ANATOLV

THE
OPEN
GAME . %::
IN

�----
ACTION c
-----
__
FOR CHESS ... READ BATS FORD FOR CHESS ... READ BATSFORD
�----- ,
\

The open games, which commence 1 e4 e5, are some of the most exciting and
important in chess. In this, the first of a four-volume series covering all of the
major openings, former World Champion Anatoly Karpov presents a survey of
the Ruy Lopez, the Russian (or Petroff), the Scotch, the Italian (or Giuoco
Piano) and the Four Knights. Complete games from recent Grandmaster play
are analysed in depth, so that the reader sees the logical connection of the
opening to the middlegame (and beyond).

This is not only an excellent book for players wishing to improve their
understanding of the opening; it is also highly entertaining, and all players will
find much to enjoy.

Anatoly Karpov was World Champion from 1975 to 1985, during which time he
dominated the chess world, becoming the most successful tournament player
in history. His recent epic struggles with Gary Kasparov have proved that he is
the only player in the world capable of successfully challenging Kasparov for
<
the World Championship. 1

Other popular titles from Bats ford

The Semi-Open Game in Action Winning with the Griinfeld
Anatoly Karpov Andras Adorjan and Jeno Dory

Learn From Your Defeats Winning with the Queen's Indian

Anatoly Karpov Zoltan Ribli and Gabor Kallai

Batsford Chess Openings

Cary Kasparov and Raymond Keene
For a complete list of Batsford Chess Books
How to Play the Ruy Lopez
please write to:
Shaun Taulbut
BT Batsford Ltd
Playing to Win 4 Fitzhardinge St
James Plaskett London W1 H OAH

ISBN 0-7134-6096-2

1 1 11111
9 780713 460964
The Open Game in Action
The Open Game in
Action

ANATOLY KARPOV

Translated by V. Kalinin

B. T. Batsford Ltd, London

First published 1 988
© Anatoly Karpov 1 988

ISBN 0 7 1 34 60962

All rights reserved. No part of this publication

may be reproduced, by any means, without prior
permission of the publisher.

Printed in Great Britain by

Dotesios Printers Ltd, Bradford-on-Avon, Wiltshire
for the publishers
B. T. Batsford Ltd, 4 Fitzhardinge Street,
London WI A OAH

A BATSFORD CHESS BOOK

Adviser: R. D. Keene GM, OBE
Technical Editor: Ian Kingston
Contents

Introduction to the four-volume series vii

Introduction to the first volume XI
I The Spanish Game I
2 The Russian Game 55
3 The Scotch Game 93
4 The Italian Game 101
5 The Four Knights' Game 106
Index of Complete Games Il l
Index of Main Variations 113
Introduction to the four-volume
series
.

This publication consists of four volumes: The Open Game in A ction,

The Semi-Open Game in Action, Closed Openings in Action and Semi­
Closed Openings in A ction.
What sort of books are they? It is difficult to define them exactly.
Three aspects come to mind instantly.
1 . These books contain games from the past few years, especially the
period between 1 984 and 1 987, which have been most interesting and
valuable for opening theory. The games included are mainly by well
known grandmasters, including the most important games between
Kasparov and myself in our first three world championship matches.
Also included are games that developed further innovative and original
opening ideas first used in these matches.
So, first of all, these books are a selection of grandmaster games from
the past few years with interesting opening ideas.
2. As we have already noticed these games are chosen according to
the openings. But it is not only a matter of choice but of the form of
presentation. Each of the main games concentrates on one particular
opening, and within the notes to each game there are many references
and even other complete games. So some of the main games look like
an opening note and some even like a whole article. The notes to the
main games often refer to earlier or later games, so the information on
the opening idea can be presented in the context of its development,
how it was perfected and refined. While commenting on the main
games I am often "distracted" by my recollections of the examples of
other grandmasters or of my own examples. I hope that these "lyrical
deviations" will not confuse the reader.
Thus, secondly, the books are a selection of opening discussions that
are devoted to the most critical openings, variations and plans played
by grandmasters during the past few years.
3. The book includes many games played by the author. This is not
surprising, as in less than ten years I have played six world champion­
ship matches-two with Korchnoi and four with Kasparov. As is well
known, in any duel for the crown, the best developments of opening
theory are used, and new ideas are born that attract huge attention and
become very popular. As I have already mentioned, these books
contain the most valuable theoretical games from my encounters with
Kasparov. Apart from these, I have included games played by me in
between these world championship matches, none of which have been
previously published with my notes in English.
viii Introduction to the four-volume series

Therefore, the third aim of these books is to be an addition to the

selections of the games of the twelfth world champion during the years
1 984-- 1 987.
Thus, the main idea of this series is to cover the modern state of
opening theory, the most popular variations which occurred in the
tournaments and matches of the mid-80s using grandmaster games of
recent years. Having become acquainted with the games in any one of
the four volumes you will have sufficient knowledge of the most critical
openings, variations and innovations which have attracted the close
attention of theoreticians and players. You will be able to follow the
development of the opening ideas and additionally their realisation at
the highest grandmaster level. In this sense these four volumes are a
true reflection of their titles. By going through the games you will
acquire a lot of useful information on the openings in action and learn
the latest developments. At the same time it is difficult to guarantee that
you will be able to overpower any grandmaster in the chosen opening.
Nobody has managed to write this sort of universal book and I doubt
that it is possible at all. Playing successfully in the opening depends not
only on the information you possess about the opening, or how many
variations you remember, but more on your ability to find new opening
ideas and exploit them over the board. You also need a mastery of
playing the middlegame and the endgame.
Although the attention in these books is concentrated on the opening
stage of the game all the main games are given from the beginning to
the end. I think that going through the whole game is very beneficial
because it allows the reader to trace the connection between the
different stages of the game. The opening reference books consist of
thousands and thousands of variations which usually come suddenly to
an end in the most interesting place. I recall how sometimes I look
through a game and it breaks off with an assessment of "unclear play",
and regret that it is not possible to see how it ended. In this series this
sort of disappointment is not going to befall the reader.
There is another consideration. Imagine that when the game is
interrupted like this we are told that White has the better chances. But
the question of the explanation of this advantage is left open. At the
same time, studying the connections between the middlegame and
endgame is most important for an improving chess player. That is why
the idea of this series seems good to me. On the one hand, the reader
can enhance or broaden his opening repertoire and on the other hand
the actual reading will not be a boring memorisation of a countless
number of variations. The value of studying chess as a preparation for
tournaments will increase considerably if you do not limit yourself by
just getting acquainted with an opening but if you also analyse the
 Introduction to the four-volume series ix

grandmaster games to the end. The use of a small opening advantage,

the art of transition from an opening to a middlegame, the technique of
defending a difficult position-all these methods of play have great
importance and much attention is devoted to them in these games. In
the end I took into consideration that the actual acquaintance with
exciting grandmaster games brings pleasure in itself!
The apparent difference between this series and specialised opening
books is, firstly, that I have covered only fashionable openings and,
secondly, that even from them, I have picked out the most popular
variations and systems. And this is understandable, because modern
opening theory is so fully developed that detailed analysis of some
individual variations would take as much space as any one of the four
books. At the same time the games that I have annotated, as was
mentioned before, were chosen with the aim of embracing the majority
of the fashionable variations including many from the world cham­
pionship matches.
I want to say some words about the structure of the four volume
series. As is well known, chess openings are normally divided into three
groups- open, semi-open and closed. This classification was formed as
far back as the end of the last century and the beginning of this one.
There was a time when I e2-e4 was almost obligatory. If Black replied I
. . . e7-e5 it was called an open opening, if I . . . e7-e6 then semi-open,
and all the rest (there were not many) were called closed. The situation
has changed substantially in recent years and the popularity of closed
openings has been continuously growing. The number of games played
with closed openings these days is probably higher than the number
played with open openings and semi-open openings together. Thus it is
possible to assume that there is a need to split closed openings into two
classes-closed and semi-closed. Consequently, closed openings
become the symmetrical systems- ) d4 d5 and I c4 c5, and semi-closed
openings become the asymmetrical systems- I d4 (I c4) � f6. Although
this classification hasn't been settled yet I believe that it will be official
quite soon. Anyway, this classification is used for the four-volume
series.
Here are the statistics of the openings of the first three matches
between Kasparov and myself:
open games- 1 2 games,
semi-open games- 14 games,
closed openings-41 games,
semi-closed openings-29 games
(96 games in all from the three matches). As you see, closed openings
(mostly the Queen's Gambit) are played more frequently even in the
x Introduction to the four-volume series

new classification and, using the old classification, their predominance

would be overwhelming.
I want to say a few words about my opening repertoire. It consists of
the main open openings (the Spanish, the Russian, the Italian*), semi­
open (the Sicilian, the French, the Caro-Kann, the Pirc), closed (the
Queen's Gambit, the Slav, the English) and the semi-closed (the
Nimzoindian and the Queen's Indian). Among the popular openings,
only the King's Indian quite seldom occurs in my games. I don't play it
as Black and nobody plays it against me when I am White. All the other
modern openings, as you see, are included in my repertoire and most of
them I play as White and as Black. Therefore I am compelled to
become an expert in opening theory. I hope that after all this is said the
reader will not be surprised by the fact that there are so many games
played by the author included in these books. It is understandable that
when talking about the modern state of an opening it is always
convenient to take your own game as a basis. But I want to point out
that in the notes to my games it is possible that I refer to no less
valuable games and in this case they are thoroughly discussed in the
text of the main game.
Each book of the four-volume series consists of about 30 main games
and the notes cover about another 1 50 games. Thus in the whole series
about 700 games are considered, most of them from the 1 984--1 987
period.
Almost all grandmaster games and games from the world champion­
ship matches were annotated in dozens of publications and so in my
notes I have tried to put different sources together. This series is not
scientific but written in a light style with some lyrical reminiscences.
Therefore, to sum up, that is why I decided not to name the authors of
all the suggested variations, especially as quite often the same moves
were suggested by different commentators.
In conclusion I would like to say that I haven't previously tried to
write this kind of literary work which is based on the modern state of
opening theory and practice. In the four-volume series the period 1 984--
1 987 is covered, including my first three matches against Kasparov. If
the reader finds my books interesting and useful, then I will probably
write another series, on "The Openings in Action", on the material
from the next "cycle", say 1 987-1 990.

A. Karpov

•Throughout this book the accepted international terminology, Spanish,

Russian and Italian, has been preferred to Ruy Loperz, Petroff and Giuoco
Piano.
Introduction to the first volume

Open games have different names in the opening textbooks and

monographs. Apart from those mentioned in this book there are the
King's Gambit (accepted, declined and some other modifications), the
Vienna Game. the Hungarian Gambit, Philidor's Defence, the Two
Knights' Defence and others. All these openings appear quite seldom at
grandmaster and master levels nowadays, so we shall leave them out of
this book. In fact, there are only two open openings which steadily
maintain their popularity. These are the Spanish Game and the
Russian Game. Games where these openings occurred are the main
theme of this volume and only at the end shall we see a few games with
the Scotch Game, the Italian Game and the Four Knights' Game.
There is no doubt that the most popular open game is the Spanish
Game and it is played considerably more often than the Russian Game.
That is why the reader might be surprised that the Spanish has only
slightly more examples in this book (the main games: the Spanish - 1 6,
the Russian- I I ). This can be explained by the fact that the theory of
the Russian Game is developing rapidly and my games with Kasparov
have played a major part in this. In the Spanish the main weight of the
struggle is shifted to the middlegame and so I would say that the theory
of the middlegame is developing rather than the theory of the opening
itself.
There is one drawback to the Russian Game from Black's point of
view. The problem is that if White chooses the ancient variation with
the queen exchange- I e4 e5 2 -df3 -2)f6 3 -dxe5 d6 4 -2)f3 -2)xe4 5 "i!i'e2
JYie7 6 d3 -df6 71L.g5 'i!iixe2 + 8 kxe2-with the aim of finishing the
game with a quick draw, Black has little chance of avoiding this result
(assuming that the opponents' strengths are approximately equal).
That is why if Black wants a lively game he should meet I e4 with I . .
e5 2 -21 f3 -2)c6 or even chose one of the semi-open openings. On the
.

other hand, as is well known, a draw would not be a big achievement

for White.
Thus this book starts with games played with the Spanish Game. Of
course it was not possible to cover all the variations and systems. First
of all the opening discussions between Kasparov and myself are
described here. A lot of attention is paid to the Open Variation. Its
popularity has not decreased after the Baguio and the Merano world
championship matches. We look at the present state of the sharp
Marshall Gambit, and at some rarer continuations, 3 . . . g6 and 3 . . .
-2)d4.
In conclusion, I want to add a little bit more about the proportion of
xii Introduction to the first volume

the Spanish and the Russian games in this book. In the first three
Kasparov-Karpov matches the open openings occurred twelve times.
And what is significant is that there were six Spanish games and six
Russian! So the relative balance between these two openings reflected in
this book is justified.
1 The Spanish Game
Kasparov-Karpov
Game 5, world championship II
Moscow, 1985
Both players are great fans of the
Spanish Game, but we played it
only twice in each of our first
three encounters (I played Black
in all six games) and consequently
we favoured some other openings.
In the first match we played the
Spanish in the final stages in
games 44 and 46. Despite a peace­
ful result in both games, I didn't
like the way they developed and I
was better prepared for the next
match. The following game hap­
pened to be the best of all six
played with this opening. I men­
tion the nuances of the 44th and
46th games of the first match in
the following game, and the 1 4th
and 3rd games of the third match
are mentioned in the analysis of
game 2 of the second match.
I e4 eS
l � f3 � c6
If Black is aiming for a draw 2
. . . � f6, leading to the Russian
Game, is more solid. However,
after my victory in the previous
game, I decided not to object to a
more intense Spanish battle.
3 �bS a6
4 �a4 �f6
S 0-0 A.e7
6 �el bS
7 A_b3 d6
8 c3 0-0
9 h3 A_b7
10 d4 j.e8(1)
Practically everything is played
here - 10 . . . t:td7, 10 . . . �a5, 10
. . . ed, I 0 . . . h6. I chose a varia-
tion worked out by my trainer
Zaitsev many years ago. This plan
is now well established in tourna­
ment practice. Black strengthens
his centre without wasting time on
the preventive . . . h6. Of course,
there is a "danger" that White will
repeat moves by I I �g5 � f8 1 2
� f3 �e8, but grandmasters sel­
dom finish the game in such a
straightforward way. However,
this was the way that Hubner
forced a draw with me in a tour­
nament immediately after the
third match (Tilburg, 1 986) and a
year later Ljubojevic followed suit
(Brussels, 1 987).
11 �bd2
Kasparov played the imme­
diate I I a4 in game 44 of our first
match and after I I . . . h6 1 2
� bd2 ed 1 3 cd -E:�b4 1 4 'l\ii'e2 �f8
1 5 e5 � c6 obtained a significant
2 The Spanish Game
advantage. Black's last move was
a mistake. Commentators sug­
gested 1 5 . . . de 1 6 -dxe5 -2Jfd5!?
However, the exchange in the
centre is premature, in my opi­
nion. 1 2 . . . il..f8 was played in the
9th game of our second match,
and in the 1 4th and 1 6th games of
the third match.
ll il..f8
12 a4 iiitd7
I played 1 2 . . . h6 1 3 il_c2 -2Jb8
in game 9 of our second match.
It's not that I was not satisfied
with the opening result of the 5th
game, but I simply wanted to try
the Breyer Variation (see the next
game). In games 14 and 1 6 of our
third match, instead of retreating
the knight to b8 I chose a different
plan: 1 3 . . . ed 1 4 cd -2Jb4 1 5 it.b l
c5 1 6 d5 -2Jd7 1 7 �a3 c4. I think
that I achieved an equal position
in both games after the opening.
In fact, despite his successful re­
sults in both games, Kasparov
didn't play this opening with
White in the rest of the match.
There is more about these games
later.
13 ab
1 3 lL.c2 has been played. In
Hiibner-Portisch (Brussels, 1 987)
White advanced the d pawn im­
mediately before the exchanges on
the queenside: 1 3 d5 -de7 14 c4
-dg6 15 i._c2 and White had a
standard "Spanish" initiative.
13 ab
14 �xa8 j_xa8(2)
Now if 1 5 -2:}g5 then 1 5 .. . �e7
followed by . . . h6. If 1 4 . . . .!i xa8
1 5 -2lg5 Black's knight is forced to
2
w
retreat to d8 and White captures
the initiative in the centre.
15 dS
This opening variation is dis­
liked by some, but it has still
taken its place in the opening
repertoire. This position was
reached in game 46 of our first
match and on that occasion I
played 1 5 . . . -2!d8 and after 1 6
-21ft h6 1 7 -2! 3h2 -d b7 1 8 11..c2
-2!c5 19 b4 -2la6 20 -2! g4 -2lh7 2 1
.£)g3 c6 2 2 de j_xc6 2 3 jJ_ b 3 .£)c7
24 "lli!f'f 3 -2! e6 25 h4 'i!i!!'d8 26 J;;;;d l
White achieved a considerable ad­
vantage.
Black's knight has four other
possible retreats and three have
been played before. A year later in
Popovic-Smejkal (Zagreb, 1 985)
after 15 . . . .£)e7 16 c4 � b8 1 7
)4e3 c6 1 8 de -2lxc6! 1 9 cb .£)d4 20
il_c4 .axb5 21 b3 l/1a7 22 i.b2 h6
23 � e l .il..c6 White maintained a
slight advantage. Smejkal lost this
game and, obviously, for that rea­
son he played 1 5 . . . .£\ b8 against
Kasparov (Dubai, 1 986). The
game continued as follows: 1 6
-d fl -2:}a6 1 7 .JLg5 JJ..e7 1 8 -dg3 g6
1 9 ·""d2 it_b7 20 �a l ,gaS 2 1
.il..c2 c6 2 2 de $t.xc6 23 �d l �d8

