You are on page 1of 162

alekhinels

defence

by Nigel Davies

EVERYMAN CHESS
Everyman Publishers pic www.everyman.uk.com
First published in 2001 by Everyman Publishers pic, formerly Cadogan Books pic,
Gloucester Mansions, 140A Shaftesbury Avenue, London WC2H 8HD

Copyright 2001 Nigel Davies

The right of Nigel Davies to be identified as the author of this work has been as
serted in accordance with the Copyrights, Designs and Patents Act 1988.

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

British Library Cataloguing-in-Publication Data


A catalogue record for this book is available from the British Library.

ISBN 1 85744 253 9

Distributed in North America by The Globe Pequot Press, P.O Box 480,
246 Goose Lane, G uilford, CT 06437-0480.

All other sales enquiries should be directed to Everyman Chess, Gloucester Man
sions, 140A Shaftesbury Avenue, London WC2H SHD
tel: 020 7539 7600 fax: 020 7379 4060
email: dan@everyman.uk.com
website: www .everyman.uk.com

To Louise

EVERYMAN CHESS SERIES (formerly Cadogan Chess)


Chief advisor: Garry Kasparov
Commissioning editor: Byron Jacobs

Typeset and edited by First Rank Publishing, Brighton.


Production by Book Production Services.
Printed and bound in Great Britain by The Cromwell Press Ltd., Trowbridge, Wilt
shire.
CONTENTS I

Bibliography 4
Introduction 5

1 Classical with 4 ....i.g4 7


2 Classical Kengis System: 4 ... dxe5 5 ttJxe5 g6 25
3 Classical with 4 ... dxeS 5 ttJxe5: 5...ltJd7 and 5...c6 42
4 Classical Divergences: 4 ... c6, 4...ll'lc6 and 4... g6 58
5 Exchange Variation with 5...cxd6 77
6 Exchange Variation with 5 ...exd6 93

7 Four Pawns Attack: Main Line 105


8 Four Pawns Attack: Divergences 1 17
9 The Chase Variation 132
10 Other Lines 142

Index of Complete Games 1 57


BIBLIOGRAPHY I

Books

107 Great Chess Battles, Alekhine, Winter {Dover 1980)


Alekhine's Defence, Eales & Williams, (Batsford 1973)
Alekhine'sDefence: 'JheFourPawnsAttack,Christiansen,Raingruper&Joseph(fhink
ers' Press, Davenpon 1988)
Aljechin- Verteidigung, Bagirov (Schachverlag Rudi Schmaus, Heidelberg 1979)
Developments in theAlekhine Defence, Dunworth (Square One Publications 1987)
Encyclopaedia of Chess Openings Volume B, Jrd Edition (Sahovski Informator 1997)
Fischer-Spassky: The Chess Match of the Century, Gligoric {Fontana)
Learn from the Grandmasters, Keene et al (Batsford 1975)
New Ideas in the Alekhine Defence, Burgess (Batsford 1996)
The Alekhine for the Tournament Player, Alburt & Schiller (Batsford 1985)
The Complete Alekhine, Burgess (Batsford 1992)
Trends: Alekhine Defence, McDonald (Trends 1990)

Periodicals

Chesspublishing.com
Jnformators 1-79
The Week in Chess 1-355
INTRODUCTION
I

Reykjavik 1972, Game 13. The Soviet Whilst the Alekhine continued to en
World Champion, Boris Spassky, joy its dubious reputation, strong
opened the game with 1 e4. He was Grandmasters such as Larsen,
losing the match by a 7-5 scoreline but Korchnoi, Bagirov, Timman, Jansa and
had won brilliantly against Fischer's Alburt kept winning with it. But how
favourite Sicilian Najdorf in Game 1 1. could they keep winning with this
And the analytical spotlight ofthe entire dodgy defence? And why were they
Soviet chess machine was now focused even playing it in the first place?
on Fischer's favourite openings... Some years later, not much has
What could Fischer do? changed. The Alekhine's reputation is
He totally wrong-footed Spassky by probably even worse, and now it's the
playing l...lbf6, the so-called Alekhine likes of Miles, Morozevich, De Firmian,
Defence. By the 12th move White was Shabalov, Agdestein, Kengis and Ba
already in a critical position and after a burin who are playing this defence and
tremendous struggle Fischer won. winning. Don't they know any better?
The Fischer-Spassky match was the Eventually there came a point at
event that fanned the flames of my in which I realised that the wisdom of the
terest in chess. I was 12 years old and frog pond was at odds with reality. It
wanted to play like Bobby Fischer, but was nothing but talk, which people re
when I asked more experienced players peated without ever looking at the posi
about this funny knight move, they in tions for themselves. When I finally
variably advised me against playing it. started to check things out I discovered
Their wisdom was confirmed when that the Alekhine is a tough, fighting
most of the books seemed to agree. defence which creates unbalanced posi
This Alekhine Defence was a dubious tions from the outset. And if anything
opening against which White could eas its dodgy reputation was actually quite
ily secure a large advantage. helpful; a lot of people were overconfi-

5
A lekhin e 's De fen c e

dent and sloppy when playing against it. taken the position that most Whites
As far as results are concerned, in should probably want to play either the
practice the Alekhine scores just shy of Exchange Variation or 4t:bf3; all the
50% on my database, which is better other variations are covered mainly
than the French, Caro-Kann, Pirc, from Black's point of view.
Scandinavian and l . eS. Only the Sicil
.. If you are using this book to learn the
ian is marginally ahead, but that in Alekhine, here's how I suggest you go
volves being booked up to the gills and about it:
treading the same paths of so many
other players. From a stylistic point of 1) Play through the games very
view the Sicilian is not for everyone; quickly and ignore the notes and sub
play ohen proceeds at an unremittingly variations. This is to give you a basic
fast pace in which Black's king ohen familiarity with the lines and positions.
comes into the firing line. The Alekhine, 2) Play a few quick games with the
on the other hand, tends to take less Alekhine, either against your computer,
forceful paths in which the greatest on the internet or at your local chess
danger for Black is that of falling into a club, making a brief note of any points
passive position. of interest.
Is the Alekhine the opening for you? 3) Look up the lines which occurred
There is certainly scope within this de in these games and find out exactly
fence for many different styles of what you should have done.
player, though it seems to me that the 4) Repeat steps 2 and 3 until you get
ideal Alekhine player should have indi a 'feel' for the defence.
viduality, flexibility, good nerves, fight 5) Start playing it in selected serious
ing spirit and tactical ability. In my games, again using the book as a source
opinion this is not a defence for those of reference.
with either a methodical, systematic way
of thinking or those with very conserva Readers who are already playing the
tive tastes. Alekhine should also find something of
In writing this book I have not tried interest. First of all, I made a decision
to produce either a complete or a schol not to hold anything back, so the notes
arly work. The emphasis has been on contain a number of new ideas and
getting someone up and running if they points at which I disagree with other
want to play the Alekhine as either authors. Secondly, I have tried to point
Black or White, and I have therefore out the lines which I believe are the
concentrated on what I regard the most most logical and promising rather than
promising lines for players of either just list all the possible variations and
colour. Accordingly I have devoted rela adorn them with selected soulless sym
tively little space to the traditional 4 bols.
t:Df3 g4 or even Alburt's 4t:bf3 g6,
concentrating instead on the modern Nigel Davies
treatments with 4 ... dxe5. I have also Southport, October 2001

6
I CHAPTER ONE I
Classical with 4 ... lL.g4

1 e4 lbf6 2 e5 lbd5 3 d4 d6 4 lbf3 Game 7 shows White using plan 'b'


.i.g4 in its most direct form, arguing that
The Classical Variation is generally Black's bishop has gone to g4 prema
recognised as the main line of the turely. Usually White will not take such
Alekhine and probably the hardest thing direct action and first play 5 i.e2. Black
for Black to fight against. then has a choice between Flohr's 5 ...c6
By playing the modest 4lL'lf3, White and the standard 5 ... e6.
sets about developing his kingside and Flohr's move actually contains a mi
puts the onus on Black to develop some nor positional threat of ...i.xf3 followed
counterplay. Depending on Black's re by ... dxeS, which brings about a posi
action he can either try to maintain the tion in which the eS-pawn needs con
cramping pawn on eS or capture on d6 stant surveillance and White's bishop
to reach a kind of Exchange Variation. pair has little scope. In Game 6 we see
Pinning the knight on f3 with 4....i.g4 White's various strategies for combating
certainly looks like the most natural way this plan, which include switching to a
to intensify pressure against eS. For kind of Exchange Variation.
many years it was regarded as the way to Against S ... e6 White has to make a
play against the Classical and as such a major decision about whether or not to
large body of theory has developed interpolate the moves 6 h3 i.h5. The
around it. pros and cons of this depend on which
Essentially White has two different plan he intends to adopt later.
plans; either to try and maintain the The plans which require 6 h3 are Be
cramping pawn on eS or effect a timely logan's treatment {Game 4), the 1 1 cxd5
liquidation in the centre with exd6. line, given as a note to White's 1 1th
Black tries to force the liquidation move in Game 2 and Geller's treatment
whilst trying to make sure that his of 12 i.xf3, given as a note within the
pieces are well placed when it happens. same game. Having 6 h3 i.hS included

7
A lekhin e 's Defen c e

is also useful for White if Black opts for with 12 ...f5 but this has the drawback of
an early ... fi)c6 (as in Game 5); if, after being very passive. A good example of
d4-d5 plus multiple exchanges, White's how to play White was given by Kiril
queen lands on d5, it will gain a tempo Georgiev in his game against Popov
by hitting Black's bishop on h5. from the 1986 Bulgarian Championship:
On the other hand, 6 h3 is definitely 13 h 1 g6 14 b4 lilc6 15 l:tg1 h8 16
not required in the 'main line', in which :.g2 l.tg8 17 a3 i..h4 18 .i.f3 ltl8e7 19
White recaptures on f3 with a pawn. A "it'd2 a6 20 liJe2 llg7 21 ltlct ltlg8 22
comparison of Games 1 and 2 shows liJd3 and White will gradually advance
that h3 is often a useful square for his queenside pawns.
White when he tries to attack Black's 1 3 .i.d3 g6 14 f5!?
king. Blocking it with a pawn reduces If White tries for a slower build-up in
his options. this position, Black can at least bring a
.------. knight to f5 and might one day be able
Gamel to lever open the f-file with ... f7-f6.
Kobalija-Nalbandian Nevertheless, White must be better, one
Geller Memorial, Moscow 1999 example being 14 h 1 fi)e7 15 l:g1
.___________..
h8 16 1i'f3 fi)bc6 17 a3 fi)f5 18 fi)e2
1 e4 IC!f6 2 e5 IC!d5 3 d4 d6 4 IC!f3 b6 19 l:ac1 with an edge for White in
.i.g4 5 .i.e2 e6 6 0-0 .i.e7 7 c41Clb6 Kruszynski-Gruen, Hamburg 1984 .
8 IC!c3 0-0 9 .i.e3 d5?! 10 c5 .ixf3 14... exf5
1 1 gxf3 After 14 ... gxf5 15 'i'h5 White will
As this game was played in the have a very dangerous attack.
'Geller Memorial', it would have been 1 5 'iVf3 c6 16 >h1 >h8?
more fitting to use Geller's own treat In such a critical position Black must
ment with 1 1 i..xf3. For a discussion of play with the utmost accuracy. He
this plan, see the note to White's 12th should probably take the bull by the
move in Game 2. horns and undermine White's e5-pawn
with 16 .. .f6!?, after which Finkel gave
the line 17 e6 .i.g5 18 l:ae1 .:es 19
.i.xf5!? gxf5 20 i..xg5!? fxg5 21 'i'xf5
lif8 22 'il'g4 "it'f6 23 f4 'i'xd4 24 'i'xg5+
Wh8 25 f5 when it's anybody's guess
what's happening.
1 7 J:g1 IC!a6?1
Black seems to be unaware of the
imminent danger to his king. He had to
try 17 .. .f6!?.
18 .i.xf511Cle7
Accepting the sacrifice with 18 ...gxf5
1 1 /ClcS 1 2 f4 .i.h4
. . loses immediately to 19 i..h6 l:g8 20
Black has also closed the position Axg8+ 'i'xg8 21 l:g1 etc.

8
Classical with 4 ... .i.g4

19 .i.c2lllg8 road.
Or 19 .. .f5 20 'iih 3tbg8 21 lLle2tbc7 29 l:lh4+ ! gxh4 30 l'Oe6+ h5 31
22 lLlf4 with a sacrifice on g6 coming .i.d 1 mate ( 1-0)
up. When White plays these gxf3 lines
he often makes use of the h3-square. Game 2
This is one argument in favour of omit Aseev-Bagirov
ting the moves h2-h3 and ... i.h5. Berlin 1990

1 e4 lllf6 2 e5 ltld5 3 d4 d6 4 lllf3


i.g4 5 .i.e2 e6 6 0-0 .i.e7 7 c4 ltlb6
8 h3 .i.h5 9 ltlc3 0-0 10 i.e3 d5
Black cannot keep the tension too
long as 10 ...lLlc6 is met by 11 exd6 cxd6
12 d5 exdS 1 3 lLlxdS lLlxdS 14 'iixdS,
with the better game for White because
of the weakness on d6.
1 1 c5
White has another interesting plan in
11 cxd5 and now:
20 1i'h3 lt:lc 7 21 l:lg4 .i.e7 22 l:lag 1 a) After 1 1...lLlxd5 White can play 12
1i'd7 23 f4! ll'b3 lbb6 13 dS!? (131lfd1 is also pos
The advance of the f-pawn increases sible) 1 3 ...exd5 (13 ...lLlxd5 is met by 1 4
the strength of White's attack to deci l:.fd1 c6 1 5 ll'xb7) 1 4 .ixb6 axb6 15
sive proportions. lLlxdS .ic5 16 l:lad1 'iic8 17 a3lLlc6 18
23 l:lae8 24 f5 .i.d8 25llle2 b6 26
.. 1i'c3 l:le8 19 l:tfe 1 with the better game
lt:lf4 g5 for White according to Kremenietsky.
b) 1 1 ...exd5 12 g4!? (12lLlet .ixe2 13
9 xe2 tbc6 14 lLld3 'i'd7 15 llad1 lL\d8
16lt:\c5 9c8 17 f4 fSleft Black with an
ultra-solid game in Plachetka-Bagirov,
Kirovakan 1978) 12 ...i.g6 13 lLlel f6
(1 3 .. .f5 14lLld3 c6 15lLlf4 'ifd7 16 'it>h2
gave Black serious problems on the g
file in Ornstein-Alburt, Reykjavik 1984;
if Black wants to play .. .fS he should
wait until White plays f2-f4) 14 f4 fxe5!?
(Setting up a blockade with 14 .. .f5
would be more solid, especially since
27 1i'xh7+ ! ! xh7 28 f6+ h6 White can no longer put a knight on f4)
Or 28... h8 29 %lxg5li:lli6 (29... .UxeS 15 dxe5 c5 16 f5 .ie8 17 i.f4 .ic6 18
only delays the inevitable after 30 dxe5 i.f3 d4 19 lLle4 lLld5 20 lLlg2 lLld7 21
lLlxf6 3 1 exf6) 30 l:th5 is the end of the i.g3 and both sides have chances,

9
A lekhin e 's De fenc e

King-Baburin, British League 2000. 15 ... b6 (15....i.g5 16 2 .i.xf4 17llW4


1 1 ....txf3 1 2 gxf3 e7 18 .i.g4 ac6 19 1i'd2 was mar
This has been the most popular way ginally better for White in Hiibner
to play it, preventing Black's knight Hort, Biel 1984 whilst 15.. .'i'd7 16 i.e3
from coming to c4 . But Efim Geller f6 17 exf6 .i.xf6 18 'i'd2 b6 19 a4
considered the pawn recapture rather e7 20 .i.e2 li)fS was pretty much
dubious, instead preferring his patented equal in Geller-Bagirov, Tbilisi 1978) 16
12 .i.xf3!? c4 13 .i.f4 a4 b5?! (according to GeJier Black
should have played either 16 ... .i.g5 or
16 .. . 6) 17 c3 b4 18 li)e2 and White
was clearly better in Geller-Bronstein,
Petropolis Interzonal 1973.
1 2 . .. c8
The knight hopes to come to f5 via
e7.
13 f4
This has been played almost auto
matically but in McShane-Baburin,
Cambridge 1995 White took some
space on the queenside with 13 b4!?,
and now: after which 13 ...c6 14 'i'a4 .i.h4 15
a) 13 ... b6 14 b3 aS 15 l:.cl (15 i.d3 f6 16 f4 6e7 17 e2 f5?! (a
a4!? d7 16 1i'd2 c6 was played in preliminary 17...fxe5 is better) 18 .i.xf5
Ciocaltea-Williams, Nice Olympiad exf5 19 'i'c2 7 20 h2 1i'd7 21 l:lg1
1974 and now 17 .i.g4!? was given as gave him pressure on the g-file and a
White's best by Ciocaltea; 1 5 .d2!? is passed pawn on e5.
also interesting) 15 ... bxc5 16 dxc5 c6
17 lle1 .i.g5 18 xdS!? exd5 19 .i.xg5
xg5 20 .i.xd5 (Schmid also mentioned
20 'lrxd5!? l:d8 2 1 'lrc4 l:ld4 22 1i'c3,
assessing the position as 'unclear'
20... h8 was a game Kavalek-Schmid,
Nice Olympiad 1974, in which White
should now have played 21 11fe2 aS 22
1i'e4 l:la6 23 f4 with two pawns for the
piece and badly placed black pieces.
b) 1 3 ...6 14 b3 4a5 15 :ct
(Geller played 15 11'd2 b6 16 llacl bxc5
17 dxc5 in his game against Hecht from 13 . . .c6
Budapest 1973 but during the subse With the moves h2-h3 and ...i.h5
quent game against Bronstein he be White can no longer use the h3-square
came concerned about 17...'i'b8!?) in some of his attacking lines. For this

70
Classical with 4 . . . .i.g4

reason Black has slightly more justifica Black can get an excellent position with
tion in playing 13 ...i.. h 4 here. It would Volzhin's suggestion of 17, ..l'jj8e7 18
then be pointless for White to play 14 l'jjxe7+ 'ifxe7! 19 .txc6 bxc6 20 11ff3
.id3 g6 15 f5?! exf5 16 11ff3 c6 17 Wh1 'ife6 {21 .txf4? 11ff5 wins a piece).
h8 18 l:tgl l'jje7. Instead White 1 7 ...J.xf4 18lDxf4lD8e7!
should play 15 11fg4 but Black is still And not 18...li)xd4 because of 19
doing okay after 15 .....t>h8 (Agzamov's .txb7 llb8 20 .i.g2! llxb2 21 'l'g4! with
suggestion of 15 ... h5 16 'i'f3 li)e7 a strong attack.
would also leave White regretting the 19 b3 l:tb8 20 J.e4?!
fact that he couldn't put his queen on
h3) 16 b4 /1)c6 17 :abl h5! 18 'iff3
li)8e7 19 b5 lba5 20 <li>h2 b6 (Diesen
Vaganian, Hastings 1974/75) and now
21 f5 missed the mark after 21.. .exf5 22
f4 {jjg8 23 :g1 4 24 e6 h7 25
exf7 l1xf7 etc.
The stodgy 13 .. .f5 would reduce
White's attacking chances but leave
Black without counterplay. A good ex
ample of how to play these positions
with White was given in the Georgiev
Popov note within the previous game. Overestimating his chances. He
14 f5 should play 20 li)e2 11fd7 21 i..g4 'i'dS
22 f4 (and not 22 'ifd2l'jjxe5) 22 .. .:.fd8
23 i..f3 'ird7 24 i.g4 1i'd5 with a draw
by repetition.
20 ...Wxd41
Black is quite right to want to ex
change queens as after 20...lt:lxd4? 21
1Wg4! White obtains a strong attack.
21 'i*xd4 xd4 22 J:.fd1 l:tfd8 23
J:ld3?
23 l'jjd5 was relatively best, though
Black is still better after 23, l'jje2+! 24
.

fl l'jjxd5 25 :xd5 llxd5 26 i.xd5


14 . . . exf5 1 5 J.f3 lCJf4! 27 .ig2 f8 because of his strong
If White can just recapture on d5 and knight on f4 and White's weak pawns.
then advance his central pawns he will 23 ...lDdc6 24 e6?
have a huge advantage. But Black is able And this is probably the losing move.
to stop this plan. He has to play 24 i..xc6 llxd3 25l'jjxd3
15 J.g5! 1 6 l.i:lxd5 f4! 17 J.xf4
with chances to make a draw.
17 i..c l has also been played but then 24. . .f5! 25 J.g2 l:txd3 26lDxd3 J:.d8

77
A lekhln e 's Defenc e

27 /0f4 g6!
Intending to bring his king to f6 from
where it puts pressure on the weak e6-
pawn.
28 l:le1 g7 29 .i.xc6

39 ...o!Od5 40 e7 xe7 41 f5 .!Of&!


42 4 86 43 g3 l0e4+ 44
xh3 o!Oxf2+ 45 h4 d5 46 g5
0-1
46 gS l2Jd3 47 h6l2Jct will mop
29 ... bxc6! up White's queenside pawns.
And not 29 ...l2Jxc6? which fails for a
moment to keep White's passed pawn Game 3
'under lock and key'. White could then Mainka-Yusupov
draw with 30 e7! l:Z.e8 31 l2Je6+ Wf7 32 German Ch., Bremen 1998
l2Jxc7 %be7 33 lb:e7+ xe7 34 f4
(Volzhin). 1 e4 .!Of& 2 e5 l0d5 3 d4 d6 4 o!Of3
30 g2 .i.g4 5 .i.e2 e6 6 0-0 .i.e7
Trying to prevent 30...g5 with 30 h4 It's also interesting for Black to play
loses to 30.. J:td4. 6 ... a6!? at this early stage. The game De
30 g5! 31 o!Oe2 6 32 h4 gxh41
.. Firmian-Baburin, Copenhagen 1996
33 o!Of4 l:ld4 34 3 l:le4 35 l:lxe4 continued 7 c4 l2Jb6 8 1Wb3 J.e7 9 lldl
fxe4+ 36 xe4 o!Od5 37 o!Oe2 llJSd7 10 J.f4 0-0 1 1 l2Jbd2 dxeS 12
The pawn ending is lost because of dxeS 1We8 with a cramped but playable
Black's distant passed pawns. The position for Black.
passed h-pawn is also the deciding fac 7 c4 o!Ob6 8 lDc3 0-0 9 .i.e3 a6
tor in the knight endgame. An interesting move which threatens
37 ... o!Oe7! 38 o!Od4 10... .ixf3 (after 1 1 i.xf3 l2Jxc4 12
If White tried to repeat the position .i.xb7 Black has the a7-square for his
with 38 l2Jf4 I'm sure that Bagirov rook) and can be useful against plans
would have varied this time with which involve an advance of White's
38...l2Jg6. queenside pawns.
38 h3! 39 l0f3
. One good reason for White insen the
Black is also winning after 39 f3 moves h2-h3 and ... .ihS is that Black
WeS. can now play 9 ...llJc6. The point is that

12
Classical with 4 . . . .t.g4

10 exd6 cxd6 1 1 d5 exd5 12 xdS after 10 ... dxe5 he should resist the
l0xd5 13 ...xd5 is not as good for temptation to play 1 1 cS?! (1 1 lbxeS
White as it might have been because he .i.xe2 12 xe2 li)Sd7 with equality is
is not hitting the bishop on hS. This better) 1 1...exd4 12 cxb6 cS 13 llfd1
allows Black to play 13 ....i.f6 (or maybe 1fe8 14 xd4 cxd4 15 .i.xd4 .i.xe2 16
13 ....i.e6). lbxe2 l0c6 with the better game for
Black in Rabiega-Jansa, Austrian Team
Ch. 1995.
c) 10 exd6 cxd6

10 d2
This allows Black to obtain a very
comfortable position. White's more
challenging options are as follows: and now:
a) 10 b3 and now: cl) The standard 1 1 b3 is fine for
at) 10 . ..Ii)8d7 1 1 h3 (11 llct l:tb8 1 2 Black after 1 1...d5 12 cS (or 12 h3 .i.bS
l:te 1 dxeS 13 xeS xeS 14 .i.xg4 13 eS .ixe2 14 1i'xe2 ll)c6 1 5 llfdl
xg4 15 1Wxg4 16 .i.h6 also gave i.f6 with equality as in Anand
White the freer game in Oratovsky Yusupov, Wijk aan Zee {7th match
Zilberman, Israeli Team Ch. 1996) game} 1994) 12 ...l0c8 13 b4 l0c6 14
l l. .. .i.fS 12 exd6 cxd6 13 l:tel !? h6 14 1i'b3 .i.f6 15 l:lfd1 8e7 16 l:ld2 li)fS
i.d3 i.xd3 15 1i'xd3 l:lc8 161:lad11:le8 as in Dolmatov-Morozevich, Moscow
17 .i.f4 and White had a slight space (rapid) 1995.
advantage in Chandler-Hort, Surakarta c2) 1 1 d5!? i.xf3 {1 1...e5 produces a
1982, though it is difficult to achieve position very much akin to Hamdouchi
anything because Black's structure is Baburin, but having the pawn on a6 is
flexible and sound. of doubtful value) 12 ..ixf3 lllxc4 1 3
a2) Alternatively Black can play dxe6 fxe6 14 ..tg4 1id7 15 1i'e2 lbe5 16
10...d5, after which 1 1 c5lb6d7 12 li)d2 ..th3l0bc6 17 f4 lilf7 18 ..tb6 fd8 19
.i.fS 1 3 f4 f6 1 4 li)f3 (14 g4!? is more dS and White had more than enough
aggressive) 14 ... fxe5 15 fxeS li)c6 16 for the pawn in Ulibin-Kengis, Pinsk
.i.d3 b6 gave him equality in Spasov 1986.
Ehlvest, Biel Interzonal 1993. 10 ..i.f5
.

b) 10 1ib3 is a reasonable move but Better than 10.....txe2?! 1 1 'ifxe2

13
A lekhin e 's D efen c e

dxeS 12 dxeS when White has a useful 2S :d2 l:lb4 26 a37!


space edge.
1 1 de4 dxeS 1 2 dxeS lbsd7 13 f4
f6

A slight mistake which creates weak


nesses in the queenside. The simplest
way to hold the position is with 26 b3
14 'ilb3!? followed by a transfer of the bishop to
This leads to great complications. c2 via dt.
White could also play 14 exf6ll\xf6 15 26 ...l:lb3 27 'iPf2
tlJcS with a fairly even game. And here 27 cS is a better try.
14...fxeS 1 S fxeS ..ixe4 27 ...cS 28 c3 ..igS 29 l:c2 c6
Simplifying the position. Black could 30 bS?
also play lS ...lL\xeS but then White has The losing move. White had to play
compensation for the pawn. 30 lL'Idl although admittedly Black has
1 6 xe4 xeS 1 7 llad1 l:txf1 + 1 8 some pressure after 30...ll\d4 31 l:tc3
..txf1 'ilea llb7 etc.
Once again aiming for a small but 30 ...d4 31 xd4 .i.e3+ ! 32 'iPf3
clear advantage rather than risk life and .i.xd4+ 33 e4 .i.xb2 34 a4 ..id4
limb in continuing complications. After 3S ..id3 l:a3 36 l:c1 l:lxa4 37 l:lb1
18 ...9c8 White can play 19 cS tbds 20 .J:r.a2 38 h4 a4 39 g4 a3 40 l:lb8+
llxdS!? exdS 21 'ifxdS+ lL\7 22 .i.c4 'iPf7 41 l:b7+ 'iPf6 42 l:la7 h6 0-1
with dangerous compensation for the
exchange. Game 4
1 9 ..ixb6 cxb6 20 'ilxb6 Wc6 21 Hamdouchi-Baburin
-.xc6 bxc6 Saint Vincent 2000
The arising endgame should be as
sessed as slightly better for Black, but 1 e4 f6 2 eS S 3 d4 d6 4 f3
Black's advantage has symbolical char ..ig4 S ..ie2 e6 6 h3 .i.hS 7 c4 lDb6
acter and it's really hard to believe that 8 exd6
White will lose this position in ten In the game Polgar-Yusupov, Dort
moves! mund 1995 White tried 8 ll\c3 il.e7 9
22 ..ie2 l:f8 23 g3 aS 24 'iPg2 l:lb8 dS but got nothing more than a fairly

14
Classical with 4 . . . .t.g4

even endgame after 9 ...exd5 10 cxd5 ing weak.


dxe5 1 1 g4 .ig6 1 2 lLlxe5 lLl8d7 1 3 1 1 g4!?
lLlxd7 1rxd7 (13 . . .lLlxd7 is also very An improvement on Bologan
reasonable, but Yusupov likes end Tischbierek, Vienna 1996, in which Bo
games) 14 .ib5 c6 15 dxc6 bxc6 16 logan played 1 1 .ie3 and obtained the
1Wxd7+ dl 17 .i.a6 .id6. Both sides initiative after lt.. .Lf3 12 .i.xf3lLI8d7
.

have some pawn weaknesses here. 13 'ife2 l:.c8 14 b3 fS?! (simply 14. . 0-0.

8 . . cxd6 9 c3 J/.e7
. looks like a possible improvement) 15
In view of the apparent strength of g4!. Bologan introduced 11 g4 in a sub
White's reply, Black might also consider sequent game against Agdestein.
9 .lLlc6!?, after which 10 dS exdS 1 1
.. 1 1 ....i.g6 1 2 h4 h5
cxdS .ixf3 1 2 .ixf3 5 leads t o posi Bologan-Agdestein, Yerevan Olym
tions similar to the next game, Nguyen piad 1996 continued 12 ... h6 13 .id3!
Anh Dung-Anastasian. 'ifc8!? (13 ...i.xd3 14 'ifxd3 'ilc8 15 b3
10 d5! 11'xg4 16 ltg1 hS 17 llxg7 fS is
strongly met by 18 ltlbS! e4 19 1fd4
etc.) 14 i.xg6 fxg6 15 l[)d2 0-0
(15 ...l[)xc4? 16 'ifa4+ wins the knight)
16 b3 lZ.f4 17 gS! and White stood
clearly better because of his mighty grip
on the e4-square.

An interesting new approach to the


position that was introduced by Bolo
gao. White simply takes some space in
the centre and challenges his opponent
to find enough counterplay.
10 ...e5
After 10...exd5 1 1 lbxd5 0-0 12 0-0 1 3 g5lDSd7 14 J/.e3 Ilea
White is simply better because of his In Vescovi-Leitao, Itabirito 1998,
strong knight and the weakness of Black played 14...a6 but still found him
Black's d-pawn. The attempt to win a self in difficulties after 15 b3 1llc7 16
pawn with 12 ... .ixf3 1 3 J.x3 lLlxc4 .id3 0-0 17 lLle4 (17 i.xg6 fxg6 18
would then be bad because of 14 'ifc2! l[)d2 followed by 19lLide4 was another
5 1 5 .ie4! ltlbc6 (15 ...g6 16llk7) 16 good line) 17 ... i.xe4 18 .ixe4 g6 19
.ixh7 + h8 17 f4, with White having lLid2 ilkS and now Volzhin suggested
the two bishops and Black's king look- 20 .ic2 (instead of Vescovi's 20 'ifc2)

,5
A lekhin e 's De fenc e

with a clearly better game for White


aher 20... a5 {20...1lac8 2 1 b4!? lDcd7 22
.tb3) 2 1 1lbl.
14 ...i.. f5 has also been tried but aher
15 lDd2 g6 16 lDde4, intending lDg3,
White was clearly better in Leonarda
Rodrigues, Maceira 1997.
1 5 b3 c5 1 6 %tg1 1?

28 ...:cxe4
The only move. Both 28 ... axb6 29
i..xc4 9xc4+ 30 'i'e2 and 28 ... i..xe4 29
i..xc4 1i'xc4+ 30 'i'e2 are hopeless for
Black.
29 xe4 .txe4 30 .txa7 'irf5 31
'ird4! tOes 32 :g3 i.xd5 33 .tg4
'irxg5 34 .tf3 'irxg3!?
Although this is partly as prophylaxis With both sides in time-trouble,
against Black moving his f-pawn, the Black tries his last chance. After
rook may also become an attacking 34 ...1lre5 35 'ifxe5 :XeS 36 i..xd5 1lxd5
piece in some lines. 37 i..xc5 dxc5 38 1lb1 Black's position
1 6 ...bd7 1 7 b4!? e4!? is hopeless.
Both sides play sharply and the game 35 fxg3 .txf3 36 i.xc5
gets thrown out of its 'normal' pattern. 36 lle1 is probably simpler though
The alternative was 17...lDe4. White should be winning in any case.
1S d2 d3+ 1 9 Wt1 lt:lxb4 20 36 ... dxc5 37 'irf4 .i.xh5 3S :e1
dxe4 0-0 21 .td4 :as 22 a3 a6 :xe1 + 39 xe 1 i.g6
23 .i.xh5 :xc4 24 .te2 'ireS!?
Black must act with urgency before
the storm breaks on the kingside. After
24 ...1lc8 25 h5 i..xe4 26ltlxe4lLlac5 27
lLlg3! White's knight will come to f5
with a ferocious attack.
25 h51
Ignoring Black's offer of the ex
change as aher 25 i..xc4 ...xc4+ 26 Wg2
lL!dcS Black obtains some counterplay
25 ....tf5 26 :g3 .i.fS 27 :e3 b6
28 .i.xb6

16
Classical with 4 . . . J..g 4

40 a4? Saanen 1994, White found another way


On the last move before the time to develop his queen's knight. After 13
control White lets the win slip. He a4 0-0 14 f4lLlg6 15 a5lLld7 he played
should first play 40 'ifc4!, preventing 16ltkl2!? and had the better game after
Black's bishop from coming to a6. 16 .. ."l'c7 17 4 b5 18lLle3 because of
40 ....i.d3! 41 a5 .i.b5 42 Wc7 ..ta6 his bishops and space.
Building an impenetrable fortress. A 1 3...0-0 14 a4 a5 1 5 l%a3 lLled7 1 6
draw is now inevitable. .i.b5 lLlc5 1 7 lLle2 .i.g5 1 8 lLlf4?!
43 g4 g6 44 g5 c4 45 >d2 .ib4+ White starts to lose his way and over
46 >c2 >g7 47 1i'd8 >h7 48 11ff6 the next few moves launches a some
c,t>g8 %-% what wild 'attack' on the kingside. Ac
r----- cording to Finkel, he should have
Game 5 played 18 f4 i.f6 19 g4!? g6 20 g5 i.g7
Nguyen Anh Dung-Anastasian 21 fS!? f6 22 fxg6 hxg6 23 gxf6lbf6 24
Budapest 1999 J:lg3 when Black's king is the weaker of
..______________. the two.
1 e4 lLlf6 2 e5 lLld5 3 d4 d6 4 lLlf3 1 8 l%c8 19 h4 .i.h6
.

.ig4 5 ..te2 e6 6 0-0 lLlc6 1 c4 lLlb6 Snatching the h-pawn with 19...i.xh4
8 exd6 cxd6 9 d5 exd5 1 0 cxd5 is obviously quite risky after 20 l:r.h3,
..txf3 1 1 .i.xf3 but it's by no means clear that this
White can also consider 1 1 gxf3!?, af would be bad.
ter which 1 1 ...5?! 12 i.bS+ d7 20 Wg4?!
{12 ...lLlbd7 loses a piece after 13 f4lLlg6
14 fS lLlgeS 15 'ife2 followed by f2-f4)
13 1i'd4 1i'f6 14 l:e1 + i.e7 15 1i'xf6
gxf6 16 3 gave White the better end
game in Vogt-Uddenfeldt, Skopje
Olympiad 1972.

Continuing with his 'plan' at the cost


of his d-pawn but this proves to be a
highly speculative venture. White
should have played the consolidating 20
g3.
20 ..txf4 21 .i.xf4 lLlxd5 22 .ig5

1 1 ...lLle5 1 2 ..te2 .i.e7 1 3 lLlc3 16 23 ._f3 d5 24 l%d1 lLlce4 25


In the game Oratovsky-Kaenel, llad3 'iVb6 26 .tel e6 27 .i.d4

17
A lekhine 's Defen c e

J:te2 28 t4 :tea 29 f3? 43 J:txb5!


.

Probably missing Black's reply. After The simplest.


29 l:le 1 the position would be very 44 axb5 l0d5 45 .i.d6 /Oe3 46 .i.e7
messy; Black is a pawn up but White's /Oxb5 47 .i.xb6 a4 48 .i.e3 a3 49
bishops are very strong. .i.e1 a2 50 .ib2 o!Oa3 51 4 ltle2
29 ... /0h5 52 e4 a,._ 53 .i.xa 1 /Oxa1 54
d4 ltlb3+ 0-1

Game 6
Zarnicki-Malbran
Argentine Ch., Buenos Aires 1998

1 e4 o!Of6 2 e5 o!Od5 3 d4 d6 4 /Ot3


.i.g4 5 .i.e2 c6

30 -.e5
Unfortunately for White he is forced
to enter what is probably a losing end
game. After 30 'ife3 there is 30 ... 'ifg6!
3 1 g4 ...d6! when Black obtains a win
ning attack.
30 ...-.xe5 31 .i.xe5 ltlf2 32 J:txd5
ltlxd1 33 J:txd1 J:te1 34 J:txe1 J:txe 1 +
35 h2 l0f6 36 .i.e3 b6 37 .i.e6 Salo Flohr's favourite line, which re
lld 1 38 g4 o!Od5 39 .i.e5 f6 40 .i.b8 ceived enthusiastic support in Burgess's
%ld2+ 41 g3 l087 42 .i.b5 J:txb2 books on the Alekhine.
43 .i.e7 6 e4
The simplest approach, which gives
White an edge without getting involved
in murky complications.
White's sharpest line is 6 lLlgS .tfs
(6....txe2 7 1Wxe2 is better forWhite) 7
e6 fxe6 8 g4 (8 J.. hS+ g6 9 g4 has also
been played but it seems fine for Black
after 9 ... J..xc2 10 ...xc2 gxhS l t lLlxe6
11fd7) 8 ..tg6 9 .td3 J..xd3 10 9xd3
..

lLlf6 (10...g6 1 1 c4 ltlf6 12 lLlxe6 'ifc8


13 9e2 6 14llk3 lbc7 15 d5 tl\xe6
16 dxe6 followed by J..f4 and 0-0-0

18
Classical with 4 . . . i.g4

gave White attacking chances in Keres fence of his kingside) 8 1Fxe2 h6 9ltlf3
Bhend, Zurich 1968) 1 1 e6 1i'd7 12 dxe5 10 dxe5 e6 1 1 0-0 ltl8d7 12 i.f4,
'i'e2 f7 13ltlg5+ g8 14 l:lg1ltla6 15 which was rather good for White in
t!t.1c3 lieS, De Firmian-Bw-gess, Gausdal Emst-Urday, Manila Olympiad 1992.
1995, and now White's best is probably 7 ...8d7
16 f4, as De Firmian suggested aher After 7 ... dxe5 8 ltlxe5 White has a
the game. pleasant space advantage and the at
One of the points of Black's system tempt to grab a pawn with 8 ...i.xe2 9
is that the 'natural' 6 0-0 can be met by 'iVxe21ixd4 gave White a strong attack
6... xf3 7 xf3 dxe5 8 dxe5 e6 after 10 ltldf3 1i'c5 11 0-0 f6 12 i.e3
producing a position in which White's 'WaS 13 i.d2 a6 14 b3! fxe5 15ltlxe5
bishops have little scope and the e5- lbsd7 16 11"h5+ g6 17 ltlxg6 ltlf6 1 8
pawn is weak. One example is lvanovic '4ih3 i n Marczell-Krecak, correspon
Kovacevic, Yugoslav Ch., Subotica dence 1986.
1984, which went 9 'ife2 liJd7 10 g3
1Fc7 1 1 l:le1 i.c5 1 2 ltld2 0-0 13 ltlb3
.i.b6 14 c4ltle7 {this is often a key ma
noeuvre for Black; the knight can come
to g6 from where it hits e5) 15 d2 aS
16 i.c3 a4 1 7 ltld2 i.a5!, exchanging
off White's dark-squared bishop, which
is a crucial defender of e5.
6 ...b6

8 exd6
The simple move, aiming for an ex
change variation in which Black's pieces
are rather passively placed- Black is
unable to aim for the traditional coun
terplay against d4 based on ...ltlc6.
Once again White can play the sharp
8 g5 i.f5 (8 ... i.xe2?! allows 9 e6! after
which 9 ... fxe6 10 'Wxe2 is clearly better
7 bd2 for White) with unclear consequences
White has an interesting alternative in after 9 e6 fxe6 10 g4 i.g6 1 1ltlxe6 (this
7 ltlgS!? i.xe2 (7 ... i.f5?! is not as good time 1 1 i.d3 is unavailable) ll ...'WcS
with the moves c2-c4 and ...ltlb6 etc.
thrown in as after 8 e6 fxe6 9 g4 i.g6 The line 8 h3 i.xf3 9 f3 dxe5 10
10 i.d3 i.xd3 1 1 xd3 Black's knight dxeS e6 is better for White than the
cannot come to f6 to help with the de- positions arising from 6 0-0 i.xf3 be-

19
A lekhin e 's Defenc e

cause his knight on f3 lends support to .i.g5! i.xg5 {2l. ...i.xf3? 22 J.xf6 gxf6
the e5-pawn. Even so, rd prefer to 23 gxf3 is even worse) 22 ll:lxg5 1i'e7
avoid them with White because Black's (or 22 ... g6?! 23 J.g8!) 23 h4 etc.
pieces would operate quite nicely on the 19 ...J..xf3 20 Wxf6??
dark squares. This is the point ofWhite's previous
8 ... exd6 9 0-0 J..e 7 1 0 a4!? move, but unfortunately he has missed
An interesting move which aims to something. It wasn't too late to bail out
create weaknesses in Black's queenside with 20 'ffxf3.
and bring the at-rook into play via a3. 20...gxf6 21 :xeS+ g7
Black's reply creates a 'hole' on b4, but Is this what White had missed? Of
his knights are too far away for this to course Black cannot play 21. ..:xe8??
be a problem for White. because of 22 i.h6 mate.
1 0 ... a5 1 1 .J:e1 0-0 1 2 .J:a3 :es 1 3 22.J:8e7 Wc8
.J:e3 Wc7 Black has a queen for a rook and
The game Yudasin-Timoshenko, Po everything seems to be clear, but the
dolsky 1989 went 13 ...lDf8 14 1i'b3! fun is just beginning!
.i.e6 1 5 .i.d3 :bs 16 1i'c2 d5 (maybe 23 J..d 3 'W'g4 24 J..n :9&!? 25 h3
16 ...1i'c7 was better) 1 7 c5 lDbd7 18 'W'h5 26 gxf3 Wxf3?
lDb3, intending 19 .i.d2, with a strong 26 ...h8+ 27 Wh2 1i'g6 forces mate.
initiative. Another possibility is 13...ll:lf6 27 .J:7e4 h8+ 28 .J:g4 .J:[xg4+ 29
but this still looks nice for White after hxg4 Wxg4+ 30 2 Wh4+ 31 g 1
14 1i'c2 .i.h5 1 5 .i.d3 etc. 'W'xd4 32 b3 Wg4+ 33 J..g 2 l0d7 34
14 J..d 3 lOt& 1 5 Wc2 J..h 5 1 6 lCJe4 J..b2 lC!e5 35 :e3 'W'd1 + 36 J..f 1
lC!xe4 1 7 J..xe4 J..f6 1 8 J..xh7+ 8 lCJf3+ 37 Wg2 lCJh4+ 38 Wg 1 Wg7
1 9 Wf5?! 39 .J:[g 3+ Wf8 40 J..xf6 lCJf3+ 41
Wg2lCJe5

White has seen an apparently attrac


tive 'combination', but there's a big hole Black has been making heavy weather
in his calculations. According to Finkel, of winning this game, but he's still got
White is basically just a pawn up after his extra queen.
19 :xeS+ :xeS 20 ltxe8+ xeS 2 1 42 J..d8 lOg& 43 J..g5 'W'h5!? 44

20
Classical with 4 . . . g4

.i.d3 lL!e5 45 .i.f1 lL!g4 46 .i.e3 ..tg4 5 c4


..Wh2+ 47 f3 lL!xe3 48 xe3 f5 49 5 h3 is very rarely played these days;
.i.d3 f4+ 50 xf4 'Wxf2+ 51 ltf3 after 5 ....ixf3 6 'ft'xf3 dxeS 7 dxeS e6
'Wd4+ 52 g3+ .1 53 .i.n 91 + the position is known to be very com
54 4 'ird4+ 55 g3 d5 56 cxd5 fortable for Black because of the lack of
cxd5 scope for White's bishops and the
weakness of his e-pawn. An example is
Zhuravlev-Aiburt, Daugavpils 1974,
which went 8 .ic4lDd7! (I personally
would prefer this to 8 ...lDc6 because it
keeps open the option of playing a later
...c7-c6) 9 ...g3lDe7 10 0-0lDfS 1 1 f4
.i.cS 12 lDd2 ...h4 13 lDf3 ...xf4 14
.i.xf4lDd4 when the e5-pawn was a big
problem.

The beginning of the end. Eventually


Black will be able to push his passed d
pawn.
57 ..tb5 1lg1 + 58 4 f6 59 J:ld3
Wf2+ 60 J:lf3 'ird4+ 61 g3+ >e5
62 J:ld3 11f4+ 63 g2 e4 64 J:lh3
Wd2+ 65 f1 1ld1 + 66 g2 1td2+
67 1 d4 68 J:lh4+ e3?! 69
l:lh3+ .4 70 lth4+ f3 71 lth3+
4 72 J:ld3 -.c1 + 73 e2 '*c2+ 5 ... lL!b6 6 exd6
74 f1 e4 75 lth3 'ird1 + 76 g2 Transposing into a kind of Exchange
Wg4+ 77 llg3 Wff4 78 lth3 d4 79 Variation, though one in which Black
l:.d3+ e5 80 J:lf3 'ird2+ 81 Wf1 d4 has developed his bishop to g4 rather
82 ..tc4 d6 83 ltd3 Wb2 84 l:lh3 early.
e5 85 ..te2 tlc1 + 86 f2 Wfc2 87 6 dS was a speciality of the Latvian
l:lh5+ d6 88 J:lh6+ c5 89 llh5+ player Vitolins; it looks playable and
b4 90 lth3 a3 91 l:ld3 b2 92 interesting but not dangerous for Black
ltxd4? Wc5 0-1 if he knows what he's doing. The most
...----. reliable defence is 6 ...e6 (both 6 ... dxe5 7
Game 1 h3 i.xf3 8 'ft'xf3 and 6 ...lD8d7 7 e6 fxe6
Mortensen-Baburin 8 h3 .i.xf3 9 'ft'xf3lbe5 10 'ft'b3 exdS
Copenhagen 2000 1 1 cxdS g6 t2lDc3 give White reason-
._____________.. able compensation forthe pawn) 7 exd6
1 e4 lL!f6 2 e5 lL!d5 3 d4 d6 4 ltlf3 1i'xd6 (7 ... ..txd6 is very messy after 8

21
A lekhin e 's D efen c e

'ifd4! i.xf3 9 'ifxg7 l:tf8 10 gxf3lD8d7 off one of his attacking pieces with 10
1 1 .i.gS i.e7 12 llgl, Vitolins-Bagirov, llJxd7} 9 ...lD8d7 (or 9 ...e6 10 llJc2) 10
J urmala 1985} 8 lDc3 exdS 9 cxdS c6 llJf3 with good play for the pawn.
and now: 6... exd6

a) 10 h3 .i.hS 11 g4 ..ig6 12 i.g2 7 tbc3 e7 8 .i.e3 ltlc6 9 h3 xf3


lDxdS 1 3 lDxdS cxdS 1 4 0-0 i.e7 As White has not castled kingside yet,
(14...lbc6!?} 15 11fd4 lbc6 16 'ifxg7 the plan of a kingside pawn storm is
0-0-0 gave Black good play in Vitolins something that Black must be very wary
Kengis, Riga 1984. of. After 9 ....thS White can play 10 g4
b) 10 .i.e2! (this is stronger} i.g6 1 1 d5 llJeS 12 lDxeS dxe5 1 3 c5
10 ... lbxd5 1 1 lbd4 .i.e6 12 0-0 i.e7 13 with a strong initiative.
lCixe6 fxe6 14 .i.hS+ g6 15 .i.g4 lDxc3 10 'tlfxf3 0-0
16 bxc3 1i'xd1 17 llxd1 lba6 18 .i.xe6 The attempt to exchange dark-square
llJcS was okay for Black in Vitolins bishops with 10....i.g5 is embarrassingly
Kengis, Jurmala 1985. answered by 1 1 1i'e4+.
Alekhine himself played 6 i.e2 on 1 1 h4
several occasions, though this should Preventing ....i.gS and starting a king
amount to no more than a transposition side pawn advance.
into normal lines after either 6...c6 or 1 1 . a5
..

