the kingside.
Game3
Mednis-King
Stavanger 1989
1 c4 b6 2 d4 e6 3 e4 Ji..b 7 4 il..d3 lLlc6 5
lLlf3 lLlb4 6 0-0 lLle 7 7 lLlc3 lLlxd3 8
"i'xd3 lLlg6 9 J:!.e1
13
 M a in Lin e with 3 e 4 !ii. b 7 4 !ii. d3: Th e 'n e w ' 4.. .tiJc6
1 8 c5 bxc5
Unfortunately, opening the a-file
doesn't work as well now: 18... axb4?! 19
cxd6 cxd6 20 axb4 followed by tLlc4, at­
tacking the b6-pawn.
1 9 bxc5
Black still has some attack, but it is not
as strong without an open file for the rook
on the queenside.
1 9 . . . .lag6 20 �g 1 !ii. a 6 2 1 'ifc2 f4 22 Ji.f2
�h6 23 a4 'it'e8 24 liJb5 liJf5
24...i.xb5 25 axb5 'ili'xb5 is stronger.
25 h3
Not 25 exf5?? l:txh2+ 26 Wxh2 �h5+.
25 . . . ii.xb5 26 axb5 liJe3 %-%
I offered, a draw here and was lucky
that it was accepted (my opponent was in
time pressure). After 27 i.xe3 fxe3 28 tLlc4
'iVxb5 29 cxd6 cxd6 30 tLlxe3 the attack is
all but over, but Black's rook is stuck out
on the edge of the board, cut off from the
queenside. The a-pawn gives me some
counterplay, but there is no doubt that
White stands better. For me, a disappoint­
ing conclusion after a successful opening. I
repeat, I don't think that Black should
have any difficulties if White plays in such
a 'normal' fashion.
It is worth seeing how two of the great­
est exponents of the English Defence have
played this system. First, in this game,
Edvins Kengis, and in the next, Tony
Miles.
....----..
Game4
Ba rkhagen -Kengis
Gausda/1991
1 d4 e6 2 c4 b6 3 e4 Ji.b7 4 Ji.d3 liJc6 5
liJf3 liJb4 6 0-0 liJxd3 7 �xd3 d6
A divergence from the usual 7...tLlg6,
which we saw in the first three games.
8 liJc3 Ji.e 7
see following diagram
9 b3
Kengis has faced two other moves here:
a) In Cmiel-Kengis, German Bundesliga
1992, White tried 9 li'e2 tLlf6 10 e5, but I
think this is a fundamental error as it
opens up the diagonal for that wonderful
bishop on b7. Black reached a superior
ending after 10...dxe5 11 dxe5 tLld7 12 tLle4
0-0 13 i.f4 tLlc5 14 tLlxc5 i.xc5 15 tLlg5 h6
16 l:tad1 'iVe8! 17 tLle4 'iVc6! 18 l:tde1 'ii'xe4
19 'iVxe4 i.xe4 20 .l:i.xe4 .l::!.ad8. Black con­
trols the only open file.
b) The most challenging move for
Black at this point is 9 d5!? e5 10 c5! tLlf6
11 'iVb5+ <t>f8 12 c6 i.c8. It takes a bit of
time for Black to sort out his king, and
from White's point of view it is a definite
plus to have forced the pawn to c6, as
Black is more cramped. Nevertheless,
Kengis proved that his system is still viable
after 13 a4 a5 14 'iVe2 g6 15 tLle1 <t>g7 16
tLld3 i.a6 17 f4 exf4 18 llxf4 h5 19 h3 h4
20 i.e3 tLlh5 21 i.d4+ <t>g8 22 l:tf2 l:th7 23
tLlb5 tLlg3 24 'i'f3 i.g5 25 lle1 i.xb5 26
axb5 'ili'e7 27 e5 l:Ie8 28 i.c3 (28 e6!? tLlf5
29 i.c3 fxe6 30 l:txe6 \lff7 31l:tfe2 .Uxe6 32
l:txe6 tLlg3 offers Black sufficient counter­
play) 28...'iVd8 29 'iig 4 dxe5 30 tLlxe5 f5 31
'ili'd1 .Uhe7 32 tLlc4 l:txe1+ 33 i.xe1 <t>h7 34
i.c3 tLle4 35 .l::!.f3 i.e7 36 i.e5 i.c5+ 37
<t>h2 l:txe5 38 tLlxe5 'iVd6 39 .l:i.xf5 gxf5 40
'iVh5+ <t>g7 41 'i'f7+ Wh6 42 'ili'xf5 'ifxe5+
0-1 Shirov-Kengis, Gausdal (Arnold Cup)
1991.
15
En g lish D e fe n c e
9 . . . lbf6
As can be seen from these games, Ken­
gis likes to develop his knight to f6. I
don't think it greatly changes the assess­
ment of the position (I believe chances are
balanced) but it demonstrates another
sound option for Black.
1 0 �b2 0-0 1 1 d5 e5
1 2 b4
Perhaps because Black appears cramped,
White underestimates the potential of his
opponent's position. For instance, the
bishop is sitting on b2 - but it is now just
biting on the solid pawn on eS; it would
be stronger on cl where it covers the f4-
square. And this last move has actually
created a serious weakness in White's posi­
tion. As we are about to discover, the
monster bishop on b7 is about to enter
the game. First Black establishes some
pressure on the a-file:
1 2 . . . a5!
White cannot contemplate capturing on
aS as it would leave the a-pawn isolated,
and the cS-square available for a knight.
1 3 a3 lbh5
The knight heads for the outpost on f4
and makes way for the f-pawn...
1 4 �fe 1 lDf4 1 5 'ilfc2 f5 1 6 lbd 1
1 6 exfS would bring the light-squared
bishop into the game, as White is unable
to hold onto the pawn: 16... i.c8! 17 g4?!
hS and the kingside breaks up.
16
1 6 . . .fxe4 1 7 J:txe4 axb4 1 8 axb4 �xa 1
1 9 �xa 1
1 9 . . . b5!
Undermining the pawn chain. White is
lucky to scrape out of this position with
just the loss of the exchange, though he
does go down eventually anyway.
20 lbe3 'i'a8 2 1 �b2 c6 22 c5 cxd 5 23
cxd6 dxe4 24 dxe7 l:tc8 25 eS'i!f + l:txe8
26 lbxe5 'i'a2 27 h3 l:tc8 28 'itkd2 'ifb 1 +
2 9 �h2 lbd3 3 0lDxd3 'ii xd3 3 1 'ife 1
In spite of the clear material advantage
Black has great difficulty in winning this
position as White is able to construct a
kind of fortress. It takes great patience to
crack it, but Kengis succeeds in the end by
concentrating on White's weaknesses,
notably the b-pawn, and thereby forces
the exchange of queens; thereafter it isn't
too difficult.
3 1 . . . l:tf8 32 �g 1 � d5 33 �e5 �c4 34
'iic 1 'ife2 35 �g3 .:.ds 36 'i'a 1 'ifd2 37
'ifa5 l:tf8 38 �h2 'iiid7 39 'ifb6 �d3 40
'ifc5 lieS 4 1 'i'e5 :ea 42 'i'c5 h6 43
'itkb6 'iii e 6 44 'ifc5 �h7 45 'it'd4 � f 8 46
'ifc5 l:tf7 47 �g 1 'i'a2 48 Wd4 'ifa7 49
'ife5 Wie7 50 'i!id4 'i'f6 5 1 'i'c5 'it'b2 52
'it'd5 'ifa 1 + 53 �h2 'f/a7 54 'i1tg 1 'i'd7 55
'itkc5 'ii e6 56 'i'd4 ·'i'c6 57 'itoh2 l:.d7 58
'iii e 5 'iff6 59 'iiic5 'iiie6 60 'itog 1 �c4 61
'itoh2 l:tf7 6 2 'i'd4 'iic6 63 'iiie5 �e6 64
Wib2 l1a7 65 'i'd4 :.d7 66 'irb2 :.d3 67
�e5 'i'd7 68 �c3 'i'a7 69 �e 1 'i'd4 70
 Main L in e with 3 e4 i. b 7 4 i. d3: Th e 'n e w' 4
'i'c 1 i.c4 7 1 Wg 1 l:!.b3 72 0.f5 'i'e5 73
0.e3 'i'b2
Mission half-accomplished.
74 'i'xb2 .=:xb2 75 i.c3 l:!.b3 76
l:!.b 1 + 7 7 �h2 l:!.b2 78 i.e 1 .=:e2 79
l:!.xf2 80 <ot>g3 l:!.f7 8 1 i.d4 i.e6 82
'it>g6 83 �h4 l:!.d7 84 �g3 'it>g5 85
l:!.f7 86 i.b6 Wh5 87 h4 .:.d7 88
.:.d3 89 �f4 �xh4 90 �xe4 l:!.d2 91
l:!.d7 92 �e5 .li.g4 0- 1
Mission accomplished.
Game 5
Polovod in -Miles
Los A ngeles 1991
1 e4 b6 2 d4 .li.b7 3 i.d3 e6 4 c4 0.c6 5
0.f3 0.b4 6 0.c3 0.xd3+ 7 'iWxd3 .li.b4
I remember looking at this variation
over ten years ago and coming to the con­
clusion that if Black were given the chance
to pin then he should do it. An exchange
of the bishop for knight clears more space
for Black (he is manoeuvring behind three
ranks for the moment) and puts more
pressure on White's centre. It is worth
noting that if White wishes to avoid this
pin then he could castle on move six in­
stead of playing the knight to c3. Of
course if he does castle kingside quickly
then White somewhat reduces his own
options - in the note below the king goes
left instead of right with success.
. . . 0.c 6
8 d5
White has also tried 8 .i.d2, indicating a
desire to castle queenside. After 8 ... lt:Je7 9
i.d2 a3 �xc3 10 iLxc3 0-0 1 1 d5 Black has:
.Jic3 a) A safe option for Black is 1 1...f6,
i.b6 closing the diagonal. After 12 0-0-0 e5
i.c5 (12 ... exd5!?) Black not only has a very
i.c5 solid position, but there are also chances
i.f8 to attack on the queenside.
b) Also playable is 1 l...c6, not fearing
12 d6 lt:Jg6 13 c5 b5. In the game Sherba­
kov-Bischoff, Linares 1996, White tried 13
h4!?, and after 13 ... cxd5 14 cxd5 (or 14
exd5 b5!) 14 ... exd5 15 e5 lt:Jg6 a very un­
clear position had arisen.
c) In the game Levin-Teske, Bad
Worishofen open 1998, Black tried the
rather risky 1 l...f5? and suffered on the
long diagonal after 12 'it'd4 .l:tf6 13 0-0-0
fxe4 14 'il'xe4 exdS 15 cxd5 .l:td6 16 'it'e5
'il'f8 17 �b4 winning the exchange.
8 ... 0.e7 9 a3 i.xc3+ 1 0 'iWxc3 0-0 1 1 0-0
White's centre is ripe. In classic style
Miles hits out with the f-pawn.
1 1 ...f5!
1 2 dxe6
12 exf5 would have been a more dy­
namic way of playing the position:
a) 12 ...lt:Jxf5 is a solid option, when af­
ter 13 dxe6 (13 .i.g5 "i!Ve8 14 .l:tae1 "i!Vg6 just
swings the queen to a good attacking
square) 13 ... dxe6 Black's active pieces
compensate for the weak pawn on e6.
17
En glish D e fe n c e
b) 1 2... exd5 would b e more daring for
Black (he might just win a pawn) but after
13 .ltg5 White has good development and
free lines for his rooks.
1 2 fxe4 1 3 lt:lg5 'iVeS 1 4 J:le 1 'iVg6 1 5
.• .
exd7 J:.ad8 1 6 it'e5
White could have tried 16 l!d1 1i'f5 17
i..e3 .l:hd7, though the position of the
knight on g5 is still unfortunate.
1 6 . . . J:.xd7 1 7 'i'e6+ �xe6 1 8 lt:lxe6 l:.f6
1 9 li:lf4 li:lf5 20 lt:le2
White runs into trouble if he plays a
'normal' move, e.g. 20 i.e3 tZ:lxe3 2 1 fxe3
.l::td2 and Black controls the file.
20 . . . e3!
A pleasing sacrifice to open the line of
the bishop on b7.
21 fxe3 li:lh4 22 e4 .ll xe4 23 lt:lg3 .llx g2
24 i.g5 li:lf3+ 25 �xg2 lt:lxg5
With an extra pawn Black should have
won, but he blows it right at the end.
26 .l:!.ad 1 l:: df7 27 l:.e8+ l:HS 28 l:.xf8+
�xf8 29 .tld8+ <3Je 7 30 .l::t a 8 a5 3 1 c5
.tlc6 3 2li:lf5+ 'it>t6 33 cxb6 .l:tc2+ 34 �g3
cxb6 35 J:.f8+ 'it>e5 36 lt:lxg7 lt:le4+ 37
�h3 .l:!.xb2 38 .l:!.f5+ �d4 39 lt:le6+ r;i;>c4
40 li:lf8 .tlb3+ 41 r;i;>g4 l:.xa3 42 lt:lxh 7
J:.a 1 43 J:.t4 �d3 44 h4 a4 45 lt:lg5 lt:lc3
46 lU3+ �c4 47 h5 l:.g 1 + 48 'it>h4 a3 49
h6 a2 50 h7 a 1 'if 51 hS'if J:txg5 52 .l:!.f4+
'ot>b3 53 'i'f6 .l:!.b5 54 'ife6+ 'it'a3
54...�b2! 55 .l::tf2+ �a3 56 'iWd6+ .l::tb4+
would have won for Black.
18
55 'ir'e7+ �b3 5 6 'ir'e6+ li:ld5 5 7 'iih3+
lt:le3 58 i*'e6+ Wb2 59 'i'f6+ Wb1 60
'it'g6+ 'Df5+ %-%
The conclusion is that Black has a com­
fortable game if he is able to pin the
knight on c3 - but White need not allow
it.
Game6
Arlandi-Yefimov
Asti 1 995
1 d4 e6 2 c4 b6 3 e4 .ll b 7 4 .lld 3 li:lc6
Black faces a sterner test if instead of 5
tZ:lf3 White replies...
5 lt:le2
At first glance it is hard to imagine that
playing the knight to e2 rather than f3 can
make such a great difference. However,
with the knight on e2 White has an extra
option: he can use the f-pawn as a batter­
ing ram to open Black's position. We can
see straightaway in this game how this
affects the way in which Black is able to
develop. (By the way, I should add that in
practice most players of the white pieces
have preferred 5 tZ:lf3.)
5 . . . lt:lb4
Not the only legal move. Speelman has
tried 5 ...e5 and Kengis has experimented
with 5 ... g6. For these options, see Games 9
and 10 respectively.
6 0-0
 Main L in e with 3 e 4 i. b 7 4 i.d3 : Th e 'n e w ' 4
The continuation of the game Anasta­
sian-Dausch, Cappelle la Grande open
1996, is a good example of the f-pawn in
action: 6 l2lbc3 l2lxd3+ 7 �xd3 g6 (for
7... i.b4!? see Game 8) 8 0-0 i..g7 9 f4 f5?!
(9 ...l2le7 strikes me as more sensible) 10
exf5 gxf5 11 d5 'tlfh4 12 Cllb 5 0-0-0 13 d6
'it'g4 14 .l:.f2 c6 15 l2lxa7+ 'iii'b 8 16 i.e3 and
White already had a winning attack.
6...l'Dxd3 7 'iVxd3 l'De7
For 7... d6, see the next main game.
8 l'Dbc3 g6
Black must change his method of de­
fence. If the knight were on f3, then in­
stead 8... l2lg6 would be quite acceptable
here, as we have seen. However, in this
position 8 ... l2lg6 would be met by 9 f4!,
when Black is in danger of being overrun.
Note, however, that 8...d6 transposes to
the next game.
9 i.g5
9 �h3 is an interesting alternative, hop­
ing to exchange bishops on h6: 9... i.g7
(9...d6 is more flexible. Black can go for
...'ifd7 and . . 0-0-0, or revert back to cas­
. is also strong) 18 ... :r.fd8 19 �f3 gives
tling kingside, depending upon how
White reacts) 10 i.h6 0-0 1 1 i.xg7 ( 1 1
:r.ad1! would have prevented Black's de­
fensive manoeuvre) 1 1...'iii'xg7 12 .l:.ad1 d6
13 tllt4 ii::lg 8 and even here White enjoyed·
a pleasant space advantage in Kohlweyer­
Gulko, Geneva 1997.
9 . i.g7
. .
. . . l'Dc6
1 0 f4
There it goes again. The f-pawn spells
danger, though Black constructs a sturdy
defence.
1 0... f6 1 1 i.h4 0-0 1 2 d5
If White were to be entirely consistent,
then 12 f5! ? is the move, with an exchange
sacrifice in mind: 12 ...exf5 13 exf5 l2lxf5 14
:r.x£5 gxf5 15 Cllg3. However it does not
look sound to me; for instance, 15... i..e4!,
getting rid of one of the knights and using
the bishop before it gets locked out of the
game with d4-d5.
1 2 .. . .lta6 1 3 J:ad 1
13 b3 looks stronger to me, so that if
Black continues as in the game with
13...�e8 White can play 14 d6! cxd6 15
�xd6 with a clear advantage. Likewise,
13...exd5 14 l2lxd5 l2lxd5 15 �xd5+ .l:.f7 is
obviously more pleasant for White than
his opponent.
1 3 .. .'i'e8 1 4 l'Dd4 exd5 1 5 exd5 ikf7 1 6
b3 c5 1 7 l'Ddb5
17 dxc6!? dxc6 18 f5 (18 'iff3 i.b7 19 f5
White some advantage.
1 7 ... i.xb5 1 8 l'Dxb5
With a series of accurate moves Black
now manages to equalise the position, but
I feel that White has had the better of the
opemng.
1 8 .. . l'Df5 1 9 .tf2 d6 20 g4 a6 21 l'Dc3
l'Dh6 22 h3 ikd7 23 .:de 1 l:tae8 24 l:txe8
79
English Defence
::Xe8 25 �e1 f5 26 g5 l?Jf7 27 �xe8+
xe8 28 �e2 �8 29 �xe8+ �xe8 30
.:i..e1 Y,-Y,
Camel
Gelfand-Short
Novgorod 1997
1 d4 e6 2 c4 b6 3 e4 Si..b 7 4 il.d3 l?Jc6 5
�2 l?Jb4 6 l?Jbc3
ot the most accurate continuation as
it gives Black the chance to pin the knight
on c3 if he wishes to with 6 ...lZ'lxd3+ 7
'i'xd3 �b4 (see Game 8). To avoid that
possibility White could simply castle king­
side straightaway.
6 . . . l?Jxd3+ 7 'li'xd3 d6
Since 7...lZ'le7 and 8 ...g6 is lacking, this
has taken over as the move of choice for
the those in the know.
8 0-0 l?Je7!?
8 ...lZ'lf6 has also been played, though it
is less dynamic: 9 d5 �e7 10 lZ'ld4 'i'd7 11
f4! (11 b3, as in Epishin-Ehlvest, Novo­
sibirsk 1993, is playable but obviously less
critical) 11...c5 12 lZ'lde2 exd5 13 cxd5 h5
14 a4 h4 15 h3 0-0 16 b3lZ'le8 17 i..b2 lZ'lc7
18 f5 i.f6 19 �f4 'i'e7 20 l!g4 �h7 21 �d2
tt'le8 22 �h1 g5 23 lZ'lg1 l!g8 24 lZ'lf3 llg7
2S Irfl �g8 26 e5! dxe5 27 lZ'le4 with a
lethal attack in Anastasian-Kengis, Ka­
towice open 1993.
9 d5
20
9 'ii'h3 was tried in Bareev-Kengis, Pula
1997, but the queen manoeuvre can be
immediately countered by 9 ...'i'd7! (the
queen is good on a light square when
White has no bishop to challenge it) 10
Ji.e3 g6 11 d5 i.a6 12 dxe6 'ilixe6 13 'ii'h4
�g7 14 b3 0-0 15 i.d4 f6 161!ad1 l:!ae8 17
l:!fe1 c5 18 i..e3 f5 19 �h6 i..xh6 20 'i'xh6
fxe4 21 lZ'lg3 lZ'lf5 22 lZ'lxf5 'tixf5 23 �e3,
when a draw was agreed as 23 ...l:!.e6 24
lZ'lxe4 .ltb7 25 f3 .ltxe4 26 fxe4 leads to an
equal position.
9 ...�d7!
Black retains flexibility. He still might
castle on the kingside, but he is also pre­
pared to go queenside; and the queen cov­
ers the sensitive e6-square. Black could
close the centre with 9 ... e5, but I would be
concerned about a direct assault with 10
f4.
10 f4!
Although White's strategy rebounds in
the end, I feel that if he is to get anything
from this variation he must play aggres­
sively, so this one gets my approval. Al­
ternatively, 10 i..e3 lZ'lg6 (10...g6!?) 11 f4!
i..e7 12 i..d4 0-0 13 dxe6 fxe6 14 f5 e5
(14 ... exf5!? 15 exf5 lZ'lh4 is good fun: 16
'ikg3 i..d8 17 f6 c5 18 fxg7 l!f5 19 i..e3
'Yi'xg7 20 'i'xg7+ �xg7 is a bizarre varia­
tion. Okay I admit it, I used the computer
to check this one out but I don't see how
White can improve - Black is doing fine in
this position.) 15 �e3 lZ'lh8 16 lZ'ld5 i.d8
17 J:!ad1 lZ'lf7 18 b4 c6 19 lZ'ldc3 "Ylt'e8 20
lZ'lg3 .ltf6 21 'it>hl .l:f.d8 22 a4 .l:td7 23 'iVe2
'it>h8 24 .l'!d2 'ike7 25 b5 and White was a
bit better in Dautov-Kengis, Hockenheim
rapidplay 1997, though the game was
drawn in 44 moves.
10 ...g6 1 1 l?Jd4
11 "YlVd4 l:!g8 is satisfactory for Black,
but in his annotations in Informator 70
Short suggests that the continuation 11
.lte3 i..g7 12 i..d4 would have slightly fa­
voured White.
 Main L in e with 3 e4 iJ.. b 7 4 JJ.. d 3 : Th e 'n e w ' 4
1 1 . . . 0-0-0
Advisable. At first I thought it might be
preferable to delay castling for a move
with 1 L .i.. g7, but White can stir up
enough trouble with 12 f5! gxf5 13 exf5,
and now after either 13 . . . i..xd4+ 14 'i'xd4
0-0-0 (14 . . . e5 15 'i'f2) 15 fxe6 fxe6 16 i.. g5
or 13 . . . lbxf5 14 lbxf5 exf5 15 i.. g5 White
has taken controL
1 2 b4
After 12 dxe6 fxe6 13 'iVh3 i.. g7!? (after
13 . . . e5 14 lbe6 .l::.e 8 15 lbd5 White has an
attack) 14 lbxe6 i.. xc3 15 bxc3 (not 15
lbxd8?? i.. d4+ 16 �h 1 'i'xh3 and wins)
15 . . . .Ude8 Black has excellent compensa­
tion for the pawn - the pawns on e4 and
c4 are extremely weak.
However, the move which concerns me
here is 12 a4. White's attack is potentially
lethal. For instance:
a) 12 . . . a5 13 b4 axb4 14 lbcb5 followed
by a4-a5 and a big explosion on the queen­
side. Black is so cramped that it is ex­
tremely difficult to bring his pieces across
to defend the king.
b) 12 . . . i.. g7 is stronger, e.g. 13 a5 exd5
14 exd5! (14 cxd5 f5! 15 lbe6 fxe4 16 lbxe4
lbxd5 17 lbxd8 ltxd8 18 axb6 axb6 with
good compensation for the exchange)
14 ... lbf5 and now White can go in for
insane complications with 15 lbdb5 a6 or
play for an endgame advantage based on
Black's weak pawns after 15 lbxf5 'i'xf5 16
. . .ti:Jc6
'i'xf5 gxf5.
c) Finally, Short suggests 12 ... c5!? 13
dxc6 lbxc6 with an unclear position.
1 2 . . . JJ.. g 7 1 3 b5?
All very fine if White could land the
knight on c6, but ...
1 3 . . . JJ.. xd4+!
. . . stops that idea, and then it just looks
like pushing the b-pawn has wasted two
moves.
1 4 'it'xd4 f5
Once again the f-pawn saves the day,
undermining White's centre and letting
Black's bishop breathe again. Without this
move Black has no play.
1 5 l:te 1
If 15 exf5 then Stohl recommends
15 . . .lbxf5 (I prefer 15 ... exd5 16 cxd5 'i'xf5
17 l:td1 lthe8 18 a4 tt:lg8! 19 a5 lbf6 20
axb6 axb6 21 lta7 tt:le4 with good coun­
terplay) 16 'i'd3 i¥g7 17 ltd1 .l::.he8, which
he assesses as unclear.
1 5 . . . l:the8
Black has to build up first before dis­
solving the centre. For instance, 15 . . . exd5
16 exd5 and the bishop on b7 is perma­
nently locked in, while 15 ... fxe4 16 'i'xe4
loses the pawn on e6.
1 6 iJ.. b 2
16 'i'g7 fxe4 17 tt:lxe4 tt:lg8! covers eve­
rything.
1 6 . . . fxe4 1 7 ltlxe4 ltlg8 1 8 ltlc3 'it't7 1 9
l:te3
21
En g lish D e fe n c e
I would still b e concerned about White
going for an attack on the queenside with
19 a4, though in this case Black has
enough counterplay after 19 ... tt:Jf6!
1 9 . . . exd5
20 ll:lxd5
Black has good counterplay whatever
White decides to capture, e.g. 20 cxdS
l::txe3 2 1 it'xe3 iLxdS, winning a pawn, or
20 l::txe8 l::txe8 and now:
a) 21 cxdS tZ:le7 22 hld1 (22 l:te1 tZ:lxdS!)
22 ... tt:Jf5 and an invasion on e3.
b) 21 tZ:lxdS tZ:le7 (21....l:i.e2!?) 22 tZ:lxe7+
!Ixe7. White has a couple of vulnerable
points in his position (g2 and c4) whereas
Black has no real weaknesses. The oppo­
site-coloured bishops make life difficult
for White as he will be unable to dampen
the attack with exchanges.
20 . . . l:!xe3 2 1 �xe3
Understandably, Gelfand decides to
ditch a pawn rather than lose the initia­
tive. If 21 tZ:lxe3 then 21....l:i.e8 and the bal­
ance is shifting in Black's favour.
2 1 . . . .i.xd5 22 cxd 5 �xd5 23 a4 �b8 24
a5?
This offer of a second pawn is just too
much. Black would still have had some
work to do to convert his extra pawn if
White had played 24 :!! c l, reminding
Black that c7 is vulnerable. Everything
would be fine for Black if the knight could
just find a decent square, but it is not easy
22
to re-deploy it.
24 . . . 'iWxb5 25 axb6 axb6 26 .i.d4 �c6
27 l:!c1 �b7
Now Black is winning.
28 f5 gxf5 29 �g5 l:!e8 30 �h5
Or 30 iVxfS it'e4 and wins.
30 . . .'i'e4 3 1 .i.f2 lLlf6 32 'iWf7 'Wie7 33
'lia2 ll:lg4 34 l:!a 1
34 SLd4? it'el+! 35 .!:txe1 l:txe1 is mate.
34 . . . 'iWe4 35 �a7+ Wc8 36 l:!c1 'i'b7 37
'Wia4 l:!e7 38 'Wib3 0-1
White gave up here. There's not much
hope, e.g. 38 ... tt:Jxf2 39 'ii'g8+ ..t>d7 40
Wxf2 it'e4. An excellent game by Short.
He played with great accuracy, but I feel
that this whole line needs checking.
White's queenside attack concerns me.
Game 8
Franco-Teske
Havana 1998
1 d4 e6 2 c4 b6 3 e4 .i.b7 4 .i.d3 ll:lc6 5
ll:le2 ll:lb4 6 ll:lbc3
6 0-0 is more accurate as it avoids the
pin, though in approximately half the
games I've seen White plays his knight out
first, as in the text game.
6 . . . ll:lxd3+ 7 'i'xd3 .i.b4
This isn't quite as effective compared to
when White has the knight on f3 (the
knight on e2 is ready to recapture) . Never­
theless, the exchange still helps Black to
free his game.
8 0-0 ll:le7 9 a3
9 tt:Jd1!? is worth considering, intending
to trap the bishop. Here are a few possi­
bilities:
a) 9 .. .f5 10 eS tLlg6 11 cS bxcS 12 a3
SLaS, when it is messy but Black should be
fine.
b) 9 ... tLlg6 10 cS (10 tLle3!?) 10 ... bxc5 1 1
a3 �aS 1 2 dxcS c6 1 3 b4 Ji.. c7 and I prefer
White's position.
c) 9 ... c5 10 a3 SLaS 11 l:Ib 1 and Black
has not solved his problems.
 Main L in e with 3 e4 i.. b 7 4 i.. d 3 : Th e 'n e w ' 4
9 . . i.. x c3 1 0 ll:lxc3 0-0 1 1 d5 d6 1 2 i.. g 5
. pawn structure familiar from the first few
1 2 . . . �d7
Why not 12 .. .f6... ? The bishop retreats,
13 i.. e3, and after 13 ... e5 Black gets ready
for .. .f6-f5.
1 3 llad 1 .:!.ae8 1 4 ll:lb5 a6 1 5 ll:ld4 'h - 'h
After 15 lL:ld4 Black has no difficulties.
For example, 15 ... e5 16 �xe7 "i¥xe7 17
lL:lfS 'iYgS followed by ... i.. c 8; when the
knight retreats, .. .f7-f5 comes. 7 ... i..b 4,
pinning the knight on c3, is a sensible con­
tinuation which gives Black good chances
to equalise; though if White is canny he
can avoid the debate altogether just by
castling before developing the knight.
Game 9
Lobron-Speelman
German Bundesliga 1997
1 d4 e6 2 c4 b6 3 e4 i.. b 7 4 i..d 3 ll:lc6 5
ll:le2 e5
'This invites White to shut the b7-
bishop out of the game for the foreseeable
future. But it does at least stop White
from rolling Black up in the centre with a
quick f2-f4.' - Jon Speelman. Speelman
goes his own way, as usual. No one else
has tried this idea before, or since. I'm
grateful to him for supplying notes to this
game and where the comments are his, I
have said so. Black forces his opponent's
hand in the centre, arriving at a closed
. . .li:Jc6
games in this chapter. This is the only
reasonable method of getting this struc­
ture, since playing the knight to g6
straightaway runs into the path of White's
f-pawn (see the notes to Game 6) . All this
is possible because of the knight's modest
placement on e2. Now let's see whether it
is any good or not.
6 d5
White must close the centre if he is to
hope for anything from the opening. 6
dxeS lL:lxeS gives Black free and easy de­
velopment.
6 . . . ll:lb4
It is possible to flick the check in first,
but the position is too easy for White to
play, e.g. 6 ... �b4+ 7 lL:ld2 lL:lce7 8 0-0 lL:lg6
9 a3 J.xd2 10 J.xd2 and White can play
on both sides of the board. Forget the
closed pawn structure - with a space ad­
vantage the two bishops are extremely
powerful.
7 0-0 ll:le7 8 ll:lbc3
The question that must be asked is
whether or not White can use the f-pawn
attack to disrupt Black's system. Let's try:
8 f4 lZ:lxd3 9 'ii'xd3 exf4 10 i..xf4 lL:lg6.
Black should be able to develop satisfacto­
rily as he has a check on cS for his bishop
to speed the process - a big advantage of
omitting ... d7-d6.
8 . . . ll:lxd3 9 'Wi'xd3 ll:lg6
23
En g lish D e fe n c e
10 'Llg3
White makes a virtue out of the knight
on e2, which heads straight towards the
dangerous outpost on fS. Instead of that,
we should check out the f-pawn attack: 10
f4 exf4 11 tt:lxf4 .ltd6 and once again Black
has the situation under control.
1 0 . . . ..te7
10 ... -ltcS!? would have been more en­
terprising. The bishop can be distinctly
irritating on that diagonal, and White
must always watch out that the bishop
doesn't find a secure home on d4.
1 1 'Llf5 ! 0-0 1 2 g3
Speelman says that 12 d6 .ltxd6 13
tt:lxd6 cxd6 14 �xd6 may be good for
White, 'but it would be hugely anti­
positional to open up the b7-bishop's di­
agonal; and Black has possible counterplay
with .. .f7-f5.'
1 2 . . . d6 1 3 h4
White's kingside mmauve develops
quickly and it is difficult to find meaning­
ful counterplay. Speelman aims for a pol­
icy of containment on the kingside com­
bined with a pawn break on the queen­
side. Strong nerves are required.
1 3 . . . ..tf6 1 4 'it>g2 .tea 1 5 .l:th 1 .l:lea 1 6
..td2 a6 1 7 a4 ..td7 1 a :ag 1
Perhaps 18 l:Iafl!? (Speelman) . The sig­
nificance of the rook's position will be­
come apparent later on.
1 a . . . l:!.ba
24
If White becomes too preoccupied with
his kingside attack, then Black can open
up the queenside with ... b6-b5.
1 9 h5
'I was much more concerned about a
semi-waiting move on the queenside, 19
b4 say, when I would have to decide
whether to carry out my "threat" to initi­
ate queenside action - or much more
likely find a waiting move myself.' -
Speelman.
1 9 . . . lZJfa 20 h6
Lobron must keep going forward or
Black will establish a blockade. For in­
stance, 20 g4 h6 followed by ... tt:lh7.
20 . . . g6
21 'Llg7
White sacrifices a pawn to open some
lines, but he doesn't quite manage to make
a decisive breakthrough. It is close though.
21 . . . ..txg7 22 hxg7 'it>xg7 23 ..th6+ Wga
24 f4
'Now we can see why the rook would
have been better on f1 than g1.' - Sped­
man.
24 . . . f6 25 f5 J:!.e7 26 lZJd 1 ..txa4 27 lZJe3
.tea 2a lZJg4
After 28 .i.xf8 'it>xf8 29 tt:lg4 I!g7 30 l:Ifl
gxfS 3 1 tt:lxf6 f4 Black wins (Speelman) .
2a . . . lZJd7 29 .l:!.f1 g5 30 .txg5 l:lg7 3 1
..th6
Instead, Speelman recommends 3 1
tt:lh6+ �f8 3 2 .lth4 b S 3 3 b3! with the
 M a in L in e with 3 e 4 if.. b 7 4 if.. d 3 : Th e 'n e w ' 4
long-term plan of 'ith3, g3-g4, .l:i.hg1 and
g4-g5.
3 1 . . .lhg4 32 'iff3 J:!g5 33 if..x g5 fxg5 34
l!h6 l2Jf6 35 l:!fh 1 c5
So as to defend along the second rank
with the rook if necessary.
36 'ir'a3 a5
37 b4! ?
A n ingenious 'randomiser' i n his own
time pressure, though it still fails.
37 . . . 'ife7 !
37 ... cxb4? 38 "ii'c l! breaks the defence.
38 bxa5 bxa5 39 'iix a5 l!b2+ 40 �g 1
'ii' b 7 41 1:.6h2 l:.b 1 + 42 �g2 'i'b2+ 43
�h3 g4+ 44 �h4 h6 0-1
Speelman's system was successful
against Lobron's rhinoceros attacking
style, but with a steadier approach I would
say that White is doing well. He has a
space advantage and Black has no clear
method of counterplay. Nevertheless, I
hope that we haven't seen the last of this
system. I would recommend 10 ... .ltc5!? as
an improvement. At least it is active.
Game 1 0
Y rjola-K engis
Yerevan 1 996
1 d4 e6 2 c4 b6 3 e4 if.. b 7 4 if..d 3 l2lc6 5
l2le2 g6
Another invention of the Latvian
Grandmaster Edvins Kengis. Black doesn't
. . .ti:Jc6
go for the bishop on d3 with the knight,
but aims at a curious double-fianchetto.
6 l2lbc3
Orthodox development has to be
White's best policy - leave it to Black to
sort out the mess he has put his pieces in.
For that reason I'm not too keen on
White's strategy in Kaunas-Kengis, Riga
1995: 6 h4 .ltg7 7 J(.e3 hS 8 ti:Jbc3 tt::l ge7 9
�d2 dS 10 0-0-0 tt::l b4 1 1 cxd5 tt::lxd3+ 12
�xd3 exdS 13 f3 'ii'd7, when Black had a
promising position (two bishops and pres­
sure on White's centre) and his opening
has been completely justified (Black won
in 27 moves). However, there was no need
for White to give up the bishops and castle
on the queenside.
6 . . . if.. g 7 7 .ie3 l2lge7 8 l:!.c 1
8 a3 is a good alternative. Kachiani Ger­
sinska-Freckmann, Baden Baden open
1993, continued 8 ... a5 (8 ... d5!? 9 cxdS exdS
10 eS �d7 would have been more true to
the Kengis method of playing the open­
ing) 9 f4 fS 10 eS d6 1 1 0-0 dxe5 12 fxeS 0-0
13 J(.c2 �d7 14 J(.a4 .l:i:ad8 15 "Yi'e1 1i'c8 16
.l:Id1 and White had a dominating position
and went on to win.
8 . . . d5
I would love to play 8 ... e5 here but,
sadly, it just loses a pawn: 9 dS tt::l d4 10
.ltxd4 exd4 11 ti:JbS. 8...0-0 could be con­
sidered instead, though, keeping Black's
options open for a moment. If White cas-
25
En g lish D e fe n c e
des, 9 0-0, then 9 ... e5! ? 10 d5 l2Jd4 is possi­
ble as 1 1 ..ixd4 exd4 12 l2Jb5 c5 saves the
pawn. 1 1 f4 is stronger, when after 1 1...c5
(1 1...d6 12 f5 gives White a strong attack)
12 'ii'd2 White has the better prospects,
though Black's knight in the middle gives
him some counter-chances.
9 cxd5 exd5 1 0 e5 'i'd7 1 1 0-0 0-0
1 2 li:Jf4
Here a draw was agreed in Budnikov­
Kengis, Reykjavik open 1984.
Instead 12 f4 is critical:
a) 12 ... l2Jf5? 13 .ixf5 gxf5 (or 13 ... 'iVxf5
14 lLlg3 'ti'd7 15 f5 with a powerful attack)
14 lLlg3.
b) 12 .. .£5! 13 exf6 (or 13 a3 lLlaS!? with
chances for both sides) 13 ...ltxf6 14 'iVd2
when I prefer White's chances (I like his
three to two pawn majority on the king­
side), though having said that Black's
pieces are not too badly placed.
1 2 . . . a6
I don't know how necessary this move
is. As Black moves the knight on c6 on the
next turn, why not move it straightaway?
For instance, 12 ... l2Jd8 13 i.. e2 (13 l2Jb5
.ic6 is satisfactory for Black) 13 ... l2Je6 14
..ig4 .l:tad8 is a slight improvement on the
game, but I still prefer White's position -
he has pressure on the c-file and a juicy
pm.
1 3 i.. e 2 li:JdS 1 4 i.. g4 li:Je6 1 5 b4 l:tad8
1 6 li:Jce2
26
This allows Black to free his game at
the small cost of a pawn. 16 a4! would
have kept Black's bishop locked in behind
its own pawns and guaranteed White a
good game.
1 6 . . . i.. c 6! 1 7 'ifc2 Ji.b5 1 8 lZJxe6 fxe6 1 9
'ifxc7 lZJf5 20 'ifxd7
20 'iVxb6 .l:tb8 21 'iVaS lLlxe3 22 fxe3 h5
23 l:txf8+ .l:r.xf8 24 ..if3 �xf3 25 gxf3 i.xe2
26 l:c7 'i\Ve8 is okay for Black - he will get
enough counterplay on the kingside with
his queen.
20 . . . .t:!.xd7
It is difficult to make anything of the
extra pawn in the ending, and Black holds
the draw comfortably.
2 1 l:rfe 1 i.. c4 22 li:Jf4 �f7 23 a3 .!::.a S 24
Ji.e2 b5 25 liJd3 a5 26 lZJc5 J:!.da7 27 a4
Ji.xe2 28 .t:!.xe2 axb4 29 axb5 .t:!.bS 30
J:!.b2 .t:!.xb5 31 �1 l:.c7 32 l:.cb 1 i..x e5 33
lZJa6 J:!.a7 34 dxe5 1lxa6 35 .t:!.xb4 J:!.xb4
36 J:rxb4 lZJxe3+ 37 fxe3 g5 38 g4 �g6
39 h3 h5 40 �2 .t:!.c6 4 1 �f3 Y:z- Y:z
Kengis's idea is interesting (and cer­
tainly different) though I have to say I
remain sceptical. To my eyes the knight
does not belong on c6 when Black plays a
double fianchetto.
Game 1 1
Ahundov-Bagirov
Yerevan 1996
1 d4 e6 2 c4 b6 3 e4 i.. b 7 4 i..d 3 li:Jc6 5
d5
This is the only game that I have come
across where White plays this blunt push.
This advance is really what you are hop­
ing for if you are playing Black - it is eas­
ier to get to grips with the centre if it
lurches forward.
5 . . . li:Je5 6 Ji.e2
Provocative! White could still com­
promise with 6 lLlf3, when after Black
whips off the bishop chances are balanced.
6 . . . f5
 M a in L in e with 3 e 4 Ji.. b l 4 Ji.. d 3 : Th e 'n e w ' 4
I am a little surprised by this one. Black
exchanges off the e-pawn, but I would like
to attack it! For instance, 6 ... 'Yi'h4 is more
in the spirit of the English Defence as
played by the English, and now 7 'Llc3
.ltb4 (7. . . i.c5 8 g3 �e7 is also possible) 8
ikd4 d6 9 'Llf3 .ltxc3+ (9 ... 'Llxf3+!? 10 .ltxf3
e5) 10 bxc3 'Llxf3+ 1 1 .ltxf3 e5 12 'ii'd3 'Llf6
is about equal.
Apart from 6 ... 'i¥h4, Black also has
6 ... .lib4+ 7 .ltd2 Wke7 and 6 ... exd5 7 cxd5
'Llf6. In both cases White's centre is under
some pressure.
7 exf5 exf5 8 l:tJh3
Why not 8 'Llf3 ... ? After 8 ... .ltb4+ the
position looks about equal, but White
certainly isn't in any danger. In the game
Black is given more chances than he de-
serves.
8 . . . .Jtb4+ 9 l:tJd2
9 .ltd2 is more sensible. Black uses his
dark-squared bishop to good effect in the
game.
9 . . . 1:tJf6 1 0 a3 .Jtd6 1 1 0-0 0-0 1 2 l:tJf3
l:tJe4!
Black assumes the initiative, or at least
White is convinced enough to fall on the
defensive, and his position begins to take a
turn for the worse.
1 3 l:tJxe5 .Jtxe5 1 4 i.. d 3 ? !
14 f3 'Lld6 (14...'Llc5 is also fine) 15 .ltg5
.ltf6 16 iVd2 is about equal.
1 4 . . .'iff6 1 5 l:ta2 c6!
. . . 1:tJ c 6
That's a crucial move. If the bishop
doesn't break out from behind the pawn
chain then Black won't stand a chance of
getting an advantage.
1 6 f3 l:tJc5 1 7 i.. b 1 cxd5 1 8 .Jtg5 'iff7 1 9
b4 l:tJe6 20 cxd5 l:tJxg5 2 1 l:tJxg5 'ifh5 22
l:tJh3 i..d 6 23 :e 1 'ifh4 24 .l:.ae2
.
This is a classic demonstration of the
power of the two bishops. White can do
very little with his pieces because of the
monstrous piece on d6.
24 . . . a5!
Increasing the scope of said monster.
25 b5 .l:.ac8 26 i.a2 h6 27 f4 �h7 28 a4
.:.c3 29 �h 1 l:tf6 30 l:te3 .:.xe3 31 :xe3
l:.fS 32 l:tJg5+ �hS 33 g3 'ili'g4 34 �xg4
fxg4 35 l:tJe4 i.. b4 36 l:tJf2 l:tcS 37 .Jtb3
i.. c 5 38 .l:.e2 .Jtxf2 39 l:txf2 .l:.c3 40 i.. a 2
l:ta3 4 1 �g 1 �xa4 42 .l:.c2 :d4 43 �f2
.Jtxd5
... and there goes the final remnant of
White's lovely centre.
44 .!:[cS+ 'it>h7 45 .Jtb 1 + g6 46 f5 gxf5 47
i..xf5+ �g7 48 .Jtxd7 a4 49 l:tc7 �f6 50
�a7 �e5 51 l:ta6 �d6 52 .tc6 .txc6 53
l:txb6 �c5 54 l:txc6+ �xb5 55 l:!.xh6 a3
56 l:te6 a2 57 l:te 1 .:.a4 58 �e3 a 1 'if 0-1
Although Black won smoothly I didn't
find his treatment of the opening too con­
vincing. Nevertheless, 5 d5 does not pose a
threat to Black. Quite the opposite in fact
- I would be delighted to face it over the
board.
27
En glish D e fe n c e

Summary
4 ... tLlc6 has rightly taken over as the main response to 4 .td3. Against 5 tLlf3 Black
should not encounter any difficulties, as the first five games of this chapter demonstrate -
and that is still the most commonly played move. 5 lLle2 is a more severe test, and in
particular Game 7, Gelfand-Short, is worthy of close examination. Even though he got
an uncomfortable position, I liked Speelman's treatment of the opening in Game 9; I
hope someone else tries it. Kengis's double fianchetto in Game 10 is dubious; and 5 d5 as
in Game 1 1 is exactly what Black wants.

1 d4 e6 2 c4 b6 3 e4 iLb7 4 iLd3 lLlc6 (DJ

5 l2lf3
5 tLle2
5 ... tLlb4 (D)
6 0-0 lLlxd3 7 'it'xd3 tLle7 8 tLlbc3
8 ... g6 - Game 6
8 ... d6 - Game 7
6 lLlbc3 lLlxd3+ 7 'Yi'xd3 i.b4 - Game 8
5 . . e5 - Game 9
.

5 . . g6 - Game 10
.

5 d5 - Game 11
5 . . . l2Jb4 (DJ 6 0-0
6 tLlc3 lLlxd3+ 7 'ifxd3 i.b4 - Game 5
6 . . . lt:Jxd3 7 11ixd3 lt:Je7
7 ... d6 - Game 4
8 lt:Jc3 lt:Jg6 (DJ 9 d5
9 b3 - Game 2
9 .l:.e1 - Game ]
9 . . . 1Le7 - Game 1

4 . lLlc6
. .
5. . . lLlb4 8. . lZJg6
.