24 ·i>o�e3 '!i-i'b7 25 _kh6 �c7 26
�f5. White had achieved a clear
advantage and it was a disaster
for Smejkal again.
In my game I chose a com­
pletely different place for the
knight . . .
15 �aS
This wins a tempo in compari­
son with the other retreats; at
least it did in game 46 of the first
match.
16 _ia2 c6
17 b4 -2!b7(3)
3
w
Grandmasters considered 1 7
. . . �c4 here when analysing the
46th game of the first match. But
then after 1 8 -2\xc4 be 1 9 �g5 ( 1 9
kxc4 cd 20 ed �c8 2 1 iitb3 iitc7
22 -2)d2 with a slight advantage
for White) 19 . . . cd 20 � xf6 de 2 1
�xe5 �a7 22 -2)g4 'l\lf"xa2 23
-2)h6+ 'lrh8 24 �h5 gf 25 �xf7
_ixh6 26 'iitfxe8+ 'lrg7 27 'fite7+
White has an advantage due to
the fact that Black's bishops have
few prospects.
It seems that Black is now suf­
focating because of lack of
breathing space. But this impres­
sion is deceptive. Grandmaster
The Spanish Game 3
Ta,imanov, when commentating
on this game, cited Nimzowitsch
who said that a piece which has
been restrained for a long time
can obtain formidable strength
when it becomes free. And, in­
deed, in this game my dormant a8
bishop, after being hemmed in by
the b7 knight and c6 pawn, will
liberate itself and make its pres­
ence strongly felt. So in due
course, Black will get the upper
hand.
18 c4
After the natural IS �fl Black
can play 1 8 . . . c5 or 1 8 . . . cd 1 9
ed h6 20 � 3h2 �d8 2 1 � g4 �h7
intending to obtain counterplay
by . . . f5.
18 �c8
It is sensible to occupy a more
promising file before retreating
the knight to d8.
19 de
The surrender of the centre im­
mediately allows Black to achieve
an equal position. An interesting
alternative was 1 9 '\ii'e2 ( 1 9 �b2
or 1 9 c5 are also possible) 1 9 . . .
-2!d8 20 �b2 be 21 �xc4 '\ii'a 7! 22
�a I cd 23 ed �xd5 24 �xd6
.itxd6 (24 . . . �xf3 25 �xc8
l!lf"xa2 26 gf is unclear) 25 �xd5
'1\lf"xal + 26 �xa l gel+ 27 * h2
(27 -E)e I J.. xb4) 27 . . . � xa l (if 27
. . . e4+ 28 � e5) and Black has
sufficient compensation for his
slight material loss.
19 l!lf"xc6
20 c5
Rather a reckless move. Better
was 20 j_ b2. The threat (20 . . . de
2 1 J.. xn+ ) is easily repulsed and
4 The Spanish Game

Black takes over the initiative. 27 �db7

20 -E)dS 2S "llf e2 .:Qd6
21 �b2 de! 29 � xeS jj'xeS
22 be '@!l'g4 �eS
30
Black also had better chances 31 g dl kg6
after 22 o2lxeS llja6! 23 \I{{ a 1 c4 24 32 '8"f4 "i'J;'b4!
Ac3 llja3 2S � e3 �xb4 26 33 '8"cl �e4
-E)exc4 be 27 A xf6 c3! 28 _g g3 34 �el )!it' aS
o2le6 or 22 �xeS o2ld7 23 �b2 c4. 3S j_b3 '!WaS
22 l!f'xeS 36 "il!J'bl b4
23 �xeS -E)d7 37 � e3 �g6
24 A b2 "W'b4!(4) 38 �xeS 'iiii' xeS
This is more accurate than 24 39 ii(el �e4
. . . \\!i'c2 2S "iia l \\!i'a4 26 �d4 40 AdS <2leS
�cS 27 � e3 with sharp play. 41 o2lb3 � d3
Black's last move was sealed.
4 His passed pawn guarantees vic­
w tory. Kasparov decided that there
was no way of escape in the
adjourned position and resigned
without resuming the game.

Kasparov-Karpov
Game 9, world championship II
Moscow. 1985
2S -E)b3?
After 2S l�J'b l White could still I e4 eS
have held the position. White's 2 <2lf3 � e6
desire to play actively puts him 3 j_bS a6
into a difficult position. 4 �a4 -EJf6
2S • . • <2leS S 0-0 �e7
26 � at 6 � el bS
White is practically forced to 7 Ab3 d6
sacrifice this pawn since after 26 S e3 0-0
�xeS �xb2! 27 � e2 llja3 28 9 h3 i_b7
o2ld3 �xe4 29 A xf7 + <2lxf7 30 10 d4 � eS
g xe4 b4 Black has a clear advan­ II -E) bell A fS
tage. ll a4 h6
26 . . • �xe4 As mentioned earlier, I decided
27 �fd4 not to repeat 1 2 . . . "l!lld 7, which I
Black was clearly winning after had played in game S, and chose
27 <2lgS Ac2! 28 "iJlj'xc2 "iJlj'xe l + 29 the Breyer Variation which I had
tlfh2 Ad6 + 30 g3 �e4. played with Black and White
 The Spanish Game 5

many times before. It is interest­

ing that Kasparov played this var­
iation with Black twice in a train­
ing match against Timman
(Hilversum, 1 985) not long before
our match. The Dutch grandmas­
ter played the quiet 1 2 a3, with the
idea of b2-b4, _icl -b2 and c3-c4,
instead of I I a4. Despite the fact
that Kasparov won one game and
lost the other one he had the
better position in both games.
13 .i_c2
The other possibility was 1 3 d5,
but the bishop retreat looks more
solid.
13 -2Jb8
The exchange on d4 and ad­
vance . . .E) b4 was played in the
. The immediate 1 9 'fllc2 (which
1 4th and 16th games of our third
match, which are covered below. I
temporarily stopped playing this
variation because I lost those two
games (not, in fact, because of the
opening) and preferred to play 1 3
. . . � b8 (see Timman-Karpov,
Tilburg, 1 986) after the match.
14 R_d3 c6
15 -2Jfl -2Jbd7
16 -2Jg3 'f!lc7
In Sax-Rivas (Rome, 1 984)
Black played 1 6 . . . g6 im­
mediately and after 1 7 R_d2 � h7!
18 l!rc 1 h5 19 �d l R.g7 20 ..tc2
-2Jb6 21 b3 ba 22 ba a5 23 �d3
� a6 24 J.. xa6 .!;1 xa6 25 de de the
position became absolutely equal.
Possibly, 1 7 h4 h5 (otherwise h5 is
unpleasant) was more energetic
and White is better because of the
weakness of the g5 square.
17 �d2 g6
18 "i!ii' cl *h7
19 b3(5)
1 9 h4, forcing 1 9 . . . h5, was
also worth considering here.
Black's counterplay in the centre
would be insufficient after 1 9 . . .
c5 20 ab c4 2 1 b6! (2 1 ba cd 22 ab
g xa l 23 "l!!i'xa l 'f!lxb7 with good
play for a pawn) 2 1 . . . -2Jxb6 22
_ic2 with an advantage to White.
White played next move) would
now be premature because of 1 9
. . . c5! 20 ab c4.
19 J.. g7
20 "*c2
Preventing . . . d5, for instance:
20 . . . d5 2 1 -2Jxe5 -2J xe5 22 de
-21 xe4 23 -21 xe4 de 24 1:1 xe4 c5 25
J;tg4 'fllxe5 26 _t xg6 + ! fg 27
"\i1xg6 + '!lfg8 28 �e 1 and Black is
worse.
20 -2Jf8
21 � e3 -2Je6
22 � ad 1 � ac8
23 J.. fl J.. f8
24 Kd2
White's main plan involves the
advance of the b pawn with the c
pawn joining it afterwards to
create pressure on the centre.
White could have carried out this
plan at once: 24 b4 \\li'b8 25 "l!!i'a2
figS 26 de de 27 c4.
24
25 'l)!(b1
6 The Spanish Game

26 b4 j_b7 regarded the adjourned position

27 ab as already better for Black, but
It is interesting that the first Kasparov managed to find a
exchange did not happen until the study-like draw.
27th move. 43 hg +
27 ab 43 'l!tc4 gh 44 �xd4 'l!te5 led to
28 �edt 'f!lc7 Black's advantage.
29 �cl j_g7 43 fg
30 gcdl 44 'if(c4 hS(6)
Now 30 de de 3 1 c4 was still
possible. White postponed the c
pawn advance again and I was
able to consolidate my position.
30 .!!t cd8
31 de de
32 gxd8 �xd8
33 � xd8 � xd8
34 c4 be
3S j_ xc4 �e8
Black could more easily have
equalised by 35 . . . _tc8 36 "t��t"a l
�e8 37 'l'lta5 J.. e6. Now after 45 -d xd4 J.. xe4 46
36 --·2 -dd6 -dxe4 'i¥Jxe4 47 -de6 '!'¥le i + 48
37 i.b3 � bS �fl 'i¥Jd2 Black retained some­
If 37 . . . J.. c8 then 38 "tllf"a 5 is what better chances. However,
unpleasant. Kasparov had prepared an unex­
38 h4 pected surprise and now it was I
This advance looked more dan­ who had to take precautions.
gerous twenty moves ago. Now 4S eS!
Black has sufficient counterplay. By opening the b l -h7 diagonal
38 � d4 White creates the unpleasant
39 j_xd4 eel threat of 46 -dxhS! gh 47 *g8 +
40 hS 'Wt e7 'lrh6 48 i_c2. The knight sacrifice
41 'i!i!!' d2 is also possible after 45 . . . h4-46
White has lost an important -dh5! gh 47 i*g8 + *h6 48 j_c2
tempo and the position suddenly forcing mate, or 46 . . . j_xf3 47
becomes sharper. Correct was the i*g8 + 'lrh6 48 -d xg7 iltxg7 (48
immediate 41 *c2 and after 4 1 . . . i_g4 49 f4 i*xg7 50 iitxd8) 49
. . . j_a6 42 j_c4 j_xc4 4 3 "t��t"xc4 *xd8 with a big advantage for
White maintains the initiative. White in the endgame. Admit­
41 cS tedly, Kasparov showed in his
42 i*c2 cb commentaries later that Black
This was the sealed move. We could have escaped in this case as
 The Spanish Game 7

well: 46 . . . .il_xe5! 47 -E)xe5 *xeS 1 e4 e5

48 *g8 + frh6 49 f4 ifte l + 50 2 -21 f3 -2!c6
r;t>h2 gh! (50 . . . ffxh5 51 il..d l + ! 3 .i, b5 a6
'jj6'xdl 52 � xd8) 51 iftg5 + frh2 4 _ta4 -E)f6
52 ifi'xh5 + 'l!ig7 with a draw. 5 0--0 1L_e7
45 il.. xf3 6 �el b5
46 gf il_xe5 7 .1 b3 d6
47 f4! .il_xf4! 8 c3 0--0
After 47 . . . il..g7 48 f5! Black's 9 h3 il.. b7
king is in trouble. 10 d4 �e8
48 'i!!t' g8 + 'i1i'h6 11 -E)bd2 il..f8
49 il_c2 * g7! 12 a4 h6
After 49 . . 'iiif f6 50 -E)e4 or 49
.
13 iLc2 ed(7)
. . . d3 50 il..xd3 'iiiff6 51 -E)e4
�a l + 52 frg2 k.c7 (52 . . . -E)c6 7
53 'ii6'f8 + ) 53 -E)c3! •aS + 54 w
.,adS 'i¥t'c6 55 �1'8 + 'l!ih7 56 il_e4
Black is completely lost.
SO jj· xd8 j_ xg3
51 fg �e5
52 �f8 + <frg5
53 * g2
The storm has settled and we
agreed to a draw: 53 . . . *e2 +
54'1!ih3 'l\li'g4 + (54 . . . 'il!t'xc2?? 55 Since this game and the next
iWf4 mate) 55 <frf2 'l!lte2 + leads to complete our analysis of my
perpetual check. games with Kasparov played with
the Spanish, I want to remind you
Kasparov-Karpov that in the forty-fourth game of
Leningrad, 1986 our first match we played I I a4 h6
Game 14, World 12 -21bd2 ed 13 cd �b4 iWe2, in
Championship Ill the forty-sixth game of our first
match and the fifth game of our
The 1 4th and 16th games of our second match - I I �bd2 11..1'8 1 2
third match are so complicated a4 'l!ltd7, and i n the ninth game of
that it's unlikely that their analy­ our second match - II -E) bd2
sis will be ever complete. For in­ il._f8 1 2 a4 h6 1 3 il_c2 -2!b8. But in
stance, Kasparov has devoted the fourteenth game of our third
several dozen pages of notes on match and sixteenth game of our
them and still hasn't covered all 3rd match we adopted a new sys­
possible variations. We are going tem.
to concentrate only on the open­ Despite the fact that I couldn't
ing part of these games. complain about the opening re-
8 The Spanish Game
suits of the two games played with
the Spanish in our third match,
the defeats in these games affected
me psychologically and I tempor­
arily switched to another well
known move, 1 3 . . . � bS, after
this match.
I want to divert your attention
for a while to demonstrate two
recent games where I played this
move.
Timman-Karpov (Tilburg,
1 9S6): 14 ab ( 1 4 i:_d3 j_cS 1 5 � fl
j_d7 1 6 �g3 'liteS 1 7 �fl 'lltb7 I S
a5 l!ffcS 1 9 -2)h2 -2)e7 20 f4 ef 21
.&xf4 -2)g6 led to equality in
Belyavsky-Giigoric, Sochi, 1 9S6)
14 . . . ab 15 i_d3 JLcS 1 6 � fl ( 1 6
-2)b3 and 16 d5 have also been
played) 1 6 . . . b4 (this was a new
move; after 16 . . . ed 1 7 cd -2) b4
I S k_bl c5 1 9 �f4 White is better)
1 7 -2)g3 be I S be ed 1 9 cd -2) b4 20
j_bl c5 2 1 iLf4 � b5 22 'lltd2 Ma5
23 � xa5 l!ffxa5 24 d5 'i!i!i'dS 25
�d l �d7. The position is ap­
proximately equal. In search of
the initiative, White worsened his
position, as often happens, but in
the end the game was drawn.
Hjartarson-Karpov (Dubai 01,
1 9S6): 16 . . . it_d7 (this was
another new move, instead of 1 6
. . . b4 a s in the previous game) 1 7
-2Jg3 WlCS! I S il_e3 (better was I S
�d2 l!ff b7 19 b4! �aS 20 '!i�ib3
� xa I 21 � xa I );::( aS 22 � xaS
l\otxaS 23 'i'li'b l l!ffeS 24 l\ota2! with
initiative for White, Geller-Gii­
goric, Sochi, 1 9S6) I S . . . �b7 1 9
de (after 1 9 d5 i t is equal) 1 9 . . .
-2lxe5 20 "ga7 '!'¥fcS 2 1 -2) xe5 de 22
'i'fi· f3 ;.:!. e6! 23 1L c2 c5. Black had
the better chances but White mir­
aculously managed to escape in
the end . . . after 50 moves.
14 cd
14 -2)xd4 has also been played
but it is unlikely that the resulting
central pawn formation offers
White any advantage.
14 -2) b4
15 .il_b1 c5
Black has many other possibili­
ties here, 15 . . . �d7, 1 5 . . . ba
and 15 . . . g6, all of which I have
tried at some time. However, the
statistics show that Black has
more chances of counterplay after
the text. Remember that 1 5 . . . ba
was played in the famous Kas­
parov-Belyavsky game in the
interzonal (Moscow, 1 9S2).
16 d5
This is considered to be almost
obligatory. After 1 6 b3 cd 1 7
-2)xd4 ba l S ,l!Xa4 a5 1 9 il_b2 g6
20 il_c3 .:!CS 21 1Lxb4 ab 22
�xb4 il..aS 23 �a4 il_g7 24 JLd3
'i'ii'b6 25 1La6 � cdS Black
obtained the initiative in Tsesh­
kovsky-Balashov (Minsk, 1 9S2).
16 -2)d7
Grandmaster Dorfman, who is
one of Kasparov's trainers, regu­
larly plays 16 . . . g6. But in his
game against Aseev (Lvov, 19S4)
after 1 7 -2) fl il_g7 I S �a3 ba 1 9
Jo;; xa4 a5 20 � a 3 il_a6 21 -2)g3
il_b5 22 �f4 ,ah7 23 �d2. White
had a dangerous initiative.
17 Jo;; a3 (8)
mine the centre favourably by 1 7
. . . f5! l S e5 ( l S ef -2! f6! with
splendid prospects) l S . . . JLxd5
19 il_xf5 il_xO! 20 l!ffx O -E}xe5 2 1