6 ... e6. The move with independent sig Without having a light-squared
nificance is 6 ... dxe5 7llJxe5 (Aiekhine bishop on the board, l l . .. dS? is very
played 7 cS e4 8 cxb6 exf3 9 i.xf3 bad for Black because of 12 c5lL!c4 13
i.xf3 10 1i'xf3 in an earlier game against 0-0-0 and the dS-pawn is lost (Wed
Euwe, but in his notes pointed out that berg-Sandstrom, Stockholm 1998).
Black should have answered this with An idea worth considering is
10 ...llJc6) 7 ... i.xe2 8 1i'xe2 1i'xd4 1 1 ...llJa5 12 b3 c6 (maybe even
(Aiekhine-Reshevsky, Kemeri 1937}. 12 ...llJc6!?, arguing that the weakening
Now Alekhine felt that he should have of White's queenside with b2-b3 is an
played 9llJa3! (in the game he played 9 achievement), after which Mortensen
0-0 and aher 9 ...lL!Sd7 had to exchange gives 1 3 i.d3 d5 14 cS llJd7 1 5 tiffS!?

22
Classical with 4 . .. g4

g6 16 'i'h3 with attacking chances. Calmly evacuating the d-file aher


1 2 0-0-0 a4 1 3 c5! lL!d7 which he can feed his knights to the e5-
square and the cS-pawn is weak.
1 6 .i.b5 .i.xc5 1 7 .i.xa4 lL!de5 1 8
'ire4 .J.xe3+ 1 9 'ifxe3 'irg4 20 .tc2

14 a3?!
After this Black can equalise with ac
curate play. Mortensen claimed that he
could have kept a slight edge with 14 20 l:lad8
..

cxd6 i.xd6 (or 14...cxd6 15 a31Z.c8 16 Snatching a pawn with 20...1i"xg2?


b1/i)b6 1 7 i.d3 d5 1 8 hS threatening would be bad for Black because of 21 f4
19 'i'fs) 15 a3 b6 16 i.d3 1We7 17 g4 22 .d3 /i)f6 23 llhg1 1Wf2 24
bS, stopping any sacrifices on a3 and /i)dS (Mortensen).
keeping a useful bishop pair_ 21 f4 lL!g6 22 g3 lL!ge7 23 l:lxd8
14 dxc5 1 5 dxc5 'ireS
l:lxd8 24 l:ld1 l:lxd 1 + %-Y.z

23
A lekhin e 's Defenc e

Summary
Whilst 4 . .i.g4 remains more or less playable, it has come under pressure from so
. .

many different directions that I find it very difficult to recommend. For anyone
playing White I'd suggest adopting either Monensen's 5 c4 or Bologan's 9 dS. If
Black plays the Flohr Variation with 5 c6 then Emsts 6 c4 li)b6 7 gS!? looks
...

very prom1smg.

1 e4 lilf6 2 a5 lild5 3 d4 d6 4 lilf3 .i.g4 5 .i.a2 (D)


5c4-Game7
5... a6
5 .c6- Game6
..

6 0-G
6 h3 .i.hS 7 c4 b6 8 exd6 cxd6 9 3 .i.e7 10 dS (D) - Game 4
6 ....i.a7
6...6 7 c4 b6 8 exd6 cxd6 9 c:l5 Game 5 -

7 c4 lilb6 8 lilc3
8 h3 .i.hS 9 li)c3 0-0 10 .i.e3 dS 1 1 cS .i.xf3 12 gxf3 c8- Game 2
a . o-o 9 .ia3 d5 (D)
..

9 a6 - GameJ
..

10 c5 .i.xf3 1 1 gxf3 lOeB - Game 1

5 .ie2 10 d5 9... d5

24
I CHAPTER TWO I
Classical Kengis System:
4...dxe5 5 ttJxe5 g6

1 e4 lbf6 2 e5 lbd5 3 d4 d6 4 c!Jf3 Kengis set-up via 4 ... g6 5 ..tc4 c6 fol


dxe5 5 c!Jxe5 g6 lowed by a later ... dxe5. Black obtains
In the Kengis Variation Black solves excellent counterplay with a timely ex
the problem of the cramping e5-pawn change of his light-squared bishop fol
in the simplest and most direct way. He lowed by playing for the freeing ...c6<5.
simply removes it from the board! Games 8-1 1 show the kind of play that
Whilst classical theories about the cen might arise.
tre maintain that the resulting 'little cen Games 13-15 illustrate aggressive at
tre' formation gives White a space ad tempts by White to use the knight on e5
vantage, in practice it is no simple mat as the basis for ambitious and forceful
ter for White to do anything with it. play. Frankly I find it difficult to believe
The position could be compared with in the merits of the primitive-looking 6
lines of the Caro-Kann in which Black 1Wf3 (Game 15) but the space-gaining 6
plays ... dxe4 (1 e4 c6 2 d4 d5 3 !0:3 c4 (Games 13-14) needs very accurate
dxe4 4 liJxe4 etc.). On the face of it handling by Black. This may be a strong
White's outpost on the half-open e-file argument in favour of a 4... g6 5 .i.c4 c6
is more valuable than the d5-point on move order.
the half-open d-file. Yet Black's posi
tion is incredibly solid and by challeng Game S
ing the knight on e5 he will probably Mukhametov-Bagirov
force it to retreat with a loss of tempo. Bern 1995
Meanwhile Black will attempt to free his ,______________...,.
position with ... c7<5 or sometimes even 1 e4 c!Jf6 2 e5 lbd5 3 d4 d6 4 c!Jf3
... e7-e5. dxe5 5 c!Jxe5 g6 6 .i.c4
The main line of the Kengis is 6 c4, This has become established as the
though this may be at least partly due to main line though it's by no means clear
the fact that Black can also reach a that it's White's best move in this spe-

25
A lekhin e 's D efen c e

cific position. One of the reasons it b) 8 f3 .i.g7 9 lle4 (9 0-0 0-0 10


might have gained w idespread accep lte1lld7 1 1 h3lbxe5 12 dxe5 ti)xc3 13
tance is the fact the positions it leads to 'irxc3 .i.xc4 14 'irxc4 'W'dS was equal in
can be reached by transposition from Hort-Panchenko, Bern 1992) 9 . .0-0 10
.

4...g6 5 .i.c4 c6. Many white players 0-0 and now Psakhis's recommendation
may want to cut down their workload of 10...lLic7 (10 ...lLid7 is also playable)
by playing just one set-up against both 1 1 .i.xe6 lLixe6 12 c3 ds looks fine
of Black's move orders. for Black.
7 0-0 J.g7
Black has also tried 7 ...lLid7 8 ti)f3
lLI7b6 9 i.b3 i.g4 but the knight on b6
is not very well placed. The point is that
Black will usually want to advance his
queenside pawns in this type of posi
tion, to compliment the activity of his
bishop on g7.
8 l:le1 0-0 9 c3

6 ...c6
Black has to be careful to protect the
knight on dS. 6....i.g7?? loses to 7lL\xf7
r.txf7 8 f3+ e6 9 e4+ etc.
Having said that Black does have an
other move in 6 ....i.e6 and aher 7llX3
(given an exclamation mark by most
sources) to play 7 ... c6 (aher 7 ....i.g7?! 8
4 White threatens to put his knight
on g5 and 8 ....i.xe5 9 dxe5llc6 10 b3! 9 ...J.e6
llxe5 1 1 j.b2 f6 12 i.xe5 fxe5 13 lDc5 Black has a major alternative in
wins back the pawn with a huge advan 9...lLid7 aher which Ostojic-Bagirov,
tage). Aher 7 ...c6 White has a choice: Neu lsenburg 1992 continued instruc
a) The immediate 8 lle4 can be an tively with 10 lLI3 (exchanging knights
swered by 8 ...lDc7, when Godena with 10 lbxd7 eases Black's position)
Vaganian, Reggio Emilia 1994/95 went 10 ...lLI7b6 11 i.fl .i.g4 12 lLibd2 e6
9 i.xe6 lL\xe6 10 f3 f6 1 1 Wb3 ds (with the light-square bishop ex
12 xb7 xe4+ 1 3 .ie3 .i.g7 14 1i'xa8 changed, Black puts his pawns on light
fxe5 1 5 1i'xb8+ f7 1 6 'l'b4 'irxg2 17 squares) 1 3 h3 j.xf3 14 ti)xf3 lieS
0-0-0 exd4 1 8 i.d2 1l'd5 19 'irb3 lidS (playing for ...c6-c5) 15 c4li:Je7 16 b3
and Black has good compensation, even lLif5 17 .ie3 and now Black should
in the endgame. probably play 17 ...li:Jxe3 (in the game

26
Classical Kengis S ys tem: 4 . . . dx e5 5 tll x e 5 g 6

his 1 7...c5 was met by 1 8 i.g5 i.f6 19 After this Black has to be very care
i.xf6 'iixf6 20 dxc5 l:txc5 21 'iid2 with ful.
an edge for White) 18 fxe3 c5 when 1 6 lbd5?!
..

Black has excellent counterplay. Bagirov suggested that the immediate


10 lbd2 16 ...c5 would have been better; White
10 i.b3 transposes into the next would be unable to answer with 17lbb3
game, Howeli-Kengis. because Black can take it and the cap
1 o .. lbd7 1 1 lbxd7
. ture on d4.
The exchange of knights definitely 1 7 .ig3 c5 1 8 lbb3! 'it'c8 1 9 'it'e2
eases Black's game and 1 1 llJdf3 llJxe5 cxd4 20 lbxd4 J..d 7 21 'it'f3 e6 22
12llJxe5llJc7 is also fine for Black. Of l:lad1 a6 23 J..d 3 l:la7 24 .ic2 b5 25
the alternatives, 1 1 llJef3 i..g4 12 h3 J..h4 l:le8 26 J..b3 lbb6
i.xf3 1 3 'it'xf3 e6 leads to play similar Finally Black is out of the woods and
to Ostojic-Bagirov in the note to can claim full equality, if not more. He
Black's 9th, whilst 11 lbd3 llJc7 lets is starting to take some space on the
Black exchange bishops. queenside and his central pawn majority
1 1 :xd7 1 2 lbe4 b6 1 3 h3
later comes into play.
A slightly more testing way for White 27 J..f6 lbc4 28 J..xc4 'W'xc4 29
to play it is with 13 llJg5, after which J..xg7 xg7 30 'it'e3 'ilc7 31 .l:ld2 e5
van der Wiel-Bagirov, Yerevan 1996 Now I definitely prefer Black; the
continued 13 ... i..f5 14 h3 f6 15 lbf3 knight has difficulty finding a good post.
1lae8 16 i..fl i..e6 17 c4 llJc7 18 i.f4 32 lbf3 J..c8 33 .l:ld5 f6 34 a3 wn
i.f7 19 1i'a4lbe6 20 i.e3 llc8 21 llad1 35 l:led1 l:lae7 36 l:ld8 .ib7 37
'h-'h. The way Bagirov regrouped his .l:l1d2 l:lxdB 38 .l:lxd8 Wa2 39 l:ld2
forces in this game was very interesting 'it'b 1 + 40 >h2 'it'f5 41 .l:ld6 .l:lf7 42
and in the final position it is difficult for lbg1 h5 43 lbt3 g5 44 h4 Wg4
either side to make any significant pro
gress.
1 3 ... lbc7 14 .i.f1 h6 1 5 .i.f4 l:lfd8
16 lbd2

45 lbe1
Time-trouble? After 45 hxg5 i..xf3
46 'i'xf3 (and not 46 gxf6+? llxf6 47
gxf3 'i'h4+ 48 itg2 llxd6 etc.)

27
A lekhine 's Defen c e

46...'irxf3 47 gxf3 fxg5 48 lba6 :.Xf3 it Of White's other possibilities at this


should probably be a draw. point, 10 h3 is sensible as after 10...li)d7
45 ... l:ld7 46 l:lxd7+ Wxd7 47 Wc5 1 1 ltlf3 Black cannot pin the knight on
og6 48 We3 Wd5 49 Wd3+ Wxd3 f3 with ....i.g4. On the other hand he
50 l0xd3 a5 51 l0c5 .lc6 52 g3 f5 can try to exchange light-squared bish
53 oh3 0-1 ops with 1 1 ...ltlc7. White's critical reply
This was probably on time. Black is is 12 c4 (12 ltlc3 .i.xb3 13 axb3 ltle6
definitely better here but White can cer was very comfortable for Black in Del
tainly play on. Rio Angelis-Soppe, Santa Clara 2000;
r------ after 12 c3 .i.dS 13 .i.g5 ltle6 14 .i.e3
Game 9 {Zso. Polgar-Kengis, Vienna 1991}
J Howeii Kengis
.
-
Black should play 14...ltlb6 with equal-
London 1991 ity, according to Kengis) after which I
,____________..
like 12 ... ltlb6 13 'ire2 aS 14 a4 ltlc8,
1 e4 l0f6 2 e5 li:ld5 3 l0f3 d6 4 d4 intending either ...b7-bS or ...ltlc8-d6-f5.
dxe5 5 l0xe5 g6 6 .tc4 c6 7 0-0 After 10 ltld2 Black once again sets
.tg7 8 l:le1 0-0 9 .tb3 out to exchange a pair of minor pieces
White's most popular choice at this with 10...ltld7 1 1 ltlef3 (or 1 1 ltldf3
juncture, which keeps the option of ltlxeS 12 dxeS aS 13 c3 .i.g4 with a
pushing his c-pawn two squares instead comfortable game) 1 t...llk7 (1 t.. ..i.g4
of one and avoids Black winning a is als0 possible; after 12 h3 .i.xf3 13
tempo with ...ltlb8-d7-b6 in some lines. ltlxf3 Bagirov suggested the line 13 .. .a5
9 ....te6 14 .-g5 %te8 15 c4 1tl5b6 with counter
Black can also play 9 ...ltld7 but after play for Black), when Christiansen
10 ltlf3 ltl7b6 (Bagirov has suggested Kengis, Manila Olympiad 1992 contin
1 O aS) White can prevent ...i.g4 with
ued in interesting fashion with 12 c3 (12
1 1 h3 and leave Black struggling to find ltlc4 aS 13 a4 .i.d5 was also fine for
a plan. For 9 . . .a5 see the next game, Black in Milos-Llanos, San Luis 1995)
Short-Agdestein. 12 ... c5 13 .i.xe6 ltlxe6 14 d5 ltlc7 15
ltle4 ltlf6 16 c4 (after 16 ltlxc5 ltlcxd5
17 ltlxb7 there is 17 ...11'b6) 16 ...ltlxe4
17 l:lxe4 e6 18 .i.g5 f6 19 d6 ltle8 20
l:lxe6 fxg5 21 'ifd5 hS 22 llaet ltlf6
23 'irxb7 11'b6 24 lle7 Wxb7 25 llxb7
l:tfeS and White had enough compensa
tion for his piece, but probably not
more than that.
10 ...l0d7 1 1 l0f3 l0c7
As usual 1 1.. ..ig4 is very reasonable
for Black. Wahls-Maus, Hamburg 1991
continued 12 h3 .ixf3 13 Wxf3 e6 14
10 c3 ltld2 b5 15 ltle4 aS 16 a4 'irb6 1 7 ltld6

28
Classical Kengis S ys tem: 4 . . . dxe5 5 liJ x e 5 g 6

with a fighting game in which White's queenside because of the pressure


chances may be slightly preferable. against the a2-pawn.
1 2 .i.xe6 lJxe6 1 3 9b3 'i'b6 14 1 9 !Jcxd5 20 .i.g5 h6! 21 .i.d2
.

'ifc4?! White may have been hoping to play


21 i.e7 but then 2l...l:fc8 22 l:ladl
J:lc6! (threatening ... f7-f6 followed by
...f7) 23 i.h4 f5 puts White's bishop
in trouble.
2 1 ...!Ja4! 22 l:lab1 l:lfe8!
And not 22...J:tfd8 because White can
then eject the knight from a4 with 23
l:le4.
23 Wf1 a6 24 lJc4

White is following some old theory,


but in doing so he is drifting into trou
ble. Volzhin suggested just 14 bd2
with equality.
1 4 c5!
..

Opening lines, which certainly makes


sense for the side with the better devel
opment.
1 5 d6
Or 1 5 dxc5 dxc5 with play on the 24. .. b5?!
d- and c-files and a juicy d3-square wait Kengis later feh that 24...l:lxe 1 + 25
ing for his knight. ltJxel b5 26 ltJe3 l:ld8 would have been
1 5 . . .lbc7 1 6 l:lxe7 'i'd6 1 7 l:le1 more precise. I n playing the immediate
'i'xd5 1 8 !Ja3 !Jb6! ... b7-b5 he might have missed the fact
The last time Kengis reached this po that 25 ...ltJf4 did not work.
sition he was White against Tsesh 25 11e3
kovsky in Riga 198 1. On that occasion After 25 lLlceS Black has a strong
his opponent played 1 8 ...tl'xc4?! and move in 25 . ..lte6.
aher 19 l1xc4 stood much worse. Ken 25...l:led8
gis plays a much better move, simply Perhaps now Black saw that 25...lhf4
keeping White's knight on a3 out of the is answered by 26 ltJf5! when suddenly
game. White manages to free his position.
1 9 'ifxd5 White now has a fleeting moment in
Avoiding the exchange of queens also which he might have eased the defence.
leaves White in trouble. After 19 tl'h4 When he misses it Kengis gives him
ltfe8 he will have trouble developing his little in the way of further opportunities.

29
A lekhin e 's D e fenc e

26 l:tec1?1 b111f threatens mate with 54...'ifdl .


White had to play 26 lDxd5 l:.xdS 27 52 .'b5 0-1
.

:.ect with good chances of saving the In the line 52 ...b5 53 :xe4 fxe4+
game. He soon regrets the opportunity 54 e2 Black queens first and delivers
to exchange this knight as it promptly mate with 54 ... b2! 55 a7 b1'il' 56 aS'il'
heads for the e4-square. 'il'cl3+ 57 Wet c3.
26 l0f6! 27 c4 lOe4 28 l:tc2 l:ld3

29 i.e1 l:te8 30 lLld2 lLld6l 31 e2 Game 10


After 31 lDb3 there is 3 1 ...lldxe3! 32 Short-Agdestein
fxe3 lDxc4 with White's pawns drop Isle ofLewis (rapid) 1995
ping like flies, whilst 31 b3 lbc3 is really
excruciating. 1 e4 t:Llf6 2 e5 t:Lld5 3 d4 d6 4 lLlf3
31 . . .J:ldxe3 +! 32 fxe3 lLlfS 33 .tf2 dxe5 5 xeS g6 6 .tc4 c6 7 i.b3
t:Lld4+ 34 d1 t:Llxc2 35 Wxc2 f5 36 i.g7 8 0-0 0-0 9 l:te1 a5
a3 f7 37 b3 t:Llc3 38 :n h5 39 h3
t:Lle4 40 l0xe4 llxe4 41 cxbS?
41 cl3 would have been more tena
Cious.
41 axb5 42 l:td1 i.h6! 43 Wd3
..

we6 44 We2 i.gS! 45 Wt3 h4 46


l:td2 i.f6 47 l:ta2 dS 48 a4 b4! 49
aS c4 50 bxc4+
After 50 a6 cxb3 51 a7 :es 52 aS'if +
:xa8 53 l:xa8 b2 Black gets a new
queen.
so.. .wxc4 51 a6 b3
Immediately setting about his space
gaining operations on the queenside.
10 c3
There is an alternative in 10 a4 but
then 1 o....te6 1 1 c3 'il'c7 12 lL:Icl3 lL:Id7
13 lL:Id2 l:tfe8 14 lL:Ie4 h6 15 1Wf3 .ifS
16 .td2 "M>6 17 .tc2 :ads 18 :.e2
lL:I7f6 brought about exchanges and
equality in Godena-Kveinys, European
Tearn Ch., Debrecen 1992.
10 ...i.f5
I'm not sure that I like the bishop on
52 l:ta4+ this square but it certainly worked in
There is no longer any defence. After this particular game. In Korchnoi
52 a7 bxa2 53 aS'il' at 'it' White will run Vaganian, Horgen 1995 Black played
out of checks, and 52 liaS b2 53 a7 the more natural-looking 10...lL:Id7 with

30
Classical Kengis S ys tem: 4 . . . dx e 5 5 IC.xe5 g 6

the game going 1 1 ltlf3 l17b6 12 a4 1i'e2 but then 19....i.xc3! 20 bxc3 (or 20
'irc7 (12 ...i.g4 is possible) 13 i.gS .i.g4 .ixdS xdS 21 l:c4 'irb5 22 bxc3
14 ltlbd2 c5 (one of Black's key levers) ltlxc3 etc.) 20... lL1xc3 21 'ire 1 l:tdl is
1 5 h3 .i.xf3 16 'irxf3 e6 17 c4 (starting a simply terrible.
sequence which wins material but crip 19 e6 20 .i.g5 l:tdc8 21 l:tae1 Wa5
..

ples White's queenside) 17...ltlb4 18 22 We2 l:te8


.i.f4 1i'c6 19 'irxc6 bxc6 20 dxc5 ltld7 Being wary of any combinations
21 i.d6 ltld3! 22 i.xf8 xf8 23 lle2 based on llxe6.
:bs 24 l:lb 1 l17xc5 25 i.c2 ltlxb2 with 23 -*.d2 'irc5 24 ltle5 'irc7 25 g4
excellent compensation for the ex ltlf6 26 ltlxf6+ .txf6 27 .i.b1 J:la51
change.
1 1 h3
In Arnason-Knezevic, Grindavik
1984 White played 1 1 ltld2 and after
1 t...ltld7 12 ltlef3 Black could consider
12 ....i.g4 (in the game he played
12 ...lL17f6 13 a4 'ifc7 14 li)c4 ladS with
an okay position) 13 h3 hf3 14 lL1xf3
e6 with a typical position for this line in
which both sides have chances.
1 1 ... ltld7 1 2 ltlf3 ltl7b6 1 3 a3 'irc7
1 4 ltlbd2 a4
A nice way to bring the rook into
play.
28 h4 We& 29 g3 .i.e7 30 .i.f4 ltld7
31 .tc2 .i.c5 32 .tg5 g7 33 'ird2
f6 34 .th6+
34 lld1 lL1b6 35 .i.h6+ has been
recommended, but Black is fine there
too after 35 ...f7 36 c4 l:aa8 etc.
34 ...7 35 h5?! f5 36 l:th4 ltlf6 37
hxg6+ hxg6 38 .te3 -*.xe3 39 Wxe3
J:laa8
39 ...e5 looks better, as now White
1 5 -*.a2 should take the opportunity to put his
After 15 1Lc2 .i.xc2 16 'irxc2 c5 queen on that square. The remainder of
Black would stand very well because of the game was played under quite serious
his active pieces and the weaknesses in pressure from the clock.
White's queenside pawn structure. 40 .i.d1 ?! e5 41 .tf3 e4 42 .te2
1 5 c5 1 6 dxc5 'irxc5 1 7 ltle4
. J:lad8 43 l:th6 J:lh8 44 J:lxh8 l:txh8
.i.xe4 1 8 :Xe4 l:lfd8 19 'irc2 45 J:ld1 g7 46 'irf4 l:te8 47 Wd&
White would have liked to play 19 e3 48 f3?

31
A lekhin e 's Defen c e

The losing move. White has to play f6 15 .i.d3 e6 with a very solid posi
48 ...xc6 bxc6 49 .tfJ with a draw as tion and chances to get counterplay
the likely outcome. with a later ...c6-c5.
48 .. :ifxd6 49 .l:lxd6 lllh5 50 .l:ld7+ 8 lhdf3 0-0 9 0-0 li:ld7
h6 51 g2

10 l0d3
51 ...f4 52 gxf4 lhxf4+ 53 1 g5 The most difficult move for Black to
54 .l:lxb7 llh8 0-1 meet; White avoids any freeing ex-
After 55 l:b5+ f6 56 llb6+ eS 57 changes and posts the knight on a
llb5+ d6 Black's king escapes the square from which it inhibits ...c6-c5.
checks and 58 llgS is met by 58 ...lth2 After 10 llel xeS 1 1 xeS (1 1 dxeS
59 :g4 l:tf2+ etc. .i.g4 12 h3 i.xf3 13 ...xf3 e6 14 i.b3
,...----. c7 15 1i'e4 llad8 followed by dou-
Game 11 bling rooks on the d-file gave Black a
Nijboer-Vaganian very easy game in Magomedov-Kengis,
Dutch League 2001 Frunze 1989) 1t. ...te6 Black has a very
....______________. comfortable game, for example:
1 e4 lllf6 2 e5 l0d5 3 d4 d6 4 l0f3 a) Adams-Agdestein, Oslo 1994
dxe5 5 l0xe5 g6 6 .i.c4 c6 7 l0d2 continued 12 .tb3 a5 13 c3 c7 14
The knight comes to the support of d3 .tfS 15 i.gS llfe8 16 'Wd2 a4 with
its compatriot on eS. 7 0-0 .tg7 8 d2 Black already starting to take the initia
is less precise as after 8...d7 White tive.
cannot play 9 d3 because his d-pawn b) The bishop is not very well placed
would hang. on b3 so it might be better to play 12
7 . . ..i.g7 .tft 'Wc8 13 .td2 lld8 14 c3 with
Black can also regroup with 7....te6 8 equality, as in Conquest-Maus, Copen
4 c7, before White is ready to hagen 1990.
meet ....i.e6 with gS. Egger-Petrov, c) On the other hand 12 c3 allows
World Junior Ch., Buenos Aires 1992 Black a some tactics with 12 ...i.xe5! 13
continued 9 i.e2 .idS 10 i.d3 i.g7 1 1 dxeS c3! 1 4 ...xd8 l:Uxd8 1 5 i.xe6
0-0 0-0 12 c3 1Lld7 13 f4 i.xe4 14 i.xe4 2+ 16 'it1 xc1 17 i.xf7+ xf7 18

32
Classical Kengis S ys tem: 4 . . . dxe5 5 liJ x e 5 g 6

l:laxcl e6 with much the better rook 16 a4 'W"b6 17 ..i.c4 ltae8?! 18 l:e2 'W"d8
endgame. 19 ..i.d2 and now 19 ...e5? 20 ..i.xd5 cxd5
1 0 ...7b6 1 1 .tb3 21 11rxd5 exd4 22 ltxe8 ltxe8 23 cxd4
lLlb6 24 'i'xd8 ltxd8 25 ..i.xaS leh Black
a pawn down.
1 5 .a5 16 a4 l:le8 1 7 l:le1 9b6 1 8
..

.tc4 'flc7
The knight on d3 is making life diffi
cult for Black. If 18 ...c5? there is 19
JlbS, winning a pawn. Black decides to
bail out into an endgame in which the
bishops slightly favour White.
1 9 'ifxc7 xc7 20 .tf4 l:lec8 21
l:lad1 b5! ?

1 1 ....tg4?!
This may not be the best and I tend
to agree with Burgess's sentiment that
Black should probably want to ex
change the d3-knight. For this reason
l l...aS 12 a4 ..i.fS! is probably better,
aher which Arnason-Kengis, Jurmala
1987 continued 13 lte 1 ..i.xd3 14 ..xd3
e6 15 ..i.gS Wc7 (aher eliminating the d3
knight, Black will find it easier to play
...c6-c5) 16 11'e4 lbc8 17 c4 lLld6 18
11'e2 lLlb4 19 lladl llfe8 20 11'd2 lLlfS Despite the dangers posed by the
with excellent counterplay. bishops, Vaganian plays in aggressive
1 2 c3 e6 and double-edged fashion. Obviously
12 . a5 would give rise to similar play
. . there are deep risks associated with this
to the game. kind of play and Vaganian has to walk
1 3 h3 .txf3 1 4 Wlxf3 d7 1 5 Wlg3 the tightrope in the later stages of this
There is a major alternative in 15 game.
lte1, aher which Kolev-Panchenko, 22 .tb3 .tf8 23 lla1 5 24 .tg3
Villalba 1996 continued in somewhat l:ta7 25 lle2 llca8 26 f3 llb7 27
artificial vein: 15 ...a5 (15 ... b5 16 g3 a5 . .i.f2 llc7 28 g3 5b6 29 Wg2 xa4
17 a3 :e8 1 8 h4 hs 19 .tgs i..f6 20 I'm not sure Black should have done
..i.d2 a4 21 ..i.a2 'ile7 was an earlier this as his queenside pawns are now
Panchenko game as Black against split. He probably got tired of waiting;
Kruppa from the 1990 Ukrainian in any case this isn't much for White.
Championship; no doubt he got the 30 .txa4 bxa4 31 l:lxa4 lOb& 32
idea to play ...e6-e5 aher this encounter) lla2 a4 33 f4 .i.d6 34 llc2 llb7 35

33
A lekhin e 's D e fen c e

Wf3 f5 36 We2 Wf7 37 c4 lbxe5 8 dxeS c6 9 0-0 .tg7 10 'i'e2 i.e6


1 1 b3 'ifc8 12 i.b2 .th3 was equal in
Wolff-De Firmian, New York 1996)
7... .tg7 8 .tg2 0-0 9 0-0 cS 10 .Ue1
cxd4 1 1 llxd4 lbes 12 lla3 .tg4 13
lb3 e6 14 c4 i.xf3 15 .txf3 b4 16
i.xb7 bd3 and Black won this wild
game, though it's anybody's guess
what's happening at present.
b) There's nothing wrong with the
'standard' 6 ...i.g7 either, after which 7
i.g2 0-0 8 0-0 c6 9 l:le 1 .tfs 10 c3 d7
1 1 lbf3 lle8 12 lbbd2 was Anand
It's now quite clear that Black's risky Adams, Linares {1st matcbgame) 1994,
play has backfired. He manages to hang and now Black should have played
on by the skin of his teeth. 12 ...e5 13 c4 i.g4 with close to com
37 .../.0d7 38 c5 c7 39 .J:lc3 .J:lba7 plete equality.
40 .J:lca3 I.Of6 41 e1 We8 42 b3
I.Od5 43 .J:lxa4 .J:lxa4 44 .J:lxa4 .J:lxa4
45 bxa4 Wd7 46 I.Ob2 h5 47 /.Oc4
h4 48 Wf3 hxg3 49 xg3 Wc8 50
/.Ob2 d8 51 f2 Wb7 52 /.0c4 f6
53 e3 Wa6 54 We2 g7 55 Wd3
I.Ob4+ 56 Wd2 I.Od5 57 Wd3 I.Ob4+
58 Wd2 %-%

Game 12
Miles-Pons
Andorra 1996
...._
.. _________...... 6 ...g7
1 e4 I.Of6 2 e5 I.Od5 3 d4 d6 4 I.Of3 Once again it's possible to play
dxe5 5 /.Oxe5 g6 6 I.Od2 6...d7, after which 7 lbxd7 'ifxd7!? 8
With a huge slab of irony, Miles f3 .tg7 9 i.e2 1i'd6 10 0-0 0-0 1 1
awarded this move two exclamation lle1 cS 1 2 dxcS xeS 1 3 c4 lbb4 1 4 a3
marks in Informator 67 and splattered c6 15 b4 was a bit better for White in
more exclamation marks over the rest Ricardi-Speelman, FIDE World Ch.,
of the game. Later on he switched sides Las Vegas 1999.
and showed how to equalise with the 7 1.0df3
black pieces. 7 i.c4 0-0 8 M3 c6 9 0-0 dl
Another quiet line is 6 g3 and now: would have transposed into the previ
a) Plaskett-Hillarp Persson, St Helier ous game, Nijboer-Vaganian.
1999 continued 6 ... lbd7!? 7 lbf"3 (1 .tg2 7 ...0-0 8 c4 I.Ob6 9 e2 I.Oad7 1 0

34
Classical K engis S ys tem: 4 . . . dx e 5 5 li:J x e 5 g 6

.i.f4 xe5
When Miles played Black in this posi
tion (K.ofidis-Miles, Agios Nikolaos
1997} he preferred 10 ...c5 and got
slightly the better of it after 1 1 dS liJxeS
12 .i.xe5 .i.xeS 13 ltJxeS 9d6 14 liJf3
e6 etc. The implication is that 6 liJd2
doesn't refute the Kengis Variation after
all.
1 1 xe5 d7 1 2 'it'd2 xe5 1 3
.i.xe5 .i.xe5 14 dxe5 'it'xd2+ 1 5
xd2 .i.e& 1 6 c3
36 b4
Surprisingly the pawn endgame is a
draw. Miles gave the variation 36 b6
d6 37 f3 g5 38 b3 e6 39 xa6 c6 40
a7 'l;c7 41 a6 e4 42 fxe4 e5 43 aS
c8 with no further progress possible.
36 . . .cxb4 37 c5 b3 38 c6+ d6 39
c7 b2 40 cS'it' b1'it'+ 41 a7 'it'b5
42 'it'xa6+ 'it'x86+ 43 xa6 c6 44
f3 g5 45 a7 c7 46 86 c6 47
87 c7 Y.z -Yz

In the endgame, White's active king Game 13


and greater control of terrain give him a J .Polgar-Agdestein
clear edge. Isle ofLewis (rapid) 1995
1 6 . . .f6 1 7 .i.f3 c6 1 8 llh81 7 1 9
a4 llad8 20 a 5 a6 21 llad1 llxd1 1 e4 f6 2 e5 l0d5 3 d4 d6 4 l0f3
22 .i.xd1 lld8 23 .i.f3 .i.f5 24 b4 dx85 5 85 g6 6 c4
lld3 25 c5 llb3 26 lle2?
The move which allows Black to es
cape. According to Miles he should
have first played 26 exf6, after which
26 ... exf6 (26...lhb2 27 lbe7+ f6 28
llxh7 also wins for White) 27 lte2, fol
lowed by lld2 and .i.d1, will win easily
enough.
26 ...fxe5 27 lld2 86 28 .i.82 h5
29 h3 h4 30 .i.d1 lld3 31 llxd3
.i.xd3 32 .ig4+ .if5 33 b6 .ixg4
34 hxg4 d7 35 xb7 c5!

35
A lekhin e 's Defen c e

A tough line for Black to meet. White ter which Grischuk-Baburin, Torshavn
takes the lion's share of the centre and 2000 went 9 ...c5!? {9....i.f5 is also possi
puts the onus on Black to find decent ble but relatively passive) 10 .i.e2 cxd4
counterplay. 1 1 .ixd4 i.h6 12 lta1 6 13 ltJg4 i.xg4
6 ...l0b6 7 l0c3 .ig7 8 .ie3 14 i.xg4 c6 15 .i.c5 lbe5 16 .i.e6+
In the game Gallagher-Kengis, Bern g7 17 lbds lL'Ixds 18 cxds 'ifaS+ 1 9
1992 White varied with 8 1Lf4 0-0 9 b4 'ti'a6 20 'ife2 d3+ 21 cla>f1 xeS 22
1Le2 aS!? 10 h4?! (simply 10 0-0 is bet bxc5 and now 22 ...'ifa5 looks interest
ter), after which 1 0...llk6! left White ing (in the game Baburin was much
struggling for equality. The game con worse after 22 ...111xe2+ 23 cJilxe2 b6 24
tinued 1 1 lt.Jxc6 bxc6 12 .i.e5 i.xe5 1 3 c6 .i.f4 25 d3).
dxe5 i.e6! 14 b3 ...xd1+ 15 l:txd1 a4 9 ... f6
and now, according to Volzhin, 16 0-0
was probably the best chance as after
16 ... axb3 17 axb3 lta3 1 8 l:lb1 .i.xc4 19
i.xc4 lL'Ixc4 20 bxc4 .:xc3 21 l:.fcl!
l:lxc 1 + 22 lbc1 White should be able to
draw.
8 ...0-0
The alternatives are 8 ... c5, 8 ... c6
and 8 ...i.e6, all of which will be looked
at in the next game.
9 .f3
A typical judit Polgar move, intend
ing to castle long. 10 l0d3
Another sharp move is 9 f4, support 10 xg6 doesn't work after 10 ... hxg6
ing the knight on e5: 1 1 c5 lb6d7! 12 i.c4+ ltf7 (and not
a) Gdanski-Tomaszewski, Naleczow 12 ... h7? 13 1i'h3+ followed by mate).
1988 continued 9 ...lL'I6d7 10 'it'f3 c5 1 1 1 0 ...e5!
0-0-0 ...aS 12 ltJxd7 xd7 1 3 .i.f2 ltb8 Quite right. Black has the better de
14 i.e2 and now 14 ... b5!? looks velopment so he opens the game up.
interesting (in the game Black played 1 1 dxe5 fxe5 1 2 ..d1 l0c6 13 l0c5
14...cxd4 1 5 i.xd4 eS 16 fxe5 ltJxe5 but e8
was slightly worse after 17 'ifd5). A slightly odd-looking move, though
b) Another possibility is 9 ... c6, after it turns out to be okay. Black can also
which Pokojowczyk-Tomaszewski, Pol put his knight in the middle with
ish Ch. 1980 continued 10 3 f6 1 1 13 ...d4. In any case he has an excel
ltJg4 i.e6 12 b3 .i.xg4 1 3 1i'xg4 f5 14 lent game.
...g3 e6 1 5 l:d1 ...e7 16 i.e2 ltJ8d7 17 14 lbd5 f7 1 5 l0e4 .if5 1 6 lOgS
0-0 a5 with counterplay on the queen 'ifd7 1 7 l0xb6 'ifxd1 + 1 8 ltxd1
side. axb6 1 9 c5
White's other possibility is 9 :let, af- Opening up the c4-square for the

36
Classical Kengis Sys tem: 4 . . . dx e5 5 t0xe5 g 6

bishop. Accurate play by both sides re


sults in perpetual check.

9 dxc5 'tl'xd 1 + 10 ltxd1 i.xe5 1 1


cxb6 .t.xc3+
1 9 ... h6 Black should take the chance to dou
Obviously not the rash capture ble White's pawns. Aher l l...axb6 12
19 .. .llxa2? because of the forking reply i.d4 J.xd4 13 llxd4 White was much
20 i.c4+. better in Campora-Abreu, Las Palmas
20 .i.c4+ h8 21 h4! b5 22 ll:lf7 + 1993 .
h7 23 ll:lg5+ h8 24 ll:lf7+ h7 1 2 bxc3 axb6 1 3 :d2 ll:ld7 14 i.e2
25 ll:lg5+ %-% ll:lc5 1 5 0-0 .i.f5
15 . 0-0 might be an improvement for
. .

Game 14 Black, but this endgame is worse for


Ernst-Bagirov him in any case.
Helsinki 1992 1 6 i.f3

1 e4 ll:lf6 2 e5 ll:ld5 3 d4 d6 4 ll:lf3


dxe5 5 ll:lxe5 g6 6 c4 ll:lb6 7 ll:lc3
.i.g7 8 .i.e3 c5
8 ...c!tJc6 is a known idea, allowing
doubled pawns but eliminating the
knight on e5. Unfortunately it doesn't
seem to work aher 9 c!tJxc6 bxc6 10
1id2 aS 11 e2 a4 12 lld1 0-0 13 h6
when White was much better in lvanov
Men, US Ch., Durango 1992. 8...e6 is
a solid move but does little to challenge
White's space advantage. Bode-Kengis, 1 6 .i.e4??
..

German Bundesliga 1991 continued 9 A blunder. Black should play 16 ..0-0


.

f4 c6 10 1ib3 0-0 1 1 ..te2 8d7 12 0-0 aher which 17 llb2 1la6 18 xeS?! (18
llJf6 13 lladl with the better game for lldl) 18 bxc5 19 lbb7 e6 leaves him
..

White. without further problems.

37
A lekhin e 's De fen c e

1 7 ..i.xc5 ..i.xf3 1 8 .id4 e5 19 i.xe5 54 We8 lla2 55 l:[g6+ '15 56 llf6+


f6 20 lle1 fxe5 Wg5 57 llf8 1 -0
Black has little choice but to jettison r------.
pawns and hope he can hang on. Game 15
20... 0-0 21 .i.d6 would win the ex- Ernst..Josefsson
change. Swedish Ch., Karlskrona 1983
21 llxe5+ 7 22 gxf3 llhc8 23
lld7+ Wf6 24 llb5 llxc4 25 l:txb6+ 1 e4 lbf6 2 e5 lbd5 3 d4 d6 4 lbf3
Wg5 26 llxh7 l:lxc3 27 h4+ Wf5 28 dxe5 5 lbxe5 g6 6 'irf3
l:ln+ We5 29 l:xg& l:lxa2 30 lle7+ One of White's two most bludgeon
Wd5 31 lld7+ We5 32 llxb7 llxf3 ing moves. The other one is 6 h4?!,
which is well met by 6...d7 7 xd7
1i'xd7 8 hS .i.g7 9 c3 cS 10 dxcS ..c6
1 1 d2 .i.fS as in Reeh-Loeffler, Am
stelveen 1994. Undaunted by this earlier
experience, Reeh later tried 6 .i.c4 c6 7
h4?! against Maus in the 1992/93 Ger
man Bundesliga season. Once again
Black had a good game aher 7...d7 8
xd7 .i.xd7 9 hS .i.g7 10 c3 'irc7 1 1
d2 0-0-0 1 2 4 eS etc.

Sometimes you can't win rook end


games with extra f- and h-pawns, but in
this case Black's king has to go too far
away.
33 llb5+ Wd4 34 llg4+ Wc3 35
llg2 Wc4 36 llbg5 lla1 + 37 Wh2
llf1 38 h51 llf6
It turns out that Black can't get his
pawn back. White wins easily after
38 .. .1Uxf2 39 h6 1hg2+ 40 xg2 l:tf6
41 l:.hS l:.f8 42 h7 l:th8 43 f3 etc. 6 ... i.e6
39 Wg3 Wd4 40 'g4 'e4 41 llg6 In my view this is an only move. If
l:l 1 xf2 42 l:lxf2 llxf2 43 l:te6+ Black has to play the ugly 6 . . f6 he can
.

Cutting Black's king off. White's h forget about the Kengis Variation alto
pawn carries the day...just! gether.
43 ...Wd5 44 l:le8 llh2 45 Wg5 'd6 7 c4
46 h6 llg2+ 47 '16 llf2+ 48 'g7 The critical line. If White plays 7 .i.c4
llg2+ 49 Wh8 Wd7 50 llg8 llh2 51 there follows 7...J.g7 {7...c6 8 0-0 ltld7
h7 we& 52 'g7 l:tg2+ 53 Wt8 llf2+ 9 3 .i.g7 10 l:td1 .i.xeS 1 1 .i.xdS

38
Classical Kengis S ys tem: 4 . . . dxe 5 5 (i)xe5 g 6

i.xclS 12 lC!xclS .tg7 w as also very 1 1 . .. c5!? is the other possibility and
comfortable for Black in Sherzer looks massively complicated.
Panchenko, Chicago 1992) 8 li)c3 (8 ().() 9 1ixa8
c6 9 l:te1 lC!d7 10 lbxdl xd7 1 1 c3 White has to take the rook if he
0-0 12 i.g5 l:tfe8 13 h3 lbc7 14 lC!d2 wants anything out the opening. 9
.i.d5 1 5 i.xdS cxclS and Black had an 11fxb4 Wxe5+ 10 .i.e2 lL'Id7 1 1 lbc3
excellent game in Luther-Bagirov, Len .tg7 12 0-0 0-0 13 .i.e3 llab8 gave
ingrad 1989) 8 ... c6, which transposes Black excellent play in Ziska-Flindtholt,
into the 6....te6 note in Mukhametov Festuge 1991.
Bagirov (see Game 8).
7 ltJb4 8 'irxb7
.

White has an interesting alternative in


8 d5!?. Vilar Lopez-Regan, Saint Vin
cent 1999 continued 8 ... lbc2+ 9 'it'dt
lL'Ixal 10 lDd2 i.fS (and not 10 ...i.g7
1 1 dxe6 0-0 because of 12 exf7+ h8
13 lL'Ixg6+ hxg6 14 ...h3+ ..th6 15
..xh6 mate) 1 1 .i.d3. Now Black
should play l l...lL'Idl (in the game Black
played 1 1 ...11Vd6 and after 12 lDxf7!?
xf7 13 g4 Black should still have
played 13 ...l0d7), after which 12 lbxf7 9...'irxe5+1
xf7 13 g4 is bad because of 13 ... lDe5 In New Ideas in the Alekhine Defence
etc. Burgess only mentions 9 ...lL'Ic2+. I like
the knight where it is because it can
now keep White's queen out of play.
Meanwhile Black will race ahead with
the rest of his development, with good
compensation for the exchange because
White's king is in the middle.
1 0 Wd1 lb4c6 1 1 ltJc3
Kavalek cops out of assessing this
position by calling it 'unclear'. l'll stick
my neck out and claim that Black is bet
ter. It's only an exchange and Black is
developing a massive attack.
8 .. 'irxd4!
. 1 1 .. .tg7 1 2 .i.e3 0-0 13 'irb7 AdS+
.

This feels right to me; it keeps Black's 14 We1 ltld4 1 5 .i.d3 ltJbc6 16 ltld5
pieces centralised and all his options %1b8?
open. 8 ...lbc2+ 9 dl lDxa1 10 Wxa8 A serious mistake. Black should play
.i.g7 1 1 ..td2 (1 1 ... b7, mentioned by 16 .. ..txcl5 17 cxclS WxciS with excellent
.

Fleck, is another idea for White) play for the exchange.

39
A lekhin e 's Defen ce

26 f4
Ernst is fighting for his life. 26 g3
lDxg3 is not an improvement.
26 .loxf4 27 c5 We5 28 l:lf3
.

17 'ita&?
White in turn misses his chance. He
can simply play 17 lLlxe7+! lLlxe7 {or
17...'f8 18 lLlxc6) 18 11t'xb8+, winning
on the spot. 28 ... Wxe3
1 7 ... .ih6 1 8 1 .ixe3 19 l:le1 .if5 This certainly ought to be good
20 .ixf5 'itxf5 2 1 lDxe3 enough but 28 ...lDxg2+ butchers White
After 2 1 lbe3 there is a good old on the spot.
fashioned mating combination with 29 :hf1 l:lxg2+ 30 Wh1 l:lg1 +?
2 1...'ifb1+ 22 l:te1 'ifd3+ 23 g1 + A jolly amusing way to draw a win
24 fl (or 24 l::xe2 11'd1+ etc.) ning position. 30 ...11'd2 is decisive.
24. ../t)g3+ 25 g1 11'fl+ 26 .Lf1 lDe2 31 >h2 l:lg2+?
mate. And here 31. ..11'd2+ 32 lUf2 l:tg2+
21 'itd3+ 22 Wg1 l:lxb2 23 h3
. WlnS.
'itd2 24 :n ltle2+ 25 Wh2 'itd&+ 32 Wh1 l:lg1 + 33 >h2 l:lg2+ %-%

40
Classical K eng is System: 4 . . . dx e5 5 lfjxe5 g 6

Summary
The Kengis Variation has become the workhorse of most of the Alekhine profes
sionals. The primitive 6 ..f3 looks very dangerous for White and against 6 .i.c4
Black gets a solid position with fighting chances. My greatest concern is about the 6
c4 lines, but Black can actually sidestep these by adopting a 4 g6 5 .t.c4 c6 move
..

order.

1 e4 l0f6 2 e5 l0d5 3 d4 d6 4 f3 dxe5 5 xe5 g6 6 J.c4 (D)


6 ..f3 - Game 15
6 ltld2 .t.g7 7 ltldf3 ()..() 8 c4 - Game 12
6 c4 ltlb6 7 ltlc3 J.g7 8 J.e3
8 0-0 Game 13
-

8...c5 - Game 14
6 ...c6 7 0..0
7 tnd2 ..tg7 8 llldf3 ()..() 9 ()..() end? lO lbd3 (D) - Game 1 1
7 . . ..ig7 8 D.e1 0-0 9 J.b3
9 c3 - Game S
9 ... a5
9 ...te6 - Game 9
..

10 c3 .if5 (D) - Game 10

6 .ic4 1 0 llid3 TO. . J.f5

41
I CHAPTER THREE I
Classical with 4 . dxe5 5 ltJxe5:
. .

5 ... ltJd7 and 5 c6 . . .