28
 CHAPTER TWO I
M ai n Li ne w ith 3 e4 i..b 7
4 ..td3: Other Fou rth M oves
for Black
1 d 4 e6 2 c 4 b6 3 e 4 �b 7 4 �d3
Let me remind you what I said in the
first chapter about Black's options after 4
i.d3: 4 .. .f5 is risky but fun (and is the sub­
ject of Games 12-19); 4 ... ..1tb4+ is a bit wet
but sound (you'll find that in Games 20-
23); and 4 ... 'iVh4 is just mad - see Game
24. Having considered 4 .. .'�Jc6 in the first
chapter, we can move on to look at these
alternatives. First of all:
.-------,
Game 12
Beliavsky-Short
Groningen 1997
._______________..
1 d4 e6 2 c4 b6 3 e4 �b7 4 �d3 f5 ! ?
This i s the consistent move. Black ex-
ploits weaknesses on the long diagonal and
immediately attacks White's centre. In the
early days of the resurgence of the English
Defence (the seventies and early eighties)
theoretical attention centred on this move
following investigations and games from
Miles, Plaskett and, originally, Basman.
When the variation proved to be dubious
(at least at international level) then alterna­
tives were explored, notably 4 .. .liJc6.
Nevertheless, the complexities of 4 .. .f5
were never fully resolved - there are still
many uncharted branches to explore, and
it is partially a measure of the success of
4 ... tZ:lc6 that the pawn push has fallen into
disuse - though of course it might just be
because it is unsound. It is interesting to
see Short returning to the move in this
game, particularly against such an illustri­
ous opponent, though the game was, I
believe, played at a fast time control, so
there were 'tactical' reasons for his choice.
5 exf5
The general feeling is that if White is to
get anything from the opening then he
must capture the pawn. In the early days
of the English Defence, when this move
often came as a shock, White players
would often 'decline' the offer of compli­
cations, preferring a continuation such as
29
En g lish D e fe n c e
5 ll'lc3 (Chapter 3 , Games 25-33), 5 'ii'h5+
g6 6 'iie2 (Game 18) or 5 d5 (Game 19).
They do not offer White anything special.
Nowadays if you see 4 .i.d3 in front of
you, you can be fairly sure that your op­
ponent will be ready and willing to plunge
into the mire - and he'll take you with
him.
5 . . . ..tb4+
5 ... .i.xg2 introduces random complica­
tions and is dealt with in Games 15-17.
6 lt>f1
Forced. The g-pawn must be protected.
Black has certainly gained by displacing
the king, though the bishop on b4 is out
on a limb and this provides tactical com­
fort for White. If the check is blocked,
then Black may capture on g2 with impu­
nity. For instance, 6 ll'lc3 .lixg2 7 'iih5+
�f8 8 fxe6 'iie 8 9 'iif5+ ll'lf6 10 d5 dxe6 1 1
dxe6 'ii'g6 12 'iixg6 hxg6 1 3 ll'lge2 .lixh1
and the extra rook came good in Lopez
Colon-Miles, Gran Canaria open 1996.
6 . . . lLJf6
To me, this developing move appears
the most natural (there is no need to close
the f-file with White's king at the end of
it) though 6 ... exf5 has also been played -
see Game 14.
7 i.e2!
'A very strong novelty that buries the
whole line. White's idea is 7 c5 bxc5 8 a3,
but it did not work yet because of 8 ... c4!,
30
so White prepares this deadly threat anci
combines it with moves such as 8 .ihS
and 8 .if3.' - Alexander Beliavsky and his
trainer Adrian Mikhalchishin writing in
New in Chess. Sadly, from Black's point of
view, I think their assessment of the posi­
tion is correct - though it is interesting to
note that Dreev declined to go in for this
line in a later game at the Elista Olympiad
(see the next game) . Dreev subsequently
wrote that he is not completely convinced
by 7 .ie2, but that leaves us none the
wiser!
Prior to Beliavsky's innovation White
had thrashed around with varying degrees
of success, though opinion had not crystal­
lised as to the strongest continuation. For
alternatives to 7 .ie2, see the next game.
7 . 0-0
. .
7... exf5 8 c5 bxc5 9 a3 .iaS 10 dxcS c6 is
good for White, though Black has grovel­
ling chances: the bishop goes back to c7,
followed by ... a7-a5 , ... .ia6, and there is
hope. If White's king were on a sensible
square then he would simply stand beauti­
fully, but on fl, it is enough to keep Black
gomg.
I thought about 7 ... .lid6 to solve the
problem of the misplaced bishop, but then
one of the points of Beliavsky's idea is
revealed: 8 .i.h5+! causes serious disrup­
tion:
a) 8 ... �e7 9 .ig5 with a clear plus for
White.
b) 8 ... tZ:lxhs 9 'iixh5+ �f8 10 .ig5.
c) 8 ... g6 9 fxg6 0-0 is an idea known
from the King's Gambit, but here it is just
too silly: 10 .ih6 wins.
Since writing this analysis, the follow­
ing game has come to light:
d) 8 .. :.tif8 9 fxe6 dxe6 10 .lif3 ll'lc6 1 1
ll'lc3 'iie 8 1 2 ll'lge2 �d8 1 3 'ii'a4 �f7 14
tZ:lb5 �f8 15 ll'lxd6+ cxd6 16 .lig5 �g8 17
.l:te1 'iig6 18 h4 ll'le4 19 .l:th3 tZ:laS 20 dS
tZ:lxg5 21 hxgS exd5 (21...'iixg5 22 ll'ld4) 22
cxd5 'ii'xg5 23 'iWe4 h6 24 b4 .l:tde8 25 'iid3
 M a in L in e w i t h 3 e 4 i.. b 7 4 i.. d 3 : O th e r Fo u r t h M o v e s fo r Bla c k
i.. c 8 26 �h5 �x£3 27 'ii'x£3 fid2 28 'ii'c3
and White was clearly better in Meessen­
Bunzmann, Leuven 1998.
8 c5
8 . . . bxc5
Julian Hodgson made an interesting at­
tempt at rehabilitating this whole varia­
tion when he played 8 ... tt:lc6!? in a recent
game, and his opponent duly trapped the
piece with 9 a3 (it would have been
stronger to play 9 tt:l£3!?, delaying the cap­
ture of the bishop for a move; in that case
I have great doubts that Black will have
enough for the piece: 9 ... bxc5 10 a3 .iaS
1 1 dxc5 tt:le4 12 b4 tt:lxb4 13 axb4 i.xb4 14
fxe6 and wins) . However, after the surpris­
ing sacrifice 9 ... tt:lxd4!? Black generated
some play with 10 axb4 tt:lxe2.
While I was working on this manu­
script two other games have appeared with
this line:
a) 1 1 "i!Vxe2 tt:ld5 12 fxe6 dxe6 13 b5 (13
tt:lf3!?) 13 ... a6 14 tt:lf3 axb5 15 l:.xa8 'i'xa8
16 h4 'i!Va1 17 'it'xe6+ 'it>h8 18 tt:le5 tt:lf6 19
tt:lf7+ 'iiitg 8 20 tt:le5+ 'iiith 8 21tt:lf7+ 'iiit g8 22
tt:ld6+ �h8 23 �e7 �g8 24 �e6+ �h8 25
'Wie7 Wg8 26 'i'e6+ 'iiih 8 27 tt:lxb7 'ik'xb 1 28
'i'e1 tt:le4 29 .l:th3 (29 f3!?) 29 ... l::i.xf2+ 30
'it>g1 I:tc2 3 1�f3 tt:lf6 32 'i'e8+ tt:lg8 33 ti:ld8
S.xcl+ 34 Wh2 h6 35 cxb6 cxb6 36 h5 .U.e1
37 I1e3 !Ih1+ 38 'it>g3 'i'£5 39 'i'g6 .U.fl 40
.:e8 fif4+ 0-1 Dautov-Tischbierek, Ger­
man Championship 1998. Mind boggling.
Obviously White could have taken a draw
by repetition on a couple of occasions, but
I suspect he was doing well anyway.
b) 1 1 tt:lxe2 tt:lg4 12 1Wd4 (much better is
12 tt:lg3! 'ii'h4 13 'iid2 ex£5 14 'fif4 !Iae8 15
h3 'i'e7 16 .id2 tt:le5 17 .l:txa7 and White
was on the way to consolidating his extra
piece in Vaisser-Sulava, Corsica 1998)
12 ... 'ii"h4 13 h3 l:tx£5 14 g3 'ii'e7 15 hxg4
J::i.d5 16 1:th5 l:i.xd4 17 tt:lxd4 'i!Vf6 with a
complicated struggle ahead in Ippolito­
Hodgson, Mermaid Beach 1998. It will
not be easy to contain Black's queen, par­
ticularly with the presence of opposite­
coloured bishops.
9 a3 i..a 5
9 ... cxd4, 10 axb4 e5 11 tt:lf3 should be
good for White as the pawns can't really
advance.
1 0 dxc5 lLld5
10 ... c6? saves the bishop, but 11 b4 .ic7
'leads to a completely lost position' - Be­
liavsky/Mikhalchishin. Of course, it is not
terribly good for Black, but he can fight
on (as in a similar variation above) with
... a7-a5.
A burst of activity starting with
10 ... tt:le4 does not reap rewards: 1 1 b4
.l:i.x£5 (or 1 1.. .1Wf6 12 .Ua2 tt:lx£2 13 'iitxf2
'ii'x£5+ 14 tt:lf3 'ii'xb 1 15 .Ub2) 12 tt:lf3 'if£6
13 .l:Ia2 i.d5 14 �b2 and White has every­
thing under control.
1 1 lLlf3
Fine, but Beliavsky actually considers
1 1 fxe6 to be even stronger. The idea is
that if 1 1...'i'Hf6 then 12 tt:lf3 'with a huge
advantage' - Beliavsky. At least Short now
gains some compensation for the piece,
though Beliavsky is still in control.
1 1 :xt5 1 2 b4 lLlxb4 1 3 axb4 i.xb4 1 4
. . .
i.b2 a 5 ! ?
If 14 ... .ixc5 Stohl recommends 15
tt:lbd2 with the idea of i.d3 and 'it'c2;
whereas Beliavsky favours the attractive 15
.i.ta4! and !Ig4. Black is going to be hard­
pressed to defend on the kingside.
31
En glish D e fe n c e

1 5 h4 33 ...tt:Jxb2 34 "ii'g8+ �g6 (or 34 ... �e7 35

Afterwards Beliavsky felt that it might 'iWe8 mate) 35 'ii'e8+ �g5 36 Wih5 is mate.
have been stronger to play either 15 tLla3 34 'iith 3!
or 15 tt:Jbd2 followed by either tLlc2 or
tLlc4.
1 5 . . . .l:!.d5 1 6 �b3 l2Ja6 1 7 lLlc3 l2Jxc5 1 8
'ifc2 l:tf5 1 9 J:.h3 �e7 20 �g 1 l:taf8 2 1
l:tf1 d6 22 l2J g 5 h 6
Short makes a practical decision to give
up more material so as to wrest the initia­
tive away from White - and he almost
succeeds in turning the game around. Ob­
jectively though, White must still be win­
ning. A full examination of this tremen­
dous game is beyond the scope of this
book, though I've added a few light notes
to indicate some of the major turning 34 . . Jbf3 + ?
points. There was a debate in the magazme
23 g4 hxg5 24 gxf5 l:txf5 25 .ltg4 New in Chess over the following variation
Perhaps 25 f4!? gxf4 26 .lig4 - Beliav­ of Beliavsky's: 34 ...l:txg4 35 'ii'g8+ �g6 36
sky. 'ii'e8+ �f5 37 fxg4+ �e4 38 'ii'xe6+ (38
25 . . . :t4 26 f3 gxh4 27 'ifh2 i-xf6 [Mikhalchishin] 38 ... tt:Jf4+! 39 .ttxf4+
If 27 'i!Vg6 e5 28 'ili'h5 g5 with compensa­ �xf4 and Black wins - Hebert in New in
tion - Beliavsky. Chess) 38 ... i-e5 39 i.xe5 dxe5 40 �g3 'i¥c3
27 . . . 'ii'g 5 28 l:txh4 l2Jd3 29 .l:th8+ �f7 41 'iff5+ �e3 42 Ith2 and wins. This varia­
tion of Beliavsky and Mikhalchishin is, at
the time of writing, the final word.
35 .ltxf3 lLlf4+ 36 'it>g3 'ifxb2 37 �g8+
'it>g6 38 �h7+ 'iitf7 39 'ifg8+ �g6 40
'ire8+ 'ilo>f5 41 .ltxb7 g5
Or 41...1i'c3+ 42 i.f3 and White is win­
mng.
42 l:th6! 'i'c3+ 43 i.f3

30 "irg3 !
30 'ii'h 5+ 'ii'xh5 31 i-xh5+ g6 32 l:th7+
�g8 33 i-xg6 tLlxb2 34 .l:i.xc7 is equal -
Beliavsky.
30 . . . �c5+ 3 1 'it>h2 .ltxc3 32 �h4!
Not 32 i-h5+? �e7 33 .lixc3 'ii'xc3 and
Black is on top.
3 2 . . . ..tf6 33 'ilfh7 "ifc2+

32
 M a in L in e with 3 e 4 i.. b 7 4 i.. d 3 : O th e r Fo u r t h M o v e s fo r 8/a.c k

43 . . . l2Je2 + Plaskett, Copenhagen 198 1.

Or 43 ... g4 44 l:rxf6+! �xf6 (44 ... �xf6 45 b) 7 fxe6 dxe6 8 �e3 0-0 9 h3 lt::lbd7 10
�h8+) 45 .lixg4+ �g5 46 'ikg8+ 'i'g6 47 lt::lf3 e5! 11 lt:'Jxe5 lt:'Jxe5 12 dxe5 lt:'Je4 13
\Wd8+ and wins. �g1 ltlxf2 14 .lixh7+ Wxh7 15 Wih5+ 'iii>g 8
44 'it>g2 tLlf4+ 45 'iii> h 1 d5 46 "Wiit7 1 -0 16 �xf2 l:r.xf2 17 Wxf2 �d4+ 0-1 Klein­
46 ... a4 (or 46 ... g4 47 �g2) 47 .l:!.xf6+ platz-Lempert, Hyeres open 1992.
�xf6 48 �g4+ We5 49 l:re1+ is terminal.
7 �e2 is undoubtedly a powerful new
move. Objectively, I feel that White
should emerge from the opening with a
clear advantage. Nevertheless, there are
certain factors which cloud the issue. First
and foremost is the fact that White's king
stands on a poor square. That gives Black
additional time to develop his pieces and
he may be able to go on the attack. Sec­
ond, the position is highly unorthodox
and more rigid players might not be able
to get a grip on the position. Even Beliav­
sky almost lost this game. So has this en­ 7 . . . bxc5 8 a3 c4
tire variation been put out of business? I 8 ... �a5? 9 dxc5 followed by b2-b4 is
don't think so. My advice is to pick your unpleasant.
opponent and situation carefully. 9 i..xc4 i..a 5

Game 13
Dreev-Sha balov
Elista Olympiad 1998

1 d4 e6 2 c4 b6 3 e4 i.. b 7 4 i.. d 3 f5 ! ? 5
exf5 i.b4 + 6 'it>f1 tLlf6 7 c5
The most popular move before Beliav­
sky's 7 �e2 turned up.
Other moves for White are less effec­
tive:
a) 7 .ltg5 0-0 8 a3 �d6 9 ltlc3 exf5!? (if
9 ... ltlc6 10 lt::l f3 lt::l e7 1 1 fxe6 dxe6 12 'ife2
"iVd7 13 .l::te 1 h6 14 �xf6 .l:f.xf6 15 lt::l e4 and 10 tLlf3
White was clearly on top in Goldin­ The critical test is 10 fxe6 dxe6
Gofshtein, Rishon Le Zion 1997) 10 d5 (10 ... �b6 is too much after 1 1 exd7+
(what about 10 �xfS!? - it is counterintui­ 'ii'xd7 12 lt:'Jf3 lt:'Jc6 13 'ii'e2+ 'ii>f8 14 ..te3
tive to open the f-file, but I don't see what when White was clearly better and went
Black's response should be; White has on to win in Arlandi-Sulava, Asti open
opened useful diagonals towards Black's 1995) 1 1 'iVa4+ (1 1 �xe6?! 'Wie7 12 d5
king) 10 ... h6 1 1 �d2 lt:'Ja6 12 b4 c6 13 dxc6 lt:Jxd5! 13 �xd5 ..txd5 14 ltle2 �f7 gave
dxc6 14 'ii'b 1 �xb4 15 axb4 lt::lxb4 with a Black excellent compensation for the
powerful initiative for Black in Brondum- pawn in Sideif Zade-Lempert, Nabereznye

33
En g lish D e fe n c e
Chelny 1993) 1 1...lZ:lc6 1 2 lt:Jf3 (12 .tb5?!
is weaker: 12 .. .'i'd5 13 CZJc3 i.xc3 14 bxc3
0-0 15 'i!Yc4 lZ:le4 with an initiative for
Black in Lauber-Czebe, Budapest 1997)
12 .. .'ii'd6 ? (12 ...0-0 is stronger, when Black
does have some compensation) 13 t'bc3
0-0-0 14 lZ:lb5 'iWd7 15 lt:Jxa7+ t'bxa7 16
'i'xaS and White won quickly in Dautov­
Teske, Vienna open 1996.
1 0 . . . 0-0 1 1 .i.d2
In lnfonnator 72 Dautov suggests 1 1
t'bc3 and if 1 1 . . .d5 1 2 i.d3 with a clear
plus for White.
1 1 . . . li:lc6 1 2 .txa5 lt:lxa5 1 3 .ta2 lt:le4
1 4 b4
1 4 . . J:l.xf5! ?
An enterprising piece sacrifice. In his
annotations in lnfonnator 73 Dreev also
suggests 14 ... t'bc6!? 15 d5 lt:Je7 with an
unclear position.
1 5 bxa5 lt:lxt2! 1 6 �xf2 �h4+ 1 7 �f1
17 Wg1 would have been met by
17 ... i.xf3 1 8 gxf3 llb8! with a strong at­
tack - Dreev.
1 7 . . . .i.a6+ 1 8 'it>g 1 �xf3 1 9 gxf3 .liltS 20
li:ld2 'it'g5+ � - �
..-------..
Game 14
Seirawan-Schussler
Malmo 1 979
._______________...,.
1 c4 b6 2 d4 e6 3 e4 .tb7 4 .td3 f5! ? 5
exf5 .tb4 + 6 �1 exf5! ?
34
If Black is to play 5 ... .tb4+, then this is
the main alternative to 6 .. .'�Jf6. My per­
sonal view is that 6 ... lt:Jf6 is a more dy­
namic continuation (it appears natural to
keep the centre fluid when White's king is
stuck there) but as that is under something
of a cloud following Beliavsky's novelty of
Game 12 it is perhaps worth examining
this older move again.
7 c5!
This ruse to misplace the bishop should
be familiar by now from the last two
games. Alternatively, I cannot imagine too
many players going in for 7 i.xf5. If noth­
ing else, it would give Black considerable
encouragement to see the f-file opened.
7 . . . bxc5 8 a3 c4!
8 ... i.a5 would be far worse according
to Seirawan, and I believe him: 9 dxc5 c6
(or 9 .. .'ii'f6 10 .l:ta2 c6 1 1 b4 i.d8 12 l:te2+
CZJe7 13 i.b2 with a big plus for White) 10
b4 i.c7 11 i.b2 tiJf6 12 iie2+ and White
has a clear advantage.
9 .txc4 .td6
Black has managed to bring the bishop
back to a sensible position, but White has
gained a great deal over the last few moves.
Most notably, the a2-g8 diagonal is wide
open - if Black could retract one move
then he would pick up the f-pawn and put
it back on f7 - but passing over illegal
moves, we can say that Black is suffering
from chronic weaknesses. Even White's
 M a in L in e with 3 e 4 i. b 7 4 i. d 3 : O th e r Fo u r t h M o v e s fo r Bla c k
displaced king (normally the straw to
which Black clutches) is feeling happier
with the f-file closed and the e-file open; a
rook can suddenly land on e1 to join the
attack.
1 0 lt:lc3
10 'ii'b 3 was tried in Christensen­
Nielsen, Danish Championship 1992, and,
while not as powerful as the game con­
tinuation, it does also have its merits:
10 ... .i.a6 1 1 .i.xa6 tt.Jxa6 12 ti.Jf3 (if 12 't!Vb5
"i!Vc8 13 'it'xf5 ti.Jf6 then Black has compen­
sation for the pawn) 12 ... c6 (12 ... ti.Jf6 did
not fare well in Summerscale-Williams,
British Championship, Torquay 1998: 13
ti.Jc3 c6 14 .i.g5 'ii'b 6 15 I:!.e1+ 'it>d8 16
.i.xf6+ gxf6 17 1lfc2 and White was clearly
better) 13 ti.Jc3 "iib 6 14 'ilic4 ti.Jc7 15 g3
'ii'a6 16 'ii'xa6 ti.Jxa6 17 d5 ti.Je7 18 dxc6
dxc6 19 �g2 and although the game was
drawn, White does have a small but defi­
nite advantage here. By the way, 10 ii.xg8
doesn't bring any rewards: 10 ... ii.a6+ 1 1
We1 "ti'e7+ and Black assumes the initia­
tive.
1 o . . . lt:lf6 1 1 li:lf3
1 1 ti.Jh3!? also isn't bad: 1 1 ...ti.Jc6 12
.i.f4 ti.JaS 13 .i.a2 .i.a6+ 14 �g1 .i.c4 15
ii.xc4 ti.Jxc4 16 'ii'e2+ 1-0 was a bit of a
disaster for Black in Kaiser-Carton, Ger­
man Bundesliga 1993.
1 1 .. :fie7
After this game Seirawan recommended
1 1...ti.Je4!?, but 12 'iib3! is powerful. For
instance, 12 ....i.a6 13 i..xa6 ti.Jxa6 14 'iib 5
"i�Vc8 15 'ii'xf5 and Black is a pawn down
for nothing.
1 2 i.g5! <3o>d8
The king was never going to get to the
kingside, so it nudges out of the line of fire
on the e-file - but into the line of fire of
the bishop on g5.
1 3 li:lh4! ?
Crafty. Black is forced to play ... g7-g6
and so the diagonal is weakened, and the
bishop on g5 gains in power. 13 ti.Jd5!?
.i.xd5 14 .i.xd5 c6 15 'iih 3 is also strong.
1 3 . . . g6 1 4 d5
It is difficult for Black to move any of
his pieces after this one.
14 . . . i.c5 1 5 lt:la4! i.xf2
15 ... .i.b6 is stronger, but that just
shows the poor state of Black's position:
16 ti.Jxb6 axb6 17 d6 cxd6 18 'iid4.
1 6 lt:lf3 i.b6 1 7 lt:lxb6 axb6 1 8 d6 cxd6
1 9 'i'd4 �f8 20 �e 1
White could have won immediately
with 20 "i!Vxb6+ �c8 21 .l:i.e1 'it'd8 (or
21....i.e4 22 ti.Jd4 .l:i.e8 23 ti.Jb5 i..d3+ 24
�f2 ti.Jg4+ 25 �g3) 22 'ii'xd6 ti.Jc6 23 .l:i.e6!
20 . . . i.e4 21 'ifxb6+ �ea 22 i.f4 lt:lc6 23
Wic7 lt:le5 24 lt:lxe5 dxe5 25 i.xe5 d5 26
'i!Yc6+ �f7 27 i.xf6 iixf6 28 i.xd5+
i.xd5 29 �xd5+ �g7
Black has survived at the small cost of a
pawn. White steers for a draw before his
35
En g lish D e fe n c e

own king lands in trouble. is difficult to get a handle on the positions

that arise - they are so far away from any­
thing that is 'normal' that reference to
similar positions doesn't come into it; it
simply isn't possible. In many variations
White gives up a whole rook for the at­
tack, while Black has a wobbly king and a
pawn on the seventh rank to deal with.
Great fun.
In recent years this variation has hardly
been seen as attention has turned towards
4 .. .tbc6. After games such as this one,
many players felt it was simply too
chancy to be subjected to such a violent
30 it'e5 "i'xe5 3 1 llxe5 I:.fb8 32 I:.e2 :l:.b3 attack. One false move and you get your
33 'it>f2 I:.ab8 34 lla 1 .l:txb2 35 a4 Wf6 head blown off. Nevertheless, I hope I will
36 aS :!:.2b5 37 llc2 J:ia8 38 J:ic6+ �e5 prove that there is still plenty of room for
39 a6 lta7 40 l::ta 3 'it>d5 41 �f6 V2 - V2 exploration, self-expression and discovery.
Unless a big improvement is found, And it is not only risky for Black: there
then in my opinion 6 ... exf5 is even worse are plenty of games where White has come
than 6 ... lt.Jf6. Black regains a pawn but to grief.
loses dynamism. Incidentally, 5 .. Ji'h4 was suggested by
Basman, but as Plaskett points out 6 t:Llf3
Game 1 5 is a good answer: 6 ... 'ii'g4 7 0-0 and White
Browne-Miles is simply better.
Tilburg 1 978 6 "i'h5+ g6
6 ... \t>e7 would be a fine move if it
1 c4 b6 2 d4 e6 3 e4 .i.b7 4 .i.d3 f5 ! ? 5 weren't for 7 'ii'g5+, winning the bishop
exf5 .i.xg2 on g2.
7 fxg6 .i.g7
Forced. 7 ... lt.Jf6 isn't quite good
enough: 8 g7+ t:LlxhS 9 gxh8'ii' lt.Jf6 (not
9 ... i.xh1 10 'ii'xh7 and wins) 10 i.h6 �f7
1 1 .txf8 'ii'xf8 12 ifxf8+ �xf8 13 t:Lld2
i.xh1 14 f3 t:Llc6 15 �f2 t:Llb4 16 i.b 1 c:Ji;e7
17 tbe2 l:i.g8 18 t:Llg3 i.xf3 19 Wxf3 and
White was winning, although Black man­
aged a swindle (0-1 in 39 moves!) in Jakob­
sen-Jepsen, Danish Team Championship
1990.
8 gxh7+ 'it>f8

see following diagram

If 5 ...i.b4+ had some semblance of san­
ity, then this move strips all that away. 9 tbe2!
The capture of the pawn leads the game This became established as the best
into random and bizarre complications. It move after this game (but note that 9

36
 M a in L in e w i t h 3 e 4 i.. b 7 4 i.. d 3 : O th e r Fo u r th M o v e s fo r Bla c k
i.g5+ lt:Jf6 10 'ii'h4 i.xh1 1 1 lt:Je2 comes to
the same thing) .
Prior t o this game 9 hxg8� + was
played, but Black had been doing well
after 9 . . . Wxg8 10 'it'g6 (10 'tlVg4 i.xh1 1 1
i.g5 'ii'e 8! is clearly better for Black, e.g.
12 h4 lt:Jc6 13 lt:Jc3 lt:Jxd4! and Black won
quickly in Lehmensick-Basman, British
Championship, Brighton 1972) 10 ... i.xh1
1 1 i.g5 'ii'f8 when Black holds a clear plus.
Nobody plays like this any more.
Alternatively, 9 lt:Jf3 was played in one
obscure game in England (Hazel-Hardy,
Peterborough 1979) but has never been
repeated. Plaskett considers that the best
continuation for both sides then is 9 ... lt:Jf6
10 'it'h4 i.xh 1 1 1 lt:Je5 d6 12 i.h6 dxe5 13
dxe5 'iixd3 14 'ii'xf6+ We8 15 'iixe6+ 'it>f8
16 'iif6+ �e8 with a draw by repetition.
9 . . i.. x h 1 1 0 �g5!
. i.xh7 i.xb2 20 'ilih5+ �f8 21 'i'g6 i.xa1
10 hxg8'ii' + �xg8 is still good for
Black; and 10 lt:Jf4 is met by 10 ... lt:Je7.
1 o . . . tt:Jt6
Not 10 ... lt:Je7? 11 lt:Jf4 'ii'e 8 12 lt:Jg6+
lt:Jxg6 13 i.xg6 'i&'c8 14 'i&'h4! and wins;
while 10 ... i.f6? 1 1 h4! looks good to me.
1 1 'iih 4 lt:Jc6
This natural developing move has been
played in the vast majority of games, but
there are two alternatives which are worth
exploring: 1 1...'ii'e 7!? and 1 1 ...i.f3!? For
these, see Game 17.
1 2 li:Jf4
For 12 lt:Jd2!, see the next main game.
1 2 . . . �f7 ?
Black gets walloped after this one, but
improvements were found later:
a) Not 12 ... lt:Jb4? 13 i.g6 fie? 14 lt:Jh5
tbc2+ 15 i.xc2 'ii'b4+ 16 lt:Jd2 lt:Jxh5 17
0-0-0 1-0 Forintos-J.Fernandez, Ca­
pablanca memorial 1979, Havana 1979.
b) Basman recommends 12 ... e5! 13
lt:Jg6+ �f7 14 dxe5 .i:txh7 and this does
look good for Black: 15 'il'f4 .l:.h3 16 exf6
i.xf6 17 i.e2 �xg6 18 'i&'g4 �h8 and wins
- a very 'Fritzy' variation!
c) 12 ... lt:Jxd4 is also playable: 13 lt:Jg6+
�e8 (or even 13 ...�f7 14 lt:Je5+ �f8 15
�xd4 [15 lt:Jg6+ �f7 16 tbe5+ 'lt>f8 17
ti:Jg6+ 'lt>f7 is a draw by repetition] 15 ... d6
16 tbg6+ 'it>f7 17 lt:Jxh8+ fixh8 with an
unclear position, e.g. 18 'i&'h4?! lt:Jxh7 19
22 �fl i.c6 23 h4 �b8 24 ti:Jd2 i.e8 25
'iVd3 i.f7 26 i.e4 'ii'c3 27 �e2 a6 28 i.d3
'ii'e5 29 .i.. e4 'iih2 30 �f3 �e5 3 1 i.f6
�xf6 0-1 Stassans-Sandler, Riga 1979) 14
'i'ixd4 �xh7 15 lt:Je5 and now John Nunn
states that Black should play 15 ... �xh2,
since Miles's 15 ... �h3 16 i.g6+ �f8 17
lt:Jc3 d6 18 0-0-0 i.b7 favours White ac­
cording to Ftacnik.
1 3 i.. g 6+ �e7 1 4 lt:Jh5 'ii'fS 1 5 li:Jd2
White's attack is now decisive.
1 5 . . . e5
Alternatively, 15 ... d5 16 0-0-0 i.e4 17
37
En g lish D e fe n c e
tt:Jxe4! dxe4 1 8 tt:Jxg7 Vi'xg7 1 9 .lixe4; or
15 ... tt::l b 8 16 0-0-0 .lib7 17 .l:te1 (or even 17
dS) with a winning position for White in
each case.
1 6 0-0-0 t2Jxd4 1 7 l:txh 1 t2Je6 1 8 f4!
18 .l:te1 tLlxgS 19 Vi'xgS �d8? 20 .l:i.xeS
�c8 2 1 .l:tfS. After the text White is win­
mng.
1 8 . . . d6 1 9 t2Je4 t2Jxg5 20 'ifxg5 i.h6 2 1
'ilfh4! i.g7 22 fxe5 dxe5 2 3 .l:!.f1 'it>d7 24
t2Jexf6+ i.xf6 25 tLlxf6+ 'it>c8 26 i.e4 c6
27 'ilfh3+ 'it>b 7 28 i.xc6+ 1 -0
28 ... �a6 (28 ... �xc6 29 Vi'd7+ �cS 30
'ilidS+ �b4 31 'fibS mate) 29 ..tbS+ �aS 30
a3 is hopeless for Black. A crushing vic­
tory, though not totally convincing. Im­
provements were later found - for White!
r------�
Game 16
Magerramov-Psakhis
Riga 1 980
1 c4 b6 2 d4 e6 3 e4 i.b7 4 i.d3 f5 ! ? 5
exf5 i.xg2 6 'ii'h 5 + g6 7 fxg6 i.g7 8
gxh7 + 'it>t8 9 i.g5 t2Jt6 1 0 'i!Vh4 i.xh 1
1 1 t2Je2 t2Jc6 1 2 tLld2!
This is superior to the 12 tt:Jf4 of the
prev10us game.
1 2 . . . e5
Basman's 12 ... b5!? was the most inter­
esting attempt to rehabilitate this varia­
tion. Black gains some control over the
dS-square, which gives his minor pieces a
38
crucial foothold in the centre of the boarc
13 cxbS! tt:Jb4 14 i;g6 .tb7 15 tt:Jf4 �e7 1
tt:Jhs Vif8 17 ds tt:JbxdS 18 tt:Je4 �d8 1
tt:Jexf6 �c8 20 tt:Jxg7 and the game Flea�
Plaskett, British Championship, Torqua:
1982, was drawn in 62 eventful move!
However, 20 tt:JxdS is stronger accordin
to Hardy and Basman: 20 ... .lixd5 21 tLlxg
'ii'xg7 22 .lid3 �b7 23 0-0-0 :hf8 24 .lih1
'tieS 25 .lixf8 :xf8 when Black is fightin:
hard but that pawn on h7 endures.
1 3 0-0-0!
The best. 13 tt:Jg3?! runs into 13 ... e4! 1·
.lixe4 .lixe4 15 tt:Jgxe4 :xh7 16 Vif4 tt:Jxd•
(16 ... Vi'e7! - Nunn) 17 .lixf6 (17 tt:Jxf6! -
Nunn) 17 ... i.xf6 18 tt::lxf6 Vie7+ 19 tt:Jde'
l:th4 20 tt:Jg4+ Vif7 2 1 Vig3 .l:Ie8 22 �f
'i&'xc4+ 23 �g2 tt:JfS 24 Via3+ d6 0-1 Akes
son-Short, World Junior Championship
Dortmund 1980.
1 3 . . . e4 1 4 i.xe4 i.xe4 1 5 t2Jxe4 l:!.xh7
1 6 'i!Vt4 'it>t7 1 7 tLl2c3
17 �g1 is also quite good: 17 ... Vih8 18
i.xf6 iLxf6 19 tt:JgS+ We7 20 'ti'e4+ �d6 21
tLlxh7 .lixd4 22 l:tg6+ �cS 23 'i*'d5+ 1-0
D.Cramling-Gausel, Gausdal 1982.
1 7 . . . l:!.h5 1 8 h4 t2Jb4 1 9 a3 d5 20 t2Jxf6
i.xf6 21 axb4 'ifd6 22 'ilfg4 llah8 23
tLlxd5 i.xg5+ 24 hxg5 .l:!.h 1 25 'it'f3+ 1 -0
It is this encounter, more than the pre­
vious main game, which has dampened
enthusiasm for 5 ... i.xg2. But what about
the next game ... ?
 M a in L in e w i t h 3 e 4 i.. b 7 4 i. d 3 : O th e r Fo u r t h Mo v e s fo r Bla c k
Game 1 7
V egh-Ziatilov
St Augustin 1990
1 d4 e6 2 c4 b6 3 e4 i.. b7 4 i.d3 f5! ? 5
exf5 i.xg2 6 'i'h5 + g6 7 fxg6 i.g7 8
gxh7 + <;t>t8 9 tLle2 i..x h 1 1 0 i.. g 5 tLlt6 1 1
li'h4 'Wie7
New move! Most players have opted
for 1 1...tt:lc6 instead (as in the previous
two games) but Mr Zlatilov has had a little
think about the position. 1 1 . ..i.f3!? has
been suggested by the English player Otto
Hardy, but to my knowledge never tried
in practice. It is well-motivated. The
bishop comes back into play, threatening
to exchange itself for one of White's main
attacking pieces, and if the knight moves,
then White is unable to castle on the
queenside. Until we see a few games with
it the jury remains out.
1 2 tLlf4
12 tt:ld2!? followed by castling queen­
side is critical.
1 2 . . :ilif7 1 3 tLld2
Alternatively, 13 i.g6 :xh7 14 i.xh7
tt:lxh7 15 'i¥xh7 i.xd4 looks okay for
Black; and 13 tt:lg6+ �e8 14 tt:le5 'i'e7 (or
even 14 . . .'ii'f8!? 15 ..li.g6+ 'iit> d8 16 tt:lg4
'iit> c8) 15 i.. g6+ 'iit>d 8 16 tt:lg4 'iit>c8 17 tt:lxf6
�b4+ 18 tt:ld2 ..Wxb2 19 l::tb 1 ..Wc3 is totally
unclear.
1 3 . . . tLlc6 1 4 i.. g 6
14 0-0-0! ? was perhaps an improvement.
1 4 . . . .U.xh7
This move, returning the rook, must
have come as a shock to White:
1 5 i.. x h7 tLlxh7 1 6 'i!Vxh7 tLlxd4 1 7 tLlg6+
<;t>e8 1 8 tLlh8 tLlf3+ 1 9 <;t>t1
Or 19 tt:lxf3 i.c3+.
1 9 . . . tLlxg5 20 'ifh4 tLlf3 0-1
I do not present this as the answer to all
Black's problems, but it is evidence that
there is far more to the rook sacrifice than
meets the eye. Hardy's 1 l . ..i.f3 must also
be worth a try. Go on, be brave; and
above all, pick the right opponent!
Game 18
T oth-Ornstein
Oslo 1978
1 d4 e6 2 c4 b6 3 e4 i.. b 7 4 i.d3 f5 5
'i'h5 +
By flicking in the check White hopes to
create a weakness on the kingside, but in
this game Black makes a virtue out of it. 5
..We2 tt:lf6 6 i.g5 ..tb4+ (6 ... h6!?) 7 tt:lc3
fxe4 8 i.xe4 i.xc3+ 9 bxc3 i.xe4 10 Jt.xf6
'i'xf6 1 1 ..Wxe4 tt:lc6; as in Sosonko-Keene,
Haifa 1976, also leads to a satisfactory po­
sition for Black (the game was drawn in 26
·
moves).
5 . . . g6 6 'ife2 tLlf6 7 i.. g 5 h6
More adventurous than 7... fxe4 8 Jt.xe4
39
 M a in L in e with 3 e 4 JJ.. b 7 4 JJ.. d 3 : O th e r Fo u r t h M o v e s fo r Bla c k
exd7 lLlxd7 1 3 0-0-0 lLld4 1 4 'i'd 3 lLlc5
1 5 'i'f1 :adS 1 6 .i.xb 7 :xt2 1 7 lLlgf3
lLlxf3 1 8 lLlxf3 'ilff4+ 1 9 �b 1 :xt 1 0-1
Another one that didn't make it past
move 20! This is typical of the mess that
an unprepared White player can make of
his position when faced with the shocking
4 .. .f5!?
Game 20
Jo. Horvath-G u l ko
Nova Gorica 1997
1 d4 e6 2 c4 b6 3 e4 JJ.. b 7 4 .i..d 3 JJ.. b4+
After the chaos of 4 .. .f5!? it is almost a
relief to return to tranquil waters with
4 ... ..ib4+. Naturally, there are different
interpretations of the move, but for the
most part this is a solid option. By going
for an exchange of bishops Black eases
some of the congestion in his position.
The danger with this approach is that it
robs Black's position of some of its dyna­
mism; and if he is unable to challenge
White's centre successfully, he could find
himself getting squashed.
5 .i..d 2 !
5 tbc3 transposes to Chapter 3, Games
25-33; and 5 tbd2, to my knowledge, has
never been played. Both moves invite
5 .. .f5 ! Finally, 5 'it>fl is dealt with in Game
23.
5 . . . 'i'e7
For me, this is the best option, but
S ... l2Jc6 and S ... ..ixd2+ are both possible ­
see Games 2 1 and 22.
6 lLlc3
At first I thought that White could
count on some advantage if he simply
went for exchanges with 6 ..ixb4!? 'ifxb4+
7 ii'd2 "iixd2+ 8 tbxd2, but 8 .. .f5, as is of­
ten the case, nibbles away at the centre
and gives Black enough counterplay to
equalise: 9 tbgf3 fxe4 10 tbxe4 tbc6! 1 1
Wd2 (1 1 0-0-0 tbb4) and now all three
knight moves, 1 1 . ..l2Jge7, 1 1...tbf6 and
1 1 . ..l2Jh6, give Black a reasonable game.
On the other hand, if 6 tbe2 then Black
should take the opportunity to get in
6 .. .f5! (6 .....ixd2+ 7 ii'xd2 tbh6 8 tbbc3 0-0
9 0-0 d6 10 f4 l2Jd7 1 1 dS eS 12 fS f6 13 b4
gave White a massive space advantage in
Avrukh-Speelman, Elista Olympiad 1998)
with all the usual complications: 7 exf5
..ixg2 8 �g1 ..ib7 (8 ... ..if3!? is worth con­
sidering; compare with Schneider-Forintos
later on) 9 tbbc3 tbc6 10 fxe6 dxe6 11 a3
..ixc3 (1 1.....id6!?) 12 ..ixc3 l2Jf6 13 'ii'c2
0-0-0 14 0-0-0 is more comfortable for
White, but Black effortlessly notched up a
victory anyway in Vasiliev-Lempert, Mos­
cow 1990.
Finally, 6 t2Jf3 tbf6 (but why not
6 .. .f5 ... ? If White captures then his king
has an awkward check to deal with: 7 exf5
exf5+� 7 a3 ..ixd2+ 8 tbbxd2 d6 9 0-0 eS 10
d5 0-0 11 b4 t2Jbd7 12 "iic2 and, although
Black's position is solid enough, White,
with his space advantage, had a small but
comfortable advantage in Malisauskas­
Litus, Katowice 199 1.
6 . . . f5!
Alternatively, 6 ... ..ixc3 was tried in
Poltl-Ta.Horvath, Gleisdorf 1989, though
I wouldn't be keen on giving up the
bishop pair so readily: 7 i..xc3 f5 8 'tifhS+!
g6 9 �e2 li:Jf6 10 d5 g like the way Black
has attacked the centre with .. .f7-f5 but
would be concerned about the bishop on
41
En g lis h D e fe n c e
c3) 1 0. . .'�Ja6 1 1 f3 (too tame; 1 1 exfS!)
1 1. ..fxe4 12 fxe4 0-0 13 0-0-0 tt:lcS 14 tt:lf3
exdS 15 exdS iixe2 16 i.xe2 tt:lce4 17 .i.d4
cS 18 dxc6 dxc6 19 .l:the1 cS when Black
has equalised and a draw was agreed.
7 lZJge2 fxe4
A solid move, breaking down White's
centre and trading pieces. Gulko equalises
by precise play. 7 ... tt:lf6!? leads the game
into more random territory, e.g. 8 exfS
.i.xg2 9 �g1 .i.f3 10 l:tg3 .i.b7 1 1 fxe6 dxe6
12 'it'a4+ .i.c6 13 'iic2 i.d6 lh-lh Schneider­
Forintos, Hungarian Championship 1979.
A pity. The game was just starting to
warm up.
8 .i.xe4 .i.xe4 9 lZJxe4 lZJt6 1 0 lZJ2g3
10 tt:lxf6+ gxf6 would have unbalanced
the position. Black would castle on the
queenside and hope for the best, and
White would do the same on the kingside.
1 o . . . lZJxe4 1 1 lZJxe4 �h4 ! ? 1 2 lZJg3
12 tt:lf6+?! gxf6 13 i.xb4 �c6 14 i.. c3
'i¥e4+ is irritating.
1 2 . . . ..1td6 1 3 �h5+ �xh5 1 4 lZJxh5 0-0
1 5 lZJg3 lZJc6 1 6 .i.e3
The game is balanced - but not drawn.
Gulko finds a way to create some chances.
1 6 . . . lZJb4 1 7 �d2
By keeping White's king in the middle
there are more tactical possibilities open to
Black.
1 7 . . . �f4 1 8 .l:hc 1 .:.ae8 1 9 .bf4 :xt4
20 �e3 �h4 2 1 h3 d5 22 c5
42
After 22 a3 tt:lc6 the king is caught in
the crossfire of Black's rooks.
22 . . . b5 23 c6 e5 24 a3 lZJa6 25 dxe5
:xe5+ 26 �d3 �c4 27 .:.xc4 dxc4+ 28
'it>d4 :e6 29 a4 :d6+ 30 'it>c3 J:ld3+ 3 1
�c2 lZJb4+
3 1 ...b4!? followed by ... b4-b3+ and
... tt:lb4 is dangerous.
32 �c1 c3 33 bxc3 :xc3+ 34 'it>b2 .:.c5
35 lZJe4 l;tc4 36 f3 lZJd3+ 37 'il.ob3 lZJc 1 +
3 8 'it>b2 bxa4 39 l:!.a3 l::tx c6 4 0 .l;txa4
l2Jd3+ 41 'iit a 3 a6 42 g4 h6 43 l:!.d4 lZJe 1
44 l;td8+ 'it>h 7 45 f4 l;te6 46 l2Jc5 :e3+
47 �a2 �g6 48 .:d7 .=.xh3 Y2 - Y2
After 49 �e4 White has enough coun­
terplay to draw. Gulko's opening play was
sound, and it was fascinating to watch him
create something out of nothing in the
ending, even though it wasn't quite
enough to win.
Game 21
Nogue i ras-Velez
Cienfuegos 1983
1 e4 e6 2 d4 b6 3 c4 .i.b 7 4 �d3 i.. b4+
5 i..d 2 l2Jc6
.
This idea turns out well here, though I
do harbour some doubts as to its effec­
tiveness.
6 lbt3 .i.xd2+ 7 l2Jbxd2
After 7 'iixd2 Black could play solidly
with 7 ... d6 and ... e6-e5, but that doesn't
 M a in L in e w i t h 3 e 4 i.. b 7 4 i.. d 3 : O th e r Fo u r t h M o v e s fo r Bla c k

challenge White's central position. Instead

7 .. .'�f6!? is more interesting, e.g. 8 eS (8
dS!?) 8 . . .'ti'e7 9 t'Llc3 f6 with some pressure
on the centre.
7 . . .'iVf6 8 e5
8 dS is more testing, e.g. 8 ... t'Llb4
(8 ... t'Lld4!?) 9 i.e2 'ili'xb2 10 0-0 t'Lla6
(10 ... t'Llc2 1 1 �b 1 'i'xa2 12 �d3 is clearly
better for White, while after 10 t'Llxa2 1 1
.•.

l::i.b 1 'i' a3 12 l:Ib3 'i¥a4 1 3 cS White has

good compensation for the pawns) 1 1
t'Llb3 when White has compensation for
the pawn, though Black's position is play­
able. I prefer Black's position. The pin is un­
pleasant for White, and there are chances
to attack on the kingside.

Game 22
Petursson -Wauthier
San Bernardino open 1991

1 d4 e6 2 c4 b6 3 e4 i.. b 7 4 i..d 3 i.. b4 +

5 i..d 2 i..x d2 +
I feel that this makes life too easy for
White, but it isn't a bad move.
6 t:Dxd2
After the solid 6 'ili'xd2 fS 7 t'Llc3 t'Llf6 (I
8 . . .'it'f4! prefer 7 ... fxe4 8 i.xe4 i.xe4 9 t'Llxe4 t'Llf6
A nice finesse, forcing White to weaken and Black is close to equalising) 8 f3 fxe4 9
himself on the long diagonal. fxe4 d6 10 t'Llf3 t'Llbd7 1 1 0-0 'iie7 White
9 g3 �h6 1 0 0-0 f5 had a slight plus in Kaufman-Mouzon,
I would prefer 10 .. .f6 so that the f-file Atlantic open, Washington 1997.
can be used.
1 1 d5 t:Dce 7 1 2 �e 1 exd5 1 3 cxd5 i..x d5
1 4 �c1 t:Dc6 1 5 .txf5 t:Dge7 1 6 .te4 .tf7
17 h4
17 t'Llc4! followed by t'Lld6 would have
given White a clear advantage according to
Vilela in Infonnator 35.
1 7 . . . 0-0 1 8 t:Dg5 .tg6 1 9 .tg2 �h8 20
e6
20 t'Llde4! would still have been at least a
little better for White.
20 . . . d5 21 i..x d5 t:Dxd5 22 �xc6 i.. h 5 23
'i'b3 �f6 24 t:Dde4 �f5 25 �c2 h6 26 e7
!?lxe7 27 t:De6 .tf7 V2 - V2 6� . . t:Dh6

43
En g lis h D e fe n c e

6 . . .f5 is more forthright: 7 ii'h5+ g6 8 l:If4 tt:le7 20 IIxf7 ttJf5 21 ttJf4 IId7 22
ii'e2 tt:lh6 9 tt:lgf3 'ii'f6 10 0-0-0 tt:lc6 1 1 e5 !!xd7 'it>xd7 23 tt:le4+ ct;e7 24 IId3, and
'i!le7 12 a3 ttJaS 13 h3 c5 14 !Ihe1 0-0 15 now if Black had played 24 ... tt:lh4! 25 tt:ld2
'iit>b 1 !Iac8 16 g3 'iit> g7 17 'iit>a2 .l:i.c7 18 l:tcl .l::tf8 26 tt:lh5 tt:lxf3 27 tt:lxf3 .l:i.xf3 28 ltxf3
IIfc8 with a balanced position in Forintos­ i.xf3 29 tt:lf4 the game would have ended
Chetverik, Zalakaros open 1996. in a draw.
However, 6 ... d6 isn't terribly inspired: 8 . . . lL'lc6 9 e5 'i'f4 1 0 g3 'ii'g4 1 1 'i'xg4
7 tt:le2 e5 8 d5 tt:le7 9 0-0 tt:ld7 10 b4 a5 1 1 1Llxg4 1 2 0-0 f6 1 3 h3 1Llh6 1 4 exf6 gxf6
a3 and White had a typical advantage for 1 5 a3 1Llf7 1 6 b4 0-0-0 1 7 �fe 1
this opening in Van der Vliet-Ree, Am­
sterdam 1983 .
7 �h5
If White develops normally, then 7
tt:lgf3 0-0 8 0-0 f5 with the usual kind of
play for Black, as in Byrne-Sandler, Aus­
tralian Championship 1995.
7 . . .'iif 6! ?
7. . 0-0 8 tt:lgf3 fS 9 0-0 tt:lc6 is also play­
.

able - the queen isn't too threatening on

h5.