8
B 20 -d fl .!;lxa3 21 ba �d3 22
'iiid
i' l c4 22 � e3 'i!lei'f6 23 �g3 d5.
The pawn "fist" in the centre was
a strong force, so Black won
quickly (Dvoirys-Kruppa, USSR
ch !-final, 1 986).
17 • • • c4
This move concedes White the
important d4 square but it is not
known yet how to create queen­
side counterplay in any other way.
In Kozlov-Gofstein (Aktjubinsk,
1 985) after 1 7 . . . g6 1 8 h4! � f6 1 9
� fl "i!i'd7 20 � g3 �g4 2 1 _ad2
ba 22 'illc l h5 23 eh2! "i!i'd7 24 f4
a5 25 f5 _aa6 26 � f3 fig? 27 _ac3
Black was subjected to a very
strong attack.
18 ab
Here after 1 8 . . . ab White
played 19 �d4 but in the 1 6th
game he played 1 8 -dd4 im­
mediately. It seems that there is
not much difference, but as we
shall see below the game de­
veloped differently. I want to
point out that the natural
exchange on b5 was, in fact,
played for the first time in this
game.
18 ab
19 �d4 �xa3
19 'lll!f' b6 was interesting.
The Spanish Game 9
After 20 � f5 g6 2 1 !!;.g3 ,ga l , or
fi xd3 iii xd4 23 _a e2 'iit't xd I 24
� xd I �c5 Black has the initia­
tive. However, White should play
20 �df3 ec5 21 .it_e3! and the e4
pawn is taboo: 21 . . . -2)xe4 22
.!! xa8 .it. xa8 23 .fl. d2!
20 ba ad3
21 Axd3 cd
22 _ab2 (9)
If 22 � xb5 then 22 . . . .1a6, 22
. . . 'i11i'b6 or 22 . . . �a5 23 �d4
-dc5 and Black has sufficient com­
pensation for the pawn.
Only recently, in the Candi­
dates' Match Sax-8hort, Canada,
1988, the Hungarian grandmaster
played a new move: 22 � e3.
The game continued: 22. . . -2)e5
23-2)xb5 m-a5 24 -2)d4 Mc3
25-2)2b3 i_a6 26 kd2 'i¥i'b2
27kb4 g6 28 f4. Very risky;
28-Md2 looked safer, offering the
immediate queen exchange. 28 . . .
-2)c4 29�xd3 �xe4 30'M'f3 );;( e8
3htth2 kg7 32 -2)c6 'i'ire2 33
abd4 �xf3 34 � xf3 �e4 35
-2)b3 ):;t.e2 36 acd4 );l;e4 37 ac6
;.;; e2 38 'itig3. Sax declines the
repetition and makes the fatal
mistake. 38 . . . -de3 39 h4
;;;;( xg2+ 40 'itih3 i_c8+ 41 f5
10 The Spanish Game
jj_ xfS + 42 � xf5 gf 43 1i. xd6
�b2 44 �e7 + 'itPh7 45 �c5 i_f6.
White resigned.
This position (a) should be
regarded as a starting point for
further analysis. The theory of the
Spanish Game already goes so
deep.
22 "Wt'aS
Some commentators consider
this move to be the best, but it is
also worth considering 22 . . .
�cS, preventing White develop­
ing an attack by 23 � fS or 23
'Wt"g4. After 23 .axbS Black has a
choice between 23 . . . A_a6 with
the initiative for a pawn, and a
forced variation which nearly
guarantees a draw: 23 . . . 't)!(b6 24
a4 _txdS 2S ed �xe l + 26 'Wt"xe l
-dxa4 27 _txg7 *xbS (27 . . .
*xg7 28 'Wt"al + ) 28 _txf8 *xf8
29 il{e3 'il!f"xdS 30 *xh6 + * g8.
Apparently, White should play 23
� e3 with a threat of 24 -2!4b3 .
23 <21f5!
It was impossible to surround
the advanced pawn immediately:
23 J.#.e3 �eS 24 f4 -2Jc4; 23 -2!2f3
-deS 24 -db3 � xb3 2S iii xb3 d2
26 �e2 'itf"a4! 27 il{xa4 ba 28
-dxd2 _txdS; 23 � 2b3 iiia4 24
iii'xd3 jj_xdS and Black gains the
advantage in every line. White
exploits the fact that Black's
queen has been distracted and
tries to organise an attack against
Black's king. The immediate
threat is 24jj_xg7 jj_xg7 2S ii!r'g4.
23 -deS ( 10)
I underestimated the following
exchange on eS when making this
move. Apparently, 23 . . . g6 was
safer. Kasparov considered 24
-21 b3 as the main reply and gave
the following line - 24 . . . 't)!(a4 2S
� xd3 �eS 26 jj_xeS (26 't)!(g3
Jl.. xdS!) 26 . . . 14 xeS 27 f4 � e8 28
�g3 with an unclear position.
Many commentators recom­
mended 24 J.#. e3 when 24 . . . gf
allows 2S J.#.g3 + 'itPh7 26 't)!(hS
but 24 . . . -deS is quite playable.
Incidentally, my move 23 . . . -deS
wasn't such a bad mistake yet.
24 jj_xeS!
It looked more natural to con­
tinue with the kingside offensive
by 24 f4. The quiet 24 . . . -dg6
would have led to a difficult posi­
tion for Black after 2S J.#. fl -de7
26 -de3 or 26 -db3 jj-b6 27 il_d4
't)!(a6 28 -dxh6 + gh 29 fS. But
after the active 24 . . . -dc4! a tense
struggle would lie ahead, e.g. 2S
-2!xc4 be 26 il_xg7 _tc8! 27 il_xf8
jj_xfS 28 _txh6 � xe4 29 � xe4
_ixe4 30 �g4 + _ig6 3 1 fS (3 1
'IM!'c8 + ? 'itPh7 32 'il\'f8 jj-cS + 33
* hI 't)!(d4 with an advantage for
Black) 3 1 . . . 'il\'xdS 32 fih2
'i/lf'eS + with complete equality, or
26 . . . jj_xdS! 27 'il\'g4 'IM!'xe l + 28
* h2 jj-e2! 29 � g3 �e6! and
Black starts a counter-offensive.
 The Spanish Game 1 1

24 de I t i s this move, made i n time

After 24 . . . � xeS 2S � f3 � e8 trouble, which is the decisive mis­
26 ..-xd3 Black has no compensa­ take. 3 1 . . . §J..g7 was also bad
tion for the pawn. after 3 1 . . . �g7 32 d6 j_ xe4 33
25 �b3 \'ili"b6 � c8 + t�Ph7 34 't'td l and the d
Black could also play 2S pawn would have been unstop­
iif"xa3 26 iif"xd3 g6 27 �e3 (27 pable. However, according to
iii"x bS itb4!) 27 . . . iJII'a6 28 .!!la l Kasparov, Black could put up
*b6. stiffer resistance by 3 1 . . . f6 32
26 � xdJ �f3 g a6 (32 . . . *g7 33 I;;.c 6!, 32
The d3 pawn has been "eaten" . . . §J_g7 33 �cS), and it would be
at last but the a3 pawn is not difficult to break through Black's
going to escape either. position.
26 � a8 32 �cS
27 g el g6 33 I;;. xc4
It would have been premature 34 e xb7
to take the a3 pawn: 27 . . . � xa3 35 e xd6 � xd6
28 d6 g6 29 d7! .!!l a8 (29 . . . "ti!Yd8 36 •n rfJg7
30 "tl!'fxbS Aa6 3 1 'i!if"xeS � xb3 32 37 f3 fS
..-es I;;. b8 33 I;;. c8 and so on) 30 38 � fl d2
�e3 I;;.d8 3 1 �g4 §J..g7 32 �cS 39 *e2 �b4
hS 33 �e3 � f8 34 � xb7 "i'i!r"xb7 3S 40 edJ �cJ
� dS! with a considerable advan­ 41 � cS .

tage to White. The game was adjourned in this

28 � e3 §J_xaJ position, but Black resigned
After 28 . . . � xa3 29 �g4 because he was losing the d pawn
Black has serious difficulty and there were no chances of sav­
defending the pawn e.g. 29 . . . ing the ending.
iii"d6 (29 . . . !JLg7 30 d6!) 30 'ili'xbS
� a6 3 1 "i'l!fe8! I;;. xb3 32 � c6! or 29 Kasparov-Karpov
. . . Ata2 30 'i�i'b1 'ili'a7 3 1 � c7 hS Leningrad, 1986
32 �xb7 iii"xb7 33 �f6 + and Game 16, World
White wins. Championship Ill
29 � a1 � a4
It was possible to defend more 1 e4 eS
obstinately by 29 . . . hS 30 iii"c3 2 � f3 -E)c6
f6. 3 JLbS a6
30 � g4 1L.f8 4 ..t a4 -df6
31 � cl s 0--0 j_ e7
Of course, if 3 1 � xeS? §J_g7! 32 6 �e 1 bS
�d7 �xa l + 33 �xa1 'ili'c7 and 7 ..tb3 d6
Black is winning. 8 c3 0--0
31 ·ifrd6 9 h3 A b7
12 The Spanish Game

10 d4 Ke8
12
11 �bd2 ..us w
12 a4 h6
13 J.. c2 ed
14 cd �b4
IS j_b l cS
16 dS -dd7
17 �aJ c4
18 �d4 ( 1 1)

think of giving up his knight in

II such a simple fashion to clear the
B
obstacles in the way of an attack
on the enemy king? 20 � xc4 • . .

This move was played for the
first time in Sokolov-Psakhis,
(Volgograd, 1 985). After 18 . . .
.'2)e5 1 9 ab "11i' b6 20 � 2f3 � bd3
White sacrificed the exchange by
21 j_e3 �xe l 22 �xe l "iftc7 23
�ef3 ab 24 � xb5 "11i'd 7 but didn't
gain sufficient compensation.
However, I had prepared an
opening surprise for this game, IS
. . . "11i'f6!? Thus, the final assess­
ment of the variation 1 8 � e5
• . .
19 ab iitb6 ( 12) was postponed.
A year after our third match this
position occurred in a grandmas­
ter game. It was so interesting that
I must distract myself and demon­
strate the complete game here.
Sax-Short, Subotica, interzonal,
1 987: Here White played the
incredible 20 -E}xc4!!? Lightning
from a clear sky! Who would
21 � g3 �c8 22 1i_xh6 ab 23 � f3
� al Short took the line o f least
resistance and gave up his rook
for the dormant b I bishop, level­
ling the material but not effecti­
vely reducing the strength of
White's attack. 24 �gS � xbl 25
\ff x bl gh 26 �e6 + 'itih8 27 � xf8
;;; xf8. 27 . . . � dB 28 "lli'c I �f6
was bad owing to 29 �g6 + !, But
perhaps Black should have played
27 . . . "11i'd4, with the aim of
blocking the c I-h6 diagonal for
White's queen, before capturing
White's trapped knight on ffi.
Now 28 "lli'c i would be met by 28
. . . "iii'd2 or 28 b3 by 28 . . . �d2.
Even so, in the second variation
after 29 'i'li'c I !;!. xffi (29 . . . � a2 30
'ff;c7 � xffi 3 1 � xd6 winning) 30
� e2, White would have a slightly
better position. 28 � c l \trh7 29
� c3 �eS 30 \ff xb4 �d7 31 �d2
f)c4 32 �dl b4 33 b3 � eS 34
<�ih2!? White intends to advance
the f pawn and, if it is taken,
activate the second rook. But the
pawn sacrifice shouldn't work. 34
. . . �xf2 35 i4 fl �b2 36 � hS f6