1 e4 ll!f& 2 e5 llJd5 3 d4 d& 4 f3 kingside fianchetto, which is in fact


dxe5 5 xe5 what happens in Games 19, 21 and 22.
From a structural point of view, the But given a chance, Black can also
positions that arise here bear an obvious adopt a set-up in which he develops his
similarity to the Kengis Variation. Once queen's bishop (....ig4 for example) and
again it brings about a 'little centre' then simply plays ...e7-e6. This plan of
structure in which Black plays for an development was used in Games 18 and
eventual ...c6-c5 to free his position. 20.
The difference is that he is not neces In Game 23 White tries 6 c4, which is
sarily committed to a kingside fi certainly a very interesting line for
anchetto. White in the Kengis Variation. But here
An immediate challenge on White's Black is not committed to a kingside
knight on e5 with 5 ...llxl7 invites a scary fianchetto and Miles was doing weU in
sacrifice on f7. Black's king gets caught this game. Very well, even.
in the centre but Games 16 and 17
show that he has many defensive re Game 16
sources. I even wonder whether White Rozentalis-A .Sokolov
should sacrifice the knight at all; a more Bern 1992
circumspect choice would be the quiet
retreat back to f3. 1 e4 f& 2 e5 d5 3 d4 d& 4 f3
Black can eliminate these dangers by dxe5 5 xeS d7
first defending the knight on d5 with As first played by Bent Larsen in his
5 ...c6. Britain's Tony Miles has been 1965 match against Mikhail Tal. The
making a living out of this move and it 'magician from Riga' would have seen
is most definitely a tough nut to crack. the knight sacrifice on f7 in the .twin
The game can easily transpose back into kling of an eye, but strangely enough
a Kengis Variation if Black plays a later decided against it. His decision was no

42
Classical with 4 . . . dx e5 5 l0xe5: 5 . . .li::J d 7 and 5 . . . c6

doubt prompted by the thought that The only move. 7 .. .<g8?? 8 1llxd5+
Larsen must have analysed the sacrifice leads to mate and 7 ...g6 8 'irxdS+ wins a
at home. After the game analysts pored pawn for nothing.
over the sacrifice and Tal was criticised 8 c4
for having avoided what was probably a During the Larsen game, Tal spent
'winning' line. But more than twenty some time analysing 8 g3 but the posi
years later, the English amateur Peter tion is far from clear after 8... b5! and
Sowray started playing 5 ...ltXI7, and now:
suddenly people weren't quite that sure.

a) 9 b3 b4 10 a3 lD7f6 1 1 .ih3+ d6
6 liJxf7l? 12 1i'e5+ c6 13 .ig2 was played in
Tal played 6 .ic4 after which van der Angelov-Orev, correspondence 1961,
Wiel's suggestion of 6...llJxe5 7 dxeS c6 and now Black should have defended
8 'irf3 'irc7 may be Black's best. himself with 13 ...1i'd6!.
Many strong practical players have b) 9 a4 c6! (Greiner-Sowray, corre
decided simply to avoid an exchange of spondence 1988 varied with 9....ib7 10
their knight on eS with 6 lDf3, after i.h3+ d6 1 1 1i'f7 cS 12 llJc3 lDxc3 13
which 6 ... e6 (6 ... c5 7 c4 lD5f6 8 d5 llJe4 1i'e6+ r:llc7 14 .if4+ r:llc8 15 bxc3 gS,
9 .i.d3 llJd6 10 0-0 g6 1 1 l:le1 ..tg7 12 but now 16 dxcS gxf4 17 0-0-0 would
.igS llJf6 13 llJcJ was also nice for have won on the spot; Burgess has sug
White in Pavosovic-Hillarp Persson, gested 9... b4 but this awaits practical
European Team Ch., Batumi 1999) 7 g3 tests) 10 axbS g6 1 1 'ite2+ r:l/f7 12 bxc6
.ie7 (7... b6 8 c4 llJSf6 9 ..tg2 ..tb7 10 lD7b6 1 3 .ig2 was played in Ernst
0-0 lLe7 11 llJc3 0-0 12 J..f4 was also Komarov, Dortmund 1992, and now
slightly better for White in Psakhis 13 ...lLg7! 14 c4 lDb4 15 dS .i.fS 16 .ie4
Komarov, Benasque 1995) 8 .ig2 0-0 9 i.xe4 17 'itxe4 lD4xd5! would have
0-0 cS 10 c4 lb.sf6 1 1 llJc3 cxd4 12 equalised according to Ernst.
llJxd4 gave White had a comfortable One practical drawback of S llJcll is
...

edge in Groszpeter-Pitschka, Pardubice that White can take a draw by repetition


2000. at this point with 8 'irg4+ f7 9 1rh5+
6 . . .xf7 7 h5+ e6 etc.

43
A lekhin e 's D efen c e

8 . lD5f6 9 d5+ d6 1 0 'iff7


. . overwhelming position) 16 'ire1 1i'f5 17
g3 g6 18 d3 h6+ 19 rj;b1 'i'f6 20
h3 and White was winning in Nguyen
Miroshnichenko, Budapest 1999.
1 1 .i.f4 c5
Forced, in view of the threat of 12
c5+.
1 2 lDc3
White has also tried 12 li)d2 but then
12 ... g5! (and not 12 ...1i'e8? 13 ltle4+) 13
i.g3 g4 14 Q.().Q h6 15 llel llf8 16
xeS+ d7 17 e6+ e8 18 d6!
exd6 19 i..xf6+ d7 20 xd8 xd8
The obvious move is 10 f4+? but brings about an equal position accord
then 10...e5! 1 1 c5+ (1 1 dxe6+ <iftxe6 ing to Bagirov.
leaves White with nothing) 1 1. ..e7 12 The other try is 12 b4!?, which we'll
gS 'IreS consolidates for Black. look at within the context of the next
1 0 lDe5
.. game.
Putting the knight in the centre is 1 2 . .86
.

thought to be the best, though in such a Black had to defend b5 and


complex position it is difficult to be 12 .....td7? would be a blunder because
totally confident about this. of 13 li)bS+ i.xbS 14 e6+ etc.
The main alternative is 10...tbb8, af. 1 3 0-0-0
ter which 1 1 c5+ d7 (l l. .xcS 12
. This fails because of Black's crahy
i..e3+ d6 13 tba3! a6 14 tbc4+ 'itd7 reply, which exploits the position of
15 d6! wins} 12 i.f4 (12 bS+ c6 13 White's king on cl. For this reason Yu
dxc6+ bxc6 14 0.0 is unconvincing aher dasin suggested 13 lld1 rather than 1 3
14 ... 1i'a5! 15 lld1+ rj;c7 16 f4+ 'itb7 O.O.Q. White's main alternative is 13 b4,
etc.) 12 ... 1i'e8 13 1ie6+ rj;dS 14 es which will be covered in the next game.
'i'd7 15 tbc3 tbg4 1 6 1i'e2 1i'f5 17 'i'd2 1 3 g6!
..

li)e5 (both 17 ...g5 and 17 ...li)d7 are


worth considering) 18 0-0-0 li)bd7 19
c6 gave White dangerous compensation
for the piece in Kopec-Sowray, corre
spondence 1990-92.
The other move is 10...ltlb6 but this
also looks good for White aher 1 1 li)c3
es 12 f4+ rj;d7 13 1ie6+ d8 14
es d7 15 0-0-0 tlg4 (aher the pas
sive 15...li)e8 White plays 16 d3! llxl6
17 cS lllf7 18 1i'e3 gS 19 J.g3 J.g7 20
i.. bS!, followed by 21 l:lhe1, with an

44
Classical with 4 . dxe 5 5 lll x e5: 5 . . . lll d 7 and 5
. . . . . c6

Suddenly endangering White's queen 10 Wf7 lOeS 1 1 ..tf4 cS 1 2 l0c3


with the threat of 14....th6. White has also played the immediate
14 .i..xeS+ xeS 1 S d6 ..th6+ 1 6 12 b4 but this seems to give Black an
c2 WeB! 1 7 :IdS+ lOxdS 1 8 additional defensive possibility in
9xdS+ ci>f6 12 ......b6 13 c3 'irxb4 (13 ... a6 would
transpose back into the game).
Batakovs-Aleksandrovs, Latvia 1991
continued 14 l::tc 1 gS 15 .i.xe5+ xeS
16 .id3 'itd6 17 0-0, followed by 18
.D.fel, with a strong attack.
12 . . . a6 13 b4

1 9 ..td3
Rublevsky-Hauchard, Oakham 1992
varied from this with 19 t'lle4+ 'itg7 20
1i'e5+ f7 (and not 20...g8? 21 lilf6+)
21 .td3 (21 dxe7 is answered by
21 ... .tf5 22 .id3 1i'xe7 23 d6+ 'irxd6
24 'irxd6 llad8) 21.. ..ig7 (2 1.. ..tf5 22 1 3 . . .1llb61
g4 .ixe4 23 .i.xe4 e6 24 llel 'ira4+ 25 Counterattack proves to be the best
d3 llhe8 26 h4 0-1 was De Firmian form of defence. The cowering 13 ...b6
Rohde, US Ch. 1989; did Rublevsky is effectively answered by 14 llb1 g5
have an improvement ready?) 22 'iVf4+ (Bagirov pointed out that 14 ...cxb4 15
g8 23 llel 1Wf8 24 'irgS exd6 25 J:r.xb4 'itc5 is answered by 16 llb1!!
'iVd5+ 'iVf7 26 t'llxd6 1Wxd5 27 lle8+ liJxf7 17 liJa4+ 'itd4 18 f3 followed by
.tf8 28 cxdS c4 29 .txc4 b5 30 i.b3 19 lld1 mate) 1S .ig3 .th6 16 bxcS+
g7 3 1 t'llxc8 .ic5 32 l:le7+ 'it>h6 0-1. bxcS 17 llb7! .td7 18 .id3! 'ireS (after
1 9 ... exd6 20 9xd6+ f7 21 l0e4 18...J:r.b8 Wolff demonstrated that
9c6 22 9es :ds 0-1 White can win with 19 llxb8 1Wxb8 20
.----- liJb5+! axbS 21 .i.xe5+ xeS 22 1i'xe7+
Game ll Wd4 23 1We3+ Wc3 24 .i.e2+ b4 25
Mysliwiec-Krzyzanowski 'W'dl+ followed by mate) 19 .i.f5! 'ifxf7
Correspondence 1995 20 li:)e4+ t'llxe4 21 l:bd7 mate.
Keres pointed out that 13 ...cxb4?
1 e4 l0f6 2 eS lOdS 3 d4 d6 4 l0f3 Loses to 14 cS+ 'itxcS 15 t'lla4+ d6 16
dxeS S 4:\xeS lL\d7 6 l0xf7 xf7 7 t'llb2 b5 17 t'lld3 etc.
'irhS+ e6 8 c4 l0Sf6 9 dS+ d6 1 4 0-0-0

45
A lekhin e 's D e fence

At one time it was thought that 14 14. . . cxb4


bxc5+ 'ifxc5 15 J:.d1 was strong but this This is supposed to be an 'only move'
assessment was turned upside down by but the reason seems to be questionable
the game Wydrowski-Krzyzanowski, to say the least. After 14 .. .'i'xb4 (sup
correspondence 1993. Black uncorked posedly .?, Volzhin has given 15 c2,
the remarkable 15 ...'ifa3! 16 .U.d3 'ifcl+l but then just lS ....tfS+ 16 .i.d3 'ifxc4
17 .i.xcl liJxfl and went on to win after looks good for Black.
18 c5+ c7! (and not 18 ... xc5? 19 1 5 a4 11'xf2 1 6 c5+
ltla4+ d6 20 li)b6 1lb8 21 .tf4+ ltle5 Gipslis has suggested 16 i.xe5+
22 ltlc4+ etc.) 19 .i.f4+ d8 20 ltla4 xeS 17 i.d3, when White intends to
.tfs 2 1 l:td4 c;;Pe8 22 .tc4 ltd8 23 0-0 bring his other rook into play.
.i.c2 24 ltlb6 eS 25 d6 .i.xd6 0-1. 1 6 . . .'d7 1 7 .i.xe5
White has also played 14 J:.cl but this After 17 c6+ Burgess gives 17...bxc6
can be met by 14 ...g5 15 .i.g3 {15 {17...d8? 18 c7+) 18 dxc6+ xc6 19
.i.xe5+ xeS 16 i.d3 would be an i.xeS 1i'e3+ 20 J:.d2 'i'e1+ 21 ltdl
swered by 16 ...cxb4) 15 ... h5 and now: 'i'e3+ with a draw. Actually, this would
a) 16 1i'g6 1i'xb4 17 'ifxg5 ltlfg4 18 have been White's best option; after the
'irg6+ d7 19 i..xe5 ltlxe5 20 'ire6+ move chosen Black's king finds safety.
'ittd8 21 'it'xeS J:.g8 left White facing the 1 7 11'e3+ 1 8 Wb1 1fxe5 19 b6+
..

threat of 22 ....tg7 in Elburg-Krantz, Wc7 20 d6+


correspondence 1991. After 20 ltlxa8+ b8 21 ltlb6 .i.g41
b) Szilagyi-Krantz, correspondence 22 ltlc4 'ttc3 23 i.d3 .i.xdl 24 J:.xdl
1992 continued 16 bxc5+ 'ifxc5 17 1!i'g6 'ifd4 Black is winning.
..d4 18 cS+ q;c7 19 1i'xg5 ltlfd7 20 c6 20 . . .Wb8 21 d7 .i.xd7 22 l:lxd7 Wa7
bxc6 21 dxc6 .i.h6 and now White
should probably play 22 lDds+ 'ifxd5 23
cxd7+ xd7 24 :C7+! q;xc7 25 1i'xe5+
'ifxeS+ 26 i..xe5+ 'ittc6 27 .i.xh8 i.e6,
when Black wins his pawn back with a
likely draw.

Made it! Now it's White king that


comes under pressure; he's standing on
b1 with the wind in his hair.
23 lld1 e4 24 xa8 c3+ 25
Wc1 xd1 26 Wxd1 'irxc5 27 .i.d3
Wxa8

46
Classical with 4 . . . dxe5 5 li:Jxe5: 5 . . .ll:J d 7 and 5 . . . c 6

Black has three extra pawns and dxe5 ltld7 10 0-0-0, all of which is very
White's king is still vulnerable. The rest impressive if you play bad moves for
requires little comment. Black!
28 l:tf1 e5 29 l:te1 e7 30 Wxg7 b) Tal-Larsen, Bled (6th matchgame)
lidS 31 e2 Wd5! 32 l:td 1 continued 6...1Wf6 7 1Wg3 h6 8 ltlc3
Black has calculated that 32 'i'xe7 is ltJb4 (here Black should consider
met by 32 ...'ifxg2+ 33 e3 'iih 3+ 34 8 ...i..b4 9 i..d2 ltJd7 which seems to me
f2 'ifxh2+ 35 fl l1g8 36 i.g6 11ff4+ to be quite playable} 9 .ibS+ c6 10 .ia4
37 e2 lbg6 with the win in sight. ltJd7 1 1 0-0 (Tal suggested that 1 1 ltJe4
32 . . ..i.c5 0-1 1if5 12 f3 ltlxe5 1 3 dxe5 .id7 14 a3
...-----. would also be good for White, which

Game 18 seems true enough at this stage}


Van der Weide-Miles 1 1 ...lL\xe5 12 dxe5 1Wg6 13 'iff3 1Wf5 14
European Ch., Saint Vincent 2000 'ife2 JJ..e7 15 a3 li:)dS 16 li:)b5! cxb5 17
'ifxb5+ d8 18 c4 and Tal had con
1 e4 lbf6 2 e5 tl:Jd5 3 d4 d6 4 lbf3 jured up one of his magical attacks.
dxe5 5 lbxe5 c6 6 c4
Arguably White's most natural move.
6 1Wf3 .ie6 7 .ic4 llJd7 8 lLid3 g6 9 0-0
.ig7 10 c3 0-0 1 1 li:)d2 was Rabiega
Loffler, Austrian Ch., Melk 1999, and
now 1 1 ... a5!? looks like Black's most
interesting plan.
6 tl:Jd7
. .

Miles has favoured this move though


Black has also tried the slightly clunky
looking 6 ... .ie6.
7 tl:Jf3
Exchanging knights abandons any
Tony Miles has been playing this a hope of obtaining an opening advan
lot; it's a very reasonable move which tage. The game Cao-Miles, European
theory has virtually ignored Rather than Ch., Ohrid 2001 continued 7 lLixd7
challenge White's knight immediately, .ixd7 8 0-0 .ifS 9 .id3 (or 9 c3 e6 10
Black first protects the horse on d5. 'ife2 J.d6 11 lL\d2 0-0 lh-lh, Kobalija
This precludes the possibility of a sacri Miles, Capablanca Memorial, Varadero
fice on f7. 2000) 9 ...i..xd3 10 ..xd3 e6 1 1 c4 ltJE6
Bent Larsen also experimented with 12 lL\d2 J.e7 13 ltJf3 0-0 14 J.d2 'ifc7
5 ... e6 in his 1965 match against Tal in 15 llfel llad8 16 1i'e2 llfe8 17 .ic3
Bled. After 6 11ff3 we have: lh -\.7.
a) 6...lLif6 was dismissed by the varia 7 ...tl:J7b6
tion 7 i.e3 lL\bd7 8 lL\c3 lL\xeS? (both Intending to develop his bishop on
8... .i.b4 and 8 ...i.d6 are better tries) 9 g4. 7 . . .ltJ7f6 is not as good after 8 h3,

47
A lekhin e 's Defen c e

but 7...g6 is worth considering. 31 b6 32 Wc4 b 5 3 3 Wb3 i!)d5 34


. .

8 .i.b3 .i.g4 9 lbbd2 e6 1 0 0-0 .i.e7 i!)e3 Wd7


1 1 h3 .i.h5 1 2 c3 0-0 1 3 tbe4 .i.g6
1 4 'ife2 tbd7 1 5 l:.e1 a5 16 a4 J:l.e8

35 l:.a1 ?
Losing. White had to play 35 tbxdS
Black has a very comfortable posi but after 35 ...cxd5 Black is obviously
tion, which sums things up from a theo better because of the weakness of the
retical point of view. Over the next few backward c3-pawn and Black's kingside
moves Miles aims to free his position pawn majority.
with ...e6-e5, this being one of Black's 35 l:.g6 36 h1
.

two major pawn levers in such struc If White had chosen 36 <iftfl Black
tures {the other is ...c6-c5). would launch a similar attack to the
1 7 .i.c4 'ifc7 18 .i.g5 .i.xg5 1 9 game with 36...l2Jf4 37 llxa4 bxa4 38
i!)exg5 h6 20 tbe4 J:l.ad8 2 1 l:.ad1 ?! 1i'xa4 1lxg2! 39 l2Jxg2 1i'xh3 etc.
After this Black is definitely better. 36 ...tbf4 37 h2
White should probably play 2 1 llJed2, After 37 .z:tgl there follows 37...tbxh3
applying some restraint to Black's im so White no longer has a reasonable
pending ... e6-e5. defence.
21 i!)5b6 22 .i.b3 e5 23 tbg3
..

After 23 dxeS llJxeS 24 lbd8 exd8


White would find Black's e-file pressure
quite unpleasant.
23 e4 24 tbh4 .i.h7 25 'ifh5 i!)f6
.

26 Wxa5 :as 27 'ifc5


After 27 1i'b4 there follows 27 ...llJfdS
28 .ixdS llJxdS 29 1i'b3 1i'f4, hitting
the knight on h4.
27 tbxa4 28 .i.xa4 l:.xa4 29 i!)hf5
..

.i.xf5 30 tbxf5 l:.e6 31 b4


Black was ready to meet 31 d5 with
31...l2Jxd5 32 l:r.xdS b6 etc. 37 ...l:.xg2+ 0-1

48
Classical with 4 . . . dx e 5 5 t'iJxe5: 5 . . . li:J d 7 and 5 . . . c 6

c) 6 d3 d7 7 d7 xd7 8 0-0
Game 19 g6 9 lle1 g7 10 c3 0-0 1 1 gS lle8 12
Tiviakov-Van der Werf d2 ti'c8 13 e4 f5 left White with
Dutch Ch., Leeuwarden 2001 nothing special in Macieja-Kuczynski,
Polish Ch., Warsaw 2001.
1 e4 o!bf6 2 e5 o!bd5 3 d4 d6 4 o!bf3 6 ..o!bd7
.

dxe5 5 o!bxe5 c6 6 o!bd2

7 lbdf3
Aiming to maintain his knight on eS, 7 d3 would avoid the exchange and
in the hope that this will guarantee a the knight is placed on a square from
slight space advantage. The drawback is which it inhibits either the ... c6-c5 or
that Black gets to exchange a pair of ... e7-e5 pawn levers. On the other hand,
minor pieces which does relieve his po it stops White's bishop coming out to
sition. its most natural posts on either d3 or
Other possibilities are as follows: c4. Black's best is probably 7... g6, after
a) 6 c3 g6 7 g3 g7 8 g2 d7 9 which 8 e2 g7 9 f3 0-0 10 0-0 b6,
xd7 xd7 1 0 d2 0-0 1 1 f3 1i'c8 intending ...c6-c5, looks about equal.
12 0-0 h3 left Black very comfortably 7 ...ltlxe5 8 o!bxe5 g6 9 Wf3 .i.e& 1 0
placed in Shur-Bagirov, Baku 1996. c3 .i.g7 1 1 e2 a5
b) 6 h3 looks innocuous but the idea One of Black's most common plans,
is quite good; when Black challenges the staking out some territory on the
knight on e5 White plans to drop it queenside. In the game Dutreeuw
back and not have to worry about a pin. Loeffle, Tanta City 2001 Black played
Jansa-Burkar, Bonnevoie 1999 contin 1 1...0-0 12 d3 f6 13 'i'g3 c4 with
ued 6 ... d7 7 f3 g6 8 c4 g7 9 0-0 a solid position.
0-0 10 llel e6 (the immediate 10 ... b5 1 2 0-0 0-0 1 3 o!bd3 a4 14 a3 f5
loses a pawn after 1 1 xd5 cxdS 12 1 5 .l:te1 xd3
3) 1 1 b3 b5 12 g5 eel 13 bd2 Giving up the bishop pair to elimi
b7 14 c4 bxc4 15 xc4 cS 16 l:lct nate the knight that controls the e5 and
and now 16 .. Jlac8 17 dxc5 xeS looks c5 squares. In the following play White
fine for Black. tries to keep Black tied down whilst

49
A lekhin e 's Defen ce

gradually trying to weaken Black's king 'ifh2 41 .i.f4 'ifh8 42 J:td2 e7 43


position. Black wants to free his game .i.xf5 l%xf5 44 .i.d6+ WeB 45 'ifg2
with an ultimate ...c6-c5 or e6-e5, but
.. J:td5 46 J:txd5 exd5 47 'ifxd5 'ii'h3+
without allowing White's bishops to 48 e1 We&+ 49 Wxe6+ fxe6 50
become too dangerous. White is slightly We2 Wd7 51 .i.fB we& 52 Wd3 1 -0
better, though it's not very much.
1 6 .i.xd3 e6 1 7 g3 'ii'b6 Game 20
17 ... b5 looks preferable, cementing Ljubicic-Zelcic
the position of the knight on dS before Croatian Ch., Slavonski Brod 1995
preparing ... c6-c5.
1 e4 lbf6 2 e5 lbd5 3 d4 d6 4 lbf3
e6
4 ...dxe5 5 liJxeS c6 6 J.e2 would
transpose back into the game. For more
about the independent aspects of 4 ... c6,
see Chapter 4.
5 .i.e2 dxe5 6 lbxe5 lbd7 7 lbxd7
This gives White nothing, though the
way Black creates unbalance in the posi
tion is interesting. For 7 liJf3 see the
next two games.
7 ... .i.xd7 8 0-0 .i.fS 9 a3 'ife7 10 e4
1 8 h4 h5 lbf6 1 1 lbe3 e6 1 2 .i.e3 h51
I think that Black should avoid this The start of an old-fashioned attack
weakening and instead focus on playing on the kingside. The immediate threat is
... c6-c5. However, the immediate 18 ...c5 13 ...liJg4.
is bad because of 19 c4 liJe7 (19 ...liJf6 1 3 h3 .i.d6 14 f4?! 0-0-0 15 'ifa4
20 dxcS 'it'xcS 2 1 'it'xb7) 20 dS. All of g5
this supports the case for 17 ... b5.
19 g2 J:tfe8 20 J:te2 lbe7 21 .i.gS
lbf5 22 :ae1 :as 23 g4
Starting the process of prising open
Black's kingside. This was definitely
made easier by Black's 18th move.
23 ... hxg4 24 'ii'xg4 'ife7 25 l:tg1
'ifd7 26 .i.e2 J:tea8 27 1 e5 28
dxe5 J:txe5 29 h5 gxh5 30 'iff3 f8
31 .i.d2 l:ld5 32 .i.e1 :ca?
Overlooking White's 34th move.
33 'ifg2 .i.f6 34 .i.xa4! b5 35 .i.e2
J:te4 36 'iff3 J:tg4 37 J:txg4 hxg4 38 This surprising blow rips open files
'ifxg4 'ife7 39 .i.e4 :es 40 .i.d2 on the kingside. Black does not bother

50
Classical with 4 . . . dx e5 5 li::J x e5: 5 . . . li::J d 7 and 5 . . . c6

to defend a7. lld7 Wf4 37 d6


1 6 d5
White in tum plays for his own attack
and the race is on. Capturing on gS
would open the flood gates; aher 16
fxgS there follows 16 ...lbg4! 17 hxg4
i.h2+ 18 h1 (18 'iPf2 1i'g3 mate)
18 ... hxg4 threatening a lethal discovered
check.
1 6 ... xf4 1 7 xf4 gxf4 18 dxc6
bxc6 19 f3 l:r.d2

37 ...f5 38 f7 g3 0-1

Game 21
Degraeve-Miles
Mondariz Zonal 2000

1 e4 f6 2 e5 d5 3 d4 d6 4 f3
dxe5 5 xe5 c6 6 e2 d7
Black does have an alternative here in
6 ...g6, after which 7 c4 lbc7 8 c3 (8
20 J:r.ad 1 ? .ie3 i.g7 9 lbd2? cS! was good for
This looks like a case of desperation. Black in Gi.Garcia-Miles, Matanzas
White should go ahead with 20 ilxc6. 1994) 8...i.g7 9 i.e3 lbd7 10 f4!? 0-0 1 1
Then 20.. .f3 21 i.xf3 lbg4 22 i.xg4? 0-0 tbxe5 1 2 dxe5 (12 fxe5 c5)
hxg4 23 lbb5? ...c5+ 24 h 1 1hh3+ 25 12 ...1i'xd1 13 J:laxd1 f6 was equal in
gxh3 ..i.e4+ is mating, but 22 1i'a6+ clS Dolmatov-Tseshkovsky, Russian Ch.,
23 i.xg4 hxg4 24 J:lad1! is totally un Tomsk 2001 .
clear. 7 f3
20 .....b6+ 21 'ith1 c2 Keeping pieces on the board makes
Winning material. White gives up the sense because White has more space.
queen in the hope that there's some 7 ...g6 8 c4 c7 9 c3 g7 10 0-0
compensation. But it never looks any 0-0 1 1 i.f4
thing like enough. Varying from De Firmian-Miles, Chi
22 J:r.xd2 xa4 23 xa4 as 24 cago 1994, in which Black obtained
c3 J:r.d8 25 l:txd8+ xd8 26 l:r.d 1 good counterplay aher 1 1 J:lel c5 12 d5
b6 27 l:r.d6 xb2 28 J:r.xc6+ d8 b5! 13 cxb5 lbb6, when Black recovers
29 b5 d7 30 xa7 e5 31 the dS-pawn with a good game. Of
l:r.d6+ e7 32 c8+ Wf6 33 xh5 White's other moves, Black would
f3 34 xf3 xf3 35 gxf3 e5 36 probably meet 1 1 .te3 with 1 1 ..e5 and.

51
A lekhin e 's D e fen c e

11 .igS with 1 1 ...c5 12 dS h6, and if 13 15 a3 .tts 16 :b1


J.h4 then 13 ...g5 14 J..g3 eS etc. White could also consider the imme
diate 16 b4!?, after which 16 ... cxb4 17
axb4 J.xb4 18 ltle4 gives him some
initiative for the pawn.
1 6 .../tJeS 1 7 b4 b6 18 2
Another possibility was 18 ltle4 but
then 18 ...ltld6 (and not 18...f5 t9 ltleg5)
19 ltlxd6 .ixd6 looks fine for Black.
18 ...liJd6 19 de4 f5
Heading for the d4-square. White de
cides that it's now or never.
20 d6 -*.b7
Miles in turn has to switch plans and
1 1 ... c5!? 1 2 d5 a short tactical skirmish ensues. Follow
In the next game we will see 12 dxcS. ing 20...ltld4 White would immediately
1 2 ...e5!? try to undermine the position of the
Obtaining a kingside pawn majority knight with 21 f4, after which 2 1...f5
but giving White a passed d-pawn. (21....ib7 22 fxeS) is still met by 22
Black will try to blockade the passed ltlgS etc.
pawn by bringing his knight to d6, and 21 bxc5 xe3 22 fxe3 f5 23 c6!
meanwhile White can play to open files .i.xc6
on the queenside. In shon, it will be a
full-blooded and double-edged game.
1 3 .tg5 f6 14 .i.e3 :t71

24 c5!
With White threatening to bring his
bishop to c4, the struggle reaches its
A very deep move. Miles envisages climax.
that he must hold on to cS, so he pre 24 .:g7 25 .i.b5 cs
..

pares to bring his bishop to f8. The Black cannot accept the offer of a
immediate 14 ...ltle8 is met by 15 ltle4 piece as after 25 ... .ixe4 26 ltlxe4 fxe4
b6 16 a3, followed by b2-b4. there follows 27 1Wd5+ h8 28 c6 ltlf6

52
Classical with 4 . . . dxe 5 5 li::J x e5: 5 . . JiJ d 7 a n d 5 . . c6

{or 2 8...lbc5 2 9 c 7) 29 llxf6 11fxf6 30 d7 ening 16 xeS.


11fd6 3 1 1ld1 etc. 14 bxc3 dxc5 1 5 /Od4 .i.d7 1 6 f4
After 25 ... 11fc8 the complications fi e4
nally fizzle out into a draw. Baburin pointed out that 16 .. .f5 was
26 ..b3+ Wh8 27 g5 xc5 28 possible both here and on his next
f7 + Wg8 29 h6+ Wh8 30 t7 + move.
*g8 31 6+ % -% 1 7 .if3 6c5 18 b3 l:tc8 19 d4!

Game 22
W.Watson-Baburin
Kilkenny 1994

1 e4 f6 2 e5 d5 3 d4 d6 4 f3
c6
After a brief detour, we soon return
to the paths of the previous game.
5 .A.e2 dxe5 6 xe5 d7 7 f3 g6
8 0-0 .A.g7 9 c4 c7 1 0 c3 0-0
1 1 .A.t4 c5 1 2 dxc5
Suddenly making it very difficult for
Black to maintain the position of his
knights. He only manages to do so with
some clever tactical play.
1 9 ....ic6! 20 f51?
Every move increases the tension.
After 20 ll)xcS ll)xcS 21 11'xd8 {or 21
WxcS .ixf3} 21...1lfxd8 22 xeS .ixf3
23 xe7 lld7 the game fizzles out into
a draw, whilst after 20 lladl Black de
fends himself with 20...b6.

A totally different treatment to the 12


d5 we saw in the last game. White
opens up the centre and hopes to use
his advantage in development to trouble
Black.
1 2 l086 1 3 .A.e3 .A.xc3!
.

It looks risky to give up this bishop,


but Black wants to secure cS for his
knight. After 13 ...xc5 14 1l'c2 White
threatens b2-b4, and preventing this
with 14 ... a5 is met by 15 1ladl, threat-

53
A lekhin e 's D e fen c e

20 . . J:le8
Black gets out of the way of the Game 23
threatened 2 1 h6. According to Ba Short-Miles
burin, 20 .. .f6? would have been bad European Ch., Ohrid 2001
because of 21 fxg6 hxg6 22 lLlxc5 lLlxc5
23 1i'xd8! {23 9g4 g7! 24 xeS xf3 1 e4 tL!f6 2 e5 tL!d5 3 d4 d6 4 tL!f3
allows Black to defend) 23 ... 1lfxd8 24 dxe5 5 tL!xe5 c6 6 c4 tL!c7
J.xcS xf3 25 xe7 lld7 26 l:lxf3
In Adams-Burkart, European Club
l:lxe7 27 .llxf6 with a winning endgame. Cup, London 1996, even Michael Ad
But there was another possibility in ams failed to get much against the na
20...1i'b6!?. ive-looking 6...lLlb4. The game went 7
21 .i.h6?! e5 22 fxe6 tL!xe6 23 e3 {7 a3? 9xd4} 7...J.f5 8 lba3 lLld7
9xa7 9h4! 9 lLlxd7 li'xd7 10 J.e2 g6 {10 ...e6 also
Suddenly finding a way to exploit the looks very reasonable) 1 1 dS g7 12
looseness of White's pieces. In time d4 lLlc2 + 13 li)xc2 J.xc2 14 1!6'd2
trouble Watson does not find the best .ixd4 15 1i'xd4 0-0 16 0-0 cxdS 17 cxdS
defence. and now the immediate 17...1i'd6 would
have left things fairly even.
7 tL!c3 lbd7

24 9e3?
24 e3 was mandatory. Now Black
is winning and doesn't give White any 8 tL!xd7
further chances. It's not clear that this is White's best.
24 ... tLI6g5! 25 .i.xg5 tLlxg5 26 9d4 In Apicella-Miles, Linares Zonal 1995,
tL!xf3+ 27 gxf3 9g5+ 28 9g4 White preserved his knight from ex
9e3+ 29 g2 9xc3 change with 8 lLlf3 g6 9 .ie2 J.g7 10
Black re-established material parity 0-0 0-0 1 1 .ie3 a6 {l t...lLlf6 12 h3
but now has an overwhelming posi lLlce8 13 b3 lL!d6 14 .lladl was un
tional advantage. comfortable for Black in Nunn-Ghinda,
30 J:tf2 J:tcd8 31 llc1 9e3 32 J:tcf1 Hamburg 1984 whilst l l...e5 leaves
J:te5 33 h4 h5 34 9g3 J:tf5 35 g1 ? White better after 12 d5 cxdS 13 cxdS
J:txf3 0-1 lLlb6 14 J.cS l:le8 15 d6 lLle6 16 a3.

54
Classical with 4 . . . dx e5 5 l0xe5: 5 . . . l0d 7 and 5 . . . c6

according to N unn) 12 a4 and now this move, but Black reacts with a pow
Black should probably play 12 ..a5 fol
. erful exchange sacrifice. 13 cxbS cxbS
lowed by bringing his c7-knight to b4 14 lDxbS can be met by 14 ... lDd5 (or
via a6. 14 ....tb4+) when White's king would
8 ... .i.xd7 9 .i.e3 .i.f5 have difficulty finding a safe haven
anywhere.
1 3 . .. bxc4 14 .i.xc6+ .i.d7 1 5 1la4
Snatching the material with 15 .txa8
'l'xa8 would have given Black excellent
compensation for the exchange, but
now he is simply better at zero cost.
Had Short missed Black's next move?
1 5 ...:ca! 16 11rxc4 d51 1 7 .i.xd7+
1Wxd7 18 1lrd3 .i.b4 19 .i.d2 0-0 20
0-0 .i.xc3 21 bxc3 1Wa4
In this position Black has more than
enough for his pawn; he can firmly
10 g4?! blockade White's c- and d-pawns on d5
Short loves to throw his pawns for and c4, and in addition White's kingside
wards like this, though very few other has a gaping hole in it.
grandmasters would have played this 22 f4
way. On this occasion Miles makes it
look rather dubious.
1 0 ...-*.cB 1 1 .i.e2 e6 1 2 1fc2 b5!
Well played! With White having cre
ated weaknesses, Miles reasons that this
thematic pawn sacrifice will be stronger
than usual.
1 3 .i.f3

22 ...f5!?
I don't especially like this move
which creates some weaknesses in
Black's own camp. My own preference
would be for 22 ...:c6 intending ...:fc8.
Black has very strong pressure on c3
and a2, and should White try to open
the f-file with f+fS, he would activate
Perhaps Short had been relying on the black rook along the third rank.

55
A lekhin e 's Defen c e

The other move that deserved con


sideration was the immediate 22 ...11fc4,
for example 23 .xc4 {23 ltf3 is better)
23 ...ltxc4 24 l:lf3 l:.fc8 25 llcl l:la4 26
llc2 :Xd4 and Black certainly stands
better.
23 gxf5 exf5 24 ltf3 -.e4 25 a4
xd3 26 llxd3 llc4 27 a5
There has been a turn around and
suddenly the position looks dangerous
for Black. Should White mop up the a
pawn, the pawn on a5 would be very
menacing. Meanwhile, Black's pieces are 2S. ..f6 29 d5 e4 30 .i.e3 l:lb7
tied down to the blockade of the passed 31 a6 l:ld7 32 llb1 6 33 llb4
pawns on d4 and c3. Ilea 34 .i.d4 W 35 lld1 wea 36
27 ...1lb8 2S lle1 l:le1 + Wd8 37 lle6 8 3S e4 d6
Threatening to land on e5. The re 39 .i.xa7 xe4 40 llxe4 l:lxa7 41
mainder of the game looks as if it was l:lxeS+ Wxe8 42 Wf2 Wd7 43 e3
most likely affected by a bout of time :as 44 Wd4 llbS 45 We5 llb2 46
trouble. a7 lla2 47 Wb6 1 -0

56
Classical with 4 . . . dx e5 5 ltJxe5: 5 . . . CiJ d 7 and 5 . . . c 6

Summary
Inviting a sacrifice on f7 (with S ...lbdl) is cenainly not for those of a nervous dis
position. White gets a very dangerous attack, but then again there is nothing clear.
This makes it into a very interesting weapon for Black, especially against stronger
players.
For anyone playing White against 5. .., I would actually recommend that you
withdraw the knight to f3. There's no point getting embroiled in complications that
your opponent has probably studied. Keep the space and play quickly.
Tony Miles's S...c6 is a very tough and interesting line for Black, which has so far
been linle explored by theory. On the next move Black can challenge White's knight
on eS without fearing a sacrifice on f7. Depending on how White plays it, Black has
the option of a kingside fianchetto.
The lines in this chapter constitute an excellent way for Black to play against the
Classical. For extra variety practical players might want to switch between 4...c6
(Chapter Four) and Miles's 4 ...dxe5 S ltlxeS c6. In these days of huge computer da
tabases it is useful to keep your opponent on his toes.

1 e4 l'tlf6 2 e5 l'tld5 3 d4 d6 4 l'tlf3 dxe5 5 l'tlxe5 c6 (D)


s ...ltld7 6 ltlxf7 xf7 7 1i'h5+ 'ite6 8 c4 ltl5f6 9 dS+ d6 10 'i'f7 ltle5 1 1
J.f4 cS 1 2 l0c3 a6 (D)
13 b4 - Game 17
13 0-0-0 - Game 16
6 .i.e2
6 ltld2 - Game 19
6 c4 - Game 23
6 J.c4 - Game 18
6 lbd7 7 l'tlf3
.

7 ltlxd7 - Game 20
7...g6 8 0..() ..tg7 9 c4 l'tlc7 10 l0c3 0-0 1 1 .tf4 c5 (D) 1 2 d5
12 dxcS - Game 22
1 2 e5 1 3 .tg5 f6 - Game 21
.

5. . . c6 1 2. . . a6 1 1 . . . c5

57
CIIAPTER FOUR I
Classical Divergences :
4 c6, 4 liJc6 and 4 g6
. . . . . . . . .

1 e4 lllf6 2 e5 llld 5 3 d4 d6 4 lllf3 Games 27 and 28 feature the ambi


In this Chapter we will consider some tious but risky 4 ... lL!c6. This certainly
of Black's more uncommon ideas does the job of challenging White's cen
against the Classical. tre but the drawback is that it allows a
Some readers may object to the fact dangerous pawn sacrifice with 5 c4 lL!b6
that the Alburt Variation (with 4...g6) is 6 e6!?. The correspondence player Mar
represented only by Games 24 to 26, cinkiewicz has discovered many im
though this really reflects the declining provements for Black against the stan
popularity of 4...g6, at least in conjunc dard lines, but my concern is that the
tion with 5 .i.c4 lL!b6. Although the little-known 7 .id3 is far more danger
bishop on g7 puts pressure on e5, it ous than is generally thought. Alekhine
does seem that White can maintain this himself thought highly of this and the
pawn and its cramping effect on Black's existing theory relies heavily on White
game. In the line 5 .ic4 lL!b6 6 .ib3 not making the most of his chances.
.i.g7 7 a4, the exciting 7...dxe5 looks In Game 29 we look at 4 ... c6, which
rather dubious for Black because of the supports the knight on d5 (in anticipa
improvements pointed out within tion of 5 .ic4) and vacates the c7-
Game 24. The alternative {7.. a5) looks
. square in case of White playing c2-c4.
rather passive and even Howell's ingen The knight is probably better placed
ious plan in Game 26 looks better for there than on b6, especially if White
White. In any case, whether or not later exchanges pawns in the centre with
Black manages to improve on these two exd6. The point is that it can often
games, he still has quite a passive game come to e6, from where it pressurises
after 7 'ire2. Maybe Black can keep the vulnerable d4-pawn.
things more or less equal if he defends The drawback to 4 ... c6 is that it is
very accurately, but for me the Alekhine relatively passive and does not immedi
should be all about counterplay. ately increase the pressure on e5. It fol-

58
Classical Divergences: 4 . . . c 6, 4 . . .li:J c 6 a n d 4 . . . g6

lows that the logical thing for White to been quite good for Black.
do would be to maintain this pawn and Black should also take account of the
it does in fact seem that White is better fact that White can transpose into an
after 5 .i.e2 dxe5 6 dxe5. Exchange Variation with 5 c4 tl.\b6 6
,-------. exd6 cxd6, though in this case his
Game 24 knight is already committed to f3 which
Oral-De Firmian precludes some of the more dangerous
Reykjavik 2000 set-ups.
.___________.. 5 i0b6
..

1 e4 i0f6 2 e5 i0d5 3 d4 d6 4 i0f3


g6

5 . . .c6 leads to positions very similar


to the important Kengis Variation and it
Putting the bishop on g7 is very logi can transpose directly if Black plays
cal as it helps to support undermining .dxe5 at a later stage and White recap
operations against White's d4 and e5 tures with the knight. The independent
pawn chain. The American grandmaster lines occur when White captures on d6
Lev AIburt did most of the spadework after 6 Q.O .i.g7 7 exd6 xd6, for ex
on this line so it rightly carries his name. ample 8 tl.\bd2 .i.g4 9 lte1 0-0 10 h3
5 i.c4 .i.xf3 1 1 tl.\xf3 e6 (lt. .tl.\d7) 12 .i.fl
.

5 tl.\g5 is not much more than a (White has also tried 12 .i.g5 but the
cheap trap which is well met by 5 ...c6 quiet text move seems to make more
(but not 5... h6 6 tl.\xf7 xf7 7 ..f3+) sense} 12 ... b5 {in order to sidestep any
and now: possible preparation, Vaganian varied
a) 6 c4 tl.\c7 7 ..f3 f6 8 exf6 exf6 9 with 12 ...c5 against Tischbierek in the
'ife3+ .i.e7 10 tl.\f3 0-0 was very com 1994/95 Bundesliga season and
fortable for Black in Ljubojevic-Tal, achieved a satisfactory position after 13
World Cup, Brussels 1988. c4 l'il.e7 14 dxc5 'W'xc5 15 llbl 'Wc7 16
b) 6 f4 .i.g7 7 .i.c4 0-0 8 0-0 tl.\a6 9 b4 li:lbc6 17 ..c2 li:lf5 18 c5 1lfd8 etc.)
.i.b3 tl.\ac7 10 c4 li:lb6 1 1 li:lc3 dxe5 12 13 a4 b4 1 4 c4 bxc3 15 bxc3 11fc7 16
fxe5 was Kosikov-Khmelnitsky, Kiev .i.a3 ltd8 17 c4 li:le7 18 lta2 a5 1 9 ltd2
1989, and now just 12 . .f6! would have
. li:!a6 20 g3 li:lb4 2 1 .i.g2 l:tab8 22 'We2

59
A lekhin e 's D e fen c e

fS 2 3 :edt cS and Black had ob a better version of this because of his


tained a fully equal game in lvanchuk cramping pawn on aS.
Vaganian, Manila Olympiad 1992. 9 .i.xf7+!
6 .i.b3 .i.g7 7 a4 The only serious attempt to gain the
Other 7th move alternatives for advantage. 9 'ife2 0-0 10 dxeS cS is
White will be examined within the notes very comfortable for Black whilst both
to Game 2S (Leko-Timman). 9 dxeS xeS 10 1Wxd8+ cRxd8 1 1 /l)gs
e8 12 f4 h6 and 9 gS e6 10 .i.xe6
0-0! 1 1 0-0 exd4 leave White a pawn
down with no compensation.
9 .'xf7 10 g5+ g8 1 1 IC!e&
..

'8'e8 1 2 xc7

7 .dxe5!?
..

A razor-sharp move which leads to


wild complications. For the relatively
sedate 7 ..aS see Games 2S-26.
.

8 a5
After 8 xeS .ixeS! 9 dxeS 'ifxdl+ 1 2 .".d8!
.

10 xdl , I think that 10....i.g4+ Lev Alburt's move, which stops


(10 ... c6 1 1 a5 d7 12 e6 fxe6 13 White's knight getting back from aS.
.txe6 deS is what Black has played so 12 ...1Wf7 1 3 a8 exd4 14 0-0 6 1 S
far, and it also gives very decent play) 1 1 c 3 allows White t o conduct a successful
f3 .ie6 1 2 .ixe6 fxe6 is a good idea, rescue.
when Black has a nice outpost for his 1 3 xa8
knight on d5 or a hole on b4, should From a practical point of view this
White play a later c2-c4. The doubled e line has a serious drawback for Black in
pawns should not be a problem in this that White can draw by repetition with
position. 13 e6 1i'e8 14 C&7 etc.
8 1C!6d7
.. 1 3 exd4 1 4 c3
..

Allowing the following sacrifice, This has been Whites knee-jerk reac
which Alburt has worked hard to show tion, trying to get play before Black
is just about okay for Black. After wins his knight. But it is starting to look
8 ... dS 9 xeS we get a position simi as if White can play less nervously with
lar to the Kengis Variation (4...dxe5 5 14 0-0! tbc6 15 c3 c5 16 b4 and now:
eS g6 6 .ic4) but with White having a) After 16 ...6?! Volzhin gives 17

60
Classical Divergences: 4 . . . c 6, 4 . . . Q)c6 and 4 . . . g6

Wb3 dxc3 18 lLlxc3 lLlcd4 19 Wc4 .i.d7 Christiansen-Alburt, US Ch.,Jackson


20 .i.e3 lLlf5 (20...Wxa8 21 .i.xd4 ..ixd4 ville 1990 in which White ended up in
22 .J:ladl! won quickly in Gubanov serious trouble after 16 0-0 e5 (Black
Romanovich, Chigorin Memorial 1995) can also consider both 16 lbc6 and ..