White has some space on the queenside,

but Black's position is solid.
1 7 . . . 1Llg5?
It was better to play 17 ... IIdg8! with
chances for both sides.
1 8 l2Jxg5 fxg5 1 9 d5 exd5 20 cxd5 lL'lb8
21 d6 J:tde8 22 .ixh7

a · lL'lgf3
Alternatively, 8 eS Vi'f4 9 tt:le2 l!Vg4 10
'iii'xg4 tt:lxg4 11 f3 tt:lh6, as in Flear­
Wauthier, San Bernardino open 1991, is
quite similar to the game, but here White
played the outre 12 g4 creating profound
weaknesses in his own kingside. Although
he won the game, I don't think it was
particularly convincing: 12 ... tt:lc6 13 0-0-0
0-0-0 14 l::th g1 d6 (perhaps 14 .. .f6!? 15 f4 White now has a clear advantage.
fxeS 16 dxeS gS) 15 g5 tt:lg8 (or 15 ... tt:lf5!? 22 . . . .l:!.e6 23 �xe6 dxe6 24 .te4 .id5 25
16 .it.xf5 exf5 17 exd6 IIxd6) 16 exd6 IIxd6 dxc7 'iitx c7 26 .ixd5 exd5 27 lL'lf3 I:.xh3
17 IIg4 tt:le5 (17 ... tt:lge7) 18 dxe5 IIxd3 19 28 1Llxg5 .:lh6 29 f4 Wc6 30 l:.d 1 :td6 3 1

44
 M a in L in e with 3 e 4 i.. b 7 4 i.. d 3 : O th e r Fo u r t h M o v e s fo r Bla c k
lLlf3 Wd7 3 2 �f2 We7 3 3 g 4 a 6 34 lLlh4
<M7 35 l2Jf5 l:l:d8 36 llc 1 .l:th8 37 �c7+
�6 38 lLld4 J:ld8 39 �f3 �g6 40 l:lb7
lLld7 41 lLlf5 1 -0
...----.
Game 23
Joyce-Speelman
Bunratty Masters 1 998
1 d4 e6 2 c4 b6 3 e4 i.. b 7 4 i..d 3 i.. b4+
5 'iit>f 1
I find it extraordinary that this move
has ever been played by White, but as I
have discovered several practical examples
of this move, I thought I ought to cover it.
I just don't see the point. Of course, it
takes a bit of time to re-deploy the bishop
from b4, but it has caused some disrup­
tion. I don't think White is worse, but I
still can't find a good answer to the ques­
tion, 'Why?'
5 . . . lLlc6
Perhaps even 5 ... .i.f8!? (the bishop is go­
ing round to g7) 6 lt:lc3 d6 7 d5 e5 8 i.e3
lt:lf6 9 g4 g6 10 h4 .i.c8 1 1 f3 h5 12 g5 lt:lh7
13 b4 .i.e7 14 c5 with complications,
though White should be better as his
queenside attack is well advanced, in J ans­
sen-Rogers, Sonnevanck 1996.
But there is no need to play the bishop
all the way back to f8: 5 ... .i.e7 is possible
(Speelman played in similar fashion in the
main game) . Now after 6 lt:lc3 d6 7 'tl¥g4
.i.f6 8 lt:lf3 lt:lc6 9 .i.e3 e5 10 dxe5 h5!?
(both 10 ... dxe5 and 10 ... lt:lxe5 are sensible)
11 "Yig3 h4 (1 1...lt:lxe5!?) 12 "Yih3 i.c8 13 g4
lt:lxe5 14 lt:lxe5 instead of 14 ... dxe5 (Dan­
ner-Barle, Ljubljana 1981) I would prefer
14 ... i.xe5!? 15 .tiel c6 16 f3 "iff6 with the
more promising position for Black.
6 lLlf3 .i.e7 7 a3
Black was threatening .. .'�:Jb4, bagging
the bishop.
7 . . . d6
I imagine Speelman was considering
... .i.f6, similar to the game in the note
above, but White leaps in first.
8 d5 lLle5 9 lLlxe5 dxe5 1 0 i..e 3 lLlf6 1 1
lLlc3 0-0
1 2 dxe6? !
There are too mariy holes in White's
position after this. 12 g3 should be about
level.
1 2 . . . fxe6 1 3 .tc2 'ife8 1 4 �g 1 .l:l.d8 1 5
�e 1 lLlg4 1 6 h4 l:!.d7 1 7 l:!.d 1 "iid 8 1 8
�xd7 '1Wxd7 1 9 '1Wd 1 '1Wxd 1 + 20 lLlxd 1
J:!.d8 2 1 �f1 .txe4 22 i.. b 3 J:ld3 0-1
5 Wfl actually gives more trouble to
White than his opponent. Personally, I
like to pay my king more respect.
Game 24
Shirov-Prie
Val Maubuee 1 990
1 d4 e6 2 c4 b6 3 e4 .tb 7 4 .td3 'i'h4
45
En g lish D e fe n c e

10 ... lt::le 8? 1 1 i..xh7+ �xh7 12 � g5+

�g8 13 'i'hS.

Mad. Although this move is seen regu­

larly in the English Defence, in this par­
ticular position it isn't justified. 1 1 i..x h7+ 'it>xh7 1 2 l2Jg5+ 'it>g8 1 3 Vwxg4
5 d5 f6 1 4 exf6 Wiixf6 1 5 l2Je4 Wilf7 1 6 tLlbc3
This gives Black some hope. There are i..d4 1 7 f3 i..xc3 1 8 l2Jxc3 exd5 1 9 cxd5
two alternatives which are stronger: i..x d5 20 l2Jxd5 'i\Vxd5
a) 5 lt::l d2! i.. b 4 6 lt::l gf3 'i'g4 (or
6 ... i..xd2+ 7 ..txd2 'i'g4 8 'ii'e2 fS 9 h3
'it'xg2 10 0-0-0 li:Jf6 l l li:Jh4 ii'xh1 12 .l::i.xh1
fxe4 13 .l:i.g1 exd3 14 liVeS '>t>f7 15 d5 lt::la6
16 i.. c3 .l:raf8 17 dxe6+ dxe6 18 !!xg7+ 1-0
Papacek-Prie, Clichy 1990; another set­
back for the former Paris Champion, Erik
Prie, with his dubious 4 ...'i'h4) 7 0-0 i..xd2
8 lLlxd2 (8 h3! ? 'i'g6 9 l2Je5 'i'f6 10 i..xd2 -
two bishops, big centre, lead in develop­
ment; sorted!) 8 .. .'iVxd1 9 !!xd1 d6 10 b3
lt::l d7 1 1 i.. b 2 l2Je7 12 i.. c3 lt::l g6 13 d5 e5 14
b4 0-0 15 lt::l b 3 and White soon won in
Mortensen-Brondum, Copenhagen 1980. Black has done remarkably well to get
b) 5 lt::l f3! ? 'i'g4 6 0-0 (Keene, Plaskett this far - but he still stands worse.
and Tisdall point out that White can even 21 i.f4 l:tf7 22 1:rhd 1 Wilb5 23 l:td2 l2Ja6
force a draw with 6 h3 �xg2 7 �g1 'it'xh3 24 lite 1 d6 25 'ii'g 5 'ii'c 6 26 h4 l:te8 27
8 l:tg3 'ii'h 1+ 9 l:tg1 'i'hs 10 Itgs 'ir'h6 1 1 'ii'd 5 'iid 7 28 l:tde2 l:txe2+ 29 J:.xe2 tLlb4
�g1 etc.) 6 . . ...txe4 7 i..xe4 'i'xe4 8 lt::lc3 30 'ti'e4 aS 31 a3 tLlc6 32 h5 l:te7 33
'i'b7 has been played by the Hungarian "i'd5+ �f8 34 :d2 l:te6 35 :c2 l2Je5 36
grandmaster Gyoz6 Forintos as Black but, l:txc7 fkxc7 37 'i'xe6 lLlf7 38 g4 'i'c2+
as yet, not by anyone else. White has big 39 �g3 'i'xb2 40 g5 iid4 41 i..e 3 it'c3
compensation for the pawn. 42 'ii'e4 Wiie 1 + 43 lio>g4 l2Je5+ 44 lio>f5
5 . . . i.. b4+ 6 'lt>f1 i.. c 5 7 g3 Wile7 8 lLlf3 'ii'h 1 45 Wiia8+ We7 46 Wilb7+ l2Jd7 47
lLlf6 'it>g6 Wilh3 48 i..d4 d5 49 �xg7 'ii'xf3 50
8 . . .f5 is met by 9 ..tgs. i.f6+ �d6 51 h6 'ti'e3 52 h7 l2Jxf6 53
9 �g2 0-0 1 0 e5 l2Jg4 gxf6 'i'g5+ 54 <t>t7 'ti'h5+ 55 <t>g8 1 -0

46
 M a in L in e with 3 e 4 ii. b 7 4 ii. d 3 : O th e r Fo urth Mo v e s fo r Bla c k

Summary
Although the position after 4 .. .f5 5 exf5 .ib4+ is quite unstable and for many moves
quite unclear, my feeling is that Black is suffering in Games 12-14. Even if improvements
are found in Beliavsky-Short, the older lines in Game 13 aren't appealing anyway. The
rook sacrifice 4 .. .f5 5 exf5 i..xg2 in Games 15-17 has not yet been fully explored. If you
put the effort in here then you could be rewarded. On the other hand, attempts by
White to run from the sacrifice in Games 18 and 19 are simply poor. I find 4 ... i..b4+ in
Games 20-23 a bit tame; and, as I said before, 4 ... 'iih4 is just daft.

1 d4 e6 2 c4 b6 3 e4 ii.b7 4 ii.d3

4 . . f5 (DJ
.

4 ... .ib4+
5 i..d2
5 ... 'i¥e7 - Game 20
5 ... tt:Jc6 - Game 21
5 ... i.. xd2+ - Game 22
5 'it>fl - Game 23
4 ...'iVh4 - Game 24
5 exf5
5 tLlc3 - Chapter 3, Games 25-33
5 'i\Vh5+ - Game 18
5 d5 - Game 19
5 . . . ii.b4+
5 ... i..xg2 6 'it'h5+ g6 7 fxg6 i.. g7 8 gxh7+ �f8 9 i..g5 tLlf6 10 'it'h4 i..xhl l l tLle2 (DJ
1 1...tbc6
12 tt:Jf4 - Game 15; 12 tbd2 - Game 16
1 1 ...i¥e7 - Game 11
6 'it>f1 l"Llf6 (DJ
6 ... exf5 - Game 14
7 ii.e2
7 c5 - Game 13
7 . . 0-0
. - Game 12

1 1 tbe2 6. . . tb f6

47
 CHAPTER THREE I
Main Li ne w ith 3 e4 i.b 7
4 d 5 , 4 lb c3 i.. b 4 5 d 5 and
4 lb c3 �b4 5 i.. d 3

1 d 4 e 6 2 c4 b 6 3 e4 i.. b 7 4 d5 is quite a common alternative - see

In this section we shall deal with games Games 35 and 36, while 4 lLlc3 f5 5 d5 is
where White puts his three pawns in the the subject of Game 4.
centre, and then supports it with either 4 . . . f5 5 lbc3 i.. b4
lLlc3 combined with i.d3 (Games 25-33)
or an early d4-d5 (Games 34-36) . In the
following two chapters we shall then
move on to look at games where White
plays the three pawns forward but chooses
to protect the centre with f2-f3 (Chapter
4) or iVc2 (Chapter 5) .
Beliavsky used to play with both i.d3
and lLlc3 before he realised that he could
take the pawn on f5 (see Game 12 in the
previous chapter) . No wonder he felt he
needed to try something new, for al­
though Beliavsky gets the advantage in
Game 25, Gulko didn't play the opening In principle this should present Black
as vigorously as he might have. Inciden­ with few difficulties: White's centre is
tally, this common position can, and often under enormous strain.
is, reached via a different move order 6 d5
whereby the knight comes out to c3 and is For 6 'iih 5+ see Games 29-32; and for 6
pinned (4 lLlc3 i.b4) before the bishop iVe2 see Game 33.
gets to d3 (5 i.. d3) . 6 ... fxe4 7 i.. xe4 lbf6
A solid treatment, but 7... iVh4! ? is
Game 25 more ambitious - as Gulko himself later
Beliavsky-G u l ko realised - see Games 27 and 28.
Polanica Zdroj 1 996 8 i..f3
Beliavsky mentions 8 i.g5!?, but the
1 d4 e6 2 e4 b6 3 c4 i.. b 7 4 i..d 3 following game demonstrates that Black

48
 Main L in e with 3 e4 i.b 7 4 d5, 4 li:J c 3 i. b 4 5 d5 a n d 4 li:J c 3 i. b 4 5 i.d3

has few difficulties: 8 ... .ltxc3+ 9 bxc3 tZ:la6

10 tZ:le2 'LlcS (as is so often the case in this
opening, once Black's queen's knight finds
a decent square, then he has normally
solved his opening problems) 1 1 �f3 0-0
12 0-0 'ii'e 8 13 tZ:lg3 'iig6 14 .lte3 tZ:lfe4 and
Black already had the more active position
in Bozic-Lempert, Bled 1994.
8 . . e5
.

This turns out well, but Beliavsky men­

tions that 12 cxdS! is even stronger, with
the idea that 12 ... tZ:la6 13 tZ:lg3 tZ:lc7? 14 d6
wms a p1ece.
1 2 . . . l2Jxd5 1 3 i.xd5+ i.xd5 1 4 cxd5
I think I would have just taken the ex­
change here: 14 �xb4 �xc4 15 ii.xf8 �xf8
16 b3 -te6 17 f4 just looks good for White
to me. Having said that, the game con­
I do think it is a pity to close the centre tinuation is also strong.
like this: Black's bishop on b7 looks stu­ 1 4 . . . i.xd2 1 5 'i¥xd2 li:la6
pid. The next main game shows that Black Normal moves don't help, e.g. 15 ... d6
does have better alternatives. 16 �ac1 tZ:ld7 17 l:tc6 'Llb8 18 �c4 with a
9 i.d2 clear advantage.
Not 9 tZ:le2? e4, but 9 tZ:lh3 deserves a 1 6 d6 li:lcS 1 7 �d5+ Wh8 1 8 b4 li:Je6 1 9
closer look: 9 ... 0-0 10 0-0 �xc3 1 1 bxc3 �xe5
tZ:la6 12 �a3 d6 13 tZ:lgS and although
Black eventually won in Reyes­
Yermolinsky, Seattle 1994, at this point I
prefer White's chances.
9 . . . 0-0 1 0 li:lge2
White has a solid position, his pieces
harmoniously placed; while Black's open­
ing has not been a great success. The idea
of blocking the centre doesn't fit in with
Black's set-up - at least not in this posi­
tion. The bishop on b7 is blocked in and
the knight on b8 has yet to find a good
square. Conclusion: White has the advan­
tage. White is a clear pawn up, but ... Gulko
10 . . . c6 1 1 0-0 cxd5 1 2 li:Jxd5 does well to get some play for it. · The
see following diagram
point is that finally Black's knight has
found a stable square in the middle of the

49
En glish D e fe n c e
board, and it is hard for White to break
through.
1 9 . . . a5 20 b5?
Beliavsky believes that 20 a3 would
have retained his advantage: 20 ... axb4 2 1
axb4 Itxa1 2 2 l:txa1 'ii'h 4 2 3 f4.
20 . . . 'ilfh4 2 1 'iie 3
Or 21 f4 !Iae8 with compensation.
21 . . . 'ilfc4 22 l:!.fc 1
If 22 a4 tLlcS and Black obviously again
has compensation.
22 . . . 'iix b5 23 l2Jd4 l2Jxd4 24 'ilixd4 l:!.f6
25 J::!. a b1 �e2
26 :b2
If 26 'ili'xb6 'tixa2 is possible, e.g. 27
'ii'b 8+ �f8! with equality (but not
27 ... .U.xb8? 28 l:!xb8+ "ii'g 8 29 l::i.cc8 and
wins).
26 . . . �e6 27 'ii'x b6 �xd6 28 'ilfxd6 J::!.x d6
29 g3 'it>g8 30 .=.c4 .l:d5 31 h4 J::!.a 6 32
'it>g2 'it>f7 33 J::!. e 2 J::!. e 6 %-%
7 ... tt:Jf6 is a sound way of dealing with 6
d5 - if Black plays in the same fashion as
Bagirov in the next game.
,....-----.
Game 26
Dgebuadze-Bagi rov
Linares 1 997
.________________.
1 c4 b6 2 d4 e6 3 lLlc3 i.b7 4 e4 i.b4 5
i.d3 f5 6 d5 fxe4 7 i.xe4 lLlf6 8 i.f3
This is the point where Gulko tried
8 ... e5. Bagirov's move is stronger:
50
8� . . i.xc3+
This isn't the only decent option:
a) 8 ...exd5 9 cxd5 c6!? (9 ... lba6 10 lbge2
lbc5 1 1 0-0 i..a6 12 'ii'd4 i..xc3 13 bxc3 0-0
14 c4 was slightly better for White in Mi­
nasian-Tratar, Cannes open 1995) 10 d6
i..xc3+ 1 1 bxc3 0-0 12 lLle2 lba6 13 0-0
lbc5 14 'it'd4 ..ta6 15 c4 lZ'le8 16 ..te3 'ii'f6
17 "ii'g4 'ili'f7 18 l:Hd1 ..txc4 and Black was
a clear pawn ahead in Lautier-David,
Hameln Voba 1987.
b) 8 ... 0-0 9 lt:Jge2 b5!? 10 b3 bxc4 1 1
bxc4 lba6 1 2 0-0 l:!b8 1 3 ..te3 i..c5 14
..txc5 lbxc5 15 lZ'ld4 ..ta6 16 lbdb5 e5 17
.I:i.e1 ..txb5 18 cxb5 d6 19 a4 'i*'d7 20 h3 g6
21 lZ'le4 lZ'lfxe4 22 ..txe4 a6 23 bxa6 l:i.b4 24
..td3 !Id4 0-1 Johansson-Bagirov, Jy­
vaskyla open 1996, was a successful ex­
periment, but it is interesting to note that
by the time he came to play this main
game, Bagirov had thought better of this
idea and essayed the safer 8 ... ..txc3+.
9 bxc3 0-0
Alternatively, 9 ...exd5 10 cxdS 0-0 1 1
tLle2 lZ'la6 12 0-0 lbc5 13 c4 lZ'le8 14 ..tb2
lt:Jd6 15 !!cl i.. a6 16 'ii'd4 'ii'g5, when al­
though White has pressure on the long
diagonal, Black has it covered, and his
knights have found excellent squares.
Chances were balanced in Sakalauskas­
Gorbatow, Polanica Zdroj open 1996.
1 0 lLle2 e5 1 1 l2Jg3 d6 1 2 0-0 lLlbd7
The exchange of bishop for knight on
 Main L in e with 3 e4 i.b 7 4 d5, 4 I:D c 3 i.. b 4 5 d5 a n d 4 I:D c 3 i. b 4 5 i. d 3
c3 means that Bagirov can aim to put a
horse on the beautiful c5-square; or at least
force White to give up the bishop pair to
prevent this.
1 J i..e 3 �e8 1 4 'i'c2 'ilff7 1 5 i.. e 2 ll:ae8
1 6 i.d3 i.c8 1 7 J:!.ae 1 ti:Jc5 1 8 i.xc5
bxc5 1 9 'i'a4 a6 20 'ii'a 5 g6 21 f3 'i'd7
Y. - Y.
A curious move to offer a draw with!
Bagirov has a plan: to play the queen back
to d8, then swing the knight over from f6-
d7-b6. That would secure the queenside
{thereby releasing Black's queen) and put
some pressure on c4. In that case I would
prefer Black's position. The question is,
can White prevent it? Let's continue the
game a little further: 22 i.. c2 �d8 23 .ta4
�e7 24 .l:tb 1 l2Jd7 (24 ... e4! ?) 25 i..xd7 l!Vxd7
26 l:tb8 l:tef7 followed by ... W/e7 and
chances are level.
Game 2 7
Kostin-Grabuzova
Moscow 1996
1 d4 e6 2 c4 b6 3 l:i:Jc3 i.b7 4 e4 i.. b4 5
i..d 3 f5 6 d5 fxe4 7 i.xe4 'i'h4!
-
That's more in the spirit of the open­
ing. Let's see if it is any good. 7....�:Jf6 was
played in the previous two games.
8 'ilfe2
White has two alternatives here:
a) 8 'iid3?! is provocative:
a1) 8 ...'2Jf6 9 .i.f3 e5 (9 ... '2Ja6!? is also
promising) 10 i..e3? (10 l2Jge2 is playable)
10 ... e4 1 1 'ti'd4 exf3 12 �xh4 fxg2 13 0-0-0
gxh1'ii' with a winning position for Black
in Simonsen-Nicolaisen, Hillerod 1979.
a2) 8 ... exd5 9 cxd5 l2Jf6 10 i..f3 i.. a6 1 1
'ike3+ 'it>f7 1 2 'iif4 l:te8+ 1 3 'it>d1 'ikxf4 14
i..xf4 i..xc3 15 bxc3 d6 16 l2Jh3 (16 l2Je2
i..c4!) 16 ... h6!, containing White's pieces
before cashing in. Black duly converted
the point in Whiteley-Keene, Cambridge
1976.
b) 8 'Wd4 l2Jf6 9 i..f4 i..xc3+ 10 bxc3
l2Jxe4 1 1 'iixe4 0-0 12 g3 'i'f6 13 l2Je2 exd5
14 cxd5 l2Ja6 15 0-0 l2Jc5 and Black already
had a powerful initiative in Bauer­
Chetverik, Zalakaros open 1996.
8 . . .ti:Jf6 9 ..tf3
Alternatively, 9 i.d3 l2Ja6 (for the
greedy among you, why not 9 ... 0-0!? 10
51
En g lish D e fe n c e
lb f3 'iig4 1 1 0-0 i.. xc3 1 2 bxc3 exd5) 10
l2Jf3 'ii'h 5 11 i.d2 0-0-0 12 0-0-0 Ithe8 and
Black stood well in Kalantarian-Bricard,
Cannes open 1995.
9 . . . i.. a 6
If a defence is later discovered for
White, then Black might also try 9 . . . 0-0 -
see the next game.
1 0 dxe6 d 5
Good move. White is struggling to get
his king out of the middle, and in the
game he immediately goes wrong.
1 1 a3?
White ought to play 11 'ii'e 5, though af­
ter 1 1..:i!Wxc4 12 lbge2 0-0 chances are bal­
anced
1 1 . . . i.. xc3+ 1 2 bxc3 i.. xc4 1 3 �c2 lZ'le4
1 4 i.. e 3
White has to defend f2 before develop­
ing the knight. If 14 g3 'it'f6 and the attack
52
rolls on, while 14 i.xe4!? dxe4 15 lbe2 0-0
16 0-0 lbc6 is also miserable for White - he
is going to come out a pawn down even if
he does manage to avoid being check­
mated.
1 4 . . . lZJc6 1 5 lZ'le2 lZJe5 1 6 l::td 1 0-0 1 7
i.. xe4 dxe4 1 8 0-0 lZ'lf3+ 1 9 gxf3 exf3 20
e7 �g4+ 21 lZ'lg3 'i!Vh3 22 ext8'i!V + :xt8
0-1
Can't argue with that; or the next
game. 6 d5 is in trouble.
Game 28
Marchand -Gulko
Geneva 1997
1 d4 e6 2 c4 b6 3 e4 i.. b 7 4 i..d 3 f5 5
lZJc3 i.. b4 6 d5 fxe4 7 i.. xe4 'ii'h4 8 'ii'e 2
lZ'lt6 9 i..f3 0-0
While 9 ... i.a6 was successful in die last
game, I have to say that castling is the first
move that would enter my head: the king
exits, the rook enters and carnage is on the
cards. From this position I have three
games on my database, with White manag­
ing one measly half-point - and that with
some luck.
1 0 dxe6
10 i..e3 lbe4 1 1 g3 was played in Spigel­
Sutter, US open 1994, and now 1 1 . .:�f6
would have been powerful.
1 o . . lZ'lc6 !
.
That's the trick. There is no need to ex-
 M a in L in e with 3 e 4 il. b 7 4 d5, 4 tb c 3 il. b 4 5 d5 a n d 4 tb c 3 il. b 4 5 il. d 3

change bishops (which would only help �d4 �a 1 + 27 �d2 'iixb2+ 28 �e3 'i'c 1 +
White develop) and at a stroke Black's 0-1
rooks are connected.
1 1 g3
1 1 J.e3 was played in Kaplan-Miles, Sao
Paolo 1977. It is interesting to see that
7 ... 'i!Vh4 was played so long ago, which
makes it all the more curious that players
started experimenting with 7 ... tLlf6. That
game continued 1 1 . . .tLle4 12 0-0-0 tLlxc3 13
bxc3 �a3+ 14 Wb 1 dxe6 15 'i!Vc2 J.cS!? (an
interesting decision; Miles allows his
pawns to be shattered and in return gets
tremendous piece play) 16 .ixcS bxcS 17
tLle2 !Iab8 18 'it>a1 "ikxc4 19 J.e4 .l::!.xf2 20
�xh7+ �h8 21 Itd2 iVh4 22 .ig6 .l:i.bf8 It is time to conclude our look at 6 dS -
(22 .. Jhg2!?) 23 l:tb l. There is no doubt and my advice to players of the white
that Black is better here, but it is a difficult pieces would be: avoid it ! 7 ... tLlf6 is a solid
position to control and White eventually response, but on the evidence of the last
succeeded in scraping a draw. two games, there is every chance that
White will be blown out of the water if
Black goes in for 7 .. .'ifh4.

Game 29
Gartner-Dey
Berlin 1994

1 d4 e6 2 c4 b6 3 e4 il.b 7 4 tbc3 il.b4 5

il.d3 f5 6 'iih 5 +

11 'iWd4 1 2 il.d2 il.a6 1 3 exd7

. . .

The capture of another pawn does not

really help matters. It would have been
oetter to castle queenside at once.
13 il.xc4 1 4 'iWe3 'iix d7 1 5 0-0-0 Sae8
. . .

16 'iit4 tbe5 1 7 'ifh4 'iie 6

see follo wing diagram

What a picture. White could have re­

signed here with a clear conscience. This check has been quite popular in
8 �b1 il.xc3 1 9 il.xc3 il.xa2+ 20 �a 1 practice. Instead of protecting the e-pawn
il3 21 �e 1 b5 22 �xe5 'i'a6+ 23 �b1 straightaway, White forces one of Black's
a2+ 24 �c 1 �xe5 25 il.xe5 :da 26 pawns forward on the kingside, creating a

53
En g lish D e fe n c e
slight weakness, and only then drops back
to e2. All very sophisticated; but in my
opinion, it is not up to much (but keep it
quiet!) . The immediate 6 'ifke2 is consid­
ered in Game 33, while 6 'ifkc2 is well met
by 6 .. .'it'h4 with tremendous pressure on
the white centre (see Chapter 5, Game 49).
6 . . . g6 7 �e2 lt:lf6 8 i.g5
White still cannot move the e-pawn; or
at least he ought not to, e.g. 8 eS .ltxg2 9
exf6 'ifxf6 10 �f4 lt:lc6 1 1 0-0-0 .lixh1 12
d5 .ltxc3 13 dxc6 .ltxc6 14 bxc3 'i'xc3+ 15
'it>b 1 0-0-0 16 lt:\f3 d6 17 ..icl eS and Black
was well on the road to victory in Knott­
Speelman, British Championship, Ayr
1978. However, it is possible to bolster the
centre with 8 f3 - see Game 32.
8 . . . fxe4
This is the most sound and safest way
of handling White's system. Black should
not encounter any difficulties if he plays
accurately. However, a more enterprising
choice is 8 ... h6, for which see Game 3 1 .
9 il.xe4
The correct move order. A couple of
inattentive players have tried 9 .ltxf6 at
this point, failing to spot the intermezzo
9 ... exd3! 10 "f*'eS (10 .ixd8 dxe2 wins as
the g2-pawn is en prise) 10 ... tbc6 1 1 'figS
.ie7 12 .ixe7 'ii'xe7 13 'i!Vxe7+ �xe7 14
tbf3 tbaS 15 li:ld2 (or 15 b3 i..xf3 16 gxf3
�af8) 15 ... .ixg2 16 .l:tg1 .ih3 17 �g3 i.fs
with a winning ending for Black in T az-
54
hieva-Chiburdanidze, Moscow Women's
Olympiad 1994.
Finally 9 .ic2?! (an unnecessary gambit)
was tried in Rodriguez-Keene, Alicante
1977, but Black was up to the task:
9 .. .'f/e7 10 0-0-0 .ixc3 1 1 bxc3 tbc6 12 f3
�a3+ 13 �b 1 0-0 14 ..icl 'ii'xc3 15 i..b 2
'ii'aS and after some complications White
lost the game.
9 . . . ..txe4
The speculative 9 ... tbxe4 is the subject
of the next main game.
1 0 il.xf6 'ii'xf6 1 1 'ii'xe4
1 1 . . . il.xc3+
It isn't absolutely necessary to capture
on c3 immediately, though the bishop
isn't going to retreat either. In this kind of
position Black is able to use the f-file to
generate play, though in Sorin-Garbarino,
St Martin 1995, he tried a different strat­
egy: 1 1 ...tbc6 12 tbf3 �fS!? (this is what is
slightly unusual - normally there is a rook
supporting the queen) 13 'iixfS .ixc3+ 14
bxc3 exfS (14 ... gxf5!? would have unbal­
anced the position) 15 'it>d2 0-0 16 h4 h6
17 'it>d3 .l:!.ae8 18 .l:!.ae1 d6 and the chances
were even.
1 2 bxc3 lt:lc6 1 3 lt:lf3 0-0 1 4 0-0 'i'f4
see following diagram
1 5 l:.fe 1
If the queen backs off then Black
should not have too many problems. Here
 Main L in e with 3 e 4 i.b 7 4 d5, 4 t'iJ c 3 i.. b 4 5 d5 a n d 4 t'iJ c 3 i. b 4 5 i.. d 3
are a couple of examples, with Black em­
ploying differing strategies:
a) 15 'i'd3 iVfS 16 'ii'e2 l::t ae8 17 l:i.ae1
�e7 18 'ii'd2 l2Jd8 19 .Ue3 d6 20 :l.fe1 lHe8
(it strikes me that Black should transfer his
heavy pieces to the f-file 20 ... �ef7, as the
knight on d8 holds the e6-pawn com­
fortably; Black can follow up with ... g7-g5
and if necessary ... h7-h5, when Black's
rooks attack and defend the king at the
same time) 21 'i*'b2 l2Jf7 22 'ii'b S �g7 with
equal chances in Elguezabal Varela­
Yewdokimov, Madrid 1994.
b) 15 'i!Ve2 is similar, though Black
played more boldly on the kingside in
Iskusnyh-Lempert, Orel 1995 - and reaped
the reward: 15 ... .Uf5 16 t2Jd2 'it'h6 17 l2Je4
�af8 18 l::t ad1 l2Je7 19 .Ud3 gS! ? 20 l:th3
li'g7 2 1 'iid 1 l2Jg6 22 l::t g3 h6 23 .Ue3 tiJf4
24 ttJg3 nsf7 2s ttJhs 'it'g6 26 l2Jg3 \itlh7 27
lib 1 'i'xb 1 28 l::txb 1 \itlg6 29 l!bS tiJhS 30
ttJxhs �xhs 31 f3 'it>g6 32 h3 lifs 33 :!:teeS
a6 34 l:i.xfS .l::txfS 35 !:i.xfS �xfS (the ending
is winning) 36 'it>f2 hS 37 g3 h4 38 gxh4
gxh4 39 'it>e3 dS 40 cxdS exdS 41 a3 aS 42
f4 a4 43 \itlf3 c6 44 \itle3 cS 0-1. A fine
positional achievement.
15 . . . 'i'xe4 1 6 .!:.xe4 :ts 1 7 .1;1d 1
Compare this game with Game 33,
where White does not bother to check on
5 and plays the immediate 6 'il'e2. The
position is the same apart from the pawn
on g6 which there stands on g7. Keene
played 17 ... l::ta5 and held the draw com­
fortably. That's not to say the move in the
game is a mistake.
1 7 . . . .!:!.af8 1 8 d5 exd5
I prefer 18 ... tiJd8 - the knight does not
stand badly, and it is a pity to undouble
the c-pawns; moreover the knight can be
re-deployed (after exchanging pawns) via
b7 to d6 or cS.
1 9 cxd5 t'iJaS 20 d6 cxd6 2 1 .!:!.xd6 lt:Jb7
22 .!:!.d2
Not 22 .lhd7? liJcS.
22 . . . lt:Jc5 23 .!:l.e3
It is unfortunate that Black's queenside
·
pawns have been split, but his knight has
found a good square on cS. But White is
better thanks to his strongly placed rooks
and his three beautiful kingside pawns.
23 . . .l:l.f4
23 ... l:t8f7 is probably stronger.
24 h3 .!:!.8f6 25 lt:Je5 .!:!.a4
Perhaps 25 ... a5!?
2 6 lt:Jxd7 t'iJxd7 27 ,l;txd7 l:txa2
27 ...l:i.f7 would have been better.
28 l:.e8+ .l:l.f8 29 l:txf8+ �xf8 30 .!:!.xh7 aS
3 1 c4 l:.a4
see following diagram
3 1 ...l:Ic2 32 I!b7 a4 33 .Uxb6 a3 34 l:ta6
a2 35 �h2 l:txf2 would have offered more
drawing chances.
32 l:.c7 lib4 33 c5 bxc5 34 l:.xc5 a4 35
.!:!.c7 a3 36 .!:!.a7 .l:.b3 37 g3 .l:.c3 38 �g2
55
En g lish D e fe n c e
l::t b 3 3 9 h4 .!::t c 3 40 g4 l::t b3 4 1 f3 �ea
�g3 �f8 43 �f4 .!::t c 3 44 �g5 l:.xf3
�xg6 l:tb3 46 g 5 l:b6+ 47 �h5 �b3
�h6 l:tb8 49 �xa3 �g8 50 g6 �h8
h 5 .!::t c 8 52 �a5 �b8 5 3 �g5 l:. b 1
.l:!.a8+ �g7 55 h6 mate 1 -0
Black was doing fine in the ending; he
only lost through poor technique. Not
only that, I do not consider that Black is
merely fighting for a draw in this kind of
ending. With his superior pawn structure
Black can play for a win, as the game
Iskusnyh-Lempert in the notes demon­
strates. The plan demonstrated in this
game represents a solid way to counter 8
i.gS. Now let's look at some less sound,
but possibly more interesting methods.
Game30
G runfeld -Prie
Paris 1990
1 c4 b6 2 lt:lc3 e6 3 e4 �b7 4 d4 ..tb4 5
�d3 f5 6 'ii'h 5+ g6 7 'ii'e 2 lt:lf6 8 i.g5
fxe4 9 �xe4
Everything the same as the previous
game so far, but here, just when White
thought he was going to get some nice
exchanges, Black chooses to sacrifice his
queen.
9 . . . lt:lxe4 ! ? 1 0 i.xd8 lt:lxc3 1 1 bxc3
..txc3+ 1 2 �f1 �xa 1 1 3 �xc7 �xd4
A dynamic balance has been reached:
56
42 Black has wonderful minor pieces, but his
45 king could be better protected. Queens are
48 superb pieces in situations where they
51 have numerous targets to attack.
54
1 4 lt:lt3 i.c5
This way White's king comes under
immediate fire. I prefer this move to
14 ... i.g7, though Black also has reasonable
chances here too, e.g. 15 h4 lLlc6 16 hS
l::tc8 17 i.d6 i.a6 18 lLJd2 lLld4 19 'iig4
with a tense struggle ahead in Bick­
Welling, Koop Tjuchem open 1996.
1 5 'ii'e 5 0-0
1 6 �xb8
It seems a pity to give up the bishop for
a knight, especially a knight that hasn't
even entered the game, but the prospect of
the queen being hounded by the minor
pieces was too worrying for White.
1 6 . . . .!::ta xb8 1 7 h4 l::t bc8 1 8 h5 l:.f5 1 9
 M a in L in e with 3 e 4 i. b 7 4 d5, 4 ti'J c 3 i. b 4 5 d5 a n d 4 ti'J c 3 i. b 4 5 i. d 3
'i'c3 i.xf3
19 ... g5! ? 20 h6 i.. f8 was worth a glance.
20 gxf3 g5 2 1 'ifd3 l:tc7
am surprised that Black wishes to de­
I .li.d3 f5 6 'i'h5 + g6 7 'iie 2 4:Jf6 8 .ltg5
�end in such a painstaking and careful
:nanner, though it has to be said that his
rategy pays off in the game. 2l...d6 with
ilie idea of activating the rooks straight­
away on the f-file appeals to me.
2 2 �g 1 i.e 7 23 .l::. g 4
This was the moment to throw in 23
6!? to confuse the issue. Black's rooks are
:Jot co-ordinated, so with the queen about
- e play is tricky.
23 . . . �f7 24 .l::. d 4 We8 25 .l::.g 4 h6
With a huge sigh of relief. Black's posi­
:ion is more secure and he can start to
:-egroup for an attack in safety.
26 Wg2 .ltt6 27 .l:te4 rJiie 7 28 'it>g3 .i.e5+
29 Wg2 d6 30 �e2 �7 3 1 .l:te4 Wg7 3 2
.l:te3 l!cf7 33 'ife4 g 4 ! 3 4 f4 .i.xf4 35
'ifxe6 i.xe3 36 fxe3 J:.xh5 37 'i'xd6 :Xg5
38 e4 g3 39 e5 l:.f2+ 40 �g 1 g2 0-1
The queen sacrifice is interesting, and
most likely sound. That's not to say it
gives Black a winning position; chances
are balanced. It really depends on one's
style of play - and one's opponent. It
might be just the thing to shock the more
gentle-minded player.
Game31
Gamota -Karasev
Moscow 1 996
1 d4 b6 2 c4 .ltb7 3 4:Jc3 e6 4 e4 i.b4 5
h6
If you are looking for a 'sound' way to
play for a win, then this is the move.
8 ... fxe4 is fine, but it releases the tension;
and the queen sacrifice from the previous
game is great fun, but can go wrong at a
stroke. 8 ... h6 holds the tension but bags
the bishop. It makes a virtue out of having
the pawn on g6: the fS-pawn is bolstered,
and the queen can slip back to g7 if neces­
sary.
9 i.xf6 'i'xf6 1 0 ti'Jf3 4:Jc6
In most of the games played in this line
Black has castled queenside, but there is
nothing wrong with going the other way
with 10 ... 0-0, particularly as there is no
57
En g lish D e fe n c e
dark-squared bishop to be concerned
about; the queen also does a good job de­
fending the king. I have found one game
which follows this course: 1 1 0-0-0 ..ltxc3
(1 1...tZ'lc6!? is more challenging) 12 bxc3
fxe4 (likewise 12 ... tZ'lc6!?) 13 i.xe4 'i!Vf4+
14 tt:Jd2 i.xe4 15 'iixe4 'iixe4 16 tZ'lxe4
tZ'lc6, when a familiar ending has been
reached and chances were equal in Ali­
Freeman, Asian Team Championship
1995.
1 1 0-0
Alternatively, Hubert-Stanetzek, Porz
1989, continued 1 1 e5 'iif7 (I prefer
1 1 . . .Wig7; if White ever tries to play d4-d5,
which can sometimes cause serious disrup­
tion, then it is nice to have the queen
trained on the e5-pawn) 12 0-0 ..ltxc3 13
bxc3 g5 (13 ... 0-0 is stronger, and only then
some action on the kingside) 14 c5! tZ'le7
(or 14 ... bxc5!? 15 ltab 1 with some com­
pensation) 15 i.a6! i.c6 16 c4 i¥h5?! (it
was better to play 16 ... 0-0 with chances for
both sides) 17 d5 exd5 18 e6 with the at­
tack.
11 • • • 0-0-0
As I said above, I like 1 1...i.xc3 12 bxc3
0-0. For me, Black's king is safer on the
kingside. But the big question here is
whether Black can capture the pawn with
1 1 . ..tZ'lxd4!? 12 tt:Jxd4 11i'xd4:
a) 13 tt:Jb5 11i'e5 14 exf5 ifxe2 15 i.xe2
0-0-0 (not 15 .. .'�d8 16 fxe6) 16 fxg6 lthg8
58
(or 16 ... i.e4) 17 tZ'lxa7+ �b8 18 tt:Jb5 ltxg6
with compensation for the pawn.
b) 13 exf5 gxf5 14 .ltxf5 (but not 14
tt:Jb5? 'i¥g7) 14 ... 0-0-0 with a complicated
struggle ahead.
The verdict on 1 1 . . .tZ'lxd4 ... ? Worth a
try.
1 2 e5 "i'g7
12 .. ."iif7 was tried in Schebler-Kalka,
German Bundesliga 1985, and now the
move which would worry me is 13 c5! ?
bxc5 1 4 .lta6 with a vicious attack.
1 3 d5
Although this creates some confusion,
it destabilises the centre. As in the note
above, I prefer 13 c5! when White gets a
thumping attack. Here is a taste of what
might happen: 13 ... bxc5 14 i.a6 cxd4 15
tt:Ja4 g5 16 ltfd1 d3 17 Wkxd3 g4 18 'ii'b 5
i..xa6 19 'i'xa6+ �b8 20 tZ'ld4 tt:Jxd4 21
.l:!xd4 and I cannot see Black's king surviv­
ing too long (though I must admit this
was a co-operative variation).
1 3 . . . l2Je7 1 4 l2Jb5 a6 1 5 a3 axb5 1 6
axb4 exd5 1 7 cxd5 l2Jxd5 1 8 i..x b5 Wle7
1 9 :a7
see following diagram
Black's king is quite safe. The position
reminds me of the Sicilian Dragon: the
fianchettoed bishop holds the fort.
1 9 . . . l2Jxb4 20 l:tfa 1 W'c5 2 1 l:t 1 a4 �bS
22 "i'e 1 l2Jc6 23 i.. xc6 i..x c6 24 :a3
 M a in L in e with 3 e4 i.b 7 4 d5, 4 t'D c 3 i.. b 4 5 d5 a n d 4 t'D c 3 i.. b 4 5 i.. d3
.:he8 25 b4 �d5 26 �a 1 .tb7 27 b5 d6
28 �xb7+ �xb7 29 :a7+ �c8 30 e6
.:xe6 31 t'Dd4 'ifxd4 0-1
Game32
Burger-Ehlvest
St Martin open 1993
1 d4 e6 2 c4 b6 3 t'Dc3 i.. b4 4 e4 i.b7 5
�d3 f5 6 it'h5 + g6 7 Wie2 t'Df6
And now, instead of 8 i.. g5 from the
previous three games . ..
8 f3
. . . White attempts to maintain his big
?awn centre. Speaking from Black's point
of view, just what we want!
8 . . . t'Dc6!
I like this move. White has bolstered
::he e4-pawn but it is now its d4-
counterpart which is the sensitive spot.
After 8 .. .fxe4?! 9 fxe4 i.xc3+ (9 ...lt:Jxe4 10
i.xe4 i.xc3+ 11 Wd1! was good for White
in Gillani-Williams, British Champion­
ship, Norwich 1994) 10 bxc3 lt:Jxe4 1 1 tt::l f3
li:Jf6 12 .ig5 'i!Ve7 13 0-0 d6 14 lt:Je5 0-0 15
lt:Jg4 li:Jbd7 16 Itf2 White had a winning
position in Botvinnik-Wallis, simultane­
ous display, Leicester 1967.
9 i..e 3
9 e5? is an amusing blunder: 9 ... lt:Jxd4!
10 'ii'f2 lt:Jh5 and Black wins a pawn as 1 1
1i'xd4 i.c5 traps White's queen. This was
actually the continuation of Adorjan­
Spassky, Toluca Interzonal 1982, which
Black won in only 23 moves.
9 . . . fxe4
Ehlvest has tried this twice (with suc­
cess) but it is not Black's only option, as
Plaskett has demonstrated:
a) 9 ... 0-0 10 lt:Jh3 fxe4 1 1 fxe4 e5 12 d5
li:Jd4 with advantage to Black in Crouch­
Plaskett, Hitchin 1977.
b) 9 .. .f4!? 10 .ixf4 (10 Si.f2 e5!? is un­
clear - Plaskett) 10 ... lt:Jxd4 1 1 'ilff2 lt:Jc6 12
lt:Jge2 0-0 13 ilh4 i.e7 14 .ih6 lt:Je5 15
0-0-0 lt:Jfg4 0-1 Flear-Plaskett, Blackpool
1982.
1 0 fxe4 e5 1 1 d5 t'Dd4 1 2 ii'd 1
The following year, Gorbatov-Ehlvest,
St Petersburg open 1994, continued 12
.ixd4 exd4 13 a3 i.xc3+ 14 bxc3 dxc3
(although White's centre appears impres­
sive, it is difficult to do anything with it; it
59
En g lis h D e fe n c e

takes time to regain the pawn on c3, and

in the meantime Black heaps on the pres­
sure) 15 ti:Jf3 0-0 16 0-0 'ir'e7 17 e5 .l:i.ae8 18
kf.ae1 'i'c5+ 19 �f2 'ir'xf2+ 20 �xf2 tt:Jh5 21
.l:lc2 c6 22 d6 c5 23 kf.xc3 i.xf3 24 gxf3 �f6!
25 i.e4 .l:.xe5 26 i.d5+ .l:lxd5 27 l:.e8+ 'it>g7
28 cxd5 l:txd6 and this should have been a
comfortable win for Black, but the game
was drawn in 44 moves.
1 2 . . . l2Jxe4!