37 g f4 'l!l'td2?? This was an appall­
ing time trouble mistake. After 37
. . . �e8 38 i!tf5 + �g6 39 'i!lf"e6
iirc 1 40 � xf6 � xf6 4 1 iirxf6 iirc7
Black's pos1t1on would be
impregnable. 38 iirxh6 + ! Every­
thing is finished by a checkmate.
Black resigned. This game needs
more detailed study.
18 iirf6
19 -2!20 f)c5 ( 13)
Putting the queen on f6 could
have already been j ustified now by
1 9 . . . �d3 20 J. xd3 (not 20
l:i(xd3 cd 2 1 iirxd3 �c5 22 iirc2
g6) 20 . . . b4! (an important Zwis­
chenzug) 2 1 Jtxc4 (2 1 � a l cd 22
'i!ii'xd3 -Elc5 23 iirb l lt_xd5, 23
iirc4 a5 24 �b5 );;;(ac8 with an
advantage for Black) 21 . . . ba 22
b3 (22 b4 � ac8) 22 . . . �c5 with
good play for Black. Incidentally,
it is clear from this variation why
Black's queen is better placed on
f6 than b6 (if 1 8 . . . iirb6); in the
second instance 19 . . . �d3 is no
longer dangerous owing to the
Zwischenzug 20 a5.
20 ab ab
21 �xb5
Not 2 1 �xa8 Mxa8 22 �xb5
.i;g,a l 23 -tJc3 -tJbd3 24 )i:;.fl � b3.
The Spanish Game 13
21 J�xaJ
22 .2)xa3
If 22 ba Black's queen would
penetrate to a 1 .
22 • . •J.. a6
It would have been premature
to play 22 . . . � bd3 owing to 23
J.xd3 � xd3 24 J�e3! For in­
stance 24 . . . � xb2 25 j_ xb2
iirxb2 26 � xc4 or 24 . . . j_ a6 25
'i!i;"a4 g aS 26 J.d2 lltxb2 (26 . . .
-E:�xb2 27 t'rc2! �d3 28 � xc4
'jjef-a l + 29 � e l ; 28 . . . _txc4 29
'/!ll'xc4 K a l + 30 �e l ) 27 � xc4
'l!l'tb l + 28 ..te l ! �c5! 29 iirc6
i!c8 30 iirb6 * xb6 3 1 � xb6 � b8
32 J..a 5 and Black has no com­
pensation for the pawn. That is
why I defended the c4 pawn first.
23 � e3 � b8!?
This move is part of the plan of
putting pressure on the queenside.
However, 25 . . . .2)bd3 was
already possible. Kasparov gave
the following variation: 24 J. xd3
cd 25 b4 � xe4 26 b5 j_ b7 27
� xd3 �c3 28 J. b2 <dxd l 29
il._xf6 -E:�xf2 30 '\tPxf2 gf 3 1 � c4
with an advantage for White.
However, after 27 . . . ·iita l (in­
stead of 27 . . . .2)c3) Black has
sufficient compensation for the
pawn, and it's not even clear if
White will be able to keep this
pawn.
24 e5!?
Now any quiet move like 24
� h2, 24 <de l or 24 � c3 would
have been followed by 24 . . .
� bd3!, so White prepares to at­
tack on the kingside to counter
Black's operations on the queen­
side. But this move has the draw-
14 The Spanish Game
back that Black's queen and pas­
sive f8 bishop are given more
scope.
24 de
25 � xeS (14)
This is another critical position.
It was time to put a knight on d3,
but unfortunately I chose the
wrong knight. The correct choice
was 25 . . . �cd3!, and Black
could look with hope into the
future. Indeed, after 26 -E)g4 (26
Axd3 �xd3 27 � xd3 cd 28 -E)d7
'i!�t'd6 29 <2lxb8 llfxb8 30 llt'a4
would lead to a draw) 26 . . . 'l!l'fd4
(26 . . . 'i!l!rb6 was recommended,
but then 27 .i!!: g3 g6 28 J.,e3 with a
dangerous initiative) 27 �c2
� xc2 (of course not 27 . . . lltxd5
28 <2l f6! + gf 29 � g3 + and 30
•g4 mating) 28 J..xc2 Ac5 with a
complicated position.
Actually, having summed up
the results of our opening discus­
sion, it is possible to finish here.
The variations considered show
that Black's new plan has passed
the test in a fitting manner. Now
we shall have a cursory look at the
remaining dazzling moves of this
game.
25 .abd3?
26 � g4?
This is a return mistake by my
opponent. 26 '1)!jc2! would have
led to gloomy consequences for
me. The f2 pawn would be
defended, the queen would reach
the dangerous diagonal, and, as a
result, my knight would be
pinned. Now Black can play
neither 26 . . . � b3 27 �d7, nor
26 . . . � d8 27 -E)exc4 � xd5 28
b4, nor 26 . . . � b4 27 �c6 � b7
28 )4e8 g6 29 �e3 )4xb2 30 J.,d4
with a clear advantage for White
in each case, though the game
wouldn't be finished yet. The
position becomes dynamically
equal after White's inaccuracy.
26 'lli!tb6
27 � g3 g6
This is the safest reply,
although 27 . . . * h8, or 27 . . .
-E)e4 were also possible.
28 j_xh6 'lli!f xb2
29 iitf3 -E)d7
30Axf8 *xf8
31 .h2 � b3!
This move has been criticized
by many commentators, but in
reality it was stronger than 3 1 . . .
il!txa3, 3 1 . . . -.ci or 3 1 . . . ffg7.
32 J.,xd3 cd?
This is a terrible mistake and I
shall try to explain it. First of all, I
admit that I regarded the position
as already better for Black. In­
deed, White's queenside had been
eliminated, and his kingside at­
tack seemed to me to be ineffec­
tive. Naturally, I started my cal­
culations with 32 . . . cd. I spent
plenty of time and in the end
rejected this move. I considered
 The Spanish Game 15

other continuations but, alas, Only this move, made in ter­

every time I came to the conclu­ rible time trouble, finally loses the
sion that Black didn't have an game. It was found afterwards
advantage. Having established that 33 . . . d2! still maintained
this sad fact, I glanced at the clock chances of escape. All the same I
and realised that the hand was in regard 32 . . . cd in this game as
the danger zone. And I automati­ the most dramatic mistake of the
cally made the move on which I whole match.
had spent most time . . . Of 34 eh6 'fjfe7
course, I saw the following simple 3S � xg6 *eS
variation: 32 . . . g xd3 33 i/rf'f4 36 K g8 + 'lle7
ilfr'xa3 34 �h6 i/rf'e7 35 K xg6 i/rf'e5 37 d6 + ! 'lle6
36 ilfr'xe5 (this time the combi­ 38 � e8 + *dS
nation which occurred in the 39 � xeS + e xeS
game wouldn't be possible: 36 40 d7 g b8
� gS + 'lle7 37 d6 + � xd6 3S 41 � xf7.
� f5 + 'llf6 39 'lilt xeS 'II xeS 40 Black resigned.
.:axd6 * xd6) 36 . . . � xeS 37
J;;. xa6 J;;.x d5 3S � aS + 'lle7 39
e fS + *e6 40 �e3 when White Timman-Karpov
would have only a symbolic ad­ Bugojno, 1986
vantage in the ending which
would be impossible to utilise. 1 e4 eS
Even stronger was 32 . . . g xa3, 2 � fJ �c6
Kasparov gave the following long 3 �bS a6
variation: 33 'i!lt'f4 g xd3 34 4 _ta4 � f6
iitd6 + 'II g7 35 iitxd7 � xg3 36 fg s 0-0 _te7
� b7! 37 h4! � aS!! 3S 'Wt'dS '!1i1i'd4 6 � e1 bS
39 • xaS -wrxg4 40 '!!\ta l + *f8! 4 1 7 �b3 d6
d 6 •es 42 iita4 + 'lidS 43 )'!lta5 + 8 c3 0-0
*eS 44 '!'ii' bS + iitd7 45 'Wt'e5 + 9 h3 _tb7
"Wte6 with a draw. The classical Tchigorin Varia-
33 'i!il'f4 iltxa3( 15) tion 9 . . . .:aas 10 �c2 c5 II d4
'Wtc7, 1 1 . . . cd 1 2 cd 'fltc7 or 1 1 . . .
-2Jc6 is not too popular nowadays
and Romanishin is the only pro­
minent grandmaster who plays it
all the time. The statistics of the
last few years are not very encour­
aging for Black.
10 d4 .!;le8
11 -21 bd2 � f8
12 a3
16 The Spanish Game
We covered 1 2 a4 in great detail
in the notes to earlier games. The
quiet pawn move 12 a3 is no less
popular now. White protects the
b4 square from the Black knight
and intends to develop an initia­
tive on the queenside. Of course,
this move is not necessary and, for
instance, 1 2 i.._c2 was possible.
This position arose after ll -2)g5
�f8 12 -E:�fJ �e8 13 -2!bd2 kf8 14
JLc2 in Sokolov-Karpov, Bugojno,
1986. After 14 ... -2!b8 15 a4 I
tried an innovation which was not
quite successful, 15 . c5, and
. •
White closed the queenside by 16
d5 -dbd7 17 b4 c4 18 -2)0 and
carried out an energetic attack on
the kingside: 18 ... -2Jh5 1 9-2)3h2
g6 20 ke3 i.._ e7 21 'ii<i" d2 ;,; f8 22
jj_h6 -2)g7 23 -2Jg3 1trh8 24 -2Jg4
-2)(6 25 -d xf6 i.._ xf6 26 � n �·d7
27 f4 aS 28 f5 ab 29 cb ba 30 ;.;;; f3
1trg8 31 'ii<i" f2 JLh4 32 .iJ... xg7 JLxg3
33 ;.;;; xg3 'lixg7 34 f6 + 1trh8 35
J;:;ga3 �b5 36 'ii<i" e3 �g8 37 h4 g5
38 hg )4 g6 39 lti'fl h6 40 ;,:; hi
<lih7 41 ii_dl � ag8 42 'ii<i" h3 � h8
�h8 43 JLh5. Black resigned.
This game was quite impressive
and was deservedly named the
best game in /nformator 41. How­
ever, I think this fact served the
young grandmaster badly; it
seemed that he overestimated his
abilities before the world candi­
dates' superfinal . . .
The correct move in my game
with Sokolov was 1 5 . . . -2)bd7,
retaining the possibility of under­
mining the centre, instead of 1 5
. . . c5. In a later round a t Bugojno,
1986 there was a quick draw in
Ljubojevii:-Portisch: 15 ... -dbd7
16 i... d3 d5!? 1 6 . . . c6 is normal.
17 ab de 18 -dxe4 -d xe4 19 JLxe4
JLxe4 20 � xe4 ab 2 1 .iL_g5 f6 22
i.. h4 � xal 23 'ii<i" x a1 iJ.. d6 24
l¥ta2 + <11 h8 25 de -2)xe5. Draw.
After 12 .iL_c2 -2Jb8 ( 1 2 . . . g6 is
also often played), apart from the
blunt 13 a4, the quiet 13 b3 is also
played . Here is a recent example
of this from my own experience.
Ljubojevii: -Karpov, Tilburg,
1986: 13 ... -2!bd7 14 il.b2 g6 15
a4 )4 b8 (this was a theoretical
novelty; the standard moves are
1 5 . . . c5, 1 5 . . . ]Lg7 and 1 5 . . .
c6) 16 kd3 c6 17 'ii<i" c2 -2Jh5 18
-2)h2 (after 1 8 c4 ed 1 9 -2Jxd4 b4
Black takes over the initiative, but
better was 18 kfl e.g. 1 8 . . . -df4
1 9 g3 -de6 20 h4) 18 .. . -2)f4 19
it n -2Je6 20 -2Jg4 kg7 21 de.
Nothing is gained by 2 1 -df3 h5
22 -2Je3 ed 23 cd (23 -2Jxd4 -2Jec5
with a better position for Black)
23 . . . c5 24 d5 (24 de JLxe4!) 24
. . . -2!d4 25 -2)xd4 cd 26 -2Jd I 8 c8
and Black has a clear advantage.
21 ... de 22 -210 h5 23 -2Je3 fftc7
24 � ad1 s bd8 25 h4. Draw. 25
. . . -2Jdc5 26 g3 ;.;;; xd I 27 ;,:; xd I
�d8 with total equality.
12 h6
13 JLc2 -2)b8
13 . . . d5!? was also worth con­
sidering.
14 b4 -2!bd7
15 kb2 g6 ( 16)
After 15 . . . -21 b6 16 de de 17 c4
and White is better.
16 c4
In Psakhis-Razuvaev (Mos­
cow, 1986) White tried a new

move- 1 6 'illr' b l ! The aim of this
move was to overprotect the e4
pawn and free the d2 knight for
relocation via -E:J d2-b3-aS. After
1 6 . . . j_ g7 1 7 -E:J b3 c6 1 8 � aS
'fi{c7 1 9 Jtb3! � adS 20 'i/lf a2 dS 2 1
ed -E:�xdS 2 2 de � xeS 2 3 -2\ xeS
� xeS 24 c4 White achieved a con­
siderable advantage. More accur­
ate was 1 7 . . . � c8 ( 1 7 . . . ;..:, b8 1 8
-E:�aS j_a8 1 9 dS) 1 8 �aS j_ a8 1 9
d S � b6 20 a4 'illr' d7 2 1 a b a b 22
Jtd3 -E:�hS 23 c4 be � xc4 -d f4 2S
� xb6 cb 26 _a c t with approxi­
mate equality (Psakhis-Portisch,
Sarajevo, 1 986).
16 ed
1 7 cb ab
IS � xd4 c6 ( 1 7)
19 a4
An improvement was played in
The Spanish Game 1 7
a game between two Hungarian
women chess players: Madi-Ver­
oci Hungary, 1987; 19 .i.d3! this is
the idea of the international mas­
ter Perenyi 19 . . •1l,g7 20 � cl
'i!it'b6 2 1 ..-b3 -2)e5 22 Jt fl � adS
23 -2)4f3 � xf3 + . It was better to
strengthen the centre by 23 . . .
� fd7. 24 'i!i{xf3 � e6. 24 . . . � xe4
2S it_xg7 -2) xd2 26 'M' f6 * h7 27
i'i!xn. 25 -EJ b3 � deS 26 kd4 'illr' a6
27 ..-c3 c5. Black couldn't resist
the pressure along the c file, but
Black would also lose after 27 . . .
;s xe4 28 il._ xf6 � xe I 29 � xe I
.i;;( Xe l 30 j_xg7. 28 be it_xe4 29
i0i' b4 'i'li' aS 30 cd .i_ f8 31 il_c5
�d5. The only way to continue
resistance was 33 . . . -2)d7. 32
ii1d4 k xg2? This was a second
time trouble mistake. 33 ;,:; xe6
k xfl 34 14 xeS and Black resigned
after a few more moves.
19 ba
20 .zt._ xa4 'ii!i' b6
This move was played for the
first time by Kasparov in the first
game of his match with Timman
(Hilversum, 1 98S). Before then 20
. . . ;.:;. c8 was preferred. After 20
. . . ..-b6 2 1 bS cb 22 il._xbS dS! 23
� xa8 jj_ xa8 24 'l\lf a4 -2\cS 2S 'l\lfc2
� b8 Black took over the initia­
tive and won the game. Timman
varied with 21 -2)c2 in the third
game of the match. He repeated
this move in our game as well.
21 -del 'i'li' c7
Black could also try 2 1 . . . 1t. a6
with the intention of meeting 22
iL_ b3 with 22 . . . .i_ d3 when 23
l4 xa8 � xa8 24 iOi' f3 .it_ xc2 2S
JLxc2 'i!iot' xb4 26 ii_c3 'i!i1CS 27 -d b3
18 The Spanish Game
llj'c4! lead to an advantage for
Black.
22 it. b3 i;g, xal
In the third game of the Tim­
man-Kasparov match, the world
champion chose 22 . . . JLa6, and
after 23 � c l JJ.. g7 24 -2Je3 .,ib5
the position was equal. However,
Timman later managed to win
this imperfect game, although he
was on the verge of losing.
23 JJ.. xal
After 23 il!r"xa l 1t. g7 24 -2)c4, 24
. . . d5 leads to equality, but 24 . . .
c5 is worth a try e.g. 25 'fWi'a5 (25
be � xc5, and Black is better, but
25 e5! leads to an equal position)
25 . . . � xa5 26 ba ."E) xe4 27 iL_ xg7
*xg7 28 f3 d5 29 fe de 30 JJ.. xc4
);;;. xe4 and Black has the initiative.
23 i!. g7
24 -2)e3
The position would also be
equal after 24 -2)c4 �xe4 25 � xe4
(25 f3 k xa l 26 il1xal d5!) 25 . . .
� xe4 26 ii!r"xd6 i11 xd6 (not 26 . .
l*c8 owing to 27 -21 b6! threaten­
ing 28 \¥f'xg6) 27 -2) xd6 Jl. xa l 28
-2)xe4.
24 cS
This move leads to a relief of
tension. It was very tempting to
take the pawn by 24 . . . �xe4. In
fact, after 25 � xe4 );;;. xe4 26
il. xf7 + 'l/xf7 27 'iii' f3 + -2) f6 28
it_xf6 l;;;. xe3! 29 \\ilixe3 i!.xf6 30
� xh6 c5 Black would be better.
However, 25 ."E) g4! gave White a
dangerous attack for instance: 25
. . . -2)ef6 (25 . . . d5 26 1J. xg7
'lt xg7 27 �xe4 de 28 -wtd4 + ."E)e5
29 f4) 26 l4 xe8 + -E:J xe8 27 il_xg7
(but not 27 'l'¥f' f3? d5 28 il.. xg7
* xg7 29 ii!r"e3 h5 30 ii(r"xe8 hg 3 1
hg -E:J f6 32 -wte2 � f4 with an ad­
vantage for Black) 27 . . . ff xg7 28
."E)e4 h5 29 JL xf7! 'il!rxf7 30 -E:J g5 +
*e7 3 1 -wtf3! -2)ef6 32 �ef6 32
-2)xf6 -E:J xf6 33 i11 e 3 + winning.
25 be -2) xc5
26 1t. xf6 JJ.. xf6.
Draw.
After 27 -dd5 1l.. xd5 28 il.. xd5
the position would be absolutely
equal.
Geller-Tal
Sochi, 1986
The Open Variation of the
Spanish Game was one of most
popular openings in the Baguio
and Merano matches, and in the
second, it is fair to say, it played a
decisive role. However, this varia­
tion never occurred in my
matches with Kasparov. It has
been quite a rare guest in my
recent games- I can single out
. only two games with Yusupov in
1 983. But I still follow closely all
the developments in this varia­
tion, the more so, because nearly
all the games covered below de­
velop ideas born in the matches
between Korchnoi and me. Inci­
dentally, Korchnoi hasn't stopped
playing his pet line and still often
uses it in tournaments, despite its
failure at Merano. But the most
fervent supporter of the Open
Variation is grandmaster Yusu­
pov. He plays it regularly (the
Russian Defence is the only
choice he has . . . ), and, most of
the time, successfully. It wouldn't
be an exaggeration to say that