21 llad1 l0d6 22 Wb3 .i.xc3 (22 ... Wxa8 16...lLlba6) 17 .i.e3 (Christiansen later
23 lLlb5! .i.xb5 24 1lxd6 exd6 25 1i'xe6+ suggested 17 lLlc3 .i.5 18 lLlb5 i.d3 19
f8 26 xd6+ f7 27 .d5+) 23 .i.c5! lLlxd4 exd4 20 i.4 and claimed an ad
xa8 24 .i.xd6 exd6 25 l:txd6 We8 26 vantage for White) 17...lLlba6 (17...lLlc6
l:lfdl! .i.a4 27 lbe6 .i.xb3 28 l:txe8+ 18 .i.xd4 exd4 19 c2 lLle6 20 a6 'ifd5
f7 29 l:lxh8 .i.xdl 30 llb8 with White 21 lLld2 g7, as in Cooper-G Smith,
having winning chances in the endgame. Walsall 1992, is also interesting) 18 lLlc3
b) 16... dxc3 17 Wxd8+ lLlxd8 and .i.f5! and now:
now Volzhin gives 18 1la3! lLle4 19 lie1 a) The game continued 19 llcl g7!
c2 20 lLld2 lLlxd2 2 1 .i.xd2 .i.f6 22 lLlc7 20 ..ixd4 'i'xd4! 21 b4?! (White should
when Black has insufficient compensa play 2 1 tixd4 exd4 22 lLlb5 after which
tion for the exchange. In Z.Aimasi-de la 22 ... ltxa8 23 lLlxd4 <itr>6 leaves Black
Villa Garcia, Pamplona 1999 White only slightly better) 21...lLld3 22 lLle2
played less well with 18 bxc5?! which 1id7 23 .J:lc3 llxa8 24 g4 lLlb2! 25 'ifcl
gave Black good compensation for the .i.d3 26 'i'xb2 Wxg4+ 27 lLlg3 .i.xf1 28
exchange after 18 ... c2 19 lLlc3 i.xc3 20 xfl lld4 with a winning position.
lla3 (20 llal!?) 20...-i.6 21 lLlc7 lLlc6 b) According to Christiansen, 19
22 lLldS 7 etc. lLlb5! .i.d3 20 .i.xd4 exd4 21 b4 L:f1
1 4 llx:5
.. (21. ..lfub4 22 llct!) 22 bxc5 i.xb5 23
tib3+ g7 24 tixb5 'i'xa8 25 'it'd?+
leads to a draw by perpetual check.
These positions are clearly very com
plicated and the reader would be well
advised to take existing 'theory' with a
pinch of salt.
16 ...e5 1 7 .i.xd4 exd4 18 0-0 ba&

1 5 cxd4
The tempting 15 b4?! doesn't work
after 15 ...lLle4 16 Wb3+ e6 17 cxd4
lLlc6! intending 18 ....i.d7, as pointed out
by Dlugy.
1 5 ....i.xd4 1 6 .i.e3
An attempt to Improve on

61
A lekhin e 's Defen c e

Cutting off the a8 knight's exit White loses a piece aher 34 lth2
squares. But can this knight actually be 11'f4+ 35 g1 lLJxd5 36 ..d7+ ..f7 etc.
captured when White's other pieces
start coming into play? Game 25
1 9 lbd2 e6 20 %le1 f7 21 lbe4 Leko-Timman
d3 22 . b4 lbxe4 23 %lxe4 Wg7 24 Wijk aan Zee 1996
l:la3 d2 25 %le2 l:le8 26 l:lxe8 .i..xe8
27 l:la2 'iba8 28 l:lxd2 1 e4 lbf6 2 e5 lbd5 3 d4 d6 4 lbf3
The knight has been lost but White g6 5 c4 lbb6 6 b3 g7 7 We2
has a rook and pawn for the two minor
pieces. Add in the factor that Black's
king is exposed and White's prospects
are slightly preferable.
28 .....c8 29 %ld8 e6 30 h3
Perhaps 30 'ifd4+ was more precise.
Aher 30...g8 3 1 h3 Black can't take
the b4-pawn with 3 1...11t'e1+ 32 h2
lLJxb4 because of 33 'ifc4+ <iPg7 34
l:ld4! lllc6 35 lle4! 'ifxf2 36 ltxe8 etc.
Now Black's queen comes to a key cen
tral square.
30.....e5! 31 %ld5 'ii'c3 32 1ie2 A logical move which avoids the
lbxb4 complications of 7 a4 dxe5 (see Game
24}, whilst aiming to maintain his ad
vantage in space. Alburt has also had a
long-running battle against 7 lLJg5 e6
(7. . dS?! gives Black a cramped position
.

because he can no longer challenge the


e5-pawn} and now:
a) In the game Pupols-Alburt, Port
land 1987, Alburt defused 8 f4 with
8 dxe5 9 fxe5 c5 10 0-0 0-0 1 1 c3 cxd4
..

12 cxd4 lDc6 13 lDf3 f6 14 lDc3!?


(White's best try as 14 exf6 ..xf6 15
i.e3 lDd5 16 J.f2 lLJf4 produces a com
33 1ixe8? plex position in which both sides have
Blundering the game away. 33 11t'e7+ chances} 14 ... fxe5 15 i.g5 1i'd7 16 dxe5
f7 (33 ... h6?? 34 1i'f8+ 1i'g7 35 11ff4+ lLJxe5 17 lLJxe5 :Xf1+! 18 1i'xfl ..d4+
gS 36 11fxb4} 34 l:ld81 leaves Black with 19 h 1 'ii'xe5 20 i.e7 i.d7 (Alburt
nothing better than a clraw by perpetual mentioned that 20...1tc7 is an al
check aher 34 11fc1 + 35 h2 Wf4+ etc.
.. ternative) 21 .Uet ...f5 22 ..e2 1Z.e8 with
33 1ic 1 + ! 0-1
. a virtually even game.

62
Classical Divergences: 4 . . . c 6, 4 . . . /;jc6 and 4 . . . g6

b) 8 11f3 1i'e7 {8...11d7 is less good alternative approach for White is the
because of 9 lDe4 dxe5 10 dxeS .i.xe5 simple 7 exd6 but then 7...cxd6 8 0-0
1 1 Ci)f6+ i.xf6 12 11xf6 l:tg8 13 .h4 f5 0-0 9 lite 1 liJc6 10 h3 liJaS gets the
14 Cl)cJ, when White has excellent com bishop pair as compensation for Black's
pensation for his pawn; 8...0-0 is also lack of space.
uncomfortable for Black after 9 1i'h3 h6 7 0-0
.

10 Ci)f3 or even 10 lDe4) 9 Ci)e4 dxe5 10 The fact that the moves a2-a4 and
i.g5 'ti'b4+ 1 1 c3 1i'a5 12 liJf6+ {12 a7-a5 have not been played gives

i.f6 i.xf6 13 xf6 0-0 14 1i'xe5 9xe5 Black an additional possibility in 7...lbc6
15 dxe5 .i.d7 gives Black equality in the 8 0-0 dxe5 9 dxeS liJd4 10 lbxd4 9xd4.
endgame according to Alburt) 12 ...Wf8 Black's knight on b6 is better protected,
13 d5 e4 14 1i'g3 Ci)a6! (14... liJ8d7 15 White no longer has support for a piece
Ci)xd7 + Ci)xd7 16 dxe6 5 1 7 e7+ We8 on bS and the queenside is now quite
18 0-0! was very dangerous for Black in habitable for Black's king. Play can con
Short-Alburt, Foxboro { 1st match tinue 1 1 e6 (after 1 1 l:te1 .i.g4 12 9ft I
game} 1985} 15 dxe6 'iff5 (15 ...h6!? 16 think that 12 ... .i.e6 13 .i.xe6 fxe6 looks
e7+ Wxe7 17 Ci)dS+ e8 18 Ci)xc7+ fine for Black because of his active
Ci)xc7 19 11xc7 11xg5 20 11xf7+ d8 pieces) l l....i.xe6 12 .ixe6 fxe6 13
21 9xg7 l:te8 is also very interesting 9xe6 llc4 14 xc4 (14 11h3 0-0 was
and leh Black with compensation for also very comfortable for Black in Vo
the pawn in van der Wiel-Blees, Dutch robiov-Chekhov, Moscow 1999}
Ch., Hilversum 1990} 16 e7+ xe7 14 ...ltlxc4 15 c3 (if both sides had
moved their a-pawns White could de
fend b2 with :ta2 at this point}
15 ...0-0-0 and Black's active pieces fully
compensated for his marginally inferior
pawn structure, Dolmatov-Neckar,
Bern 1994.
8 h3 a5
After White's reply this transposes
back into the 7 a4 line. Black can also
play 8...lbc6, after which 9 0-0 liJaS 10
lbc3 actually transposes into a line of
the Pirc Defence. Kveinys-Speelman,
Moscow Olympiad 1994 continued
17 Ci)d5+ (17 li)g8+ e8 18 li)h6 . 10 ... h6!? l t liJe4 ltlxb3 12 axb3 f6!? 13
i.xh6 19 .txh6 .i.e6 20 0-0 f6 leaves c4 fxe5 14 dxe5 lbd7 with complex play
White with some compensation for his in prospect.
pawn but 1 7 lbh5+? is bad because of 9 a4 c6 10 0-0 dxe5 1 1 dxe5
17 .. .f6 18 lbxg7 WxgS} 17 ...Wf8 18 ltle3 d4 1 2 xd4 "ii'xd4 1 3 J:le1
Wd7 and Black will consolidate with Arriving at a critical position in which
19 .. .f6 according to Nigel Short. An White temporarily has a slight advan-

63
A lekhine 's D efen c e

tage because his pawn on e5 shuts the Black can equalise after 14 c3 with
g7 bishop out of play. Having said that, 14 ...1i'h4 15 lld2 i.c6 16 ll:le4 (or 16
there is nothing terribly wrong with ll:lf3 'i'h5 17 lLld4 'i'xe2 18 l:txe2 i.d5}
Black's position and if he successfully 16 ...i.xe5 17 i.g5 1Wxe4 18 11t'xe4 i.xe4
completes his development he will 19 l:lxe4 lld7 20 l:td1 i.d6 21 i.xe7
probably be about equal. ll:lc5!, as in Stefansson-Egger, Moscow
1 3 ... d7 Olympiad 1994.
13 ...e6 was Timman's first idea and it 14 ...c6
may in fact be playable if Black meets At the time of the game this was a
14 llcl2 with 14....i.d7 intending to new move but it's not at all clear that
transfer the bishop to c6 (in the game anything's wrong with the old line. Af
Short-Timman, Tilburg 1991, Black ter 14...e6 15 i.g5 h6, Chekhov gave 16
played 14...lld5 but after 15 ll:lf3 'i'cS i.e7 (White should play 16 1lad1, after
16 'ife4 'ifb4 17 .t.c4! ll:lb6 18 b3! which 16 exe5 17 Wxe5 i.xe5 1 8
..

ll:lxc4 19 bxc4 l:te8 20 l:td1 found him llxe5 hxgS is just equal} 1 6...l:tfe8 17
self completely tied up, White's pawn llad1 but then, as Burgess pointed out,
weaknesses being insignificant in this 17 ...1i'xe5 just wins a pawn for Black.
position). 1 5 ltlb51
Black would really like to exchange As Leko pointed out in his notes,
the light-squared bishops but 13 ...i.e6 Black can meet 15 e6 with 15 .. .f5 16
14 i.xe6 fxe6 15 lld2, followed by i.g5 Wb4, producing a position in
llf3, will support the key pawn on e5 which both sides have chances.
and leave Black with a miserable game. 1 5 .txb5 16 'lrxb5 c6 1 7 'lre2
..

He can, however, play 13 ... i.f5 14 lld2 ltld5 1 8 c3 'lrb6 1 9 .tc4 1lad8
'i'd7, after which 15 lLlf3 i.e6 16 l:td1
1Wc8 17 1Wb5 i.xb3 18 11fxb3 is only
slightly better for White.

20 .tg5?!
According to Leko this was an error.
He later preferred an initial 20 h4! and
14 3! only after 20...h5 to play 21 i.g5. This
Leko finds the most challenging line, position would be rather unpleasant for
planning to plant the knight on b5. Black.

64
Classical Divergences: 4 . . . c 6, 4 . . . lDc6 and 4 . . . g6

20... h6! 21 .i.c1 g6 5 .i.c4 llJb6 6 J.b3 .i.g7


After 2 1 ..ih4 Black would play In Volzhin-Davies, Dhaka 2001, I
2 1 .....c5, when the bishop on h4 is mis tried 6 ...liJc6?! but obtained a very bad
placed. But now Black gets some coun position from the opening after 7 exd6!
terplay on the d-file. cxd6 {Black also has a difficult position
21 86 22 h4 l:ld7 23 g3
after 7...1i'xd6 8 c3 .ig7 9 lbe4 'i'd8
Playing for mate with 23 h5?! g5 24 10 c3 because his only pawn lever, ...e7-
9e4 {threatening 25 .id3) would give e5, is very hard to achieve due to the
Black counterplay after 24 .. .f5! 25 exf6 danger of White playing .icl-g5 in re
l:.xf6, threatening f2. ply) 8 d5 ttla5 9 9d4 f6 10 ..td2! /1)xb3
23 l:lfd8 24 Wg2 Wc5 25 .i.b3
. 1 1 axb3 .ig7 12 .iaS! etc.
Wb6 26 .i.c4 Wc5 27 .i.b3 Wb6 28 7 a4 a5
Wc4 l:iJe7
Unveiling his fire-power on the d-file.
Black is out of the woods.
29 J.e3 Wc7 30 We2 l:iJf5
After 30. ..111xeS Leko intended to
play 31 .ib6!, after which 31...9xe2 32
llxe2 llf8! (and not 32 ...11.a8 33 l:txe6Q
33 .ixa5 {33 llxe6? ll'k8!) 33 ...liJd5 34
.ixd5 l:lxd5! 35 .ib6 lla8 36 a5 ..tf6
brings about an equal endgame.

The solid move, preventing the fur


ther advance of White's a-pawn.
7...dxe5 was covered in Game 24 (Oral
De Firmian).
8 'We2
At this point in the proceedings, 8
exd6 is well worth considering as after
8...cxd6 {White is also better after either
8 .....xd6 9 ltlc3 or 8...exd6 9 0-0 0-0 10
lte1) 9 0-0 0-0 tO h3, White is certainly
31 .i.f4 'Wb6 32 .i.c4 'Wc5 33 l:lab1 better off than in the 7 exd6 line be
.i.f8 34 .i.b3 Y.z -% cause of the weakness of b5. Adorjan
.-------. Smejkal, Wijk aan Zee 1972 continued
Game 26 10... ltlc6 11 ltlcJ ds 12 .if4 .ie6 13
Nunn-J. Howell liJbS with much the better game for
Port Erin 1994 White.
,____________...... 8 ... 0-o
1 e4 l:iJf6 2 e5 l:iJd5 3 d4 d6 4 l:iJf3 Black has also played 8...ll'k6 9 0-0

65
A lekhin e 's D efen c e

dxe5 (9...g4 is met by 10 .i.xf7+ etc.) better than 12 xe3 13 xe3.


10 dxe5 /i:}d4 1 1 /i:}xd4 'ifxd4, but with Nunn prefers to give up his light
the moves a2-a4 and .. a7-a5 insened
. squared bishop.
the position is very difficult for Black. 1 1 ...l0c5 1 2 :ld1 'it'e8 1 3 l0c3
Van der Wiel-Santa Roman, Cannes l0xb3
(rapid) 1992 continued 12 e6 (12 :e1 is Nunn pointed out that trying to win
also very interesting as aher 12 ....i.g4 I the a-pawn with 13 ... .A.d7 is far too op
think that Whitecan play 13 .A.e3! i.xe2 timistic; aher 14 lLld4 lLlbxa4 15 lLlxa4
14 .A.xd4 i.c4 15 .A.xb6 .i.xb3 16 cxb3 a4 (or 15 ....i.xa4 16 .i.xa4 xa4 17
cxb6 17 /i:}d2 followed by lLlc4, when b5 etc.), White plays the powerful 16
White simultaneously guards e5 and e61.
puts pressure on the b6-pawn) 1 4 cxb3 .te6
12 .. ..i.xe6 13 xe6 fxe6 14 1fxe6 1i'c4
15 1i'xc4 /i:}xc4 16 :a21

Black's main problem in this position


is in finding some sort of active plan.
(without a2-a4 and ...a7-a5 this He can also develop his pieces with
would be impossible) 16 ...l:r.f8 17 b3 14....i.d7 15 4 :d8, but where does
/i:}d6 18 .i.a3 :fs 19 c4 i.f6 20 lle2 he intend to go from there?
and White started to develop quite seri 1 5 l0d4 i.d5 1 6 l0xd5 l0xd5 1 7
ous pressure. l0f5! gxf5
9 h3 dxe5 1 0 dxe5 l0a61? Black had an unpleasant choice be
For the standard 10.../i:}c6 see the tween weakening his kingside like this
previous game, Leko-Timman. The text and allowing the exchange of his dark
is an interesting attempt to generate squared bishop. Personally speaking I
some counterplay which Howell had to would have preferred 17 ...c6; aher 18
face with on the white side before xg7 xg7 19 l:td4 1i'd7 20 :h4 g8
springing it on Nunn. 21 .i.h6 Black can play 2l. ..l:lfe8, in
1 1 0-0 tending ...1i'e6 and ...f7-f6 (Nunn gave
White can prevent the knight from 2 1 ...ltfd8 22 ltel when he felt that
coming to cS with 1 1 i.e3 but then White's attacking chances - based on
1 1...d5 leaves White with nothing e5-e6 or f4-f5 - were more significant

66
Classical Divergen c es: 4 . . . c 6, 4 .. .fi)c6 a n d 4 . . g6

than the strength of Black's knight on J.xf4 with a piece and two pawns for
ciS). the rook.
1 8 :Xd5 22 .z:lb1 J.g7?
22 ...J.e5 was more tenacious but
Black is still in serious trouble after 23
l:lcl.
23 tlg3 'ifc3 24 Wxc3 .i.xc3 25
.i.xf8 l:lxfB 26 .z:lxc7 ..tb4 27 .z:lxb7
.z:ld8 28 g3 Wg7 29 .z:lc1 1 -0

Game 27
Wydrowski-Marcinkiewicz
Corrondence 1997

1 e4 f6 2 e5 d5 3 d4 d6 4 f3
1 8 Wc6 1 9 Wd3
. ll:!c6
The only way to maintain control of
the d-file, though this involves the sacri
fice of the e-pawn. Black is okay after
either 19 llbS b6, intending ... 'tie6, or
19 l:ld3 :ads 20 l:lg3 1i'e6.
1 9 e6 20 .z:ld7 .*.xeS 21 .i.h6
..

A highly provocative move, which


encourages White to go for broke. I do
not recommended it to players with a
nervous disposition!
5 c4
The critical line, and in fact probably
2 1 ...J.xb2? the only way to get something. After the
According to Nunn this was the deci quiet 5 .te2 dxeS 6 xeS xeS 7 dxeS
sive mistake. Apparently Black can still ..tfs Black developed very comfortably
defend himself with the paradoxical in the game Kengis-Grigorian, Togliatti
21. ..i.d6! 22 f4 (or 22 1l'd4 f6 23 llg7+ 1985.
h8 24 h4 1i'e4 25 1i'h5 f4! defends White has also played 5 ..tbs but this
against the threat of l:lxh7+) 22 ...h8 fails to achieve anything after 5 ... a6 6
23 i.xf8 Lf8 24 11'c3+ 1i'xc3 25 bxc3 .i.a4 (aher 6 .txc6+ bxc6 7 0-0 Black

67
A lekhin e 's D efen c e

achieved a solid position with 7...e6 8 a) 7 h4 e5 8 d5 llJd4 9 llJxd4 exd4


liJbd2 .i.e7 9 lik4 a5 10 l:le1 a4 1 1 and now:
J.d2 J.d7 12 'ii'c t 0-0 13 liJg5 9b8! in a1) 10 J.d3 'i'd7 1 1 J.gS h6 12 J.d2
the game Shamkovich-Larsen, Moscow 'i'g4 13 J.e2 'ii'e4 14 0-0 J.fs 15 liJa3
1962) 6 ...liJb6 7 J.xc6+ bxc6 8 0-0 J.g4 was played in Tal-Larsen, Eersel {8th
9 'i'e2 'i'd7 10 h3 J.h5 1 1 1ik3?! (1 1 b3 matchgame) 1969, and now Black
looks better) 1 1 ...e6 12 i.g5 d5 and should have played 15 ...g6 (after his
with ...c6<5 coming Black had a good actual choice of 15 ...1i'xh4?! White
game in Lastin-Dyachkov, Russian Jun could have obtained a strong attack
ior Ch., Ekaterinburg 1996. with 16 liJbs d3 17 llJxc7+ ds 18
5 lbb6 6 e6
llJxaS dxe2 19 'ii'xe2 xa8 20 c5!).
The sharpest line, sacrificing a pawn a2) 10 'ii'xd4 eS 1 1 dxe6 i.xe6 12
for a strong attack. White can also try to J.g5 'i'd7 13 i.e2 was played in Honfi
exploit the early development of Black's Westerinen, Wijk aan Zee 1969 and
queen's knight by transposing into an now 13...1i'f7 looks fine for Black.
Exchange Variation with 6 exd6 exd6 b) 7 J.e3!? d7 (the alternatives look
(6... cxd6 7 d5 llJe5 S liJd4! exc4 9 a4! even less satisfactory: 7 ...g6?! 8 h4 will
llJes to liJc3 a5 1 1 J.bS+ i.d7 12 f4 give White a very strong attack and
i.xb5 13 fxe5 J.c4 14 e6! gave White a 7...e5 8 dS leaves Black without a good
powerful attack in Adorjan-Polgar, Bu place to put his knight) 8 lbc3 f6 9
dapest 1973) but after 7 d5 (7 llJc3 J.e7 J.d3 g6 10 h4 J.g7 1 1 lbgS eS 12 dS
8 .te2 0-0 transposes into positions li:\d4 13 h5 gxh5 (13 ...lbxh5 14 LhS;
considered in Chapter 6) 7...llJe5 8 liJd4 13 ... i.g4 14 f3) 14 i.xd4 exd4 was
c5! 9 dxc6 llDcc6 Black can equalise ac Vogt-Cibulka, Trencianske Teplice 1974
cording to Pogorelov. and now Plachetka's suggestion of 15
6 ... fxe6 li)ce4 (the game saw 15 l1e2) 15...J.g4
16 'ii'c2 looks very good for White.
c) 7 J.d3!? will be considered in the
next game, Reed-Danielsson.
7 ... e5

7 tbg5
The most direct line but not neces
sarily the best. White has several inter
esting alternatives:

68
Classical Divergen ces: 4 . . . c 6, 4 . . .loc6 and 4 . . . g6

8 .id3 ll'lxf8 xf8 13 0-0 c5 is another possi


White has a major alternative in 8 d5 bility suggested by Komarov; Black has
lbd4 9 i.d3 and now: compensation for the exchange) 12
a) 9 ...g6 looks very dangerous for ltlxc4 (if 12 0-0 there is 12...ll'lxa3 1 3
Black after 10 h4!? ..tf5 1 1 h5 i.g7 12 bxa3 i.c2 1 4 1Wh5+ g6, or if 1 2 1Wa4+
hxg6 i.xg6 (after 12 ... hxg6 there fol there follows 12 ...b5 13 ll'lxb5 i.d7 etc.)
lows 13 llxh8+ i.xh8 14 i.xf5 lbxf5 15 12 ... ll'lc2+ 13 4iii>f1 ltlxa1 14 ll'lxf8 xf8
'ifd3 with an edge for White because of 15 ll:le3 1rd7 16 1Vf3? (the theoretical
his control of e4) 1 3 lbxh7! (13 lbc3!? move is 16 i.d2 after which Burgess
c6 14 i.e3 e6 15 i.xg6+ hxg6 16 llxh8+ gives 16...ll'lc2 17 ll'lxc2 ...a4 18 ll'le3
i.xh8 17 dxe6 1re7 18 'ii'd3 also gave i.d3+ 19 c,i>g1 ...xa2 without an as
White a strong attack in I.Drozdov sessment; I propose that it's 'unclear',
Nosenko, Ukrainian Ch., Kharkov Black having three pawns for the piece)
1988) 13 ...i.xh7 (or 13 ...e4 14 11fg4!) 14 16 ... g6 17 g4 'i'c6 18 'i'xc6 i.d3+ 19
i.xh7 with a strong attack. el bxc6 20 <iftd2 i.bl and Black was
b) Black should probably play winning in Runowiecki-Marcinkiewicz,
9 ... i.f5! 10 i.xf5 lbxf5 1 1 lbe6 (after 11 correspondence 1997.
i.e3 too, Black should get ready to eject 9 ...g6 10 .ixg6+ hxg6 1 1 'iVxg6+
a White knight from e6 by playing d7 1 2 llJf7
1 1...g6, intending ...lbg7) l l .. .'it"d7 1 2
i.e3 (after 1 2 lbc3 the right move is
12 ... g6 as usual) 12 ...g6 13 i.xb6 axb6
14 lbc3 lbg7! 15 1i'g4 lbxe6 16 dxe6
1i'c6 17 h4 i.g7 18 lbcl5? b5! and Black
seized the initiative in Sorsak-Stratil,
Bratislava 1992.
8 ...llJxd4

1 2 llJxc4! 1
.

This amazing queen sacrifice com


pletely alters the assessment of this po
sition. 12 ...'i'e8 13 lbxh8 is known to
offer Black inadequate compensation.
13 llJxd8
Black intended to answer 13 Q)xh8!
with 13 ...c6!?, when it's anybody's guess
9 1Vh5+ as to what's happening.
The alternative is 9 i.xh7 :Xh7 10 1 3 .'xd8 14 b3

ll'lxh7 i.f5 1 1 ll'la3 lbxc4 (11...e6!? 12 Both 14 ll'ld2 and 14 Q)c3 allow

69
A lekhine 's D e fen c e

14....i.f5 but the position isn't clear in after 22 Wb5+ c6! 23 11xb6 ltb4 etc.
either case. White can also prevent this 22 J.g6 23 'it'd5 llXI4+! 24 Wd1
.

with 14 g4!?; all this requires more c& 25 'it'c4 b5! 26 xb5
analysis. Retreating the queen doesn't help.
1 4...J.f5 After 26 111ft there follows 26 ...ll:)xb3
27 We2 ll:)d4+ 28 Wdl lbc2 29 e2
llc4! etc.
26 J.c2+ 27 we1
..

1 5 'it'f7 l0b6 1 6 .i.e3


This meets with an amazing refuta
tion but there is very little White can do.
Aher 16 ..tb2 there follows 16....i.e6 17 27 ... cxb5! 0-1
...g6 lth6 18 11fg3 li)c2+ etc. If 28 11fc3 there follows 28 ....i.xb3! 29
1 6 ...l:lh7! 1 'li'h3 .te6 30 7 .tf5 31 ..h5 ltxg2.

Game 28
E.Reed-G .Danielsson
Buenos Aires Olympiad 1939

1 e4 l0f6 2 e5 l0d5 3 d4 d& 4 l0f3


lDc6 5 c4 l0b6 6 e6 fxe6 7 .i.d3!?

1 7 'it'g8
The point is that after 17 ...xf8+ d7
White's queen is trapped.
1 7 l0c2+ 1 8 We2 Wd7 1 9 J.xb6
..

axb6 20 l0c3 l:lg7 21 'it'c4 l:lg4 22


'it'f7
White also loses his queen mid-board

70
Classical Divergences: 4 . . . c 6, 4 . . Ji)c6 and 4 . . . g6

Alekhine himself liked this natural llxd4 12 lla3 es (there's nothing


move, which to me indicates that it de better) 13 116+ , but that after 13 ...exf6
serves a closer look. 1 4 11fxh8 11fg6 Black has good compen
7 li:Jd7
sation.
s . ltlt& 9 J:te1
..

Burgess, in The Complet.eAiekhine, sug


gests that 7 e5 is the answer but after 8
.. 9 ...e5?
d5 lDb4 White should keep his light Returning the pawn without easing
squared bishop with 9 i.e4! (9 lDg5 the defence. According to Alekhine
lDxd3+ 10 11fxd3 e6 1 1 lDxh7 11fe7 was Black should have played 9 ... g6, after
unclear in Letelier-Penrose, Moscow which the great man suggested that
1956) and now: White should sacrifice another pawn
a) White obtains a strong attack after with 10 d5 exd5 1 1 cxd5 llxd5 12
both 9 ... e6 10 lDg5 and 9 ... i.g4 10 a3 llg5!. White would certainly have very
lDa6 1 1 11fc2. strong pressure though this position
b) 9 ...g6 10 llg5 .i.g7 11 lllxh7 is would require further analysis.
also good for White. 10 dxe5 li:Jxe5 1 1 /Dxe5 dxe5 1 2
Black's best may be to challenge the J:txe5 d& 1 3 .*.f4 .tg4 1 4 d2
bishop immediately with 7 ...lDb4, after ..d7
which 8 i.e2 (8 1lg5 lDxd3+ 9 11fxd3 After 14...0-0-0? 1 5 ltd5 White wins
11fd7 10 lDxh7 c6 1 1 g6+ d7 pro on the spot.
duced a messy position in Lehmann 1 5 li:Jc3
Bogolyubov, Munich 1950) 8 ... g6 9 a3
see following diagram
lDc6 10 b4 .tg7 1 1 .tb2 0-0 12 0-0
11fe8 1 3 b5 d8 gave White good com In his notes Alekhine suggested that
pensation for the pawn in Vogt-Rogulj, 1 5 h3! was stronger as after 15 . 0-0-0 he ..

Trencianske Teplice 1979. obtains a winning attack with 1 6 aS


8 0-0 11fxd3 17 ltc5 etc.
Alekhine pointed out that White can 1 5 J:td8?
.

win the exchange with 8 lllg5 lllf6 9 After this Black's king gets caught in
.txh7 llxh7 10 11fh5+ d7 1 1 tDxh7 the centre. Alekhine pointed out that

71
A lekhin e 's D efen c e

Black should play 15 ...0-0-0 an d then 1 9 hxg4 'ii'xg4 20 J.t5! 'ii'xf4 2 1


gave a brilliant means of continuing the .ie6+ e8 22 'ii'a4+ l:td7 23 l:txd5
attack with 16 llb5! {Black is okay aher 'flc7 24 llxd7 ll:lxd7 25 l:td1 1-0
either 1 6 lld5 e6 or 16 l:le3 eS 17 .ixeS A very elegant game.
.ib4) 16...'irxd3 (lf 16 ... c6? then 17
llxa7+ wins, but 16 ...e6 is more tena Game 29
cious) 17 WaS d7 (17...1i'xc4 1 8 l:.cS) Ulibin-Baburin
18 l:te3 1i'c2 19 1i'xc7+ e8 20 l:.ael Vienna 1998
etc.
1 e4 ll:lf& 2 e5 ll:ld5 3 d4 d& 4 ll:lf3
c&

16 ll:ld5 c6
After 16...e6 White wins with 17
llxc7+ 1i'xc7 18 llxe6+ etc. 5 J.c4
1 7 'ii'c21 White can also gain space with 5 c4
but in this position Black's knight can
retreat to an arguably superior square
on c7 from where it may later harass
White's d-pawn with ... lbe6. So 5 ...llk7
and now:
a) White can prevent a pin on his f3
knight with 6 h3 but then Black can
revert to a Keogis Variation type plan
with 6 ...dxe5 7 llxe5 lld7 8 llf3 g6 9
llc3 .ig7 10 i.e2 bS! 1 1 0-0 (1 1 cxbS
llxb5 12 i.xb5 cxb5 13 llxb5 i.a6
would give Black excellent compensa
Threatening 18 i.g6+ followed by tion for his pawn) 1 l ...bxc4 12 .ixc4
mate. 0-0 13 lle1 llb6 14 i.b3 - Ramesh
1 7 /17 1 8 h3 cxd5
. Ibragimov, Sangli 2000.
Or 18 ... i.e6 19 l:txe6! 'i'xe6 20 ..if5 b) 6 llc3 ..ig4 7 exd6 (in the game AI
winning the queen. Sayed-Varga, Budapest 2001 White

72
Classical Divergences: 4 . . . c 6, 4 . . . li) c 6 and 4 . . . g6

carelessly played 7 h3?! and had his cen 5 .i.g4 6 h3 i.h5


..

tre demolished after 7...i.xf3 8 ..xf3 6...i.xf3 7 'it'xf3 dxe5 8 dxe5 e6 is


lLle6! 9 i.e3 dxe5 10 dxe5 lLld7 1 1 'it"g3 also very reasonable.
'it'aS 12 f4 g5! 1 3 f5 lLlg7 14 1Wxg5 7 g4 .i.g6 8 e6 fxe6
'ifxe5 15 g4 h5! with Black obtaining a
strong initiative) 7 ...exd6 (7...11fxd6 is
also possible) 8 h3 i.hS (and here Black
could consider 8 ...i.xf3 9 'i'xf3 i.e7,
intending to castle and then mount
pressure on d4) 9 g4 i.g6 10 ..b3 b6?!
(10 ...'it'c8 looks better, avoiding weak
nesses in his queenside pawn structure)
1 1 i.e3 i.e7 12 i.g2 0-0 13 h4 h5 14
gxh5 i.xh5 15 lLlg5! and White had
taken the initiative, McShane-Baburin,
Kilkenny 1998.
The quiet 5 ..i.e2 only has independ 9 .i.d3
ent significance if Black plays 5 lLld7!?,
.. Black can meet 9 ..e2 lLla6 10 lLlg5
(after 10 .ixa6 there is 10...11'a5+ fol
lowed by ...'it'xa6) with 10 lhac7 1 1 ..

lbxe6 lbxe6 1 2 11'xe6 1l'd7 with equal


ity. After the immediate 9 lbgS Black
can play 9...e5.
9 .i.xd3 10 1i'xd3 g6
.

but then 6 0-0 dxeS 7 dxe5 lLlc5 8 c4


lLlc7 9 11'xd8+ xd8 10 lLlgS! e8 1 1
.i.e3 lLl7e6 1 2 lLlxe6 lLlxe6 1 3 f4 gives
White the better endgame according to
Stoica. Also, after S ... dxeS White can
play 6 dxe5!? (6 lLlxeS transposes into
the 4 ...dxe5 5 li)xe5 c6 line) 6 ... .tf5 7 1 1 c4
lLld4!? e6 8 lLlxf5 exf5 9 0-0 li)d7 10 U lib in and Lisenko mention a couple
.i.c4 lLl7b6 1 1 .i.b3 .i.cS 12 'iff3 g6 13 of alternatives. The attempt to charge
lLlc3 0-0 14 l:ld1, which gave him strong down the h-file with t 1 h4 can be met
pressure in the game Smirin-Zelcic, by 1 l ...'l'a5+ 12 .i.d2 1l'a6 13 11'e4 lLld7
Struga Zonal 1995. 14 lbg5 lLlc7, whilst 1 1 lbgS allows

73
A lekhin e 's D e fence

1 1...e5 12 c4 (12 clxeS dxeS 13 c4 is met 25 cxd6


by 13 ...1ib4!) 12 .../l:lf6 13 dxeS 1Wa5+ 14
ll:lc3 1i'xe5+ 15 d1 1l:la6 16 1lel lhc5
17 'i'd2 'lrh2 18 b4 with a full-blooded
struggle in progress. These lines would
require _further analysis before any kind
of conclusion could be reached.
1 1 ...li)f6
1 1 .../1Jc7 is strongly met by 12 h4, in-
tending h4-h5.
1 2 li)c3
12 /l:lgS would once again allow Black
to play 12 ... e5, though once again the
resulting position would be far from 25 .. Jlxd4?
clear. A mistake. Black should play simply
1 2 .. .loa6 1 3 i.f4 25 ...exd6, after which 26 1l:lxe6 /1Jxe6 27
llxe6 .ieS 28 ltxg6+ hxg6 29 dxeS is
unclear.
26 .b3?
White in turn makes a mistake which
comes close to costing him the game.
He should play 26 'i'a3l, after which
26 .L4 (or 26.../l:lbS 27 1i'b3) 27 'ifb3
.

exd6 28 i.xd6 ltxg4+ 29 hxg4 'irxd6 30


h3l is probably winning.
26 ...exd6

1 3 ...o!Oc7
An interesting alternative is 13 ...c5!?,
undermining White's last remaining
central pawn. Baburin's move is aimed
more at solid defence, though it remains
to be seen how Black will use his extra
pawn.
14 0-0 i.g7 1 5 l:.fe1 0-0 16 i.g3
d7 1 7 l:.ad1 b5 1 8 b3 bxc4 1 9
bxc4 l:.ab8 20 l:.e2 l:.b4 21 Dde1
cs 22 c5 d7 23 lL!g5 lLifd5 24 27 Wt'b7!
lL!xd5 cxd5 And not 27 ll:lxe6? because of
The intermediate 24...Ld4 is bad 27 ... lDxe6 28 llxe6 .ieS winning the
because of 25 1ixd4 .ixd4 26 /1Jxc7 etc. exchange.

74
Classical Diverg ences: 4 . . . c 6, 4 . . .lll c 6 and 4 . . . g6

27 e5 28 l%c2 l:l.c4 29 l%ec1 Wb5


. . 36 ltxd6 1!rb7+.
35 . . .Wxa2 36 l:l.xd6 d4 37 :ea l:l.f6
38 g5

30 l%xc4
Giving up the queen is cenainly the
best practical chance. After 30 1!rxb5 38 ...:xt2+?
xbS 3 1 ltxc4 dxc4 32 lbc4 d4 the A serious mistake which lets the win
endgame is very bad for White. slip. Black should eliminate public en
30 ...Wxb7 31 l%xc7 Wb2 32 tile& emy number one, the knight on e6.
l:l.f7 33 l:l.c8+ f8 34 l:l. 1 c6 38 ... ltxe6! 39 ltdxe6 f7 wins without
After 34 ltlc7 Black can play 34 ... h6! much trouble.
aher which 35 ltxf7 36 xf8 d4 39 .i.xf2 'ii'xf2+ 40 1 Wf1 + 41
makes the d-pawn difficult to stop. h2 f2+ 42 h1 e1 + 43 *g2
34. . .Wb1 + 35 h2 Wd2+ 44 'i>g3 e3+ 45 *g2 We2+
35 g2? is even worse after 35 ... d4 46 g3 Wd3+ % -%

75
A lekhin e 's De fenc e

Summary
If White wants something against 4 lbc6 he is almost forced to play a critical pawn
..

sacrifice with 5 c4 li)b6 6 e6!?; I don't believe that either 5 J.b5 or 5 J.e2 really of
fer him anything. This certainly makes 4 ...li)c6 an interesting weapon for games in
which Black is prepared to take risks to win, but I strongly suspect that 7 J.d3 is a
strong move that gives White lasting compensation for his pawn.
Personally speaking, I feel that neither 4...c6 nor 4...g6 5 J.c4 li)b6 do enough to
challenge White's centre, and if White maintains the cramping e5-pawn Black finds
himself without a decent plan. The main advantage of 4...g6 might actually be to
transpose into the Kengis variation by meeting 5 .ic4 with 5...c6, and only reach
lines of the Kengis in which White's bishop is on c4 already. With 4...dxe5 S li)xe5
g6 6 c4 looking like a problem for the Kengis Variation, this move order should
certainly be a consideration for Black.

1 e4 f6 2 e5 d5 3 d4 d6 4 tLlf3 g6
4...c6 5 c4 li)b6 6 e6 fxe6 (D)
7 .id3 - Game 28
7 lbgS Game 27
-

4 ...c6 Game 29
-

5 .i.c4 b6 6 .t.b3 .t.g7 7 We2


7 a4 dxe5 8 a5 li)6d7 9 J.xf7+ (D) - Game 24
7 . . . 0-0 8 h3 aS 9 a4 tLlc6
9 dxeS 10 dxe5 lba6 - Game 26
...

1 0 0-0 dxe5 1 1 dxe5 /0d4 1 2 tLlxd4 Wxd4 1 3 l:le1 (D) - Game 25

6. . . fxe6 9 Jl.xf7+ 1 3 l:le 1

76
I CHAPTER FIVE I
Exchange Variation
with 5 cxd6 . . .

1 e4 c!iJf6 2 e5 c!iJd5 3 d4 d6 4 c4 than 9 . . . e5. This definitely gives White


l.l:lb6 5 exd6 cxd6 initiative aher 10 dxeS dxeS 1 1 'ii'xd8
The Exchange Variation has always l:txd8 12 cS! (Game 34) but these posi
been known as a solid line but in the tions might be playable for Black.
last couple of years it has come into its r------

own. The so-called Voronezh variation Game 30


with 6 ltJc3 g6 7 .te3 .tg7 8 :te l 0-0 9 Levacic-Rogulj
b3 has been giving Black all sorts of Croatian Ch., Pula 1998
trouble. It looks innocent enough, but
the mass-evacuation of the a1-h8 di 1 e4 l.i:lf6 2 e5 l.i:ld5 3 d4 d6 4 c4
agonal makes it very difficult for Black l.l:lb6 5 exd6 cxd6 6 ll:lc3
to generate his thematic counterplay.
Game 30 shows how Black's coun
terplay usually develops; White is unable
to play 10 d5 because of the strong re
ply, 10...ltJa5!. Thus Black gets time for
central action with 1 1...e5 or even
1 1 ...d5 12 cS ltJc4. However, when
White has already played the 'Voronezh'
9 b3, as soon as Black plays 9. ..ltJc6 he
gets hit with 10 dS! {Game 31).
Black has tried to combat the Vo
ronezh in a number of different ways;
perhaps the most intriguing of these I successfully revived Hartston's old
being 9 .. .f5 as in the notes to Game 32. idea of 6 .te3 g6 7 h4 (7 dS gives Black
Yet if this fails to pass the test posed by good play aher 7 ....tg7 8 .td4 .ixd4 9
10 g3, there may be little option other ...xd4 0-0 10 ltJc3 eS!) in a critical last

77
A lekhin e 's D e fenc e

round game which I needed to win for on the knight is to continue developing
my first GM norm. Davies-Westerinen, with 7 i.cl3 i.g7 8 ge2, a plan used
Oslo 1987 continued 7... .tg7 8 h5 ltlc6 by the great Bobby Fischer. Black can
9 ltlc3 and now, rather than 9 ... d5, then obtain adequate counterplay with
Black should have played 9 .. 0-0 10
. 8 ... 0-0 9 0-0 e5! (9... ltlc6?! 10 d5 lt)e5 1 1
hxg6 fxg6, which I found Westerinen b3 leaves White with an annoying space
analysing with Jansa shortly after the advantage) 10 i.e3 6 (10...5!? 1 1
game. Black's pawn structure isn't great, dxeS dxe5 1 2 c5 l2)6d7 1 3 (3 a6 14 b4
but he is well ahead on development. ltlc6 {Trapl-Hoticka, Ceske Budejovice
6 ...g6 7 h3 1993} is also interesting) 1 1 d5 ltlb4
(1 1...ltle7 12 b3 ltld7 13 ltle4 ltlf5 14
i.g5 (6 15 .td2 was good for White in
Fischer-Berliner, US Ch., New York
1962/63) 12 b3 ltlxd3 13 1i'xd3 lieS!?
{the 'obvious' 13...f5 is well met by 14
f4, shutting Black's light squared bishop
out of the game) 14 llac 1 ltld7 15 b4 a5
16 ltlb5 e4 17 1i'd2 ltle5 and Black had
taken the initiative in Jhunjunwala
Timman, Teeside 1974.
Having lost faith in the h-pawn push
I used against Westerinen, I subse
White wants to develop his knight on quently switched to an advance of my a
f3 without having it pinned by ... i.g4. pawn with 7 a4.
7 ltlf3 .tg7 8 i.e2 0-0 9 0-0 i.g4 10
i.e3 ltlc6 1 1 b3 is known to be fine for
Black after 1 1. .. d5 12 c5 c8, a recent
example being 13 b4 a6 14 llb1 e6 15
a4 ltl8e7 16 b5 axb5 17 axb5 i.xf3
(17...a5 18 i.f4 i.xf3 19 i.xf3 ltlf5
20 ltle2 h5 2 1 h3 lt)h4 22 1i'd.3 lt)c4
also gave Black good counterplay in
Glauser-Varga, Zurich 2001) 18 gxf3?!
{trying to keep control of c4, but the
weakening of the kingside carries more
weight) 18 ... lt)aS 19 i.cl3 ltlc4 20 1i'e2
lt)xe3 21 fxe3 e5 22 dxe5 .txe5 23 lt)d 1 The game that made me lose faith in
lla3 24 llcl d4 25 e4 11fc7 26 llc2 this plan was Davies-Chekhov, Gausdal
i.xh2+ 27 1i'xh2 1i'xh2+ 28 1990, in which I got nothing from the
llxd.3 and Black went on to win in opening after 7...a5 S lt)f3 (the sharp 8
Zufic-Zelcic, Pula 2001. c5!? was successfully neutralised in
Another means of preventing the pin Perovic-Begovac, Yugoslavia 1985:

78
Ex change Varia tion with 5 . . . cxd6

8 ... dxc5 9 .i.b5+ tn6d7 10 .i.4 .i.g7 1 1 g5 13 i.e3 d5 14 c5 lL!c4 15 i.xc4 dxc4
lL!d5 e5 12 dxe5 0-0 13 lL!f3 lLlc6 14 16 d5 lL!b4 17 h4 ..td3 18 hxg5 hxg5 19
.i.g5 f6; Burgess has suggested that .i.xgS lL!xdS was good for Black in
White should try to get this line with the Adorjan-Eales, Groningen 1970) 12...e5
moves h2-h4 and ...h7-h5 inserted, but 13 .ie3 e4 14 lL!d2 d5 15 cxd5 lLlxd5 16
Black should probably meet 8 h4 with lL!xdS xdS 17 .ic4 d8 Black had
8 ... .i.g7 9 h5 0-0 10 hxg6 fxg6 with a the better game in Geller-Vaganian,
lead in development that offsets Black's Moscow 1985.
weakened structure) 8 ... .i.g7 9 i.e3 0-0 b) 1 1 .igS!? h6 12 i.e3 dS 13 c5 c4
10 .i.e2 lbc6 1 1 d5 (without the moves 14 .tel b6 15 b3
a2-a4 and ...a7-a5 inserted this would be
strongly answered by 1 1 ...lLla5; the
negative side of White's plan is that
Black gets an invulnerable outpost on
c5 for his other knight) 1 1...lLle5 12
lL!xe5 ..txe5 13 0-0 lL!d7 14 lla3 lL!cs
15 .i.d4 .i.xd4 16 xd4 i.d7 17 lL!bs
b6 1 8 :let lle8 19 ..tg4 f5 20 .if3 l:lc8
and a draw was agreed a few moves
later.
7 !i:.g7 8 /t)f3 0-0
.

with a further split:


b1) It's starting to look as if the
tempting 15 ... bxc5?! is good for White
after 16 bxc4 cxd4 17 lLlxd5 and now:
bll) 17...e6 18 .ia3 exd5 (18...l:le8
19 lM4) 19 ..txf8 xf8 20 llc l dxc4 21
.ixc4 and Black had inadequate com
pensation for the exchange in Sermek
Sutter, Biel 1994.
b12) 17 ...d3 18 .ixd3 ..txd3 19 .xd3
i.xa1 20 i.xh6 .ig7 21 J..xg7 xg7 22
9 !i:.e3 c3+ g8 (22 .. .f6 23 g5 11rd6 24
There's a strong argument for not lbc7 was also good for White in
putting this bishop on e3 so soon as it Khmelnitsky-Pesotsky, Kiev 1989) 23
gets hit after the sequence ...d6-d5, c4- lle1 l:le8 (or 23 ...'1t'a5 24 'lt'e3 threaten
c5, ...lL!c4. In fact there is a major alter ing 25 h6, as in Khmelnitsky-Kozlov,
native for White in 9 .ie2 after which Naberezhny Chelny 1986) 24 e5
the main line runs 9 ... lbc6 10 0-0 .if5 xeS 25 'IfxeS and White had powerful
and now: threats in Hervet-Cimolai, correspon
a) After 1 1 .i.f4 h6 12 llcl (12 'iid2 dence 1993.

79
A lekhine 's Defen c e

b2) 15...ltJ4a5 1 6 cxb6 axb6 17 .i.e3 clxc4 13 0-0 .i.f5.


(White can also play 17 .i.f4 when Ser
mek-Moscatelli, Vienna 1991 cominued
17 ...g5 18 .i.g3 ltJb4 19 Wd2 ltJac6 20
.J:fcl .i.e4 21 ltJb5 .i.xf3 22 .i.xf3 llc8
23 h4 with an edge for White) 17...Wh7
18 :ct .i.e6 19 11d2 and I, for one,
prefer White due to the awkward posi
tions of Black's knights and the outpost
on b5 {Brodsky-Pesotsky, Ukrainian
Ch., Kharkov 1988).
In view of the fact that the main line
looks rather good for White, I suggest
taking at look at Black's treatment in the Lev Alburt has analysed this position
game Rechel-Movszeszian, Germany with White's queen's rook on a1 rather
1995; after 9 ... e5!? 10 .i.g5 f6 1 1 i.e3 than cl, but it seems to me that it
.i.e6! 12 d5 .i.f7 13 0-0 ltJa6 he ob doesn't make a massive difference:
tained a compact and harmonious posi a) In The Complete Alekhine Burgess
tion. gave Gipslis-Ciocaltea, Bucharest 1968
9 /0c6
as better for White after 14 b3 .i.d3 15
:tel ltJxd4 16 ltJxd4 .i.xd4 17 bxc4
.i.xe3 18 :Xe3 i.xc4 19 xd8 llfxd8
20 llxe7 but I think that Black can
equalise with 20... b5! 21 c6 (or 21 ltJe4
f8) 21. ..llac8 22 c7 .J:d6 etc.
b) 14 a4 and now:
bt) Not 14...i.d3 15 :tfd1 ltJxd4?
(15 ...e5 16 dS ltJd4 17 ltJel!) 16 ltJxd4
i.xd4 17 :Xd3 and White wins mate
rial.
b2) 14 ...e5 15 llfd1 (15 clxe5 is met
by 15....i.d3 16 llfdt ltJxe5, and 15 d5
10 :c1 by 15...ltJd4) 15 ...exd4 16 ltJxd4 ltJxd4
The immediate 10 d5?! allows 17 i.xd4 i.xd4 18 1i'xc4 i.xf2+ 19
10 ... ltJa5! 1 1 i.d4 e5! 12 clxe6 .i.xe6 13 xf2 and now 19...Wg5!? produces a
.i.xg7 xg7 14 'ifd4+ Wf6 15 b3 dS! 16 sharp position in which both sides have
cS ltJd7 17 Wxf6+ ltJxf6 when Black chances (after 19...W'f6 20 ..d4! Wxd4+
had the initiative in Mikac-Khmelnitsky, 21 .J:xd4 llac8, the move 22 ltJdS takes
Sibenik 1989. advantage of the fact that White's rook
10 . .. 85 is on c1 already).
I am not convinced that Black is 1 1 dxe5 dxe5
worse after 10... d5 1 1 c5 ltJc4 12 .i.xc4 The fighting move, but Black may

80
Ex change Varia tion with 5 . . . c x d6

have a simpler route to equality in 0-0 i.h6 are messy) 19.../.l)cS 20 i.e4
1 1...e5 12 llxe5 .i.xe5 13 .i.e2 .i.e6 b5 21 i.xf8 lbf8 and Black had a
14 b3 d5 15 cxd5 .A.xc3+ 16 llxc3 winning attack in Grzesik-Hanman,
llxd5 17 lld3 llxe3, as in Serper German Bundesliga 1985.
Yermolinsky, Tilburg 1994. 14 a3 e4 1 5 lt\d2
1 2 c5 l0d7 1 3 .i.c4 15 lLld4 gives White nothing after
15 .../.l)xd4 16 i.xd4 i.xd4 17 11xd4
11'xc5 etc.
1 5 lL\de5!?
..