Instead of throwing in a check on h5,

I have a feeling that White overlooked
this. Black has a winning position.
1 3 il.. x e4 �h4+ 1 4 �1 il.. xc3 1 5 bxc3
'ifxe4 1 6 il.. x d4 exd4 1 7 'ifxd4 0-0 + 1 8
l2Jf3 �ae8 1 9 �2 'ife2+ 20 �g3 �e4 2 1
llhe 1 J:txd4 22 J:txe2 llxc4 2 3 'P.e7 .i.xd5
24 li[xd7 .i.xf3 25 gxf3 .!::tx c3 26 !:!.f1 'P.f7
27 .l:!.d8+ �g7 28 .!:!.f2 a5 0-1
Although I'm sure that 8 f3 isn't a poor
move, in practice White has scored badly
with it. There is so much pressure on the
white centre that it is difficult to know
where the next hit is coming from. What
is encouraging about the position from
Black's viewpoint is that he has several
promising continuations at his disposal.
White plays the queen to e2 directly. Al­
though this is less popular than 6 'ifh5+, I
think it is stronger as the f5-pawn no
longer has the protection of the pawn on
g6. Nevertheless, Black has few difficulties
as Keene demonstrates.
6 . . . l2Jf6 7 il.. g 5
7 f3 ti:Jc6! is very similar to the previous
game.
7 . . . fxe4
7 ... h6 is also playable here, when play
will follow the lines of Game 3 1.
8 .i.xe4
Not 8 i.xf6? as 8 ...exd3 9 ikh5+ g6 10
'il'h4 .i.e7 1 1 i.xe7 'i'xe7 12 'ir'xe7+ 'it>xe7
gave Black a winning ending in Raz­
myslov-Kengis, Bad Zwesten open 1997.
8 . . . i.xc3+ 9 bxc3 il.. xe4 1 0 ii..xt6 Wixf6
1 1 'i'xe4 l2Jc6 1 2 l2Jf3 0-0 1 3 0-0

Game33
Garces-Keene
Lausanne 1977

1 d4 e6 2 c4 b6 3 e4 il.. b 7 4 l2Jc3 il.. b4 5

ii.. d 3 f5 6 'i'e2

60
 Main L in e with 3 e4 i.b 7 4 d5, 4 EU c 3 i. b 4 5 d5 a n d 4 t'iJ c 3 i. b 4 5 i. d 3
1 3 . . .'i'f4!
Keene's improvement on one of his
own games (13 ... tt:'la5?!, as in Sosonko­
Keene, Haifa Olympiad 1976, could have
been met by 14 li:Je5!) . Compare the posi­
tion after 13 .. .'ii'f4 with Game 29 earlier in
this chapter where the only difference is
that the pawn is on g6 rather than g7.
Keene's handling of the position is alto­
gether more forthright than in that en­
counter.
1 4 J:Iae 1 'ilt'xe4 1 5 J:Ixe4 l:If5 1 6 .l:!.d 1
1 6 . . J:ta5 ! ?
Forcing White t o 'do something' or he
will fall into a passive position.
1 7 d5 t'iJdS 1 8 t'iJg5 lha2 1 9 dxe6 t'iJxe6
20 t'iJxe6 dxe6 21 .!'.:!.xe6 �f7 22 l:.c6 J:!.e8
23 .l:!.xc7+ �e7 24 .l:txe7+ �xe7 25 �f 1
.:a4 26 l:td4 .l:.a3 Y,- Y,
,....----.
Game 34
Zsu . Polgar-Speel man
Dutch League 1993
.._______________..
1 d4 e6 2 c4 �b4 + 3 t'iJc3 b6 4 e4
After 4 dS .i.b7 White has nothing bet-
er than a transposition to the main game
with 5 e4, as 5 .ii.d2 li:Jf6 6 e4 0-0 7 e5
�xc3 8 .ixc3 tt:Je4 9 tt:Jf3 d6 was already
'ery good for Black in Wohl-Lalic, Ubeda
open 1998.
5 . . . �b7
And now for something a little differ-
ent. So far we have majored in 5 .ii.d3 f5.
Here we look at what happens if White
blocks out the bishop with ...
5 d5
However, much to the chagrin of the
unsuspecting, Black is able to upset
White's plans straightaway with ...
5 . . .'ilt'e7 !
The threat is simply to take on d5,
winning a pawn. So White must block the
e-file.
6 i.e2
If 6 �e3 then Speelman in his notes to
this main game recommends 6 ...li:Jf6;
while I favoured 6 ...f5 when faced with
this position in one of my own games.
The centre must crumble: 7 exf5 exd5 8
cxd5 li:Jf6 9 .ic4 (9 li:Jh3 tt:Jxd5 10 't!Vh5+
'it>d8 1 1 'iti>d2 lt:Jxe3 12 fxe3 lbc6 13 'iikg5
.:.f8 14 i.d3 .:.f6 15 li:Jf4 tt:Je5 16 .:.he1 .ie4
17 .:i.ad1 tt:Jxd3 18 tt:'lxd3 .ixd3 0-1
O.lvanov-Chernyshov, Frydek Mistek
open 1996) 9 ... 'fi'e4 10 .ie2 'Llxd5 1 1 .if3
and now in Castricum-King, Bunratty
1997, I played 1 1.. .'1i'xf5, which did the
job, but 1 1 . ..tLlxc3! 12 bxc3 .ixc3+ 13 �fl
Wic4+ 14 .ie2 'il'd5 15 �xdS .ixdS would
have left me with a winning ending.
Instead of 6 .i.e3 White may play 6
tt:'lge2, though he still has problems devel­
oping his pieces after 6 ... exd5 7 exd5 li:Jf6.
Black is ready to batter down the d5-pawn
with either ... b6-b5 or ... c7-c6.
61
En g lish D e fe n c e

6 . . . lt:Jf6 1 2 ..tg5
White's problem is that if she makes a
normal move such as 12 0-0 then after
12 ... i.xc3 13 bxc3 both 13 ... k!.xd5 and
13 ....�:Jxd5 are strong.
1 2 . . . .l:!.he8 ! 1 3 .txf6 gxf6 1 4 lZ'lf4 'i'e5 1 5
'i'd2 ..txc3 1 6 bxc3 li:lb4

7 f3
White's pawn chain is extremely brittle.
It is an easy task for Black to knock it
down and in that case f2-f3 is not a helpful
move: White's knight on g1 has its best
square taken away and another diagonal is
opened around the king. 1 7 �f2?
Instead 7 1Wd4 exdS 8 exdS �e4 gave Losing immediately, though White's
Black the edge in Tartakower-Reti, Goth­ position is unenviable. For a detailed
enburg 1920, but both 8 ... c6 and 8 ... cS are analysis of this game I would direct you to
more dynamic. ]on Speelman's Best Games.
Finally, 7 i.gS is more solid, and if 1 7 . . . li:lxd5 1 8 lZ'ld3 li:lxc3 0-1
White wishes to limit the damage then White resigned as, for instance, 19 .l:the1
after 7 .. h6 it is best to swap: 8 ..ixf6 'i'xf6
. 'iid4+ 20 Wfl .l:txe2 21 .l:txe2 lbxe2 22
9 l:.cl and White ought to be able to come �xe2 ..ia6 wins material. An excellent
out of the opening with his head on his example of how quickly an overextended
shoulders - hardly a great achievement. centre can turn rotten.
7 . . . exd5 8 cxd5 c6 9 dxc6 li:lxc6 1 0 lZ'lh3
d5 1 1 exd5 0-0-0 Game35
Rahman-Speelman
Calcutta 1998

1 c4 b6 2 d4 .i.b7 3 d5 e6 4 e4
It is a different story if White plays d4-
d5 without the knight on c3. After the
standard...
4 . . . ..tb4+
...White does not need to stick the
knight on c3, but can play the more
solid...
5 ..td2
In Allen-Almeida, European Team

62
 M a in L in e with 3 e 4 i.. b 7 4 d5, 4 lb c 3 ii.. b 4 5 d5 a n d 4 lb c 3 ii.. b 4 5 i.. d3
Championship, Haifa 1989, White played
5 li:\d2 W/e7 6 'ii'c2, and now I don't know
why Black didn't play 6 ... exd5 7 cxd5
..ltxd5 8 'il¥xc7 'ii'xe4+ winning a pawn.
5 . . . �e7 6 i.. x b4 'i'xb4+ 7 �d2
7 li:\d2 is more combative, though Black
can capture the pawn with impunity:
7 ... 'ii'xb2 8 �b 1 'ii'f6 9 li:\gf3 li:\e7 10 ..ltd3
li:\g6 1 1 0-0 li:\a6 12 �e1 (Milovanovic­
Chernyshov, Djakovo open 1994) and
now if 12 ... li:\c5 Black stands well.
7 . . . 'iYxd2+ 8 0Jxd2
Black has had few problems against 8
'it>xd2. Now it all depends on your style.
Miles once played 8 .. .f5 (while Spassky
chose the calmer 8 ... li:\f6 9 li:\c3 d6 10 i.. d3
0-0 1 1 ct:Jf3 c6 12 dxe6 fxe6 13 e5 dxeS 14
li:\xe5 ct:Jbd7 15 li:\xd7 li:\xd7 16 f3 .l:i.ad8
and a draw was agreed in Timman­
Spassky, Tilburg 1983) 9 f3 ctJa6 10 li:\c3
li:\e7 1 1 ..ltd3 0-0 12 exf5 exd5 13 �e 1 ct:Jxf5
14 cxd5 li:\b4 15 li:\h3 li:\d6 16 ..lte4 a5 17 a3
ctJa6 1 8 i.. c2 b5 19 li:\e4 ctJc4+ 20 'it>cl
..ixd5 21 .l::i.d 1 li:\e3 22 �d2 i..xe4 23 i..xe4
ctJcS with a clear extra pawn in
H.Williams-Miles, BBC Master Game
1976.
8 . . . f5 ! ?
Thematic, though there are two decent
alternatives:
a) 8 ... li:\e7 (the advantage of this move is
that Black can wait and see how White
commits his pieces before deciding on a
method of attacking the centre) 9 g3 (hor­
rible! White is making too many pawn
moves) 9 ... li:\a6! 10 dxe6 dxe6 1 1 0-0-0 l:td8
12 f3 ctJc6 13 li:\h3 cJJe7 14 ..lte2 li:\d4 15
.l:ihe1 ct:Jb4 16 a3 li:\bc2 17 .l:i.h1 li:\e3 0-1
Hiebel-Lau, German Championship 1994.
Another miniature. There is something
about this opening that induces utter
brainlessness in some players of the white
pteces.
b) 8 ... c6 led to a slightly better position
for Black after 9 dxe6 dxe6 10 e5 c5 1 1 f4
li:\e7 12 li:\gf3 0-0 13 ..ltd3 li:\bc6 14 i.e4
�adS in Sjodahl-Kengis, Vienna open
1996.
9 exf5?
This spoils White's structure. It would
have been better to play 9 f3, which isn't
as serious when the queens are off the
board, though obviously White is under
pressure.
9 . . . exd5 1 0 0Jgf3 0Je7 1 1 g4 h5! 1 2 J:lg 1
dxc4 1 3 i..xc4
13 h3!? hxg4 14 hxg4 b5 is good for
Black.
1 3 . . . hxg4 1 4 l:lxg4 0Jbc6 1 5 ii.. d 3
Or 15 �xg7 li:\xf5 and Black has a clear
plus. Once he castles on the queenside his
pawns are covered and safe, while White's
pawns on h2 and f2 are bound to drop.
1 5 . . . 0-0-0 1 6 b4 l::.df8 1 7 b5 0Jd8 1 8
0Jh4
Or 18 li:\d4 �xh2.
63
En g lish D e fe n c e
1 8 . . . lt:Jf7 1 9 f4 lt:Jd6
Not 19 ... tLlh6 20 .:.g5!
20 0-0-0 lt:Jdxf5 2 1 �xf5 lt:Jxt5 22 lt:Jg6
lt:Je3 23 l!dg 1 lt:Jxg4 24 J:xg4 'it>d8 25
lt:Jc4 .lid5 26 lt:Jce5 �e6 27 l:g3 l:thg8 28
.!:!.a3
28 . . . d6?
Black gets there in the end, but he
could have taken a shortcut to victory
with 28 ... .:.f5 29 l!xa7 i.. d5! and wins.
Black's opening has been a success and we
can finish our analysis of the game. It was
a hard slog to victory.
29 lt:Jc6+ 'it>d7 30 lL:!xf8+ :xt8 3 1 .!ha7
l!xf4 32 lt:Jb8+ Wc8 33 lt:Jc6 .!:If1 + 34
'it>b2 J::!.f2+ 35 Wb1 �f5+ 36 'it>c1 g6 37
a4 .l:!.xh2 38 .l:!.a8+ Wd7 39 l:d8+ �e6 40
J:e8+ Wf6 41 l!f8+ Wg7 42 l:!.e8 l!c2+ 43
Wd 1 l:!.c4 44 :e7+ Wt6 45 l:!.xc7 .lic2+
46 Wd2 .lixa4 47 l:td7 We6 48 l:te7+ '.t>f5
49 'it>d3 J:!.c5 50 lt:Jd4+ \it>g5 51 !lb7
.lixb5+ 52 Wd2 J:!.d5 53 'it>e3 Wg4 54
l:ixb6 �c4 55 J:!.c6 �f1 56 .!:!.c 1 l::te 5+ 57
�2 :e4 58 J:!.d 1 �c4 59 l!d2 d5 60 lt:Jc2
Wf4 61 lt:Je 1 d4 62 lt:Jg2+ �g4 63 lt:Je 1
:t4+ 64 �g2 �f1 + 65 �g 1 �g3 66 lt:Jg2
�xg2 67 l:txg2+ Wt3 68 Wf 1 �e4+ 69
.t>e2 Wt5 70 :g 1 g5 71 :a 1 g4 7 2 :a5+
We4 73 .l:!.a3 I:.f3 74 l:.a4 .t>t4 75 l'rxd4+
Wg3 76 J:!.a4 'it>h3 77 l:ie4 l:!.f8 78 .t>e1 g3
79 :e7 g2 0 - 1
4 ... ..1i.b4+ gives Black easy play, though
if White plays sensibly he should be able
64
to maintain equality. In the final game in
this chapter we shall look at more dy­
namic attempts to break down d4-d5.
Game36
Serper-Hodgson
Groningen 1993
1 c4 b6 2 d4 e6 3 e4 �b7 4 d5 lt:Jt6
If I were looking to embroil my oppo­
nent in wild complications in the hope of
tripping him up then this is the move that
I would choose in place of 4 ... .tb4+ -
which is strong enough as we saw in the
previous game. Apart from anything else,
4 ... tbf6 has the merit of developing a
ptece.
The other ways of having a go at
White's centre are less reliable. Let's just
have a quick run through these alterna­
tives:
a) 4 ... b5 5 a3!? (5 cxb5 is worth a look)
5 ... bxc4 6 i..xc4 c6 7 tbc3 cxd5 8 exd5 tLlf6
9 tLlf3 exd5 10 ..li.a2 .ii. a6 1 1 tLlxd5 ..li.e7 12
..li.f4 0-0 13 tLlc7 was good for White in
Bagaturov-Chachibaia, Ml.Boleslav open
1993.
b) 4 .. .'iih4?! just isn't appropriate here:
5 tbd2 .fi.b4 6 ..li.d3 e5 7 tLlgf3 'fle7 8 a3
..li.xd2+ 9 i..xd2 tLlf6 10 0-0 with two bish­
ops and a pleasant space advantage in Al­
exandrov-Simonenko, Ashkhabad 1990.
c) 4 .. .f5?! is also a bit too gung-ho, as af-
 M a in L in e with 3 e4 i.b 7 4 d5, 4 lb c 3 i. b 4 5 d5 a n d 4 lb c 3 i. b 4 5 i. d 3

ter 5 exf5 .tb4+ 6 .td2 .txd2+ 7 'ikxd2 Black) 19 ... c4 and Black held the initiative.
exd5 8 cxd5 �e7+ 9 .te2 lLlh6 10 lLlf3 c) By the way, 6 e5 doesn't help too
lLlxf5 1 1 0-0 0-0 12 l2Jc3 White had the much: 6 ... bxc4 7 exf6 cxd3 White's centre
freer game in Meister-Shabalov, Podolsk has gone.
1990. 6 . . . exd5
5 i.d3
If 5 e5 then Black's knight hops happily
into the middle, 5 ... l2Je4, and White's cen­
tre is seriously overextended.
5 . . . b5!
This is the move that gives Black's posi­
tion its kick. Having lured the bishop to
d3, Black can gain a tempo when he cap­
tures on c4.
6 cxb5
White makes no attempt to hold his
pawn centre together and grabs a pawn. It
is a compromise, though looking at the
alternatives it may well be a wise decision: 7 e5 l'De4 8 lt.Jf3 a6 9 0-0 axb5 1 0 i.xb5
a) 6 tLlc3 (the most sensible move in the i.c5 1 1 lt.Jc3 0-0
position, though Black has a good answer, 1 1...tLlxc3 12 bxc3 0-0 13 .ltd3 would
which gives me faith in the system) only benefit White. By keeping the knight
6 ... i.. b 4! (6 ... b4?! 7 l2Ja4 takes the pressure on e4 Black retains the initiative.
off the centre and the bishop is prevented 1 2 i.f4
from moving to c5) 7 �f3 bxc4 8 .txc4 Not 12 tLlxd5? c6 and wins.
exd5 9 exd5 'ike7+ 10 Wfl (if 10 l2Jge2 1 2 . . . f6! ?
'ifc5) 10 ... 0-0! 1 1 d6 'ikxd6 12 'ikxb7 .txc3 Black is in a hurry to attack, and why
13 i.. e2 .taS 14 'ii'xa8 'ii'b 4 15 l2Jf3 l2Jc6 16 not? It is best to use the knight on e4. The
a3 'itb3 17 'i¥xf8+ �xf8 18 g3 'ii'c2 19 h3 game plunges into a big mess.
i..b 6 and Black was well on top in Gal­ 1 3 exf6 'ifxf6
liamova-Muhutdinov, Nabereznye Chelny
1993.
b) 6 'i¥b3 looks extremely dodgy. Black
has an excellent response: 6 ... tLla6! 7 tLlc3
(7 'iWxb5 l2Jc5 8 i.. c2 c6 wins the queen)
7 ... l2Jc5 8 �c2 lLlxd3+ 9 �xd3 bxc4 10
�xc4 exd5 11 exd5 and Black's position is
fine with two bishops and the isolated
pawn on d5 to attack. It is possible just to
play ... i.. e 7 and ... 0-0, but in the game
Kamp-Tischbierek, Bad Worishofen 1996,
Black played more ambitiously, aiming at
the d-pawn straightaway: 1 1...�e7+ 12
..lte3 'ii'b 4 13 'it'e2 .te7 14 0-0-0 0-0 15 .l::!.d4 1 4 i.xc7?
'ikd6 16 'i¥d2 c5 17 .1:1d3 l2Jg4 18 .tf4 'ikg6 Serper thinks that 14 lLlxd5 is stronger
19 l2Jh3 (19 h3 l2Jxf2 20 ng3 l2Je4 favours as 14 ... 'ilt'xb2 15 .tc4 l2Jxf2 is met by 16

65
En g lish D e fe n c e

lbe3+ winning; but 1 S ... �h8 would be a
better idea.
1 4 . . . d6 1 5 lL'lxd5 'it'xb2 1 6 i.. c4 lL'lxf2 1 7
'ii' b 1 'ii'x b1
Not 17 ... lbe4+? 18 lbe3+! and wins.
1 8 .l:taxb 1 i.. a 6
18 ... lbe4+ 19 �hl i.a6 is equal.
1 9 i.. x a6 ltJe4+ 20 �h 1 ltJxa6 2 1 i.. b 6
i.. x b6 22 lbxb6
The position has stabilised and it is
about equal, though that doesn't stop
both sides from going for the win.
22 . . . J:!.a7 23 lL'ld4 J:!.e8 24 .l:tb2 ttJac5 25
ltJd5 h6 26 lbc6 J:!.f7 27 l:txf7 Wxf7 28
g3 J:!.c8 29 lL'ld4 J:!.a8 30 lL'lb6?!
30 �g2 is stronger.
30 . . . .l:ta3 3 1 lL'lb5 J:!.d3 32 �g2 J:!.d 1 33
�f3 d5 34 �e2 .l:th 1 35 �e3 J:!.e 1 + 36
Wd4 J:!.d 1 + 37 'ii.?e 3 d4+ 38 �e2 J:!.h 1 39
ltJc4 J:!.xh2+ 40 c;io>t3 ltJg5+ 4 1 'iSi>g4 J:!.xb2
42 lbxb2 d3 43 'it>f4 d2 44 ltJc3 �e6
Y. - Y.
4 ... lbf6 is a funky alternative to
4 ... i.b4+; not only that, there is a chance
that White will find a way to go seriously
wrong. When White plays his pawn to dS
his intention is to block out the bishop on
b7 and in general to squash Black's coun­
terplay. In practice, the opposite often
occurs: pushing the pawn helps Black to
get to grips with the white centre.

66
 M a in L in e with 3 e4 i.b 7 4 d5, 4 t'iJ c 3 i. b 4 5 d5 a n d 4 t'iJ c 3 il. b 4 5 il. d 3

Summary
None of White's ideas in this chapter should cause Black problems; proof of which is the
range of responses available. For instance, against 4 �d3, 5 t'Dc3 and 6 d5 in Games 25-28,
there are several solid and reliable moves (see Game 26), as well as the sharper 7 ... 'ii'h 4! in
Games 27 and 28. Likewise 6 'i'h5+ can be met by the super solid endgame of Game 29,
the queen sacrifice of Game 30, and the halfway house of Game 3 1 . Game 32 only looks
like good news for Black - in practical play the score is heavily in his favour; Game 33
does not differ significantly from 6 'ii'h5+; and when the d4-pawn advances to dS, as in
Games 34-36, Black can have a great time throwing stones at it and watching it crumble.

1 d4 e6 2 c4 b6 3 e4 i.b7
4 t'iJc3
4 d5
4 ... �b4+ - Game 35
4 ... t'Df6 - Game 36
4 . . . i.b4 5 i.. d 3 (D)
5 dS - Game 34
5 . . . f5 6 d 5
6 'i\Vh5+ g6 7 'i\Ve2 t'Df6
8 �g5 (D)
8 ... fxe4 9 �xe4
9 ... �xe4 - Game 29
9 .. .<!2:Jxe4 - Game 30
8 ... h6 - Game 31
8 f3 - Game 32
6 'iie2 - Game 33
6 . . . fxe4 7 i.. x e4 t'iJf6
7... 'i!Vh4 8 'iie2 t'Df6 9 id3
9 ... i.. a6 - Game 27
9 ... 0-0 Game 28
-

8 i.f3 (DJ e5
8 ... �xc3+ - Game 26
9 i.d2 - Game 25

5 i.d3 8 i.g5 8 i.f3

67
 CHAPTER FOUR
Main Line w ith 3 e4 �b 7
4 tt:Jc3 �b4 5 f3 and 4 f3
1 d4 e6 2 c4 b6 3 e4 ..tb 7
In this chapter White meets the attack
on the e-pawn by the black bishop on b7
with the solid-looking f2-f3 either imme­
diately (Games 37-38), or after the knight
goes to c3 (Games 39-42) . As we shall see,
however, Black still has plenty of ways to
upset his opponent.
Game 37
Babu-M i les
Sakthi 1996
._____________..
1 d4 e6 2 c4 b6 3 e4 ..tb 7 4 f3
Superficially, this gives White's centre a
solid look: what can possibly go wrong
68
I
now that the bishop on b7 has been
blunted by this little pawn move? Well,
loads! The shift of the f-pawn opens more
diagonals: there is a check on h4 to bear in
mind; and White can also find himself
embarrassed along the gl-a7 line. Black has
a couple of ways to upset White's centre
of which the most lively is ...
4 . . . f5
Miles first played this as long ago as
1979, so it is appropriate that we look at
another of his games played nearly twenty
years on. Another idea for Black is 4 ... e5,
as in the next main game, where his other
alternatives are also discussed.
5 exf5
5 . . .l2lh6
 Main L in e with 3 e4 i.b 7 4 ti:J c 3 i. b 4 5 f3 a n d 4 f3
This was Miles's original idea. Black
ambitiously sacrifices a pawn, speeding up
his development and hoping to catch
White's king in the middle of the board. It
is possible to play more solidly just by
recapturing, and results have shown that
this is a reliable continuation. For exam­
ple, S ... exfS 6 lbh3 i..b4+ 7 lbc3 'ii'h4+ 8
g3 �f6 9 Jl.gS Wif7 10 Wid3 lbe7 1 1 0-0-0
0-0 12 lbbs lba6 13 .i.f4 lLlg6 14 a3 c6 15
lbc3 lbxf4 16 lbxf4 i.d6 with equal
chances in Vukovic-Schussler, Smederev­
ska Palanka 1979.
6 fxe6
Black was threatening to recapture the
pawn on fS with the knight, so something
must be done. Alternatively, 6 �xh6 has
been tried, though Black then has little to
fear as this helps his development. For
instance, 6 ... 'ii'h4+ 7 g3 Vi'xh6 8 �d2 (or 8
fxe6 Jl.b4+ and ... 0-0 with quick develop­
ment) 8 ... lLlc6!? so that if White exchanges
on h6 the bishop can come to g7 to attack
the d4-pawn. After 9 lbe2 i.b4 10 lbbc3
'ifxd2+ 1 1 'it>xd2 lLJaS 12 lLlf4 Jl.xf3 Black
was already better in Semkov-Bricard,
Sofia 1990.
6 . . . ti:Jf5 !
A remarkable conception. Pawn num­
ber two is thrown in the pot - if White is
foolish enough to accept it. Black threat­
ens a big check on h4.
7 i.f4
The best way to counter the threat,
though two other moves have also been
tried:
a) The original game with this idea con­
tinued instead 7 lbe2 �d6! (the check is
threatened again; in the notes to Game 39
we shall be looking at the same position
but with the bishop already on b4 pinning
the knight on c3 - in which case this pos­
sibility isn't open to Black) 8 h4 (an outra­
geous move, but how else can the check
be stopped? 8 g3 allows 8 ... i.xf3; 8 i.f4
gets hit by 8 ... i..xf4 9 lLlxf4 'i*'h4+ 10 g3
lbxg3; while Fritz suggests 8 exd7+ lLlxd7
9 'ii?f2 - go on, make my day!) 8 ... 0-0 9
lbbc3 'ifif6 (9 ... dxe6 and 9 .. .'ii'e7 have both
been suggested as possible improvements
over the game; in both cases I think that
Black has a tremendous attack - what is
that pawn doing on h4?) 10 cS! i.. e7
(10 ... bxc5? 1 1 e7! 'ik'xe7 12 'ii'b 3+) 1 1 exd7
'il'f7 12 'il'b3 �xb3 (12 ... lLlxd7!) 13 axb3
lbxd7 14 b4 i-xh4+ 15 'it>d1 l:tfd8 and after
more adventures the game Ree-Miles,
Wijk aan Zee 1979, eventually ended in a
draw.
b) It is interesting to see what happens
if White does not prevent the check on h4
with 7 i-d3? �h4+ 8 <,i;>fl lbxd4 9 i.. e4
i.xe4 10 'iVxd4 i.cS 1 1 exd7+ 'it>f7 12 �d2
�d3+ 0-1 Pergericht-Boudre, Val Maubuee
1988. I can't guarantee that if you play the
English Defence you are going to win as
quickly as this, but let's just say your
chances increase!
7 . . . dxe6
A bit tame for Miles. Perhaps he was
thinking back to his game against Ree
where he went just a bit too far in the
opening. I would prefer 7 .. .'i'h4+ 8 g3 'iff6
when White is a long way from bringing
his king to safety.
8 tt::le 2 c5 9 d5!
A good move. Unusually, White keeps
his head and prefers to return his extra
pawn to speed up his development.
69
En g lish D e fe n c e
9 . . . exd5 1 0 cxd5 'i'xd5 1 1 "ii'x d5 �xd5
1 2 tt::l b c3 �b 7 1 3 tt::l b 5 ti:la6 1 4 tt::l g 3
tt::lx g3 1 5 hxg3 l:!.dS
Black gives up a pawn to ease the pres­
sure. A wise decision. 15 ... .i.e7 16 0-0-0
would be wonderful for White.
1 6 �f2 �e7 1 7 l:!.e 1 �f8 1 8 �c4 �d5
1 9 �xd5 l:!.xd5 20 l:!.d 1 tt::l b4 21 l:!.xd5
tt::lx d5 22 tt::lx a7 �f6 23 tt::l b 5 �f7 24 b3
tt::l b4 25 a4 adS 26 l:!.b1 g5 27 �d6 tt::l c 2
Y2 - Y2
White is still better, but Black timed his
draw offer well as the position is still a bit
murky. Nevertheless, I don't think the
opening was at fault. Miles's sacrifice has
more to it than shock value.
Game 38
Pytei-Piasetski
Buenos A ires 1978
1 c4 b6 2 d4 e6 3 e4 �b 7 4 f3
In this position 4 .. .f5, from the previ­
ous game, has a good (or perhaps I should
say an 'interesting') reputation, but there
are some alternatives to consider.
4 . . .e 5 ! ?
To my knowledge this bizarre move
has been played just once, though, curi­
ously, I was recently flicking some pieces
around on my chessboard and came up
with the same idea independently. Before
dealing with this move let's look at the
70
less interesting alternatives:
a) 4 ... .i.b4+ has been played on a num­
ber of occasions but I find it uninspired. If
White plays 5 .i.d2 then he takes the sting
out of the position. The exchange of
pieces robs Black's position of its dyna­
mism.
b) A few games have been played with
4 ... d5, with some success from Black's
point of view. However, in none of these
games did White find a convincing answer.
If he plays 5 cxd5 exd5 6 e5 c5 7 f4 cxd4 8
tt:Jf3 followed by recapturing on d4, then I
think Black has a horrible position.
White's kingside pawn majority is potent.
5 d5
Obviously this is not the most severe
test of 4 ... e5. Let's take a look at what
might happen if the pawn is taken with 5
dxe5 lLlc6:
a) 6 .i.f4 .i.b4+ (6 ... g5 7 .i.g3 .i.g7 also
recovers the pawn) 7 lLlc3 'iie7 and Black
regains the pawn with a good position.
b) 6 f4 is critical. White hangs on to the
booty. Black has several possible options:
b1) 6 ... d6 7 exd6 .i.xd6 8 ltJc3 'iie7
(8 ... 'iih4+ 9 g3 'iie7 10 .tg2 seems satisfac­
tory for White; but 8 ... .i.b4 9 'i'xd8+
l:Ixd8 10 .i.d2 tt:Jf6 offers Black some com­
pensation) 9 .i.d3 lLlf6 10 lLld5 lLlxd5 1 1
cxd5 tLlb4 12 lLle2 0-0-0 with excellent
compensation for the pawn.
b2) 6 ... .i.b4+ 7 .i.d2 li'h4+!? 8 g3 'iie7
 Main L in e with 3 e4 i.b 7 4 [jj c 3 i. b 4 5 f3 a n d 4 f3
followed by .. .f7-f6 or ... d7-d6 depending
on how White reacts.
b3) 6 .. .f6 7 exf6 tLlxf6 8 e5 (or 8 tLlc3
.i.b4) 8 ... tbe4 9 tbf3 .i.c5 with good at­
tacking chances.
I don't suppose this analysis will get
further until someone is silly enough to
play 4 ... e5 again - and their opponent is
mad enough to take the pawn.
5 . . . i.c5
I rather like Black's position (though
that doesn't mean he stands better) . The
queenside is secure and he can prepare to
play the pawn break .. .f7-f5 just like in a
King's Indian; and before White can castle
he must deal with that bishop.
6 [jj c 3 aS
6 ... .i.xg1!? 7 l:i.xgl 'ti'h4+ 8 g3 'ii'xh2 9
I!:g2 'ii'h l would have been interesting!
Apart from the immediate 10 tbb5, caus­
ing a limited crisis, I imagine that White
has good compensation for the pawn if he
simply plays .i.e3, 'ii'd2 and 0-0-0.
7 {jj g e2 {jja 6 8 g3 [jje 7 9 i.h3 0-0 1 0
{jja4 i.b4+ 1 1 [jj e c3 [jj c 8
Obscure. Black is doing just fine if he
sticks to more traditional strategy: 1 1 . ..d6
12 a3 .i.xc3+ 13 tbxc3 tbc5 (or 13 ... .i.c8)
14 .i.e3 .i.c8 (or 14 .. .£5 15 0-0 .i.c8) 15 .i.g2
f5 16 0-0 (if 16 exf5 tLlxf5 17 .i.f2 tbd4!?)
16 .. .f4 17 .i.f2 tbg6. I prefer Black. The
immediate 1 1 . ..f5 is also possible.
1 2 a3 i.. e 7 1 3 0-0 [jj c 5 1 4 i.e3 i..a 6 1 5
b3 {jjxa4 1 6 {jjxa4 b5 1 7 cxb5 i.xb5 1 8
J:!.f2 i.xa4 1 9 bxa4 i.d6
20 f4
This move merely leads to weaknesses
on the kingside and centre. It would have
been better to concentrate on the queen­
side with 20 'ii'd3 tbb6 2 1 'ikb5 'ii'e7 22
l:i.c2.
20 . . . [jj b 6 2 1 fxe5 i.xe5 22 J:!.c 1 d6 23
'ieb3 J:!.b8 24 'iii'b 5 i.f6 25 J:!.fc2 i.g5 26
i.f4 i.xf4 27 gxf4 'ili'f6 28 J:!.f2 1lid4 29
'iexa5 {jjxa4 30 'iii'd 2 'iii'xe4 31 J:!.xc7 [jj c 5
32 i.g2 .=.b1+ 33 J:!.f1 l:Xf1 + 34 'ii< xf1
"ifb 1 + 35 �f2 [jje4+ 36 i.. xe4 'iib6+ 37
'i'e3 "Yi'xc7 38 'iih 3 g6 39 f5 'iii'e 7 40 "Yi'f3
'iii'h4+ 41 �g2 l:!.b8 42 i.c2 'ii'g 5+ 43
'it>h3 J:!.e8 44 i.e4 h5 45 fxg6 fxg6 46 a4
'ii'g4+ 0-1
Even though the 'modern' English De­
fence is now around 25 years old, much of
the theory and many of the ideas are still
very much in their infancy. I hope that
4 ... e5 isn't merely destined to be a foot­
note in history. Go on, play it! And if you
do, you do so with my heartiest recom­
mendation. Golly, I'm so impressed I
might even play it myself.
Game39
Sadler-Kengis
Koge 1997
1 d4 e6 2 c4 b6 3 e4 i.b7 4 {jj c J i.. b4 5
71
En g lish D e fe n c e
f3 f5
The standard move, but other moves
are also seen from time to time. 5 . . .'Yi'h4+
is discussed in Game 41 and 5 ... tt:ie7 in
Game 42.
6 exf5
The only testing move as 6 eS tZ:lh6! is
fine for Black.
6 . . . lt:Jh6
Compare this position with Game 37
after five moves. It is exactly the same
apart from the bishop on b4 pinning the
knight on c3. This sacrifice has been
played on quite a few occasions - but the
verdict remains wide open. Naturally, in
place of 6 ... tZ:lh6 Black can simply recap­
ture the pawn, but that is not the reason
why most of us play S .. .fS. After 6 ...exf5 7
tZ:lh3 tt:ic6 8 i.. gS tZ:lg6 9 'i!Ve2+ 'iii>f7 10 0-0-0
White had a promising position in Piket­
Timman, Wijk aan Zee 1999. The highly
risky 6 .. .'Yi'h4+ is the subject of Game 40.
7 fxe6
7 i..xh6 'ii'h4+ 8 g3 'Yi'xh6 9 'ii'd2!?
(Sokolovs-Litus, Moscow 1992) is a safe
option for White, though Black should be
able to equalise. For instance, 9 ... 'ii'xd2+
10 Wxd2 exfS (or 10 ... tZ:lc6!?) .
7 . lt:Jf5
. .
7 ... 0-0 has been played on a number of
occasions, and it is complicated, but I
think White stands better. For instance,
the game Arbakov-Mihalko, Budapest
72
199 1, continued 8 tZ:lh3 iih4+ and now
instead of 9 tZ:lf2 tZ:lg4!, White should have
played 9 g3 f6e7 10 ..ie2 ik'xe6 1 1 tt:lf4
followed by 0-0 with a clear extra pawn.
8 .tf4
This has become established as White's
best move. The alternatives are less effec­
tive:
a) Hardly anyone goes the whole hog
and takes the second pawn - and rightly
so. The following is a blitz game, so I sup­
pose we can forgive White, but it shows
what Black can do with a massive lead in
development: 8 exd7+?! tZ:lxd7 9 i..f4 (9
'i'e2+ 'it>f7 only encourages Black)
9 .. .'ti'h4+ 10 g3 'iVe7+ 1 1 i.e2 g5 12 'ii'd3
l:!.f8 13 i.. d2 0-0-0 14 0-0-0 tZ:leS 15 'ikb 1
tt:ixd4 16 tZ:ldS .l:txdS 0-1 Hager-Lempert,
'DataGeneral' 1992.
a) I have two games on my database
with 8 tZ:lge2 0-0. In both cases White lost,
i.e. 9 'iVd3 (9 Vib3?! c5! 10 exd7 tZ:lxd7 1 1
dS tt:ieS with a massive attack in Ree­
Morozevich, Tilburg rapidplay 1994)
9 .. .'�'h4+. As ever this is a useful 'destabi­
liser'. White can block the check with 10
g3, which weakens the long diagonal, or
play 10 cJi>d1 dxe6 11 Wc2 tZ:lc6 with ob­
scure complications in Gislason-Budnikov,
Reykjavik open 1994, though with the
king wandering all over the board the
odds are in Black's favour.
8 . dxe6
. .
 Main Lin e with 3 e4 ii..b 7 4 !D c 3 1i.. b 4 5 f3 a n d 4 f3
In view of the poor standing of this
move (currently!), it is worth looking at
alternatives:
a) 8 ... �h4+!? 9 g3 �e7 (9 .. .'fii f6!?) 10
ii'd2 'Yi'xe6+ 11 .lte2 d5 12 0-0-0 0-0 13 a3
.lid6 14 .ltxd6 cxd6 15 l2:lxd5 .lixd5 16
cxd5 1\Vxd5 17 l2:lh3 l2:lc6 18 1i'c3 (not 18
l2:lf4?! 1i'a2 19 1i'c3 l2:la5) 1 8 ... l2Ja5 19 �b 1
ltac8 20 'ii'd3 l2:lc4 2 1 ltcl (it would have
been better to play 2 1 l2:lf4! "i¥b5 22 'ii'c3
l2:lfe3 23 .l:Icl) 2 1 . . .b5 with counterplay in
Szeberenyi-Czebe, Budapest 1997.
b) Sadler had faced 8 ...0-0 a year before
the main game: 9 1\Vd2 dxe6 10 0-0-0 l2:lc6
1 1 d5 lLla5 (or 1 1 . ..exd5 12 1i'xd5+ 'ikxd5
13 l2:lxd5 and wins) 12 lLlh3! 'ike8 13 �el
.i.xc3 14 11ixc3 11ia4 15 b4! 'iVxa2 16 bxa5
exd5 (if 16 ... bxa5 17 .lid3) 17 .lid3 d4 18
1i'b2 'it'a4 (18 ... 'i¥xa5 19 l:Ie5 is no better)
19 lbg5 bxa5 20 l2:le6 �fb8 21 lLlc5 'i¥c6 22
'i¥b5! 'i¥xb5 (or 22 ... 'i¥f6 23 l2:ld7; 22 .. .'ti'g6
23 lbxb7) 1-0 Sadler-Luther, Hastings
1995/96. 23 cxb5 l2:ld6 24 �e7 is hopeless
for Black.
9 'ifa4+
The original game with this line,
Panno-Miles, Buenos Aires 1979, contin­
ued 9 l2:lge2 0-0 10 'iVd2 Wkh4+ 1 1 l2:lg3 (or
1 1 g3 'iVh5) 1 1. . .i.. d6 12 i..xd6 cxd6 13
lLlce2 lLlc6 14 0-0-0 lLlcxd4 15 lbxd4 l2:lxg3
16 hxg3 i¥xh1 17 lLlxe6 and Black was
better, but the position is difficult. Miles
eventually won on time. From a theoreti-
cal point of view this game isn't very rele­
vant as the text move is so strong.
9 . . lt::l
. c 6 1 0 0-0-0
10 d5 is also quite dangerous for Black.
1 0 . . lt::l
. fxd4
10 ... .ltxc3 1 1 bxc3 0-0 12 l2:lh3 l2:la5 13
.ltd3 1i'h4, as in Muhvic-Markovic, Croa­
tian Team Championship 1995, is still a
bit messy, but should be better for White.
1 1 lt::lb 5!
11 lbh3 was successful in Conquest­
Plaskett, Hastings open 1987, but Sadler's
move is more direct and, above all,
stronger.
1 1 . . . 0-0 1 21i..xc7
Not a random capture. White has calcu­
lated that he is winning material.
1 2 . . . 'it'e7
12 .. .'it'g5+ (12 ...'ti'd7 13 l2:lxd4 wins a
piece) 13 f4 'i¥g6 14 l2:lxd4 l2:lxd4 15 �xd4
.ltxg2 16 .ltd3 'i¥e8 17 'ti'xb4 .ltxh1 is win­
ning for White.
1 3 lt::lxd4 lt::lxd4 1 4 �xd4 1i..c 5 1 5 :d7
ilg5+ 1 6 �c2 l:!.acS 1 7 lt::lh 3 'i'g6+ 1 8
�d2
Perhaps 18 �b3!?
1 8 . . . 'i'le8 1 9 l:!.xg7+
19 b3 �f7 20 �xf7 �xf7 must have
looked rather terrifying from White's
point of view. Instead, Sadler goes for the
ending - a safe bet.
1 9 . . . �xg7 20 1i..e5+ WgS 21 VlixeS l:!.cxeS
221i..d 3
73
En g lish D e fe n c e
· With two pawns for the exchange and a
solid position White has a clear advantage.
22 . . . .l:.d8 23 Wc3 ..ltxf3 24 �xh7+ Wxh7
25 lLJg5+ 'ith6 26 lLJxf3 .l:.g8 27 g3 �df8
28 �f4+ �h5 29 lLJe5 ..ltd6 30 .l:!.e 1 l:!.f5
3 1 lLJd3 �xf4 3 2 lLJxf4+ �g4 33 �xe6
. White should be winning, but it isn't
easy and Black just manages to hang on in
the end.
33 . . . l:rh8 34 J:.e2 b5 35 b3 bxc4 3 6
Wxc4 J:tc8+ 3 7 �b4 'lt>f3 38 .l:. d 2 a5+ 39
Wa4 We3 40 l:.e2+ �f3 4 1 .l:.d2 We3 42
l:.e2+ �f3 43 .l:.b2 We3 44 a3 �d4 45
.l:.d2+ <,!;c3 46 .l:.d7 >i'c2 47 lLJd5 .l:.f3 48
b4 axb4 49 axb4 .l:.a8+ 50 Wb5 l:tb8+ 5 1
�c6 .l:.f2 52 b5 .l:.xh2 5 3 b6 :h6+ 54
�c7 l:!.bxb6 55 lLJxb6 l:tg6 56 lt:Jd5 l:txg3
�-�
Sadler's handling of the opening was
powerful. If improvements aren't found
74
earlier then Miles' s pawn sacrifice with
.. .f7-f5 and ... tLlh6 appears dubious - at
least against a well prepared opponent. As
can be seen from the other games in the
notes, those White players who aren't
familiar with the theory often manage to
go horribly wrong at a very early stage.
Gam e40
Piket-Speelman
An do rra Zona / 1998
From a theoretical standpoint this next
game is not particularly relevant, but I
wanted to include it for its sheer verve. It
was played in the last round of a Zonal
tournament where Speelman absolutely
had to play for a win to have a chance of
qualifying for the next stage of the World
Championship. His method was brutal,
but successful. I'm grateful for his notes
on which I have based my comments.
1 d4 e6 2 c4 b6 3 e4 ..ltb7 4 lLJc3 �b4 5
f3 f5 6 exf5 fih4+
Never been seen before, and probably
won't be seen again.
7 g3 'i'f6 8 fxe6lt:Je7 9 �h3 h 5 ! ?
Improvisation. Speelman realised that
his original intention of 9 ... dxe6 would
have just left him with a rotten position:
10 tLlge2 (threatening to castle, when
White has an extra pawn and a good posi­
tion) 10 ... iixf3 l l l:tfl 'it'hS 12 tLlf4 iixd1+
 Main Lin e with 3 e4 il.b 7 4 tD c 3 il.b4 5 f3 and 4 f3
13 \t>xd1 when White wins the pawn on
e6 with an excellent position to boot.
1 0 il.f4
The point of ... h7-h5 was to discourage
White from developing the knight: 10
lbge2 h4 with irritating pressure.
1 0 . . . dxe6
Now if 10 ... h4 then 11 ..ie5 is strong.
1 1 'ii'd 2 lDbc6 1 2 0-0-0 il.xc3 1 3 'i'ixc3
13 bxc3 liJaS 14 'ii'e2 ..ia6 gives Black
some play.
1 3 . . . g5 1 4 il.e3
If 14 ..ixc7 then Black can create some
confusion with 14 .. ..l�.c8 15 ..id6 g4,
though I'm sure White is still better after
16 ..ig2.
1 4 . . . g4! ?
Tricky.
1 5 fxg4 hxg4 1 6 .il..x g4
If 16 ..ig2 lbb4 17 'ii'd2 'ii'g6 18 b3?!
lLld3+ Black wins.
1 6 . . . lDb4 1 7 'ifxb4 il.xh 1
At least Black has the exchange for his
trouble but his position is a wreck.
1 8 lDh3
After 18 h4 �g8 19 ..ih5+ \t>d7 Black is
still in the game - just about.
1 8 . . . 'i'g6 1 9 il.g5 c5!
The only move to keep going. After
this White misses several good continua­
tions, but Speelman always manages to
keep the position unstable enough to give
himself chances.
20 dxc5
For what it is worth, 20 iid2 �d8 21 d5
..ixd5 22 'i*'f4 looks good for White.
20 . . . �xh3 21 'ii'd 2 �h7
22 l:txh 1
It might have been better to play 22
'iVd7+ \t>f8 23 �fl+ lbf5 24 �d6+ \t>f7 25
75
En g lish D e fe n c e
h4 i.. e4 26 'i!Ve5 i.. d3 (26...i.. g2 27 l:i.xf5+
exf5 28 i..xf5 wins for White) 27 i..xf5
i..xf5 (not 27 ... exf5 28 'ifd5+ l!Ve6 29
'ifxd3) 28 g4 with an attack.
22 . . .'ii'e4 23 J:itd 1 ?
23 'ii'd 1 'ii'xc4+ 24 'itib 1 is still messy. If
White can organise himself then Black's
king will be in trouble.
23 . . .'ii'xg4 24 �d7+ 'itis
25 �f4
25 i..xe7+ l':.xe7 26 .l:tfl+ 'it;g8 27 'ii'xe7
'ii'xc4+ 28 'itid2 'ii'xf1 29 ii'g5+ �f7 30
'iih 5+ 'itte7 31 'i!Vh7+ 'i'f7 32 'i!Vh4+ 'itie8 33
'i'h8+ 'i!Vf8 34 'iih 5+ 'itid7 also wins for
Black.
25 . . Jbh2 26 c6 l'ie2 27 ..id2 'ifxc4+ 28
�b1 l'ie4+ 0-1
29 'itia1 .l::th 1 is the simplest. An inspir­
mg game.
Gam e4 1
Sashikiran-Speelman
Britis h Ch., Torqua y 1998
1 d4 e6 2 c4 b6 3 e4 .ltb 7 4 tl:\c3 .ltb4 5
f3
If one is going to play 5.. .f5 with the
idea of sacrificing a pawn then I think one
needs a special mentality: do or die. That's
why it might not appeal to everyone. And
that's the reason...
5 . . . �h4+
...has fou�d more advocates over the
76
years. It has the merit of causing some
confusion in White's position without the
dangers which the ...f7-f5 pawn sacrifice
implies.
6 g3 'i'h5
The check has created a weakness on
the long diagonal, which Black would like
to blast open with ...f7-f5 - as usual. In­
stead of retreating the queen there is an
argument for throwing in the exchange on
c3 first with 6... i..xc3+ 7 bxc3 'i!Vh5.
Normally I would not advocate exchang­
ing on c3 without being pushed by a2-a3
(the dark-squared bishop can be useful
elsewhere) but here White can prevent the
exchange, if he wishes, with 7 i..d2 (see the
next note).
7 tl:lh3
The most forceful move, attempting to
gain time on the queen. White has two
main alternatives:
a) I once faced 7 i.. g2 but this hardly
causes Black any problems. I was able to
get on with attacking White's centre
without delay by playing 7...i..xc3+ 8
bxc3 f5 9 e5 lLlc6 10 lLlh3 lLlaS (White's
pawns are chronically weak and the posi­
tion has great similarities to the Nimzo­
Indian Defence) 1 1 lLlf4 'i'f7 12 c5 lLle7 13
0-0 g5 14 tLld3 f4 15 1if2 lLlf5 16 gxf4 gxf4
17 ..ixf4 �g8 18 'itih1 0-0-0 and White was
getting done over on the kingside in
Vaidya-King, Dhaka 1993.
 Main Lin e with 3 e4 i.. b 7 4 liJ c 3 il.. b 4 5 f3 and 4 f3

b) The reason why I was suggesting that pablanca memorial, Havana 1998.
it might be more accurate to exchange on 8 d5
c3 a move before is that White can play 7 8 dxeS .ixc3+ 9 bxc3 'fixeS is good for
Sl.d2. It doesn't look the most active Black. White's pawns are terribly weak.
move, and so perhaps for that reason it has 8 . . . liJe7 9 liJf2 0-0 1 0 i..e 2 "iig 6 1 1 liJd3
only been tried once. Nevertheless, it is i..xc3+ 1 2 bxc3 d6
slightly irritating for Black, e.g. 7 .. .f5 8
exfS 'iixfS 9 liJbS! (the point) 9 ... �xd2+ 10
1i'xd2 liJa6 11 0-0-0 lLle7 12 .id3 �f6 (the
knight on bS is awkward for Black to deal
with, but then again White has difficulties
developing his kingside - unless he wishes
to sacrifice a pawn) 13 liJh3!? 'iixf3 14 lLJgS
fihs 15 lthfl h6 16 liJf7 (16 liJe4!?) 16 .. 0-0
.

with bizarre complications in Ree-Miles,

Amsterdam 1978.

The posttlon is very similar to the

Nimzo-Indian Defence: Black has the
pawn on c4 to attack; while White has a
space advantage and the bishops. Chances
are balanced.
1 3 g4
Preventing Black from playing .. .f7-f5.
1 3 . . . liJd7 1 4 i..e 3 liJc5 1 5 liJf2 ! ?
If 15 lLlxcS dxcS followed by . . . liJe7-c8-
d6, and . . a7·a5 and ... .ia6 to attack c4.
.