every one of his games makes a
valuable theoretical contribution.
Several dozen grandmaster games
have been played with this varia­
tion since Merano and the most
important ones are covered be­
low. Some games are singled out
for detailed treatment and the rest
mentioned in passing.
1 e4 eS
2 -21fJ l£1 c6
3 .ibS a6
4 .1 a4 �f6
S 0-0 -2J xe4
6 d4 bS
7 .1b3 dS
8 de .1e6
9 �bd2
The immediate 9 c3, which is
mentioned below in the notes to
Popovic-Timman, is also popu­
lar. As for the ancient 9 11!fe2, it is
temporarily out of fashion. I want
to remind you of a long forced
variation which was known some
thirty years ago -9 i�te2 .1e7 1 0
� d l -2J c S I I ,i xdS j_xdS 1 2 �c3
j_c4 1 3 � xd8 + � xd8 14 iite3 b4
I S b3 j_e6 1 6 � e4 � d l + 1 7 � e l
l£)d4 1 8 .1 b2 e xc2 1 9 11!fe2 � xa l
20 ,ixal 2 1 � xeS ,i xcS 2 2 -2!d3
.ib6 23 -2J xb4 0--0 24 ec6 f6 2S
h4 fe 26 ili'fxeS � f6. Now the well­
known continuations 27 g4 and
27 ed4 have been joined by
another one . . .
Timman-Yusupov (Montpei­
Iier candidates', 1 985); 27 .'El d8
.1 f7 (27 . . . .tc8 28 'iiit d S + • rs
2 9 .'Elc6!) 2 8 �xf7 •xf7 ( 2 8 . . .
� xf7 29 'i/ltxa l � xf2 30 <ll h2 3 1
'i!\i'e5) 29 "Wtxa l � xf2 30 *h2 aS
31 'i!!i' e S h6 32 a4 (32 a3!) 32 . . . g6
The Spanish Game 19
33 *dS + * g7 34 ..-es + *f7 3S
hS � fS! with equality. So, the
assessment of the forced play
from the opening to the endgame
hasn't changed throughout all
these years. Black's position is
safe.
Another move worth mention­
ing is 9 ,ie3 and then 9 . . . .1e7
10 c3 0--0 I I � bd2 (the l Oth and
I I th moves can also be played in
reverse order). This position is
well-known to theory, and the fol­
lowing continuations are con­
sidered: I I . . . l£)xd2, I I . . . '/ili'd7
and I I . . . .,;l g4. A recent game,
Uim-Kengen (9th Correspon­
dence Chess Olympiad, 1 982-
1 98S) continued I I . . . ecS 1 2
j_c2 '!lli' d 7. After 1 3 b4 l£)a4 1 4
j_xa4 ba I S 'il(( xa4 -2J xeS 1 6 'i!!i' xd7
-2J xd7 17 -21 b3 c5 1 8 e xeS l£l xcS
19 be � fc8 20 ed4 j_ xcS 2 1
� fb i 1t,xd4 2 2 ,i xd4 � ab8 23
1;4 b3 � bS! with equal chances.
Talking about 9 .1e3, it re­
mains to be said that contribu­
tions have been made to the varia­
tion 9 .1e3 ,ie7 10 c3 '!lli' d 7 I I
-21 bd2 during the past few years.
In Timman-Korchnoi (Reykja­
vik, 1 987) after I I . . . .i::l d8!? I I
• . . -21 xd2 is normal) 1 2 .'El xe4 de
1 3 'il(( x d7 + j_xd7 1 4 -2J gS -2) xeS
I S .1d4 _txgS 16 _t xeS 0--0 1 7
.1 xc7 J;g, c8 1 8 .1 b6 M fe8 the posi­
tion was completely equal.
9 -2)cS
10 c3 d4 ( 18)
The most popular move these
days is the ancient 10 . . . j_e7
which was played in Hiibner­
Korchnoi.
20 The Spanish Game
11 � xe6
Before going any further, it is of
course worth considering l l
� gS!? This forceful move was
found by my trainer Zaitsev while
preparing for the match in
Baguio. The knight is deliberately
put en prise to Black's queen -it is
not the sort of move that every­
body would find! In fact, Korch­
noi didn't accept the knight. Let's
look at this game in more detail.
Karpov-Korchnoi, game l 0,
world championship, Baguio
l 97S; I I . . . de 1 2 � xe6 fe 1 3 be
ti-d3 1 4 �f3 (if 1 4 �c2 ilt' xc3 I S
iithS + g6 1 6 � xg6 + hg 1 7
'i!iJ'xhS tl'xa l and White has
no compensation for the lost
material, but 1 4 t\' g4 was worth
considering) 1 4 . . . iit'xd l I S
�xd l j_e7 1 6 j_e3 �d3 1 7 j_ b3
* f7 I S J Ud i eJdxeS 1 9 eJ xeS
� xeS 20 JJ.. f4 �c4. Not 20 . . .
j_d6 2 1 �xeS � xeS 22 � fe l
j_ xc3 23 I;� xe6 * f8 24 I;� d7! etc.
Here I played 2 1 j_ xc4 and after
2 1 . . . be the chances soon became
level. Tal recommended 2 1 I;� d7
cS 22 I;� e l � adS 23 I;� b7 � d3 24
j_gS K eS 2S 1. e4 g xc3 26 h3
and White maintains dangerous
threats. However, instead of 23
. . . � d3, 23 . . . .!;i dS is stronger!
and Black can be confident of the
future.
But the question is whether it
was possible to take the knight!?
A year after Baguio Timman­
Smyslov (FRG, 1 979) continued
I I . . . \'lf'xgS 1 2 � f3 0--0---0. Better
than 1 2 . . . �d7 1 3 j_ xf7 + *e7
14 �b3 ! \'lf'xeS I S � xeS il!(xcS 1 6
K el + \li d S 1 7 cd 'fjif'd6 I S Jl g S +
*cS 1 9 ii ac l ! -pointed out by
Keene-or 1 2 . . . *d7 1 3 �dS!
j_ xdS 14 'iWf' xdS + j_d6 I S cd
�xd4 1 6 � c4! �e2 + 1 7 w h l
)ll(( fS I S � xd6 '1ll&' d 3 1 9 'i!ii' x f7 +
*c6 20 JLe3 K af8 2 1 'iiei' e 7! ..-ds
22 K ad l �d3 23 e6 and White
obtained a strong attack (Bron­
dum-Brinck-Ciaussen, Denmark,
1 979). 1 3 j_ xe6 + fe 14 'i!¥;'xc6
.-xeS I S b4 'i!!i' d S 1 6 'iirJ' x dS ed 1 7
be de I S -21 b 3 d4 1 9 j_ a3 j_e7 20
j_ b4 Jl f6 2 1 a4 *d7 22 ab ab 23
I;� a6 c6 24 � d l *e6 2S � xc6 +
*dS 26 K xf6! *c4! and the
chances can be considered equal,
despite the fact that White has got
two extra pieces at this moment.
So far, it seemed that taking the
knight passed the test. However,
A. Lilienthal later suggested a
substantial improvement 23
K fd l ! \li e6 24 g ae l *f7 2S \li ft
K heS 26 � d3 � e4 27 g3, Black's
central pawns have been stopped,
and he faces difficulties.
I think that the theory of the
knight sacrifice on gS is far from
complete. Incidentally, it has been
known for a long time that White
gains nothing by I I cd -d xd4 1 2
-d xd4 Q xd4 1 3 j_ xe6 � xe6 1 4

iJ!;'f3 � d8 1 5 a4 �b4! 1 6 ab ab 1 7
-crtc6 + 'tit;'d7 1 8 -crtxd7 .!!1 xd7 with
equality (an old variation by V.
Nenarokov).
Back to I I �xe6 which has
nearly always been played in re­
cent games.
II �xe6
12 cd
1 2 a4 and 1 2 .£Jb3 have also
been played, but White didn't
achieve anything.
12 -Elcxd4
( 19)
/9
w g c8 20 ii!(e4 � bd4 2 1 � xd4 cd
This pos1t10n is the starting
point for the modern theory of the
Open Variation. Now 1 3 .a xd4
'ii6- xd4 1 4 'l!l'tf3 � d8 1 5 a4 reaches
the position from the old game
Capablanca-Lasker (St. Peters­
burg, 1 9 1 4). I tried two dangerous
new moves here in the Merano
match: 1 3 �e4 ( 1 4th and 1 6th
games) and 1 3 a4 ( 1 8th game). 1 3
�e4 became popular afterwards,
which is why the following discus­
sion is devoted to this move. But
first let's have a look what hap­
pens after 1 3 a4.
13 a4 }Le7. (We also considered
1 3 . . . j_c5 and 1 3 . . . j_ b4. 13• . .
El b8 was played in Nunn-Tim-
The Spanish Game 21
man, Amsterdam, 1985, and after
14 ab ab 15 �e4 �e7 16 -Eld6 cd
17 �xd4 -2l xd4 18 'iV{xd4 de 19
l!lfxe5 0--0 20 �f4 S! b7 21 'tit;'e4
� d7 22 �c6 a draw was agreed. I
still believe that 1 3 . . . �e7 is the
most logical move.)
14 � xd4. 14 ab was worth con­
sidering, though in Sax-Yusupov
(Sofia, 1 984) after 14 • • . -Elxb5-
after 14 . . . ab 1 5 g xa8 'jif xa8 1 6
-2:)xd4 -2:) xd4 1 7 "il{(g4 � e6 1 8 f4
White is better- 1 5 'i!i'c2-the
immediate 1 5 .E) b3!? should be
tried - 1 5 . . . 0--0 1 6 ,a b3 c5 1 7
Jl.. e3 )lllj" b6 1 8 K a4 � fb8 1 9 ,a a5
22 �xd4 the chances were equal.
In Hiibner-Yusupov, Tilburg,
1979, White played differently on
move 1 5; 15�e4, but didn't
achieve anything. After 15 . 0--0
. .
16 J.. e3 c5 17 � a4 ii!�'b8 18 "il{(a1
abd4 19 j_ xd4 cd 20 .E) xd4 a xd4
21 � xd4 �xeS 22 � fdl .i;i fd8 23
b3 i4 xd4 24 ·tl1 xd4 )'i!J' xd4 a draw
was agreed.
1 4 . . . -crtxd4. The right capture.
In the 1 8th game of the Merano
match Korchnoi took with the
knight: 1 4 . . . � xd4 1 5 �e4-
also possible was 15 ab ab 1 6
� xa8 'Wrxa8 1 7 'il!i'g4 .£J e6 1 8 f4 -
l 5 . . . �e6-if 1 5 . . . 0--0 1 6 ab
,a xb5 1 7 �e3 Black still has
problems, and if 1 5 . . . ,a f5 then
the manoeuvre 1 6 'i!lff3 -E)d4 1 7
'i/lfg4 would become even
stronger- 1 6 it. e3 0--0 1 7 f4 'WI' xd l
1 8 � fxd l .i;I fb8 1 9 � d7 . White
achieved a considerable advan­
tage which soon became decisive.
This victory was my sixth and
22 The Spanish Game
brought the Merano match to an
end.
1 5 ab *xe5 1 6 ba 0--0 1 7 � f3
�b5 1 8 '1\lri'a4 -.:txa4 1 9 � xa4
�c5. The chances are equal. This
was proved by Psakhis-Dolmatov
from the 49th USSR champion­
ship, 1 98 1 , soon after the Merano
match. Despite the fact that this
game is quite old, there have been
no new ideas suggested since then.
13 �e4 JJ.. e7
Recently Black has almost
automatically chosen this move,
but 1 3 . . . *d5 14 � xd4 � xd4
may also be playable. The posi­
tion after 1 5 -2l c3 '!\ltd? 1 6 JJ.. e3
1J.. c5 ( 20) occurred twice in the
49th USSR championship 1 9 8 1
and i n both cases Black success­
fully overcame his opening diffi­
culties.
Romanishin-Yusupov: 1 7 'lliJ' h5
( 1 7 ii!'d2 � d8 18 � ad l 0--0) 1 7
. . . 'il'fe6 1 8 � ad l g d8 1 9 l;l d2
0--0 20 X fd l g6 2 1 ·!Wfh4 Ja.e7 22
�e4 <E:J f5.
Belyavsky-Dorfman: 17 -EJe4
,la7 1 8 Ii c I 0--0 19 �c5 .it, xc5 20
.:1 xeS g fd8 2 1 ..1 xd4 'CJi'(xd4 22
'il't'xd4 � xd4 23 K xc7 K d2 24
g fc l K e8 25 � 7c2 � xc2 26
� xc2 f6.
Perhaps 1 5 X e 1 was more
accurate than 1 5 �c3, for ex­
ample: 1 5 . . . J.. b4 ( 1 5 . . . JJ.. e 7 1 6
� f6 + !) 1 6 �c3 -.d8 (after 1 6 . . .
'ltd7 1 7 JJ.. e3 JJ.. c 5 White has an
extra tempo in comparison with
the position of the diagram,
although it is not clear how im­
portant it is) 1 7 J.,g5! * xg5 1 8
'Wfxd4 J., xc3 1 9 'Wfxc3 0--0 20
l;l ac l (Malchikov-Odeev, USSR,
1 983). The weakness of Black's
pawns on the queenside is
obvious.
14 JJ.. e3 �fS
This position occurred for the
first time in the 1 6th game of the
Merano match. In game 1 4
Korchnoi mistakenly played 1 4
. . . � xn + and after 1 5 * x f3 0--0
1 6 l;l fd l 'iii' e8 1 7 � f6 + ! J., xf6 ( 1 7
. . . gf 1 8 e f J.,d6 1 9 � d4 * h8 20
� h4 � g8 2 1 g xh7 + * xh7 22
)'l(( h 5 mate) 1 8 ef 'jjf c8 1 9 fg � d8
20 h4! White had a clear advan­
tage.
15 'il'fcl 0-0
16 g ad1
In the 1 6th game of the Merano
match, I played 1 6 �eg5 J., xg5 1 7
-2l xg5 g6 (it was easier to achieve
equality by 1 7 . . . � xg5 1 8 'i!ii' x f5
-2le6 1 9 'jjf e4 c5 20 f4 f5) 1 8 � xe6
fe ( l 8 . . . � xe3? 1 9 'il't'c6) 19 l;l ae l
(the immediate 1 9 .it,c5!? was pro­
bably better) 19 . . . 'i!ii' d 5 20 b3
� ac8 21 1l_c5 x fd8 22 h3 'i!ii' c6 23
b4 � d7 24 Iii d 1 � cd8 and Black
overcame the opening difficulties
this time.
16 � f6 + JLxf6 17 j(;'xf5 JJ.. e7
 The Spanish Game 23