15 ...lbxc5 16 b4 'irxa3 17 i.xcS


i.xc3 18 i.xf8 xf8 19 b5 was good
for White in Zagrebelny-Remizov, St
Petersburg 1994.
16 0-0 l:ld8

1 3 85

It isn't quite clear what Black's


strongest move is at this stage. He has
tried two other moves with reasonable
results:
a) 13 ...h6 14 0-0 h7 15 /.l)d2 f5 16
f3 /.l)d4 17 llb3 llb8 18 /.l)b5 llbc6
(18 ...eDxb5 19 i.xb5 llc6 20 'ffd6 11'e8
21 lla5 llf6 22 11'd2 was slightly better 1 7 1Wb3
for White in Zagrebelny-Varga, Buda A mistake, according to Finkel. An
pest 1993) 19 /.l)3xd4 lbxd4 20 /.l)d6 b6 earlier game (Glauser-Varga, Zurich
2 1 b4 a5 22 a3 produced a very complex 2000) went 1 7 llcxe4 lbxc4 18 llxc4
struggle in Pergel-Nataf, World U16 i.xb2 {18...i.e6!? is a natural alterna
Ch., Szeged 1994. tive) 19 i.g5 lle8 (19 ...l:ld5 20 b3
b) 13 .../.l)d4 14 lLle4 {14 /.l)xd4 exd4 i.g7 21 l:la4 wins material and 19...l:lf8
15 i.xd4 .Ue8+ 16 eDe2 aS+ recovers 20 i.f6 i.eS 21 /.l)f3 i.xf6 22 llxf6+
the cS-pawn with equality) 14 . . b6 15
. g7 23 'flat threatens a deadly discov
i.gS 'ilc7 16 i.e7 i.b7! 17 c6 (17 i.xf8 . ered check) 20 .i.f6 (20 b3 i.g7 21
i.xe4 gives Black compensation for the lLld6 llf8 22 eDxb7 c7 is fine for
exchange) 17...'ffxc6 18 lLleg5 h8 19 Black) 20.../.l)e5 21 l:lc2 - . though
i.d3? (19 /.l)x7+? llxf7 20 i.xf7 e4+ according to Finkel White is much bet
2 1 fl i.a6+ 22 g1 eDe2+ 23 h2 ter in this final position.
11'f4+ wins for Black, but both 19 i.xf7 1 7 lt\d4 18 Wb4 xb4 1 9 axb4
.

llxf3+ 20 gxf3 and 19 i.xf8 Lf8 20 lt\d3 20 l:lb1

81
A lekhin e 's D efen c e

After 20 i.xd3 exd3 21 .i.xd4 l:r.xd4 The so-called 'Voronezh Variation',


22 b5 Black keeps an edge with which has been giving Black some seri
22...l:r.b4. ous headaches.
20 ...lbxb4 21 lbdxe4 .i.e& 22 .i.xe6 9 l0f3 gives Black excellent play after
lbxe6 23 lbb5 9....i.g4 10 h3 .i.xf3 1 1 11xf3 (1 1 gxf3?
Or 23 llJd6 b6. d5 12 c5 t:bc4 13 i.xc4 dxc4 14 0-0
23 ...lbd5 l0c6 15 t:be2 e5 16 d5 tbb4 was good
for Black in Shovunov-Solozhenkin,
Russian Ch., Elista 1996) 1 1 ...l0c6 12
1td1 d5 13 c5 t:bc4 14 i.xc4 dxc4 15 d5
llJb4 16 0-0 t:bd3 17 llc2 lieS, as in
Levin-Shainswit, Ventnor City 1944.
9 lbc6?! 10 d5
..

24 .i.g5
After this White loses a pawn. The
most tenacious line was 24 b4 a5 25
bxa5 l0xe3 26 fxe3 f5 27 l0ed6 l0xc5
28 :fcl with drawing chances.
24. . . f6 25 .i.c1
25 .i.h4 is no better as 25 ... g5 26 The point of the Voronezh set-up
i.g3 f5 27 llJed6 f4 shuts White's and White's mass evacuation of the al
bishop out of the game. h8 diagonal. Usually this advance would
25 f5 26 lbg5 lbxc5 27 b4 lbe4
. be met by 10 ...llJa5, with embarrassing
28 lbe6 l:td7 29 l:d1 a6 30 lbxg7 pressure on the c-pawn. In this position
axb5 31 .i.b2 'iti7 32 g4 fxg4 33 there would no pressure whatsoever,
hxg4 l:ad8 34 J:ld4 'Ddc3 35 J:lxd7 + just a badly placed black knight.
J:lxd7 36 .De1 J:ld 1 37 l:lxd1 lbxd1 10 lbe5
.

38 .i.d4 lbdxf2 0-1 1 O llJbS is no better; White keeps an


--------. edge after 11 i.e2 e5 12 dxe6 i.xe6 13


Game JJ llJf3 l0c6 14 0-0.
Nedev-Oney 1 1 .i.e2 lbed7
Heraklio 1997 One of several alternatives, but in
.______________. every case White's space advantage is
1 e4 'Df6 2 e5 lbd5 3 d4 d6 4 c4 the position's most important factor:
ltlb6 5 exd6 cxd6 6 lbc3 g6 7 .ie3 a) 1 1 ...f5 12 f4 tbg4 (12 ....i.h6 13
.i.g7 8 J:lc1 0-0 9 b3 d2 llJf7 14 llJf3 e5 15 dxe6 .i.xe6 16

82
Exchange Varia tion with 5 . . . cxd6

0-0 was also good for White in Peder


sen-Hjorth, Aalborg 1995) 13 .i.d4 e5
14 dxe6 .txe6 1 5 liJf3 ll.e8 16 .i.xg7
xg7 17 0-0 and White had an edge
because of the weakness of d6, Howell
Panchenko, Hamburg 1995.
b) 1 1..Jie8!? 12 /iJf3 e6 13 dxe6
.txe6 14 /iJxe5 (14 /iJd4!?) 14 ....txe5 15
0-0 d5!? 16 .txb6 1Wxb6 17 cxd5 l:lad8
18 .tc4 .tf5 19 1i'd2 a6 20 a4 1i'd6 21
g3 h5 22 :fe 1 produced a position in
which it is difficult for White to convert
his extra pawn, though he's certainly on 17 b4
the positive side of the position, Kuc Starting to advance his queenside
zynski-Nokka, Nordic Team Ch., Pohja pawn majority, but there was an argu
1985. ment for further suppression before
c) 1 1 ...h5!? 12 f4 (the most direct line, trying to do anything active. White
but the simple 12 /iJf3 also looks good) could also restrain Black's next move
12 .../iJg4 13 .td4 .th6 14 liJh3 e5 15 with 17 llcd1 and an earlier improve
dxe6 fxe6 16 .txg4 hxg4 17 1i'xg4 with ment might have been 15 1i'd2 followed
a huge edge in Howell-Trifunovic, by 16 l:tedl.
Hastings 1995. 17 e6 18 .!Dd4
..

d) 1 1...e6 12 f4! liJed7 (12 ....th6 13 White could also consider 1 8 dxe6
1i'd2! liJg4 14 .i.xg4 'i'h4+ 15 1i'f2 .i.xe6 19 l:.ed1 but this might have been
1i'xg4 16 h3 1i'f5 17 g4 1i'd3 18 /iJge2 psychologically difficult to play because
exd5 19 l:ld1 trapped Black's queen in it frees Black's pieces to some extent.
Raecki-Velicka, Apolda 1992) 13 dxe6 18 a6 19 .!Db3
.

fxe6 14 1i'xd6 e5 15 /iJf3 exf4 16 .txf4 Finkel recommendedthat White first


llx:S 17 0-0 .tg4 18 h3 .i..xc3 19 hxg4 play 19 dxe6 /iJxe6 and only then 20
left White with a good extra pawn in the liJb3, after which 20 ....i.c6 2 1 l:lcd1
encounter V.lvanov-Bagirov, Moscow 'i'c7 22 a4! (and not 22 'i'xd6? 1i'xd6 23
1995. l:lxd6 i..f8} 22 ...l:lad8?! 23 a5 is very
e) 1 1...a5 12 f4 /iJed7 13 /iJf3 /iJc5 14 unpleasant for Black. But in any case
0-0 .tg4 15 .td4 .txf3 16 1lxf3 .i.xd4+ the position looks better for White.
17 1i'xd4 liJbd7 18 .i.fl 1i'b6 19 l:le1 1 9 ....z:tc8 20 c!Da5 b6 21 c!Db3 exd5
left White with pressure against e7 in 22 c!Dxd5 c!De4 23 'lrd1 .z:tb8 24 .i.d4
the game Emms-McDonald, Hastings f6 25 .i.d3 .!Dg5 26 .z:txe8 .i.xe8 27
1997/98. 'l'e2 c!Dge6 28 .i.c3 .i.f7 29 a4!?
1 2 lbf3 lbf6 1 3 h3 .!Dd7 30 a5 f5
13 /iJd4 is also good. 30 ... bxa5 31 /iJxaS llc8 looks better,
1 3 lbbd7 14 0-0 %le8 1 5 %le1 lC!fS
. . avoiding any further weakening of the
1 6 'ifd2 .i.d7 dark squares.

83
A lekhin e 's Defence

lbxc4 41 ltlxa6 l:lb6 42 lbc7 etc.


36 ...Wf6! 37 'ile2 ltxb4 38 .i.c2
:b21 39 d5 .i.xd5 40 cxd5 :Xc2
41 'ilxc2 f3+ 42 Wf1 xe1 43
xe1 -.a1 + 0-1

Game 32
Sofronie-Ziatic
Brasov 1998

1 e4 f6 2 e5 d5 3 d4 d6 4 c4
b6 5 exd6 cxd6 6 c3 g6 7 .te3
31 .i.xg7 Wxg7 32 l:te1 .i.g7 8 :c1 0-0 9 b3 .i.f5
32 1i'h2+!? would force 32 ...h6 {not Black's most popular alternative to
32 ...g8? 33 axb6 lbxb6 34 lbf6+) but 9...e5 is the interesting 9...f5!?, which
then it isn't easy to capitalise on the aims for active counterplay at the cost
position of Black's king. of structural weaknesses. After this
32 ...'ilg5 33 g3?! move we have the following possibili
With the time control approaching, ties to consider:
White starts to lose his way - this weak a) 10 d5 is adequately met by 10 .. .f4
ens f3. Finkel suggested 33 axb6 lbxb6 1 1 d4 e5 12 dxe6 xd4 13 1i'xd4
34 1i'b2+ h6 35 lbxb6 .U.xb6 36 g3! lbc6 14 1td2 .ixe6 15 lbf3 'it'e7 16
when 36 ...llxb4? is met by 37 h4 fol .ie2 d5 17 cxd5 l:lad8 when Black re
lowed by a fork from d2. covered the pawn with an okay position
33 ...bxa5 34 l0xa5 e5! 35 b2!? in Milu-Ignatescu, Romania 1995.
c5 b) 10 lbh3 looks logical but the
knight can become misplaced on h3.
The game Malisauskas-Kupreichik,
played in the Moscow Olympiad 1994
continued 10 ...h6 1 1 f4 e6 12 dS .if7
13 .ie2 lte8 14 Q.O e6 15 dxe6 llxe6 16
.id4 .ixd4+ 17 1txd4 lbc6 18 'iff2
lLld7 19 f3 ltlf6 with a reasonable
position for Black.
c) 10 lLlf3 is probably best met by
Burgess's suggestion of 10...f4 1 1 .id2
e5!? 12 dxeS dxe5 1 3 c5 e4!? 14 lbxe4
lte8 15 cxb6 lbe4+ 16 e3!? 'ifxd1+
36 f4?? 17 d1 .if5 with messy complica
A blunder. White could force a draw tions.
with 36 :XeS! lbxd3 37 lhf5+ lbxb2 38 d) White's most promising reply
.U.xg5 e6 39 lbc7 .ixc4 40 lbxc4 seems to be 10 g3

84
Ex change Varia tion with 5 . . . cxd6

dS, with White keeping a solid space


advantage after 10...t0a6?! (I prefer
lO...eS!? 1 1 dxe6 .i.xe6 intending ...d6-
d5) t 1 li.)f3 .i.g4 12 .i.e2 ..txf3 13 .i.xf3
llX5 14 b4 llXd7 15 b3 l:lc8 16 .i.e2
li.)f6 17 0-0 in Yagupov-Petit, Ubeda
1996.
10 ...d5?!

10 ... e5 1 1 dxeS dxeS (1 1...i.xe5?! 12


li.)h3 would see White's pieces converge
on the weak d5-square) 12 ..xd8 l:txd8
13 cS f4 14 ..td2 li.)6d7 15 .i.c4+ h8
(15 ...8 16 li.)dS li.)xcS 17 li.)c7 b6 1 8
li.)xa8 .i.b7 19 li.)c7 fxg3 20 hxg3 .i.xh 1
21 ..tb4 .i.h6 22 l:lc2 l:lc8 23 6+
e8 24 .i.xcS bxcS 25 f3 .i.e3 26 l':h2
.i.xgl 27 l:lxh1 is good for White ac The idea behind 9.....tf5, but White
cording to Ardeleanu, but such long can now force a clear advantage.
variations may contain improvements 1 1 c5 l0c8 12 .i.f3! c6
for both sides) 16 lilb5 lilxc5 17 lilc7 Attempting to counterattack. 12 ...e6
b6 18 lilxa8 i.b7 19 f3 .i.xa8 20 b4 can be met by 13 g4 and 12 ....i.e6 is
4 21 fxe4 ..i.xe4 22 lilh3 i.xh 1 23 answered by 13 lilge2, intending lilf4.
lOgS and the threat of lilf7+ had Black 13 .i.xd5 e5 14 .i.xc6 bxc6
in trouble in the game Pavasovic Or 14 ... exd4 15 .i.xb7 l:lb8 1 6 .i.f3
Bawart, Bled 1998. dxe3 17 'irxd8 exf2+ 18 Wxf2 1bd8 19
I'm not sure that all of Black's 9th lilge2 etc.
move alternatives have been properly
investigated. One further possibility is
the super-solid 9 ...e6, intending to de
velop with ...lilc6 and ... ..td7 and then
eventually play ...d6-d5. Another idea is
9 ... ..td7, intending either ... e7-e5 (with
out allowing the exchange of queens) or
going back to the solid plan of ...e7-e6
and ...ll:x:6. I think that these ideas merit
further investigation.
1 0 .ie2
Another possibility for White is 10

85
A lekhin e 's D e fen ce

15 dxe5?! For 1 1 'ii'xd8 see the next game.


Allowing Black to get some compen Rowson's move was also highly re
sation for his pawn. After 15 lLlge2 he garded until Baburin's 12th move nov
has nothing. elty.
1 5 ..txe5 1 6 'ifxd8 l:lxd8 1 7 lt:lf3
.. 1 1 ...lt:l6d7 1 2 ..tc4 'ifa5!
.i.g7 1 8 0-0 lt:le7 1 9 .i.g5 lld7 20
l:tfe1 lt)d5 21 lt:lxd5 J:txd5 22 J:ted1
..te4 23 l:lxd5 ..txd5 24 lLld2 f5 25
f1
Black's bishops dominate and the po
sition is even becoming dangerous for
White. 25 h4 would have been a better
idea, preventing the following kingside
pawn advance.
25 h6 26 ..te7 J:te8 27 .i.d6 g5 28
..

f3 g4 29 ..tf4 h5 30 a4 ..td4 31 b4
a6 32 fxg4? fxg4 33 g3? h4!
An important new move which
forces White to develop his knight on a
far more passive square than he'd like.
After 12 ...lLlc6 White can play 13 lLlf3.
13 lt:lge2
The point of Black's move order is
that 1 3 lL!f3 can now be answered by
13 ...e4 14 lLld4 lLlxc5.
1 3 ...lt:lc6

With the terrible threat of 34 ... h3 fol


lowed by 35... .1g2 mate - and there's
nothing White can do.
34 J:te1 .i.g2+ 35 xg2 0-1

Game 33
Rowson-Baburin
Port Erin 1999

1 e4 lt:lf6 2 e5 lt:ld5 3 d4 d6 4 c4 14 a3
lt:lb6 5 exd6 cxd6 6 lt:lc3 g6 7 ..te3 An attempt to exploit the position of
..tg7 8 l:tc1 0-0 9 b3 e5 10 dxe5 Black's queen, which meets with a
dxe5 1 1 c5 strong reply. Another possibility is 14

86
Ex change Varia tion with 5 . . . cxd6

J.d2 'irxcS 1S lbbS (threatening i.f7) variation 22 lbxe8 :E.xe8 23 h4 ll:\hS! 24


1S ...1We7 16 lbc7 ltb8 17 lbd5 9h4 and 'ifh6 hc4+ 2S bxc4 ll.e2 26 gl llxf2
although White has the initiative it's 27 l:th2 ll:lf4 28 1i'gS h6! when Black
hardly worth a pawn. wms.
1 4 l2Jd41
.. 22 J:tad8 23 h4 .i.xc4+
..

14 ... 'ilxa3?? would lose the queen af Volzhin pointed out another possibil
ter lS llal 'il'b4 16 lla4, but now Black ity for Black in 23 ...1i'eS!?, after which
threatens both a3 and cS. 24 'fixeS .txc4+ 2S bxc4 llxeS gives
1 5 lbxd4 exd4 1 6 .i.xd4 J:te8+ Black a winning ending.
24 Wxc4 .!Lle4?
A serious mistake. Volzhin pointed
out that simply 24 ...lte6 is strong,
threatening both ... l;)e4 and ...l;)e8.
25 'tixf7 + h8 26 'tif3 .!Llxd6
26 ...lbd2+ would also lead to a draw
after 27 llxd2 11'xd2 28 11'f6+ g8 29
'l'f7+ h8 30 'l'f6+ g8 etc.
27 cxd6 J:tf8 28 1Ve3 J:tfe8 29 'tif3

1 7 1
The loss of castling rights shows that
things have gone wrong for White,
though Rowson certainly makes a fight
of it. 17 J.e2 would have been worse,
losing a pawn for inadequate compensa
tion after the sequence 17 ...i.xd4 18
xd4 'ilxa3 19 ltdll li'xcS 20 'fixeS
ll:lxcS 2t llJdS lleS! 22 f4 (or 22 b4 J.e6
23 ll:lf6+ g7 24 f4 .:Xe2+ 25 xe2 29 We5?
..

xf6 26 bxcS .tg4+) 22 ...1Z.xe2+ 23 Playing on through momentum


c;i;>xe2 .tg4+ etc. rather than any objective merit of the
1 7 .i.xd4 1 8 'tixd4 'tixa3 19 'tif4
. position. Black should repeat the posi
l2Jf6! 20 l2Jb5! 'tib2 21 l2Jd6 tion.
It looks as if White has obtained a . 30 h5! J:txd6 31 hxg6 J:txg6 32 J:th5!
powerful initiative for his pawn, but Finally developing the king's rook.
Baburin's next move turns the tables. Now White is better because of Black's
21 .i.e6! 22 l:td1
exposed king.
If White accepts the exchange sacri 32 ...9e7 33 J:tf5 g8 34 g3 9e6
fice it isn't clear how he should com 35 g2 a6 36 ltd4 'tic& 37 'tixc6
plete his development. Volzhin gave the bxc6?

87
A lekhin e 's D e fen c e

Black's nerve is going. Allowing Af4 l:tg1 79 l:lf5 llc1 80 l:lh3 llc7
White's rooks to the seventh rank is 81 5 llf7 82 l:le3 :as 83 l:lfe5
certainly a lesser evil than the crippling l:[ff8
of his queenside pawns. After 37...lbc6
38 %ld7 Black defends himself with
38 ... b6 39 llff7 h6 40 l:tg7+ h8 41
%lh7+ g8 42 lldg7+ f8 43 %la7 g8
etc. - the rooks can check but they can't
deliver mate.
38 :as l:lbB 39 b4 :as 40 ltd1 l:lf6
41 l:d7 l:laf8 42 f4 l:l6f7 43 l:ld6
l:lb8 44 l:lxc6 l:lxb4 45 l:lcB+ g7
46 l:lg5+ f6 47 l:lc6+ e7 48
l:le5+ fB 49 l:lc8+ g7 50 Ag5+
51 l:lc6+ e7 52 l:lxa6
84 Wh6?
Missing a win with 84 lle7 llal
(84...llg8 85 f7 l%gf8 86 lte8 g7 87
:Xa8 ll.xa8 88 l:J.e8) 85 l:.e 1 lbe 1 86
l:.xel, followed by lte7 and h6.
84...l:la1 85 l:lh3 l:lg8 86 1le7 l:lg6+
87 h5 l:lg1 88 lle8+ llg8 89
l:lxg8+ xg8 90 l:la3 l:lh1 + 91 Wg4
l:lg1 + 92 4 llf1 + 93 g4 llg1 +
94 f4 l:lf1 + 95 l:[f3 1la1 ?!
Black could already force a draw with
95 ....:.xf3+! 96 q;xf3 f7, followed by
If there were only one pair of rooks ... h7-h6. Next time round he grabs this
on the board then Black should be able opportunity.
to draw. But now his king is a constant 96 l:lb3 llf1 + 97 e5 l:le1 + 98 4
source of worry, which adds considera llf1 + 99 l:lf3 l:lxf3+ 100 wxf3 wn
bly to the burden of defence. 101 Wf4 h6 102 gxh6 %-%
52 ...:Z.b8 53 h3 f8 54 h4 ltg7
55 l:lc5 g8 56 g4 l:lgb7 57 :Z.cc6 Game 34
l:lf7 58 f5 l:ld7 59 l:c3 :Z.db7 60 Honfi-Varga
l:lcc6 l:ld7 61 g5 g7 62 l:c3 Budapest 1995
l:ldb7 63 h5 l:lf7 64 l:lac6 l:la8 65
g5 l:lb8 66 ll3c5 l:[bf8 67 llb6 1 e4 lllf6 2 e5 llld 5 3 c4 ll:lb6 4 d4
:a7 68 f6+ g8 69 Wh6 :a1 70 d6 5 exd6 cxd6 6 l/Jc3 g6 7 .i.e3
:9s+ hB 11 l:lhs :n 12 9s :9s .i.g7 8 l:lc1 0-0 9 b3 e5 10 dxe5
73 :h4 l:.f5 74 :94 :n 75 l:.bb4 dxe5 1 1 'ifxd8 l:txd8 1 2 c5!
lth1 + 76 llh4 :9 1 11 llbg4 :n 78 The critical line. 12 ltlb5 is answered

88
Ex change Varia tion with 5 . . . cxd6

by 12 ... ltk6! 13 l0xa7 lbd4! 14 fOxeS satisfactory results.


fOxeS 1S .ixd4 exd4 16 a4 d3!, when a) 14.../lld4 15 /llgS llfS 16 ltke4
Black had a very strong initiative in the lllfs 17 0-0 {17 lhxf7!? llxf7 18 lllgs
game Chernisbov-Friesler, Pardubice lllh6 19 1lle6 {Olsson-Zetterberg, Bor
1992. lange 199S} should be met by 19 ... bS!
20 .idS l06! 21 .ixa8 .ixe6 with
complex play) 1 7.../llf6 18 /lld6 lhxd6
19 cxd6 .td7 20 llfd1 a6 21 f3 and
White bad an edge because of the
passed d-pawn in Dzindzichashvili
Alburt, US Ch. 1996.
b) 14...b6 15 llle4 {Emms has also
suggested 15 0-0) 15 .../llaS {15 ...l0f8 16
lhd6 %ld7 17 O-O ile7 18 l0xc8 1lxc8 19
llfd1 left White slightly better in Gross
Bagirov, Berlin 1996, whilst 15...l0d4
can be met by 16 lhd6 lhxf3+ 17 gxf3
1 2. lb6d7
.. llf8 1S llg1, threatening 19 Jlxg6, ac
And not 12 ... l0ds? 13 lld1 .ie6 14 cording to Raetsky) 16 .td5 (16 i.e2
.ic4, winning a piece. l0f8 17 lhd6 also gives Black a difficult
1 3 J.c4 lbc6 1 4 lbf3 game according to Sanakoev) 16 ...l0f6
Another possibility is 14 l0e4!? lllf8 17 lhxf6+ .ixf6 1S lld1 g7 19 0-0
15 l0d6 l0e6 16 l03 l0cd4 17 lOgS lteS 20 lhd2 and White had a definite
lllxgS 18 .ixgS lld7, which also gave advantage in Finkel-Drazic, Nova
White a slight edge in Yagupov-Ukolov, Gorica 1997.
Moscow 1996.

1 5 .i.g5!?
14 lba5
.. Certainly a tempting move, but pos
The bishop on c4 is public enemy sibly not the best. Both Emms and
number one. Black has tried a couple of Sanakoev suggest that the quiet 15 .ie2
alternatives but neither bas produced is much more difficult for Black, and it

89
A lekhin e 's Defen c e

does seem as if he is struggling in all .ib7 1 8 0-0 dS 19 Jlfd1 was played in


variations: Varga-Llanos, Budapest 1999, and now
Finkel suggested 19 ...xe3! 20 fxe3 e4
2 1 lbfd4 f8! with equality) 16 ... bxc5
17 0-0 c4 1 8 b4 lbc6 19 Lc4 lbxb4 20
a3 lLla6 2 1 lLlg5 (threatening both the
f7-pawn and .tdS) was deeply unpleas
ant for Black in the game Kiik-Hautala,
Tampere 2000.
1 5 . .1:lf8
.

Black has another possibility in


15 ... l:r.e8!?, the game Sanakoev-Honfi,
correspondence 1993-5 continuing 1 6
lbbS e4! 17 li1c7 ex3+ 1 8 lLlxe8 fxg2 19
a) 15 ... h6 16 0-0 lbf8 17 lbe4 lbc6 18 ltg1 lLlxc4 20 Jlxc4 5 2 1 ltc3 i.g4
d6 llb8 19 a3 f5 20 Jlfd1 e6 21 b4 22 lbxg7 lbf3+ 23 1lxf3 .txf3 24 f6
staned the key push of White's queen llc8 25 d2 {and not 25 b4 lk6 26
side pawns in Hunt-Schnabel, Oxford i.d4 f6 27 8 lte6+) 25...Jixc5 26
1998. i.d4, reaching a highly complex end
b) 15 lbc6 16 b5 8 17 6
.. game which is probably about equal,
e6 1 8 i.c4 cd4 19 lbgS also gave according to Sanakoev.
White strong pressure in Mitkov-Toth, 1 6 .i.e7 IC!xc4
Rio de Janeiro 2000. After 16 ... l:le8 there follows 17 lLld5
c) 15 ...lbf8 16 0-0 c6 17 llfdl i.fS xc4 1 8 ltxc4 e4 19 lbg5 with numer
1 8 bs lbe6 19 d6 ltab8 20 gS ous threats.
/i)xgS 21 i.xgS was again pleasant for
White in Mitkov-Dischinger, Sitges
1997.
d) 15...b6

1 7 .i.xf871
According to Sanakoev, Black is also
slightly worse after 17 bxc4 e4 18 i.xf8
exf3 19 J.xg7 fxg2 20 llg1 xg7 2 1
16 /i)a4!? {16 cxb6 lbxb6 17 lbb5 l:.xg2, though admittedly he would have

90
Ex change Varia tion with 5 . . . cxd6

good drawing chances because White's lbxd4 l:lxd2 30 lbxc6 bxc6 31 1


pawns are split. Now Black obtains ex Zlxa2
cellent play for the exchange.
1 7 .'xf8 1 8 bxc4 ltlxc5 1 9 0-0 e4
.

20 ltlg5 ltld3

Winning a pawn and maintaining his


rook in a dominating position on the
seventh rank.
21 lbcxe4? 32 Zlb1 lla4 33 l:lb7 aS 34 l:la7
After 21 :c2 Black can repeat the J:lxc4 35 Zlxa5 l:lc2 36 l:la7 g7 37
position with 2t...li:}b4 22 :cct li:}d3, h4 g5 38 hxg5 hxg5 39 g4 f6 40
but this would have been White's best e1 e6 41 l:lc7 f6 42 l:lc8 l:lc5 43
course. Inexplicably he gives back the e2 e5 44 We3 l:lc3+ 45 d2
exchange and then has to face the wrath l:lc4 46 d3 lld4+ 47 e3 :td6 48
of the two bishops. f3 d5 49 wd3 Wc5+ so c3 d5
21 ltlxc1 22 llxc1 h6 23 ltlf3 i.d7
.. 51 Wd3 c5 52 c3 :ta6 53 :td8+
24 ltJc5 i.e& 25 ltld2 l:ld8 26 ltlf1 e5 54 c4 l:la3 55 Zld5+ f4 56
l:le8 27 ltlb3 l:le2 28 lbfd2 i.d4! 29 :tf5+ g3 57 :txf6 :txf3 0-1

91
A lekhin e 's D e fence

Summary
The Voronezh is a problem, a big one in fact. Right now I'd only recommend this
line to White, though it is possible that Black may find a major improvement in the
sharp lines arising from 9 .f5 10 g3. He can probably get a solid enough game with
.

9 e6 (or even 9 .td7) but he is unlikely to achieve full equality in these lines.
... ...

The ost natural' move is 9 e5, but defending against the initiative White ob
..

tains is probably a rather thankless task. It could be that someone like Baburin will
eventually show how Black can draw; until that happens I'd preferto recapture with
the e-pawn on move 5.

1 e4 lilt& 2 e5 llld 5 3 d4 d& 4 c4 lllb& 5 exd& cxd& 6 lllc3 g& 7 .1e3 {D)
7 h3 - Game 30
7 .1g7 8 1lc1 0-0 9 b3 e5
..

9 ..tf5 - Game 32
.

9 lik6 {D} - Game JJ


...

10 dxe5 dxe5 1 1 1Wxd8


1 1 c5 - Game 33
1 1 llxd8 1 2 c5 lll&d7 1 3 .1c4 lllc& 14 lllf3 {D) - Game 34
. .

7 .1e3 9 . . . lllc 6 14 lllf3

92
CHAPTER SIX I
Exchange Variation
with 5 exd6. . .

1 e4 lt:lf6 2 e5 lt:ld5 3 d4 d6 4 c4 ried out. A plan that often crops up is a


lt:lb6 5 exd6 exd6 kingside pawn advance, usually with h2-
The simple 5 ... exd6 has always been h3 and g2-g4, often after White has cas
known as a solid line for Black, but the tled kingside! The idea is to deprive
rise of the Voronezh Variation (Chapter Black's pieces of good squares.
5) has thrown it into the limelight. The Game 35 shows a very standard treat
symmetry of the pawn structure is ment for both sides, though in this par
maintained with White claiming a space ticular game White used a slight twist in
advantage and a badly placed knight on that he delayed kingside castling. In
b6. On the other hand it can be argued Game 36 Black also used some subtlety
that White has slightly weakened his d in that he developed his queen's bishop
pawn. The pawn on c4 cannot move to fS rather than g4. Here the idea is
back to c3! that h2-h3 by White does not gain a
White has a wide choice of different tempo.
piece formations and move orders that Game 37 features an aggressive new
make an understanding of plans and try by White - an early ..f3 followed by
ideas more important than a slide-rule castling long. Games 38 and 39 show a
knowledge of exact variations. Black's more mature treatment with i.d3 fol
traditional method of counterplay has lowed by l'llge2.
been to play i.e7, castle short, de
..

velop his b8 knight and c8 bishop (the Game 35


latter to either f5 or g4) and then play Velicka-Freisler
...d6-d5. When White meets this with Czech Ch., Zlin 1997
c4-c5, the knight on b6 is recycled to f5
via c8 and e7. Pressure against d4! 1 e4 lt:lf6 2 e5 llJd5 3 d4 d6 4 c4
Of course White does not sit still l0b6 5 exd6 exd6 6 l0c3 .te7 7
whilst these manoeuvres are being car- i.e3!?

93
A lekhin e 's D e fence

doubled pawns and maintain a slight


space advantage. In the game he played
12 i.e3, with unclear complications
resulting from 12 ...i.xc3 13 bxc3 ll:le5
(13 ...lbf8!? looks like a good move to
me) 14 llxe5 lbe5 15 a4 'i'h4.
7 0-0 8 l0f3 .i.g4 9 .i.e2 l0c6
..

Once again it's worth considering


Agdestein's flexible 9 ...c6!? approach.
10 b3 .i.f6
Black can also first play 10...1le8,
when Djurhuus-Tisdall, Eikrem Memo
This is actually quite a subtle move rial, Gausdal 1996 continued 1 1 1fd2
order. White intends to delay castling so i.f6 12 0-0 d5 1 3 cS lbc8 14 h3 i.e6
that a later ... Ji.g4 by Black can be met (14...i.f5!?) 15 b4 a6 16 b5! axb5 17
by h2-h3 and g2-g4 without worrying ll:lxb5 ll:la5 18 i.f4 lte7 19 llfe1 b6 20
about weakening the king's defences. llac 1 c6 21 ll:ld6 ll:lxd6 22 .i.xd6 llb7
After the standard 7 lbf3 0-0 8 i.e2 23 ll:le5 bxc5 24 i.xc5 i..g5 25 f4 with a
i.g4 (8 ....tf5 9 0-0 transposes into the definite edge to White.
next game, Yagupov-Balashov} 9 0-0
Black can consider the flexible 9 ...c6!?
(9 ... lbc6 10 i.e3 i.f6 1 1 b3 d5 trans
poses into the note to White's 1 1th
move} 10 b3 lieS, when Mortensen
Agdestein, Reykjavik Zonal 1995 con
tinued 1 1 h3 .th5 12 i.f4 aS 13 l:tel
lba6 14 i.fl lbc7 15 l:tcl lbe6 16 i.e3
d5 17 c5 lbd7 with a complex struggle
Black can put pressure on d4 and break
up White's queenside pawns with ... b7-
b6.
7 h3 is an attempt to prevent ... ii.g4 1 1 h3!?
altogether, though this gives Black time Giving the game an independent fla
to adopt an alternative plan based on vour compared with the standard plan
...c7-c5. V Atlas-Loffler, Austrian Ch. of castling short.
2000 went 7...0-0 8 lbf3 c5!? (8 ... i.f5, After 1 1 0-0 d5 12 c5 lbcS 13 h3 (13
seems very reasonable here, with similar b4?! is dubious because of 13 ...lb8e7 14
play to Game 36, Yagupov-Balashov) 9 b5 lbaS 15 h3 i.xf3 16 i..xf3 c6 17
d5 i.f6 10 i.e2 lle8 (the immediate 'ifd3 ll:lc4 when Black's pieces find ex
1 o. . i.xc3+ also looks interesting) 1 1
. cellent squares and the d-pawn comes
0-0 ll)Sd7 and now White should under pressure, Gipslis-Larsen, Sousse
probably have played 12 i.. d2! to avoid Interzonal 1967} 13 ...i..e6 (more or less

94
Exchange Varia tion with 5 . . . exd6

the only move for Black as 13 ...i..xf3 is d) The other interesting move for
met by 14 i..xf3 8e7 15 g4 g6 16 ltc1 Black in this position is 14 ... g6.
and 13 ...i..f5 by 14 W"d2 h6 15 llacll White doesn't have much here which
8e7?f 16 g4 .th7 17 h4f ltlg6 18 g5 explains Velicka's attempt to introduce
hxg5 19 hxg5 i..e7 20 g2, with the a finesse based on delayed castling.
simple but highly effective plan of dou 1 1 e6
.

bling on the h-file, Kurajica-Hort, Som And not 1 1. ...i.h5 12 0.0 lte8 1 3
bor 1968) 14 W"d2 1Wd2 d5 1 4 c5 ltlc8 15 l:tad1, transpos
ing into Kurajica-Hort above.
1 2 li.le4!?
12 0-0 d5 13 c5 c8 would transpose
into Mortensen-Kengis above, which
was very reasonable for Black.
1 2 ... d5 1 3 li.lxf6+ xf6 14 c5 lt.lc8
1 5 ..d2 .i.f5
Compared to the positions arising
from 1 1 0-0, White has exchanged his
knight for Black's dark-squared bishop.
Probably this is in White's favour, for
although Black's position has been
and now: slightly freed he will find it difficult to
a) 14... h6 15 .td3 ltl8e7 16 g4 g5 17 put effective pressure against the d4-
e2 .i.g7 1 8 ltlh2, intending f2-f4, gave pawn.
White a strong attack in Armas-Graf, 1 6 b4 a6 1 7 0-0 lOSe7 18 a4 ll:lg6
West Germany 1989. 1 9 b5 axb5 20 axb5 ll:lce7 21 g41?
b) 14. . .b6 15 a4 J.d7 16 ltacl lle8 .i.d7 22 l:ae1
1 7 l:tfel ltl8e7 18 g4 h6 19 b4 g6 20
a3 ltlxd4 2 1 ltlxd4 .txa4 22 b5 put
Black in serious trouble in the game
Brynell-Baburin, Copenhagen 2000.
c) Black is doing okay after the flexi
ble 14 ... ltl8e7. Mortensen-Kengis, Mos
cow Olympiad 1994 continued 15 g4
g6! 16 ltle1 (after 16 b4 it looks quite
interesting to play 16...xb4 17 llab1
ltlbc6 1 8 llxb7 lLla5 19 l:tbb1 c4 20
i..xc4 dxc4 with the idea of ... h7-h5 in
the air) 16 ..i.g7 17 ltlg2 b6 18 cxb6
.

cxb6 19 f4 1Wd6 20 ltacl .Uac8 2 1 It now looks as if White is better; he


ltfdt ltlbS 22 ltlb5 1Wd7 23 l:txc8 l:txc8 has more space and Black's pieces are
24 ltct llxcl+ 25 W"xct ltlbc6 26 W"d2 awkwardly bunched on the kingside.
'h-'h. The downside is that his king is fairly

95
A lekhine 's D efen c e

exposed, which certainly presents Black 41 g3 l:la4 42 h4 b4 43 e6 fxe6


with practical chances. The following 44 .i.xe6 6 45 .i.f5 l:la3+ 46 2
moves look as if they were played under e5 47 l:lb1 l:lf3+ 48 e2 b3 49
pressure from the clock. d2 .i.d5 50 l:ld1 l:lh3 51 h5 .i.f3
22 h6 23 lbe5!? lbxe5 24 dxe5
. 52 l:lb1 .i.d5 53 l:ld1 l:lh1 54 l:lxh1
h4 25 g2 c6 26 f4 lbg6 27 .i.d4 .i.xh 1 55 e3 .i.d5 56 .i.b1 .i.e4 57
l:la3 28 l:lf3?? g5 hxg5 58 h6 6 59 h7 g7 60
A blunder which should have lost d4 .i.e2 0-1
immediately. Finkel suggested 28 .ic3 After 60....ie2 61 c3 .id 1 62 .ie4
cxbS 29 l:la1 llfa8 30 l:.xa3 l:lxa3 31 fS .ic2 63 .ixb7 xh7 Black will win the
0Je7 32 11'b2 with a complicated strug bishop with his g-pawn and then pro
gle. mote the b-pawn.
28 ...l:lfa8??
Missing an immediate win with Game 36
28 ... l::txf3 29 .ixf3 0Jxf4+! etc. Yagupov-Balashov
29 .i.f2 l:lxf3!? 30 .i.xf3 Moscow 1996
And not 30 .ixh4?! lCJxh4+ 31 gl
llaa3 32 bxc6 .ixc6 33 l:ldl l::txh3 with 1 e4 .!Llf6 2 e5 .!Lld5 3 d4 d6 4 e4
a ferocious attack. .!Llb6 5 exd6 exd6 6 ll:lf3 .i.e7 7 .i.e2
30 . ....e7 31 d4?1 exb5 32 .i.g3 0-0 8 0-0 .i.f51?
l:la4 33 d2 xeS This is often the reaction to an early
Suddenly Black is material up whilst h2-h3 by White, but in any case it seems
White's kingsicle advance is no further like a very sensible alternative to the
forward. development of the bishop on g4. If
34 f5 lbf8 35 f6 lbg6 36 fxg7 'xg7 White chooses the typical plan of a
37 xd5 xd5 38 .i.xd5 .i.e& 39 kingside pawn advance (h2-h3, g2-g4
l:ld 1 .!Llf4+ 40 .i.xf4 l:lxf4 etc.), it saves a tempo for Black.
9 .!Llc3 .i.f6 10 .i.e3 .!Lle6 1 1 b3 d5
12 e5 .!Llc8 13 ll:le1 h6 14 g4 .i.h7

The endgame offers Black excellent


winning chances, though there are some
practical difficulties. Compared to the Kurajica-Hort

96
Exchange Varia tion with 5 . . . exd6

game, given in the notes to Game 35, "ird7 27 llc1 "irb5


Black has an extra tempo through hav
ing played ... .i.f5-h7 rather than ..i.g4-
_

f5-h7. This cenainly improves his pros


pects because his counterplay kicks in a
move earlier.
1 5 l0g2 b6 1 6 l0a4
Ganging up on Black's d-pawn fails
to produce the desired effect: 16 cxb6
axb6 17 ll)f4 ll)Se7 18 .i.f3 ll)b4 threat
ens 19 ...ll)c2.
16 l0b4 17 1.0&1
.

Preventing 17...ll)c2.
1 7 bxc5 1 8 l0xc5 l0d6 1 9 "ild2 aS
. . Winning a pawn.
28 "ifxb5
After 28 IZ.c3 there is 28 ...11'xd3 29
IZ.xd3 llxe3! 30 llxe3 .i.xd4 etc.
28 l0xb5 29 J.f2
..

Or 29 ll)c2 ll)c3, threatening ...2+


and ... llxb3.
29 l0xd4 30 llxc7 J:lxb3 31 llc5

llxa3 32 J:lxd5 l0e2+ 33 Wh1 lla1


34 J:ld6 J.c3 35 J:lc6 .tb4 36 J:lc4
g5 37 J:le4 J:lxe4 38 fxe4 l0f4 39
.i.g3 0-1
After 39 ...ll)d3 Black wins a piece.
20 a3?!
This creates more weaknesses on the Game 37
queenside, notably the b3-square. Ac H.Hagesaether-Wohl
cording to Chekhov, White could con Ubeda 2000
solidate his position with 20 f3
{preventing ...li)e4 by Black) 20...lte8 21 1 e4 l0f6 2 e5 l0d5 3 d4 d6 4 c4
IZ.cl 'ile7 22 ll)g2, when both sides lOb& 5 exd6 exd6 6 "iff3?!
have chances in a complex game. A very sharp plan introduced by Vic
20 l0c6 2 1 llc1 Ilea 22 l0b77!
. . tor Korchnoi; White aims for queenside
It's too late for 22 f3 because of castling and an all-out onslaught against
22...lbe3!. Black's king.
22 l0xb7 23 llxc6 l0d6 24 f3
. 6 l0c6
..

And not 24 .i.d3?! because of An alternative treatment is 6 ....i.e7 7


24 ....i.xd3 25 ll)xd3 'ild7, forking c6 ll)c3 and now:
and g4. a) The odd-looking 7 ...11'd7?! was
24 . 1lb8 25 -t..d3 J.xd3 26 "ifxd3 tried in Saltaev-Marinkovic, Ikaria 1994

97
A lekhin e 's Defen c e

but after 8 i.e3 t:Lla4?! (8 .. .'ifg4 is 10...1i)c4!? (not many grandmasters


stronger, but slightly better for White would play such a loose-looking move,
after 9 ...xg4 i.xg4 10 h3) 9 1lld5 0-0 but the passive 10...1llc8?! gave White
10 b3 t:Llb6 1 1 ltJxe7+ exe7 12 .i.d3 the better game after 1 1 i.d3 d5 1 2
White had a clear edge because of his il)ge2 i.g5 1 3 h4 i.xe3+ 1 4 fxe3 'ifd7
useful bishop pair. 15 h5 in Osadchenko-Donchenko,
b) 7 ...0-0 8 i.e3 c6 9 0-0-0 {9 i.d3 d5 Moscow 1996), after which 1 1 i.f4 {1 1
1 0 c5 lllc4 1 1 i.cl b6 gave Black excel d5 tD6e5Q 1 1 ...'ifc8 12 'Wg3 6a5 13
lent counterplay in Grosar-Raspor, Por b3 tba3 14 cxd6 cxd6 15 'iPb2 b5 gave
toroz 1996) 9...d5 10 c5 1ll6d7 1 1 i.d3 Black an attack in Buenermann
(or 1 1 g4 b6 1 2 cxb6 axb6 13 i.d3 b5 Waechtler, German Bundesliga 1994.
14 a3 llla6 with a strong queenside ini 10 . . .85
tiative, Lakos-Zhukova, Women's Black can also interpose 10 ... 'i'c8 1 1
Olympiad, Yerevan 1996) 1 1 . .. b6 12 h4 h3 before playing 1 1 ...a5, at which point
lllf6 13 lllge2 (Camacho-diaz Perez, Seul-J Horvath, Budapest 1995 was
Pinar del Rio 1995), and now 13 ...bxc5 rather unhelpfully agreed drawn.
14 dxc5 1llbd7 would have given Black
a good game.
7 .i.e3 .i.e7 8 ltlc3 0-0 9 0-0-0 .i.e&

1 1 c5
Deciding to punch, rather than duck.
Alternatively White can try to blockade
10 b3 the queenside with 1 1 a4, when 1 1 ...d5
A logical and possibly superior alter (or 1 1 ...'ifc8 12 i.e2 lllb4 13 h3 dS
native is the move 10 d.S, after which 14 c5 t:Lld7 15 f4 1llf6 which also led
10 ... 1lle5 1 1 e4 .tg4 12 f3 f5 13 'iff4 to complex play in Gadjilu-Miles,
lllg6 14 ...d4 f4 15 .tf2 i.6 16 'ifd2 Linares 1998) 12 c5 IlleS 1 3 lllh 3 (13
i.f5 17 g4 fxg3 18 hxg3 was slightly lllge2 lllb4 14 f4 would transpose)
better for White but agreed drawn in 13 ...ltJb4 14 1llf4 11d7 15 1llxe6 fxe6 16
the game Donev-Zlochevsky, German 1Wh3 i.f6 17 i.b5 c6 18 i.e2 liJe7 19
Bundesliga 1996. g4 eS was good for Black in Mas-Wohl,
After the sharp 10 cS!? Black should Sydney 1999.
probably react with the risky looking 1 1 ...d7 1 2 .tbS?

98
Exchange Variation with 5 . . . exd6

ifa3+ 21 Wc2 1l\b4+

A poor move which helps Black ac


celerate his play on the queenside. 22 Wd2!
White should play 12 d5 /llce5 13 'i'e2 Trying to escape via el.
.*.fS 14 h3, intending g2-g4, with 22 llJxd3 23 xd3 ifxb3+ 24 ll::lc3
..

chances for both sides in a highly com If 24 d2 there follows 24 ....b4+ 25


.

plex and tactical position. lilc3 dS!, threatening ... d5-d4.