7 . . . e5 1 5 . . . f6 1 6 h4 'ilea 1 7 �b1 a5 1 8 :h2

Most people have tended to reach for i.. a 6 1 9 'i'd2
the f-pawn in this situation, but Speelman
has shown a penchant for this move in the
English Defence. For instance, think back
to his game against Lobron in the first
chapter (Game 9), and check out the notes
to move six to the next main game, Hodg­
son-Bischoff. Instead, here's an example of
the f-pawn in action: 7 .. .f5 8 lLJf4 'iif7 9
dS?! (better was 9 exfS ..WxfS 10 �d3 ..Wf7
1 1 0-0 with an unclear position) 9 ... fxe4 10
fxe4 eS 11 lLJg2 liJf6 12 ..td3 ii.xc3+ 13
bxc3 d6 14 cS bxcS 15 0-0 0-0 16 g4 1i'e8 17
.Ub 1 ..tc8 18 h3 liJbd7 19 �gS "ikg6 20 ..te3
liJb6 with an extra pawn and the advan­ 1 9 . . . liJg6
tage in Borges Mateos-Arencibia, Ca- I'm not sure why Speelman didn't just

77
En g lish D e fe n c e
go for the c-pawn straightaway with
19 . . . 'i'a4.
20 h5 lt::lf4 2 1 ..txf4 exf4 22 'iixf4 Wka4
23 Wid2 i..xc4 24 i..xc4 'i'xc4 25 l:!.b2 a4
26 Wid4 a3 27 J:!.c2 �b5 28 c4 �a4 29
lt::ld 3 "ifeS 30 l:the2 l:ta4 31 �f2 g5 32
hxg6 hxg6 33 �g2 g 5 34 .l:e 1 lt::lx d3 35
'ii'x d3 'ife5 3 6 .:lh 1 �g7 3 7 'i'b3 .:laa8 38
'i'c3 J:th8 3 9 .l:xh8 �-�
The final position is about equal,
though who knows what was going on
earlier? Extremely complicated. Anyway,
it is clear that Speelman's 7... e5 is a viable
continuation for Black.
,...-----.
Gam e4 2
Hodgson-Bischoff
Linares 1996
.1 c4 b6 2 d4 e6 3 e4 ..tb 7 4 lt::lc 3 ..tb4 5
f3
Because 5 .. .f5 and 5 . . .'i'h4+ are such
thematic 'English' moves, it doesn't seem
entirely necessary to me to look any fur­
ther. However, a little research has re­
vealed that there is more than one player
who has played the calm ...
5 . . lt:Je7
.
... in this position. Black is treating the
opening like a Nimw-Indian. He might
still go for .. .f7-f5, hitting the centre, or he
might play against the doubled c-pawns (if
he manages to double them that is) .
78
6 ..td3
White can make an attempt to avoid
the doubled c-pawns if he wishes with 6
'i'b3, for example, though he then runs
the risk of falling behind with his devel­
opment. I have seen one game where
Black tried 6 ... tl'la6 and then ... f7-f5, but I
prefer 6 ... tl'lbc6 - there is more chance of a
'hit' against the queen. 6 ... c5 also isn't bad.
Finally, 6 i..dl is less good as 6 .. .f5 is a bit
embarrassing.
6 . . . lt::lb c6
Instead, twenty years ago Speelman
tried his favourite move 6 ... e5!? 7 a3 (if 7
dxeS tl'lbc6 8 f4 d6 9 exd6 'i'xd6 Black has
compensation for the pawn) 7 ... exd4 (I
prefer 7... i..xc3+ 8 bxc3 tZ'lbc6 9 tl'le2 d6
followed by attacking the c4-pawn in tra­
ditional 'Nimw' style with ...'i'd7, ... tl'laS
and ... i.. a6) 8 axb4 dxc3 9 bxc3 tl'lg6 10
tZ'le2 tZ'lc6 1 1 0-0 tl'lceS 12 i..c2 0-0 13 cS d6
14 cxb6 axb6 15 l:f.xa8 i..xa8 16 f4 and
White was a bit better in Christiansen­
Speelman, Hastings 1978.
7 lt::lg e2 ..txc3+ 8 bxc3 d6 9 0-0 'i'd7
If we compare this position with one
from the Samisch variation of the Nimw­
Indian, then Black is doing fairly well. He
doesn't have to worry about his knight
being booted away from f6, and his solid
centre enables him to attack the c4-pawn.
1 0 f4
Or 10 i.. a3 i.. a6 1 1 cS i..xd3 12 'i'xd3
 Main Lin e with 3 e4 il.b 7 4 liJ c 3 Ji.. b 4 5 f3 and 4 f3
dxc5 13 .i:tad1 lt:Ja5 14 'ii'c2 'i'c6 15 d5 exd5
16 exd5 'i!Vd7 17 c4 0-0 with a clear extra
pawn for Black in Gropp-Ruzele, Berlin
1997.
1 0 . . . liJa5 1 1 f5 exf5 1 2 exf5 0-0-0
The only place to be now that the f­
pawn has advanced.
1 3 d5
Perhaps White had missed earlier that
13 f6 gxf6 14 .l:Ixf6 lt:Jxc4 would be rather
strong for Black. The g2-pawn is going to
cave m.
1 3 . . . ii..a 6 1 4 c5 ii..x d3 1 5 c6 'lieS 1 6
'i¥xd3 l2Jxd5 1 7 ifxd5 'i\Vxe2 1 8 'ifxf7
a.hf8 1 9 'iid 5 a.de8 20 ii..f4 'i\Ve4 2 1
'i¥xe4 l:txe4 22 g 4 l2Jxc6 2 3 \i;>g2 llc4 24
a.t3 h5 25 h3 hxg4 26 hxg4 g6 0- 1
The loss of pawn number two is the fi­
nal straw for White.
5 ... lt:Je7 is a respectable alternative to
the (potential) madness of the alternatives
we have already considered.
79
En g lish D e fe n c e

S u mmary
I relish games where my opponent plays f2-f3 - all those lovely open diagonals! Miles's
sacrifice in Game 37 seems sound, though when repeated with the knight on c3 and the
bishop on b4 (Game 39) it is less clear; Sadler's play in that game is convincing. But Black
does not have to go in for this pawn sacrifice if he doesn't wish to, as proved by Games
38, 41 and 42.

1 d4 e6 2 c4 b6 3 e4 �b7

4 lLlc3
4 f3
4 .. .f5 - Game 37
4 ... e5 - Game 38
4 . . . �b4 5 f3 (DJ f5
5 .. .'it'h4+ - Game 41
5 ... tbe7 - Game 42
6 exf5 (DJ lLlh6
6 . .'ii'h4+ - Game 40
.

7 fxe6 (DJ - Game 39

5 f3 6 exf5 7 fxe6

80
 CHA PTER FIVE I
1 d4 e6 2 c4 b6 3 e4 i.b7
In this chapter we shall look at games in
which White defends the e-pawn with
'i!Vc2, either immediately (Games 43-46) , or
combined with lLlc3 (Games 47-49) .
Again, Black does not lack ways to create
counterplay, though I must say that from
a theoretical viewpoint the immediate 4
'ii'c2 is perhaps the most solid way for
White to play once he has advanced his
three pawns into the centre.
,...----..,
Gam e43
Remlinger-Rogers
Phila delphia 1986
.______________..
1 d4 e6 2 c4 b6 3 e4 i.b7 4 "ii'c 2
This solid move has the advantage,
compared to 4 f3, of keeping the rest of
the pawns intact and healthy - though the
queen is sometimes open to a hit on c2;
and the pawn on d4 is a little vulnerable.
From this position, attention has centred
on .. .
4 . . . 'l/Hh4
... ever since it was first played in the
Polugayevsky-Korchnoi Candidates match
in 1977, and that is the move I shall be
concentrating on.· For alternatives to
4 ... "ii'h4, see Game 46.
s tt:ld2
The least troublesome method of de-
fending the e-pawn - in theory. The alter-
natives are:
a) The thoughtless 5 ..td3?! produced
another miniature in Brunk-Berebora,
Berlin open 1998: 5 ...lLlc6 6 d5 (6lLlf3 'it'g4
7 0-0 lLlb4 8 'i*'e2 lLlxd3 9 �xd3 .txe4 is
obviously good for Black) 6 ...lLlb4 7 'iie2
lLlf6 8 e5lLlg4 9 �d2 'if'g5+ 10 �d1lLlxf2+
1 1 'if'xf2 'i*'xcl+ and White could have
·resigned here.
b) 5 d5 engages prematurely, e.g. 5 .. .f5
(or 5 ...lLlf6!? 6 ..td3 lLla6!) 6 exf5 exdS 7
cxd5 .txd5 8 'i*'xc7 with an unclear posi­
tion in Koukolik-Tratar, Ceske Bude­
jovice open 1995.
81
En glish D e fe n c e

c) 5 tt::l c3 allows a pin with 5 ... �b4, 21 i..xf5 tt::le7 22 i..d3 'bg6 23 f3 'i'h1+ 24
though this isn't too bad (see Games 47- �f2 .l::tf8 25 'i'g5 tt::lh4 26 �e2 'i'xf3+ 27
49) . <3id2 tt::le4+ 28 i..xe4 Vi'xe4 29 \\Vg4 l:.f2+
5 . . . �b4 0-1 Quinn-Speelman, Dublin Zonal 1993.
The pressure increases. Finally, 8 'Dxd2 isn't very testing, as af­
ter 8 .. .'�Jf6 White has problems on the
long diagonal.

6 �d3 f5
6 .. .'ii' g4 is a decent alternative - see
Game 45. 8 . . . �g4 9 lt:Je5 'iix g2 1 0 0-0-0 fxe4
7 lt:Jgt3 �xd2+ The laudable attempt to exchange off
This was the move originally played in some pieces with 10 ... i..xe4 unfortunately
Polugayevsky-Korchnoi. For the alterna­ loses to 1 1 l:f.hg1 i..xd3 12 'i!Vxd3 'ife4
tives, 7. . . 1!Vh5 and 7 .. .'ii'g4, see Game 44. (12 . . .'�'xf2 13 'Df3 and Black must jump
8 .txd2 through hoops to save the queen) 13 'ii'g3
By far the best move. Instead Po­ g6 14 'Dxg6, as in Vujatovic-Le Blancq,
lugayevsky played the ugly 8 �fl?! He Lloyds Bank open, London 199 1 .
must have been completely psyched out 1 1 �e2
by Korchnoi's opening. After 8 .. Ji'h5 9
it.xd2 tt::l £6 10 exf5 ..ixf3 1 1 gxf3 tt::l c6 12
..ic3 Black has a pleasant choice
a) 12 . . . 0-0 13 .l::te 1 'ii'h 3+ 14 �e2 .l::tae8 15
<3id1 e5! 16 dxe5 tt::lxe5 17 i.. e2 tt::lxf3 18
'ifd3 l:i.xe2 19lhe2 'ifg2 20 kthe1 'Dxe1 21
Wxe1 '*'xh2 (21 . . .Vi'h1+! 22 �d2 Vi'xh2 is
even stronger) 22 l:!.e7 'iYg1+ 23 �e2 Vi'g4+
24 �e1 h5 25 'ii'g3 'ii'xg3 26 fxg3 and
White had some drawing chances, though
he eventually lost in Polugayevsky­
Korchnoi, Candidates match 1977. The
first time that the English Defence had
succeeded at the highest level. This posttlon was analysed by
b) 12 ... 0-0-0 13 l::te 1 l::th e8 14 .l:.e2 Vi'h3+ Korchnoi's team in the 1977 match im­
15 �e1 'i'xf3 16 .l::tg 1 'i'f4 17 �xg7 exf5 18 mediately after the game against Po­
'ii'd2 .l::txe2+ 19 �xe2 .l::te8+ 20 �fl 'ifxh2 lugayevsky (see above) . According to

82
 M a in Lin e with 3 e 4 ii.. b 7 4 li:J c 3 ii.. b 4 5 ii c 2 a n d 4 ii c 2

Messrs Keene, Tisdall and Plaskett in their
book The English Defence Korchnoi was
reluctant to repeat the whole line as he
thought this position was too dangerous
for Black. In that book some fairly de­
tailed analysis of this position is given,
though without any examples from actual
play. Since then a few players have been
brave enough to take on the black side
(with mixed results) and those games help
us to form an assessment of the variation.
1 1 . . li:Jc6
.
.l:txf2 exf2 17 �xg7 �e4 18 'ii'd 1 .l:tg8 19
�xf6 l:tg1 20 �f1 with a clear edge for
White) 16 .l::i.h g1 'iVh2 17 �xg7 l:i.g8 18
'i'd1 tbe4 19 �h5+ �e7 20 .l:tf7+ 'it>d6 2 1
tbf3 and White is winning. Some very
'Fritz'-inspired variations, I admit; but
anyway, my gut feeling here is that Black
is lost.
1 2 li:Jxc6 ii..xc6 1 3 d5

Recommended in the book above as

one of the best continuations, but still
doubtful. Alternatively, 1 1 . ..tbf6 is a plau­
sible move:
a) 12 �e3 'ii'h3 13 .U.dg1 tbc6 (13 ... h6
did not turn out well in Rogers-Lau, Wijk
aan Zee 1989, after 14 .l:txg7 tbc6 15 tbg6!

0-0-0 16 tbxh8 .l:txh8 and White ought to

convert) 14 .l:tg3 'ii'f5 15 .l:tg5 tbxd4 16 'ii'd 1
tbxe2+ 17 ifxe2 'ii'h 3 18 �hg1 d6 19 l:txg7
0-0-0 20 tbf7 'iVxh2 when in Webster­ 1 3 . . . ii..b 7
Adams, Prestwich 1990, with so many Perhaps 13 ...exd5!? 14 cxd5 �xd5! 15
pawns for the exchange, Black shouldn't l:Ihg1 'i'xh2 (the lads above didn't consider
be worse, but the position is still ex­ this one, but 15 ... 'iVxf2 16 �h6 is hopeless)
tremely complex. and Black is still in with a shout, e.g. 16
b) 12 .l:thg1 was recommended in the �c3 tbf6 with an unclear position.
above book as practically winning for 1 4 ii..h 5+ g6
White:
b 1) 12 .. ."ii'xf2 was the only move they
analysed, and that is indeed good for
White: 13 �h6! 'ii'xh2 (13 ... g6 14 .l:tgf1
'ii'xh2 15 �g7; or 13 . . . gxh6 14 �h5+) 14
�xg7 �g8 15 �xf6 !Ixg1 16 �h5+ it'xh5
17 .l:.xg1 and White is winning - Keene,
Plaskett and Tisdall.
b2) However, they failed to consider
12 .. .'it"xh2, when it looks like .White
should be able to win somehow, though it
is difficult to prove conclusively, e.g. 13
l!h1 (13 �xg7 tbc6 14 �g5 0-0-0!)
13 . . .11Vxf2 14 �h6! (this is the trick that 1 5 ii..c 3
makes the difference) 14 ... e3 (not Instead, 15 'iic3! looks like a winning
14 ... gxh6? 15 �h5+) 15 l::tdf1 (rather than move, e.g. 15 ... tbf6 (or 15 ... 0-0-0 16 'ili'xh8
15 .l:thf1 ..te4!?) 15 . . .'it'g3 (or 15 ... �xh1 16 gxh5 17 .l:thg1 and White is winning) 16

83
En glis h D e fe n c e
'Wxf6 :f8 1 7 'Wg7 :f7 (if 1 1. . .0-o-o 1 8
d:xe6) 1 8 'Wh8+ �e7 1 9 'We5 gxh5 20 i.. g5+
and wins.
1 5 . . . 1\fg S+ 1 6 'it>b1 e5 1 7 f4 'ifxf4 1 8
ltdf1 'ii' g 5 1 9 'ifxe4 d 6 20 .i.d 1 0-0-0 2 1
:hg 1 'fl/e7 22 .i.d2 ltf8 23 'i'g4+ c,t;b8 24
�g5 'I/Vg7 25 '11Vd 7 'ii'x d7 26 ltxf8+ .i.c8
27 �g4 'i'g7 28 lbc8+ \t>b7 29 ltf 1 h5
30 ltcf8 hxg4 31 lt 1 f7 'llixf7 32 J:!.xf7
lbh6 33 l:Ig7 lbf5 34 lbg6 J:!.xh2 35 .i.c 1
g3 36 l:Ig5 g2 0-1
7 ... i.. xd2+ is interesting but, if you play
it, you (and your opponent!) must be tac­
tically alert at all times. Before trying it
out I would do some thorough research
on the pawn sacrifice played in this game.
My feeling is that it is good for White,
though the evidence so far would suggest
'case not proven' .
Gam e44
Remlinger-Kengis
Ga us dal 1991
1 d4 e6 2 c4 b6 3 e4 �b7 4 'ifc2 'flih4 5
lbd2 �b4 6 �d3 f5 7 lbgf3
This is all as in the previous game, but
instead of 7... i..xd2+, Black plays:
7 . . . 'ifh5
7 .. .'ii'g4!? is also playable, e.g. 8 0-0
i..xd2 9 tbxd2 (not 9 i..xd2? fxe4 10 tbe5
exd3 and wins) 9 ... tbc6 10 f3 'Wh4 1 1 "ikc3
tbf6 (1 1 . . .'iif6!? 12 e5 'Wh4 13 f4 tbh6 14
84
tbf3 �h5 15 b4 tbe7 with counterplay was
already quite strange in Wilson-Avni,
Hastings Challengers 1995/96) 12 b4 fxe4
13 fxe4 0-0 14 d5 tbg4 15 h3 tbf2 16 .tc2
tt:Jd4 17 'iVxd4 tLlxh3+ 18 gxh3 'ii'g3+ 19
'it>h 1 �xh3+ 20 'it>g1 "i!Vg3+ 21 'oith 1 lh-lh
Knechtl-Heilinger, Austrian Team Cham­
pionship 1996.
8 a3
The critical 8 0-0!? tbf6 9 exf5 i..xd2 10
tbxd2 "iVg4 is unclear according to Kengis.
8 . . . .¥i.xd2+ 9 �xd2 lbf6 1 0 exf5 �xf3 1 1
gxf3 lbc6
1 2 �c3
Or 12 ..te3 e5 (12 ... "iVxf3 is stronger, as
White's king is caught in the middle, e.g.
13 l:i.g1 exf5 14 .txf5 0-0-0 with an attack)
13 d:xe5 tLlxe5 14 0-0-0 tt:Jxd3+ 15 �xd3
"iVxf3 16 nhg1 0-0-0 17 nxg7 nhf8 with
good counterplay in Webster-King, British
Championship, Eastbourne 1990.
1 2 .. .'ifxf3 1 3 J:l.g 1 0-0-0 14 'ife2 'iff4 1 5
'ife3 'i'xh2 1 6 0-0-0 exf5 1 7 .l:!.h 1 'iVd6
1 8 �xf5
Black is a pawn up, but matters are
complicated by the presence of the two
bishops. However, the knights have solid
protection and weak squares to aim for, so
Black is already better.
1 8 . . J:tde8 1 9 'fl/g3 'i'xg3 20 fxg3 lba5
21 .txa5
see following diagram
 Ma in Lin e with 3 e4 i.. b 7 4 ti'J c 3 i.. b 4 5 Wi c 2 a n d 4 'ii c 2
After 2 1 cS lZ:lc4 Black also has the bet­
ter game.
21 . . . bxa5 22 �de 1 g 6 23 i.c2 �d8 24
�e5 �xe5 25 dxe5 ti'Jh5 26 g4 ti'Jf4 27
�d2 h5 28 �e3 g5 29 gxh 5 �xh5 30
.Uxh 5 ti'Jxh5 3 1 �3 ti'Jg7 32 �g4 ti'Je6 33
c5 <i;;e 7 34 i.f5 a4 35 c6 dxc6 36 i.c2
ti'Jd4 37 i.xa4 �e6 38 'it.>xg5 �xe5 39
'Oi>g6 c5 40 �f7 �d6 4 1 �e8 ti'Jc6 42
i.xc6 'it.>xc6 43 �e7 c4 44 �e6 a5 45
�e7 �b5 46 �d7 �a4 47 �xc7 �b3 0-1
A successful game from Black's point of
view, but 8 a3 is not the best move. 8 0-0
must be investigated.
Gam e4 5
Parker-Sher
Co pen hagen 1996
1 d4 e6 2 c4 b6 3 e4 i.b7 4 �c2 fih4 5
ti'Jd2 i.. b4 6 i.. d 3
Instead of 6 .. .fS, in the past five years
there have been some games with ...
6 . . :iVg4
. . . tempting White into weakening his
structure with g2-g3, but in every game I
have seen White has made the conces­
ston ...
7 �f1
Although White cannot castle, he will
gain time by attacking Black's queen. 7 g3
is met by 7 .. .f5!
7 . . .f5
Black must stir it up to get anything at
all from the position.
8 h3
White has two main alternatives here:
a) 8 lZ:lgf3 i..xd2 and now:
a1) 9 i..xd2 fxe4 10 lZ:leS exd3 and Black
wms.
a2) 9 lZ:lxd2 and:
a2 1) 9 ... lZ:lc6 10 f3 (not 10 'iii'c3 'ii'd 1
mate!) 10 .. .'ii'h 5 (or 10 ... 'ii'h4 11 'ii'c3 lZ:lf6
with counterplay) 1 1 Vi'c3 fxe4 12 lZ:lxe4
0-0-0 and Black has sufficient counterplay.
a2i) 9 ... ll:Jf6 10 h3 (or 10 f3 'ii'h4 1 1 exfS
lZ:lc6 12 lZ:lb3 lZ:lb4 13 'i"c3 lZ:lxd3 14 'iixd3
0-0 with counterplay) 10 ... 'ii'h4 (perhaps
10 ... 'ii'g6!?) 1 1 exfS 0-0 12 fxe6 'iixd4 13
lbb3 'iVeS 14 exd7 lZ:lbxd7 and again Black
has enough counterplay.
a3) 9 lZ:leS 'iih4 10 lZ:lf3 'iihS 1 1 i..xd2
ll:Jf6 12 exfS i..xf3 13 gxf3 lZ:lc6 14 i.. c3,
when via quite a different move order we
have transposed to Game 43. Here
Korchnoi played 12 .. 0-0 and achieved a
.
good position, while Speelman went the
other way and achieved the same .
b) 8 f3 was, from a theoretical view­
point, successful in the game Levitt­
Ehlvest, New York 1994, so it is worthy
of some close investigation. After 8 .. .'ii'h4
9 exfS we reach the first major crossroads:
b 1) 9 ... lZ:lc6. In principle, this is the
move I would like to play, just developing
quickly to get some attack, but it appears
85
En g lish D e fe n c e
to fail: 1 0 fxe6 dxe6 1 1 l2Je2 0-0-0 1 2 a3
�xd2 13 �xd2 lLlxd4 14 l2Jxd4 'ii'xd4 15
i.. c3 'iixd3+? (perhaps 15 .. .'ikd7! 16 i..xh7
l2Jf6 and Black has some compensation for
the pawn, though for some people's taste
it may not be sufficient! - DK) 16 'iixd3
.l:txd3 17 i..xg7 and White is winning -
Jonathan Levitt.
b2) 9 ... exf5!? 10 �xf5 l2Jc6 1 1 'iVe4+
(not 1 1 lLlb3 Vi'e1 mate!) 1 1...'ii'xe4 12
i..xe4 lLlf6 with some compensation in
view of Black's superior development.
b3) And finally, the actual game,
9 .. .'.Wxd4 10 lLle2 (10 a3 �c5 1 1 lLlb3 is
even stronger, according to Levitt)
10 .. .'�h4 1 1 fxe6 lLlf6 12 lLlg3 0-0 13 lLlde4
i..xe4 14 �xe4 l2Jc6 15 exd7 lLlh5 16 i.. d5+
Yl-Yl Levitt-Ehlvest, New York 1994.
Here, as Levitt himself points out, instead
of agreeing a draw, he should have played
16 ... Wh8 17 i.. f4! with a clear advantage.
8 . . .'i!Vg6
The most sensible retreat, but 8 .. .'ii'h5 9
exf5 lLlc6 (or 9 ... l2Jf6) is also possible.
9lt::lg t3 i.xd2 1 0 lt::le 5 �f6 1 1 i.xd2 lt::le 7
1 2lt::lf3 h 6 1 3 :e 1 0-0
If White were castled then he would
simply have a great position, but freeing
the rook from h1 is not going to be easy.
1 4 d 5 ! lt::la 6 1 5 i.c3 'iit 7 1 6lt::le 5 �e8 1 7
b4 c5 1 8 a3
Or 18 b5!? l2Jb4 19 �xb4 cxb4 with
counterplay.
86
1 8 . . . cxb4 1 9 axb4 d6 20 lt::lt3 fxe4 2 1
:xe4 e 5 22 'iie 2 :11 2 3 :h4 lt::lf 5 24
:h5 g6 25 i.xf5 gxh5 26 i.e6 lt::lc 7 27
lt::lx e5 tt:lxe6 28 lt::lxf7 �xf7 29 dxe6 'iig 6
30 t3 :e8 31 �f2 .l:txe6 32 'iid 2 h4 33
:e 1 'i'g3+ 34 'it>t1 :xe 1 + 35 'iixe 1
'i'xe 1 + 36 i.xe 1 i.a6 37 b5 i.c8 38
i.xh4 i.e6 39 i.e7 i.xc4+ 40 �f2 i.xb5
4 1 i.xd6 a5 42 i.c7 a4 43 i.xb6 a3 44
i.d4 i.d7 45 g4 �-�
White played the opening sensibly (in
general, quite rare in the English Defence)
and it was Black who was struggling to
counter White's initiative throughout the
game.
Gam e4 6
Levitt-Short
Calcutta 1998
1 d4 e6 2 c4 b6 3 e4 i.b7 4 'iic 2 g6
A logical move. The cl-pawn is now
unprotected so Black aims to attack it. In
the next chapter which deals with a2-a3 by
White, we will see that the double fi­
anchetto (hippopotamus system) has come
into fashion again for Black.
Let's take a quick look at the alterna-
tives to 4 .. .'ikh4 and 4 ... g6:
a) It is a pity that 4 ... l2Jc6 doesn't quite
work due to 5 l2Jf3 lLlb4 6 'iVe2 when
White holds the e-pawn and next move
boots the knight back.
b) 4 ... �b4+ 5 l2Jd2 l2Jf6 6 �d3 c5 7 dS
b5 8 dxe6 bxc4 9 exf7+ �f8 10 'ii'xc4 d5 1 1
'i*'c2 dxe4 1 2 i..e2 l2Jc6 1 3 a3 l2Jd4 1 4 'ii'd 1
e3 15 fxe3 i..xd2+ 16 �xd2 �xg2 and
Black was winning in the game Ro­
gozenko-Teske, Dresden open 1996. That
was jolly good fun, but after 4 ... �b4+, life
is far less interesting if White blocks with
the bishop. In that case White is left with
his big space advantage and Black's coun­
terplay is less dynamic. Compare with 4
i..d3 �b4+ from Chapter 2 - it is really
rather similar.
 Main Lin e with 3 e4 i.. b 7 4 l:i'J c 3 i.. b 4 5 'il c 2 a n d 4 '¥i c 2
5 l:i'Je3 .li.g7 6 i.. e 3 l:i'Je7 7 0-0-0
If 7 tt:lf3 0-0 and Black is ready for .. .f7-
f5.
7 . . . 0-0 8 f4 f5 9 e5
Or 9 tt:lf3 b5!?
9 . . . d 5 1 0 e5 bxe5 1 1 dxe5 g 5 ? !
This seems pointless as Black isn't going
to use the kingside anyway. Instead,
l l.. .tt:lbc6 doesn't look too bad for Black­
compare with the game.
1 2 g3 tt::lb e6 1 3 l:i'Jf3 g4 1 4 l:i'Jd4 l:i'Jxd4
1 5 i.. xd4 l:i'Je6 1 6 .li.b5 l:i'Jxd4 1 7 .l:.xd4 e6
1 8 i.. d 3 .iilf7 1 9 �b 1 .li.fS 20 l:i'Je2 i.. e S
21 .iila4
White has a clear advantage and h�s op­
ponent lacks counterplay.
21 . . . a5 22 l:i'Jd4 .iilb 7 23 l:e 1 i.. d 7 24
'i¥e3 "i/ie7 25 l:i'Jb3 .iilb a7 26 .li.e2 .li.eS 27
l:i'Jd4 .iilb 7 28 �a 1 h5 29 l:i'Jb3 .iilb a7 30
l:i'Jd4 'i'd7 3 1 h4 i.. a 6 32 .li.xa6 .iilx a6 33
'i'b3 i.. e 7 34 .l:!.e3 �f7 35 "ii'e 2 .l:!.6a7 36
l:.e3 'fifeS 37 l:.b3 .li.dS 38 'i'd3 �e7 39
�b 1 �f7 40 'it>e1 'it>e7 4 1 'it>d 1 .iilb 7 42
�e2 .:tabS 43 l:.xb7+ J:.xb7 44 a3 .iilb S 45
b4 "it'b7 46 l:i'Jb3 axb4 47 axb4
47 �xb4! is winning according to
Levitt, e.g. 47 .. .'iVa7 48 .lhb8 iixb8 49
'i¥a6 �d7 50 tt:ld4 1Wc7 5 1 ViaS.
47 . . J;[a8 48 J:!.xaS 'ifxaS 49 'it>b2 �-�
4 ... g6 wasn't too successful here, but I
don't think it was a true reflection of its
worth. White players often underestimate
this type of system.
Gam e4 7
lvanchuk-Sadler
Monaco (blin dfold) 1998
1 e4 b6 2 d4 e6 3 l:i'Jc3 .li.b7 4 e4 i.. b4 5
'i'e2
There have been quite a few games with
the queen move in this position, though I
don't think it is particularly good here.
5 . . . ikh4!
Once again the best response. White
faces immediate difficulties, and it is sur­
prising that Ivanchuk went in for this line,
though it wasn't a 'proper' game, as such.
6 d5
6 i.d3 is the subject of Game 49.
6 . . . i.. xe3+
Sadler likes to go his own way. The ma­
jority of players have gone for 6 .. .f5 here;
87
En g lish D e fe n c e
for that see the next game.
7 bxc3 li:Jf6
This treatment is more solid than 6 .. .f5
(since the pawn structure is more intact)
but it still has some bite. White's pawns
on c4 and e4 are a cause of concern.
8 Sl.d3
Not 8 e5? 'ife4+! 9 'ifxe4 tZ:lxe4.
8 . . . li:Ja6 9 li:Jt3 'i!ig4!
If 9 .. .'iih5 10 i.. a3! and the knight does
not make it to the dream square.
1 0 0-0 li:Jc5 1 1 h3 'ilig6 1 2 li:Je5 'ilih5 1 3
f4 li:Jxd3 1 4 'i'xd3 d6 1 5 li:Jt3 0-0-0
With the bishop on d3 gone the pawns
become weaker.
1 6 Sl.e3 I:the8 1 7 I:tae 1 Sl.a6 1 8 li:Jd4?
Allowing a tactic, forgivable consider­
ing it was blindfold chess; in the same
tournament Karpov left his queen 'en
prise' to Sadler. Instead 18 a4 exd5 19 exd5
BB
l:te7 followed by doubling is still more
comfortable for Black than his opponent.
1 8 . . . exd5 1 9 exd5 'iWxd5 20 'iWc2 Sl.xc4
2 1 .l:If3 c5 22 li:Jb3 �e7 23 li:Jc1 .l:!.de8 24
�h2 'iic 6 25 'i't5+ 'iid 7 26 'iig 5 li:Jd5 27
f5 f6 28 it'g3 li:txe3 29 li:tfxe3 �xe3 30
�xe3 li:Jxe3 31 'i'xe3 'i'xf5 32 'i'e8+ <3;c7
33 'i'e7+ 'i'd7 34 'i't8 'it>b7 35 a3 d5 36
'itg3 d4 37 cxd4 cxd4 38 'ii'b4 'li'c7+ 39
'ith4 'i't4+ 40 g4 'i'f2+ 0-1
Gam e4 8
Urban-Maciejewski
Lubni ewi ce 1993
1 d4 e6 2 c4 b6 3 li:Jc3 Sl.b7 4 e4 Sl.b4 5
'i'c2 'i'h4 6 d5 f5
More vigorous than Sadler's 6 ...li:lf6.
7 .td2
7 exf5 has been seen in several games
and, while the situation is messy, there is
no doubting that Black's position is fun­
damentally more sound than his oppo­
nent's due to his superior pawn structure.
For instance, 7 ... exd5 8 tZ:lf3 (or 8 cxd5
tZ:lf6! 9 i..d2 i..xc3 10 �xc3 0-0 1 1 0-0-0
'iff4+ 12 i..d2 'i'xf2 13 tZ:lh3 'iic5 14 'iixc5
bxc5 15 i.. e3 d6 16 tZ:lf4 tZ:lbd7 17 i..e2
.l:i.fe8 18li:le6 i..xd5 19 .l:i.he1 i..xe6 20 fxe6
tZ:le5 and Black was clearly on top in Link­
Kengis, Bern open 1995) 8 .. .'ii'e4+ 9
'iixe4+ dxe4 10 tZ:ld4 and now both
10 ... tZ:lc6 and 10 ... tZ:le7 are strong. Black
 M a in Lin e with 3 e4 ii.b 7 4 tiJ c 3 ii.b 4 5 fi c 2 a n d 4 ii c 2

has a nice ·pawn structure and better de­

velopment.
7 . . . fxe4 8 0-0-0
Two other moves have been tried here:
a) 8 g3 as usual, is asking for trouble,
e.g. 8 ... "iie7 9 i.g2 tZ:lf6 10 0-0-0 exdS 1 1
cxdS tZ:la6 1 2 tZ:lxe4 tZ:lxdS 1 3 tZ:lc3 �xc3 14
�xc3 tZ:ldb4 15 �xb7 tbxc2 16 �xa8 c6
and Black was winning in Behrhorst­
Birmingham, Hamburg 1986.
b) 8 tLlxe4 is reasonable, and after
8 ... exd5 9 cxdS:
b 1) 9 . . . ..ixd2+ 10 tZ:lxd2 �xdS (or even
10 ... 'iie7+!?) 1 1 'iixc7 tbc6 is unclear. . 6 . . .f5 is sound. That doesn't mea.ri. that
b2) 9 . . . ..ixd5 10 �xb4 iixe4+ 1 1 'iixe4+ Black stands better, but the pressure is
�xe4 12 0-0-0 tZ:lf6 13 �e1 when White already on White to prevent his centre
had some compensation for the pawn in from collapsing.
Rotstein-Bricard, Cannes open 1992.
Gam e4 9
Kryzius-Piesina
Ra dvilis kis 1995
1 c4 b6 2 lLlc3 i.b7 3 e4 e6 4 d4 i.. b4 5
'i'c2 'ifh4 6 i..d 3
This is seen more often than 6 d5 from
the last two games, but it is not clear
whether it "is any better.

8 � .lt:lf6
. 9 i.. e 2 �xf2
Considering that g2-g3 was a threat,
why not?
1 0 lt:lh3 'i'Vc5 1 1 4Jxe4 4Jxe4 1 2 'ifxe4
i..x d5 1 3 i.h5+ �d8 1 4 i.xb4 'i'Vxb4 1 5
'iih 4+ �ea 1 6 J:!.d4 i.. x g2 1 7 l:thd 1 i.c6
1 8lt:lg5 �b7 1 9 4Jf7 l:tf8 20 4Jd8+

see following diagram

A slightly insane sequence, but Black's 6 . . . f5

king is quite safe on b7, and he has taken
lots of pawns.
20 . . . l:txd8 21 'i'xd8 a5 22 J:!.h4 h6 23
'i'h8 g5 24 l:h3 'i'xc4+ 25 �b1 i..e4+ 26
�a 1 lZ:Jc6 27 �xh6 lZ:Jb4 28 b3 'iWc2 0-1
... and the usual problems ensue on the
long diagonal.
7 g3
Most players plump for this one, but
the critical move has to be 7 tZ:lf3 ..ixc3+ 8

89
En g lish D e fe n c e
'ii'xc3 (not 8 bxc3? 'ii'g4 9 0-0 fxe4 and
wins) 8 ... 'i\i'g4 9 0-0 (forced) 9 ... fxe4 10
tt:Je5 Vih4 (not 10 ... 'i¥h5? 1 1 ..ic2 tt:Jf6 12
..id1 �h4 13 .te3 tLlc6 14 d5 tLle7 15 dxe6
dxe6 16 ..ia4+ c6 17 'ii'a3 and White was
clearly better in Garcia-Gonzalez - Forin­
tos, Montpelier 1985) 1 1 .te2! ( 1 1 ..ic2?,
which was mentioned by Keene, Plaskett
and Tisdall, is a dubious piece sacrifice; for
instance, 1 1 . . .d6 12 .ta4+ c6 13 d5 dxe5 14
'ii'xe5 'ile7 15 dxe6 tLlf6 16 ..igS 0-0 17
l:Iad1 tLla6 and Black is winning) 1 1 ...d6!
( 1 1 . . .tLlf6? 12 i.e3 0-0 13 'iiVc l is rather
embarrassing) 12 tLlg4 tLlf6 13 tLlxf6+ 'ili'xf6
and Black is a pawn to the good.
Finally, I was once lucky enough to re­
ceive the following gift: 7 exf5?? il..xg2 8
fxe6 .txh1 with a won position in Kachi­
ani Gersinska-King, Oviedo rapidplay
1992.
7 . . . 'i¥h5
8 f3
The alternatives aren't terribly promis­
ing either:
a) 8 .te2 'ifif7 9 f3 fxe4 10 fxe4 tLlf6 1 1
d5 0-0 1 2 tLlf3 'ii' g 6! 1 3 i.d3 'i\Vh5! 1 4 0-0
tLla6 15 a3 i.xc3 16 bxc3 tLlc5 17 i.e3
lLlxd3 18 'i¥xd3 exd5 19 exdS tLlxd5 20
cxdS l:txf3 0-1 Farago-Miles, Hastings
1976/77. One of the first shocking games
with the system on the international cir­
cuit (see the Introduction).
b) 8 tLlge2? isn't too clever due to
90
8 .. .'ilkf3 9 0-0 i.xc3 10 tLlxc3 tLlc6 11 'ili'd1
tLlxd4 and Black was winning in Chi­
bukhchian-Lempert, Yerevan open 1996,
although he messed it up and only drew.
8 �c6
...
Fine, though I quite like the idea of
playing very simply, as Miles did, with
8 ... fxe4 9 fxe4 tLlf6 10 tLlge2 tLlc6 and now:
a) 11 a3 i.xc3+ 12 bxc3 e5 13 d5 tLlaS 14
c5? (White probably wasn't expecting the
fatal blow to come from this side of the
board) 14 ... tLlb3! 15 l:Ib1 (or 15 iVxb3 �f3)
15 ... tLlxc5 and, a little prematurely, but
understandably, White resigned here in
Komarek-Chetverik, Policka open 1996.
His kingside is a mess and he will proba­
bly lose his e-pawn.
b) 1 1 0-0 .td6!? 12 rJtg2 tLlb4 was good
for Black in Wiedenkeller-Berg, Sollentuna
open 1995. White's problem in these
variations is that his kingside is so weak;
the bishop on d3 can't return to cover the
light squares.
9 d5
Alternatively, 9 .te3 e5 (perhaps
9 ... fxe4!? 10 i.xe4 tLlf6) 10 dS tLld4 1 1
i.xd4 exd4 12 a3 fxe4 1 3 fxe4 ..ltxc3+ 14
bxc3 dxc3 15 'ifxc3 tLlf6 and Black already
had the initiative in Kharlov-Gretarsson,
Leeuwarden open 1995.
9 �d4 1 0 'ilt'f2 fxe4 1 1 �xd4
...
Hopeless. Black makes a simple sacrifice
and it is all over. Instead, if 1 1 fxe4 eS 12
 M a in Lin e with 3 e4 .ii. b 7 4 l'i'lc3 .ii. b 4 5 "iic2 a n d 4 "iic2
.te3 tt:Jf6 with the initiative or 1 1 i.xe4
exd5! ? ( 1 1 . . .e5 is reasonable) 12 1Yxd4 dxe4
13 'iexg7 exf3! with a stonking attack.
1 1 . . . exd3 1 2 �xg7 0-0-0
I love it when Black's king reaches the
queenside in the English Defence - four
pawns and a lump of rock on b7; it is so
safe.
1 3 'i'xhS exd5 1 4 .ii.e 3 lieS 1 5 0-0-0
l'i'lh6 1 6 �xe8+ �xeS 1 7 .ii.x h6 dxc4 1 8
i.d2 d5 1 9 l'i'lh3 d4 20 t'i'Je4 c3 2 1 l'i'lf4
.111.x
. e4 22 fxe4 'ifa4 23 t'i'Jxd3 'ifxa2 24
�c2 �a4+ 25 �c1 cxd2+ 26 �b1 .lil..a 5
27 J:lhf1 �b3 28 l'i'lc 1 dxc 1 � + 29 Wxc 1
d3 30 l:tf2 �c4+ 3 1 �b1 d2 32 l:tfxd2
.ii.xd2 33 l:.xd2 'li'xe4+ 34 �c 1 "iic4+ 35
�d 1 a5 0-1
There is no way that White would ever
plan to go into positions like these after 6
.ltd3. He is already on the back foot.
91
En g lish D e fe n c e

S u mmary
The greedy idea of grabbing the pawns in Game 43 looks unsound. However, Game 44
is a more reliable way of playing the line, while 6 ... 'i!tg4 of Game 45 seems a little dubi­
ous, though it is as yet relatively unexplored. The double fianchetto idea of Game 46 is
yet another way of coping with 4 ii'c2 and seems reasonable. Games 47-49 are only fun
for Black.

1 d4 e6 2 c4 b6 3 e4 .ltb 7

4 l2Jc3
4lli'c2
4 ... "ii'h4 5 tt:Jd2 ..tb4 6 i.d3
6 .. .f5 7 lt:Jgf3 (D)
7... i.xd2+- Game 43
7... 'i\Yh5- Game 44
6 . . ."ii' g4- Game 45
4 ... g6- Game 46
4 . . . i.. b4 5 'ifc2 'i'h4 (D) 6 d5
6 Sl.d3 - Game 49
6 . . . i.. xc3+
6 .. .f5 -Game 48
7 bxc3 (D) - Game 47

7 lt:Jgf3 5 . . . Wih4 7 bxc3

92
 CHA PTER SIX I
White Plays an early a2-a3
1 d4 e6 2 c4 b6 3 a3
In this chapter I'm going to be looking
at games where White plays a2-a3. A
clever little move. In many of the games
we have considered so far White's centre
has come under enormous pressure from
the bishop on b4; so why not prevent it
from going there? The move has a respect­
able relation: the Petrosian variation of the
Queen's Indian Defence (1 d4 l2Jf6 2 c4 e6
3 l2Jf3 b6 4 a3) which has the same aim.
Indeed, a transposition is possible after
3 ... l2Jf6 4 tLlc3 .tb7 5 tLlf3, so just on those
grounds alone I'm not sure that 3 a3 rep­
resents a refutation of the system. But
that's too gloomy an assessment! As ever
Black has plenty of ways to stir up trou­
ble.
The most modern method of dealing
with a2-a3 is to play a double fianchetto,
reaching a kind of 'hippopotamus' forma­
tion. Miles has championed this system,
though I would say it takes skill to avoid
being squashed (see Games 50-52) .
In Games 53-58 Black instead prevents
his opponent from occupying the centre
with all three pawns by playing .. .f7-f5,
reaching an unusual kind of Dutch forma­
tion. This is probably the most respectable
way of coping with a2-a3; at the very least
Black has several plausible options avail­
able if he plays in this way.
In Games 59 and 60 Black offers a
transposition to the Queen's Indian De­
fence by playing ... l2Jf6, but White avoids
this by playing his pawn to d5. However,
Black does not suffer in these games, so
White should just accept the transposition.
Let's start with a radical idea from the
true master of the English Defence.
Game 50
Hellsten-Miles
Malmo 1996
1 d4 e6 2 c4 b6 3 a3
Naturally 3 l2Jc3 can be met by
3 ... .tb4, when 4 e4 j_b7 transposes to
Chapter 3 and 4 e3 (intending tLle2) gives
Black the option of either transposing to a
Nimzo-Indian with 4 ... l2Jf6 or crippling
White's pawns with 4 ... .txc3+ 5 bxc3 .ltb7
followed by .. .f7-f5.
3 . . . g6
'I suspect that 3 ... g6 may well be best. 3
a3 really "wastes" a move to prevent
... i.b4 so Black instead aims for the best
diagonal, hoping to get a little mileage out
of ... e7-e6. ' - Tony Miles. A simple argu­
ment, but in chess sometimes one has to
93
En g lish D e fe n c e
'take a view'. Previously Miles had ex­
perimented with a double fianchetto, but
only after playing .. .f7-f5 to reach a form
of Leningrad Dutch.
4 lt:Jc3 i.g7 5 li:Jf3 lt:Je7 6 e4 i.b7 7 .te2
0-0 8 0-0
White has opted for sensible, straight­
forward development, and that can't be
wrong, but I don't think it is the way to
squash the system. Check out the next
two games for attempts at a refutation.
s . ts
. .
Miles writes here: 'Already it is hard to
suggest moves for White. Hmm. I'm not
so pessimistic about White's chances. Inci­
dentally, 8 .. .f5 isn't forced. Black may also
play a hippopotamus formation: 8 ... d6 9
i.. e3 tt::l d7 10 d5 tt::lc 5! 1 1 tt::l d4 'ii'd7 12 b4
tt::lxe4 13 tt::l xe4 exd5 14 cxd5 lbxd5 15 i.. f3
l:tae8 and, with the knight in the middle
en prise and ... c7-c5 coming, Black had
excellent compensation for the piece in
Savchenko-Teske, European Cup, Brati­
slava 1996.
9 e5
White doesn't have to play this. For in­
stance, 9 exf5 lbxf5 10 d5 transposes into
an old game of Plaskett's (he only fi­
anchettoed his king's bishop after White
had moved the pawn to d5) :
a) 10 ... i..xc3! ? 1 1 bxc3 tt::l a6 ( 1 1 . . .'ii'f6!?) .
b) 10 ... tt::l a6 1 1 'ikc2 and Black has suffi-
cient play.
94
c) 10 ... 'ife7 1 1 i.. g5 'iff7 12 'iic2 tt::l a6 13
b4 c5 14 dxc6 i..xc6 15 l:tad1 l:tac8 16 tt::le4
tt::lc7 17 tt::ld6 tt::lxd6 18 l:txd6 i..xf3 19 Si.xf3
tt::lb 5! with a tremendous attack in Akes­
son-Plaskett, l}ergsjo 198 1.
9 . . . h6
1 0 J:te1
10 h4 d5 ' .. .leaves the h-pawn looking
silly' - Miles. That's as maybe. But one
thing I am sure of is that it is best to pre­
vent Black playing ... g7-g5, which, as we
will see, gives Black a highly dangerous
attacking formation on the kingside.
Anyway, take a look at this, played a little
later than the main game and see what you
think. Miles's strategy is remarkable; it
reminds me of the old Arabic formations.
Although it appears that Black is cramped,
in fact his co-ordination isn't too bad, and
the central pawn majority a long-term
asset: 1 1 exd6 cxd6 12 ..if4 tt::lc 8 13 'ii'd2
'it>h7 14 l:tae1 tt::l d7 15 i..d 1 lieS 16 l:te3
tt::lf8 17 I:i.fe1 l':te7 18 a4 a6 19 b3 l:ta7 20
.l:Id3 l::rd7 21 Si.g3 i.. a8 22 tt::lh2 tt::le7 23 i..f3
i..xf3 24 tt::lxf3 e5 25 dxe5 dxe5 26 l:txd7
l:txd7 27 'ii'c 1 f4 28 Si.h2 tt::lc6 29 tt::ldS 'iiih 8
30 'i'b 1 l:td6 31 g3 g5 32 hxg5 hxg5 33 'ii'f5
l::i. g6 34 gxf4 gxf4+ 35 Wh1 when the situa­
tion was still unclear, but Black certainly
wasn't worse in view of the bishop on h2
in Xu Jun-Miles, Tan Chin Cup, Beijing
1996.
1 0 . . .g 5 !