1 8 � ad l 'il{(c8 1 9 �d2 � d8 (Van trate on achieving a knockout

Der Wiei-Korchnoi, Sarajevo, blow on the kingside.
1 984) is unpromising. 20 ecs
16 � xeJ 20 . . . �e8 was played in the
17 fe next game. Black requires only
It is amusing that the isolated one tempo- �g5-and the weak­
doubled e pawns have their own ness of the e pawns will count.
merits, like the control of the d4, However 20 . . . �g5 was bad
d6 and f6 squares and, additio­ now, owing to 21 )ll({ d5 'ifj-e8 22
nally, the f file is opened for man­ �g4 with a very strong attack for
oeuvres as well. White.
17 'i!!J' c8 (21) 21 iiei' f3 �e8
22 �gJ � d8
More flexible was 22 . . . � b8
21
and 23 . . . li b6, or 22 . . . aS and
w
23 . . . � a6 in order to connect the
rook to the defence of the
kingside. The second possibility
occurred in the next game.
23 �g4 �e6
24 -E)fS *h8
25 � xe7
White decided to go into the
endgame with an extra pawn but,
This position has occurred apparently, stronger was 25 h4
quite often in recent years. This (25 . . . � d7 26 I!IJ'c6) or 25 'tii' g3,
game has led us to it. It is analysed increasing the pressure on the
in more detail in the next game. kingside.
18 h3 25 ·tifxe7
I played this move for the first 26 i/6' xf7 iltf' xf7
time against Yusupov (Linares, 27 � xf7 hS
1 983). 1 8 Ji d3 soon led to equa­ 28 � fl Ai dS
lity in Hiibner-Hort (Wijk aan 29 Ji fS g6?
Zee, 1 986): 1 8 . . . c5 1 9 l2:) d6 ttb8 After 29 . . . Ji d2! it was un­
20 � fd l � a7 21 � d5 'i!!( a 8 22 a4 likely that Black risked losing.
� d8. 1 8 -E)d4 has been played The f6 square weakness would
many times. There are examples immediately become important.
in the notes to the next game. 30 e4!
18 � d8 This Zwischenzug makes all the
19 �h2 � xd 1 difference. White first defends his
20 'i!l!l'xd1 pawn with gain of tempo and only
It is sensible to leave the rook then plays � f6.
on the f file in order to concen- 30 M CS
24 The Spanish Game

If 30 . . . � d2 then 3 1 � f6 This defeat convinced Tal that

would still be extremely un- he should start playing this varia­
pleasant. tion with White. The next game is
31 � f6 )!;( xeS an example.
32 � xg6 -E)c 5
33 � c6
33 ;,:; g3 won easily . Tal-Korchnoi
33 -d xe4 Reykjavik, 1987
34 .?J xe4 ;,:\ xe4
35 1ti f2 � e5 I e4 e5
36 ;,; xc7 b4 2 -d f3 .£Jc6
37 b3 3 ..it_b5 a6
37 � b7 was more accurate, 4 k a4 -2! f6
although the rook ending turned 5 0--0 .2\ xe4
out still to be won after the move 6 d4 b5
order adopted in the game. 7 .ii b3 d5
37 *g8 8 de .;.e6
38 � b7 � a5 9 dbd2 -2!c5
If 38 . . . a5 White would gra­ 10 c3 d4
dually win by bringing his king to 11 JJ.. xe6 -E) xe6
h4 and playing g3. 12 cd -2!cxd4
39 ;.; xb4� xa2 + 13 .El e4 j_e7
40 <lr g3 *g7 14 Jt. e3 -2! f5
The game was now adjourned 15 �c2 0--0
and White easily won in the 16 � ad 1 -E) xe3
second session. 17 fe 'i!tt" c8
41 ;,. a4 ;.::, b2 17 'iii' e8 used to be con-
42 b4 *ti sidered bad because of 18 �d4. In
43 ;4 xa6 h4 + Van Der Wiel-Korchnoi, Wijk aan
44 <lr xh4 �. xg2 Zee, 198 7 after 18 g d8 19
a b6
• . .

45 .� d2 -2! xe6 g xdl 20 )4 xd1 fe White

46 <lr g4 � d4 + captured the unprotected pawn,
47 <lr g5 8 d5 + but Black turned out to have suf­
48 1ti f4 E!; d3 ficient counterplay. 21 'i!(:( xc7 'i!tf" h5
49 h4 � d4 + 22 � n � xfl + 23 ;!ti xfl 'i!(:(f5 +
50 <lr g5 � d5 + 24 -2!f2i!1b1 + 25 \tiel 'i!txb2 + 26
51 <lr h6 � d4 * f3 .i f'S 27 -2!e4 'i'Ji' xa2 28 .2\g5.
52 h5 ..,; d5 White has come very near to
53 b5 � c5 Black 's king, but Black finds a
54 � b8 very elegant way to force a draw.
55 b6 � b5 28 . . 'ioifd5 + 29 <lr g3 'iti h8!
.

56 b7 l! b6 + White's queen is tied to the e5

57 'ttr h7. pawn and there is also the threat
Black resigned. of . . . h6. So White is forced to

repeat moves- 30 -2) 17 + <ftg8 31
-E}gS *h8. Draw.
It is worth trying to find out
what happens after White plays
1 8 h3, 19 -2)h2 and so on, as he
does when the queen is on c8. In
this case, after the exchange of
one pair of rooks, Black's second
rook is left on f8. It guards the n
square here, but at the same time
is rather passive.
18 h3 (22)
22
B
As I have already pointed out,
this modest pawn move occurred
for the first time in Karpov-Yusu­
pov, Linares, 1983. Before then
the normal continuation was 18
� d4 -2) xd4 19 ed 'ii!r' e6 . There was
an unexpectedly quick finish in
Tseshkovsky-Yusupov (Erevan,
1 982): 20 -2l g3 f6 21 -2) fS fe 22
'ii!r' b3! and Black resigned because
he was losing a piece. The right
way was shown by Yusupov three
years later: 20 . . . c6 2 1 -d f5 � fe8
22 � d3 i_ f8 23 � h3 g6 24
-E}h6 + i, xh6 25 Jo;. xh6 c5 with an
equal game (Short-Yusupov,
Montpellier candidates', 1 985).
Apart from 20 -dg3, other con­
tinuations are not dangerous
for Black: 20 � d3 f6 2 1 'Wi xc7
fe 22 'i:Wfxe5 .i;l xfl + 23 \ti xfl
The Spanish Game 25
(Chandler-Yusupov, Minsk,
1 982); 20 f,tf xc7 )4 ac8! 21 'iff a5
� c2 22 .i;l f2 .i;l fc8 (H iibner-Lju­
bojevic, Tilburg, 1 982). Black had
no difficulties in either game.
18 Jo;. d8
19 -dh2
I delayed this move in the origi­
nal game: 1 9 � c l ( 1 9 � xd8 +
'i!Jxd8 20 � d l 'iiit e8 2 1 -dd4 -2l xd4
22 ed c6 and the position is equal)
19 . . . c5 (better than 1 9 . . . i!�!'b7
20 l!li'c6!) 20 l!li-f2 � b7 21 -2) h2!
'iff xe4 (2 1 . . � f8 22 'i!\f'f3! and
.
-d g4) 22 litf xfl + *h8 23 'ili' xe6
i_ g5 24 1t> h l . White maintained
the mttlatlve but Yusupov
managed to draw in the end. 1 9
-2) h 2 i s more consistent.
19 a xdl
20 'ii{f xd1 'ii!r' e8
20 . . . -2) c5 was played in the
previous game and was covered
there.
21 'i!\f'hS -deS
22 -2l g3 aS! (23)
This is an important improve­
ment. Black has found a simple
way of relocating the rook to the
kingside. If you remember, in the
previous game (with the White
queen on f3) we said that this
method of defence promised
26 The Spanish Game

equality. However, Black was too defensive problems once and for
subtle in this game and allowed all by taking control of the g4
Tal to carry out an effective plan. square.
23 �f5 Jl, a6 33 itt d8 g6?
24 �g4 K g6 In this game Korchnoi demon­
The knights are beautiful, but strated the unlimited resources of
Black's rook covers all the vulner­ the Open Variation, but he spoilt
able squares around his king . . . the finish and allowed Tal to play
25 b3 a beautiful combination winning
This is a waiting move. White's the queen. After 33 . . . K e8 the
pieces are excellently placed but it likeliest result would be a draw.
is not clear how to improve his 34 �h6 + fl g7 (24 )
position any further.
25 .td8 24
26 � fl w
White i s compelled t o beat a
retreat. His e pawns have become
a real weakness, although they
cannot yet be taken. (26 . . . itt xe5
27 � h6 + ).
26 itt c6
27 e4 llt e8
After 27 . . . l£) xe4 then 28 itt d l
l£)d6 29 �g4 � b7 the knight has 35 �fg4!
been thrown back and White has Tal's cavalry becomes fierce.
compensation for a pawn. Korch­ 35 hg
noi is betrayed by his attempt to 36 K xf7 + fl xh6
gain an advantage at low cost. 37 itt xf8 + figS
28 itt d 1 J... e7 38 h4 + ! flxh4
29 itt d2 .tf8 39 itt h6 +
Black is in a hurry to relieve the White could have finished the
tension. He apparently rejected 29 game even more beautifully by 39
. . . b4 because of White's occupa­ K h7 + flg5 40 itt h 6 + fl f6 4 1
tion of the d file. itt h4 + 1!i'g5 4 2 K f7 + . White's
30 itt xa5 queen manoeuvre is like an end­
31 itt xb5 game study . . .
32 itt b8 39
White has acquired dangerous 40 g3 +
threats- l£) g4 and, if the oppor­ 41 itt f4 + !
tunity arises, �gh6 + . Not 41 \'!!i' e3 + fl h4 42 K f2 g3
32 h5 43 flg2 \'!!i' g5
32 . . . Jl e8 was more accurate 41 fl h4
but Black intended to solve his 42 'i!!i' fl + g3
 The Spanish Game 27

43 � f4 +
25
B
44 "«f xg3 +
45 K h4
White's combination has now
triumphed and the rest was sim­
ple.
45 K xe4
46 .l;l xbS + 'ilr xbS
47 "«f xc7 �e6
48 tltf' b7 K eS
49 a4 �(4 amount of theory on it. This game
so tltf' f3 + 'ilrgS presents an opportunity to survey
51 b4 K el + the most interesting and modem
52 'ilrfl K bl developments in this line.
53 bS J� b2 + 12 "«!td7
54 'ilre l 'ilrfS Strangely 12 ... d4, which
55 "«!t c3 K b4 theory considered to be fine, has
45 b6 �ciS completely disappeared from
57 tltf' c8 + tournament play. The other op­
Black resigned. tions- 1 2 . 0--0, 1 2 . . . j_h5 and
. .

the move played in this game­

Hiiboer-Korchnoi often lead to the same position.
Tilburg, 1986 Here are a few recent examples:
1 2 . . . J.. h5 1 3 � fl 0--0 1 4 � g3
I e4 eS J., g6 1 5 � f5 tltf' d7 1 6 g4 K fd8 1 7
2 � f3 �c6 � 3d4 �xd4 1 8 cd �e6 1 9 J.,e3 c5
3 J.,bS a6 with sufficient counterplay for
4 J.. a4 � (6 Black (Thipsay-Agzamov,
s 0-0 � xe4 Frunze, 1 986). Kupreichik-Kai­
6 d4 bS danov (Kuibishev, 1 986) contin­
7 J.,b3 ciS ued: 1 2 . . . 0--0 1 3 <E:J fl J.. h5 1 4
8 de J.. e6 �g3 J.,g6 1 5 � f5 't!t'd7 1 6 g4 (this
9 '2J db2 �cS position is familiar from the pre­
10 c3 J.,e7 vious game), and after 1 6 . . .
I I J.,c2 J.,g4 ,l adS 1 7 h4! <E) e4 1 8 � xe7 +
12 K el (25) � xe7 1 9 � h2 �c5 20 J.,e3 <E)e6
This position can arise from 21 J., b3 c5 22 h5 J.. e4 23 h6!
different move orders, most often White had the initiative.
by 9 c3 (avoiding the light- There are three more important
squared bishop exchange) 9 . . . examples: 1 2 . . . 0--0 1 3 h3 j_h5
�c5 1 0 J.,c2 J.,g4 1 1 � e 1 j_e7 1 2 1 4 '2J b3 <E) e4 15 j_ xe4 de 1 6
� bd2. The variation has a long *xd8 K axd8 1 7 K xe4 K d 1 + 1 8
history and there is a considerable 'i1r h2 f6 with compensation for the
28 The Spanish Game

pawn (Mikhalchishin-Kaidanov, �c4 20 � c l h6 2 1 b3 l£l b6 22

Kuibishev, 1 986). .i_e3 c5 23 h5 � h7 24 'i!!i' c2
1 2 . . . � h5 1 3 l£! fl 'i!l!t'd7 1 4 j_ xf5 25 gf d4! and Black had
l£lg3 �g6 1 5 l£l d4 l£lxd4 1 6 cd successfully resolved all his
l£le6 1 7 �e3 c5 1 8 l£l f5 0--0 1 9 de problems (Gavrikov-Haritonov,
� xc5 20 � xc5 l£l xc5 2 1 � c l Sverdlovsk, 1 984).
� xf5 22 � xf5 l£le6 23 ». c3 g6 24 15 A g6
�g4 .1!1 ac8 25 X h3 with an ad­ 16 �d4 0-0 (26)
vantage for White (Short-Garcia,
Dubai 01., 1 986)
26
w
12 . . . �h5 1 3 l£l fl 0--0 1 4 l£l g3
�g6 1 5 �e3 lltd7 1 6 b4 l£l a4 1 7
� xa4 ba 1 8 a3 X fd8 1 9 J,.g5 � e8
20 � xe7 ,g xe7 21 g a2 �e4 22
l£l xe4 de 23 'f!lxd7 ,g xd7 24 e6!
and White's chances were slightly
better (Geller-Agzamov, 52nd
USSR championship, 1 985).
13 h3
I want to remind you that I had
already played 13 l£lb3 in the 28th Until this game the normal
game of the Baguio match, and move was 1 6 . . . l£l xd4 e.g. 1 7 cd
after 13 l£le6 14 h3 �h5 15
• . . �e6 1 8 €) f5 c5 19 J,.e3 0--0 ( 1 9 . . .
J,.f5 l£l cd8 16 �e3 a5 1 7 j_c5 a4 'l!fa7 20 J;i c l J,. xf5 2 1 J,.xf5 cd 22
1 8 j_ xe7 'iiif xe7 1 9 � bd2 c6 20 b4 � d2 'it' b6 was unclear in De Fir­
�g5 the position became sharp. mian-Agzamov, Vrsac, 1 985,
I regard Nunn-Tal, Naestved, although White could have still
1985 as important. After some tried to gain the initiative by 23
transpositions Black played 15 . • . b4) 20 J;i c l g c8 21 de A xc5 22
J,.g6 (instead of 1 5 . . . l£l cd8). � xc5 � xc5 23 l£!d6 'i!li'c6 24 J,. b3
The game continued: 16 l£lfd4 0-0 d4 25 i!td2 d3 26 f4 l£l d4 27 .1!1 xc5
17 J,.g4! l£lcxd4 18 cd a5 19 f4 h5 -.xc5 28 * h2 l£j xb3 (28 . . . o.2)e2
20 J,. xh5 J,. xh5 21 'i!i!'f xh5 a4 22 29 K xe2 was also better for
<E)c5 l£l xc5 23 de J,. xc5 + 24 J,.e3 White) 29 ab f6 30 f5 and White
� xe3 + 25 X xe3 f5 26 e6 'f!le7 27 had an advantage (Koroljov­
� ael � adS 28 g4! d4 29 g5! g6 30 Biro, correspondence game,
'f!lxg6 + and White won after ten 1 986).
more moves. 17 J,.f5
13 �h5 1 7 l£l gf5 was probably better.
14 l£l fl � d8 17 .
. . �e6
15 l£lg3 18 j_g4
1 5 j_ e3 is not dangerous for The intention of this move was
Black. e.g. 1 5 . . . o.2)e6 1 6 g4 j_g6 to advance the f pawn, but this
1 7 � f5 0--0 1 8 o.2)g3 l£j a5 1 9 h4 will not be possible.
 The Spanish Game 29