1 2 .../l\b4 1 3 1l\ge2 24.....tf6 25 exf7+ Wh8 26 ..td4?
13 9xb7? llb8 would be even worse. Having defended himself very accu
1 3 c6 1 4 .1d3 a41 1 5 /l\xa4
. rately White finally goes astray. He
should return the piece with 26 11'xc6!,
after which 26 ... L8 27 'i'dS llxc3+ 28
e2 ltc2+ 29 fl sees his king finally
escape. A draw would be likely after
29 ...'i'xd5 30 llxd5 ltxf7 etc.
26 .....txd4 27 Wxd4 d5! 28 :a 1
ifc4+ 29 We5 D.ad8! 30 l:lhc1 d4

1 5...b5!?
Attempting to decide the game with
an immediate attack. A simpler way to .
play the position was with the continua
tion 15 ...1llxa2+ 16 b2 lilb4, recover
ing the pawn with White's king still vul
nerable.
1 6 cxb6 /t)xa2 + 1 7 'b2 /t)xb6 1 8 31 ltla4
llJxb6 ifxb6 1 9 d51 Wa5 20 dxe6 Or 31 llle2 1Wb4 32 11fxc6 llxf7, once

99
A lekhin e 's D e fence

again leaving White's king hopelessly queen; 1 1 11fg3? is a blunder because of


placed. 1 1...h4) 1 1 .. xc3+ 12 bxc3 1i'f6 13
31 ...'irb4 32 l0c5 d3 33 'ire4 'irb2+ d2 ltlc5 14 11fg5 1i'xg5 15 fxg5 f5
34 Wf4 'irxf2+ 35 g5 with more than enough for the sacri
Or 35 g4 lld4, winning the queen. ficed pawn because of the weakness of
35 ... h6+ 36 Wh5 l%d5+ 37 g4 c4.
l%g5+ 38 h3 l%h5+ 39 g4 l%h4 b) White is probably well advised to
mate (0- 1 ) delay this advance. Emms-Baburin, Port
.-------. Erin 1997 continued 7 ltlf3 g4
Game 38 (7...JJ..e7 8 dS!? I.Oe5 9 lbxe5 dxe5 10
Davies-Sinha e2 0-0 11 0-0 f5 was okay for Black
Calcutta 1990 in Palliser-Miles, British Ch., Scarbor
_____________.
ough 2 00 1) 8 e2 i.e7 (8 ...xf3 9
1 e4 l0f6 2 e5 l0d5 3 d4 d6 4 c4 xf3 etlxc4? 10 1i'e2+! 1i'e7 1 1 etldS
l0b6 5 exd6 exd6 6 l0c3 1i'xe2+ 12 .i.xe2 is strong for White) 9
dS! i.xf3 10 xf3 I.Oe5 1 1 J.e2 0-0 12
b3 f6 13 i.b2 with the better game
for White.
7 J.d3 lbc6 8 l0ge2

6 ... J.e7
Black can also try 6 ...lbc6!? in an at
tempt to prevent White from adopting
the d3 plan. The drawback is that the
knight can be driven away immediately, 8 ...J.g4
though the immediate attempt at refuta For the sensible 8 ..0-0 see the next
.

tion seems to faJl short: game {Sermek-Zelcic).


a) A game between Ara and Artashas 9 f3 J.h5
Minasian, Armenian Ch., Yerevan 1999 Or 9...J.h4+ 10 lDg3 J.h5 1 1 0-0
continued 7 d5!? etle5 8 f4 ltled7 (and J.xg3 12 hxg3 0-0 {12 ... J.g6 13 1le1+
not 8...etlexc4? 9 i.xc4 lbxc4 10 11'a4+) f8 14 b3 'iff6 15 e3 lle8 16 11'd2
9 1i'd4 J.e7! 10 9xg7 f6 1 1 1i'g4 {1 1 J.xd3 1 7 11'xd3 1l'g6 18 11'd2 1i'xg3 19
1i'h6 9e7+ 12 d1 lDc5 gives Black c5 lDd7 20 J.f4 1l'g6 21 etlb5 gave
very strong play for the pawn because White a winning attack in Boleslavsky
of the offside position of White's Kopylov, USSR Ch. 1949 and

7 00
Ex change Varia tion with 5 . . . exd6

12 ... lt)xd4? loses a piece after 13 g4 lbb6 5 exd6 exd6 6 lL!c3 i.e7 7
.ig6 14 .txg6 etc.) 13 b3 :e8 14 g4 i.d3 0-0 8 lL!ge2 lL!c6
.ig6 15 .ixg6 hxg6 1 6 d5 and White Another possibility for Black is 8 ... c6,
had a nice space advantage in Panov meeting 9 1i'c2 h6 10 0-0 with lO.. .dS
Mikenas, Moscow 1942. 1 1 cS lb6d7, followed by ...b7-b6. It
10 0-0 i.g6 1 1 i.xg6 hxg6 1 2 d5 looks like a solid way to play the posi
l0e5 1 3 b3 .d7?! tion and may not leave Black suffering
Unrealistically hoping for some play from the same lack of space as the lines
on the kingside. 13 ...0-0 was better, in- in which White kicks the knight on c6
tending ...IZ.e8 and ... .tf6. with d4-d5.
9 0-0

1 4 i.e3 'iff5 1 5 lL!d4 'ifh5 1 6 h3


End of 'attack'. Now White's well 9 ...i.f6
placed pieces and space advantage start The most popular move which ap
to tell. plies pressure to the d4-pawn. There are
1 6 . . ..-h4 1 7 :e1 a6 1 8 f4 lL!ed7 1 9 several alternatives:
lL!f3 'ifh5 20 i.d4 lL!f6 21 i.xb6 a) 9 ...lbb4 sets out to 'win'the bishop
cxb6 22 2 lL!g8 pair, but White can simply retreat it with
This is not a good sign for Black. 10 .ib1 (it turns out that 10 ...lt)xc4?
23 3 0-0-0 24 Wxb6 i.f6 25 loses to 1 1 a3 llX6 12 1i'd3). Kaminski
lL!e4! ..te7 Baburin, Biel 1995 continued 1O ... aS!?
After 25 ....txa1 White mates with 26 1 1 b3 lle8 12 e3 a4!? 13 lt)xa4 lt)xa4
lbxd6+ llxd6 27 lle8+ d7 28 'ild8. 14 bxa4 llX6 15 3 lt)aS 16 .id3 c6
26 /Od4 lL!f6 27 lL!c6! 1 -0 17 1i'c2 g6 18 h3 d5 19 cxdS cxdS 20
27... bxc6 28 dxc6 wins easily. l:lab 1 and White was better because of
.----- his pressure against the d5 and b7
Game 39 points.
Sermek-Zelcic b) 9 ...lle8 is a good, flexible move
Croatian Team Ch., Makarska 1994 that has only been seen in one game
_______________. that I know of. Atkinson-Fogarasi,
1 e4 lL!f6 2 e5 l2Jd5 3 d4 d6 4 c4 Groningen 1 989 continued 10 a3?! f6

101
A lekhin e 's Defence

1 1 .i.e3 g6 12 'irc2 fi)e7 13 fi)g3 d5 14 .tg6 13 b3 J.xd3 14 'irxd3 d5 15 cS


c5 fi)d7 15 b4 fi)f8 16 bS h5 17 fi)ge2 fi)c8 16 b4 l0xb4 17 'irb1 lflc6 18
fi)fS with an excellent game. 'irxb7 lfl8e7 19 ltabl, Sermek-Zelcic,
c) 9 ....i.g4 is similar to the 8 ....i.g4 Pozega 2000.
from Davies-Sinha. White gets more
space after 10 f3 i.hS 1 1 fi)f4 .i.g6 12
.i.xg6 (12 fi)xg6 hxg6 13 d5 fi)eS 14 b3
.i.f6 15 .i.b2 aS gave Black counterplay
in Ilfeld-Zilberman, Israeli Ch., Tel
Aviv 1994) 12 ...hxg6 1 3 dS lfleS 14 b3,
Djuric-Miles, Aegina 1993, and now Zil
bermans plan of 14 ....i.f6 (Miles played
14... 'ird7?! which takes d7 away from
his knight on b6) 15 .i.b2 aS looks
reasonable.
d) 9 .. .f5 is a double-edged way of
gaining space on the kingside. 1 1 b3
Malashenko-Cistiakova, Decin 1998 Emms's suggestion of 1 1 lflg3 is
continued 10 J.e3 (10 b3 looks worth considering.
stronger) 10 .. .ch8 1 1 f3 lflb4 12 a3 1 1 ...xd3 1 2 'iVxd3 i.g4 1 3 13 .th5
fi)xd3 1 3 'irxd3 d5 14 cS lflc4 with 14 g3 i.g6 15 'iVd2 i.h4 16 d5
counterplay. Taking the familiar space advantage,
10 .tel which pretty much guarantees that
White will be better. The big question is
whether or not he will be able to make
further progress. Blacks position is
cramped but sound.

10...b4
On this occasion Black succeeds in
obtaining the bishop pair as after 1 1
.i.b1 fi)xc4 1 2 a3 he can capture the
bishop on e3. 1 6 ...:ea 1 7 :ae1 d7 18 :e2
10 ... .i.g4 is still possible but led to a i.xg3 1 9 hxg3 a6 20 :te 1 lOIS 21
plus for White after 1 1 h3 .thS 1 2 'ircl2 .td4 ..d7 22 4 i.xe4 23 :xe4

1 02
Exchange Varia tion with 5 . . . exd6

:xe4 24 :xe4 :ea 25 WVe3 :xe4 kept both his space edge and bishop
26 WVxe4 f6 27 2 7 28 g4 h6 versus knight. But it's not enough to
29 e3 WVe7 30 WVxe7+ xe7 WID.
31 e4 a5 32 c5 lbg6 33 cxd6+
xd6 34 .i.f2 lbe7 35 .i.g3+ d7
36 d4 c6 37 dxc6+ xc6 38 .i.e1
b6 39 .i.d2 b5 40 .i.f4 c6 41 a4
lbg6 42 .i.g3 fa 43 e4 e6 44
f5 lbd4+ 45 g6 b3 46 xg7
c5 47 Wxh6 l.tlxa4 48 g6 l.tlc5
49 xf6 a4 50 .i.e 1 lbd3 51 .ic3
b5 52 g5 b4 53 g6 bxc3 54 g7 c2
55 g8WV c1WV 56 WVe8+ wc7 57
WVxa4 WVh6+ 58 5 WVh5+ 59 wt6
% -%
White is still slightly better. having

1 03
A lekhin e 's D efenc e

Summary
Black must know what he's doing in these lines in order to avoid the kind ofthing
that happened in Kurajica-Hon, given in the notes to Game 35. Having said that,
he can obtain excellent counterplay if he plays the position precisely. Yagupov
Balashov is certainly worth studying, as is Mortensen-Kengis, given in the notes to
Game 35. I also l ike Agdestein's flexible ... c7-c6 treatment, also given in the notes
to Game 35.
One of the best ways to play it with White is with .td3 and ltlge2, which cer
tainly calls for accurate play on Black's part if he wants to avoid falling into a pas
sive position. In this case I like Fogarasi's play against Atkinson in the note within
Game 39. Black avoided any fruitless development of his queen's bishop, instead
concentrating on the reorganisation of his kingside.
The Exchange Variation is one of the two lines I'd recommend for White against
the Alekhine, the other being the Classical. Against 5 ...exd6 I suggest the J..d3 and
ge2 set-up.

1 e4 f6 2 e5 d5 3 d4 d6 4 c4 b6 5 exd6 exd6 6 1003 (D)


6 3 Game 37
-

6 ltl3 J..e7 7 .i.e2 0-0 8 0-0 i.5 9 ltlc3 J..6 10 J..e3 etk6 - Game 36
6 ... .ie7 7 .id3
7 i.e3 0-0 8 f3 i.g4 9 i.e2 etk6 10 b3 i.f6 1 1 h3 - Game 35
7 . . . 0-0
7 llk6 8 ltlge2 J..g4 (D) - Game 38
..

8 ge2 c6 9 0-0 .if6 1 0 .ie3 (D} - Game 39

6 l'Dc3 8 . . . .ig4 10 .ie3

1 04
CHAPTER SEVEN I
Four Pawns Attack :
Main Line

1 e4 f6 2 e5 d5 3 d4 d6 4 c4 White's ninth move; Black must decide


b6 5 f4 dxe5 6 fxe5 c6 7 .i.e3 how he intends to tackle the centre.
.i.f5 8 c3 e6 9 f3 With 9 .....ie7 he intends to castle and
The Four Pawns Attack is White's play ... f7-f6 (Game 40}, but he must also
sharpest line against the Alekhine and take account of the fact that White can
represents an attempt at outright refuta play 10 dS (Games 41 and 42).
tion. The reason that it is not at the One line that has been quite popular
front of the book is more a question of of late is 9 . ..tg4 (Game 43}, which im
..

popularity; in practice it is far more mediately puts pressure on d4 by


likely that Black will meet either 4 liJf3 threatening to eliminate the knight on
or the Exchange Variation. f3. Existing theory applauds the idea,
White has taken as much space as he but remain unconvinced after 10 ..ie2.
possibly can, sending four of his pawns Does Black get compensation for the
forward with gain of tempo. The prob two bishops?
lem with this massive expansion is that Games 44 and 45 deal with 9 . '1fd7
. .

pawns are unable to move back. If the and 9 . . ..ib4 respectively, neither of
.

centre collapses White's entire position which have received much attention but
may implode. both of which look playable.
Black's traditional method of com
bating the Four Pawns Attack is to de Game 40
velop his pieces before attempting to Yudasin-Kengis
undermine White's pawn structure. Minsk 1985
Both sides get the opportunity to de- ._______________.
velop their pieces before commencing 1 e4 f6 2 e5 d5 3 d4 d6 4 c4
battle which is how people played be lbb6 5 f4 dxe5 6 txe5 c6 7 .te3
fore they broke all the rules! .i.t5 8 c3 e6 9 t3 .i.e7
The critical position comes after This natural developing move is

1 05
A lekhin e 's D e fence

Black's most popular continuation, equilibrium, with Black's weak pawn on


though it does require a good knowl e6 being compensated for by the weak
edge of theory after the sharp 1 0 d5. pawn on d4. The more forceful 14 c5
led to equality after 14...llJd5 15 -t.f2
l:lad8 16 Z:fe1 llJdb4 17 .ic4 llJc2 18
.ixe6+ i.xe6 19 'ifxc2 lbxd4 20 llJxd4
.ixd4 21 i.xd4 ltxd4, Garma-Alburt,
New York 1993.
14 ...1lad8 15 Wfc1 h6 16 h3 .i.h7
1 7 llfe1 h8 18 .i.f1 f7 19 h1
lbc8!
Improving the position of the poorly
placed knight on b6.
20 .i.f2 d6 21 a3 e5 22 c5 exd4
23 lbxd4 .i.xd4 24 .i.xd4 lbxd4 25
10 .te2 0-0 1 1 0-0 f6! 1 2 exf6 llxd4 liJf5 26 llxd8 llxd8 27 Wff4
After the sharp 12 lbh4?! Black ob %-%
tained excellent counterplay in
Arakelov-Bagirov, Baku 1960: 12 ... fxe5 Game 41
1 3 lbxf5 exf5 14 d5 lbd4! (offering a Pegoraro-Henderson
pawn to gain strong play on the dark Ischia 1996
squares; 14 ... lbb8 15 c5 is too passive)
15 .i.xd4 exd4 16 'ii'xd4 llJd7! 17 h 1 1 e4 f6 2 e5 liJd5 3 d4 d6 4 c4
.i.c5 (17 ....i.d6 is not bad either) 18 b6 5 f4 dxe5 6 fxe5 c6 7 .i.e3
'ii'd3 'ii'g5 19 liJbS .:lae8! 20 lbxc7 .Ue3 .i.f5 8 lbc3 e6 9 ll)f3 .i.e7 1 0 d5
21 lbe6 1fh6 22 lbxcS lbxcS 23 1ic2 exd5 1 1 cxd5 b4 12 li)d4 .i.d7
lbe4 24 g1 'ii'b6 25 h1 l:t6 26 lU3
l:lx3 27 .i.xf3 lbg3+ 0-1.
1 2 ....i.xf6 1 3 d2 7 1 4 llad1

13 ..,3
This is currently White's most popu
lar choice. An alternative method of
The position is in a state of dynamic defending the dS-pawn is with 13 1fb3.

1 06
Four Pa wns A ttack: Main L in e

Tsarev-Palatnik, Kiev 1989 continued a) After 19 i.e2 Black should proba


13 ...c5 14 dxc6 (14 f3 c4 15 -'lxc4 bly play 19 ... ..td5 (if 19 ... c5 20 .tg5
lDxc4 16 'ilxc4 2+ 17 f2 lDxa1 18 1We5 21 e7 cxd4 22 exf8e + l:txf8 the
llxa1 is another possibility in which move 23 1i'xd4 is now possible because
White has some compensation for the White's bishop is no longer on b5;
exchange) 14...bxc6 1 5 0-0-0 (after 15 Black is unable to play 23 ...xa2+ be
lld1 b8 16 e6 fxe6 17 f3 a5 18 lDe4 cause of 24 lDxa2 .xe2 25 llxhl) 20
lD6d5 Black had successfully consoli ..tgS 11fe5 21 e7 IU2 (21...l:tfe8 and
dated in Polajzer-Kovacevic, Maribor 21. ..l:tf7 are also possible) 22 a3 c5 23
1980} 15 ...c7 16 ..tc4?! (16 e6 fxe6 17 axb4 cxd4 24 l:txd4 h5!, which was at
lDxe6 11fe5 18 l:txd7 is the critical line, least equal for Black in Marjanovic
according to Palatnik; I would need Cicovacki, Sombor 1978.
convincing about White's compensation b) 19 ..tb5 c5 20 ..tgs 11fe5 21 e7
- Black can anchor a knight on d5) cxd4 22 exf8+ llxf8 23 llxhl (this
16...c4 17 xc4 1i'xe5 18 l:the1 0-0 time 23 xd4 is met by 23 ... lDxa2+ 24
19 a3 d5 20 xc6 ..txc6 21 lDxdS lDxa2 exbS 25 llxhl? .c6+) 23 ... dxc3
..txa3! 22 bxa3 {Or 22 lbc3 llfc8} {this is Black's simplest option, though
22 ...l:tfc8 23 -'lc5 ..txdS 24 llxe5 -'lxc4 23 ... h6 and; 23 ...a6 are also possible) 24
25 b2 f6 0-1. exb4 cxb2+ and the exposed white
The immediate 13 e6 leads to wild king gave Black an easy game in Murey
complications aher 13 ... fxe6 14 dxe6 Alburt, Beersheba 1980.
..tc6 15 11fg4 i.h4+ 16 g3 -'lxh1 1 3 ...c5
{16.....tf6? 17 0-0-0 ..txh1 18 lDf5! is 13 ...0-0 14 a3 c5 15 axb4 cxd4 16
known to give White a winning attack) ..txd4 i.xb4 has been suggested by
17 0-0-0 (17 ..tb5+ c6 18 0-0-0 0-0 19 Alekhine guru Lev Alburt. Black evi
gxh4 h5! 20 .g3 cxb5 21 i..g5 1ib8 dently hopes that the pawns on e5 and
was okay for Black in Velimirovic dS will be vulnerable but it remains to
Kovacevic, Yugoslav Ch. 1984) 17 ... 0-0 be seen how this is so aher, say, 17
18 gxh4 f6 .td3.
14 dxc6 bxc6

and now:

707
A lekhin e 's Defenc e

14...l0xc6 is bad because of 15 e6 xd7 21 1fg4+ c7 22 axb4 lDa2+ 23


fxe6 16 i.d3, with the threat of 17 b1 lDxb4 24 i.c4, when White had a
11fh5+, and 14...i.xc6 is better for White strong attack in Velimirovic-Kovacevic,
after 15 l0xc6 l0xc6 (1 5... bxc6!?) 16 Yugoslavia, 1978 .
.tb5. 1 9 b1 i.xe6! 20 :xd8+ :xd8 2 1
1 5 e6 axb4 lbxb4
Or: Black has only a rook, knight and
a) 15 a3 c5! 16 axb4 cxd4 17 i.xd4 pawn forthe queen, but the main factor
0-0 is equal, according to Hanston. is that White's king is very exposed.
b) 15 o-0-0 o-o 16 lDf5 lD6d5 17 22 i.e2
l0xe7+ 1i'xe7 18 lDxdS cxd5 19 a3 Black is also winning after either 22
llfc8+ 20 b1 lDc2 21 .tel J:lab8 gave hS+ i.f7 23 1fg4 i.g6+ 24 at
Black a winning attack in Trkaljanov lDc2+, or 22 e4 J:ld1 + 23 .tel i.a2+.
Kovacevic, Stip 1979. 22 ...:t8 23 9h5+
1 5 ...fxe6 Or 23 1l'g3 i.fS+ 24 al l:ld5!, again
15 i.xe6 16 a3 lD4d5 17 lDxc6 1ic7
.. with a winning attack.
18 l0xe7 11fxe7 19 i.bS+ is good for 23 ...g6 24 9e5 .t.f5+ 25 <i>a1 :ds
White. 26 9b8+ *t7 0-1
1 6 0-0-0
16 a3 l04d5 17 i.d3 l0xe3 18 1ixe3
0-0 19 11fe4 g6 20 h4 e5 2 1 lDf3 .tfs 22
i.c4+ g7 23 1fe2 l0xc4 24 1fxc4
1ib6 25 0-0-Q J:lab8 26 J:ld2 i.xa3 was
winning in Velimirovic-Kovacevic,
Yugoslav Ch. 1979.
1 6 ...lb6d5 1 7 a3 lbxc3 18 li)xe6
lbca2+!

27 11'xa7 lDc2+ 28 a2 lDxe3 29


xe3 liaS+ 30 b3 ltb8+ would be a
massacre.

Game 42
Paramonov-Bratchenko
PetroffMemorial, St Petersburg 2000

1 e4 lbt6 2 e5 lbd5 3 d4 d6 4 c4
This recommendation of Bernard lbb6 5 f4 dxe5 6 fxe5 lbc6 7 i.e3
Caffeny is an important improvement i.f5 8 lllc3 e6 9 lt)f3 i.e7 10 d5
over 18...1fa5 19 lDxg7+ d8 20 :Xd7+ li)b4

1 08
Four Pa wns A ttack: Main L in e

I doubt that this will be played very


much in future. White has two good
lines.
1 1 d4!?

18 .te2
Or 18 d6 .i.d8 19 .i.e2 'ila2 20 .i.d4
.i.e4 21 lift i.xg2 22 llxf8+ xf8 23
.i.d3 .i.h4+ 24 d2 as in Buchnicek
This may be even better than the Krajnak, Stary Smokovec 1996. With
more traditional 1 1 llcl, though Black White having a supponed passed pawn
has some problems there too. Velimi on d6, Black will be struggling to hang
rovic-Marovic, Yugoslavia 1977 contin on in these endgames.
ued l l. .. exd5 (11. .. .i.g4 12 a3 6 13 1 8 ...82 1 9 J:lf1 ?
,j_e2 0-0 14 0-0 .i.cS 15 AxeS lbxc5 16 A serious mistake; for reasons known
b4 lbcd7 17 'ifd4 .i.xf3 18 ,j_xf3 was best to himself, White suddenly pre
also good for White in Vodicka sents his opponent with his own passed
L.Smejkal, Czech Team Ch. 1997/98) b-pawn. White should play 19 d6 J.d8
12 a3 c5 13 axb4 d4 14 Axd4 cxd4 1 5 20 .td4 as in the Buchnicek-Krajnak
lbxd4 1Wb8 16 lbxfS 1Wxe5+ 1 7 .i.e2 game above.
'ifxf5 18 c5 lbd7 19 ltld5 AdS 20 llc3 1 9 ...W'xb2 20 J:lxf8+ xf8 21 d6
0-0 2 1 ltle3 1l'e6 22 Ag4 f5 23 .i.f3 .td8 22 .td4 11a2 23 11a1 b3 24
with strong pressure. 11b2 xb2 25 .t.xb2 .i.e4 26 .tf3?
1 1 ... .t.g6 1 2 a3 c5 1 3 xe6!
The key move. 13 axb4 cxd4 14
11'xd4 i.xb4 1 5 dxe6 11xd4 16 .i.xd4
0-0 17 exf7+ .i.xf7 gave Black the initia
tive in Blazek-Kantorik, Slovakian Ch.
1995.
1 3 fxe6 14 axb4 cxb4 1 5 a4 o-o
..

Gipslis once suggested 15...lbd7, but


then 1 6 1id4 11'a5 17 d6 .i.d8 18 c5 b5
19 b3! (19 ...bxa4? 20 llxa4) is good for
White.
1 6 xb6 axb6 1 7 J:lxa8 xaB

1 09
A lekhin e 's D efen c e

Exchanging light squared bishops is a b) 11. ..'ii'h4+ 12 .tf2 'ii'f4 13 c5


further mistake; Black's king can come
marching in on the squares it used to
protect. 26 g3 was the right move, when
White should be able to draw.
26 .i.xf3 27 gxf3 7 28 d2?!
..

This makes it easy for Black. White


had to stop Black's king from coming in
on the kingside, and for that reason 28
f2 would have been better.
28 ...g6 29 c3 f5 30 xb3 4
31 b4 xf3 32 c5 e4 33 b5
d5
33 ... bxc5 34 xeS gS 35 d7 g4 36 with a further split:
d6 fS, followed by a march of the b1) Aher 13 . ../i]d7 14 .tb5 .te7
kingside pawns, is also winning. ( 14 .. .f6?! has been suggested by Burgess,
34 cxb6 g5 35 .i.c1 h5 36 .i.d2 g4 but it looks bad aher 15 'ii'b3) 15 0-0
37 .i.e 1 h4 38 .i.f2 xe5 39 c5 .th4 16 .txc6 bxc6 17 'ii'a4 0-0 18
e4 40 d7 f3 41 .i.d4 g3 42 hxg3 lZ.ad1 White was simply threatening to
h3 43 .i.e5 h2 44 .i.c7 h1'it' 45 take the pawn on c6 in Koch-Konopka,
.i.xd8 'ifd1 0-1 Clichy 1993.
b2) 13 ...ttJd5 14 tlJxdS exdS 15 'li'd2
Game 43 'ii'xd2+ (the endgame is certainly un
Fedorov-Baburin pleasant for Black, but in any case the
Istanbul Olympiad 2000 position seems cramped and passive;
'------_. 15 ...'flf5 16 'ii'd3 'ii'd7 17 f4 tlJe7 18
1 e4 lL\f6 2 e5 lL\d5 3 d4 d6 4 c4 :gt h5 19 i.fl g6 20 .th3 was also
lOb& 5 f4 dxe5 6 fxe5 lL\c6 7 .i.e3 good for White in Bongers-Etmans,
.i.f5 8 lL\c3 e6 9 lL\f3 .i.g4 10 'it'd2 Dieren 1998) 16 xd2 g6 17 f4 .th6 18
This is currently thought of as the .te3 tlJe7 19 .i.d3 tiJfS 20 .i.xfS gxfS 21
main line, but it is not necessarily the lZ.hg1 d7 22 :g3 lZ.ag8 23 b4 lZ.g6 24
most testing move for Black. The quiet 'iti>e2 .tf8 25 :gs e6 26 lZ.hs h6 27 bs
10 .te2 has been looking quite venom i.e? 28 a4 llb8 29 lZ.b1 and Black was
ous: 10....txf3 1 1 gxf3 and now: under massive pressure in Timman-De
a) Aher 1 1 . ..'ii'd7 White should play Firmian, Malmo 2001.
12 'ii'd2 which takes the games along 10 ... .i.b4!?
similar lines to 10 1i'd2 'ifd7 1 1 0-0-0 This is Black's most trendy move,
but with Black having played 1 1.. ..txf3 trying to provoke White into playing a2-
rather early. Instead 12 f4 l:Z.d8! 13 dS a3 so that a later ...tlJc6-a5 will have
.tb4! 14 .txb6 axb6 15 .tf3 exd5 16 greater force. He can also consider two
cxdS 'i'h3 was good for Black in van other options:
der Wiel-Vaganian, Ter Apel 1993. a) A young Jan Timman played

1 10
Four Pa wns A ttack: Main L in e

10...e7 1 1 0-0-0 f6!? and went on to 1981. Last, but not least, it isn't good to
win against Schenstok in a 1968 game in play 12 0-0-0 because of 12 ...lha5 13
Holland. The game continued 12 exf6 1i'c2 .i.f5 - yet another advantage of
.i.xf6 13 lhe4 0-0 14 e2 1i'e8 15 lbfg5 Black keeping his bishop.
f5 16 g4 xg5 17 lhxg5 g6 18 h4 1 2 1Wd7 13 b4
.

h5 19 gxh5 f5, which looks very dan White tries to profit from the move
gerous for Black but probably isn't that 1 1 a3, but these space gaining opera
easy. tions leave huge gaping spaces behind
b) 10 ...'ifd7 1 1 .i.e2 0-0-0 (the posi the ranks. The game Hiibner-Hort, Biel
tion after the voluntary 1 1.. ..i.xf3 12 1987 went 13 .i.e2 0-0-0 14 0-0-0 f5
gxf3 can also be reached via 10 e2 15 lhg3 i.g6 16 h4? (16 .i.d3 is better)
xf3 1 1 gxf3 'ifd7 12 'ifd2 and seems and now Black uncorked the startling
quite difficult for Black after the further 16 ...lbb4! 17 b3 (17 axb4 'iia4 18 .td3
12 ...0-0-0 1 3 0-0-0 b4 14 a3 e7 15 lbxc4} 17 ...lhc2 18 c5 1fc6! 19 .id3.
lhe4 f6 16 exf6 gxf6 17 :hg1) 12 c5! Here 19 ... lhxa3 20 b2 xd3 21 Wxd3
xf3 (12 ... lhd5 13 lhxd5 'i'xd5 14 b4 i.xc5 would have been the crispest way
a6 15 a4 looks very strong - White can to finish matters according to Hort.
still castle kingside!) 13 cxb6 xg2 14
b5 a6 (14...xht 15 bxa7 will be fol
lowed by promotion) 15 1i'xg2 axb5
(Shabalov-Kengis, Riga 1989} and now
Shabalov gave 16 a4! b4 17 bxc7 'ifxc7
18 lhb5 1i'd7 19 0-0 as being good for
White.

1 3 .txf3! 14 gxf3 0-0-0


..

According to Baburin, Black can play


14...l:Z.d8 15 lld1 .i.h4+ 16 lhg3 f6 17 f4
0-0, which also leaves White's game
looking very ropy.
1 5 l:ld1 .th4+ 1 6 g3 f6!
Commencing undermining opera
1 1 a3 .te71? 1 2 e4?! tions.
It's too late for 12 e2 because of 1 7 b5 e7 18 1Wa5 Wb8 1 9 .te2
12 ...lha5. After 12 l:Z.d1 Black can open f5 20 .tf2 1Wf71 21 f4 g51 22
it up with 12 .. .f6, while 12 b4 .i.xf3 13 xf5 .txf2+ 23 wxf2 exf5 24 d5
gxf3 .i.h4+ cost White his castling rights fxe5 25 fxe5 g4
in Kremenietsky-Vaganian, Moscow White's centre has remained intact

17 1
A lekhin e 's De fenc e

but his king is horribly exposed. An lbb6 5 f4 dxe5 6 fxe5 lbc6 7 .ie3
other good line was 25 ...'i'e7 26 "ffc3 .if5 8 lbc3 e6 9 lilf3 'fld7
t'Lla4 27 "ffd4 t'Llc5, with the knight Another way of putting pressure on
threatening to come into the e4-square. d4 which has many similarities to
26 Wc3 h5 9...i.g4.
10 d5
A tricky line in which the complica
tions are still unresolved. What is clear
is that White must have nerves of steel
to play this way as his king may have to
spend some time in the centre.
White's most testing line is 10 i.e2
after which the once traditional
10... 0-0-0 11 0-0 (1 1 "ffd2 t'Llb4)
1 1 ...i.g4 now seems very dangerous for
Black because of 12 c5! (12 t'Llg5!? lDxc4
13 llxf7 'i'e8 14 i.f2 h6 15 i.xg4 hxgS
27 e6 16 llxf8 l:txf8 17 "ffe2 t'Llb6 is uncon
Perhaps 27 d6 would have been bet vincing) 12 ...t'Lld5 13 t'Llxds 'i'xd5 14
ter, but in any case White is in trouble. b4!? (14 t'Llg5 i.xe2 15 'i'xe2 lDxd4 16
27 . . .'fle7 28 h3 J:lhg8 29 hxg4 hxg4 i.xd4 'i'xd4+ 17 h1 'i'd2 18 ..xd2
30 Wd4 f4 31 .id3 Wg5 32 .te4 f3! lhd2 19 llxf7 i.xc5 20 t'Llxe6 i.d4 21
33 c5 lbc8 t'Llxd4 llxd4 led to drawish rook end
Baburin also mentioned the line games in Ligterink-Gipslis, Amsterdam
33 ... g3+ 34 xf3 g2 35 llhg1 llgf8+ 36 1976 and Gipslis-Kengis,Jurmala 1983)
e2 "ffg4+ 37 <it>e3! lDxds+ 38 i.xdS 14 ...'ife4 15 1i'b3 t'Llxd4! 16 t'Llxd4 i.xe2
'i'g3+ 39 e2 llxciS, but bringing the 17 lbxe2!? lld3 18 1i'a4 1i'xe3+ 19 h1
knight round is much simpler. 'ifxe2 20 'i'xa7 ltd2 21 "iVa8+ d7 22
34 J:lh7 g3+ 35 Wxf3 g2 36 J:lg1 'i'xb7
J:tdf8+ 37 We2?
37 llf7! is White last chance.
37 .'flg4+ 38 Wd3 Ilf3+ 39 .ixf3
.

'flxf3+ 40 Wd2 Ilg4 41 'fle3 Wlxd5+


42 Wc1 Ile4 43 'flf2 'flc4+ 44 'flc2
'flf1 + 45 "ild1 'flf4+ 46 '*b1 Wlf5
0-1

Game 44
Mendes-R.Rodrigues
Figueira Foz Honra 1999

1 e4 lbf6 2 e5 lbd5 3 d4 d6 4 c4 when White had a very strong attack

1 12
Four Pa wns A ttack: Main Line

in the game Art.Minasian-Donchenko) 16 ...i.b4+.


Naberezhnye Chelmy 1988. 1 6 . . ..i.xe5 1 7 'irb3 o!Of4 1 8 .i.e3
10..Jld8 looks much safer, aher J:the8 19 'i'a3
which 1 1 0-0 (1 1 'Wd2 llJaS!? is an un
tested idea of Hellers) ll ...i..g4 12 c5
(12 llJg5 .i.xe2 13 .. xe2 llJxd4 14
i..xd4 'tlt'xd4+ 15 h1 l%d7 16 'Wf3!?
also needs more analysis) 12 ...llJd5 13
llJxd5 xd5 14 llJg5 i.xe2 15 'ifxe2
l:td7 16 'iff2 llJd8 was the ultra-soJid
continuation of Macieja-Timmer, Par
dubice 1994.

1 9 ...Wd2+!1 20 .i.xd2 .i.xb2+ 21


We3 .i.xa1 22 f2?
The only way to keep playing was
with 22 1i'xe8, forlorn though this is.
22 .i.d4 23 g3 .i.xe3+ 24 .i.xe3
.

d3+ 25 f3 e5+ 26 *t2 g4+


0-1

Game 45
10 exd5 1 1 cxd5 b4 1 2 d4
.. Kulaots-Kengis
6xd5 1 3 o!Oxf5?? Riga Zona/ 1995
White should play 13 llJxd5, aher
which 13...lbxd5 14 llJxf5 .i.b4+ 1 e4 f6 2 e5 d5 3 d4 d6 4 c4
(14... 0-0-0!? 15 ..d3 g6 is also very in b6 5 f4 dxe5 6 fxe5 c6 7 .ie3
teresting, Bullockus-OakJey, correspon .i.f5 8 o!Oc3 e6 9 f3 .i.b4
dence 1984) 15 e2! 0-0-0 16 llJd6+
i..xd6 17 1i'xd5 1i'f5 18 'Wc4 1h- leh
unanswered questions in Tomic-Gipslis,
Dortmund 1978.
1 3 . . . 0-0-0
Theory gives the simple 13 'WxfS 14
.

llJxd5 0-0-0 as the refutation. But Rod


rigues comes up with something far
more spectacular and beautiful.
14 d6+ .i.xd6 1 5 o!Oxd5 xd5 1 6
.txa7
And not 16 'Wxd5 because of

1 13
A lekhin e 's D e fence

Although this move has a poor repu 10 ...J..e7? I should point out that 1 1 d5
tation, its adoption by an Alekhine spe will be very strong because Black's
cialist such as Kengis should at least knight is unable to come to b4.
earn it a second look. 1 1 bxc3 0-0!?
10 a3?! A very simple and reasonable move,
If White plays the solid 10 J.e2, preparing to play 12 ...lLla5.
Black should try to develop counterplay In the game Faibisovich-Bagirov,
with 10...0-0 (10: .. lLla5!? 1 1 c5 lLld5 12 Baku 1969 the Alekhine maestro gained
J.d2 lLlc6 13 0-0 0-0 14 J.g5 was the advantage after l l ...'fld7 12 .i.e2
played in Rohde-Shamkovich, New lLlaS 13 lLld2 1i'c6 14 J..f3? 1i'a4 15 c5
York 1976) 1 1 0-0 lLla5!? 12 lLld2 (12 c5 'flxdl+ 16 ltxdl lLld5 17 .i.xdS exd5 18
.i.xc3! 13 bxc3 lLlbc4! is an excellent 0-0 .i.g6! 19 llde 1 d7; Black is better
possibility that was not available for due to his superior pawn structure.
Black in the Rohde-Shamkovich game Bagirov later suggested 14 0-0 as an
because White could take the knight improvement and gave the line
and play 1i'a4+) 12 ....i.g6 13 1i'el c5 14 14 ...lLlaxc4 15 lLlxc4 lLlxc4 16 d5 lLlxe3
a3 .i.xc3 15 bxc3 llc8 16 l:.cl 'fle7 and 17 dxc6 lLlxdl 18 cxb7 llb8 19 .i.b5+
the veiled threat against the a3-pawn f8 20 llaxd1 llxb7 21 l:ld8+ rJi;e7 22
gave Black excellent counterplay in .:.xh8 .:.xb5.
Goldenberg-Ghizclavu, Graz 1972.
After 10 llcl the Czech GM Pavel
Blatny suggested the line 10 ...0-0 1 1 a3
..txc3+ 12 llxc3 ...d7 13 .i.e2 llfd8 14
0-0 .i.g4 15 <itht .i.xf3 16 .i.xf3 lLlxd4
17 .i.g5 with an 'unclear' assessment.
1 o ..i.g4!? is also interesting, aiming for
..

similar play to the 9 ....i.g4 line, but with


queenside castling for White ruled out.

1 2 c5 d5 13 i.d2 f61
With White still lagging behind in de
velopment, Black prises the position
open.
1 4 exf6 xf6 1 5 .i.b5 e5 16 .i.xc6
bxc6 1 7 dxe5
After 17 lLlxe5 'irh4+ 18 rJi;e2 (or 18
g3 1i'e4+) 18...llae8! 19 h3 llxe5+ 20
dxeS ...e4+ Black gets a winning attack.
10 ...i.xc3+ 17 ...-.e7 18 ._a4
Before anyone gets the idea to play White should have tried 18 c4!?, but

1 14
Four Pa wns A ttack: Main Lin e

then 1 8...li)f6 19 .ig5 .tg4 1ooks pretty This bishop cuts through White's p<>
good for Black in any case. sition like a knife through butter. Of
18 .. JWxc5 1 9 9d4 'irxd4 20 cxd4 particular importance is the fact that it
.i.d3! controls the b1-square and thus pre
vents White from contesting the b-file.
21 .i.b4 J:r.fb8 22 Wd2 .i.g6 23 J:r.hc1
aS 24 .*.c3 .Ub3 25 a4 c5! 26 l:.a2
26 dxcS loses a piece after 26 ...li)xc3
27 llxc3 l':d8+.
26 c4 27 & 1 IUS 28 f3 f4
..

0-1
The pawn on g2 is falling and with it,
White's position.

1 15
A lekhin e 's Defenc e

Summary
I don't think that it's the right time to start playing the trendy 9 .tg4 unless Black
..

can find a way to generate meaningful counterplay after 10 e2. So unless Black
wants to study the complications arising from 9 . .te7 10 d5 he should take a look
. .

at 9 ...d7 (and after 10 .ie2 play 10 ...l:td8) or 9 .tb4.


... ..

I recommend that full-time Alekhine players use one of the solid lines given in
this chapter together with one ofthe more experimental lines given in Chapter 8. It
is probably advisable to learn the solid line first.

1 e4 f6 2 e5 d5 3 d4 d6 4 c4 b6 5 f4 dxe5 6 fxe5 /t)c6 7 .ie3 .tfS


(D) 8 c3 e6 9 f3 .te7
9 .tg4 {D) - Game 43
..

9 'ffd7 - Game 44
..

9 .tb4 - Game 45
...

1 0 d5
10 e2 0.0 1 1 0-0 f6 12 exf6 .txf6 Game 40
-

1 0 exd5
..

10 ...lbb4 Game 42
-

1 1 cxd5 b4 1 2 d4 .td7 (D) 13 1Wf3 - Game 41

3 .tb4
. . . 3 .'fle7
. . 9
. . . 0-0

1 16
I CHAPTER EIGHT I
Four Pawns Attack :
Divergences

1 e4 I.L!f6 2 e5 lLldS 3 d4 d6 4 c4 most wonderful must be with 5 ...g5!?.


lLlb6 5 f4 Despite its astonishing appearance, the
In this chapter we take a look at theoretical reputation of this move has
Black's more extreme ways of counter never been better, and that due largely
ing White's massive pawn centre. to the efforts of the correspondence
Games 46 and 47 illustrate the razor player Schirmer. Game 50 demonstrates
sharp 6 ...c5, originally championed by his treatment of the line and his many
Ljubojevic in the 1970s and more re new ideas. In fact I've added one of my
cently taken up by Shabalov. Actually own, 1 1 ...'ii'g6 instead of his 1 1.. . .i.g7.
this is starting to look like a drawing Game 51 is a gambit line invented by
line, due largely to the fact that if Black the Australian IM Wohl. His prepara
improves on Game 47 with 10 ....i.b4, tion backfired in this game but he still
he is likely to get an equal endgame but obtained strong counterplay. The worry
nothing more. In Game 48 Black com is that 1 1 .i.e3 may leave Black with
bined 6 ...c5 with 7...g6, which enters inadequate counterplay, but in such ob
really uncharted territory. scure positions it is always possible that
In Game 49 Black used an alternative an improvement will be found.
method of playing for ... c7-c5; he first The Russian player Pushkin has been
developed his queen's bishop and then experimenting with 5 ... g6, with the
brought his knight out to a6. White theoretical reputation of this move hav
should play 8 .i.e3 with chances for a ing been considerably enhanced by
better endgame. Djurhuus found him Game 52. My own feeling is that White
self in big trouble when his centre fell should actually avoid the automatic 6
apart. lbc3 as this knight may be required to
Amongst the weird and wonderful defend the c-pawn. I suspect that Black
ways Black has tried to undermine will have far more difficulty after my
White's pawn front, the weirdest and suggested 6 f3, followed by 7 .i.e2

1 17
A lekhin e 's D e fenc e

and 8 0-0. 8 ...exd5 9 cxd5 c4

Game 46
Vetemaa-Shabalov
USSR 1986

1 e4 li)f6 2 e5 li)dS 3 d4 d6 4 c4
li)b6 5 f4 dxe5 6 fxe5 c5!?

Reaching the critical position. Black


is forcing White's pawns to advance in
the hope that they will become weak.
10 a3?1
Preventing Black's threatened
10... .ib4, but losing time. White has a
number of alternatives:
One of the sharpest lines at Black's a) 10 ..tf4?! is bad because of
disposal, it's no accident that it has been 10 ... .ib4 1 1 ..ixc4 lLlxc4 12 ...a4+ lLlc6
adopted by the likes of Ljubojevic and 13 dxc6 lC.xb2, winning on the spot.
Shabalov. b) 10 'i'd4 ltlc6 1 1 'i'e4 ltlb4 12 a3
1 d5 (12 d6 g6 threatens the horrific
White has a much quieter possibility 13 .....if5) 12 ...ltl4xd5 13 lLlxd5 'i'xd5 14
in 7 f3 but 7 ...cxd4 8 'i'xd4 'i'xd4 9 ...xd5 ltlxd5 15 i..xc4 lLlc7 16 i..e3
d4 (Velimirovic-Bagirov, Palma de .ie6 was fine for Black in Ciocaltea
Mallorca 1989) and now 9 ...e6 10 b5 Ljubojevic, Malaga 1971.
a6 looks solid enough. c) 10 d6!? lLlc6 1 1 f3 (1 1 J.f4!? gS!?
7 ... e6 S li)c3 needs testing} 1 1. ..J.g4 12 J.f4 g5! 13
The seemingly strong 8 d6 ...h4+ 9 ltle4 (13 ..ig3 ..ig7) 13...gxf4!! 14 ltlf6+
g3 1i'e4+ is not quite the simple win of a 'i'xf6 15 exf6 0-0-0 and Black had more
rook that it appears to be, but in any than enough for the queen in Paunovic
case White's compensation is probably Mrsevic, Yugoslavia 1982.
inadequate. 10 'i'e2 'i'xh1 1 1 f3 is d) 10 e6 is good for Black aher
met by 1 1 ...c6! 12 bd2 d7! 13 10 ...i..c5 1 1 exV+ xV 12 lLlf3 :e8+
f2 dxe5! 14 lLlxe5 'i'xh2+ 15 .ig2 13 .ie2 .ig4.
lLld4 16 11'd1 .ixd6 17 lLlf1 (finally e) 10 ..ie3 .ib4 1 1 i..xb6 is a radical
trapping the queen) 17 ...'ilxg2+ 18 way of securing the dS-pawn, aher
xg2 .ixeS with a decisive advantage which Alexandrov-Shabalov, Riga 1987
according to Volzhin. continued 1 1 ...11'xb6 12 ...d2 0-0 13

1 18
Four Pawns A ttack: Divergences

lbf3 .i.g4 14 h3 .i.xf3 15 gxf3 dl 16 A move reminiscent of Frank Mar


f4 lbc5 17 lDe4 18 Wd4 xc3 19 shall's famous 'gold coins' move against
'i'xb6 and now van der Talt's recom Lewitsky, ...'irg3. The immediate threat
mendation of 19 ... axb6 20 bxc3 .i.xc3 is mate on b2, and the queen is immune
would give Black good counterplay. to capture.
f) 10 lbf3 is seen in the next game. 21 l:ld2
Both 21 xbS b3 and 21 .i.xbS
b3 are immediate mates, while 2 1 b4
loses to 21. ..b3+ 22 .i.xb3 ltxc3+.
21 ...lL!xc3 0-1
22 bxc3 allows 22 .....b1 mate, while
22 'irxc3 is met by 22 ...b3+ 23 .i.xb3
'irxb3.

Game 47
Grunfeld-Ljubojevic
Riga Interzona/ 1979

10 .i.c5
. . 1 e4 lOt& 2 e5 lbd5 3 d4 d& 4 c4
Preventing White from castling king lOb& 5 f4 dxe5 6 fxe5 c5 7 d5 e& 8
side, and the other flank turns out to be lbc3 exd5 9 cxd5 c4 10 lbt3
pretty dangerous too. White's usual move, which leaves
1 1 lbt3 0-0 1 2 .te2 .tf5! 1 3 .tgS Black the choice about which knight to
Wd7 1 4 "ild2 h& 1 5 .tf4 lOa& 1 6 pm.
0-0-0 llac8 1 7 h3 l0a4! 10 .tg4

The prelude to a really fantastic com Recently Black has been playing
bination. Shabalov must have foreseen 10....i.b4! which leads to an endgame in
his 20th move at this point. which he gets excellent positional com
1 8 lbd4 .txd4 1 9 "ilxd4 l06c5 20 pensation for a pawn: 1 1 .i.xc4 (aher
.txc4 Wb5!! Timman's suggestion of 1 1 .i.gS, Black
seems to survive with 1 1 .. . .i.xc3+ 12
bxc3 xdS 13 xdS xdS 14 0-0-0
e7 15 .i.xe7 e7 16 .i.xc4 .i.g4 17
llhft c6) 1 1...i..xc3+ 12 bxc3 xc4
13 'ira4+ dl 14 'irxc4 b6 15 'irb5+
11fd7 16 'i'xd7+ (after 16 11fe2 'irxdS 17
().() 0-0 18 .i.a3 lieS Black's pieces get
well entrenched on the light squares)
16 i.xd7 17 d6 llc8 18 .i.d2 (18 .i.e3
..

is more active but then 18...lbc3 19


i.xb6 axb6 20 d2 lta3 gave Black
equality in Shirov-Shabalov, Riga 1986)

1 19
A lekhin e 's D e fence

1 8 ....i.b5! (probably even more precise tion undisturbed) 1 5 h5 .id3 1 6 .txd3


than the older 18 .../iJc4; first of all Black cxd3 was played in S.Biicker-Fleck,
stops his opponent from castling) 19 Biinde 1985 and now Foisor's sugges
liJd4 .i.d3. tion of 17 l:Lxf7! hxg5 18 llxg7 liJxdS
19 'i'f3 gives White a winning attack.
1 1 . . ..txf3 1 2 gxf3 .ib4 13 .ixc4
0-0 14 Dg1
The spectacular 14 .ih6!? should be
answered by 14.../iJ8d7!, which leads to
unexplored complications after 15 .Z:g1
g6 16 e6 /iJe5!, and not 14 ... gxh6 15 e6
f6 16 d6! liJxc4 17 e7! 1i'e8 18 .Z:gl+
h8 19 1llg4 llf7 (19 ...'iff7 20 11fxc4!)
20 1llg8+! 11fxg8 2 1 llxg8+ '.txg8 22
e8111 + etc.