White is definitely worse. His centre
will crumble and his minor pieces are re­
stricted.
1 1 d5lt:lg6 1 2 i..f 1 g4 1 3lt:ld4 i..x e5 1 4
dxe6
Not 14 �xh6 I:!.e8!
1 4 . . . 'it't6 1 5 ..txh6 l1e8 1 6 lt:ldb5 dxe6
1 7 'iid 2lt:lc6 1 8 'it'g5
18 �g5 'i1Vh8 19 I:i.xe5 'i1Vxe5 20 l::i.e 1 'i¥g7
is hopeless for White.
1 8 . . . J:!.e7 ?
Miles thinks that White can resign after
18 .. hc8
J 19 a4 a6. That might be a little
premature, but it certainly isn't very
pleasant, e.g. 20 'Lla3 <it>f7 followed by
..lh8
J and it is difficult to see how White
can defend on the kingside.
1 9 J:txe 5 <to>f7 ! ?
Miles played this extraordinary move
because his opponent was in time trouble,
but objectively 19 ... 1Wxg5 20 �xg5 'Llcxe5
2 1 �xe7 'Llxe7 22 tt'lxc7 .l:tc8 23 'Llxe6
'Llxc4, with some compensation for the
pawn, is stronger.
20 J:!.ee 1 a6 21 :ad 1
2 1 tt'le4! fxe4 22 'Llc3 is much stronger,
when Black is in trouble.
21 . . . axb5 22 ifxf6+ �xf6 23 cxb5 lt:lce5
24 ..te3lt:lt7 25 ..tc4 .l:!.h8 26 g3 lt:ld6 27
..ta2 ..tt3 28 h4lt:lxh4 29 lt:ld5+ �f7 30
i.. d 4
see following diagram
Wh i t e Pla ys a n e a rl y a 2-a 3
Or 30 'Llxe7 'Llg6 3 1 �xe6+ �eS and
wms.
30 . . . l:!.h6 31 J:!.d3 exd5 32 J:!.xe7+ �xe7
33 J:!.e3+ �d7 34 gxh4 .l:!.xh4 35 .l:!.xf3
gxf3 36 ..tt6 l1g4+ 37 �f1 'it>e6 38 ..td8
tt:lxb5 0-1
Amazing stuff. When anyone says that
chess is being played out and that there is
nothing new any more, I would direct
them towards some of Miles's games.
Nevertheless, White players have managed
to come up with better replies to his sys­
tem since this first try.
Gam e 51
Milov-Miles
Bi e/ 1996
1 d4 e6 2 c4 b6 3 tt:lc3
By the way, if White doesn't occupy
the centre with his three pawns then Black
gets a very easy game. For example, 3 a3
g6 4 tl:lc3 �g7 5 ltlf3 'Lle7 6 e3 �b7 7 �e2
0-0 8 0-0 d5 9 cxd5 ltlxd5 10 ..id2 tt'ld7 1 1
l:tcl c5 12 dxc5 ltlxc5 1 3 b4 ltld7 1 4 ltlxdS
�xd5 15 �c3 �xc3 and a draw was agreed
in Li Wenliang-Miles, Tan Chin Cup, Bei­
jing 1996 .
3 . . . ..tb7 4 a3 g6 5 tt:lt3 ..tg7 6 e4lt:le7 7
i..e 3
White does better to assume a more ag­
gressive stance with his pieces than �e2
95
En g lish D e fe n c e

and 0-0, though this one might not b e the 1 3 0-0

best. The following year Milov tried 13 'i'd2!, preventing ... 0-0 and preparing
something different - see the next game. 0-0-0. would have been interesting. It
Alternatively, 7 d5 isn't convincing: would cut across Black's plans.
7 . . . 0-0 8 �g5?! f6 (surely 8 . . . h6 - this. is
generally a move that Black would like to
play, so I don't like �g5 at all) 9 �f4 c6
(too much) 10 d6 !DeS 1 1 b4 with a clear
plus for White in Sinanovic-Sandler, Yere­
van Olympiad 1996.
7 ii.d3 is a sensible move, as the kind of
King's Indian after 7 . . . 0-0 8 0-0 d6 (perhaps
8 . . .f5!?) 9 �e1 !Dd7 10 �e3 eS 1 1 dS is not
what Black should be aiming for. Black is
behind with his kingside attack, and is
weak on the queenside (the c6-square) .
The course of the game Astrom-Sandler,
Yerevan Olympiad 1996, demonstrates 1 3 . . . g5!?
this: 1 1 . . .h6 12 !Dd2 f5 13 f3 lbf6 14 b4 f4 Black could have just castled and played
15 �f2 g5 16 c5 h5 17 l:tcl lbg6 18 �e2 as in the game I mentioned above, but
�c8 19 �a4 l:tf7 20 i.bS g4 21 i.c6 i.d7 Miles obviously had good memories of his
22 cxd6 cxd6 23 lbb5 ii.xc6 24 dxc6 and game against Hellsten, played just a few
White was clearly better. weeks before, and goes for . . . g7-g5 again. It
is just a bit too much here. Nevertheless,
to refute it, White has to play with great
energy and precision.
1 4 hxg5 ltlg6 1 5 ..tg3
White has plenty of ways to go wrong:
15 gxh6? �xf3! 16 hxg7 .i':txg7 17 �g3 �b7
and the rook on a7 comes good; and 15
i.e3 �xf3 16 �xf3 hxg5 is also dangerous
for White.
1 5 . . . hxg5 1 6 .l:!.e 1 ?

7 . . . f5 8 e5 h6 9 h4!
Milov appreciates how important it is
to prevent . . . g7-g5.
9 . . . d6 1 0 exd6 cxd6 1 1 ..tf4 a6
Perhaps 1 1. ..0-0.
1 2 ..te2 .!:!.a7
This looks bizarre, but if you think
back to Xu Jun-Miles in the notes to the
last game, then you will see what Black
might be about. If 12 . . . 0-0 then 13 'i&'b3!?

96

16 �d3! would have been a more pre­
cise move order. After 16 ... g4 17 l:!.e1 play
transposes back to the game, but Black
could have exploited this inaccuracy
with ...
1 6 . . . g4?
... 16 .. .£4! 17 �d3 (not 17 �h2? g4 18
li:Jd2 Wt'h4 and wins) 17 ...li:Jf8 18 d5 (or 18
�h2 g4) 18 . . .fxg3 19 fxg3 �c8 and Black
should survive the attack and win with his
extra piece. Once again the rook on a7
comes in handy.
1 7 �d3! �f7
After 17 ... gxf3 18 lhe6+ �f7 19 d5
White has a powerful initiative.
1 8 d5 J.. x c3 1 9 dxe6+
19 :xe6!? would also have been very
strong:
a) 19 ... gxf3 20 'iiVxf3 �f6 21 �xf5 li:Jf8
22 �xd6 and White is clearly on top.
b) 19 ...�xb2 20 l':!xd6 'ii'e7 (or 20 .. :it'c8
21 li:Jg5+) 21 .l::f.e6 'ii'd8 22 �xf5 li:Jf8
(22 . . . li:Je7 23 li:Jg5+ �g8 24 �xg4 is win­
ning for White) 23 li:Je5+ ii.xe5 24 �xe5
with a powerful attack.
b) 19 ... i.. f6 20 i..xf5 li:Jf8 2 1 l':!xd6 'i'e7
22 li:Jd4 and again White is on the attack.
1 9 . . . �g8 20 J.. xf5 gxf3
After 20 . . . i..xf3 21 gxf3 j_xel (or
21..Jhh7 22 e7 .l:i.xe7 23 bxc3 and wins) 22
j_xg6 White is also clearly better.
21 �xg6 fxg2 22 �f7+ �g7 23 �g4+­
�f8
Wh i t e Pla ys an e a rl y a 2 - a 3
24 f3
24 �h5! is the simplest: 24 ... �f6 25 e7+!
�xe7 26 �f5+ j_f6 27 j_xd6+ Wg8 28
'i¥g6+ and wins.
24 . . . �c8 25 bxc3 J:!.xt7 26 'it'd4! 'i*'f6
Or 26 ... .l::f.fh7 27 e7+.
27 �xd6+ 1 -0
27 ... �g7 is met by 28 �e5.
Miles went just a bit too far in this
game, but it does demonstrate the range of
Black's possibilities. I especially like the
way the rook on a7 is able to swing into
action.
Gam e 52
Milov-Miles
New York 1997
1 d4 e6 2 c4 b6 3 a3 g6 4 e4 �g7 5
tt:Jt3
You might notice how Miles delays de­
veloping his queen's bishop, and there is a
subtle reason for this, which the following
game demonstrates: 5 li:Jc3 li:Je7 6 �e3
(but now with the bishop on e3 . . .) 6 .. d5 .
( ... is possible, as the knight can move to f5 ·
to hit the bishop) 7 cxd5 exd5 8 e5 0-0 9 f4
and in this kind of position the bishop is
better on c8, covering the f5- and g4-
squares. After 9 ... c5 10 li:Jf3 li:Jbc6 an un­
clear position was reached in Halldorsson­
Miles, Reykjavik 1998.
5 . . . '2Je7 6 l2Jc3 �b7 7 h4
97
En g lish D e fen c e
A year down the line from the previous
game, Milov is better prepared. He knows
that in certain positions ... g7-g5 can be a
key part of Black's strategy, so he moves
to prevent it from the outset (and h4-h5
can also be useful for White) ; and he posi­
tions his pieces so that he is ready to meet
.. .f7-f5 if necessary.
7 . . . h6 8 .ltd3 ! d6 9 .lte3 tt::l d 7 1 0 fke2 a6
This time Miles adopts a 'hippo' forma­
tion and awaits developments.
1 1 J:!.c1 c5
1 1...0-0?! 12 'it'd2 'it>h7 13 h5! wouldn't
be too good for Black.
1 2 d5 e5
Another King's Indian type pos1tton
and, again, not particularly good for Black
as the bishop is misplaced on b7. It should
be on c8 to cover the kingside. Milov has
cleverly avoided committing his king
98
(which would have given Black something
to aim for) but now it runs to the other
side of the board.
1 3 h 5 ! tt::lf 6
After 13 ... g5?! 14 lbh2 the knight goes
via f1 to g3 to look at f5.
1 4 hxg6 fxg6 1 5 tt::ld 2 .lieS 1 6 f3 h5 1 7
�d 1 .lth6 1 8 Wc2 .ltxe3 1 9 'ii'xe3 tt::le g8
20 iig5
White has permanent pressure on the
kingside.
20 . . . J:!.h6 21 g4 "ike7 22 'ii'e 3 J:!.h7 23
gxh5 tt::lx h5 24 tt::l e 2 J:!.a7 25 f4 .ltg4 26
fxe5 'ii'x e5 27 J:!.cg 1 tt::l gf6 28 l:!.h4 .ltxe2
29 i..x e2 tt::lg 7 30 J:!.xh7 tt::lx h7 31 J:!.xg6
tt:Jt5 32 'i'h3 lLld4+ 33 �b1 tt:Jts 34 J:t9s
>t>t7 35 J:!.g2 >io>eS 36 'ii'c8+ �e7 37 i.. h 5
1 -0
In this game Milov showed a good way
to counter the Miles system, using con­
tainment and patience.
Gam e 53
Salov-Short
Ma dri d 1997
1 d4 e6 2 c4 b6 3 a3 f5
The most usual continuation for Black
at this point has been 3 ... i.b7 (or 3 .. .f5 4
lbc3 i.b7) 4 lbc3 f5 5 d5 lbf6 6 g3 lba6
and I'll be taking a look at this line later
(Games 54-56) . But first I thought it was
worth looking at Short's individual treat-
 Wh i t e Pla ys a n e a rl y a 2 - a 3

ment of the opening - he comes up with a White could have prevented Black's
subtle idea. next move with 9 'ikc2!?, but he is falling
4 tt:Jc3 ti:Jf6 5 d 5 dangerously behind with his development
If 5 g3, inviting a transposition to the on the kingside. For instance, 9 ... c6 10
'main line' I mentioned above with i.g2 (or 10 dxe6 dxe6) 10 ... cxd5 11 cxdS
5 ... �b7 6 dS tt:la6, Black does have an al­ exdS 12 tt:lxdS tt:lc6 with the initiative.
ternative policy, as Short showed in an­
other of his games: 5 ... c6!? 6 i.g2 dS (the
game has transposed into a Stonewall
Dutch where the extra moves a2-a3 and
... b7-b6 have been thrown in - not a bad
deal for Black as his opponent can no
longer trade bishops on a3, and Black's
bishop stands well on b7, or sometimes
a6) 7 tt:lf3 i.e7 8 �g5 0-0 9 0-0 �b7 10 l:1cl
h6 11 �xf6 i.xf6 12 e3 tt:ld7 and chances
were balanced in C.Hansen-Short, Euro­
pean Rapidplay Championship, Cap
d'Agde 1996.
5 . . . i.a6 ! ? 9 . . tt:Je4!
.

The exchanges free Black's position.

1 0 �c1 tt:Jxc3 1 1 i.xc3 i.xc3+ 1 2 �xc3
'tWf6 1 3 'tWd2 i.b7
Work done, the bishop returns.
1 4 ti:Jf3 a5 1 5 0-0 tt:Ja6

. This reminds me of 4 ... i.a6 in the

Queen's Indian (1 d4 tt:lf6 2 c4 e6 3 tt:lf3 b6
4 a3 �a6) . However White defends the
pawn he must make a small but significant
concession which Black can exploit.
6 b3
If 6 'ikb3 Black could try 6 ... c5!? when
the queen is stuck defending the pawn,
while 6 ... �c5 is also possible. 6 e3 doesn't
present any difficulties either: 6 ...exd5 7
tt:lxdS (or 7 cxdS �xf1 8 'itxf1 with coun­
terplay) 7 . . . �d6 with equality.
6 . . . g6 7 i.b2 i.g7 8 g3 0-0 9 i.. g 2
Chances are about equal. Black is solid
on the queenside; the knight is about to
.leap into the beautiful cS-square; and the
queen on f6 is well placed, a legacy of the
exchanges a few moves ago.
1 6 iid4
Salov obviously didn't like the activity
of Black's queen and hastens to exchange

99
En glish D e fe n c e
it. If 16 lZld4 lbc5 1 7 b4 axb4 1 8 axb4 lbe4
19 i.xe4 fxe4 20 dxe6 dxe6 21 "ii'e3 l:ta2
Black is very active.
1 6 . . . �ae8 1 7 dxe6 .l:!.xe6
The pawn structure is very similar to a
Queen's Indian; Black has no problems.
1 8 'ifxf6 llfxf6 1 9 l2Jd4 ..i.xg2 20 �xg2
litd6 21 e3 'it>f8! 22 h4 r3;;e 7 23 li1b1 llf8
24 J:tc2 c6 25 �f3
Or 25 b4 axb4 26 axb4 lbc7 27 c5 bxc5
28 bxc5 �d5, intending . . . lZle6 with equal­
ity.
25 . . . l2Jc5 26 �e2 litdf6 27 l2Jf3 h6 28 b4
axb4 29 axb4 l2Je4 30 :l.a 1 c5! 31 bxc5
bxc5 32 l2Jd2 g5 V2 - V2
The draw becomes inevitable with
every pawn exchange: 33 hxg5 hxg5 34
l:th 1 l:ta8 is dead level.
,....----.
Gam e 54
P . Cramling-Gulko
Pam plona 1996
.______________.
1 d4 e6 2 c4 b6 3 a3 i.b7 4 l2Jc3 f5 5
d5
This is the most common move here,
although the solid 5 lZlf3 lZlf6 6 g3 is also
seen (see Game 57) and 5 f3 is also possible
(see Game 58) . Note that 5 lbf3 lZlf6 6 dS
is less effective. In general, if White is go­
ing to play d4-d5, then it is best not to
have the knight on f3, since the cl-pawn
needs protection. For example, 6 ... g6
1 00
(6 ... lba6 is 'normal' with a likely transpo­
sition to the note to White's eighth move
in the main game after 7 g3 lbc5 8 i.g2) 7
dxe6!? (7 g3 i.g7 8 i.g2 0-0 9 0-0 lZla6
transposes to the note to White's eighth
move in Game 56) 7 ... dxe6 8 'iVxd8+ �xd8
9 i.f4 i.d6 10 i.g3 and now 10 ... lbbd7
would have been fine for Black in Peturs­
son-Gretarsson, Reykjavik 1995.
5 .. )2Jf6 6 g3 l2Ja6
In these kinds of position, rather similar
to the Queen's Indian and the Dutch, the
problem piece for Black is the knight on
b8, so it makes good sense to develop it
while there is the opportunity. Neverthe­
less, the immediate 6 ... i.d6!? is worth
more than a second glance: 7 i.g2 0-0 8
lbf3 (to prevent ... i.e5) 8 .. JIVe7 9 0-0 aS 10
tt:Jd4 tt:Ja6 1 1 b3 tt:Jc5 12 i.b2 tt:Jfe4 13
tt:Jdb5 i.e5 with good play in Apel-Lau,
Dresden open 1995.
However, 6 ... b5?! is less good due to 7
lbf3 bxc4 8 dxe6 dxe6 9 "ii'a4+ 'ii'd7 10
'ii'xc4 lbc6 1 1 i.g2 i.d6 12 b4 lbe7 13 0-0
lbed5 14 i.b2, as in S.Mohr-Bischoff,
German Bundesliga 1995/96. Black's pawn
structure is a mess.
6 .. . a5 is sometimes played (it is often
useful to control White's queenside pawn
structure) but it has been established that
Black does best to develop as quickly as
possible, and actually has no need for this
preparatory move.

The other main alternative, 6 . . . g6, is the
subject of Game 56.
7 .tg2
It is important to see what might hap­
pen if White attempts to trap the knight
on the edge of the board. There are two
possibilities after 7 b4:
a) 7 ... lt:lxb4!? 8 axb4 i.xb4 9 i.. d2 (or 9
'ti'b3) 9 . . . exd5 and Black has three pawns
for the piece and a solid position. Chances
are balanced.
b) And for the less adventurous, 7. . . c5
and now:
b 1) 8 b5 lt:lc7 9 i.g2 exd5 (9 . . . i.d6) 10
cxd5 i.. d6 is fine for Black as the white cl­
pawn is a liability.
b2) 8 dxc6 dxc6 (8 . . . i.xc6 9 lt:lf3 i.. e7 10
i.. g2 llc8 1 1 0-0 lt:lc7 is unclear) and now:
b2 1) 9 'ii'xd8+ �xd8 10 i.. g2 l:!.d4!?
(10 ... e5) 11 b5 lt:lb8 12 i.f4 lt:lfd7 and
Black wins a pawn, though the game is
just starting.
b22) 9 i.. g2 'fi'xd1+ 10 �xd1 lt:lg4 1 1
lt:lh3 lt:le5 1 2 lt:lg5 0-0-0 + 1 3 'it>c2 l:i.e8 14
Wb3 h6 15 lt:lh3 lt:lc7 16 i..b2 tt::l g4 17 e4 c5
18 b5 i.e7 19 Ii.ae1 i.f6 20 f3 lt:le5 2 1 lLlf2
g5 22 tt::l e2 .l::.hf8 23 i.c3 fxe4 24 tt::lxe4
i.xe4 25 fxe4 tt::l g4 and Black was clearly
better in Lazarev-Lempert, Bled open
1994.
Conclusion: Black has nothing to fear
from 7 b4, so he does not need to waste a
move with 6 . . . a5 first.
Wh i t e Pla ys a n e a rl y a 2 - a 3
7 . . . tt:Jc5 8 t2Jh3
8 tt::l f3 isn't as good: 8 . . . tt::lce4! (the pawn
on d5 is being rounded up) 9 dxe6 (or 9
0-0 tt::lxc3 10 bxc3 i.c5 with a free game in
Apel-Blauert, German Bundesliga 1995)
9 ... dxe6 10 �a4+ �d7 1 1 'iWxd7+ �xd7 12
tt::lxe4 i.xe4 13 0-0 i.d6 14 .l::.d 1 �e7 15
.i.e3 e5 with a slight plus for Black in Ho­
ang Thang Trang-Czebe, Budapest 1998.
8 . .td6 9 0-0 .i.e5!
. .
A clever manoeuvre. Black does best to
set up pressure straightaway on White's
centre.
1 0 'it'c2
There is a sharper move available to
White: 10 lt:lb5!? 0-0 (after 10 ... a6 11 f4
i.xb2 12 lt:lxc7+ i¥xc7 13 i.xb2 0-0 14
tt::lf2 the powerful bishop on b2 ensured
White some advantage in Alexandrov­
Matveeva, St Petersburg open 1994) 1 1
i..f4 (1 1 f4!? i.d6 12 tt::l f2 is double-edged:
the pawn on d5 could become weak and
the e4-square already is, but White gets the
two bishops) 1 l . ..d6 12 i..xe5 dxe5 13 b4
tt::l ce4 14 dxe6 'it'e7 15 'ii'c2 a6 16 tt::l c3
'i'xe6 was fine for Black in Shneider­
Lempert, St Petersburg open 1993, as the
knight on h3 is misplaced.
1 0 . . . 0-0
This has been the starting point for
numerous games.
1 1 .i.e3
White has several alternatives here:
1 01
En glish D e fe n c e
a) For 1 1 i.d2, see the next game.
b) 1 1 �d1 'i*'e7 12 i.e3 tZ:lce4 13 tZ:lxe4
tZ:lxe4 and now:
b 1) 14 Si.xe4!? (White runs a great risk
by exchanging off this bishop, but there
are some strange tactics in the air) 14 ... fxe4
15 �xe4 i.xb2 (not 15 ... exd5? 16 cxd5
.l:i.ae8 17 tZ:lg5 g6 18 d6 and wins) 16 d6!?
i.xe4! 17 dxe7 l:tfe8 18 �a2 i..f6 19 �xd7
l:txe7 and Black is fine.
b2) 14 l:tacl c5 15 dxe6 dxe6 16 i.f4
i.f6 17 .l:td3 e5 18 i.e3 g5 19 l::!.cd1 .l::!. ad8
20 �xd8 �xd8 2 1 �xd8+ 'i*'xd8 22 'i*'d3 h6
23 g4 'i*'xd3 24 exd3 tZ:ld6 25 i.xb7 tZ:lxb7
26 f3 e4! was clearly better for Black in
Sher-Lempert, Yerevan open 1996.
c) 11 tZ:lf4 'it'e7 12 i.e3 tZ:lce4 13 tZ:lxe4
tZ:lxe4 14 l::t ad1 (or 14 i..xe4 fxe4 15 'ikxe4
i.. xb2 16 d6 i.xe4 17 dxe7 l:If7) 14 ... c5
(perhaps even 14 ... a5!? 15 i.cl �ae8 16
..t>h1 �f7 17 dxe6 dxe6 18 t2Jd3 i.d6 19 b4
axb4 20 aXb4 c5 2 1 bxc5 bxc5 22 i.b2 'i'c7
23 f3 tZ:lf6 24 'ir'c3 e5 with counterplay in
Franic-Doric, Croatian Championship,
Porec 1994) 15 i.. c l g5 16 tZ:ld3 i.. d4 17 e3
i.g7 18 f3 tZ:ld6 19 e4 i.d4+ 20 ..t>h1 g4
with unclear chances in Naumkin­
Lempert, Moscow 1994.
1 1 . . . t'Llce4!
As we have seen in the games above,
this is a key move for Black, blocking out
the bishop on g2.
1 2 lt:Jxe4 t'Llxe4
1 02
1 3 �fd 1
Not 13 i.xe4? fxe4 14 'iVxe4 i.xb2 15
.Ua2 (if 15 t2Jg5 .Uf5) 15 ... i.f6 with a clear
advantage for Black. White should only
exchange the bishop for some tangible
gam.
1 3 . . . 'i'f6 1 4 �ab 1 exd5
For tactical reasons 14 ... c5? isn't good
due to 15 dxc6! i.xc6 16 f3 tZ:lc5 17 f4 i.c7
(not 17 ... i.xg2? 18 <;t>xg2! i.c7 19 i.xc5
bxc5 20 l:txd7) 18 .li.xc6 dxc6 19 i.xc5
bxc5 20 �d7 with a plus for White; while
after 14 ... a5?! 15 f3 tZ:lc5 16 f4 i.d6 17 b4
axb4 18 axb4 tZ:le4 19 c5 Black has lost
control.
1 5 cxd5 �ae8 1 6 ifb3
1 6 . . . .td6
Perhaps 16 ... ..t>h8.
1 7 �bc 1 'iif7 1 8 .tf4 c5 1 9 t'Llg5 ! ? 'ikg6
20 t'Llxe4 fxe4
Or 20 ... i.xf4!? 21 tZ:ld2 i.xd2 22 �xd2
d6 23 'i!Va4!?
2 1 .txd6 'i'xd6 22 �c2 �h8? !
2 2... l:i.e5! would have been stronger.
23 'i'e3 'iie 5
23 ... i.xd5? 24 �cd2 .l:te5 25 i.xe4! �xe4
26 'i'xe4 wins for White.
24 .l:lcd2 d6 25 b4 c4?
see following diagram
25 .. J:tf6 would have been better. After
the game move White holds a clear advan­
tage.
 Wh i t e Pla ys a n e a rl y a 2 - a 3

26 !td4 �a6 27 !txe4 'i'h5 28 �f3 !txe4 1 1 . . . exd5

29 'ifxe4 'iff7 30 'iid 4 'ife8 3 1 J:tc1 h6 1 1 ...tt::lce4, as we have seen from some
32 J:lc3 'i!fg6 33 J:lc1 'iig 5 34 l:lc3 b5 35 of the other games, is quite possible.
'i!fxa7 'ifd2 3 6 �e3 'ifb2 37 'i�Vc 1 'i�Va2 38 1 2 lLJxd5
l:lc2 'i�Vb3 39 l:lc3 'iia 2 40 Wg2 �c8 41 12 'ifxfS .ixc3 13 .ixc3 tt::l fe4 is equal;
litc2 'ifb3 42 J:tc3 'iWa4 43 it'e3 ..tt5 44 but not 12 cxdS? .ixc3 13 .ixc3 tt::lxdS and
'Wie7 .l::.f 6 45 e4 ..tg6 46 e5 J:txf3 47 J:txf3 Black is on top.
�e4 48 exd6 ..txf3+ 49 �xf3 'ii'd 1 + 50 1 2 . . . lbxd5 1 3 cxd5
�g2 'ilfxd5+ 5 1 f3 ii'f5 52 d7 'ilfc2+ 53 13 i.xd5+ .ixdS 14 cxdS 'ii'f6 is bal­
'it>h3 'i!ff5+ 54 'it>h4 g 5+ 55 'it>h5 'Yiixf3+ anced.
56 �xh6 'ifc6+ 57 Wxg5 'ifd5+ 58 �96
'ilfd3+ 59 �f7 'i!Vh7+ 60 �e6 'i!fe4+ 1 -0
After 6 1 �d6 �d4+ 62 �c6 White is
wmnmg.
The system with 4 .. .f5 followed by
. . . tt::l a6-c5 is quite good, but it is highly
complex and requires a good understand­
ing of the position. The next game is an­
other example of the same set-up.

Gam e 55
M . G urevich-Kengis
Ba d Go desburg 1996
1 3 . . . 'iff6 1 4 ..tg5
1 c4 b6 2 d4 ..tb 7 3 lbc3 e6 4 a3 f5 5 14 i.c3?! i.xc3 15 bxc3 �ae8 gives
d5 lbf6 6 g3 tba6 7 �g2 tbc5 8 tbh3 Black the initiative; while 14 l:tab1 tt::l e4 15
�d6 9 0-0 �e5 1 0 'Wic2 0-0 i.b4 l:tfe8 16 l:tfd1 cS! 17 dxc6 dxc6 is un­
Everything as in the previous game, but clear.
now Gurevich tried something new and 1 4 . . . 'i'f7 1 5 ..tt4 ..txf4 1 6 lLlxf4 g 5 ! ? 1 7
different. lL'lh3
1 1 ..td2 17 tt::ld3 i.xdS 18 tt::lxcS i.xg2 19 �xg2
see following diagram
'if'dS+ offers Black good counterplay.
1 7 . . . h6 1 8 b4 lL'le4 ! ?

1 03
En glish D e fe n c e

1 8. . .'�Ja6?! 1 9 l:tad1 would have been 26 . . . tt:le4

weird! Both knights are trapped at the 26 ... d5!? 27 f3 'We3+ 28 tbf2 is unclear.
edge of the board, but it is easier to acti­ 27 1:tc2 f4
vate the one on h3 with f2-f4 and tLlf2.
1 9 'Wixc7 i.xd5

White must force a draw.

28 l:lxc8
A complex position has arisen. Black's Not 28 Wg2? �xh3+! 29 Wxh3 t2Jxf2+
pawn structure isn't very pretty but his 30 <it>g2 tbxd3 and Black wins.
minor pieces are active, while White's 28 .. Jbc8 29 f3 ! 'Wixh3
knight on h3 is out of play. 29 ... fxg3 30 �xe4 �xh3 3 1 'Wg6+ Wh8
20 l:tad 1 'Wie6 ! also draws.
Threatening ... kifc8, trapping the 30 'ifxe4 fxg3 3 1 iig6+ �h8 �-�
queen. 32 �f6+ Wh7 33 'i!Vf7+ with a perpetual
2 1 'i'c 1 �ac8 22 'i'e3 tt:lc3 ! 23 i.xd5 check.
tt:lxd5 24 'Wid3 tt:lc3 The system with ... tba6-c5 has survived
a severe test (Mikhail Gurevich is a decent
player!), so we can conclude that it is fully
viable.
Now we move on to Black's double fi­
anchetto.

Gam e 56
Hjartarson-Gretarsson
Rey kja vi k 1995
1 c4 b6 2 d4 e6 3 a3 i.b7 4 tt:lc3 f5 5
d5 tt:lt6 6 g3 g6
Instead of 6 ... tba6, which we saw in the
25 .l:!.d2 previous two games, Black fianchettoes
Not 25 'ii'xd7? �e4! threatening the the king's bishop, leading to a version of
rook and mate on e2. the Leningrad Dutch. This has been quite
25 . . Jlfe8 26 lite 1 a popular way of handling White's open­
26 'ii'xd7 tbxe2+ 27 Wg2 g4 28 'i\Yxe6+ ing so it is worth taking a look. On the
l:txe6 29 tbg1 is level. surface of it, compared to the ... tba6 sys-

1 04

terns where the bishop does a dance in the
middle of the board (... .lii. d6-e5) and can be
a little vulnerable in the process, the fi­
anchetto has much to recommend it. The
bishop reaches the crucial long diagonal
and is quite safe on g7; moreover the f5-
pawn gains some useful support. On the
downside the knight on b8 doesn't get
developed for some time, and that was
Black's undoing in this game.
7 .lit.g2 �g7 8 ti:Jh3 !
The best way to develop the knight.
Other routes are less effective:
a) 8 e3 0-0 9 tbge2 e5 10 0-0 d6 1 1 e4
fxe4 12 tbxe4 tbxe4 13 .lixe4 tbd7 14 tbc3
.i.c8 15 .i.g2 t2:lf6, as in Lautier-Topalov,
Linares 1994, is more comfortable for
White, but Black's position remains
sound.
b) 8 t2:lf3 0-0 9 0-0 t2:la6 10 .lii. f4 (not 10
e4?! tbxe4 1 1 tbxe4 fxe4 12 tbg5 exd5 13
cxd5 tbc5 14 tbxe4 tbxe4 15 .i.xe4 �f6 16
l:tb 1 iid4 with the initiative for Black in
Fioramonti-Khenkin, Geneva open 1995)
10 ... exd5 1 1 cxd5 tbc5 12 l::tc l t2:lfe4 13
.i.e5 i.h6 14 e3 d6 15 .lii.d4 .i.a6 16 b4
.i.xf1 17 .i.xf1 tbxc3 18 ..ltxc3 tbe4, when
White had some compensation for the
exchange in Kharitonov-Morozevich,
PCA open, Moscow 1994 (weak square on
e6 and c-file), but I doubt whether it is
really sufficient.
8 . . . 0-0 9 0-0
Wh i t e Pla ys an e a rl y a 2 - a 3
9 . . . ti:Ja6
9 ... a5 is possible, though Black must be
prepared for 10 c5 (10 l:!.b 1 t2:la6 also
slightly favours White), when he is fight­
ing for equality: 10 . . . 'it>h8 1 1 cxb6 cxb6 12
.lii.f4 tbxd5 13 tbxd5 .i.xd5 14 .lit.xd5 exd5
15 'i'xd5 tbc6 16 �ad1 i.xb2 17 �d3 ..ltf6
18 �fd1 'i'e8 19 �e3 'i'f7 20 'iib 5 and
White had the initiative in P.Cramling­
Gulko, Spanish T earn Championship
1996.
1 0 .l:tb1
The immediate 10 b4 may be even bet­
ter, as it appears that Black cannot exploit
the position of the white rook on the long
diagonal:
a) 10 ... t2:le4 1 1 tbxe4 .lit.xa1 12 .lit.g5 'ii'e 8
13 1/Vxa1 fxe4 14 .lit.h6 l:H7 15 dxe6! dxe6 16
lLlgS :e7 17 .lit.xe4 and White is on top
(U manskaya) .
b) 1 0 ... exd5 1 1 cxd5 tbe4 1 2 tbxe4 .lit.xa1
13 .i.g5 ..We8 14 'ii'xa1 fxe4 15 .lit.h6 .Uf7 16
tbg5 .i.xd5 17 'ii'd4! also favours White
(Umanskaya).
c) 10 ... tbxd5!? 1 1 tbxd5 exd5 12 .i.xd5+
.i.xd5 13 'i'xd5+ �h8 14 .lit.g5! 'i'e8 15
-�ad1 with a clear edge for White m
Farago-Varga, Pecs 1998.
1o . . . ti:Jc5 1 1 b4 tt:Jce4 1 2 .lit.b2 exd 5 ? !
It would have been better to keep the
position more fluid with 12 ... 1i'e7, for
instance.
1 3 cxd5 ti:Jd6 1 4 'i'b3 �h8 1 5 .:tbc1 .l:.e8
1 05
En g lish D e fe n c e
1 5. . .c5 1 6 dxc6 dxc6 1 7 l2Jf4 and it lands
on e6.
1 6 �fd 1
White has a wonderful position; all his
pieces are beautifully centralised. On the
other hand, Black lacks space, the c-pawn
is vulnerable, and opening the position
merely unleashes the full force of White's
army.
1 6 . . . lt:lg4
16 ... l2Jfe4? 17 l2:lxe4 l2:lxe4 18 .lit.xg7+
�xg7 19 .lit.xe4 .l:i.xe4 20 'i!Vc3+ 'it>g8 2 1
'iVxc7 i s winning for White.
1 7 a4 a6 1 8 e3 l:.c8 1 9 lt:lf4 lt:le5 20
lt:lce2 lt:le4 2 1 lt:ld3
21 .. ."�e 7 ?
2 1 . ..l2Jxd3 would have been a better try,
but White still has a great position: 22
..ixg7+ �xg7 23 'iVxd3 �e7 24 .l:i.c4 fol­
lowed by doubling on the c-file.
1 06
22 lt:lxe5 i.xe5 23 i.xe4 fxe4 24 �xc7 !
�xc7 25 d6 'i'g7 26 i.xe5 'i'xe5 27 dxc7
'l//ixc7 28 'l/lit7 �d8 29 lt:lf4 1 -0
29 .. .'i/Vc8 30 l2:le6! dxe6 3 1 "iVf6+ is hope­
less.
6 ... g6 is just about playable for Black (it
is not necessary to get squashed like this)
though still highly complex. Now we
shall take a look at what happens if White
delays playing d4-d5.
Gam e 57
Kislova-Minasian
Oms k 1996
1 c4 b6 2 d4 i.b7 3 lt:lc3 e6 4 a3 f5 5
lt:lf3 lt:lf6 6 g3
6 . . . lt:le4
Black does not have to leap in so
quickly with the knight. Both 6 ... g6 7
..ltg2 ..ltg7 8 0-0 0-0, as in Mednis-Plaskett,
Luxembourg open 1990, and 6 ... ..lte7 7
..tg2 0-0 8 0-0 l2Je4, as in Rind-De Jager,
Dieren open 1991, are perfectly playable.
However, 6 ...l2Je4 presents problems for
White to solve at a very early stage in the
game - and few players have been up to
the challenge!
7 lt:lxe4
This is the critical response. Others
aren't nearly as testing, generally leading
to positions with a strong resemblance to
the Queen's Indian, but where White has
 Wh i t e Pla ys an e a rl y a 2 - a 3

wasted a move on a2-a3. For example: game continuation. Black does not need to
a) 7 'iid3 iLe7 8 d5 lLlxc3 9 'ii'xc3 0-0 10 play ... d7-d6 however; 8 ... iLe7 is better.
'ii'd3 .tf6 1 1 iLg2 l'Lla6 12 lLld2 tt:Jc5 with 8 . . . i.. e 7 9 tl:lg2 0-0 1 0 tl:le3 �e8 1 1 i.. g 2
good play for Black in Bukic-Tratar, 1 1 1Lh3 (as recommended by the gen­
Slovenian Championship 1995. tlemen above) just doesn't have the same
b) 7 �c2 iLe7 8 iLg2 lLlxc3 9 'ii'xc3 i..f6 effect when Black hasn't moved the cl­
10 0-0 0-0 1 1 iLe3 iLe4 12 �acl tLlc6 13 b4 pawn: 1 1 . . .'ii'h 5 12 .i.g4 'ii'f7 13 0-0 tLlc6 14
'ii'c 8 14 �fd1 'iWb7 15 'ii'b 3 lLle7 16 iLfl a6! f3 (14 b4? l'Llxd4 15 'i'xd4 iLf6; 14 d5 tLleS
17 tLld2 .li.c6 18 f3 b5! with excellent con­ 15 1Lh3 iLc5) 14 ...exf3 15 1Lxf3 .i.f6 and
trol over the central light squares in Ja­ Black has the better development.
kobsen-Rewitz, Esbjerg 1996. 1 1 . . . d5 1 2 0-0 tl:ld7
c) 7 .td2 1Le7 8 .i.g2 .i.f6 9 i::tc l (or 9
0-0 l'Llc6 10 d5 tLlaS with counterplay)
9 ... lLlc6!? 10 iLe3 (10 d5 l'Lla5 and 10 0-0
lLlxd4 1 1 tLlxd4 .i.xd4 12 tt:Jxe4 fxe4 both
slightly favour Black) 10 ... 0-0 1 1 0-0 tLle7
12 lLlxe4 i.. xe4 13 'i!Vd2 l:Ib8! with the idea
of ... b6-b5, when Black had sufficient play
in M.Gurevich-Kengis, Bonn 1995.
7 . . . fxe4
If 7. . . iLxe4 White has an excellent re­
sponse: 8 .i.h3! 1Le7 9 0-0 0-0 10 d5! .i.xf3
(not 10 ... exd5? 1 1 tLld2! and wins) 1 1 exf3
.tf6 12 J::tb 1 tLla6 13 l::te 1 l'Llc5 14 b4 tLlb7
15 dxe6 dxe6 16 ifb3 with a big advantage 1 3 i.. h 3
due to the white bishops and pawn struc­ Black has to be a little careful of the
ture and Black's poor knight in Adorjan­ pins after 13 cxd5 exd5 14 'ii'b 3, but
Miles, Gjovik 1983. 14 ...'ii'f7! holds.
1 3 . . . l:.f6 1 4 cxd5 exd5 1 5 'ii'b 3 c6 1 6
i..g 2

S liJh4
Also possible is 8 l'Lle5 d6 9 l'Llg4
' . . . intending 1Lh3 and l'Lle3 with a clear
advantage. ' - Keene, Plaskett and Tisdall.
In fact this idea is rather similar to the
White has lost the plot somewhat with
her to-ing and fro-ing. From now on she
gets steadily outplayed. Around here,

107
En g lish D e fe n c e
chances are still equal.
1 6 . . .c,1;th8 1 7 f4 exf3 1 8 exf3 1:l.f7 1 9 1:l.d 1
ttlf6 20 ttlf 1 .td6 2 1 .tg5 h6 22 .te3
l:!e7 23 .tf2 'i¥g6 24 l:!.d2 c5 25 .l:!.c1 c4
26 'iic 2 'i'xc2 27 .l:!.cxc2 J:tae8 28 .th3
g 6 29 �g2 b5 30 �g 1 a5 31 .l:l.c1 .tc6
32 l:!.dd 1 b4 33 axb4 .txb4 34 I:!.a 1 .l::!.e 2
3 5 .l::!. d b 1 a4 3 6 g4 a3 37 bxa3 i.. c 3 38
a4 .txa 1 39 l:!.xa 1 c3 40 a5 c2 41 a6
.l:!.a8 42 a7 .l:!.xa7 43 .l::!.x a7 c 1 'if 44 .l:l.c7
0-1
Very few players have taken up the
challenge and captured the knight on e4.
However, I feel that Black's position is
strong enough; White is, after all, wasting
a lot of time by moving his knight around
the board, so he falls behind in his devel­
opment. More practical tests are needed
(there was a large discrepancy in strength
between the players here) before a firm
conclusion can be reached. One thing is
certain: if White does not capture on e4,
and most players tend not to, then Black
should equalise.
Gam e 58
Suba-Piaskett
Lon don 1991
1 c4 b6 2 d4 .tb 7 3 ttlc3 e6 4 a3 f5 5 f3
Here I'm going to be looking at games
where White plays with f2-f3. A danger­
ous move. Defensively, it blocks out the
bishop on b7 and prevents Black's knight
from moving to e4; and offensively White
can sometimes play for e2-e4, as he does in
this main game.
A similar idea is 5 e3 lbf6 6 f3 and now:
. .
a) 6 . . . c51�
.. 1s mterestmg.
. with his pawns and takes it. If he were
b) 6 . . . g6 is playable.
c) 6 ... .i.d6 7 lbh3 0-0 8 tbf2 tbc6 9 i.e2
lbhS 10 f4 tbf6 1 1 i.f3 and White was
already better in Spraggett-Yewdokimov,
Oviedo rapidplay 1992.
d) 6 ... d6 7 i.. d3 lbbd7 8 lbge2 lbhS!? 9
dS (9 0-0 'tih4) 9 . . . g6 10 dxe6 lbcS 1 1 0-0
1 08
tbxe6 12 e4 f4 13 lbds gS 14 g4 lbf6 15 b4
i.g7 16 i.b2 c6 17 lbdc3 lbd7 18 .i.c2 lbeS
with good play for Black in Galianina­
Strutinskaya, Russian Women's Champi­
onship, Elista 1997.
e) 6 ... .i.e7 7 i.d3 cS 8 ti:lge2 lbc6 9 0-0
0-0 10 b3 .i.d6 11 .i.c2 'i'b8 12 h3 lbe7 13
�h1 lbg6 14 .l:i.b 1 cxd4 15 exd4 'iid8 16
i.e3 lic8 17 ii'd2 .i.b8 18 cS 'i'e8 19 b4
ti:lh4 20 .l:i.f2 h6 21 i.f4 .i.xf4 Y2-Y2 Piket­
Adams, Linares 1997.
5 . . . ttlf6 6 ttlh3
6 e3 transposes to the previous note.
6 . . . g6
I once tried 6 ... .i.e7, but with hindsight
I don't like this move: it is too passive.
After 7 e3 0-0 8 i.e2 aS 9 0-0 lba6 10 b3
�e8 11 lbf4 1i'f7 (if 1 1 ...g5 12 lbd3 d6 13
b4 with a great position for White) 12
i..b2 dS 13 cxdS lbxdS 14 lbcxdS exdS 15
e4 c6 16 exdS cxdS 17 lbd3 White was
better in Murshed-King, Dhaka 1993.
7 ttlf2 .tg7 8 e4
Suba is an uncompromising player: he
is offered the chance to grab the centre
safely castled then White would have a
large advantage, but that is still some way
off so Black has a chance to strike back.
8 . . . fxe4 9 fxe4 0-0 1 0 .tg5 h6 1 1 .te3
ttlc6 ! 1 2 e5
White must do something about his
pawns, and once the mass advances, Black

can attack it more easily. However, 12
i..e2 is met by 12 ... e5! 13 d5 t:Dd4 14 0-0
c5.
1 2 . . . lt:\h5 1 3 'i'g4 lt:\e7 14 �d3 'ife8 1 5
0-0 l2Jf5 1 6 �d2 d 6 1 7 exd6 cxd6 1 8
.!itae 1 'iff7
The position is totally unclear: Biack's
pawns aren't in the best shape but, on the
other hand, his minor pieces are extremely
active.
1 9 d5 exd5 20 lbxd5 �xd5 2 1 cxd5
�xb2 22 J:le6 lt:\f6 23 �c4 l:!.ac8 24 'i¥b3
�e5 25 �xf5 gxf5 26 �xh6 J:1c3 27 �b 1
lt:\xd5 28 :xe5 dxe5 29 �xf8 'i'xf8 30
lt:\e4 l:!.xa3 3 1 'i¥b2 lt:\e3 32 l:!.f3 lt:\c4 33
'ilic 1 l:!.xf3 34 "ifxc4+ Wh8 35 gxf3 fxe4
36 'ii'xe4 a5 37 Wf1 'iid 6 38 h4 Wg7 39
'ilig4+ �f7 40 "iff5+ 'ii'f 6 41 'ifd7+ �g6
42 'i'g4+ Wh6 43 We2 b5 44 'ild7 b4 45
'i'b5 'iff4 46 'it'e8 iic4+ 47 �e3 'ifd4+
48 We2 Y. - Y.
Mad stuff. I feel that if White is going
to play with f2-f3 he does better by hang­
ing back with e3-e4 and seeing what for­
mation Black adopts (as in the note to
White's fifth move, for instance) .
Gam e 59
Morovic-Speelman
Cala Gal dana 1994
1 d4 e6 2 c4 b6 3 a3 �b7 4 lt:\c3
I mentioned in the introduction to this
Wh i t e Pla ys an e a rl y a 2 - a 3
chapter that if nothing else after White
plays a2-a3, Black has the option of trans­
posing back into a Queen's Indian De­
fence, Petrosian variation, with ... t:Df6. But
is that really so? In fact White can avoid
that if he wishes to as the next couple of
games demonstrate. White can also reach
these positions by playing d4-d5 before a2-
a3, so these positions have a direct signifi­
cance for English Defence players.
4 . . lt:\f6
.
Black could also play 4 ... d5, when 5
cxd5 exd5 6 t:Df3 ll:\f6 is one of the main
variations of the Petrosian variation
(which is outside the scope of this book).
It is also possible to be crafty and delay
the development of the knight on g8 with
6 i..d 6.
•. .
5 d5
A bold attempt to avoid the Queen's
Indian. Instead, 5 t:Df3 would transpose
directly.
5 . exd5
. .
For 5 ... i..d6 see the next game.
6 cxd5 b5! ?
Very much in keeping with the wacki­
ness of a great deal of the English Defence.
Black has also tried:
a) 6 ... i.. d6 7 g3 0-0 8 i..g2 .l::i.e8 9 t:Dh3 c5
10 0-0 i..f8 1 1 l::re 1 d6 12 e4 t:Dbd7, as in
Lukacs-Ostojic, Budapest 1977, is a sensi­
ble set-up for Black. He threatens ... c5-c4
and ... t:Dc5.
1 09
En g lish D e fe n c e
b) 6 ... c6 is the most direct attempt to
meet White's early (premature?) pawn
push: 7 g3 (not 7 e4? 'iie7) 7 .. .'�Jxd5 8
lbxd5 cxd5 9 .i.g2 'iif6!? 10 lbh3 lba6 1 1
0-0 i.. d6 1 2 i..xd5 i..xd5 1 3 'ii'xd5 0-0 14
.l:!.b 1 'il¥e6 15 l:i.d1 'iVxh3 16 "ii'xd6 l\Vh5 17
'iid3 lbc5 18 'iif3 'iixf3 19 exf3 �fd8 20
l:td5 l:tac8 21 i.. e3 lba4 22 i.. d4 f6 23 �d1
l:tc2 'h-'h Lukacs-Regan, Budapest 1978.
On the basis of this game Black has little
to fear.
c) 6 . . . g6 7 g3 i.. g7 8 i.. g2 0-0, as in
Bischoff-Lau, Bad Endbach 1995, is also
quite acceptable.
7 'ilid4
The most straightforward move is 7 e4,
as Speelman has faced in another of his
games: 7 . . . b4 8 axb4 .i.xb4 9 i..d3. Now I
think Black should try 9 ... 'iie7 10 "iie2 c6
(10 ... 0-0 and . . . l:te8 is also possible). The
actual game went 9 ... c6 10 dxc6 lbxc6 1 1
lb f3 0-0 1 2 0-0 .i.xc3 1 3 bxc3 d5 1 4 e5 lbe4
15 'iic2 l:Ie8 16 .i.xe4 dxe4 17 1i'xe4 ltJaS
18 'tlt'b4 .i.xf3 19 'iixaS i.. e2 20 l:te1 .i.c4 21
i.. e3 and White was somewhat better,
though the game Baburin-Speelman, Co­
penhagen 1996 ended in a draw.
7 . . . a6 8 il..f4 il.. e 7 9 e4 d6 1 0 a4 c5! 1 1
'ilfd3 b4
see following diagram
Black's play on the queenside already
gives him the advantage.
110
1 2 l2lce2 0-0 1 3 l2lg3 �e8 1 4 il..e 2 il..fS
1 5 l2lf3 aS 1 6 t2ld2 l2lbd7 1 7 l2lc4 l2lb6
1 8 l2lxb6 �xb6 1 9 f3 Ji.a6 20 'iWb3 Ji.xe2
21 l2lxe2 'iWa6 22 ikc2 c4 23 0-0 c3 24
bxc3 l2lxd5 25 cxb4 l2lxb4 26 'i'd2 d5 27
l2lc3 dxe4 28 l2lxe4 �adS 29 'ii'f2 t2ld3
30 'i'g3 f5 31 l2lg5 �d4 32 il..d 2 h6 33
l2lh3 il..d 6 34 f4 il..fS 35 il.. c 3 :de4 36
'tif3 "ifc4 37 il..d 2 l:!.e2 38 Ji.xa5 "Yii'c S+ 39
�h 1 lil2e3 0- 1
6 ... b5 is fun, but perhaps one needs to
be a special kind of player to get away
with it.
When faced with d4-d5 preventing the
Queen's Indian, most Black players have
tried the system featured in the next game.
Gam e 60
Andreasen-Pedersen
Danis h Cham pions hi p 1988
1 d4 l2lf6 2 c4 b6 3 l2lc3 il.. b 7 4 d5 e6 5
a3 Ji.d6
By far the most popular way of dealing
with White's system (but that doesn't
mean it is necessarily the best) .
6 l2lf3
White does best to prevent ... i..e 5. For
example, 6 e4 i..e5 7 i.. d3 i..xc3+ 8 bxc3
li'e7 9 lbe2 lba6 10 0-0 lbc5 1 1 f3 i..a6 12
i..c2 .i.xc4 and Black was clearly better in
Plachetka-Prie, Paris Championship 1989,
though it is still complicated and Black

finally succeeded in losing this position.
6 . . . 0-0 7 e4 exd5 8 exd5 c6
9 ..ae2
0� 9 i.g5 h6 10 i.h4 cxdS 11 tLlxd5
'ii'e8+ 12 tLle3 tLle4 13 i.. e2 f5 with chances
for both sides in Kindermann-Forintos,
Reykjavik open 1982.
9 . . . cxd5 1 0 cxd5 ttla6 1 1 0-0 J:!.e8 1 2
..ag5
1 2 . . . ttlc5
Black probably ought to flick in 12 . . . h6
as the option of . . . g7-g5 is useful. After 13
i.h4 we have two games:
Wh i t e Pla ys an e a rl y a 2 - a 3
a) 1 3. . .tLlc5 1 4 tLld4 i.e5 1 5 i..c4 ,.b8 16
f4 (16 i.. g3 i..xg3 17 hxg3 tLlce4 is fine for
Black) 16 ...i.. xd4+ (16 ... i..xf4 17 i..xf6
i.xh2+ 18 '.t>h1 gxf6 19 'i'g4+ is too risky)
17 'ii'xd4 tLlfe4 18 b4 tLlxc3 19 bxc5, and
now instead of 19 ... tLle4, as in Knaak­
Planinc, Polanica Zdroj 1979, I think
Black should play 19 ... tLle2+ 20 i..xe2
�xe2 with equal chances.
b) 13 ... �c8 14 tLld4 tLlc5 15 tLlf5 i.f8 16
b4 tLlce4 17 tLlxe4 .l:i.xe4 18 i.. d3 .l:i.e8 19
tLle3 g5 20 i.. g3 tLlxd5 2 1 tLlxd5 i..xd5 was
also fine for Black in Petrosian-Planinc,
Moscow 1975.
1 3 tL'ld4 ..ae5 14 J:!.a2
Not very clever. 14 i..c4 h6 15 i..h4
would transpose to Knaak-Planinc above.
1 4 . . . 'ii'b 8 1 5 f4 ..axd4+ 1 6 'ii'xd4 ttlxd5
1 7 tt:Jxd5 llxe2 1 8 J:!.f3 'ii'e 8 1 9 ..ah6
l::te 1 + 20 <t>t2 'iWe2+ 21 �g3 ttle4+ 22
<t>h3 f6 23 tt:Je7+ �h8 24 'ifxd7 ttlf2+ 25
J:!.xf2 fixe 7 0-1
This is a reliable system for Black; so if
you are happy playing a Queen's Indian
then after a2-a3 you can play ... tLlf6 with
confidence.
111
En g lish D e fe n c e

S u mmary
If you want to be one of the pioneers then I would recommend the double fianchetto in
Games 50-52, but be prepared for reverses - even Miles hasn't scored that well with it. A
reasonable alternative way of tackling a2-a3 is to play some kind of Dutch Defence.
Short played excellently in Game 53 and the systems in Games 54-56 are all reliable. Per­
sonally, I have always found it rather irritating playing against f2-f3 (Game 58), probably
because Black doesn't have the fun of knocking down a pawn centre, though this is still
no refutation of the opening. And lastly, if you are content to play a Queen's Indian,
then there is nothing to stop you transposing with .. .'�Jf6, as Games 59 and 60 prove.