18 <21cxd4 "{tg3 g5 + 44 fg )!I xh3 + 45 �xh3

19 cd c5 'i!i:i' xg5 mate.
20 <21 f5 �a7
21 <21 xe7 + 'fl!xe7 Popovic-Timman
22 ..1e3 cd Sarajevo, 1984
23 �xd4 li eS
Black has already gained the I e4 e5
initiative and Korchnoi is known 2 <210 <21c6
for his winning technique in this 3 �b5 a6
sort of position. 4 ..1a4 <21 f6
24 �d2 5 H <21 xe4
25 'fl!e3 6 d4 b5
26 j_ xe6 fe 7 ..1b3 d5
27 fJ 8 de �e6
An inappropriate move. Better 9 c3 �c5
was 27 a3 il!f c4 28 b3 ""c7 29 10 �bd2
� e2. 1 0 'l!ltd3 is also played, for in­
27 .M fc8 stance by Sokolov against Yusu­
28 l!l adl h6 pov in the World Championship
29 a3 'lV!e7 Candidates Final (Riga, 1 986),
30 l;l cl but 10 <21 bd2 is more logical.
More stubborn was 30 �c3 10 H
.1. 8xc3 31 be 'l!ltxa3 32 il;b6 � f5 II ..1c2 ..1f5
33 g4 "fl!e7 34 ll!r'e3 '1!lt h4 35 l!l e2 In Marovic-Yusupov (Tunis,
"fll g3 + when the result would still 1 985) Black played the well
be in doubt. known exchange- 1 1 . . . .E:J xf2 1 2
30 "fl!h4 � xf2 f6 1 3 ef ..1xf2 + 1 4 * xf2
31 I. xc2 ,M xc2 i!f' xf6 1 5 .a n -2Je5. Here White
32 gn �f5 played 1 6 * g l and after 1 6 . . .
33 Ji fl .M c4 -2Jxf3 + 1 7 gf 'i!Jll( xf3 1 8 'i!l!fxf3
34 f4 'iii" h5 ,g xf3 Black restored material
35 * h2 equality and gradually took over
This is the decisive mistake. 35 the initiative. The right move was
g d2 was necessary. 16 ..1e3! coolly leaving White's
35 'fl!dl king in the pin. In Tseshkovsky­
36 l!!l d2 '*'bl Chekhov (Rostok, 1 984) after 1 6
37 ..1c3 ,g e4 . . . � ae8 1 7 ..1 h4 iitf' h4 + 1 8 * g l
38 �fl � el -2J xf3 + 1 9 gf iitf' g5 + 20 .E:J g3
39 J;t e2 K hl + il_ h3 21 a4! g6 22 ab ab 23 ..1f2!
40 * g3 J,. xh3! White gained the initiative.
41 gh ii\Jg6 + 12 .ab3 (27)
42 *h4 "i!lff5 This is a critical position in this
White resigned
because he variation. Black has the choice
couldn't prevent mate, e.g. 43 between 1 2 . .Jl_ g4, which was
.
30 The Spanish Game
27
B
popular in the seventies, and 1 2
. . . .1g6 which became popular in
this decade. In Marovic-Murray
(Malta Olympiad, 1 984) the unex­
pected 1 2 . . . .1 xf2 + was played
(in the spirit of the 40s!) and after
1 3 l, xf2 � xf2 1 4 * xf2 .1 xc2 1 5
"Wtxc2 f6 1 6 e6 'ltd6 1 7 .1e3 'lt xe6
1 8 � bd4 � xd4 19 � xd4 White
gained an advantage.
12 .1g6
After 1 2 . . . .1g4 I used some
years ago to exchange on c5 auto­
matically. I remember that in this
way I achieved an advantage
against Belyavsky and Smyslov
(Leningrad, 1 977). However,
Korchnoi managed to get a good
game twice in the Baguio match
and there and then I switched to
1 3 h3. Since 1 3 . . . j_ xf3 14 gf
loses a piece, Black is compelled
to retreat by 1 3 . . . .t h5, and after
1 4 g4 .1 g6 1 5 .t xe4 de 1 6 <2l xc5 ef
1 7 .1 f4 White has a clear advan­
tage. Let's have a look at some
moves from Karpov-Korchnoi
(game 1 4, Baguio, 1 978): 1 7 . . .
'ltxd I (this is a forced exchange as
1 7 . . . 'f;lfe7 allows 1 8 *d5 � a5 1 9
b4 <2lc4 20 'lt x f3 <2l xe5 2 1 J_xe5
.-xe5 22 �d7!) 1 8 K axd l �d8
19 X d7 <2le6 20 � xe6 fe 21 A_e3
� ac8 22 g fd I .1e4 23 J.. c 5 K fe8
24 K 7d4 i.d5 25 b3 a5 26 flh2
K a8 27 *g3 and White's pos­
itional advantage was decisive. I t
seems that n o satisfactory impro­
vements have been found for
Black during the last decade.
13 <2lfd4
1 3 a4 has been played many
times e.g. 1 3 . . . ..1 b6 14 <2l bd4
<2l xd4 1 5 e xd4 (Van Der Wiei­
Korchnoi, Wijk aan Zee, 1 983)
and now 1 5 . . . *d7 (better than
1 5 . . . c5 1 6 <tlc6 <2l xc3 1 7 � xd8
<2l xd l 18 .txg6 l;l axd8 19 i_ f5!
with a dangerous initiative) 1 6
_te3 <2lc5 1 7 a5 .1 a7 1 8 f4 _txc2
1 9 <tl xc2 f6 20 ef J;� xf6 21 iit h5 c6
would have led to equality.
13 .1xd4
14 cd
Black would have had satisfac­
tory play after 1 4 -2l xd4 *d7 1 5
f4 e xd4 1 6 cd f6.
14 a5
15 J.. e3 a4
In an old Karpov-Savon game
(Moscow, 1 97 1 ) White achieved a
big advantage after 1 5 . . . <2l b4?
1 6 j_ b l a4 1 7 <2l d2 a3 1 8 * c l !
16 <2ld2
In the 6th game of my match
with Korchnoi (Merano, 1 98 1 ) I
retreated my knight and after 1 6
-E) c l a 3 1 7 b3 f6 1 8 ef 'i/irxf6 1 9
<2le2 <2l b4 20 JL b l 'fl/(e7 2 1 * e l
� fe8 2 2 � f4 j_f7 Black almost
equalised (although by playing 23
<2ld3! <2l xd3 24 .1 xd3 I could
have maintained slightly better
chances). Instead of 1 7 b3 White
has since tried 17 ba ( 1 7 . . . K xa3
1 8 j_ b3 �c3 19 'iir d2 b4, Ljuboje-

vic-Timman, Bugojno, 1 984) and
1 7 X b I ( 1 7 . . . f6 1 8 f3 fe 1 9 fe
,k xfl + 20 'ilit' Xfl ed 2 1 �d2,
Unzicker-Korchnoi, Beer-Sheva,
1 984) but Black achieved equal
chances in both games.
16 a3
17 -E) xe4 ab
18 � b1.t xe4
When I commentated on my
game with Savon, I assessed the
position after 1 8 . . . de 1 9 � xb2
f?je7 20 X xb5 � xa2 2 1 -wt b l �aS
22 X c l as favourable for White.
However, later, in Ivanov-Yusu­
pov (47th USSR championship,
first league, 1 979) Black played 1 8
. . . .t xe4 and equalised after 1 9
l;!; xb2 'tltd7 20 k, xe4 de 2 1 .l!l xb5
f?j xd4 22 x es x fd8. 1 8 . . . .t xe4
has been played dozens of times
since then, and, perhaps, on the
whole, White has come off better.
In this game, White also artfully
realised his positional advantage.
It seems that this variation isn't
played at grandmaster level any
more.
19 K xb2 �d7
20 .tdJ (28)
28
B 22 X tb1
This move (instead of 20 J.. xe4)
was played by Hubner for the first
The Spanish Game 31
time against Korchnoi in 1 982.
The bishops were still
exchanged - 20 .1 xd3 21
. • •
� xd3-but where White had a
more appropriate pawn structure.
They played this position twice.
In the first game after 21 . . . l!ii, fb8
22 .!;!; tb1 b4 23 a3 ba 24 � xb8 +
f?j xb8 26 �xa3 'iW c6 27 'W{e7 �d7
28 jta3 a draw was agreed
(Hiibner-Korchnoi, Chicago,
/982) although, according to
Hubner, White could maintain
better chances by playing 27 g4 h6
28 f4 -E)d7 29 f5 -E) b6 30 i,f2. In
the second game after 21 . . . b4 22
.1d2 � fb8 23 � fb i i!J'g4 24 a3
� b6 25 h3 '*'c8 26 � c2 b3! 27
� xb3 .)! xb3 28 i!J' xb3 -2) xd4
Black won the exchange and
the game (Hubner-Korchnoi,
Lucerne 01 ., 1 982). But White
didn't have to give up material
and could have kept quite tang­
ible pressure by playing accu­
rately. This game, in particular,
proves this point of view . . .
20 .t xdJ
20 . . . b4 was bad owing to 2 1
.t b 5 X fb8 2 2 X xb4 ,g xa2 23
k, xc6.
21 t;-xdJ J;� fb8
22 X xb5 X xb5 23 t;-xb5 -E) xeS
24 -Wtb7 -Wtc6 25 �xc6 -E) xc6 26
X c l � a6 would have led to equa­
lity.
22 b4
23 Xcl
In Karpov-Yusupov ( Moscow,
1 983) the moves 23 h3 h6 were
inserted and only now did White
play 24 K c l . We shall look at this
32 The Spanish Game

game later. The game Popovic­
Yusupov, Sarajevo, 1984 ended in
a quick draw after 23 h3 � b6! 24
"Wtc2 � ab8 2S � cl � 8b7 26 tlJ'cS.
Draw. Perhaps 23 � c l was more
accurate.
for the pawn but it is only illusion.
38 f]g3! fg 39 f] xh4 gh 40 f4. 40
* xh3 �e6 + 41 tlr h2 t;- f5 threa­
tening 42 . . . '(lte4 was dangerous
for White but now the rook comes
to help the king. 40 -wte6 41
. • •

23 � b6 '(lthS! (not allowing 41 . . . h5) 41

24 h6 'f!/e7 + 42 f] xhJ '(ltf7 43 1!1 b2!
.�!. a7 ( 29)
• . .

2S (preventing 43 . . . g g3 + 44 * h4
� h3 + ) 43 t;-d7 + 44 fS. Black
• . •

29 resigned. It is interesting that at

w the closing ceremony of the USSR
championship Yusupov and my­
self were both awarded the prizes
for the most beautiful game of the
championship and it was the
game between ourselves because
any chess masterpiece is created
by two authors!
26 i;� cS! � aS!?
After 2S • . . � ab8 the position Black sacrifices a pawn to acti-
from Karpov-Yusupov, Moscow vate his knight.
1983, would occur. Although this 27 � xb4 �c4
game might be known to the 28 i;� b3 K ab7
reader it is worth looking at the 29 *h2 c6 (30)
whole version.
26 � cS �d8 27 � cc2 �c6. 27

�-
. . . -E) e6 allows 28 f4 with a pawn
attack; and so Black retreats the
knight. 28 �cl � 8b7 29 cS
J?:Je7 30 *h2 �fS. Black gives p
a pawn to complicate the ga e;
he could have tried 30 . . g b5 or
.

30 . . . c6 with a defensible posi­

tion. 31 � bc2 � g6 32 � xc7
� xc7 33 � xc7 iitbS 34 g4! �h4
3S � c8 + *h7 36 '(ltdl 'CIJ'a6 37
� c2. White intended to under­
take the manoeuvre Jt.e3 - f4-g3
and if 37 . . . 'Wt'a3 38 '!\li'e2 b3 39 ab
"Wtxb3 White's position would be
strengthened by 40 � c7. 37 . . . fS.
It seems that Black has good play
It appears that Black is com­
pletely safe because the hostile c5
rook is in effect out of play. How­
ever, White has a witty trick here,
which brings the rook back to life.
30 .!;l aS!!
 The Spanish Game 33

Now White is simply a pawn many years in both systems. Inno­

up, which is the passed a pawn. vative ideas and sharp continua­
30 '!ll1 c8 tions are discovered from time to
31 s a4 �b8 time but the overall statistics of
3 1 . . . i;;. xb3 32 ab l:i1b8 33 \!!( a2 victories and defeats implacably
offered Black no joy. grow in White's favour. That is
32 � xb6 � xb6 why "solid" chess players, as a
33 'i!lf'f5 Wi' b7 rule, choose more sound struc­
34 11.. c l � bS tures when playing Black. But, of
3S 'll(f g4 * h7 course, those who favour sharp
36 .1_ a3 *g8 positions don't take into consider­
37 .1cS -E)d2 ation these general points of view.
A decisive time trouble mis­ l e4 eS
take. 37 . . . -E) xe5 was a better 2 .a n -E)c6
chance although after 38 'i!lf' f5 3 � bS a6
'iltb8 39 g3 and 40 A a7 the a 4 .1 a4 .ar6
pawn's advance would have been S 0-0 Jie7
almost impossible to prevent. 6 � el bS
38 g a7 'i�t'b8 7 j_b3 0-0
39 'ilt'f4 8 c3 dS
39 '\!ifd7 would have finished the 9 ed � xd5
game. 10 �xeS .a xeS
39 f6 l1 K xeS c6 (31)
40 '1!16' g4 The ancient 9 . . . e4 (instead of
White could have taken the 9 . . . -E)xd5) or I I . . -E) f6 (in­
.

knight, but he wanted more. stead of I I . . . c6) hardly ever

40 � b7 occur in serious tournaments
41 .�;t xb7 'ii!i' xb7 these days. It has been an uncon­
42 ef -E)e4 testable fact for many years that
43 '*'xg7 + Black doesn't get any compensa­
Black resigned. tion for his pawn.