The fact Shulman-Baburin, San Fran


cisco 2001 was agreed drawn at this
point confirms the view that this posi
tion (the last word on the 6...c5 line) is
fine for Black. The previous game in
this line was MChess Pro-Shabalov,
Boston 1994 which continued 20 f2
d7 2 1 llhe1 lthe8 22 a4 J.g6 23 .:ta2
lieS 24 liJf3 liJc4 25 J.f4 ..thS 26 llae2
.i.xf3 27 gxf3 l:taS 28 llg1 g6 29 llbl
b6 30 l:Lb4 /iJxe5, recovering the pawn
with a good position. Shabalov actually 14... g6
lost this game but this was only due to a 14 ...\lfcl? 15 e6 f6 16 .th6 1Wxc4 17
subsequent blunder. llxg7+ h8 18 l:Lg8+!! 1-0 was the con
1 1 Wd4 clusion of Ljubojevic-Honfi, Cacak
This leads to some spectacular play, 1970.
but it is not the only move. The simple 1 5 .ig5
1 1 J.e2 J.cS (1 t ...J.b4 12 0-0 i.xc3 13 In such a sharp position, possession
bxc3 11fxd5 14 1i'xd5 liJxdS 15 .ixc4 of the initiative is paramount. 15 .ih6 is
liJb6 16 i.b3 left Black facing menacing bad because of 15 ... liJc6 16 11fe4 l2Jxe5!
threats against f7 in Boudre-Werner, 17 .ixf8 'ifxf8, with a safe king and
Royan 1988) 12 liJg5 .tfs 13 llfl (13 demonic piece activity for the sacrificed
i.g4!? .id3 14 e6 {Durao-Silva, Portu exchange.
guese Ch., Lisbon 1995} is also very 1 5 .. .'iic7 16 .ib3 .ic5 1 7 Wf4
interesting) 13 ... .tg6 14 h4 h6 (14 ... h5 .ixg1
15 .if4 allows White to build his posi- Driven by necessity rather than

120
Four Pawns A t tack: Divergences

greed. 17 ...l:le8 18 .i.f6 ltl8d7 19 ltle4 Tecklenburg 1984) 23 ... ltl8d7 24 'ifh4
l:lxe5 (19 ... ltlxe5? is refuted by the dev- 11'xe5+ 25 1 h5 26 ltlxh5 gxh5 27
astating 20 llxg6+!!) 20 .i.xe5 ltlxe5 21 'iixh5+ g8 28 l:lg2 and White wins
e2 .i.d6 was played in Banaventure (Wiemer).
Renaud, Le Havre 1977, and now 22 21 l08d7! 22 e7 1i'xh2!
..

llacl would leave Black struggling. Black can afford to return some ma
17...ltl8d7 is well met by 18 d6 'i'c6 19 terial as long as he breaks White's at
0-0-0. tack.
18 d6 -.c5! 23 exfB + D.xf8 24 1i'xh2
The only defence. 18 ...11'c8 19 0-Q.O A forlorn hope, but White has noth
.i.c5 20 e6! fxe6 2 1 1i'e5 lieS (or ing better.
21...lbd5 22 d7!!) 22 i.h6 d7 23 ltle4 24.....ixh2 25 lOt&+ Wg7 26 lbxd7
ltlc6 24 ltlf6+ wins for White, as lbxd7 27 ..te7 D.bB!
pointed out by Bronstein.
19 lbe4 d4 20 D.d 1 xb2

28 Wf2
Black's precise 27th move ensured
21 e6?! that 28 .i.a4 is met by 28 ...b5.
In such complex positions it can of 28 .....ie5 29 D.c1 lbc5! 30 D.d1 .if&
ten take years to uncover the right path, 31 ..ixf6+ Wxf6 32 ..tc4 D.dB 33
and it seems that White missed his way g3 a& 34 ..tf1 e5 35 f4+ e&
at this point. 36 ..ic4+ Wf6 37 Wt3 b5 0-1
The correct line is 21 ltlf6+ hS 22
.:td2 1i'a1+ (22.. .'W'b1+ 23 e2 ltl8d7 24 Game 48
1i'h4 h5 25 ltlxh5 was also winning for Volzhinwl.Svechnikov
White in Moura-Rinaldi, correspon USSR 1988
dence 1983, while 22.....cl+ 23 e2
'ifc5 24 fl 1i'c1 + 25 J.d1 also leaves 1 e4 lbf& 2 e5 lbd5 3 d4 d6 4 c4
Black facing deadly threats to his king) lOb& 5 f4 dxe5 6 fxe5 c5 7 d5 g6!?
23 .i.dH (23 lld1 'iib2 24 l:ld2 repeats An intriguing move which takes aim
the position, while 23 e2 ltlc6 24 1i'h4 at the eSwpawn. As yet it has very little
h5 was unclear in Griinfeld-Wiemer, established theory.

121
A lekhin e 's Defenc e

enough but this loses immediately) 16


i.f6! and White won in Solomon
Kanikevich, Sydney 1991.
12...llJ8d7 also fails to gain the much
needed counterplay after 13 llJf3 l%e8
14 l%e 1. If White maintains his centre
like this, Black will be struggling.

8 .*.f4
White can also play 8 llk3 .ig7 9
.i.f4 (9 .ie3 0-0 10 .i.xc5 llJ8d7 will
recover the pawn with a good game,
while 9 c!ilf3 is met by 9 ....i.g4).
8 .tg7 9 .!Oc3 0-0 1 0 Wrd2
.

Preparing to castle long. White has


tried a couple of other plans: 1 3 l:te1 !
a) 10 .i.d3 e6 1 1 d6 is dubious be White wants to kick Black's bishop
cause it allows Black to start hammering out of g4 before developing the knight
away at e5. Ciuffoletti-Sedina, Saint on f3. After 13 llJf3 llJ8d7 Black threat
Vincent 1998 continued 1 1 ...llJc6 12 ens 14 ...l:le8 and it gets difficult for
llJf3 llJdl 13 1ie2 l0d4! 14 llJxd4 cxd4 White to hold e5.
15 llJb5 1ia5+ 1 6 J.d2 ilb6 17 b4 1 3 ...c4! 14 h3 .tf5 1 5 g4
l0xe5 1 8 c5 llJxd3+ 19 1i'xd3 1i'c6 20 15 llJf3 deserves consideration. Now
0-0 e5 with the better game for Black. Black comes up with a promising pawn
b) 10 .i.e2 e6 1 1 l0f3 exd5 12 cxd5 sacrifice.
.i.g41 13 0-0 llJ8d7 14 9e1 .i.xf3 15 1 5 ....td31 16 .ixd3 cxd3 1 7 'ibd3
.i.xf3 llJxe51 16 .i.xe5 l%e8 won a pawn lba&!
for Black in Moraru-Grunberg, Bucha The immediate threat is 18 ...llJb4.
rest 1999. 18 d& l:tc8 1 9 *b1 lbc41
c) 10 h3 is slow but nevertheless And now the idea is 20.....b6. In a
worth considering. White wants to de later game (Ilincic-Marinkovic, Vrnjacka
velop his knight on f3 without it being Banja 1989) Black played the inferior
pinned. 19 ... l0c5 but even here he had compen
10 ... e6 1 1 0-0-0 exd5 1 2 cxd5 sation after 20 1i'e2 llJe6 21 .ig3 l0c4
.tg41 22 ltlf3 1i'a5 .
12 ... c4 is not as good because of 13 20 .!Od5 'ifa51 21 lbe7+ *h8 22
llJ3 .i.g4 14 .i.h61 .ixf3 15 .ixg7 .!Df3 l:tc5
.i.xdl? {15 ...clt>xg7 16 gxf3 is miserable According to Volzhin, Black could

122
Four Pa wns A ttack: Divergen ces

also consider 22 ... lbxb2!? as after 23 game.


Wxb2 lbb4 24 t!fb3 l:tc2+ 25 b1 l:txa2 28 l:txh7+! xh7 29 l:th1 + .J.h6 30
the attack is very strong. g5+ g7 31 f5+ g6 32 h71
23 h4?
Missing Black's reply. White had to
play 23 1i'b3 lbxb2 24 i..d2 1i'a4 25
l:te3 llb5 26 1i'xa4 lbxa4+ 27 lZ.b3
lbb3+ 28 axb3 lb4c5, producing an
endgame in which both sides have
chances (Volzhin).

32 ...d2 +
Finally it is dear why White's king
went to b3 on move 25. Had he played
25 b1 he would be getting mated!
33 a4 xh7
Black decides not to push his luck
and accedes to a draw by perpetual
23 ...Wxa2+1 24 xa2 b4+ 25 check. The attempt to play for a win
b3! xd3 26 h51 with 33 ...l:tc4+ 34 b4 tbh5? is met by 35
The only chance. With both players :XhS :.Xg4 36 l:lxh6+ xfS 37 liJxf8
in time trouble it is Black's turn to make xeS 38 lbd7 +I f5 39 l:H6+ e4 40
a mistake. liX5+ e5 41 d7 (Volzhin).
26 gxh5?
.. 34 l:txh6+ g8 35 e7+ g7 36
The right way to play it was with f5+ 98 37 e7+ % -%
26 ...lbxf4! 27 hxg6 fxg6! (and not
27 ...lbxg6? 28 lhh7+! xh7 29 l:th 1+ Game 49
i..h6 30 lbg5+ g7 31 lbf5+ g8 32 Djurhuus-Agdestein
lbxh6+ g7 33 lbf5+ g8 34 l:th7 Norwegian Ch., Asker 2000
lbd2+ 35 b4 l:tc4+ 36 a3 lbb1+ 37
a2 J:.d8 38 llg7+ f8 39 l:txf7+ g8 1 e4 f6 2 e5 d5 3 d4 d6 4 c4
40 ltlh6+ h8 41 l:th7 mate). b6 5 f4 .tf5 6 c3 e6 7 f3
27 l:lxh5 xf4 After 7 i..e3 Black can play 7 dxe5
..

After 27... lbxe1 Volzhin gave 28 (7...liJa6?! would leave Black's knight on
lbgS h6 29 d7! lba5+ 30 ct>a2 li)c6 31 a6 badly placed after 8 exd6 cxd6 9
lbxc6 bxc6 32 d8'if! l:txd8 33 li)xf7 + liJf3) 8 fxe5 i..b4!? (8...lllc6 transposes
h7 34 lbxd8 tLid3 35 e6! llxh5 36 into the main lines) 9 liJf3 c5, which is
gxh5 i.. f6 37 i..c7 with a drawish end- known to give Black good counterplay:

1 23
A /ekhine 's Defen c e

a) 10 b3 cxd4!? 1 1 ltlxd4 tlh4+ 12 8 .i.d3


g3 tle4 13 f2 .txc3 14 .tg2 .xd4! 1 5 This now turns out to be very dan
.txd4 .txd4+ was good for Black in gerous for White, who shortly gets hit
Marjanovic-Suba, Novi Sad 1974. by a flurry of tactical blows. 8 exd6
b) 10 a3 cxd4!? 1 1 .xd4 (1 1 .txd4 is would also be bad because of 8 ...ltlb4!.
met by 1 1. .. .txc3+ 12 .txc3 1i'xd1+ 13 White should play 8 .te3, after which
:lxdl ltla4!, and 11 axb4 by 1 t...dxe3! 8 ... c5 9 dxcS dxcS 10 a3! (stopping
12 1i'xd8+ xd8 13 0-0-0 + e7 14 cS Black's a6 knight from getting back into
aS! 1 5 cxb6 axb4, threatening 1 6.. .:a1+ the game) 10...11Vxd1+ 1 1 :Xd1 .te7 12
{Miulescu-Ghuzdavu, Romania 1972}} .ie2 gave White the better endgame in
1 t...Wxd4 12 .i.xd4 .txc3+ 13 Lc3 Hort-Knezevic, Luhacovice 1973. He
ltla4 14 .tb4 ltlxb2 15 ltld4 (Ostojic has more space and the knight on a6 is
Marovic, Yugoslav Ch. 1972) and now badly placed.
1 5 ...a5! 16 .td6 .tg6 17 ltlbS ltla6 is 8 . . . .i.xd3 9 xd3 c5 10 dxc5
fine for Black according to Burgess. I doubt that White really relished this
7 ..86
. line, but it seems to be the only way to
After 7...dxeS 8 fxeS .tb4 White can play it. 10 d5 is bad because of 10 ...exd5
play 9 .td3! ..i.xd3 (9... c5 10 0-0! cxd4 1 1 lLlxdS dxe5 12 ltlxeS ltlxdS 13 cxdS
1 1 ltle4 gives White a dangerous attack, ltlc7, winning the dS-pawn.
and 9 ....tg4 is met by 10 0-0! ltlc6 1 1 c5 1 0 ...li:)xc5 1 1 2 dxe5 1 2 1Cixe5
ltldS? 12 M) 10 xd3 c5 1 1 0-0 cxd4 f6!?
12 ltle4 ltl6d7 (12...0-0 is met by 13 Forcing the knight from eS by book
ltleg5 g6 1 4 ltlxh7! h7 15 lLlg5+ g7 or by crook. White is forced to go for
16 9h3 l%h8 17 :lxf7+ g8 1 8 1i'xe6) broke and accept the offer of the ex
13 ltlfg5 ltlxeS 14 1i'g3 ltlbd7 15 .tf4 change.
1i'b6 (15 ...ltlg6 16 ltlxf7) 16 .txeS f6 17 1 3 h5+ g6 14 1Cixg6 hxg6 1 5
.td6 fxgS 18 .txb4 1i'xb4 19 ltld6+ xh8 .d41
e7 20 l%f7+ d8 21 1i'xg5+ with a
winning attack in lvkov-Timman, Am
sterdam 1974.

Putting the biggest lump available


right in the middle of the board. Black
has a multitude of threats and White's

1 24
Four Pa wns A ttack: Diverg enc es

queen is shut out of the game. ltlxd1 We2+ 0-1


16 2
It seems that 16 f1 would result in Game 50
a draw by perpetual check aher Mischke-Schirmer
16 ...tbe4 17 ltlxe4 11fxe4 18 1i'h3 (18 Correspondence 1995
'ii'xf6 11'd3+ 19 f2 .tc5+ 20 el
.tb4+ 21 2 .tc5+ is also a draw) 1 e4 f6 2 e5 ltld5 3 c4 lLlb6 4 d4
18 ...1t'xc4+ 19 g1 1t'e4 20 1t'g3 (or 20 d6 5 f4 g5!?
f2 .tc5+ 21 g3 'ifd3+ 22 h4 g5+
23 g4 1t'f5+ 24 g3 d3+) 20....tc5+
21 ft 1t'c4+ 22 e1 1t'e4+ 23 f1
1t'c4+ etc.
16 'ifh3 is bad because of 16 ...ti)d3+
17 e2 'iff2+ 18 xd3 l:[d8+ 19 ltld5
ltlxdS 20 xe6+ ti)e7+ 21 c3 1t'd4+
22 b4 l:[d6 23 'ilfel .Ub6+ 24 a3
'ifxc4 25 b3 11fd4, according to analysis
by Carsten Hansen.
16 .....xc4+ 1 7 Wf3 d5 1 8 l:ld1
3 1 9 WigS?
Under massive pressure, White loses You need to be very enthusiastic
his way. After 19 ltlxd5 'ifxdS+ 20 g3 about undermining pawn centres to play
f5 (20....d4 21 g8} 2 1 .te3 (or 21 this idea of GM Albin Planinc. More
h3 1ie4 22 1i'g4 .tb4) 21...e5 the po than thirty years after Planinc used to
sition would still be massively compli play it the shock value is still intact. And
cated. armed with his own ideas, Michael
Schirmer thinks it's good enough for
correspondence chess.
6 exd6
The theoretical 'refutation'. Another
Schirmer game (Gerloff-Schirmer, cor
respondence 1989) went 6 fxg5 dxe5 7
d5 (7 dxe5 'ii'x d1+ 8 d1 .ig7 9 ti)fJ
.tg4 is good for Black) 7...e6 8 ti)fJ
exdS 9 .te2 e4 10 ti)e5 'i'e7 1 1 ti)g4
.txg4 12 .txg4 1i'b4+ 13 ti)c3 'irxc4 14
.te2 'i'c5 15 'i'd2 llx:6 16 .tg4 ti)e5 17
.te2 0-0-0 18 h4 .tg7 19 llft llhe8 20
1 9...e7! 20 Wh7 We&+ 21 e2 ltlb5?? ltld3+ 0-1 - a total massacre, but
Wxg2+ White's play was unimpressive.
An old-fashioned king hunt. In the classic Eales and Williams
22 Wxd3 0-0-0 + 23 Wc4 l:lxd1 24 book on the Alekhine, 5-.g5 is relegated

1 25
A lekhin e 's Defen c e

to a side note. But neither of the sug 13 lba3 0-0-0 14 xc6 l2Jxc6 1 5 lbc2
gestions offered seem very good; 6 lbf3 .i..fS 16 lbe3 Ad3 17 d2 e4 18 lbc2
can be met by 6... g4 and 6 'it'h5 by ..ixf3 19 gxf3 lbxd4 20 l2Jxd4 Ld4+ 2 1
6... dxe5. Meanwhile, 6 d5?! e6 7 f5 exfS e2 llc4 2 2 e3 ltc2+ 2 3 'iti>d3 llxb2
8 e6 'it'f6 was better for Black in Erma 0-1 .
kov-Keene, correspondence 1971-72. 8 ...llJd5 9 i..c4
If White is in search of a good line The game Blake-Schirmer, corre
against 5 ...g5!?, then maybe he should spondence 1994 continued 9 fxg5 .i.g7
also look at simple development with 6 10 .i..c4 l2Jc6 1 1 lbf3 Wf5 12 ..ie3?
lL!c3. lbxe3 13 xe3 lbxd4 14 lbxd4 Wxg5+
6 . . .1i'xd61 15 ..td3 ..if5+ 16 ci>c3 1!fe3+, when
This was thought to be bad, with no Black recovered the piece with a win
less an authority than Vlastimil Hon ning attack.
recommending White's reply. Black's 9 . . .gxf4 1 0 'iff3
earlier try was 6 ...gxf4 but then 7 dxc7! White could also consider simple de
'ii'xc7 8 lDc3 e5 9 dxe5 lbc6 10 xf4 velopment with 10 lL!f3, after which
j.e6 1 1 lbe4 b4+ 12 'itf2 lbxc4 13 10 ... 1i'f5 could be Black's best.
i.xc4 Axc4 14 lL!f3 was horrible in 10 ...c6 1 1 i..xf4 i..g 7
Tringov-Pianinc, Varna 1970. I suggest 1 1 ... 11t'g6 (!), keeping e7 de
7 c5 'ife&+ fended, the g-file open and preparing
... .i.e6. The position looks very interest
ing and complex.
1 2 llJe2 llJd7 13 llJbc3 llJ7f6 14 h3
1i'd7 15 l:lad1 llJxf4 16 llJxf4 'iff5
16 ...0-0?! 17 l:r.he1 is poor for Black,
so he leaves his king on e8 for the mo
ment in order to defend the e-pawn.
17 g4 1i'g5 18 l:lhg 1 !?

8 2!?
At this point Hon gave 8 ...e2 but
then in Zoels-Schirmer, 1993, Black
generated powerful counterplay with
8...lbd5 9 Wxe6 (delaying the exchange
of queens with 9 fxg5 Ag7 10 lbf3 ltk6
doesn't help White) 9 ...j.xe6 10 fxg5
Ag7 1 1 lbf3 l2Jc6 12 Ab5 lbdb4! (the
key move, avoiding doubled pawns and White can also try to eliminate
opening the d-file against the d4-pawn) Black's dark-square bishop with 18

126
Four Pa wns A t tack: Diverg en c es

lllli5, after which 18...lihh5 19 1i'xf7+ 7 cxd6 exd6 8 exd6 ll:lf6! 9 We2+
d8 20 .xh5 IU8+ 2 1 'it>g2 1i'xh5 22 .i.e6 10 ll:lc3
gxh5 gives White slightly the better of 10 fS? doesn't work because of
the endgame, but 1 8 ...1i'h4+ 19 g2 10...1fa5+.
l:lg8!? is still messy. 10 g6!

18 ...0-0 1 9 l:lge1 l:ld8 20 g3 h5 Wohl had originally prepared


21 l:le5?! 10...1fa5, but changed his mind during
In his notes, Shirmer gave 21 lDe4 the game.
lLlxe4+ 22 'ifxe4 as being better. 1 1 c5
21 hxg4 22 1We3?
.. In a subsequent game (Smeets-Wohl,
After this White has nothing. He had Zwolle 2001) White played 1 1 .i.e3
to try 22 llxg5 gxf3 23 lllg6! e6 24 tbe5 .i.g7 12 0-Q.O and left his opponent
with a complex struggle ahead. with very little compensation after
22 .tf5 0-1
. 12 ... 0.0 13 tbf3 lbc6 14 1i'd2 .aS 1 5
....---- lbd4 li)xd4 16 .i.xd4 b5 17 .i.x6! .i.x6
Game 51 18 tbd5 etc. It remains to be seen if
Mamadshoev-Wohl Wohl has an improvement ready.
Yerevan Olympiad 1996 1 1 ... .tg7 1 2 Wb5+ ll:lbd7 1 3 ll:lf3
In giving the check on b5, White
1 e4 f6 2 e5 d5 3 d4 d6 4 c4 might have intended 13 1i'xb7 lllxc5 14
b6 5 f4 c5!? 6 dxc5 'i'c7, exchanging queens and staying a
6 d5 is met by 6...e6, with very strong pawn up. But by now he might have
counterplay. realised that after 14...1i'xc7 15 dxc7 Q..O
6 l06d7
Black's active pieces provide good
6 ... dxc5 7 'i'xd8+ gives Black a truly compensation for the material deficit.
horrific endgame, which is not at all After c7 drops it will only be one pawn.
what Black had in mind. Instead he 1 3 0-0 14 .i.e2 l:lc8 1 5 0-0 ll:lxc5
..

gambits a couple of pawns for what 1 6 l:ld1 .td7! 1 7 'tWb4 a5 1 8 Wd41


turns out to be quite interesting com
pensation.

Over the next few moves, some


fancy footwork keeps the queen on the

127
A lekhine 's Defen c e

g1-a7 diagonal. After 18 1lra3 Black can r------.


generate ferocious play with 18 ... b5 19 Game 52
.txbS 11rb6 etc. Brener-Pushkin
1 8 l[)g4 1 9 lOes l[)e& 20 'tWa7!
. USSR 1988
l[)xe5 21 fxe5 .i.xe5 22 'tWf2! .i.e&
Preventing 23 lDds. 1 e4 l[}f6 2 e5 l[)dS 3 d4 d6 4 e4
23 .i.h& .i.g7? lbb& 5 f4 g6!?
After his brave and imaginative ear
lier play, Wohl suddenly loses his nerve.
Black should sacrifice the exchange with
23 ....txd6! 24 .txf8 11rxf8, when the
mighty pair of bishops provide excellent
compensation.
24 .i.xg7 xg7 25 .i.b5! .i.xb5 26
lbxb5 'tWg5 27 lbe3 'ireS 28 l:ld5
'tWxf2+ 29 xf2 l:lfd8 30 l:lad1

Another interesting line in which the


theoretical 'refutation' has been ren
dered totally unconvincing.
6 lbe3
In view of the dangers to his c-pawn
in this line, White might do well to con
sider the unpretentious 6 3 .tg7 7
.te2 0-0 8 0-0, after which 8 ... .te6 can
be met by 9 bdl.
30 b6?
.. 6 .i.g7 7 .i.e3 0-0
..

The final mistake? According to GM The immediate 7 ... .te6!? is also


Ian Rogers, Black could still draw with worth considering, a game Kotek
30...ltlc5. Sergiev, Czech Republic 1997 continu
31 d7 l:le5 32 lba4! l:le2+ 33 l:l1d2 ing 8 11rb3 (8 cS is met by 8...d5;
l:lxd2+ 34 l:lxd2 l[)eS?! White's best may be 8 l0f3, after which
The last hope was 34... b5, though 8 0-0 transposes back into the game)
...

this does look rather ugly after 35 b6! 8 ... a5! 9 a4 6 10 l:td1 b4 1 1 f3
J:[b8 36 8 ltld8. dS! 12 cxdS (12 cS?! 4!) 12 ...xd5!
35 l[)xeS bxe5 36 e3 16 37 l:ld5 13 dS .txdS 14 1ic2 .txf3 15 gxf3
a4 38 d3 e6 39 e4 l:lxd7 40 dS 16 .i.ct e6 with a clear advantage
l:lxd7 xd7 41 Wxe5 f5 42 b4 f4 for Black because of his superior pawn
43 xa4 g5 44 *b3 g4 45 e3 h5 structure and grip on dS.
46 d3 h4 47 .2 1 -0 8 lbf3

1 28
Four Pa wns A ttack: Diverg en c es

White has also tried the supposed exd6 exd6, intending 1 1... IZ.e8, whilst 9
'refutation' 8 cS!?, but then 8 ...6d7 is b3 allows 9 ... c5! 10 dxc5 6d7 1 1 cxd6
not at all clear {and not 8...dxc5 9 dxcS exd6 12 1i'xd6 xeS! etc.
1i'xd1 + 10 llxd1 6d7 1 1 dS etc.). 9 ...a5! 10 g5
After the further moves 9 f3 b6! 10 The threat of 10 ... a4 is quite difficult
b4 .i.b7 1 1 i.e2 aS 12 a3 bxcS 13 bxcS to meet; after 10 a4?! 6 Black gets
dxcS 14 dxc5 ltla6, Black had excellent the b4-square as in the Kotek-Sergiev
counterplay in the game Rogers-Loffler, game above.
Wijk aan Zee 1996. 9 h4!? has been sug White's can play 10 4!?, after
gested by Volzhin and would need an which 10 . .d5 1 1 ed2 {1 1 egS?! dxc4
.

energetic response by Black; it is cer 12 xe6 cxb3 13 xd8 llxd8 14 axb3


tainly very dangerous after, for example, 6 gave Black the better endgame in
9... b6 10 1i'f3 c6 1 1 hs. Nikitin-Pushkin, USSR 1988) 1 1...dxc4
8 ...e6! 12 xc4 6 looks fairly even. White
Once again this is the key to Black's has more space which is compensated
counterplay; the attack on the c-pawn is by Black's powerful grip on the dS
surprisingly difficult to meet. 8...dxe5 9 square.
dxeS! condemns him to a cramped and 10 . . a4!
.

pass1ve game. And not 10 ....tf5 because of the


powerful thrust 1 1 cS.

9 3
After 9 cl2 dxeS 10 fxeS {10 dxeS 1 1 xe&
may be White's best, but Black has an 1 1 1i'b5?! can be met by 1 1. .. .i.d7 12
easy game after 10 ...6, intending ..b4 6 13 1i'a3 f6!, detonating
...f7-f6) 10 ... c5! 1 1 d5 {or 1 1 dxcS White's centre.
6d7} 1 t ....tfs 12 .tf4 e6! 13 d6 6 1 1 ... axb3 1 2 xd8 l:lxd8 1 3 a3
14 f3 d7, Black won the eS-pawn in c6 14 0-0-0?!
Schmidt Schaeffer-Haakert, Germany After this White gets renewed prob
1988. lems with his c4-pawn. Volzhin rec
Of White's other moves, 9 d5? is ommended an improvement for White
downright bad because of 9....tg4 10 in 14 :c1, his analysis continuing

1 29
A lekhin e 's D efen c e

14...dxe5 1 5 dxe5 ltld4 16 .td3 ltlc2+


17 .txc2 (17 e2 lbd3! 18 d3 :ld8+
19 e2 e3 20 xe3 ltlxc4+ 21 e4
ltlxb2) 17 ...ltlxc4! 1 8 e2!? bxc2 19
:lxc2 with equality.
1 4.../lla5! 1 5 c5 dxc5 16 dxc5
lbd 1 + 1 7 lC!xd1
17 xd1 ltlbc4 18 .tel ltd8+ 19
e 1 f6! puts White in even deeper
trouble.
1 7 ...lC!bc4 1 8 .i.d4 l:td8 19 .i.c3
l:td5 20 .i.b4 g51
An Alekhines Defence dream - the 25 l:te4
undermining operations finally cause If 25 ltlc3 there follows 25 .. .1td2 26
Whites centre to collapse. l:te2 l:lxe2 27 ltlxe2 .txb2 + etc.
2 1 fxg5 25 ...lC!e5 26 ltJf2 .i.f81 27 c6 lC!xc6
21 g3 gx4 22 gx4 .th6 will win the 28 .i.xf8 Wxf8 29 l:tc4 l:ld4 30 l:lc3
f-pawn. l:lxh4 31 l:lxb3 lC!a5 32 l:ld3 l:th2 33
21 J%xe5 22 .i.xc4 lC!xc4 23 l:te 1
. l:tg3 Wg7 34 ltJe4 Wg6 35 b4 ltJc6
e6 24 h4 36 lC!c5 l0d4 37 l:lg4 e5! 38 l:le4
Or 24 lbe5 .txe5 25 g3 .td4, which l:lxg2 39 AxeS l:lxg5 40 l:lxg5+
ties Whites pieces up prior to bringing Wxg5 41 lC!xb7 Wg4 0-1
the king in. The advance of Black's h-pawn will
24...l:td5! decide matters.

1 30
Four Pa wns A ttack: Divergences

Summary
The lines in this chapter are still largely uncharted, with only 6 c5 7 d5 e6 being

truly 'respectable'. I have my suspicions about the soundness ofWohl's treatment in


Game 5 1 and the counterplay available from 5 ... g6. Both of these (plus 6 ... c5 7 d5
g6) require some homework to turn them into useful lines. On the other hand the
amazing S ...gS looks quite playable on the basis of existing evidence.

1 e4 f6 2 e5 d5 3 d4 d6 4 c4 b6 5 f4 dxeS
S ... .tfS - Game 49
5 ... g5 Game 50
-

S ... cS {D) - Game 51


5 ...g6 - Game 52
6 fxe5 c5 7 d5 (D) e6
7. g6 Game 48
.. -

8 3 exd5 9 cxd5 c4 1 0 llJf3


10 a3 - Game 46
10 ....1g4 1 1 'ird4 .1xf3 1 2 gxf3 .1b4 1 3 .1xc4 0-0 (D) - Game 47

5 . . . c5 7 d5 13 . . . 0-0

131
CHAPTER NINE I
The Chase Variation

1 e4 lLlf6 2 e5 lLld5 3 c4 lLlb6 4 c5


lbd5 Game 53
The Chase Variation, thus named for Thorhallsson-Gausel
obvious reasons, is often chosen by Reykjavik 1997
players who like to force the pace early
on. Well-known exponents of the white 1 e4 lbf6 2 e5 lLld5 3 c4 lLlb6 4 c5
side include the strong grandmasters lLld5 5 lLlc3 lLlxc3
Sveshnikov and Thorhallsson, so it cer Another interesting possibility is
tainly deserves to be taken seriously as a 5 ...c6 6 .i.c4 {6 tbxdS cxd5 7 d4 d6
white weapon. gives Black an easy game) 6...e6 which
Having driven Black's knight away can also be reached via transposition
from d5, White promptly sends it back from 5 .ic4 c6 6 tbc3 e6:
again and argues that the pawn on c5 a) Black seems to be doing okay after
will have a cramping effect, if only tem 7 d4 b6 8 cxb6 axb6 9 tbge2 .ia6 10
porarily. White sometimes gambits this .tb3 d6 11 tbxdS; this was agreed
pawn whilst continuing with his devel drawn at this point in Rozentalis
opment. If he chooses to protect it with Kengis, Daugavpils 1989.
the move d2-d4, then Black should b) 7 tbe4 b6 8 tbd6+ .i.xd6 9 cxd6
probably construct a plan to undermine 0-0 10 d4 .ia6 was also fine for Black in
it with ... the advance b7-b6 at some Szabolcsi-Knezevic, Budapest 1981.
point. c) Finally, 7 'ifg4 f5 8 'ilg3 b6 9 cxb6
The positions arising from the Chase axb6 10 t2Jge2 i..a6 11 d3 ...e7 12 0-0
Variation are very sharp and Black cer 'ikf7 also saw Black complete his
tainly needs to know what he's doing. development with a good game m
My recommendation is to meet either 5 Sveshnikov-Giuzman, Bern 1992.
tbc3 or 5 .ic4 with 5 e6, as in Garnes
.. 6 dxc3 lLlc6
54 to 56. 6 ... d5 7 cxd6 exd6 8 ..if4 d5 9 i..d3 is

1 32
Th e Chase Varia tion

slightly better for White because of his with 1 1 1i'd2 achieved nothing after
lead in develoJ?ment. The text is Bagi 1 1.....i.e7 12 0-0 0-0 13 l:lfd1 d6 14 exd6
rov's idea. cxd6 in Kalikova-Hallerova, Czech
Women's Ch., Nymburk 1994.
1 1 ...i.e7 1 2 'i'd3!? 0-0
According to Finkel, the tempting
12. ..o!Da7?! leaves Black with slightly the
worse endgame after 13 .ta4 ..i.xf3 14
gxf3 llJc6 15 ..i.xc6 dxc6 16 "ffe4 "ffdS
17 l:lfdl "ffxe4 1 8 fxe4, his problems
stemming from the fact that he can't
castle without allowing White's rook
into d7.
13 l:lfd1 d5 14 exd6 cxd6
And not 14 ... ..i.xd6?! 1 5 llJg5, which
7 ll:lf3 e6 8 i.e3 b6 9 i.b51? forces a serious weakening of Black's
This is probably the only move to kingside. In the game too, White man
give White solJle pressure. 9 'iVa4?! gave ages to bring some real pressure to bear.
Black the initiative after 9 .....i.xcS 1 0
..i.xcS bxcS 1 1 0-0-0 f6!? 12 ..i.bS o!DxeS
13 l:the 1 0-0 14 o!DxeS fxeS 1 5 .txd7
'iVgS+ in Cappello-Bagirov, Tunis 1979,
while after 9 cxb6 axb6 1 0 a3 fS!? 1 1
exf6 tfxf6 12 gS 1i'f7 13 .td3 .ta6 1 4
c4 .td6 15 1i'c2 1i'h5 Black had a com
pact and well organised position in Fo
garasi-Bagirov, Budapest 1989.
9 ....i.b7 10 cxb6 axb6

1 5 We4! Wc7 16 'i'g4! :ladS 1 7


i.d3 ll:le5
Black wisely tries to simplify the posi
tion before White's threats become too
serious, though in this case he gets a
different set of problems due to his in
ferior pawn structure.
After 1 7...d5? White has 18 ..i.h6 ..i.f6
19 "ffh3!, when the mating threats (with
20 ..i.xg7) would force Black to give up
1 1 0-0 the exchange with 19...g6.
Trying to pot pressure on the d-file 18 ll:lxe5 dxe5 1 9 'i'c41

1 33
A lekhin e 's D e fenc e

Yet another queen move, but one We3?1


which confirms White's advantage. After this Black is right back into it
White's 3-1 queenside majority would and the shock of the turnaround tells on
be a big factor in the endgame, not least White's nerves. He should play 40 1ic7!,
because the mobility of Black's own keeping Black tied up.
majority has been hobbled by the dou 40 .. .Wa4!
bled e-pawns. Suddenly threatening ...a8 followed
19 ...Wb8!? by ...c6. White gives up a pawn to
Giving up a pawn rather than ex prevent this and suddenly he is equal at
change queens. The decision to play like best.
this might have been motivated by prac 41 c3 aS 42 l:lf1 'ffc6 43 .i.f3
tical considerations; White might have 'il'xc4 44 Wg2 f4 45 'ffe 1 fxg3 46
been running short of time and it would hxg3 'il'a2+ 47 Wd2 xf3+ 48 l:lxf3
be easier for him to play the endgame. 'it'a4

20 xb6 :ca 21 Wb3 Was 22 f1 49 'it'd3+


c6 23 a4 h5 24 aS h4 25 c4 f5! Probably missing the fact that Black
The only chance to create a counter can profitably use the h7-square. He
play. Black should be losing with best should play the move 49 'i'd4 immedi
play but the position still poses some ately, when the position is probably
practical difficulties. about equal.
26 Wc3 l:lf6 27 Wxe5! l:lg6 28 g3 49 ...Wg8 50 'it'd4 'ffc2+ 51 l:lf2
l:lf8 29 e3 Wea 30 f4?1 'ifh71 52 'it'f4?! l:lh2+ 53 Wg1 l:lh1 +
Thus far White's play has been ex 54 wg2 'it'h3+ 55 Wf3 'ifh5+ 56
emplary but this allows some counter Wg2 .i.c5!
play. White should push his passed Winning for Black.
pawn with the cold-blooded 30 a6!. 57 .i.d4 'ifh3+ 58 Wf3 'it'h5+ 59
30 hxg3 31 fxg3 f6 32 1rc5
Wg2 l:ld1 1 60 l:lf1 l:lxf1 61 Wxf1
xb2 33 d6 xa1 34 xf8 .*.f6 'ird1 + 62 wg2 'it'xd4 63 Wb8+ Wh7
35 d6 l:lh6 36 .i.e2 g5 37 rlb 1 64 1rb1 + g6 65 'it'b7+ Wh6 66 a6
Wh7 38 .i.e5 e4 39 l:ld1 e7 40 0-1

1 34
Th e Chase Varia tion

r----- i.d3 {9 'tlt'g4?! c5 10 dxc5 'flc7! 1 1 lDf3


Game 54 lDa6 12 i.xa6 i.xa6 13 i.e3 i.xc5 was
Sveshnikov-Solozhenkin fine for Black in Lein-Zelcic, Belgrade
Russia 1998 1988) 9...i.b7 10 f3 d6 {10....i.e7 1 1
,______________.
0-0 - was the end of the game
1 e4 /Of6 2 e5 d5 3 c4 b6 4 c5 Sveshnikov-Solozhenkin, Russian Ch.,
d5 5 /Oc3 e6 6 d4 Elista 1996) 11 0-0 .i.e7 12 lle1 d7 13
The continuation of Oim-Mikenas, i.c2 lla5 14 d2 dxe5, which led to
Palanga 1961 cenainly deserves a men complex play in Sveshnikov-Moroze
tion. Aher 6 ...g4 Black played 6...h5!? 7 vich, Alushta 1994.
'ilc4 d6! 8 cxd6 cxd6 9 xdS exdS 10 8 . . .bxc5 9 .i.g5 J..e 7 10 .i.xe7 Wxe7
'ifxdS llk6 11 .i.b5 .i.d7 12 exd6 lth6 1 1 xg7 :ts 1 2 /Of3 cxd4
13 i.c4 1le6+ with excellent compensa
tion for the sacrificed pawn.
6 l0xc3 7 bxc3 b61

1 3 J..e2
The impatient 13 g5? is answered
by 13 ...'tlt'a3! 14 e4 .i.b7 15 ll)6+
Once again this is a key move for t:lie7, when Black's threats prove to be
Black. He undermines White's pawn far more serious.
structure and prepares to develop his 1 3 f51
. .

queen's bishop via a6 or b7. Improving on 13 .. .f6 14 1Wg3 d3 15


s g4 i.xd3 f5 16 llb1 a6 17 d4 c5 18
8 'ilf3!? is also interesting: 'tlt'f3! c6 19 0-0, which left Black with
a) 8...c6?! 9 cxb6 axb6 10 lDh3 i.a6 some problems to solve in Sveshnikov
1 1 .i.xa6 xa6 12 0-0 b5 13 a4 was Basagic, Ljubljana 1997.
better for White in Mukhametov 14 Wh6
Oriamin, Moscow 1996. Keeping queens on the board is the
b) Bowden-Murshed, London 1988 only way to trouble Black; 14 1txe7+
continued 8...c6 9 cxb6 cxb6 10 i.d3 'ii?xe7 1 5 xd4 .i.b7 is equal. Now
i.b7 1 1 'flg3! 'flc7 12 2 d6 13 .i.f4 White threatens 15 g5, so there's no
and White had the better game. time to capture on c3.
Another possibility is 8 cxb6 axb6 9 14 :9s 1s o-0 .i.b1
.

135
A lekhin e 's D e fence

Wisely continuing with his develop problematic winning chances. 34.. 1id8
.

ment. After 15...dxc3 White can play 16 is probably the best, though Black still
llab1, which prevents the development has cause for concern over the position
of the bishop. of his king.
16 cxd4 .i.d5 1 7 J:fd1 l0c6 18 l0e1
J:b8 1 9 J:ab 1

34 l0c3!

Completely turning the tables.


1 9 J:lb6
.. 35 J:lxb7
Once again Black must consolidate. After 35 llb2 there is 35 ...1Wc1!.
Although his position is structurally 35 l0xe2+ 36 Wf1 Wxb7 37 Wg2+
..

quite nice, the position of his king gives c6 38 Wxe2 'ifg4+ 39 l0f3 'ife4+
constant cause for concern. 19 ...llxb 1? 40 Wd2 ds 41 'ifh 1 1ia2+ 42 We3
20 .:Xb1 .i.xa2 would be tantamount to 'ifb3+ 43 We2 Wb2+ 44 Wd3 a4 45
suicide after 21 l:tb7 d8 22 .i.b51 etc. 'ifd1 a3 46 c2 'ilb5+ 47 Wc3 :.as
20 a4 Wd8 21 Wd2 Wa3 22 "irh6 48 11a2 'ifd5 49 'ifb1 + Wc7 50 l0d2
'ife7 23 J:lb5 Wg7 a2 0-1
Finkel suggested 23 ... .i.b3, going af
ter White's a-pawn. Game 55
24 'ifh3 Ci:Je7 25 J:db 1 .i.c6 26 J:xb6 Posch-Baburin
axb6 27 aS l0d5 28 .i.f3 'ifg5 Vienna 1995
And not 28 ... bxa5? because of 29
llb8+ e7 30 1ih4+ 1Wg5 31 'ifxh7+ 1 e4 l0f6 2 e5 Ci:Jd5 3 c4 Ci:Jb6 4 c5
llg7 32 1i'h8 etc. l0d5 5 .i.c4 e6 6 l0c3
29 g3 &3 30 J:la1 Ci:Je4 31 'ifxh7 6 1i'g4 was tried in Radojevic
bxa5 32 J:b 1 Wc8 33 .i.e2 .i.b7 34 Bagirov, Trinec 1973, with Black win
'ifh3? ning quickly after 6 ...b4 7 lba.3 b6 8
A blunder in time-trouble. The cor d4 .i.a6 9 1We4 8c6 10 i.xa6 xa6 1 1
rect move was 34 f3!, after which ...d3? (1 1 cxb6 is good for Black but
34 ...1Wg7?! 35 1Wxg7 l%xg7 36 llb3! (pre not fatal) 1 l ...ab4 12 c4 bxc5 13
venting ...ltk3) 36 ... a4 37 lla3 .i.c6 38 dxcS xeS 0-1.
.i.d1 wins the a-pawn and gives White 6 ...ltlxc3

136
Th e Chase Varia tion

There are two other interesting lines:


a) 6 ... .txc5!? 7 d4 .tb4 8 11fg4 ltlxc3

9 g3
9 1i'd2 ltlxe5! 10 .ie2 ltlg6 1 1 J.xc7
a1) 9 a3 .tf8 10 .tg5 (10 bxc3!? - .ixc5 12 .tg3 1ra4 also left White with
Sveshnikov) 10 ... h5 1 1 'irf4 ltld5 12 insufficient compensation in Angelov
.txdS .ie7 13 .ixe7 Wxe7 looks like Suba, Varna 1975.
rather nebulous compensation for the 9 ....e7 10 b4
pawn. After 10 1i'e2 Black obtained the bet
a2) 9 bxc3 9 ....txc3+ 10 ft 'tie7 1 1 ter game with 10...g5! {10...1rxc5 also
l%b1 (1 1 Wxg7 1rf8 1 2 1rf6 l:lg8 also leaves me wondering if White can pos
leaves White struggling) 1 1 ...f5 12 1ih5+ sibly have enough) 1 1 .td2 'tixc5 12
g6 13 1rd1 ltlc6 and White's compensa .txg5 11'xe5 13 11'xe5 liJxeS 14 i.f6
tion was inadequate in Sveshnikov liJxc4 15 .txh8 ltlxb2 in Holzl
Khmelnitsky, Sibenik 1990. Speelman, Hastings 1971/72.
b) 6 ...ltlc6 7 d4 ltlxc3 8 bxc3 d6 9 10 /iJ3 h6 1 1 i.e3 b6 12 cxb6 axb6
cxd6 cxd6 10 exd6 .ixd6 1 1 ltlf3 0.0 12 13 'tie2 .tb7 14 0.0?! (14 0-0-0!?)
0-0 e5 was also fine for Black in Vavra 14 ...g5! was also very promising for
Pacl, Czech Team Ch. 1992. Black in Hegedus-Grunberg, Romanian
7 dxc3 .!Llc6 Ch., Bucharest 1985.
7....txc5 8 1rg4! gives White a dan
gerous initiative for the sacrificed pawn.
After 7 ...ltlc6 Black can meet 8 ltJf3
with 8 ... .txc5 because 'tig4 has been
ruled out.
8 ..t.f4 .h41?
An interesting idea of Mihai Suba
which comes close to winning a pawn
by force. Can White hang on to it or, if
not, demonstrate sufficient compensa
tion?

137
A/ekhin e 's D e fence

10 ... g5 1 1 ..te3l!Jxe5 21 i.f2 i.xf2+ 22 xf2 dxeS with three


pawns for the piece and an exposed
white king to aim at.
1 6 -.xg5 1 7 :h5 -.g& 1 8 l!Jh31
. .