1 d4 e6 2 c4 b6 3 a3 (DJ

3 . . .g6
3 .. .f5 4 l2Jc3
4 ...l2Jf6 - Game 53
4 . . . i..b 7 (D)
5 d5 l2Jf6 6 g3
6 ... l2Ja6 7 i.. g2 l2Jc5 8 l2Jh3 i.. d6 9 0-0 i.. e5 10 '*1Vc2 0-0
1 1 i..e3 - Game 54
1 1 i..d2 - Game 55
6 ... g6 Game 56
-

5 l2Jf3 - Game 57
5 f3 - Game 58
3 ... i.. b 7 4 l2Jc3
4 .. .f5 - see Games 54-58 above (by transposition)
4 . . . l2Jf6 5 d5
5 ... exd5 - Game 59
5 ... i.. d6 - Game 60
4 lt:lc3 ..li.g7 5 lt:lf3 lt:le7 6 e4 ..li.b7 (DJ 7 h4
7 .ie2 - Game 50
7 i.. e3 - Game 51
7 . . .h 6 - Game 52

3 a3 4 . . . i.b 7 6 . . . i.b 7

1 12
 CHA PTER SEVEN I
The Two-Pawn Attack
( 1 c4 b6 2 tt:Jc3 ..tb7 3 e4)

1 c4 b6 2 l2Jc3 .ib7 3 e4 ing spirit for the rest of them put to­
There are certain White players who gether!)
open with 1 c4 and then play their next
four pre-programmed moves without even Gam e 61
looking up from the board. It is the Korchnoi-Short
equivalent of the second serve in tennis, Groningen 1997
designed not to score an ace, but just to
get the ball safely over the net. 1 c4 b6 2 e4 .ib7 3 l2Jc3 e5! ?
In these final two chapters I'm going to
be examining games where White declines
to occupy the centre with all three pawns
in an attempt to refute Black's system, but
instead attempts to play a more modest,
and to them, more familiar English or
Reti opening. Here we shall look at games
where White advances two pawns, the c­
and the e- pawns, but leaves the d-pawn at
home.
Most of the games in this chapter con­
tain some interesting ideas - generated by
both colours. Although Games 62-66 are
interesting enough, I don't think Black In a way, Black's fianchetto isn't par­
should experience any difficulties. The ticularly relevant to the position (though
critical move is 4 tt:lf3 followed by .i.d3 it can of course be useful in covering the
(Games 67-70) . dS- and e4-squares) ; what Black is aiming
But first, something a little out of the for is control of the central dark squares.
ordinary from Black. Short brings some­ This was the first time this move had been
thing new to the opening. (By the way I seen at such a rarefied level, though there
wouldn't put Korchnoi in the category of are a couple of previous examples from
the 'second servers'; he has enough fight- less well-known players.

1 13
En glish D e fe n c e
4 d3
Or tLlf3 tLlc6! (4 ... d6? is a horrible
move, closing off the bishop's path to c5;
after 5 d4 exd4 6 tLlxd4 tLlf6 7 i.. d3 White
was already better in Endres-Oberem,
Waldshut 199 1) 5 d4 exd4 6 tLlxd4 i.. b4 7
tLlxc6 i..xc6 8 i.. d3 ..ixc3+ 9 bxc3 and
Black was already comfortably placed in
Deubelbeiss-Horn, Swiss Team Champi­
onship 1993.
4 . . . lt:Jc6 5 g3 ..tc5 6 ..tg2 lt:Jge7 !
The knight is far better here than on f6:
when its comrade leaps into d4 it can
move over to c6 to lend support; moreo­
ver it might be useful that the f-pawn is
free to move.
7 lt:Jf3 0-0 8 0-0 aS 9 J&.e3 d6 1 0 d4
exd4 1 1 lt:Jxd4 lt:Jxd4 1 2 J&.xd4 lt:Jc6! 1 3
i.xc5 bxc5
. White has absolutely nothing here. If
1 14
he leaves the structure as it is, he must
cover the d4-square with his knight (to
prevent Black from occupying it with his
own) but then he will be far too passive.
Instead, Korchnoi explodes the position:
1 4 e5 dxe5 1 5 lt:Ja4 'ild6 1 6 'ild5 lt:Jd8
1 7 'i'xc5 J&.xg2 1 8 'i'xd6 cxd6 1 9 'ot>xg2
lt:Je6 20 �ad 1 lt:Jd4 21 b3 f6 22 J:td2
.l:!.ab8 23 lt:Jc3 J:!.fc8 24 h4 h5 25 .l:.fd 1
J:.c6 26 lt:Jb5 Y:z - Y:z
Wisely liquidating into a drawn end­
game, as the knight on d4 was too strong.
After 26 ... tLlxb5 27 cxb5 �xb5 28 �xd6
l:hd6 29 �xd6 the game is equal. Perhaps
it is just a question of taste, but to me
3 ... e5 looks a sensible move. Certainly,
Korchnoi failed to make any headway.
Gam e 62
Razuvaev-Barle
Maribor 1996
1 c4 b6 2 lt:Jc3 e6 3 e4 .i.b7
This is the 'normal' starting position
for the two-pawn attack. Note that with
certain move orders, like here, it is not
always possible to go for Short's 3 ... e5 if
you have already played ... e7-e6; at least,
you'd be fairly silly to do so.
4 d3
White strengthens his centre, and waits
to see what Black is going to do before
committing his pieces, but this is rather
 Th e T w o - Pa wn A t t a c k ( 1 c 4 b 6 2 liJ c 3 i.. b 7 3 e 4 )

tame. Alternatively, 4 lt:'lge2 (Game 63), 4

g3 (Games 64 and 65) and the main move
4 lZ'lf3 (Games 66-70) are considered later
in this chapter.
4 . . . i.. b4
Black continues in the spirit of the Eng­
lish Defence and pins. Instead, 4 ... c5, lead­
ing to a kind of 'Hedgehog' position, is
quite acceptable. 4 ... g6 is more original: 5
g3 .ltg7 6 l2lge2 lt:'le7 7 .ig2 d6 (7 ... c5!?) 8
0-0 0-0 9 d4 c5 10 .ie3 a6 1 1 f4 and White
was on top in Borek-Kirchmayr, Austrian
Women's Championship 1996.
5 'ifg4 6 i..d 2 l'£Jc6 7 0-0-0 'i'g6 8 'iih 4 tbt6 9
White is inspired by the opening to tLlh3 'i'g4 1 0 'it'xg4 l'£Jxg4 1 1 f3 liJf6 1 2
play something original, but considering tLlb5 i..xd2+ 1 3 �xd2 0-0-0 1 4 d4 a6 1 5
the options, this was perhaps his best l'£Jc3
course of action:
a) 5 lt:'lge2 d5 6 cxd5 (after 6 a3?! .ixc3+
7 l2lxc3 dxe4 8 lt:'lxe4 lt:'le7 9 .ie2 lt:'lbc6
Black's control of d4 already gave him the
advantage in Eising-Hjorth, Amsterdam
OHRA 1984) 6 . . . exd5 7 'ii'a4+ l2lc6 8 a3
i..xc3+ 9 l2lxc3 l2lge7 10 .ig5 f6 1 1 .if4 0-0
12 i.. e2, as in McNab-Rowson, London
Agency 1998, and now 12 .. .f5 would have
given Black a promising position, e.g. 13
exd5?! (13 i.. g5 is slightly better for Black
according to McDonald) 13 ... l2lxd5 14
'ii'c4 lt:'la5 15 'ii' a2 �h8 with the initiative.
b) 5 t2lf3 lt:'le7 6 .ie2 0-0 7 0-0 f5 8 e5 It is curious how a position has been
lt:'lg6 9 d4 l2lc6 10 'ili'c2 t2lh4, and after reached that is similar to the games where
dreadfully stereotyped play from White, Black decides to check on b4 in Chapter 2
Black already held the initiative in Zlot­ (Games 20-22); but in this case Black is
nikov-Miles, Lloyds Bank open, London better developed so his central strike has
1982. more effect:
c) 5 l2lh3 t'i:Je7 6 'ili'g4 0-0 7 i.d2 f5 was 1 5 . . . e5 1 6 dxe5
better for Black in Duckers-Forintos, Or 16 d5 lt:'ld4.
Bozen open 1992. 1 6 . . . l'£Jxe5 1 7 f4 l'£Jg6 1 8 e5 l'£Jh5 1 9 .:l.f2
5 . . .'i'f6 l'£Je7 20 f5 l'£Jc6 21 i..e 2 l'£Jxe5 22 i..x h5
liJd3+ 23 r,t>d2 l'£Jxf2 24 lLlxf2 g6 25 i.f3
see following diagram
i.xf3 26 gxf3 �-�
Anyone like to try the idea of 5 .. .'�f8!? Normally, with such a material balance
followed by ... lt:'lf6 and ... d7-d5 ... ? Black I would prefer the knights, but as Black's
may be able to make something of his pawns are so compact, the beasts have no
considerable lead in development. good outposts. Here, instead of capturing

115
En glish D e fe n c e
on f5, I think Black should play 26. . J:tde8
and . . J:te5, with equal chances.
Gam e 63
Korchnoi-King
O vi edo ra pi dplay 1992
1 c4 e6 2 tt:lc3 b6 3 e4 i.. b 7 4 tt:lge2
I don't like this move. The knight goes
to an inferior square without any provo­
cation or justification. Black should have
no difficulty in equalising.
4 . . . tt:lf6!
Exposing the shortcomings of 4 lt:Jge2.
Alternatively, 4 . . . .i.b4 5 a3 .i.xc3 6 lt:Jxc3
lt:Je7 followed by castling and .. .f7-f5, was
suggested by Keene, Plaskett and Tisdall,
though no one has tried it to my knowl­
edge, probably because 4 ... l2Jf6 is such a
reliable choice. I wouldn't want to put
you off the idea though; having played
around with the pieces for a bit here I
don't see an obvious way to control
Black's counterattack, e.g. 7 d4 (or 7 .ie2
d5!) 7 . . . 0-0 8 .i.d3 f5 with the usual coun­
terplay. 4 . . . c5 is, naturally, also quite ac­
ceptable, most probably leading to a
'hedgehog'.
5 tt:lg3
The most straightforward, and the most
sensible, move. If White gets too flashy he
can end up in trouble. For example:
a) 5 f3 e5!? Oike Short in Game 61;
116
5 ... d5!? appeals to me more though - it
takes White a long time to develop with
the knight on e2, and in the meantime the
centre is exploding) 6 d4 exd4 7 l2:lxd4
.ib4 (perhaps 7 ... .i.c5 or 7 ... l2Jc6) 8 .ig5
.i.xc3+ 9 bxc3 h6 10 .th4 d6 1 1 l2Jf5 0-0 12
"i¥d4 l2Jbd7 was unclear in Jakobsen­
Keene, Aarhus 1976.
b) 5 e5, lurching forward, but it is only
going to get beaten up: 5 ... l2Jg4 6 d4 'iVh4
7 l2:lg3 .i.b4 (or 7 ... d6!?) 8 h3 l2Jh6 9 a3
.txc3+ 10 bxc3 lt:Jf5 1 1 'it'h5 'ti'xh5 12
l2Jxh5 0-0 13 a4 d6 14 exd6 cxd6 15 a5 l!c8
16 f3 bxaS 17 Iha5 l2Jd7 18 'it>f2 l2Jb6 and a
draw was agreed here in Plachetka-Kengis,
Viking open 1997, but Black is better.
5 . . . c5
Since the knight no longer protects the
d4-square it seemed like a good idea to fix
it. Instead, 5 ... h5!? is good fun, e.g. 6 h4 (6
e5 h4!) 6 ... .i.d6 7 d3 .i.xg3 8 fxg3, as in
Byway-Smith, British Championship,
Southport 1983, and now 8 ... d5! ? 9 e5
l2:lg4 is the most dynamic continuation.
6 i..e 2 l2Jc6 7 0-0 g6 8 f4 d6 9 d3 i.. g 7
10 i..e 3 0-0 1 1 i..f3 'i'd7
With hindsight 1 1 ...l2:ld7 strikes me as
far more sensible, increasing control over
the d4-square. My excuse is that this was a
rapidplay game.
1 2 l2Jge2 a6 1 3 i..f 2 llac8 1 4 J:!.c1 b5
see following diagram
 Th e T w o - Pa wn A t ta c k ( 1 c 4 b 6 2 ltJ c 3 iL b 7 3 e 4 )
.·... nevertheless, Black is still doing fine.
Up till now we had both been playing
quite steadily, but now Korchnoi cracked
off the pawn instantly, with a bang of the
clock to boot.
1 5 cxb5 axb5 1 6 ltJxb5 ltJb4 1 7 a4
If 17 ll:lec3 ll:lxa2 or 17 ll:lbc3 i.a6 with
good compensation.
1 7 . . . iLa6 1 8 e5 dxe5 1 9 fxe5 ltJfd5 20
l:!.xc5 iLxe5 21 l:!.xc8 �xc8 22 d4 iLg7 23
ltJec3 ltJxc3 24 bxc3 ltJdS 25 'il'b3 ltJxc3
26 ltJxc3 i.xf1 27 'it>xf 1 i.xd4 28 iLxd4
'it'xd4
. The position has finally settled. The a­
pawn is potentially dangerous, but it is
difficult to guide it down the board with­
out the minor pieces; if the knight or
bishop assists, then the king will become
exposed. Black also has the long-term
threat of advancing the f- and e-pawns.
29 ltJe2 'it'e5 30 h3 .!:tc5 3 1 'it>f2 'it>g7 3 2
g3 fits 33 'it> g 2 e5 3 4 'ilib7
... here we stopped recording the moves
as time was getting short. Korchnoi played
recklessly and I mated him on the king­
side. 4 tLlge2 is not a move to trouble us.
Gam e 64
Smejkai-Kengis
Prague 1993
1 c4 b6 2 ltJc3 e6 3 g3 iLb7 4 e4
A slightly unusual move order. Instead
of 4 e4 White can play 4 li::lf3 which will
often transpose to a 'hedgehog' system,
after, for instance, 4 ... c5, but Black has the
interesting option 4 ... i.xf3!? 5 exf3 cS,
attempting to gain control over the d4-
square, which is considered in the next
chapter (Games 74-77) .
117
En g lish D e fe n c e
4 . . .f 5 5 ..tg2
The system which White employs is
quite common among a group of hardy 1
c4 players who have grown up playing the
Botvinnik System (1 c4, 2 tZ:lc3, 3 g3, 4
.i.g2, 5 e4, 6 tZ:lge2 and 7 d3) against almost
anything. However, against the English
Defence they have to do a little thinking
for themselves.
5 . . . liJf6
5 . . . fxe4?! 6 tZ:lxe4 makes life a little too
easy for White. Black should hold the
tension for a bit longer.
6 d3
6 1ic2 isn't too clever as White loses
control over the d4-square. After 6 ... fxe4 7
tZ:lxe4 tZ:lc6 8 tZ:lxf6+ "ifxf6 Black was al­
ready better in Hon-King, German Bun­
desliga 1995.
6 . . . fxe4
Black could have played 6 ... i..b4 here
(and on the previous two turns) to in­
crease the pressure on e4. That is the sub­
ject of the next game (by transposition) .
7 lLlge2!
White does best to delay the recapture
of the pawn. It keeps more options open,
though 7 dxe4 has been played, e.g.
7 ... .i.b4 (instead 7 ... il.c5!? is potentially
rather dangerous for White) and now:
a) 8 i..d2 0-0 9 tZ:lh3 tZ:la6 (9 ... 1\Ve7!? 10
l\Ve2 tZ:lc6) 10 0-0 tZ:lc5 11 e5 .ixg2 12 �xg2
tZ:le8 13 tZ:la4 tZ:la6 14 tZ:lg5 il.e7 15 tZ:le4 d5
1 18
16 exd6 tZ:lxd6 17 1ig4 and White was bet­
ter in Libeau-Lau, German Bundesliga
1995.
b) 8 1\Vc2 .i.xc3+ 9 "it'xc3 i..xe4 (or
9 ... 0-0) 10 il.xe4 tZ:lxe4 1 1 "ii'xg7 "ii'f6 12
iVxf6 tZ:lxf6. Here I prefer Black's position,
but you should be aware of my prejudices:
I always like centre pawns.
7 . . . ..tc5
7... i..b4 8 0-0 would transpose to the
next game; but there is really no need to
put the bishop on b4. I like the look of
7... tZ:lc6!? - as Kengis played in a later
game: 8 dxe4 tZ:le5 (8 ... i..c5! ? is also possi­
ble, e.g. 9 0-0 0-0 10 tZ:la4 e5 1 1 tZ:lxc5 bxc5
12 tZ:lc3 tZ:ld4 13 f4 with complications in
Cekro-Hausrath, Belgian Team Champi­
onship 1997) 9 b3 il.c5 10 h3 ike7 1 1 f4
tZ:lf7 12 a3 a5 13 'ii'd3 0-0 14 g4 (rather ex­
travagant) 14 ... d5 15 e5 tZ:le4 16 �xe4 dxe4
17 'iig3 .l::!. ad8 18 i..e3 i..xe3 19 1ixe3 tZ:lxe5
0-1 Heinbuch-Kengis, Bonn GSK 1995. By
the way, no one has tried 7 ... d5 - though
it does look quite ugly, e.g. 8 dxe4 (8
tZ:lf4?! dxc4 9 tZ:lxe6 ii'd7 10 tZ:lxf8 .l::!.xf8 1 1
tZ:lxe4 tZ:lc6 12 'iie2 0-0-0 with the initia­
tive) 8 ... dxe4 9 'i'xd8+ �xd8 10 tZ:ld4 and
Black is already under pressure.
8 0-0 0-0 9 ..tg5!
Much stronger than 9 dxe4 tZ:lg4!; while
9 d4!? .i.b4 (9 ... i..e7) 10 'iic2 il.xc3 1 1
tZ:lxc3 d5 should be okay for Black.
9 . . ..te7 1 0 dxe4
.
 Th e T w o - Pa wn A t t a c k ( 1 c 4 b 6 2 t'iJ c 3 i.. b 7 3 e 4 )
1 0 . . . t'iJg4
Not 10 . . . d6? 11 tZ::l d4 and White has
clear plus.
1 1 i.. x e7 "W/xe7 1 2 t'iJd4 t'iJe5 1 3 b3 l2lbc6
1 4 l2lc2 l2lb4 1 5 l2lxb4
According to Ribli 15 tZ::l e3!? a5 16 c;i;>h 1
would have given White a slight advan­
tage.
1 5 . . . "W/xb4 1 6 �c1 .l:radS 1 7 f4 l2lf7 1 8
"Wid2 a 6 1 9 �f2 W/e7 20 .l:!.cf1 i.. c 6 2 1
11i'd3 1.\Vc5 2 2 <;ith 1 b5
. White hasn't made the most of his
space advantage and after this move, chip­
ping away at the structure, Black solves
most of his problems.
23 cxb5 axb5 24 .l:!.c2 Wib6 25 l2le2 e5!
26 l2lc3 b4 27 l2ld5 i.. x d5 28 11i'xd5
28 exd5 exf4 29 l::i:xf4 tZ::l e5 is unclear.
28 . . . c6 29 'iic 5 iixc5 30 �xc5 l:taS 3 1
i.. h 3 l!fdS 3 2 .l:!.d 1 .l:!.xa2 3 3 �xd7 �a 1 +
34 �g2 .l:!.a2+ 3 5 �f3 �xh2 3 6 nxc6
.l:!.xh3 37 �xd8+ t'iJxdS 38 �cS exf4 39
J:Ixd8+ <t;f7 40 'it>xf4 h5 41 J:ld7+ 'it>fS 42
J:lb7 h4 43 gxh4 .l:!.xb3 44 e5 .l:!.b1 45 e6
b3 46 <t;g3 �-�
Although the game concluded satisfac­
torily from Black's viewpoint, it appears
more promising to play 7 ... tZ::l c6!? rather
than 7 ... .it.c5.
Game 65
Heinbuch-Spassky
German Bun desliga 1984
1 c4 b6 2 l2lc3 i.. b 7 3 e4 e6 4 g3 f5 5
d3 i.. b4 6 l2lge2 fxe4
Black could delay capturing on e4,
though this has only been tested in one
game: 6 ... tZ::l f6 7 .i£.g2 0-0 8 0-0 tZ::l a6!? 9
exf5?! (simply 9 a3 is stronger) 9 ... .it.xg2 10
�xg2 exf5, as in Joksimovic-Bricard, Lyon
open 1990. After the exchange of bishops
Black is doing fine.
7 i.. g 2 l2lf6
8 0-0 0-0 .
8 .. ."ii'c8 is too passive: 9 dxe4 0-0 10 h3
tZ::lc6 1 1 .it.e3 t'Lle5 12 b3 a5 13 l::tc l tZ::l g6 14
f4 with advantage to White in Stulik­
Bores, Czech T earn Championship 1995.
9 dxe4
9 a3 .it.e7?! (9 ....it.xc3 10 tZ::lxc3 d5 is un­
clear) 10 dxe4 tZ::l c6 11 h3 e5 12 tZ::ld5 clearly
favoured White in Sehner-Schmitzer,
1 19
En glish D e fe n c e

German Bundesliga 1992.

9 . . . tZ:lc6
9 ... tba6?! 10 eS �xg2 11 �xg2 tbe8 12
a3 �xc3 13 lLlxc3 was better for White in
Schneider-Schonthier, German Bundesliga
1995.
1 0 h3

Gam e 66
Laqua-Cording
OLNN 1997
1 c4 b6 2 tLlc3 e6 3 e4 ..lib 7 4 tLlf3

1 0 . . .l:tb8
10 ... tbe5 is more ambitious and, above
all, tricky, e.g. 1 1 b3 .lieS 12 tbd4 (maybe
12 �h1) 12 ... 'iYe8 13 .llf4 d6 14 �e3 'iYg6
15 tbxe6 .llxe3 16 tbxf8 'iYxg3 17 tbe2
tbf3+ 0-1 Smejkal-Kavalek, Kettler Cup
(rapidplay) 1997.
1 1 f4 a6 1 2 a3 ..t.xc3 1 3 tZ:lxc3 d6 1 4
..te3 'ife8 1 5 'iie 2 tZ:le7 1 6 �h2 .taB 1 7
l:!.ad 1 b5
As in the previous game, once Black
gets this move in, fighting White's central
control, then he should equalise. The most sensible, and I think the best
·

1 8 c5 d5 1 9 ..t.d4 move in this position. The knight belongs

19 eS tbe4 is unclear.
1 9 . . . tZ:lxe4 20 tZ:lxe4 dxe4 21 ..t.xe4 ..txe4
22 �xe4 'ikc6 23 't'kxc6 tZ:lxc6
see follo wing diagram
... and now if anyone is better, it is
Black.
24 ..li.c3 !:.bd8 25 �g2 .:d5 ! 26 b4 J:!.fd8
27 l:!.xd5 l:!.xd5
Perhaps 27 ... exd5! ?
2 8 l:!. e 1 �f7 29 � 3 .l:td3+ 3 0 !:.e3 tZ:ld4+
3 1 ..t.xd4 l:txd4 32 We2 �f6 �-�
on f3!
4 . . . ..li.b4
Applying the usual pressure on White's
centre. In the early days of the English
Defence (I mean in the late 1970s) atten­
tion was focused on ...
5 'i!l¥b3
...until Black came up with an excellent
response:
5 . . . tLla6!
The greedy S ... i..xc3?! 6 'i'xc3 i..xe4
runs into 7 d3! i.xf3 8 'ii'xg7 'i'f6 9 i.h6!

1 20
 Th e T w o - Pa wn A t ta c k ( 1 c 4 b 6 2 li:J c 3 i.. b 7 3 e 4 )
'Wxg7 10 i.. xg7 and White picks up an
exchange.
6 i.. e 2
The point of 5 . . .'!i:Ja6 is that if White
now tries 6 a3 then 6 .. .'!i:Jc5! is a good
move as 7 'Wxb4?? loses the queen to 7 ... a5
8 'ii'b 5 i.. c 6. Alternatively, 6 d3 f5! 7 exf5
i..xf3 8 gxf3 'We7 9 <Ji?d1 i..xc3 10 'i1Vxc3
exfS 1 1 i.. h3 0-0-0 12 i..xf5 'iYf7 13 i.. e4
tt:lf6 14 i.. e3 d5 gave Black a great attack in
Ledger-Hodgson, Isle of Man open 1996.
6 . . . ti:Je7 7 d3
7 a3 is still met by 7 ... tt:lc5! and if then 8
'Wc2 i..xc3 9 'Wxc3 0-0, while after 7 0-0 0-0
8 e5?!, as in Uhlmann-Lau, Austrian Team
Championship 1996, Black should have
just played 8 ... tt:lf5, gaining control over
the d4-square.
7 . . . 0-0 8 0-0 d 5
. 8 . . . tt:lg6 is also strong, e.g. 9 i.. d2 d6 10
a3 (perhaps 10 'Wc2) 10 ... tt:lc5! (the same
trick as before!) 1 1 'Wc2 (1 1 'Wxb4? a5 12
'ikb5 c6) 1 1...i..xc3 12 i..xc3 f5! 13 exf5
�xf5 and Black had the initiative in Ser­
per-Yermolinsky, Lucerne 1993.
9 cxd5
9 exd5 exd5 10 a3 tt:lc5! (our familiar
trick) 1 1 'Wc2 i..xc3 12 'Wxc3 tt:lf5 13 cxd5
�e8 14 �e1 i..xd5 15 i..f4 tt:lb3 16 �ad1 c5
17 tt:ld2 tt:lbd4 18 i..fl �f6 19 tt:lc4 tt:lh4 20
l:i.xe8+ .:xe8 21 i.. g3 tt:lxg2 0-1 Pastorini­
Forintos, Forli open 1988.
9 . . . exd5 1 0 i.. g 5
After 10 exd5! ? i..xc3 (not 10 ... tt:lxd5 1 1
d4 and the knight o n a6 is out o f play) 1 1
bxc3 tt:lc5 1 2 'Wc2 �xd5 (or 1 2. . .tt:lxd5)
chances are balanced.
1 o . . . f6 1 1 i..h 4 'it>h8 1 2 J::t a c1 li:Jg6 1 3
i.. g 3 .1i.xc3 1 4 'ikxc3 c5
After this Black assumes control of the
centre.
1 5 a3
15 exd5 is met by 15 ... tt:lb4!
1 s . . . .:.c8 1 6 ti:Jd2 li:Jc7 1 1 .1i.g4 ti:Jb5 1 8
'ikb3 li:Jd4 1 9 �a2 f5 20 exf5 ti:Jxf5 2 1
.:.te 1 'i'g5 22 .1i.xf5 'i'xf5 23 .:.cd 1 .:.ce8
24 'i'b3 ti:Jf4 25 .1i.xf4 'i'xf4 26 ti:Jf3 d4
27 I:.xe8 l!xe8 28 .:.e 1 .l:tf8 29 'ii'd 1 h5
30 l:te5 i..xf3 31 �xh5+ i..x h5 32 'ii'x h5+
Wg8 0-1
5 ... tt:la6 is a good response to 5 'Wb3; so
much so that this move is hardly seen
anymore compared to the main move, 5
121
En g lish D e fe n c e
i.. d3, which we shall be concentrating on
in the remaining games of this chapter.
Gam e 67
Krylov-Zak
Groningen 1994
1 c4 b6 2 lll c 3 i.. b 7 3 e4 e6 4 .!bf3 i.. b4
5 i.. d 3
A strange move, blocking the cl-pawn,
but quite a good one too. If Black captures
the knight on c3 then White immediately
solves the problem of how to develop the
bishop on cl; and if Black doesn't capture
then he will have to lose time re-deploying
the bishop on b4. So White's bishop ma­
noeuvre, i..f1-d3-c2 followed by moving
the cl-pawn, doesn't look so costly as
Black also has to waste time. Generally
Black's response to this is to play the
knight to e7, castle and go for .. .f7-f5,
though in this first game with 5 �d3 I
would like to look at some example where
Black captures the knight on c3 and em­
ploys a different plan.
5 . . . llle 7
After the immediate 5 ... i..xc3!? 6 bxc3?!
(6 dxc3 is better) 6 ...d6 7 0-0 e5 8 �c2 ll:lf6
9 �a4+?! ll:lbd7 10 i..xd7+?! 'ii'xd7 Black
had no difficulties in Wulfmeyer-Rosch,
GIRL-NO 1997.
6 0-0
6 ll:le2, preserving the knight, is the sub-
1 22
ject of Game 70.
i..xc3 ! ?
For 6. . .0-0 see the next two games.
7 dxc3
Naturally White takes the opportunity
to open the diagonal for his bishop.
7 . . . 0-0
I prefer 7 ... d6. For one thing, Black
doesn't need to commit his king yet, but
also it is worth preventing the possibility
of e4-e5 - see the next note. For example,
8 ll:ld4 ll:ld7 9 f3 0-0 10 .liel e5 1 1 ll:lc2 f5
12 exf5 ll:lxf5 13 ll:le3 ll:lxe3 14 i..xe3 with
equal chances in Johansen-Sandler, Austra­
lian Championship 1995.
8 i.. g 5
This only helps Black to construct his
ideal pawn formation. 8 e5!? is more dan­
gerous: 8 .. ll:lg6 9 Itel i..xf3?! (9 ... d6 and
.
9 .. .f5 both spring to mind) 10 ilxf3 ll:lc6
11 'il'g3 d6 12 �g5 'ti'd7 13 �xg6 hxg6 14
l:i.ad1 d5 and now 15 'i¥g4 would have
given White a winning position in the
game Roos-Haik, French Team Champi­
onship 1993.
8 . . . f6 9 i.h4 e5
With the pawn on eS White's bishops
are blocked in and it is difficult to see how
White can create play for them.
1 0 i.g3 d6 1 1 �e2 .!bd7 1 2 .!bd2 .!bc5 1 3
b4 .!be6 1 4 .!Z:lb3 lll g 6 1 5 f3 lll gf4 1 6
'i'e3 c6 1 7 l:.fd 1 'i'c7 1 8 i.f1 f5! 1 9 ..tf2
fxe4 20 fxe4 c5 21 �d2 .!bg5 22 l:.d5
 Th e T w o -Pa wn A t ta c k ( 1 c 4 b 6 2 I:D c 3 SL b l 3 e 4 )
2 2 . . . JLxd5 2 3 cxd5 �f6 2 4 SLh4 �g6 2 5
� h 1 h6 26 bxc5 bxc5 2 7 CiJd2 �b8 28
I:Dc4 l!f8 29 h3 'iWd7 30 �h2 I:Dfxh3 3 1
gxh3 I:Df3+ 32 <t>h 1 I:Dxh4 3 3 1Le2 J:.f3 0-1
The early capture of the knight on c3 is
a solid alternative to the main continua­
tions. Indeed, in view of the complexity of
6 . . . 0-0 and 7 .. .f5, I can see 6 ...1Lxc3 being
employed more often.
Gam e 68
Fedorowicz-Fiear
Wijk aan Z ee 1988
1 c4 e6 2 e4 b6 3 I:Dc3 SLb 7 4 I:Dt3 .ltb4
5 SLd3 I:De 7 6 0-0 0-0 7 1Lc2
This has to be the most sensible move -
the first step to sorting out the queenside
pieces. 7 .l:f.el has been played a few times,
but not with great success, e.g. 7 .. .f5
(7 ... 1Lxc3!? 8 dxc3 d6 - compare with the
previous game) and now:
a) Not 8 eS? tLlg6 9 JLfl JLxf3 10 �xf3
tLlc6 and Black was already clearly better
in Hiibner-Miles, Bad Lauterberg 1977.
b) 8 a3 JLxc3 9 dxc3 fxe4 10 1Lxe4 1Lxe4
1 1 �xe4 tLlbc6 12 tt:Jd4 (or 12 'it'e2)
12 .. .'i'e8 with equality in Friedgood­
Keene, Cape Town 1976.
c) 8 exfS!? tLlxfS 9 i.e4 1Lxe4 and now:
cl) 10 lDxe4 tLlc6 (10 ... tLlh4!?) 1 1 a3 JLcS
12 b4 JLd4 13 l::i.b 1 4Jh4, when I prefer
Black's position.
c2) 10 l:txe4 1Lxc3? (10 ... tLlc6! and Black
is fine) 1 1 dxc3 tLlc6 12 JLf4 l:tc8 13 'i'd3
'iie 8 14 .U.ael (preventing ... d7-d6) was a
little better for White in Sunye Neto­
Hoffman, Argentina-World 1994.
7 . . . f5
Or 7... JLxc3!? 8 dxc3 d6 - once again,
compare this position with Krylov-Zak.
At this point White has a large range of
options. Apart from those given below,
the commonly played 8 'i'e2 is featured in
the next game.
8 exf5
Alternatively:
a) 8 .l:Ie1 is sensible, e.g. 8 ... 1Lxc3 9 dxc3
h6 (or 9 ... fxe4 10 1Lxe4 1Lxe4 1 1 :Ixe4
tLlbc6 12 'ife2 with a slight plus for White)
10 exfS tDxfs 1 1 4Jd4 'i'f6 12 tDxfs exfS 13
1Lf4 d6 14 cS! bxcS 15 JLb3+ 'ot>h7 16 'iVhS
with a great attack in Rotstein-Sulava,
123
En g lish D e fe n c e

Geneva open 1996. 1 0 d4 l2Jc6 1 1 d5

b) 8 d4 is, unsurprisingly, unpromising, Black is getting pushed around, but his
e.g. after 8 . . . .ixc3 9 bxc3 .i.xe4 10 .ixe4 position remains fundamentally strong.
fxe4 1 1 lbgS �e8 12 lbxe4 1Wg6 13 lbg3
lbbc6 Black was playing a good Nimzo­
Indian Defence in Rahman-Speelman, Cal­
cutta open 1996.
c) 8 eS is a mistake due to 8 ... .txf3 9
gxf3 (9 �xf3 lbbc6 10 1i'e3 .tcS 1 1 'i¥g3
lbg6 12 �e1 .id4 wins the e-pawn) 9 ... lbg6
10 d4 1\Vh4 (10 . . . c5 1 1 a3 .ixc3 12 bxc3
'it'h4 13 'ith1 lbc6 is also strong, as in
Contin-Vera, Saint Vincent open 1998) 1 1
lbe2 lbc6 1 2 f4 'iig4+ 1 3 lbg3 'iih3 1 4 f3
lbxd4 with a winning position in Zlot­
nikov-Shabalov, New York open 1997.
8 . . .l2Jxf5 9 l2Je4 1 1 . . . l2Jb4 1 2 �b3 l2Ja6 1 3 l2Jg3 l2Jxg3 1 4
After 9 .i.e4 'iic 8!? (or 9 ... .txe4 10 hxg3 l2Jc5 1 5 i.c2 i.f6 1 6 :b 1 a 5 1 7
lbxe4 lbh4) 10 'iVb3 .i.cS 11 'iic2 lbc6 i.e3 exd5 1 8 cxd5 Wie7 1 9 a3 l::ta e8 20
Black was already better in Uhlmann­ l::te 1 1i'd8 2 1 b4 axb4 22 l::txb4
Vera, Berlin 1982. After 22 axb4 lbe4 23 "i'd3 g6 Black has
enough counterplay.
22 . . . d6 23 "i'b 1 h6 24 i.h7+ 'it>h8 25
i.g6 i.xd5 26 i.xe8 J::!. xe8

9 . . . �e7
Much too passive. I think the right
move is 9 ... lbh4! with the following pos­
sibilities:
a) 10 lbegS? lbxf3+.
b) 10 lbfgS h6 (10 . . . lbxg2 is unclear) 1 1
lbh3 dS and Black has the initiative.
c) 10 lbxh4 'iixh4 when White doesn't
have time to build a strong centre; Black
has the attack going already.
d) 10 lbeS d6 1 1 lbg4 dS with a power­
ful initiative.
Even here, Black shouldn't be worse:
his minor pieces are well placed.
27 i.d4 i.xf3 28 J:xe8+ 'i'xe8 29 i.xf6
i.e4 30 'i'c 1 'i'g6 31 i.d4 l2Jd3 32 'ifxc7
l2Jxb4 33 axb4 b5 34 'iYb8+ 'it>h7 35
'ifxb5
It is unpleasant to play, but I don't
think Black should lose. Anyhow, this has
nothing to do with the opening ...

1 24
 Th e T w o - Pa wn A t t a c k ( 1 c 4 b 6 2 [jj c 3 JJ.. b 7 3 e 4 )
3 5 . . . i.xg2 3 6 'iitx g2 'ife4+ 37 �g 1 'i'xd4
38 'iff&r- 'iit h 8 39 b5 d5 40 'i'e6 'il'd 1 + 4 1
�g2 d4 42 'il'e8+ �h7 4 3 'il'e4+ g6 44
b6 d3 45 'il'e7+ Wh8 46 b7 1 -0
Fedorowicz's opening play turned out
well, but try 9 ... tt:Jh4! Looking at all the
eighth move alternatives, I think White
does best by remaining 'solid' with 8 l:i.e1
or 8 �e2 in the next game. The flashier
ideas tend to give Black a good attack.
Gam e 69
Rayner-Piaskett
Lon don (Lloy ds Ban k) 1993
1 c4 b6 2 {jj c 3 e6 3 e4 i.b 7 4 {jjf3 i.b4
5 i.d3 [jj e 7 6 0-0 0-0 7 i.c2 f5 8 'il'e2
The most popular move in this posi­
tion, for what that is worth.
8 : . . JJ.. x c3
In principle I would like to play
8 ... tt:Jbc6 here, but it is difficult to 'make it
work'. Koshy-Lovlu, Sakthi 1996, contin­
ued 9 d4 i.. xc3 10 bxc3 and now:
a) 10 ... 4Ja5 appeals to me, but White
has a big space advantage, and that, com­
bined with pins and threats on the king­
side gives him the better chances, e.g. 1 1
i.a3 i.xe4 (after 1 1. . J!e8 1 2 4Jd2 I would
be worried about Black's king) 12 i..xe4
fxe4 13 1i'xe4 tLlxc4? 14 tLlg5 .l:i.f5 (or
14 ... g6 15 1i'h4) 15 i..xe7 and White is
wmnmg.
b) The actual game continued 10 ... "ii'e 8
11 d5 t2Jd8 12 tt:Jd4 i.. a6 13 i..a3 f4 14 i.. a4
c6 15 dxc6 dxc6 16 .l:i.ad1 and Black's posi­
tion is miserable - though he later won.
9 dxc3
9 bxc3 is less good after 9 ... i..xe4! 10
i..xe4 fxe4 1 1 "ii'xe4 tLlbc6 12 d4 reaching,
once again, a kind of Nimzo-Indian posi­
tion which presents few difficulties for
Black, e.g. 12 ... tLlf5 13 d5 ti:JaS 14 dxe6
�e7 15 c5 'ifxe6 16 'ifxe6+ dxe6 17 l:te1
�fe8 18 i.f4 tLlb7 19 g4 tLle7 20 i..xc7
tLlxc5 21 k!.ad1 tt:Jd5 22 i..e5 b5 23 tt:Jg5
�-� Ioseliani-Matveeva, European
Women's Team Championship 1997.
9 . . . i.xe4
9 .. .'it'e8 was played in Hjartarson-Lau,
German Bundesliga 1990/91, resulting in a
very similar position to the main game
after 10 l:i.e1 i..xe4 1 1 i..xe4 fxe4 12 ii'xe4
tLlbc6 13 i..f4 k!.c8 14 .l:tad1 h6 15 h3 "ii'f7
16 i..h 2. White has an excellent position
with his heavy pieces bearing down on the
centre pawns; they are a long way from
rolling down the board, and in the mean­
time Black's pieces are seriously cramped.
Alternatively, 9 ... tLla6 doesn't help ei­
ther due to 10 i.. g5 (10 b4!?) 10 ... fxe4 1 1
i..xe4 tLlc5 1 2 i..xb7 tLlxb7 1 3 ltad1 ti:Jd6
14 tLle5 with similar pressure in Koshy­
Shantharam, Indian Championship 1994.
1 0 i.xe4 fxe4 1 1 'it'xe4 [jj b c6 1 2 i.f4
This kind of position should be familiar
1 25
En g lish D e fe n c e
t o you by now! I have a feeling that the
Black players who went in for it just imag­
ined that the centre pawns would give
them a reasonable position. If the queens
were off the board, enabling the king to
march into the middle of the board, then I
think Black would be better, but that is
just a dream.
1 2 . . . lLlf5 1 3 l:Iad 1 l:Ic8 1 4 l:Id2 wt'e8 1 5
.l:!.e 1 !
Good move, preventing ... d7-d6 and
... e6-e5.
1 5 . . . h6 1 6 h3 d6?
It is interesting that Black becomes frus­
trated with his position so quickly, and
gives up a whole pawn just to free himself.
Technically, he is lost, but in the end
Plaskett pulls it off.
1 7 ifxe6+ wt'xe6 1 8 l:txe6 l:tce8 1 9 l:txe8
l:txe8 20 g4 lLlfe7 21 �g2 ttlg6 22 .tg3
l:te4 23 b3 �f7 24 ltJd4 ttlxd4 25 cxd4
ltJe 7 26 c5 �e6 27 �f3 �e 1 28 cxd6
cxd6 29 l:Ic2 �d7 30 ..i.f4 ttlg6 3 1 .tg3
l:td 1 32 l:tc4 b5 33 llb4 �c6 34 a4 a6 35
'it>e2 .:l.b1 3 6 axb5+ axb5 37 d5+ �c5 38
l:.e4 �xd5 3 9 :tea l;Ib2+ 40 �d3 l:txb3+
41 �c2 :l.b4 42 l:tg8 l:tc4+ 43 Wb3 .l:!.c7
44 l:td8 ttle5 45 l:tb8 �c5 46 .txe5 dxe5
47 .l:!.e8 �d4 48 �b4 .l:.f7 49 .l:!.d8+ �e4
50 .l:!.d2 l:tf3 51 �xb5 l:txh3 52 .li!.d7 g5
53 J:!.f7 l:tf3 54 l:!.h7 l:txf2 55 .l:!.xh6 �f4
56 l:tf6+ �g3 57 :ea l;Ie2 58 �c4 l:te3
0-1
126
A solid approach by White has paid
dividends in the example we have looked
at. Black has to find an improvement early
on after 8 1i'e2 to avoid falling into the
kind of passive position we have seen too
often.
Gam e lO
Agrest-Zviaginsev
Kazan 1997
1 c4 e6 2 ttlc3 b6 3 e4 .tb 7 4 ttlf3 .tb4
5 .td3 ttle7 6 ttle2 ! ?
This can also b e tried after both sides
have castled. The knight runs over to the
kingside, leaving the bishop on b4
stranded and looking silly. Then again, the
bishop on d3 isn't too clever either, so it is
going to take a few moves before every­
thing straightens out and we will be able
to see who is doing what to whom.
6 . . . f5
6 ... 0-0? loses to 7 a3 .ltd6 8 eS .ltxf3 9
exd6 .ltxe2 10 dxe7 ..ixd1 1 1 exd8'ii' �xd8
12 r;itxdl.
7 ttlg3
7 'i'c2 is sensible, e.g. 7... .ltd6 (7...fxe4 8
.i.xe4 .i.xe4 9 'ii'xe4 tbbc6 10 a3 .ltd6 1 1 d4
0-0 also slightly favoured White in Eng­
Wiemer, German Championship 1984;
but not 7 .. 0-0? 8 a3 .ltd6 9 eS and wins) 8
.
a3 tbg6 9 exfS tbh4 10 .lte4 tbc6 1 1 fxe6
0-0 12 d4 and White stood much better in
 Th e T w o - Pa wn A t ta c k ( 1 c 4 b 6 2 !D c 3 i.. b 7 3 e 4 )

Mitenkov-Ivanov, Moscow open 1994.

7 . . . !Dg6 8 0-0 f4
After his game against Chemin (see
later on) Speelman made the suggestion of
8 . . . 0-0!?
9 !De2 0-0

White does best to keep the centre

fluid. Alternatively, 14 dS lL:lceS 15 lL:lxeS
and now:
a) 1S ... l2::lxe5 16 f3 (16 J..c3 f3 favours
Black; while 16 J..xf4 l2::l f3+ 17 gxf3 !:txf4
18 l2Jxf4 'ii'xf4 19 Wid3 J..d6 is unclear)
1 0 i.. b 1 16 ... g5.
I don't see why White doesn't just put b) 15 ... 1Wxe5 16 J..c3 'ii'g5.
the bishop on c2. In effect he gains a move In both cases Black has sufficient play.
on the game when it comes to developing 1 4 . . . e5 1 5 dxe5
his queenside. Not 15 d5 l2Jd4!
1 0 . . . i.. e 7 ! ? 1 5 . . . !Dcxe5 1 6 !Dxe5 !Dxe5 1 7 f3
1 0. . .l2Jh4 goes directly for the attack on
the kingside, but it is just a bit too crude:
1 1 l2Jxh4 'ii'xh4 12 f3 eS 13 d4 l2Jc6 14 a3
J.. e7 15 dxeS lL:lxeS 16 'it>h1 J.. d6 (perhaps
16 ... g5! ?) 17 J..xf4! lh£4 18 g3 l:i.xe4!? 19
gxh4 and 1-0 a few moves later in
Chernin-Speelman, European Club Cup
1997.
1 1 d4 c5 1 2 i.. d 2
If White plays 12 dS?! then after the
centre is closed with 12 ... e5 Black can go
for it on the kingside. And if 13 d6 J.. f6
and the black knight lands on d4, cutting
off the d-pawn. 1 7 . . . !Dg6? !
1 2 . . . 'ifc7 1 3 b3 !Dc6 What about 17 ... J..f6 . . . ? It is obviously
13 . . . e5!? 14 dxeS (14 dS is slightly better more desirable than the text, the only
for Black) 14 . . . l2::lxe5 15 lL:lxeS 'ii'xeS 16 question is whether it works tactically: 18
J.. c3 WigS 17 f3 is unclear. l2::lxf4 l2::l g4! ? (18 ... l2::lxf3+ 19 !:txf3 J..xc3 20
1 4 i.. c 3 lL:ldS J.. d4+ 2 1 'ii'xd4; and 18 ... l2::lxc4 19
see following diagram
lL:ldS both win for White) 19 J..xf6 l:txf6
20 fxg4 .l:txf4. As extra pawns go, it is a

127
En glish D e fe n c e
pretty rotten one. I would even say that
Black has the better chances. White is pay­
ing for putting the bishop on b 1 and not
c2 many moves ago.
1 8 lili'd2 d6 1 9 ii.c2 .l::.a d8
19 . . ..tf6 20 .l:i.adl .te5 would have been
stronger.
20 .l:tad 1 .i.f6 21 .i.xf6 ltxf6 22 e5!
Giving the bishop some life; White has
the better chances.
22 . . . lZJxe5 23 lZJxf4 lili'f7 24 lZJd5 ii.xd5
25 cxd5
25 'fixd5 is better, e.g. 25 ... 'ii'xd5 26
1 28
�xd5 tt:lc6 27 .te4 tt:ld4, though with that
knight Black should hold.
25 . . . l:.f8 26 .i.e4 h6 27 lili'e 1 g5 28 'i'e2
l:.f4 29 a4 h5 30 a5 bxa5 3 1 'i'a6 lili'c7
32 l:.a 1 g4 33 'i!Vxa5 'ii'x a5 34 .l:lxa5 gxf3
35 gxf3 lZJxf3+ 36 .i.xf3 .!':txf3 37 .:!.xf3
.!':txf3 38 .l:txa7 .!':tf5 39 b4! cxb4 40 l:tb7
.!':txd5 41 l:.xb4 �f7 42 �f2 ltd2+ 43 �g3
!le2 44 lth4 l:te5 45 �f3 d5 46 l:.a4 Wf6
47 �f4 .!':te2 48 lta6+ lte6 49 lta4 l:.c6
50 We3 J:!.c2 51 l:.f4+ �g6 52 J:!.d4 J:!.c3+
53 Wf4 % - %
After 5 3. . J:tc4 (53 . . .l':tc5 5 4 �e5 is
slightly better for White; and 53 ... l':th3 54
l':txd5 l:i.xh2 55 l':td6+ is equal) 54 �e5
l:rxd4 55 �xd4 �f5 56 �xd5 �g4 57 �e4
the game is drawn. I am sure that we will
be seeing more of 6 tt:le2. A couple of
things strike me: first, that White could
improve his position greatly by playing 10
..ltc2 instead of ..ltb 1; and that Black does
not have to close the position in the first
place - see Speelman's pawn sacrifice.
Moreover, 7 'ii'c2 is an unpretentious
move that, at first glance, gives White an
easy game.
 Th e T w o - Pa wn A t ta c k ( 1 c 4 b 6 2 ti:J c 3 iL. b 7 3 e 4 J

Summary
I enjoyed looking at the games in this chapter - quite original positions often arise when
White plays the 'two-pawn attack'. Short's solid idea in Game 61 reminds me a little of
the game Pytel-Piasetski (Game 38) where Black also brought the bishop to c5: worth a
try. White's fourth moves in Games 62-65 are not a cause for concern (I would prefer
Black's treatment of 4 g3 in Game 64; in particular 7 .. .'�Jc6!?). 5 1i'b3 has been put out of
business by 5 ... tZ:la6 - Black equalises easily; which means that 4 tZ:lf3 and 5 i.d3 is the
only proper test for Black. I would like to play as in Games 68-70 for Black, with ... 0-0
and .. .f7-f5, but if White players get wise and don't go too crazy (Game 68) then it might
be time for a rethink. There could be a way to equalise here for Black, but it isn't imme­
diately obvious to me. In which case I would recommend returning to Game 67. The
capture on c3 is positionally desirable and, on the evidence of those games, a solid way to
treat the position.

1 c4 b6 2 ti:Jc3 iL.b7 3 e4

3 . . . e6
3 . . . e5 - Game 61
4 ti:Jf3
4 d3 - Game 62
4 tZ:lge2 - Game 63
4 g3 f5 (D)
5 ..tg2 - Game 64
5 d3 - Game 65
4 . . . iL.b4 5 iL.d3 (DJ
5 'itb3 - Game 66
5 . . . ti:Je7 6 0-0
6 tZ:le2 - Game 70
6 . . . 0-0
6 . . . ..txc3 - Game 67
7 iL.c2 f5 (DJ 8 'it'e2
8 exf5 - Game 68
8 . . . iL.xc3 - Game 69

4 . . f5
.
5 i.. d3 7. . . f5

1 29
 CHA PTER EIGHT I
Systems with . . . �xf3

The first few chapters of the book dealt Indian Defence.

with White's most critical response to
l . . .b6: the occupation of the centre by Gam e 71
three pawns. If Black fails to counter effec­ King-Piaskett
tively then he could be squashed off the Lon don 1991
board. In that respect this chapter is far
less important. White doesn't attempt a 1 c4 b6 2 lLlf3 i.b7 3 g3
refutation of the opening, but merely con­
tinues with the standard 'Reti/Catalan'
development, fianchettoing the king's
bishop in the hope that Black will trans­
pose into something 'normal'. Although
there is nothing wrong with allowing a
transposition, that would not be in the
spirit of the pioneering English Defence. If
you play this opening you have already
accepted that you are doing something
slightly offbeat so as to lure your oppo­
nent onto unfamiliar territory. And this is
not the moment for orthodoxy. Instead
Black can exchange the bishop on b7 for When I see my opponent bottle out of
the knight on f3, doubling White's pawns, the three pawns challenge I feel provoked
and so gaining a slight structural advan­ into doing something 'different' and, luck­
tage. Whether this is enough to compen­ ily, there are some interesting and rea­
sate for the 'loss' of the bishop remains to sonably sound methods of knocking
be seen. White out of his complacency. Like this:
Games 71-77 show Black playing 3 . . . i.xf3 ! 4 exf3 c5!
... SLxf3 against a Reti set-up; while the Black must follow up in this manner or
final two games feature Black trying he will have just given up his bishop for
...SLxf3 to avoid the standard Queen's no benefit. The aim is to seize control of

1 30
 S ys t e m s with . . . .il. x f3

the d4-square by playing ... tbc6 and ... g7-

g6 followed by . . . 1Lg7. It is a potent
positional threat which White must coun­
ter immediately with ...
5 d4 !
By the way, here's a quick taste of the
mess White gets into if he doesn't play 5
d4: 5 tbc3?! tbc6 6 i.g2 g6 7 0-0 i.g7 8 f4
e6 9 !!e1 tt:lge7 10 d3 0-0 and Black already
stood better in Buchegger-Schneider Zin­
ner, Austrian Team Championship 1996.
He has the d4-square firmly under control
and is ready to break open with ... d7-d5.
After 5 d4, most players capture on d4, 8 .td2
and the majority of the games in this sec­ I was impressed by Plaskett's idea and
tion concentrate on that move; but there decided to try it myself at the next oppor­
are some who know better ... tunity - which came about two years
later. Instead of moving the bishop my
opponent tried 8 tbc3 g6 9 i.h3 tbxe3
(9 ... i.g7!?) 10 fxe3 i.g7 1 1 'ii'd3 tbf6 12 d6
(12 0-0 would have been more sensible,
with equal chances) 12 ... e6 and now:
a) 13 e4 was rejected by my opponent,
though obviously it is critical, e.g. 13 ... 0-0
14 e5 tbe8 15 f4 f6 16 'ir'e3 i.h6! (but not
16 ... fxe5? 17 fxe5 and the bishop doesn't
get out) and White's centre won't survive.
b) In the game he chose 13 lbbS, but
the cl-pawn is doomed: 13 . . 0-0 14 tbc7
.

l:i.c8 15 lld1 tbe8 16 tbxe8 .l::i.xe8 17 b3 �c6

5 . . . lLlc6 ! ? 6 d 5 18 f4 i.f8 19 0-0 .l:txd6 with a clear extra
The most critical response. Instead, 6 pawn in Dunn-King, Dublin Zonal 1993.
dxc5 bxc5 7 i.g2 g6 8 0-0 i.g7 9 f4 !!c8 8 . . g6 9 .il.c3 .il.g7 1 0 .il.xg7 lLlxg7 1 1 f4
.

should be okay for Black as the c5-pawn

gives him control over d4.
6 . . . lLld4
A wonderful square for the knight,
though White can exchange it if necessary.
7 .il.e3 lLlf5

see following diagram

7 ... e5! ? is possible, though it does give

White something to aim at, which isn't
quite in the spirit of the opening. Black's
centre will be attacked by f2-f4.

131
En glish D e fe n c e
Now I regain control over the d4-square.
1 1 . . . lLlh6
After 1 1...t'Llf5?! 12 t'Lld2 t'Lld4 13 t'Llf3
Black is forced to exchange with
13 ... t'Llxf3+ 14 'i!Vxf3 t'Llh6?! 15 �h3! and
knight no.2 doesn't make it to d4.
1 2 lLld2 0-0 1 3 i.h3 e6 1 4 l2Jf3 'i'f6 1 5
:b1 d6
15 ... exd5?! 16 "Yi'xd5 �fe8+ 17 '>i>fl fol­
lowed by 'it>g2 is simply good for White.
1 6 dxe6
Perhaps 16 0-0!? e5 17 fxe5 dxe5 18 Jael.
1 6 . . . "i'e7 ! 1 7 0 - 0 fxe6 1 8 :e 1 lLlh t 5 1 9
'ii' e 2 l:.ae8 20 �d2 'iff6 Y2 - Y2
In the final position chances are about
equal. Actually, I couldn't see what to do
next, and neither could my opponent. For
instance, doubling on the e-file is possible,
but then what? And I didn't like the look
of 21 b4 cxb4 22 'ili'xb4 �c8 followed by
.. Jk5 and doubling on the c-file.
5 ... t'Llc6 is chancing it, but there is no
obvious refutation and it does have the
advantage of making one's opponent
think. In view of the popularity of
3 ...�xf3, I'm surprised that it hasn't been
played more often.
Gam e 72
Donchenko-Muhutdinov
Alus hta 1993
1 c4 b6 2 lLlf3 i.. b 7 3 g3 i..xf3 ! 4 exf3
1 32
c5! 5 d4! cxd4 6 "i'xd4 lLlc6 7 'i'd2
By far the most popular retreat for the
queen.
7 . . . g6
In place of the standard 7 ... e6 (see the
next game) this fianchetto has been tried
in a few games, but it doesn't appeal to
me. When Black plays ...e7-e6 the d5-
square is covered; but in this case White's
knight constantly threatens to land on d5.
8 b3
White decides to offer the exchange of
bishop. This is okay, but not absolutely
necessary as the following demonstrates: 8
t'Llc3 �g7 9 b3 tt:lf6 10 �b2 0-0 1 1 �g2 a6
12 0-0 .l:!.b8 13 f4 b5 14 t'Lld5! t'Lle8 15 t'Lle3
�xb2 16 'i'xb2 t'Llf6 17 .l:Ifd1 'it'b6 18 �ac1
l:tfd8 19 c5 'iic7, as in Tonoli-Lau, Belgian
Team Championship 1997, and now 20 h4
would have maintained White's excellent
position.
8 . . . i.g7 9 i.b2 i.xb2
In principle I think Black should be
glad to exchange off a pair of bishops,
even though it makes his king draughty.
Instead, 9 ... t'Llf6 10 �g2 0-0 1 1 f4 �c7 12
0-0 �ac8 13 t'Llc3 d6 14 !He 1 gave White
the usual space advantage in Egeli-Madsen,
Norwegian Championship 1996. That
game continued 14 . . . .l:!.fe8 15 f5 a6 16 �ac1
tt:le5 17 fxg6 hxg6 18 t'Lld5 t'Llxd5 19 �xd5
e6 20 �g2 t'Llc6 21 h4 �xb2 22 'i'xb2 e5 23
h5 t'Lld4 24 hxg6 fxg6 25 �d5+ �g7 26

Wg2 �h8 27 �cd1 �cf8, and now 28 .!Ixd4
exd4 29 'i¥xd4+ .!If6 30 g4 wins for White.
1 0 Wkxb2 lLlf6 1 1 f4 0-0 1 2 i.. g 2 'iYb8 1 3
0-0 b 5 1 4 c 5 b4 1 5 ltJd2 'i'kb5 1 6 l:tfc 1
�ac8 1 7 i..f 1 'i'kb7 1 8 i.. g 2 l:!c7 1 9 l:te 1
Wlb5 20 l:tac 1 .l:!.fc8 2 1 �c4 ltJe8 22 f5
With practically Black's entire army
camped out on the other side of the board,
White feels justified in exploiting his space
advantage to launch a massive assault on
the king. It isn't strictly necessary though
as, for instance, the alternative 22 tLle4 is
strong.
22 . • . gxf5 23 g4 ltJa5 24 �xe7 ltJxc4 25
bxc4 'i'xc5 26 �e5 iff8 27 �xf5 'ifg7 28
'tixb4 '\Wxg4 29 l:!.f3 'ifd4 30 �g3+ ltJg7
3 1 i.. d 5 �h8 32 lLlf3 Via 1 + 33 �g2 ltJe6
34 'ifd6 f6 35 �g4 a5 36 h3 Wkc3 37
'ife7 f5 38 ltJg5 ltJxg5 39 �xg5 J:!.b8
S ys t e m s with . . . i.. x f 3
40 Wh2!? �cc8 41 .!IxfS is still better for
White.
40 . . . .1:!.cc8 41 .l:!.g3 'ii'd4 42 �d3 �b6 43
Vlixd7 .!i'!.g6+ 44 l:!.g3 �xg3+ 45 fxg3
Vlid2+ 46 �3 .l:!.g8 47 Vlixf5 'iic3+ 48
�e4 .l:!.e8+ 49 �4 'ifd4+ 50 i..e4 'i'f2+
51 i..f3 'i'e3+ 52 �g4 J:!.g8+ 53 'ith4
'i'e7+ 54 �h5 h6 55 h4 �f8 56 'i'd5
Vlie8+ 57 �g4 '\Wc8+ 0-1
The fianchetto of the king's bishop
gives White a very easy game. Black does
best to stick to a system with ... e7-e6, as
we see in the next game.
Gam e 73
Moskow-Shabalov
New York 1993
1 c4 b6 2 ltJf3 i.. b 7 3 g3 i.. xf3 4 exf3 c5
5 d4 cxd4 6 'i'xd4 ltJc6 7 'i'd2 e6 8 i.. g 2
8 tLlc3 is more canny, leaving the
bishop to defend the pawn on c4 for the
time being - see Games 74-77 (by transpo­
sition) .
8 . . . i.. b 4! 9 ltJc3 fif6
In this way Black ensures that he dou­
bles the c-pawns.
1 0 0-0 i..xc3 1 1 bxc3 ltJge 7 1 2 c5
12 �a3 is a critical alternative:
a) 12 ... 0-0?! 13 'ii'xd7 .l:lfd8 14 'i!Vb7
!Idb8 15 �xe7 tLlxe7 16 'ii'a6.
b) 12 ... tLle5?! 13 .!Ifd1 (13 f4 lLlxc4 fa­
vours Black) 13 ... �d8 14 'ii'e2 �c8 15 cS
1 33
En g lish D e fe n c e
bxcS 1 6 f4 tLl5c6 1 7 i.xcS and White is on
top.
c) 12 . . . tLla5! 13 1i'd3 (13 .l:tfd1!? .U.d8)
13 .. :iif5 14 "ii'd4 eS 15 'iVd6 �e6 16 f4 �c8
17 fxeS tLlxc4 18 ii'xe7+ 1\Vxe7 19 i.xe7
�xe7 20 !hd1 lieS 21 .:.fe1 .U.hc8 22 i..h 3
!l8c7 and Black held the advantage in
McNab-McKay, Scottish Championship
1988.
1 2 . . . bxc5 1 3 i.a3 0-0
Alternatively, 13 ... c4 14 i.. d6 0-0 15 f4
.l:i:fd8 16 i.c7 dS 17 i.xd8 �xd8 18 l:!ab 1
tLlc8 19 it'e3 tLlb6 and Black had good
compensation for the exchange in Re­
schke-Cicak, OLO-C 1996.
1 4 i.xc5 l:.fd8 1 5 l:!ab 1 d6 1 6 i.e3 h6
1 7 f4 d 5 1 8 :b7 lbf5 1 9 i.c5 .!:tacB 20
c4 dxc4 21 :d7 lbcd4 22 J:txdB + 'it'xdB
23 'i'b4 c3 24 g4 lbe2 + 25 �h1 c2 0-1
r-------,
Gam e 74
Van Wely-Zviaginsev
New Yor k o pen 1997
.________________.
1 c4 e6 2 lbc3 b6 3 g3 i.b7 4 liJf3 i.xf3
5 exf3 c5 6 d4 cxd4
6 ... tLlc6?! is not as successful here, with
tLlc3 and .. e7-e6 thrown in, as it was in
Game 71. For example, 7 dS tLld4 8 i.e3
tLlfS 9 i.h3 tLlgh6 10 'ii' a4! g6 1 1 0-0 tLlxe3
12 fxe3 exdS 13 :!ad1! with tremendous
pressure for White in Davies-Plaskett,
London 1991 .
1 34
7 fixd4 lbc6
This is the starting point for a great
many games.
8 'iid 2
The usual retreat. 8 iVdl is the subject
of Game 76 and 8 li'd3 of Game 77.
8 . . . �f6 ! ?
Alternatively, 8 . . .i.c5 isn't bad. Black
develops sensibly and unpretentiously: 9
i.g2 tLlf6 10 0-0 0-0 1 1 f4 'ii'e7 12 b3 .l:tab8
13 i.b2 �fd8 14 Itad1 a6! (it would be
highly desirably for Black to play ... b6-b5,
chipping away at White's centre) 15 a3?
(cracking up) 15 ... tLla5 16 tLle4 tLlxe4 17
i.xe4 i.xa3 and Black was on top in Re­
schke-Kulaots, Weilburg open 1995.
8 .. J:k8 is the subject of the next main
game.
9 .Jl.g2 �e5+
There is nothing wrong with 9 ... i.b4,
transposing to the previous game.
1 0 iie2 .!:tea 1 1 .Jl.d2 t5! ?
Taking away an important square from
White.
1 2 .Jl.e3 lbf6 1 3 0-0 .Jl.e7 1 4 �fd 1 0-0 1 5
h3 .Jl.c5 1 6 f4 �bB 1 7 �ac 1 liJdB 1 8
i.d2 lbt7 1 9 a3 �feB 20 b4 i.fB 2 1 .Jl.e3
J:tedB 22 i.d4 lUeS
see following diagram
Black's set-up is a rather passive, but it
is very solid and he manages to unwind
successfully.

2 3 !tb1 g6 24 !tdc 1 .1J.. g 7 25 .1J.. e 3 tt:Jc7
26 'ifd3 d 6 27 a4 d5 28 cxd5 i.xc3 29
.l:txc3 tt:Jxd5 30 I:!.xc8 'itxc8 31 I:!.c1 'ii'd 7
3 2 i.xd5 exd5 33 'ii'd 4 l:re8 34 b5 l::te 4
35 Wif6 d4 36 .1J..d 2 I:!.e6 Y:z - Y:z
37 �c7 'it'xc7 38 'it'xe6 is dead level.
,------.
Game 75
Piesina-Kengis
Riga Zonal 1995
1 c4 b6 2 tt:Jc3 i.b 7 3 lLlf3 e6 4 g3 i.xf3
5 exf3 c5 6 d4 cxd4 7 'i'xd4 tt:Jc6 8 'i'd2
I:!.c8
9 b3
Or 9 .ltg2 tDf6 10 b3 i..b4 11 a3 (1 1
i.. b 2 d5!) 1 1....i.xc3 12 'i¥xc3 d5 13 .id2
(13 cxd5 CDaS!) 13 . . . dxc4 14 bxc4 CDaS 15
0-0 0-0 16 .l:i.fd1 .l:i.xc4, pawn up, thank
you, Otto-Haubt, German Bundesliga
S ys t e m s with . . . .1J.. x f3
1994.
9 . . .tt:Jf6
In the light of experience ... ? In a game
five years previously Kengis had tried
9 .. .'tWf6 and now the best move is:
a) 10 .i.g2 when I imagine it must have
been Kengis's intention to play:
a1) 10 ... d5!? with the following possi­
bilities:
al l) 1 1 cxdS 1 l....ib4 12 .ib2 CDce7 and
wms.
a12) 1 1 0-0 .i.b4 12 .ib2 d4 13 i:tad1
'ille7 and wins.
a13) 1 1 .i.b2 dxc4 (1 l....i.b4 12 a3) 12
0-0 .i.b4 is unclear.
a2) 10 ... .i.b4 1 1 .i.b2 'ili'e5+ 12 li'e3 and
White's position is secure.
b) In the actual game White tried 10 f4?!
d5 1 1 .i.b2 d4 12 CDe4 li'g6 13 li'e2 .i.b4+
14 'iii>d 1 CDh6 15 l!Vf3 0-0 16 ..td3 f5 17 a3
CDg4 18 <t>e2 CDge5! and he was in big
trouble in Titz-Kengis, Vienna open 1990.
10 i.b2 i.b4! ?
. Clever; Black wishes to tempt a2-a3,
when the queenside pawn chain has been
weakened.
1 1 a3 .1J..e 7 1 2 .te2 tt:Ja5 1 3 'ii'd 1 0-0 1 4
lLlb5
Perhaps 14 0-0!?
1 4 . . . a6 1 5 lLld4 d5
Kengis stirs it up while his opponent's
king is still in the middle of the board, a
decision which is justified in a practical
1 35
En g lis h D e fe n c e
game, but isn't quite so sound when exam­
ined in the cold light of day.
1 6 cxd5 'i'xd5 1 7 �xa6 J:l.cd8 1 8 b4
�c5 ! ?
18 . . .'�Jc4 1 9 .ltxc4 'ii'xc4 2 0 �cl 'i!V a2 2 1
'iib 3 'ii'xb3 2 2 tt:lxb3 i s clearly better for
White.
1 9 bxc5
19 bxa5? .ixd4 20 'ilxd4 'i\i'xa5+ 21 �b4
'ii'xa6 22 .ixf6 gxf6 slightly favours Black.
1 9 . . . bxc5 20 ltlc2 'i'b3 2 1 'i'c 1 'ifb6 22
�xf6!
Not 22 .ie2? tt:lb3 and now:
a) 23 'i'b 1 'ii'a5+ 24 tt:lb4 (24 'it>fl tt:ld2+
also wins) 24 . . . tt:lxa1.
b) 23 'i\i'g5 tt:lxa1 24 .ltxf6 'i1Yb 1+.
In both cases with a winning position
for Black.
22 . . . gxf6
1 36
23 0-0?
23 .ie2! is the move, though White
needs steady nerves: 23 ... tt:lb3 24 'iib2 .l:i.d2
(24 ... 'i¥a5+!? 25 Wfl tt:lxa1 26 'ii'xa1 .l:i.d2 27
tt:le3 �xe2!? 28 Wxe2 'iWh5+ is fun, though
I don't believe it) 25 .l:i.b 1 'i\i'a5 26 'ii'xb3
.l:i.d3+ 27 'it>fl .l:i.xb3 28 .l:i.xb3 and with care
White should be able to untangle and
come out on top.
23 . . . '1/ixa6 24 'i'f4 'ifd3 ! 25 tt:le3 'i'd4 26
'ilic7 tt:lb3 27 J:l.ad 1 'i'e5 28 'ifxe5 fxe5
29 c3;>g2 f6 30 J:l.xd8 J:l.xd8 31 .J:l.b1 J:l.d3
32 a4 �g7 33 tt:lc4 l:.c3 34 tt:ld6 c4 35
J:l.b2 J:l.c1 3 6 tt:lxc4 J:l.xc4 37 J:l.xb3 J:l.xa4
Yz - Yz
It is interesting to note that Kengis
changed his mind and played 9 . . . tt:lf6 in­
stead of 9 .. .'iWf6 when faced with the same
position in a later game. Having said that,
I don't see anything wrong with 9 .. .'i*'f6.
Black has good counterplay.
Game 76
Appolonov-Lempert
Katowice 1992
1 c4 b6 2 ltlf3 .ib7 3 ltlc3 e6
3 ... c5!? is worth a thought so that if 4
g3?! ii.xf3! 5 exf3 tt:lc6 and Black already
has control over d4. Naturally, 4 g3 is not
the best move.
4 g3 �xf3 5 exf3 c5 6 d4 cxd4 7 'iixd4
tt:lc6 8 "ifd 1 ltlf6

9 i.e2
As we saw in the previous game, one of
Black's idea's is a rapid assault on the c4-
pawn, so White is understandably a little
nervous about moving the bishop to g2.
9 . . . i.b4 1 0 0-0
I don't see what is wrong with 10 i.d2,
preventing the doubling of the c-pawns.
1 o . . i.xc3 1 1 bxc3
.
The play now becomes double-edged.
Black has lots of weak pawns to aim for;
while White has open lines for his rooks
and bishops.
1 1 . . . d6
1 1. . .0-0 is also possible. In fact I think I
prefer it, e.g. 12 .i.a3 .l:!e8 13 i.d6 tZ:laS
followed either by . . . .Uc8 or ... tZ:lb7.
1 2 i.a3 We7 1 3 'i'id2 'iic 7 1 4 l:lad 1 .!:tad8
1 5 l:Ue 1 li:Ja5 1 6 f4 li:Jb7 1 7 f5 e5 1 8 f4
li:Jc5 1 9 g4 .!:the8 20 i.f3 'itf8 2 1 g5 e4
S ys t e m s with . . . i. x f3
22 i.g2 li:Jh5
2 J f6 g6 24 i.f1 'itg8 25 'i'e3 'i'c8 26
i.c 1 li:Jd3 27 i.xd3 exd3 28 'i'xd3 'iig 4 +
29 �h 1 li:Jxf4 30 'ii'g 3 .!:txe 1 + 3 1 J:rxe 1
'i!ixg3 32 hxg3 li:Jd3 33 1If1 li:Je5 34 i.f4
li:Jxc4 35 .!:te1 a5 36 .!:te7 .l:tc8 Y. - Y.
Exciting stuff, though I'm sure Black is
doing fine (particularly after 1 1 . ..0-0) .
Game 77
Seui-Kengis
Bonn 1995
1 c4 b6 2 li:Jc3 e6 3 li:Jf3 i.b 7 4 g3 i.xf3
5 exf3 c5 6 d4 cxd4 7 'ilfxd4 li:Jc6 8 'i1Vd3
l:tc8
9 f4!
The best move, preventing . . . tZ:leS. 9 b3?
tZ:leS 10 'i!Ve3 'i!Vf6 1 1 f4 tZ:lg4 12 'i!Vf3 i.cS
13 i.b2 tZ:lxf2 14 b4 .i.d4 favoured Black
137
En g lish D e fe n c e
in Hagenbach-Bandza, Hessenliga 1994.
9 . . . .i.b4 1 0 i.d2 liJf6
If 10 ... t2:ia5 then 1 1 b3 holds White's
position firm, e.g. 1 1 . . .d5? 12 cxd5 'i¥f6 13
'ii'b 5+.
1 1 .i.g2 .!Lla5 1 2 b3 0-0 1 3 0-0
1 3 . . .d 5
Perhaps 13 . . .'iie7!?
1 4 cxd5 .i.xc3 1 5 i.xc3 liJxdS 1 6 i.xa5
16 .lafcl!? t2:ixc3 (maybe 16 ...t2:ic6) 17
�xc3 �xc3 18 'i!Vxc3 is better for White
due to the poor position of the knight on
aS.
1 6 . . . bxa5
1 7 l:!.fd 1 liJc3 1 8 'ifxd8 l:!.fxd8 1 9 J:t xd8 +
l:txd8 20 �f1 �8 2 1 �e 1 rJJe 7 22 .i.f1
a4 23 bxa4 liJxa4 24 .!:!c1 .!Llb6 25 .l:!.c3
liJd5 � - �
White must take care when he plays 8
'iid3 as the queen is a little exposed. How-
1 38
ever, compared to the usual 8 'ii'd2, it does
stay out of the way of the bishop on cl so
development is smoother. On the basis of
this game, Black has some work to do
before he can equalise.
In the final two games of this book
White invites a transposition to a Queen's
Indian Defence instead of entering the
usual fun and mayhem of the English.
Here Black may also use ....ltxf3 (at the
right moment!) to keep the game off the
straight and narrow.
Game 78
Yermolinsky-Speelman
Hastings 1996
1 d4 e6 2 c4 b6 3 g3 i.b7 4 lLlf3
4 . . . i.b4 +
The immediate 4 ... i.xf3 5 exf3, without
the exchange of dark-squared bishops, is
not as strong. For instance, 5 . . . d5 6 t2:ic3 c6
7 cxd5 exd5 8 i.d3 i.e7 9 0-0 .ltf6 10 .ltc2
t2:ie7 1 1 'ifd3 t2:id7 12 l:te1 a6 13 i.f4, when
although Black has an extra pawn on the
queenside, without the light-squared
bishop his structure is weak, and White's
pieces are very active, as in Andersson­
Ljubojevic, Monaco blindfold 1997.
5 .i.d2
The best block. The alternatives are:
a) Amazingly, 5 t2:ibd2?! already lands
White in difficulties. After 5 ... i.xf3 6 exf3

there are two decent moves:
a1) 6 ... tLlc6!? 7 a3 .i.xd2+ 8 .i.xd2 (8
'iixd2 'ii'f6 is just good for Black) 8 ... tLlxd4
9 Sl.c3 c5 (or 9 . . . e5 10 f4 with counterplay)
10 .i.xd4 cxd4 1 1 'ifxd4 'iff6 12 .l:td1 'fi'xd4
13 l:.xd4 .l:tc8 14 b3 tLle7 and Black is
slightly better.
a2) 6 . . .'i!Vf6 7 .i.g2 'i!Vxd4 8 f4 tLlc6 9 0-0
Sl.xd2 10 Sl.xd2 'ii'xb2 and Black was al­
ready well on top in Kraidman­
p .Littlewood, Lloyds Bank open, London
1978.
b) 5 tLlc3 transposes to a kind of
Nimzo-Indian Defence, although as Black
has delayed the development of the knight
to f6 he has more options so, in theory, he
should have few difficulties. Here are
some ideas after 5 . . . .i.xc3+ 6 bxc3:
b 1) 6 . . ..:t:J f6 7 Sl.g2 d6 8 0-0 tLlbd7 9 a4 a5
10 tLld2 .txg2 1 1 'iitxg2 eS 12 e4 'i!Ve7 13
'i!Ve2 0-0 14 .i.a3 l:tab8 15 f4 was better for
White in Gyimesi-Bricard, Paris open
1995. Black's development was too rou­
tine.
b2) 6 . . . tLlc6!? 7 Sl.g2 (7 e4 tLlge7)
7 . . . tLla5.
b3) 6 .. .f5!? 7 .i.g2 tLlf6 8 0-0 0-0.
5 . . . i.xf3 !
Only now.
6 exf3 i.xd2+ 7 lbxd2
For 7 'it'xd2 see the next game. It makes
a massive difference to the position for
Black to have exchanged off a pair of bish-
S ys t e m s with . . . Ji. x f3
ops. It means, for example, that he may
develop his queen safely on a dark square
without fear of being hassled.
7 . . . lbf6
7... tLle7 also has a good reputation. For
example, 8 f4 tLlbc6 9 i.g2 tLlxd4 (an en­
terprising sacrifice; Black gets a pawn for
the exchange, and has excellent control
due to the centre pawns) 10 Sl.xa8 'i¥xa8 1 1
0-0 c5 12 1:i.cl 0-0 1 3 l:.c3 l:.d8 1 4 tLlf3 tLlec6
with good compensation in Grabarczyk­
Teske, OLO-B 1998.
8 f4
Or 8 .tg2 0-0 9 f4 dS 10 0-0 c6, as in Re­
lange-Degraeve, Cappelle la Grande open
1984. This is the most usual kind of struc­
ture that is reached when Black goes in for
this line. Having got rid of the light­
squared bishop, Black sticks his pawns on
light squares to block out his opponent's
bishop. Chances are about equal.
In the main game, Speelman does not
go for this structure immediately, but
maintains his flexibility and in so doing
blocks out his opponent's minor pieces:
8 . . . lbc6 9 lbf3 lbe7
To counter the threat of d4-d5, break­
ing open the position.
1 0 i.g2 0-0 1 1 0-0 c6 1 2 1lc1 1lb8 1 3
lbe5 l:tc8 1 4 'ii'e 2 d6 1 5 lbf3 'Wic7 1 6
.l1.h3 .l:!.fe8 1 7 llc3 11cd8 1 8 l:tfc 1 g 6 1 9
'ifc2 c5 20 �d 1 lbc6 2 1 .l:.e3 d5 22 dxc5
bxc5 23 cxd5 lbxd5 24 litee 1 lbcb4 25
1 39
En g lish D e fe n c e
'ilic4 ll:lb6 2 6 'ifb3 a S 2 7 ll:le5 a4 2 8 'i't3
ll:lc2 29 ltxd8 ltxd8 30 ltc1 li::ld 4 3 1 'i'e4
li::l d 5
It is interesting to see how Black has
managed to find superb squares for the
knights.
32 i.. g 2 'i'b6 33 ll:lc4 'i'a6 34 h3 li::l b 6 35
..lit1 ll:lxc4 36 :xc4 'ifa7 37 i.. g 2 'ifb6 38
.!:lxa4 'i'xb2 39 h4?
39 I!.c4 should still maintain the bal­
ance, but it is tricky with that knight on
d4.
39 . . . ll:le2+ 0 - 1
40 �h2 is met by 40 ... tt:Jc3.
Game 79
Kempinski-Miles
Groningen 1996
1 d4 e6 2 c4 b6 3 li::lf3 i.. b 7 4 g3 i.. b4 +
5 i.. d 2 i.. xd2 + 6 'i'xd2 i..xf3 7 exf3 d 5
Miles makes the logical move, playing
the pawn to d5 straightaway, but it is also
possible to wait for a move or two with
7. . . tLle7 or 7 . . . tt:Jf6. However, if you do,
you should be aware that White may try
d4-d5 to cut across your plans. For in­
stance:
a) 7 . . . tt:Je7 8 d5!? 0-0 9 tLlc3 a5 (not
9 . . . c5? ! 10 i..h 3 exdS 1 1 cxd5 d6 12 0-0
tLld7 13 .l:He1 a6 14 a4 with advantage to
White in Rohde-Kengis, Tilburg 1992) 10
i..h 3 tLla6 1 1 0-0 tLlc5 12 l:tad1 tLlg6 13 f4
140
e5 Y2-Y2 Sloth-Hoi, Danish Championship
1983. Just when things were hotting up!
White should play 14 f5 and now:
a1) 14 . . . tLle7 15 d6 (or 15 f6!?) 15 ... cxd6
16 �xd6.
a2) 14 ... tLlh8 15 d6 c6 16 f4.
In both cases with a promising posi­
tion.
b) 7 ...tLlf6 is the better of the two
moves as d4-d5 then has some drawbacks:
b1) 8 d5?! exd5 9 cxd5 0-0 10 tLlc3 �e8+
1 1 iLe2 tLla6!? (or 1 l . ..c6!?) 12 0-0 tLlc5
allows Black good counterplay.
b2) 8 tLlc3 0-0 9 iLg2 d5 10 f4 c6 1 1 b3
tLla6 12 0-0 'iid7 13 l:tad1 tLlc7 14 l:tfe1
l:Ifd8 15 'ti'e2 Y2-Y2 Ljubojevic-Speelman,
PCA qualifier, New York 1995.
8 i.. g 2
8 cxd5 would be a mistake due to
8 ... 'it'xd5 9 i.. g2 tLlc6 10 0-0 (or 10 tLlc3
Vi'xd4 1 1 f4 'iixd2+ 12 �xd2 0-0-0 + with
an extra pawn) 10 ... tLlxd4 1 1 'iid3 !id8 and
Black was a pawn up in Bock-Czebe, . Bu­
dapest 1997.
8 . . . ll:le7
8 ... c6 is rather similar after 9 b3 tLle7 10
tLla3 0-0 11 0-0 'iid6 12 tLlc2 tLld7 13 tLle3
a5 14 f4 g6 and Black's position was super­
solid in Stanec-Loebler, Austrian Champi­
onship 1995.
9 0-0 0-0 1 0 cxd5 li::lx d5
see following diagram

·In general it is desirable to recapture on
dS with the knight so as to have a go at
the pawn on d4 later on.
1 1 lLle3 lLle6!
It appears dangerous to put another
piece on the same diagonal as the bishop,
but Miles had calculated that by putting
pressure on the d-pawn he can force a
concessiOn.
1 2 lLlxd5 'ii'x d5 1 3 f4 'ii'd 6 1 4 J:!.td 1 �adS
1 5 i.. x c6 'i/Hxc6
S ys t e m s with . . . i.. x f3
The position is completely equal.
1 6 .l:.ac 1 'ii'd 6 1 7 'ilie2 J:!.d7 1 8 'ifc6 'Wie7
1 9 J:!.c4 .:.td8 20 J::.d c1 g6 21 'ii'a 4 a5 22
�c2 l:!.xd4 23 J:!.xc7 'ifid6 24 �e8 l:!.d2 25
l:!.xd8+ 'i'xd8 26 �ea �xe8 27 J:!.xe8+
'iit g 7 28 b3 b5 29 �c5 lld5 30 l:!.e7 h 5
3 1 a3 l:!.d3 Y. - Y.
If Black doesn't want to allow it, then
White cannot easily reach a Queen's In­
dian. Black's system, as played in the last
two games, is fully viable.
141
En g lis h D e fe n c e

S u mmary
5 ... tZ:lc6 from Game 71 is fun, but objectively a bit dodgy; you need to pick your oppo­
nent well. If Black captures on d4 then he should definitely play with ...e7-e6 (Games 73-
77) rather than fianchetto the bishop (Game 72) . I see nothing wrong with Black's sys­
tem in Games 78-79, so long as the dark-squared bishops are exchanged - but watch out
for an early d4-d5 from White.

1 c4 b6

2 l'Llc3
2 tZ:lf3 .i.b7 3 g3 .ixf3 4 gxf3 c5 5 d4 (D)
5 ... tZ:lc6 - Game 71
5 ... cxd4 6 iixd4 tZ:lc6 7 iVd2
7 ... g6 - Game 72
7 . . . e6
8 .i.g2 - Game 73
8 tZ:lc3 - Games 74-77 below (by transposition)
2 d4 e6 3 tZ:lf3 ii.b7 4 g3 .i.b4+ 5 .i.d2 .i.xf3 6 exf3 (D) .i.xd2+
7 tZ:lxd2 - Game 78
7 iixd2 Game 79
-

2 . . . .1i.b7 3 l'Llf3 e6 4 g3 .ixf3 5 gxf3 c5 6 d4 cxd4 7 'iixd4 l'Llc6 (D) 8 'i'd2

8 iidl - Game 76
8 'ii'd3 - Game 77
8 . . . �f6
8 .. J::i.c 8 - Game 75
9 .ig2 - Game 74

5 d4 6 exf3 7 lL\ c6
. . .

142
 INDEX OF COMPLETE GAMES I

Agrest-Zviaginsev, Kazan 1997 ................................................................................. 126

Ahundov-Bagirov, Yerevan 1996 ................................................................................ 26
Andreasen-Pedersen, Danish Championship 1988 ................................................... 110
Appolonov-Lempert, Katowice 1992........ ; ................................................................ 136
Arlandi-Yefimov, Asti 1995 .......................................................................................... 18
Babu-Miles, Sakthi 1996 68
................... ..............................................................................

Barkhagen-Kengis, Gausda/ 1991 ................................................................................ 15

Beliavsky-Gulko, Polanica Zdroj 1996 ........................................................................ 48
Beliavsky-Short, Groningen 1997 ................................................................................ 29
Browne-Miles, Tilburg 1978.......................................................................................... 36
Burger-Ehlvest, St Martin open 1993 ........................................................................... 59
Cramling.P-Gulko, Pamplona 1996 .......................................................................... 1 00
Dgebuadze-Bagirov, Linares 1997............................................................................... 50
Donchenko-Muhutdinov, Alushta 1993 .................................................................. 132
Dreev-Shabalov, Elista Olympiad 1998 ....................................................................... 33
Fedorowicz-Flear, Wijk aan lee 1988 ....................................................................... 123
Franco-Teske, Havana 1998.......................................................................................... 22
Gamota-Karasev, Moscow 1996 .................................................................................... 57
Garces-Keene, Lausanne 1977....................................................................................... 60
Gartner-Dey, Berlin 1994 .............................................................................................. 53
Gelfand-Short, Novgorod 1997..................................................................................... 20
Griinfeld-Prie, Paris 1990 ............................................................................................. 56
Gurevich.M-Kengis, Bad Godesburg 1996 ................................................................ 103
Heinbuch-Spassky, German Bundesliga 1984........................................................... 119
Hellsten-Miles, Malmo 1996 ......................................................................................... 93
Hjartarson-Gretarsson, Reykjavik 1995 ................................................................... 104
Hodgson-Bischoff, Linares 1996 .................................................................................. 78
Horvath.Jo-Gulko, Nova Gorica 1997........................................................................ 41
lvanchuk-Sadler, Monaco (blindfold} 1998 .................................................................. 87
Joyce-Speelman, Bunratty Masters 1998 ................ ; ...................................................... 45
Kempinski-Miles, Groningen 1996 . . . . .
. ....... .. . . ..
.. . . . .. . . . . 140
.............. ... ..... ........ .. ..... ... .... . .

143
In d e x o f C o mp le t e G a m e s

King-Plaskett, London 1991 ........................................................................................ 130

Kislova-Minasian, Omsk 1996.................................................................................... 1 06
Korchnoi-King, Oviedo rapidplay 1992..................................................................... 116
Korchnoi-Short, Groningen 1997 .............................................................................. 113
Kostin-Grabuzova, Moscow 1996................................................................................. 51
Krylov-Zak, Groningen 1994 ...................................................................................... 122
Kryzius-Piesina, Radviliskis 1995 ................................................................................ 89
l.aqua-Cording, OLNN 1997 ..................................................................................... 120
Levitt-Short, Calcutta 1998 ........................................................................................... 86
Lobron-Speelman, German Bundesliga 1997.............................................................. 23
Magerramov-Ehlvest, Moscow 1992 ............................................................................ 12
Magerramov-Psakhis, Riga 1980 ................................................................................. 38
Marchand-Gulko, Geneva 1997 ................................................................................... 52
Mednis-King, Stavanger 1989 ....................................................................................... 13
Milov-Miles, Biel 1996 95
...................................................................................................

Milov-Miles, New York 1997 ........................................................................................ 97

Morovic-Speelman, Cala Galdana 1994 . .......... . . .. . . .
.......................... ... ......} 09
... ... ........

Moskow-Shabalov, New York 1993 ........................................................................... 133

Nogueiras-Velez, Cienfuegos 1983 ............................................................................... 42
Parker-Sher, Copenhagen 1996 ..................................................................................... 85
Petursson-Wauthier, San Bernardino open 1991 ....................................................... 43
Piesina-Kengis, Riga Zonal 1995 ................................................................................ 135
Piket-Speelman, Andorra Zonal 1998 ......................................................................... 74
Polgar.Zsu-Speelman, Dutch League 1993 .................................................................. 61
Polovodin-Miles, Los Angeles 1991 ........................................................................ :..... 1 7
Pytel-Piasetski, Buenos Aires 1978 ................................................................................ 70
Rahman-Speelman, Calcutta 1998............................................................................... 62
Rayner-Plaskett, London (Ll(J)Ids Bank) 1993 ........................................................... 125
Razuvaev-Barle, Maribor 1996 ................................................................................... 114
Remlinger-Kengis, Gausdal 1991 ................................................................................ 84
Remlinger-Rogers, Philadelphia 1986 ......................................................................... 81
Reuben-Basman, London 1982 ..................................................................................... 40
Ruzele-Kengis, Bad Godesburg 1996............................................................................ 10
Sadler-Kengis, Koge 1997 .............................................................................................. 71
Salov-Short, Madrid 1997.............................................................................................. 98
Sashikiran-Speelman, British Ch., Torquay 1998 ....................................................... 76
Seirawan-Schussler, Malmo 1979 ................................................................................. 34
Serper-Hodgson, Groningen 1993 ............................................................................... 64
Seul-Kengis, Bonn 1995 ............................................................................................... 137
Shirov-Prie, Val Maubuee 1990 ..................................................................................... 45
Smejkal-Kengis, Prague 1993 ...................................................................................... 117
Suba-Plaskett, London 1991 ........................................................................................ 108
Toth-Ornstein, Oslo 1978 ............................................................................................. 39
Urban-Maciejewski, Lubniewice 1993 ......................................................................... 88
Van Wely-Zviaginsev, New York open 1997 ............................................................ 134
Vegh-Zlatilov, St Augustin 1990................................................................................... 39
Yermolinsky-Speelman, Hastings 1996..................................................................... 138
Yrjola-Kengis, Yerevan 1996......................................................................................... 25

1 44