.i.e&

1 2 ..td4!?
An interesting attempt to breathe
new life into White's position. 12 'ifhS
i.g7! 13 i.xgS i.f6 14 i.xf6 Wxf6 15
'ffe2 xc4 16 'ffxc4 bS! 17 Wd4 11'xd4 19 %lg5?
18 cxd4 i.b7 19 f3 aS! gave Black a So far White has played very well, but
huge endgame advantage in Schwarts this is a mistake. He should play 19
Solozhenkin, New York,1994. ll:\f4! We4 20 fl! {threatening 21 f3
1 2 .....tg7 1 3 .i.e2 d6 14 cxd6 cxd6 3 22 i.d3) 20... h6 2 1 f3, aher which
1 5 h4 the forced retreat with 21...'ffh7 leaves
15 f4? is wrong because of 15 ...gxf4 White with definite compensation.
16 gxf4 11'h4+ etc. 1 9 . . .'ife4 20 l:lxg7 l!Jf3+ 21 1
1 5 .....td7 ll:lh2+ 22 e1 ll:lf3+ 23 1 ll:lxd4

16 hxg5! 24 f3
16 f4?! still doesn't work, this time This leaves White two pawns down
because of 16 ... gxf4 1 7 gxf4 i.c6 18 without anythingto show for them. But
lth3 0-0-0 19 Wc2 i.f6 20 fxeS i.xh4+ 24 Wxd4 is hardly attractive as after

1 38
Th e Chase Variation

24...1i'h l+! 25 g1 i.g2+ 26 e1 /1Jxd2 0-0 looks fine for Black) 9 ...lbxa6
'ifxg1 + 27 d2 Wxa1 he loses most of 10 0-0 i.e7 l l ll)c3 /1Jab4 12 a3 0-0 13
his pieces. /1Je4 f5 14 /1Jc3 c6 15 i.d2 /1Ja6 16
24 lllxf3 25 *t'2 llle5 26 .!bt4
. /1Jxd5 exd5 17 'ti'cl l1Jc7; Black's knight
0-0-0 27 b5 .J.e8 28 l:lb1 d5 29 :Z.b4 is coming to the 'dream' e6-square
'Llc4 30 'ira4 Wb8 31 .J.t3 'ire3+ 32 which makes his position super-solid.
g2 1fxc3 33 .!bxe6 l:lc8! 34 .txd5 Neither 8 Wg4 f5 nor 8 c3 /1Jxc3 9
1i'd2+ 0-1 bxc3 d5! 10 .i.d3 i.a6 {Machulsky
r------....., Bagirov, Kirovabad 1973) promise
Game 56 White anything.
Potkin-Neverov 8 ...d6
St Petersburg 2000 Black has also played 8 ...i..a6, after
.______________. which Semeniuk-Mikhalchishin, Cheli-
1 e4 lllt6 2 e5 .!bd5 3 c4 lllb6 4 c5 abinsk 1975 continued 9 i.xd5!? exd5
.!bd5 5 .i.c4 e6 6 d4 b&!? 10 bc3 c6 11 0-0 .i.e? 12 :et (12 f4!?
Another possibility is 6...d6, which f5 13 g4 fxg4 14 llf2, intending g3
actually transposes into a 2 c3 Sicilian and f4-f5, is worth a second look)
after 7 cxd6 cxd6. 12 ...0-0 13 g3 d6 14 f4 f5 with a good
game for Black.
9 0-0 .i.b7 10 1fb3 dxe5 1 1 dxe5
ll:ld7 1 2 1t'g3
Making it difficult for Black to de
velop his kingside. Black tries to solve
the problem by re-routing a knight to f5
although this costs valuable time.
1 2 ll:le7 13 .!bd4 c5 14 1Llxe6!?
. .

7 cxb6
White has also played 7 1Wg4 bxc5 8
i.xd5 exd5 9 i.g5 i.e7 10 .i.xe7 xe7
1 t 1i'xg7 9f8 12 96 (Longschmidt
Zeh, correspondence 1988) and now
Burgess's suggestion of 12 ... l:lg8 looks
very good for Black.
7 ... axb6 8 .!be2
White played 8 /1Jf3 in Machulsky One of those infamous positional
Gurgenidze, USSR 1973, but failed to piece sacrifices. White gets two pawns
obtain any advantage after 8 ... .i.a6! 9 and Black's king is stuck in the centre.
.i.xa6 (9 bd2 .i.b4 10 0-0 i.xd2 1 1 Enough compensation? Who knows?

7 39
A lekhin e 's Defen c e

14 fxe6 1 5 .ixe6 ll:lc6 16 .ie3


.. l:h7 32 11fe4 recovers the c2-pawn with
1We7 1 7 1Wh3 lbd4 1 8 .txd4 cxd4 19 Black's king still exposed.
ll:la31 ltlc5 20 .tg4 h5 21 .tf5 *t7 29 1i'xb6? was bad because of
22 lLlb5 11xe5 23 ltfe1 1Wt6 29 ...11fxal!.
23 ... l:.xa2!? was possible but then 24 29 .txc5 30 ltxc2 lth71 31 1le4
.

l:acl ! would continue the attack. l:lt7 32 11e2 ltdta 33 .:n 1Wg5 34
24 b4 g6 25 .tc2 1Wc4 Wh7 35 li:ld4 11g4!

25 d37
. 36 ll:le6??
Perhaps Volzhins suggestion of A blunder in mutual time trouble. M
25 ...i.c8! 26 11fg3 tt:le6 would have been ter the superior 36 tt:\3! llf4 (and not
better, though in this case too Black's 36 ...l%xf3 37 1i'xg4 :x2 because of 38
king is far from safe. llfxf2! hxg4 39 llxcS llxf2 40 .C.c7+
26 bxc5 dxc2 27 1ld7+ g8 28 etc.) 37 1i'e6! .C.8f6 38 11fxg4 hxg4 39
1Wxb7 ltd8 29 ltac1 tt:leS .C.d6 a draw is the most likely out
According to Volzhin, another inter come, Black's pressure against f2 com
esting possibility was 29 tt:ld6!?, after pensating for the pawn.
which 29 ... i.xd6 {29... bxc5 30 'irb3+ 36 ....:Xf21 37 ll:lxf8+ ltxf8+ 38
..th7 31 tt:le4) 30 cxd6 11fxd6 31 llacl :tt2 'irxc4 0-1

1 40
The Chase Varia tion

Summary
The Chase Variation offers Black his fair share of the play in a sharp and uncom
promising struggle. The plan of 5 . . .e6 (aher either 5 t!c3 or 5 c4) followed, if
necessary, by ...b7-b6, is both sound and economical. But there are some specifics
to be learned here too.
Black should probably pay close attention to Sveshnikov-Solozhenkin and
White's alternative of 8 1i'f3. And instead of following the 8 ... d6 of Potkin
Neverov, I suspect that 8 .. ..ta6 is more promising.

1 e4 f& 2 e5 li:\d5 3 c4 b& 4 c5 d5 5 J.c4 (D)


S t!c3
5...lbxc3 6 dxc3 lik6 7 t!f3 (D) - Game 53
5 . .. e6 6 d4 t!xc3 7 bxc3 b6 - Game 54
5 e6 6 d4

6 t!c3 - Game 55
& b& 7 cxb& axb& (D) - Game 56
.

6 .t.c4 7 li)f3 7. . . axb6

141
CHAPTER TEN I
Other Lines

1 e4 lbf6 structure but obtains open lines for his


The final chapter is devoted to a vari pteces.
ety of unusual systems that have had Rozentalis has tried 3 lL!c3 but more
occasional bouts of popularity before recently he has played 3 g3 (Game 61).
slipping back into the footnotes. None Black needs to play purposefully to
of these lines seems, in principle, to avoid being slightly worse, but a king
offer White a lasting initiative, but the side fianchetto looks like a perfectly
element of surprise is likely to be quite good plan.
dangerous. Last, but not least, we come to 2
Game 57 features 4 f4, which is an lL!c3, which has proven quite irritating
interesting attempt to play a kind of for many Alekhine players. The stan
Four Pawns Attack but without having dard 2.. d5 has come under severe pres
.

committed the c-pawn as yet. The notes sure because of the ingenious im
to this game include other unusual 4th provements thought up by the Swedish
move alternatives such as Romanishin's GM Hector, and his victims in this line
4 .i.e2 and 4 .tc4 (as formerly played include Alekhine specialists such as
by Sax). All of these require accurate Kengis.
handling but the current theoretical
verdict is that they are harmless if met Game 57
correctly. Njobvu-Wohl
In Game 58 we see 3 ...li)b6, an un Yerevan Olympiad 1996
usual line that Westerinen has been
playing. Games 59 and 60 illustrate the 1 e4 lllf6 2 e5 llld5 3 d4 d6 4 f4
old Keres speciality of 3 lL!c3, which is One of several interesting fourth
actually quite a reasonable line that re move alternatives. Here's a round-up of
quires accurate handling from Black. some of the other possibilities:
White has a slightly damaged pawn a) 4 .te2

142
O ther Lines

A highly unusual and experimental


treatment. A more solid way to play
against this line is with 4...dxe5 5 fxe5
c5 (or 5...i.f5!?) and now:
a) 6 c4 ll'1b4! 7 d5 i..f5 8 ll'1a3 e6 is
very unpleasant for White.
b) 6 ll'1f3 cxd4 7 11'xd4 ll'1c6 8 i.b5
and now Kengis has suggested 8 ... i.f5,
intending 9 ...e6. This seems to be quite
comfortable for Black.
5 c3?!
After 5 dxc5 dxe5 6 fxe5 e6 Black al
4 ... dxe5 5 dxe5 i.f5 6 c3 e6 7 ll'1f3 ready has an excellent game and threat
ll'1d7 8 0-0 i.e7 9 a3 0-0 10 h3 aS 1 1 c4 ens 7...1i'h4+.
ll'15b6 saw Black generating some inter White can also try to transpose into
esting play on the queenside in Walter Wohl's line against the Four Pawns At
Mohr, Bled 1992. tack with 5 c4 after which 5...ll'1b4!?
b) After 4 i.c4 most Alekhine spe (5... ll'1b6 transposes into Mamadshoev
cialists like 4 ...ll'1b6 5 i.b3 J.fS and Wohl) is the only move with independ
now: ent significance. Mutzner-Varga, Men
b 1) 6 e6?! i.xe6 7 J.xe6 fxe6 8 ll'1f3 drisio 1999 continued 6 a3 lL14c6 7 d5
ll'1c6 9 0-0 1i'd7 10 ll'1g5 g6 1 1 11'3 ll'1d4 8 i.d3! (8 ll'1f3 is strongly met by
ll'1d8 left White struggling to justify the 8 ...J.g4) 8 ...e6 9 ll'1c3 exd5 10 cxdS
pawn sacrifice in D.Gross-Zvolanek, 1i'h4+ 1 1 g3 11fg4 12 1i'xg4 J.xg4 13 h3
Czech Ch., Luhacovice 1973. i.f5 14 J.xf5 ll'1xf5 15 f2 and White
b2) 6 11'3 is met by 6...1i'c8, though was better, though he lost because of
sacrificing a pawn with 6 ... e6!? 7 11'xb7 later errors.
d5 is also interesting, one point being
that 8 ll'1e2 i.b4+ 9 c3? can be met by
9 ...i.d3! and 10...i.a6, trapping White's
queen.
c) 4 .i.g5 was played in one of
Alekhine's own games, Steiner
Alekhine, Budapest 1921. Black gained
the advantage with 4 ...dxe5 5 dxe5 ll'1c6
6 i.b5 (6 ll'1f3 J.g4 7 J.b5 h6 8 J.d2 e6
was fine for Black in Lutikov-Kopylov,
correspondence 1968) 6...i.f5 7 ll'1f3
ll'1db4 8 ll'1a3 'ifxd1+ 9 llxdl ll'1xc2+ 10
ll'1xc2 i.xc2 11 llc1 i.e4 12 ll'1d4 i..xg2 5 ...c6 6 f3 g4 7 b5?! cxd4 8
13 l1.g1 0-0-0! etc. cxd4
4...c517 8 11'xd4 J.xf3 9 gxf3 e6 is also good

143
A lekhin e 's D e fen c e

for Black.
8 .. .1Wa5+ 9 c3 xc3 10 .i.xc6+ Game 58
bxc6 1 1 bxc3? Stefansson-Westerinen
White had to play 1 1 'ifcl2, aher Reykjavik 1997
which 1 1 . ...i.xf3 12 gxf3 1ld5 13 1lxc3
e6 is good for Black but a long way 1 e4 f6 2 e5 d5 3 d4 b6!?
from decisive.
1 1 ...1Wxc3+ 1 2 ..td2 1Wd3
12....i.xf31 was simpler.
1 3 9b3! 9xb3 1 4 axb3

Westerinen plays this quite ohen and


quite successfully. Black's position is
actually reasonably solid; combine this
with the provocative appearance of
14... .i.xf3? 3 ... b6 and it can be quite effective in
And now it becomes very difficult for practice.
Black to achieve the win. 14....i.e6!, in 4 a4
tending ... .i.d5, would keep Black's ad Another Westerinen game went 4
vantage. f3 d5 5 .i.d3 .i.g4 6 h3 J.xf3 (and not
1 5 gxf3 e6 1 6 lla6! dxe5 1 7 dxe5 6 ... .ih5? 7 e6 fxe6 8 g4} 7 'ii'xf3 e6 8
d7 1 8 e2? 0-0 c5 9 c3 when White could claim to
Missing his chance to save the game. be slightly better, Bjarnason-Westeri
18 .i.e3 would prevent Black's next nen, Torshavn 1997.
move and lead to a draw aher 4 ...85 5 f3
18 ....tb4+ 19 e2 c7 20 :Z.hal. White could also try 5 i.d3, intend
1 8 .....tc5! 1 9 lld1 c7 20 b4 .i.b6 ing to avoid a pin on his knight aher
21 .i.e3 b7 22 llaa1 llad8 5 . .d5 6 lbe2, though with the bishop on
.

Finally I can say that it's a matter of the unusual d3-square Black could also
technique. consider 5 ...d6.
23 .i.c5 lld5 24 .i.d6 h5 25 llxd5 A further possibility is 5 .tb5!? c6 6
cxd5 26 .i.c5 :ca 27 llc1 a& 28 .td3 d5 (6... g6 is more flexible, main
d3 b5 29 d4 a5 30 llb1 a4 31 taining the option of either ... d6 or ... dS)
llb2 a3 32 llc2 :as 33 llc1 a2 34 7 .tg5!? g6 8 h4 h6 9 .te3 .tg7 10
lla1 J.xc5+ 35 bxc5 lla4+ 0-1 lLla6 1 1 c3 .te6 12 lLlh3, when White

1 44
O th er L ines

had dangerous attacking chances in Tal 15 ...l0c6 16 h5 h6 17 J:th4 lDe7 1 8


Lutikov, USSR Ch. 1969. J:tf4 'tld7 19 l0d4 :c8 20 i.e3
5 ...d5 6 i.d3
Or 6 .te2 .tfs (6....tg4 is also possi
ble) 7 0-0 e6 8 b3 lba6 9 lba3 J.e7 10
.td2 lLlb4 1 1 .tb5+ c6 12 i.e2 0-0 13
c3 lLla6 with an unclear position,
Sariego-Arencibia, Linares 1992.
6 ...i.g4 7 h3 i.xf3
And not 7 ....th5 because of 8 e6.
8 Wxf3 e6 9 Wg3 c5 10 dxc5 l06d7
1 1 l0c3 l0xc5!?
1 1. ..lLlc6 12 llb5!? lLldxeS 13 i.f4 f6
14 0-0-0! 11'd7 15 i.e3 llc8 16 i.e2
11'f7 17 f4 lld7 18 llhe1 gave White 20 ...l0c6 21 l0f3 Wg8 22 J:[g4 i.f8
strong pressure in Benjamin-Aiburt, 23 i.b6!?
New York 1990. White manages to win the aS-pawn
but in doing so he loses both time and
piece co-ordination.
23 ...Wh7 24 tLld2 l0e7 25 i.xa5
lLlf5 26 Wf3 i.e7 27 l0b3 %thf8 28
i.c3 f6 29 We2 fxe5 30 i.xe5 i.f6
31 Wg 1 'tlf7 32 J:[f4 i.xe5 33 Wxe5
Wxh5 34 Wxe6?
Maintaining the extra pawn, but now
White's king comes under attack. White
should centralise his knight with 34
lLld4!.
34 . .Wg5! 35 J:tf3 J:tce8 36 'tld7 :e7
.

1 2 lLlb5?!
Giving Black an important tempo for
development. According to Volzhin,
White should first play 12 0-0, after
which 12 ...llc6 (or 12 ... llxd3 13 cxd3
lLlc6 14 lLlbS) 13 lLlbS :lc8 14 .tf4
leaves Black wondering how to com
plete his kingside development.
1 2 ...l0xd3+ 1 3 cxd3 i.b4+ 14 Wf1
Wt8 1 5 h4!?
White still has some kingside pres
sure, but now his loss of castling rights
gives Black chances. 37 Wb5?!

145
A lekhin e 's Defen c e

Probably the losing move. White has


to play 37 1Wxd5, after which 37...ltJh4
38 'i'xg5 ltJxf3+ 39 gxf3 hxg5 would
leave him with drawing chances in the
endgame.
37 ... lDd&! 38 l:lxf8 lDxb5 39 axb5
'ireS 40 d4 'ire2 41 l:lf3 'irxb5
The technical stage is fairly
straightforward; Black combines threats
against White's king and weak pawns
which eventually proves too much for
the defence.
42 l:la5 l:le1 + 43 Wh2 'ird7 44 l:la1 3 lDxc3
.

'irc7+ 45 g3 l:le2 46 lDc5 'irb& 47 Black has also tried 3 ...e6 4 llJxd5 (af
b3 Wb4! 48 l:la4 'ire1 49 Wg2 'irb1 ter which 4...exd5 5 'iff3 may be
50 l:lb4 l:le1 51 l:le3 l:lg 1 + 52 Wt3 White's best) and even 3 ...lbb6!?. But
'irf5+ 53 We2 'irh5+ 54 l:tf3 l:tc1 55 capturing on c3 has to be the critical
lDd3 l:lc2+ 56 We3 'irg5+ 57 lDf4 line.
4 dxc3
White captures 'away from the cen
tre' in order to obtain free piece play,
but it is not the only move. The game
Yudasin-Ehlvest, Biel Interzonal 1993
was a recent, high-level example of 4
bxc3 which continued 4...c5 5 f4 lbc6 6
d4 (6 llJf3 d5 7 d3 was mentioned by
Y udasin as another possible set-up for
White) 6 ... d5 7 llJf3 .i.g4 8 e2 e6
(8. ..'ifa5?! 9 0-0 'i'xc3 10 d2 'i'a3 1 1
llb1 gives White the initiative for his
57 ...'ire7+ 58 Wd3 'ire4 mate pawn) 9 0-0 e7 10 h3 h5 1 1 llb1;
.....-----.., now Black should probably play
Game 59 1 1. .'ifd7 in order to lend better support
.

Rozentalis-Volzhin to f5 with a double-edged struggle in


Polish Team Ch., Zakopane 2000 prospect. In the game he played
.____________...,.
1 1...1!fc7 but after 12 e3 c4 13 g4 g6
1 e4 lDf& 2 e5 lDd5 3 lDc3 14 llJe1 faced the unpleasant threat of
An old line favoured by Paul Keres: 15 f5.
'I know that the theoreticians consider 4 ... d& 5 lDf3
this move of little interest and not dan 5 c4 is well met by 5...lbc6! 6 lbf3
gerous for Black, but his position is by dxe5! 7 W'xd8+ ltJxd8 8 ltJxe5 f6 9 lbd.3
no means without problems.' e5 10 0-0 e6 1 1 b3 .i.d6 12 llel g5

146
O th er L in es

13 .i.e3 f7 14 f3 lbc:6, which proved stronger than 8 ..f6) 9 f3, and now
.

to be clearly better for Black in Volzhin's sensible suggestion of 9 ... e6


Nezhmetdinov-Spassky, USSR 1959. intending ... .i.cS or ... .id6 gives Black
He has safely completed his develop very comfortable equality.
ment and maintains a kingside pawn 8 ....i.d7!
majority. White's idea was that after 8 . c6? it
. .

5 ... dxe5 would be difficult for Black to complete


For S ... c6 see the next game, Toot his development - if he later moves his
hill-Davies. e-pawn the d6-square becomes very
6 1Wxd8+ Wxd8 7 l0xe5 We8 weak. A sample line is 9 .i.c4 e6 10 .i.f4
.i.d6 1 1 0-0-0 e7 12 l:lxd6! etc.
8 d7 is also difficult for Black af
...

ter 9 .i.f4 c6 10 .i.c4 (and not 10 xc6?


a6! 1 1 J.a4 cS!) 10...lilice5 1 1 .i.xe5 f6
12 .i.g3 eS 13 0-0-0; White will be able
to open up Black's king with f2-f4.
9 l0xd7 l0xd7 1 0 .i.f4 c& 1 1 .i.e2
e6 1 2 c4 .i.c5 1 3 c3 e7 1 4 l:d 1
The immediate 14 b4 J.d6 is okay for
Black, but now b2-b4 is a threat.
14 ... a5 1 5 i.d3 l:hd8 1 6 We2 l0f6
1 7 h3 l:d7 18 a4 l:ad8 1 9 .i.c2
Arriving at a critical position. Black l:xd1 20 i.xd1 l0e4 21 .i.c2 e5!
has a sound kingside pawn majority,
whilst White's queenside majority is
difficult to use because of the doubled
pawn. On the other hand Black has not
yet castled and it might be difficult for
him to co-ordinate his pieces.
8 .i.b5+ !?
A clever wrinkle which requires accu
rate defence from Black.
8 .i.c4 has long been known to give
White absolutely nothing: 8...e6 9 J.f4
.i.d6 10 0-0-0 d7 1 1 :bel was
Pfleger-Schmidt, Polanica Zdroj 1971, And not 21.../-Dxf2?? 22 1U1, winning
and now ll . . .'it>e7! would have been the knight.
fine for Black {1 1 ...xe5 12 J.xeS 22 .i.c1
.i.xeS 13 l:lxe5 e7 14 lieS was slightly Keeping the game going. 22 .i.xe4
better for White in the game). exf4 would be completely drawn.
8 .i.e3 doesn't give White anything 22 ...lbd6 23 .i.d3 h6 24 f3 f5 25
either after 8. d7! (less common but
.. J:.d1 J:.f8 26 b3 J:.f6 27 .i.c2 e4 28

747
A lekhin e 's D e fen c e

.i.f4 %-% tected by the doubled c3-pawn. It


After 28 .i.f4 gS 29 .i.xd6+ llxd6 30 blocks both the c-file and the h8-a1
l:txd6 xd6 31 fxe4 f4!? there is noth diagonal.
ing left to play for. 6 . . . g6
In the light of experience I prefer
Game 60 6...dxe5 7 ..xd8+ lDxdS 8 .i.xeS (8
Toothiii-Davies lDxeS f6, followed by ... e7-e5, is fine for
British League 2001 Black) 8... c6, with the game L.Hiibner-
.______________. Baburin, Berlin 1992 continuing 9 0-0-0
1 e4 f6 2 e5 d5 3 c3 xc3 4 f6 10 .i.c7 .i.g4 11 .i.c4 eS 12 llhe1
dxc3 d6 i..e7 13 l:td2 lhe6 and now White sacri
I'm not sure I particularly like 4...d5, ficed unsoundly with 14 lDxeS.
despite its solid theoretical reputation. 7 .i.c4 .i.g7 8 We2
After 5 11ff3 White's queen comes to g3 A few White players have been
and makes life uncomfortable. A recent tempted by 8 lDgS, not knowing that
game Hector-Jessen, Copenhagen 2000 8...dxe5 9 .i.xf7+ f8 10 1ff3 exf4 1 1
went 5 ... g6 6 .i.d3 .i.g7 7 1llg3 cS 8 lhf3 lld1 1fxd1 + 1 2 'ifxdl h6 just wins for
lDc6 and now 9 .i.f4 (Hector castled Black as in Payrhuber-Ligterink, World
short at this point) 9 ...c4 10 .i.e2 .i.fS Junior Ch., Stockholm 1969.
1 1 0-0-0 looks quite good. 8 ...0-0 9 0-0-0
5 f3 c6

9 .i.g4

6 .i.f4?1 According to my database this natural


After the normal 6 .i.bS I was intend move is new. Previously Black had
ing 6 ...g6 (6....i.d7 7 11fe2 is known to played odd moves such as 9 ...'ifd7 and
be uncomfortable for Black) 7 .i.f4 9 ...11fe8.
.i.g7, but then 8 exd6 cxd6 9 'ird2 fol 10 e6 fxe6 1 1 .i.xe6+ .i.xe6 1 2
lowed by castling long gives White in Wxe6+ h8 1 3 .i.g5 Wc8 14 l:lde1
teresting play. Black's central pawn ma Wxe6 1 5 l:lxe6 l:lf7 1 6 l:lhe1 h6!?
jority will find it difficult to advance An interesting pawn sacrifice.
whilst White's king will be well pro- 1 7 .i.d2

148
O th er L in es

After 17 J.xe7 I intended 17 ...g5, A favourite line of Rozentalis with


leaving White's bishop stranded on e7. which White aims for smooth develop
1 7 ... g5 1 8 h4 g4 1 9 d4 ment and a slight space advantage. In
I felt that 19 tt:lh2 ti:leS 20 i.xh6 the game Termeulen-Davies, Haarlem
i.xh6+ 21 l:xh6+ g7 22 l:e6 :h8 2001, White tried the even more indi
would have given Black more than vidual ) 'iff3!? but after 3 ...c6 4 'it'g3 d6
enough compensation. 5 f4 i.f5 6 c3 e6 7 ti:lf3 c5 8 .te2 tt:lc6 9
1 9...e5 20 f4 gxf3 21 xf3 xf3 0-0 .ig6 10 ti:la3 a6 1 1 d4 cxd4 12 cxd4
22 gxf3 .tf6 23 .ixh6 .ixh4 24 bS 13 .td2 'it'b6 White was already in
:h1 :t6 serious difficulty because of the pres
sure against his d4-pawn.
3 ... d6 4 exd6 cxd6
4 ...exd6 5 .i.g2 ti:lf6 6 d4 d5 7 ti:lf3
.te7 8 0-0 0-0 9 tLle5 ti:lbd7 10 ti:lc3 c6
1 1 f4 was slightly better for White in
Rozentalis-Appel, German Bundesliga
1994.
5 .ig2 f6 6 d4 g6 7 e2
The Israeli IM Alex Finkel has sug
gested an alternative plan of develop
ment for White with 7 b3, intending
.tb2, ti:ld2, ti:lgf3 and h3. It looks rea
25 J:txf6 sonable.
The spectacular 25 .ig7+ is refuted 7 . .ig7 8 0-0 0-0
. .

by 25 ...xg7 26 l:xe7+ f8 27 llxc7 Another reasonable line for Black


l:h6, threatening ....tg5+. was 8 . ..d5 9 h3 0-0 10 tt:ld2 ti:lc6 1 1 c3
25 ....ixf6 26 :g1 J:tg8 27 J:txg8+ eS, Rozentalis-Cs.Horvath, European
xg8 28 d2 7 29 b3 g6 30 Team Ch., Debrecen 1992.
.ie3 .ig5?1
30...a6 would have been better, when
Black still has some chances.
31 f4 .ih6 32 .ixa7 f5 33 d3
.ixf4 34 a4 e5 35 c4 e4+ 36 e2
c5 37 .ib6 e5 38 .ia5 5 39
.ie1 e6 40 .if2 .ie5 41 .ie1 d5
42 cxd5+ xd5 43 c3 .if6 % -"h

Game 61
Petrik-Bogdanovski
European Club Cup Heraklio 1997
9 c3?!
1 e4 f6 2 e5 d5 3 g3!? After such a passive move Black will

149
A lekhin e 's D efen c e

have an easy time of it. The logical 1 3 lLle3


move is 9 c4, as played by Rozentalis Again opting for a 'safety-first' ap
himself in this position. Black in turn proach. At this point he should play 13
should probably reply with the flexible llJf4!? (preventing ...J.h3).
9 ....!iJc6 10 llJbc3 and now: 1 3 . . ..i.h3 14 i.xh3 xh3 1 5 lLlf4
a) 10 . ..t..g4 1 1 ..t..e3 Wd7 12 f3 .ih3
.. ..d7 1 6 'Dfd5?!
13 d5 lDe5 14 b3 ..i.xg2 15 xg2 e6 16 Facing a higher rated opponent,
ltcl (Rozentalis-Yermolinsky, Rakvere White is evidently trying to play it safe
1993) and now 16 ...l:He8 would have by exchanging pieces. Although his po
left Black only slightly worse. sition is still solid enough, Black now
b) 10.....i.f5 1 1 h3 a6 12 b3! llb8 13 has whatever chances that are going. 16
..t..e3 h5 14 d5 llJa5 15 llJd4 d7 16 a4 d5 was still the best, at least getting
cs 17 h2 with an edge for White, some space.
Rozentalis-Pribyl, Liechtenstein 1996. 1 6 lLlxd5 1 7 lLlxd5 e6 1 8 lLlf4
.

c) 10 ... a6! 1 1 b3 l:.b8 12 h3 ..t..d7 13 lba5!


d5 (13 ..t..e 3!?) 13...llJa5 14 llJd4?! (14 Preparing ... b7-b5.
J.e3 is met by 14 ... b5 so White should 1 9 b3 b5 20 i.e3
probably play 14 .ib2) 14 ...Wc8 15 After 20 a4 Black should probably
h2 b5! 16 l:.b1 bxc4 17 b4 llJb7 18 play the quiet 20...a6 with continued
.ie3 llJd8 when White had only nebu pressure. 20... b4 looks attractive at first
lous compensation for the pawn in sight; 21 cxb4?! llJc6 22 .ib2 .!lJxb4 is
Rozentalis-Hjartarson, Tilburg 1994. better for Black, but White can play 21
9 lLlc6 1 0 lLla3 i.f5 1 1 lLlc4 Wc8
.. c4 llJc6 22 b2.
1 2 lte1 20 . ....b7 21 l:lc1 l:lac8 22 ._d2 lLlc6
White's eventual downfall in this
game stemmed largely from his refusal
to ever play d4-d5. Here he would be
no worse after 12 d5 llJb8 13 3 .ih3
14 .ixh3 Wxh3 15 .!lJ4 'i'c8 16 'i'e2.
1 2 l:le8
..

The knight is en route for a better


post, either f5 or d5.
23 lLld3
23 d5 is now bad because of 23 . ..le5
24 g2 llJd7 (24... .!lJg4!?) 25 f3 eS 26
llJe2 fS, intending ....!iJf6, with an active

1 50
O ther Lines

kingside pawn majority and pressure 37 c5 g5 38 l:tf3 J.xd4 39 J.xg5


against dS. :lxc5 40 l:.d1 e5 41 g2 l:c2 42
23 ...ltle7 24 h6 h8 25 g5 J.f6 J.xf6 43 llxf6 l:txa2 44 lld4
lilf5!? l:te2 45 l:.df4 l:.c7 46 g5 l:te5 47 h4
Intending to undermine the d4-pawn l:ld7 48 l:td4 l:l.e4 49 llxe4 dxe4 0-1
with ...a7-a5 and ... b5-b4. White hurries
to prevent this but in doing so seriously Game 62
weakens his position. Hector-van der Wert
26 g4 ltle7 27 f4 l:led8 28 11fe2 Berlin 1993
l:ld7 29 d2 ltdc7 30 lilf4 llld 5 31
..f3 lllxf4 32 ..xf4?! 1 e4 f6 2 3
With his kingside looking weak,
White should definitely take the oppor
tunity to exchange queens. After 32
'ii'xb7!? lDh3+ 33 g2 l:lxb7 34 xh3
a5 he is definitely worse, but the posi
tion should be tenable with accurate
defence.
32 ...d5 33 e4?
The final mistake. He had to vacate
the c-file with 33 llcd1, after which
33 . .. a5 (and not 33 ... b4?! 34 c4 'ifxd4 35
'ifxd4 i.xd4 36 ..i..xb4) 34 'ife4 'ifxe4
35 l:xe4 d5 36 lle2 still leaves him on 2 . . . d5
the board. Now he is lost. The real 'Alekhine' move, but one
33 . . ...xe4 34 l:xe4 b4! 35 c4 d5! which has been under pressure of late,
36 l:.f4 a5 due largely to the efforts of Swedish
GM Jonny Hector. 2 ...e5 would trans
pose into a Vienna Game, which theory
holds to be relatively harmless for
Black.
3 e5
3 exdS lbxdS 4 i.c4 is well met by
4 ...lDb6 5 i.b3 lbc6 (in the game Min
nulina-Davies, Osteraker 1995 I tried
the risky S ... cS 6 d3 c6 7 'ifhS e6 8
lbf3 g6, after which White should have
played 9 'ifh3 with dangerous attacking
chances) 6 lbf3 g6!? (more dynamic
Simply threatening to take the pawn than 6 ...i.f5) 7 g5 e6 8 d3 lbd4 9 0-0
on c4. White covers this one but then i.g7 10 lbce4 h6 t l lbf3 lbxb3 which
loses his d4-pawn. was very comfortable for Black in

151
A lekhin e 's D efenc e

Rozentalis-Bagirov, Vilnius 1985. pared to transpose into a Classical


French; 4 d4 e6 5 f4 c5 is the French
transposition, which might be enough
to put Alekhine exponents off this line.
White also has a sharp gambit line in
4 e6!? fxe6 5 d4 and now:
b1) After 5 ...c5 6 1Clf3 lt:'lc6, Hector
has done well with 7 .ib5!?, fighting for
control of the d4 and e5 squares. Hec
tor-Sergeev, Berlin 1995 continued
7 ...g6 8 dxc5 .ig7 9 .ie3 1i'a5 10 0-0
0-0 1 1 llb 1 a6 12 .ixc6 bxc6 13 lle 1
'it'c7 14 lt:'lg5 1Cle5 15 .id4 llf5 16 lL'lh3
3 ...l0e4 lt:'lf7 17 .ixg7 xg7 18 lla4 e5 t9 1Clb6
There are two other moves, both of llb8 20 c4 with a complex struggle in
which lead to complex play: which I, for one, would prefer to play
a) 3 ... d4 4 exf6 dxc3 5 fxg7 cxd2+ 6 White.
.ixdl .ixg7 7 1i'f3 1Clc6 8 .ib5 11fd6 b2) 5 ... g6 h4 .ig7 7 h5 lt:'lf8 s .if4!
(or 8 ...0-0 9 .ixc6 bxc6 10 .ic3 .ixc3+ (White's number one priority is to
1 1 xc3 1i'd5 12 lld1 11fe4+ 13 1Cle2 clamp down on the e5-square; with this
.ia6 14 lldl :adS 1 5 f3 1i'e6 16 f2 in mind, Bagirov's suggestion of 8 1Clf3
with a clear advantage to White, Hec is also interesting, but a dubious alterna
tor-Daillet, Cannes 1989) 9 .ic3 0-Q 10 tive is 8 .id3 tl:k6! 9 hxg6 hxg6 10
.ixc6 .ixc3+ 1 1 11fxc3 bxc6 12 1Cle2 l.txh8 .ixh8 11 .ie3 e5! which gave
.ia6 13 .J:.d1 1i'e6 14 ...e3 1i'xe3 15 fxe3 Black a good game in Bobkov-Sokolov,
and White had slightly the better end correspondence 1960) 8 ... c5 9 llb5
game in Hector-Kengis, Haninge 1992. 1Cla6 (I don't really like this move; a far
b) 3 ...1lfd7 more critical line is 9...cxd4!? 10 lt:'lc7+
f7 1 1 1Clxa8 e5 with a mighty centre
and White's knight on aS is trapped) 10
llf3 .id7 1 1 5 'it'aS+ 12 1i'd2
1fxd2+ 13 xd2 c4, Hector-Maus, Co
penhagen 1990. In The CompleteAlekhine
Burgess gives no comment on this
complex endgame. Personally speaking,
I would not want to play Black - the
extra pawn is useless and he is under
severe pressure.
4 l0ce2
This is probably the most critical
This currently looks like Black's best, move, aiming to gain time against
though in this case Black must be pre- Black's errant knight.

1 52
O th er L ines

4 11t'f3 looks primitive, but it needs winning attack.


careful handling by Black: 4...xc3 5 b2) 12 g3 i.g6 13 .dl b5 14 a3
dxc3 cS! (this natural move may be {14 .i.e2 b4 15 cxb4 xb4 16 a3 c3!
stronger than either s ... c6 or s ... g6; wins) 14... b4! 15 axb4 .i.xb4! 16 .i.xc4!
White gets some temporary pressure (16 cxb4 lDxb4 17 l:td4 c3 also gives
against d5 but it looks as if Black can Black a winning attack) 16 ... dxc4 17
handle it: 11'f3 0-0! 18 cxb4 (18 xc6 hc3!)
a) After 6 .i.bS+ Black's safest reply 18 ...lllxb4 19 1l'a3 xa3! 20 bxa3 lbd5
is 6 ... 6! (6 ... .i.d7 can be answered by 21 i.g5 l:tfb8+ 22 at (22 cl c3 fol
7 e6!? fxe6 8 .i.d3 with dangerous lowed by ...l:lb2 wins) 22...c3 23 l:le2
looking attacking chances for the pawn) l:lb2 24 fl (or 24 f3 l:lc8 25 llle4 llc6}
7 c4 e6 with a complicated game in 24 ...llab8 25 3 lllb4 26 .i.e7 l:ta2+
prospect. 27 b1 lllxc2+! 0-1 was Zahariev
b) 6 .i.f4 .i.f5 7 0-0-0 e6 8 el?! (the Stefanopoulos, Hania 1996.
critical move is 8 c4!?, after which
8 ... .i.e4 9 b3 d4 isn't clear) 8...c6 9
h4?! (and this doesn't help White either;
the calm 9 bl!? looks better) 9 ... h51 10
b1 aS 1 1 l:[d2 {the immediate 1 1
g3 is met by 1 1 ....i.g4, which is why
White was probably regretting his 9th
move at this stage) 11...c4!.

4...f6
The other method of disrupting
White's plans is with 4 ...d4 5 c3 and
now:
a) s ...lllc6 (this is regarded as the
critical line) 6 xd4 xd4 7 a4+ c6 8
xd4 9xd4 9 cxd4 g5 10 i.d3!? {10
i.c4 i.e6!? 1 1 d3 bS 12 .i.xbS cxb5 13
This simply threatens to generate an .ixgS .i.d5 14 f3 e6 gave Black very
overwhelming anack with ... b5-b4; annoying play for his pawns in Hector
White, meanwhile, will unfortunately Werner, Andorra 1988 - and both
find it very difficult to generate any se 10 ...e6 and 10 ... bS are good alterna
rious counterplay: tives) left Black with very little for his
b1) After 12 d4 there follows pawn after 10 ...6 (10....i.e6? 1 1 h4
12 ...d4 13 llxd4 (13 cxd4 .i.b4 wins wins a piece) 1 1 2 g6 12 .ie4 ll:Jc7
for Black) 13 ... .i.c5 14 l:td2 b5 with a 13 b3 .ih6 14 h4 in the game Hector-

1 53
A lekhin e 's D e fence

Anagnostou, Komotini 1992.


b) Black should probably play the
simple 5 ...dxc3!? 6 bxc3 (and not 6
'ffa4+ lDd7 7 'ffxe4 c5) 6...c5 7 d4
ca6 8 f4 e6 9 t0f3 cS, as in Nier
mann-Bagirov, Giessen 1993.
5 d3 ll:lg5 6 .txg5 fxg5

8 ll:lf4
White can also play 8 d4 and after
8 ...c5 (Pedersen-Burgess, Assens 1990)
the move 9 dxcS is critical.
8 ....tf5
An attempt to improve on the 8 ...g6
9 d4 .th6 10 d3 a6 11 .i.e2 c5 12
7 h4! .i.xg4 ofW.Watson-Dunwonh, London
Much stronger than 7 t0g3 e6 8 'fi'g4 1987, which continued 12 ...cxd4 (Dun-
g6 9 h4 t0c6!, which was good for Black worth suggested that 12 ...c4 13 cl
in the encounter Paoli-Alburt, Odessa 1i'b6 might be better, but after 14 .i.xc8
1976. l:lxc8 15 e2 1i'xb2 16 l:lb1 xa2 17
7 . g4
. . llxb7 White looks better) 13 h5 llks 14
In his book The Complete Alekhine .ixc8 1lxc8 15 t0f3 'fi'a5+ (here Dun
Burgess felt that this advance was nec worth suggested 15.....b6 but I doubt it
essary in order to keep the h-file firmly helps aher 16 0-0} 16 f1 lbe4 17 hxg6
closed. hxg6 18 xd4 with White much better
7 ... gxh4 is supposed to be bad for because of Black's vulnerable king.
Black aher 8 f4 g6 9 .:xh4 .i.g7 10 d4
c5 1 1 .i.d3 'ifaS+ 12 f1 cxd4 13 llxh7!
l:lxh7 14 .i.xg6+, the game Vorotnikov
Kengis, Riga 1983 'confirming' this as
sessment after some wild complications:
14 ...d8 15 .ixh7 .ixeS 16 1i'f3 c6
17 xd5 .ie6 18 ..f8+ 4iPd7 19 'ffxa8
'ffb 5+ 20 .i.d3 1i'xb2 21 b6+ 1Vxb6
22 lbf3 .tbs 23 c3 dxc3 24 l:lb1 9xb1+
25 .ixb1 q;c7 and now 26 e1 .id5 27
'itdl! would have won fairly easily ac
cording to Kengis.

1 54
O ther Unes

9 ge2 23 hxg6 hxg6


Here 9 d4 looks like a good alterna
tive.
9 c6 1 0 g3 'W'd7 1 1 d4 lllb4 1 2
.

e6 'W'd6
12 ... .i.xe6 13 c3 ltlc6 14 .i.e2, intend
ing to castle, eliminate the bishop on e6
and recapture the g4-pawn would leave
Black with a poor position without any
redeeming features. Unclerstandably he
tries to confuse the issue but Hector
maintains a vice-like grip.
1 3 lllxf5 'W'xf4 14 llle3 g3
Theres not much choice - 14 ...h5 is 24 l:tag1
answered by 15 c3 lLlc6 16 lLlxd5 'ile4+ End of game. None ofBlacks pieces
17 ltle3 1ixe6 18 d5 etc. can do anything.
1 5 'iff3 'ifxf3 1 6 gxf3 gxf2+ 1 7 24. . ..i.g7 25 J:lxg6 .i.f6 26 l:thh&
xf2 0-0-0 1 8 c3 llla6 1 9 h5! J:lxg6 27 J:lxg6 c5 28 lllg4 .i.h8 29
Sealing in Blacks f8 bishop. f& .i.xf6 30 lllxf6 exf6 31 dxc5 l:te8
1 9 c6?!
. 32 J:lxf6 dB 33 e3 e7 34 l:tf7+
19 ...:g8 was a slightly better try. xe6 35 l:txa7 e5 36 :xa6 J:lb8
20 .i.xa6 bxa6 21 f4! l:tg8 22 f5 g6 37 l:lb6 1 -0

1 55
A lekhin e 's D e fen c e

Summary
One of the main difficulties in dealing with these lines is of a practical nature; they
crop up so rarely that by the time you get to meet them you no longer remember
what to do! I therefore advise occasional revision of anything that has not occurred
in your tournament practice.
Against the 4 f4 of Game 57, I suggest 4 dxe5 5 fxe5 c5. My recommended an
..

tidotes to other lines are contained within the notes to this game. Westerinens
3 ... c!Ob6 might he a useful surprise weapon to have available; it slows the game
down and leads to some quite original play.
Hectors favourite 2 c!Oc3 is not a problem if Black is happy playing a Vienna
Game. The only purely 'Alekhine line I can recommend for Black is Bagirovs
2 . d5 3 e5 c!Oe4 4 l0ce2 d4 5 c3 dxc3, given in the notes to Game 62.
. .

1 e4 .!Of6 2 e5
2 l0c3 dS (D) - Game 62
2 . . .!0d5 3 d4
.

3 c!Oc3 c!Oxc3 4 dxc3 d6 5 c!Of3


5 dxe5 6 1i'xd8+ xd8 7 l0xe5 (D) - Game 59
..

5 ...c!Oc6 Game 60
-

3 g3 - Game 61
3 . d6
..

3 ...c!Ob6 (D) - Game 58


4 f4 c5 - Game 57

2 . . . d5 7 liJxe5 3. . . %6

1 56
I INDEX OF COMPLETE GAMES I

Aseev-Bagirov, Berlin 1990 9


.... . . . . . . . . . . . . .................... . . . . . ............. . . . . . . . . . . . . . . .

Brener-Pushkin, USSR 1988 .............................. ..... ........................... 12 8


Davies-Sinha, Calcutta 1990.............................. . . . . . . . . . . . ........................ 1 00
Degraeve-Miles, Mondariz Zonal 2000 ........... . . . ....... .......... ................... 5 1
Djurhuus-Agdestein, Norw;r;jan Ch., Asker 2000 . . . . . . . . ......................... 123
Emst-Bagirov, Helsinki 1992.......................... ............................... ....... 37
Emst-Josefsson, Swedish Ch., Karlskrona 1983................................ ....... 38
Fedorov-Baburin, Istanbul 0/ynpiad 2000 ............ . . . . . .............. .... ......... 11 0
Grunfeld-Ljubojevic, Riga /nterzona/ 1979 ................... . . . . . . . . . .... ........... 1 1 9
Hagesaether.H-Wohl, Ulxxla 2000 ................. ......... ....... . . . . . . . . . ............ 97
Hamdouchi-Baburin, Saint Vinant 2000 . . . . . . . . . ......... ........................... 14
Hector-van der Werf, Berlin 1993......................................... ............... 151
Honfi-Varga, Budapest 1995 .................................... . . ...... . . . . . . . . . . . ........... 88
Howell.J-Kengis, London 1991 ................... . . . ..... . . . . . . . ........................... 28
Kobalija-Nalbandian, Geller Memorial, Moscow 1999 ...... .. ................. ...... 8
Kulaots-Kengis, Riga Zonal 1995 ..... . . . . . . . . . . . . .. . . .................................... 1 13
Leko- Timman, Wijk aan lee 1996....... ................................................. 62
Levacic-Rogulj, Croatian Ch., Pula 1998 ......... ....... . . ............................. 77
Ljubicic-Zelcic, Croatian Ch., S/awnski Brod 1995 ........ . . . . . . . . . . . . . . . .......... 50
Mainka-Yusupov, German Ch., Bmnm 1998. . . . ....................... . . . . . . . . ....... 12
Mamadshoev-Wohl, Yerwm Ol)mpiad 1996. . . . . . ... ......................... . ...... 12 7
Mendes-Rodrigues.R, Figueira Foz Honra 1999........................ ........... 1 12
Miles-Pons, Andorra 1996 ........ ............................................. . . . ............ 34
Mischke-Schirmer, Com:sponJence 199 5 ......... . ...... .................... ........... 12 5
Mortensen-Baburin, Copenhawz 2000 ......................... ......................... 21

157
A lekhin e 's De fen c e

Mukhametov-Bagirov, Bern 1995 . . . . . . ..... . . . . . . . . . . ................................... 25


Mysliwiec-Krzyzanowski, Correspondence 1995 ..................................... 45
Nedev-Oney, Heraklio 1997 82
. . . . . .......... ..... . . . . . . . .......... . . . . . . . . . . . . . . . . . . . . ........

Nguyen Anh Dung-Anastasian, Budapest 1999 ................................... 1 7


Nijboer-Vaganian, Dutch League 2001 .................................................. 32
Njobvu-Wohl, Yereum Olympiad 1996 ............ ...................................... 142
Nunn-Howell.J, Port Erin 1994 ............................................................ 65
Oral-De Firmian, Reykjtnik 2000 .................... .......... ........................... 59
Paramonov-Bratchenko, PetroffMemorial, St Petersburg 2000................ 1 08
Pegoraro-Henderson, Ischia 1996 ....................................................... 106
Petrik-Bogdanovski, European Club Cup, Heraklio 1997....................... 149
Polgar.J-Agdestein, Isle ofLewis (rapid) 1995 ........................................ 35
Posch-Baburin, Vienna 1995................................................ ............... 136
Potkin-Neverov, St Petersburg 2000...................................................... 139
Reed.E-Danielsson.G, Buenos A ires Ol)mpiad 1939............................... 70
Rowson-Baburin, Port Erin 1999 ......................................................... 86
Rozentalis-Sokolov.A, Bern 1992 ......................................................... 42
Rozentalis-Volzhin, Polish Team OJ., Zakopane 2000.............. .............. 146
Sennek-Zelcic, Croatian Team OJ., Makarska 1994............................... 101
Short-Agdestein, Isle ofLe'llis (rapid} 1995.. .......... ................................. 30
Short-Miles, European OJ., Ohrid 2001 .................................. ................ 54
Sofronie-Zlatic, BraS(Jl) 1998...................................................... ........... 84
Stefansson-Westerinen, Reykjauik 1997 ................ . . . ...................... . . . . . 144
Sveshnikov-Solozhenkin, Russia 1998 ................................................ 135
Thorhallsson-Gausel, Refejauik 1997 ....... . . . .... ............. ...................... 132
Tiviakov-Van der Werf, Dutch OJ., Leeucmrden 2001.............. ............... 49
Toothill-Davies, British League 2001 ........ . . . . . . ............ . . . . . . .................... 148
Ulibin-Baburin, Vienna 1998 .................................. ............................. 72
Van der Weide-Miles, European OJ., Saint Vinant 2000........................ 47
Velicka-Freisler, Czech 0,., Zlin 1997................................................... 93
Vetemaa-Shabalov, USSR 1986.................................... . . . . .................. 1 18
Volzhin-Sveshnikov.L, USSR 1988 ................ .................................... 121
Watson.W-Baburin, Kilkenny 1994 . . . .......... ... . .. ........... . . 53
.. ............ .. .. . ... .

Wydrowski-Marcinkiewicz, Correspondence 1997 . ................................. 67


Yagupov-Balashov, MosaYW 1996 ............................................. ............ 96
Yudasin-Kengis, Minsk 1985 .............................................................. 105
Zamicki-Malbran, A rgentine OJ., Buenos Aires 1998 .............................. 18

7 58

You might also like