Georgiev-Tseshkovsky 31
Minsk, 1985 w

We shall cover the modern state

of the Marshall Gambit in the
commentaries on the next few
games. 1, personally, associate
this variation of the Spanish with
the Dragon Variation of the Sici­
lian Defence. Similar theoretical
research has been going on for 12 d4
34 The Spanish Game
Sometimes the more restrained
2 1 d3 is played and this is covered
later.
The exchange 1 2 j_ xd5 has
gone out of fashion in the last few
years, but the amusing 12 'iiei' fl !?
was chosen in Kapengut-Mala­
nyuk, USSR, 1985. White draws
his queen nearer to the king in
advance, anticipating Black's
offensive on the kingside. In fact,
this plan looks rather artificial.
After 12 . . . j_d6 13 g et 'iiei' h4 14
g3 �h5 15 d4 � g4 ( 1 5 . . . .1h3 is
also good). While 1 6 ii_e3 � ae8
1 7 -E}d2 � e6 1 8 c4 � xe3 1 9 � xe3
g h6 20 i!l'r g2 c5 leads to a level
position, Kapengut tried a new
move, 16 -E}d2, developing the
knight immediately. The game
continued: 16 • • • � ae8 17 f3 j_ h3
(The pawn recapture is favour­
able for White in this situation­
the tension is relieved and he is
left with the advantage of the two
bishops.) 18 "lll!r' fl f5 19
g xe8
� xe8 20 c4 -E}f4! 21 c5 + (It is
difficult to assess the position
after 21 gf l!l!tg6 + 22 -.g3 .k e l +
23 \lf f2 "lll!r' e8 24 -E}e4 g fl + 25
*e2 fe 26 cb + 'IPffi 27 "lll!r' xM ef +
28 "*d3 "l)ife2 + 29 frc3.) 21 . . .
frf8 22 cd? The losing move: cor­
rect was 22 gf! e.g. 22 . . . 'iit g6 +
23 "1il'g3 g, e l + 24 *f2 "lil'e8 25
-E:Je4 g fl + 26 '1ie2 fe 27 'iiri' xh3
d+ � ·� 'i!i!tcl + � ·�
'ii' e l + 30 j_d2 b4 + 3 1 'ili c4
iit'e2 + 32 'i1P xb4 "i!J!f' xd2 + 33 *a3
j_ xc5 + 34 de llfa5 + 35 j_ a4
iit'XC5 + 36 b4 'i!ri'C3 + 37 j_ b3
� xa 1 38 -wr-c8 + • e7 39 Y�tc7 +
*e8 with perpetual check . 22 . . .
g e2 23 d7 '1Pe7 24 j_c2 � xfl 25
* xfl �e6 26 <2lb3 JL. g4! and
White resigned in ten moves.
Before we concentrate on 1 2 d4
in more detail, I want to mention
another recent game where
Fischer's 1 2 g3 was played.
Braga-Geller (Amsterdam,
1 986): 12 . . . ii_ f6. (White's d
pawn has stayed on its original
square and so Black can organise
pressure on the d4 square.) 1 3
� e l c5 1 4 d4 j_ b7. (An innova­
tion. After 14 . . . cd 1 5 cd ii_ b7 1 6
-E}c3 >21xc3 1 7 be White is slightly
better; now, when the c3 square is
occupied, White is behind in deve­
lopment.) 1 5 de � e8 1 6 �d2
-E} xc3! 1 7 be j_ xc3 18 c6 (Geller
gave variations which show that
after 1 8 � b l Y�td7 Black has the
advantage.) 1 8 . . . j_ xc6 1 9
� xeS + "lll!r' xe8 20 � b 1 .k d 8 2 1
'(i!i'c2. If 2 1 'i!IJ h5 - 2 1 � b2 a5!­
there is a pretty variation: 2 1 . . .
g6 22 Y�tg5 Y�te l + 23 � fl 'i!iJ'e4 24
\i(fxd8 + '1J g7 25 f3 .- xf3 26 -E}e3
Y�t h l + 27 'ilif2 llfel mate. 21 . . .
j_ xd2 22 ii_ xd2 j_e4 23 j_ xf7 +
*xf7 24 'iiri' b3 + j_d5. Black had
an endgame advantage which he
easily converted into a win.
12 iLd6
13 � el
Perhaps White could present
Black with more problems by 1 3
g e2, but we will cover this later.
13 'iiri' h4
14 g3 "iWfh3 (32)
15», e4!?
The well known continuations
1 5 j_ xd5 and 1 5 'iiri' d 3 are not
dangerous for Black and are not

32
w tage for White (Ramirez-Velas­
played these days. The classical
position occurs after 1 5 �e3 �g4
1 6 �d3 ll ae8 (the exchange 1 6
. . . �xe3, suggested by Geller as
far back as 1 965, is not sufficient
for equality: 1 7 � xe3 c5 1 8 'iii' fl
� h6 19 �d2 � adS 20 �f3 �xf3
21 g xf3 cd 22 cd *d2 23 lit d3!
and White had better chances,
Fischer-Donner, Santa Monica,
1 986) 1 7 �d2 � e6 1 8 a4 (it was
established in the 70s that after 1 8
c4 � f4! Black gains splendid
chances) 1 8 . . . f5 1 9 'ii!f' f l 'tlifh5 20
f4 ba 2 1 g, xa4. From this very
position, which has become stan­
dard, we will start the commen­
taries on the next game . . .
Before going any further, I
want to mention one recent game
after 15 j_e3 �g4 16 ._d3 It ae8
17 €)d2 f5 (instead of 1 7 . . . X e6).
To all appearances, the pawn ad­
vance also gives Black sufficient
counterplay. Let's see.
Ulmanis-Van Der Haiden, cor­
respondence tournament, 1 986: 18
f4 tlr h8 19 �xd5 cd 20 'Wt'fl 'i/lf'h5
21 a4 ba! ECO considers the main
reply to be 21 . . . g5 e.g. 22 ab ab
23 � a6 gf 24 � xf4 �xf4 25 l!l xe8
� xeS 26 "i!!i' xf4 X e l + 27 �fl
�e8 28 � b6 JLh3 29 g b8
The Spanish Game 35
». xfl + 30 * xfl with an advan­
quez, Niceol, 1 974). The rook's
deflection from the first rank
changes matters. 22 I! xa4 g5 23
fg? White takes an unjustified
risk. In Schlosser-Nunn (Krefeld,
1 986) after the rook's retreat 23
J;t aa 1 a5 24 fg f4 25 � xf4 a draw
was agreed. 23 . . . g xe3! 24
� xe3 f4 25 gf � xf4 26 g g3 -.e8!
(Black exploits the rook's delay
on the a4 square. Certainly there
was no point in taking the queen
immediately: 26 . . . �e3 + 27
g xe3 � xfl 28 �xfl ) 27 g xg4
JLe3 + 28 *g2 g xn 29 � xn
� cl! 30 X a5 '111f e2 + 31 *g3
'ilf'xfl and Black won.
15 g5
16 -.n
Better than 16 -.n � f5 ( 1 6 . . .
f5 1 7 g e5!) 1 7 jj_c2 �e4 1 8 _t xe4
�e6 1 9 � xg5 ( 1 9 � f5 tt e 1 + 20
ftg2 'i!iJ'xc l 2 1 � a3 *d2! 22 J;t d l
�e3 + ) 1 9 . . . f5 20 � d3 h6 2 1
.1h4 g a7! 22 �d2 J� g7 23 * fl
f4 with the initiative for Black.
16 'Wr'h6
This was a novelty but it was
doomed to failure. The exchange
of queens would have given Black
a good game.
17 f3 *h8
18 €)d2 �h3
19 \'!l(fe1 �f4
20 �n
After 20 g f gf 2 1 * h i g g8 22
11 e2 *g7 23 *f2 j_e7 the threat
of . . . ..t h4 was unpleasant.
20 11t g7
21 'iir fl �d3
22 '*d2 � xcl
36 The Spanish Game

23 � xcl fS 41 �d7! J,. xd7

24 !! e6 (33) 42 �xd7 + *e6
43 -E:�c5 + *d5
33 44 �e4! h6
B 45 � g3 *c4
46 �f5 b3
47 a3 a4
48 *e4 �c7
49 �e3 + *b5
50 � d5 �d6
51 -E:Jc3 + w a5
52 f; d5 ..U8
53 *c6 j_g7
The tension has gone and 54 -E)b5! * a6
White has the initiative. 55 *c5 j_f8 +
24 . . . K ad8 56 -E:Jd6
Not 24 . . . i_ f4 25 gf gf + 26 Black resigned.
�g3 fg 27 K h6 gh + 28 * xh2
J.g2 29 !! gl f4 30 K h3!
25 K eel c5 Prandstetter-Blatny
26 '(itfl cd Czechoslovak championship, 1986
27 i/lf'xd4 �xd4
28 cd *g7 1 e4 e5
29 !! cl .!He8 l -El f3 .ac6
30 * fl f4 3 J.b5 a6
31 � xe8 K xe8 4 J. a4 -El f6
32 -E)dl! 5 0-1) �e7
Black would have had an advan- 6 � el b5
tage after 32 � c6? _a b4, but now 7 J.b3 0-0
White's knight is heading for the 8 c3 d5
centre and Black gets the worse 9 ed -El xd5
position. 10 � xe5 �xeS
32 �rs 11 K xe5 c6
33 �d� � c8 12 d4 �d6
34 K xc8 fg + 13 g el i/ll' h4
35 hg j_xc8 14 g3 i!t h3
36 -E) e4 j_ e7 15 _ae3 _a g4
37 g4 aS 16 iit d3 � ae8
38 J.c6 b4 17 �dl � e6
39 •e3 �d8? 18 a4 f5
40 -2) c5 * f6? 19 iit fl iit h5
Black made several mistakes in 20 f4 ba
a row because of his time trouble. 21 K xa4 (34)

34
B kevich suggests 22 . . . � e4! 23
That is how far the research of
the Marshall Gambit has moved:
it is possible to play more than
twenty moves without blinking
. . . White sometimes doesn't
hurry with the capture on a4 and
plays 21 j_ xdS cd 22 .-g2 instead.
Here is one recent example.
Tseshkovsky-Agapov, Kiev,
1984: 22 � fe8 23 'ilf xd5 *h8
. . •
24 j_ f2 � e2. The other moves
which have been played here are
25 * g2 or 25 g xa4 .i_ xf4! 26 gf
.i.c4 27 g xc4 (27 i/txc4 'ilfg4 + 28
* h l g xe l + with equality) 27 . . .
.�;t xe l + 28 j_xe l � xe l + 29 � fl
� xfl + ! 30 * xfl ._d l + with a
draw. White plays a new move,
which turns out badly. 25 �c4.
Putting the knight on e5 looks
attractive, but weakening the f3
square is very important. 25 . . •
§J.. c7. 25 . . . � xf4 26 g xe2 g xe2
27 gf K xf2 28 * xf2 "i!l!(xh2 + 29
* fl a3! was also good for Black.
26 �e5 ll. xe5 27 K xe2. The game
would end dramatically after 27
de .i.f3 28 iirc4 'ii!f x h2 + ! 27 . . .
'i!ii' xe2 28 de '\!lfxb2 29 g xa4 'ilfxc3
30 K c4 ili'a1 + 31 * g2 h6 32 '\!lfdJ
'\!lfa2 33 � c7 K b6 34 K c2 K b2 35
� xb2 •xb2 36 'i!l({xa6 \'!llf' c2 and
Black soon won.
The Spanish Game 37
Instead of 22 . . . g fe8 Matzu­
g xa4 (23 -E:� xe4 fe and . . . g7-g5)
23 . . . g5 24 g xa6 gf 25 -E:� xe4 fe4
26 g xd6 fe 27 g xe3 §J.. h3 28 g4
'ilfh4 29 *xh3 'i!i' f2 + 30 * h l
K a8 3 1 g d8 + � xd8 3 2 g5 g aS
33 )ie6 + * g7 with a draw.
21 g b8
This is Lilienthal's improve­
ment on 2 1 . . . IHe8 22 iltf2 *h8
23 � xd5 cd 24 c4! or 21 . . . g5 22
� xa6 *h8 23 g xc6! -E:� xe3 24
•f2 with a big advantage for
White in both cases.
22 .i_ xd5 cd (35)
White now played the novelty
23 � xa6. But before going any
further, it is useful to consider the
well known continuation 23 'iltg2
.-es 24 'iltxdS * h8 25 *fl. Some
books consider here 25 . • •g5 26
J� xa6 ;g xb2 27 K a2 gf 28 gf
g xa2 29 'iltxa2 §J.. xf4 30 � fl � e4
with good play for Black.
ECO suggests 27 � a8, which is
refuted by 27 . . . il. b8! when
Black avoids the exchange of
rooks, eliminates the pin along
the sixth rank and maintains all
his threats. However, White has
38 The Spanish Game
the more logical continuation 27
� a8! g b8 28 ii;-c6! Now, after 28
- basis of the continuation 30 . . .
. . . gf, the forced continuation 29
'lib'xe8 + 1;1 bxe8 30 gf JJ.. e7 3 1
l;l xe6 k, h4 + 3 2 *g2 !t xe6 3 3 d5
j_xe l 34 de j_ xd2 35 j_ xd2 leads
to a position with bishops of op­
posite colour, but with two extra
pawns for White. No better is 28
. . . �h5 29 h3! 'W;'xh3 30 'l!li'g2 gf
31 l!lrf xh3 il_xh3 32 gf l;l g8 33 *e2
li ge8 34 *fl � g6 35 g h l and
White won, Popolitov-Trushakov,
USSR, correspondence tourna­
ment, 1980-1981.
So, is the position in diagram
3 1 favourable for White after 23
�g2!? No, this is not the case.
The opening books simply over­
looked that after 23 . . . 'i!lte8 24
�xd5 wh8 25 * fl Black can play
25 g xb2 (instead of 25 . . . g5).
. . .
Play can then continue: 26 g, a2
� xa2 27 'i!ltxa2 g5 28 d5! (more
promising than 28 <E:lc4 gf 29
-2} xd6 fe + 30 ,!:! xe3 ....,. h5 31
'$ xe6 �xh2 + 32 * " ilifbl +
draw agreed, Chiburdanidze­
Tseshkovsky, Tashkent, 1980) 28
. � xe3 29
. . � xe3 �c5 30 -E:� fl
(36) .
Until recently, this position was
assessed as better for White on the
jl;'e4 3 1 'll(( d 2 gf 32 gf ..- xf4 + 33
w g2 �. xe3 34 �xe3 �xe3 35
.:t.} xe3 etc. However, Black
shouldn't waste a tempo on 30 . . .
'W;'e4 allowing his opponent time
to stabilise his position, but
should play 30 . • .k,h3! (sug­
gested by G. Fridshtein) which
presents White with serious prob­
lems. It seems that White then has
nothing better than 3 1 )llli' xa6
J., xfl 32 ..- f6 + forcing perpetual
check.
So, Lilienthal's idea of 21 . . .
.!:! b8 leads after 22 � xd5 cd 23
\'!lfg2 to equality which, from a
theoretical point of view, is very
satisfactory for Black. But
White's attempt in the present
game to improve by 23 g xa6
turns out to be unsuccessful.
23 g xa6 li be8!
It was bad to take the pawn
now: 23 . . l!l xb2 24 1ir'g2 ..-es 25
.
lif' xd5 fih8 26 .:t.}c4!
24 lif'fl
Black would also have had the
better chances after 24 ..-g2 g xe3
25 g xc3 .!:! xe3 26 t;- xd5 + t;-f7
27 *xf7 + * xf7 28 !;l xd6 g e l +
29 <21 fl �h3 30 'ilff2 K xfl + 3 1
*e3.
24 g5
25 .!:! xd6! ,l xd6
26 fg .!:! de6
27 h4? (37)
This move allowed Black to
start a decisive offensive. The
right continuation was 27 j!;' f4
J.,h3 28 ilif2 l;l e4 29 � xe4 fe 30
* g l g f8 3 1 g4 � xg4 32 il{(e5
 The Spanish Game 39

*n 33 \'!ltg3 with a probable 9 ed -dxd5

draw. 10 -E) xeS -dxe5
ll � xeS c6
37
12 d3
B This modest d pawn advance
(White strengthens the e4 square
and, if the opportunity arises,
leaves the d4 square free for a
rook) is unlikely to be better than
the standard 12 d4. This ancient
move competes with the two
square pawn advance only
because of fashion.
27 f4! 12 j_d6
28 gf 13 � e1 (38)
Black also retains the advan­
tage after 28 1/l'txf4 ;;,;. f8 29 i/itc7 38
l!fe8. B
28 �h3!
29 �xn
30 j'(g4
31 � xel +
32 .i, xe1 'Wt'h3 +
33 'ili'g1 'ii!t' e6
34 *" '@!i' e4
35 �d2 'i!it f3 +
36 �fl � e4! 13 �h4
37g6 h5 The main continuation accord­
38 c4 � xf4 ing to theory, is 13 JJ_f5 1 4
. . .

39 'ii!i'e 1 )ff h3 + JJ:.xd5 ( 1 4 j'( f3 � e8 1 5 � xe8+

White resigned. j'(xe8 1 6 j_d2 -d f4 1 7 iL_c2 ;;;; d8
and Black was better, Whittaker­
lvanov-Agapov Harding, correspondence tourna­
Kiev, 1984 ment, 1 977) 1 4 . . . cd 1 5 -2J d2 ( 1 5
'i!lf f3 doesn't achieve anything for
1 e4 e5 White either: 1 5 . . . � e8 1 6 iL.d2
2 �f3 -E}c6 -.d7 17 h3 .ft.c7 18 -2Ja3 jj- d6 1 9
3 �b5 a6 g3 � h3, Zaitsev-Adorjan, Polia­
4 J.. a4 -E}f6 nica-Zdroj, 1 970) 1 5 . . . ii.. xd3 1 6
5 0-0 JJ:.e7 � fl ,ig6 1 7 j'(d4 � c8 1 8 .i!_ e3 b4
6 � e1 b5 1 9 � ad ! be 20 be it_ b8 2 1 'il'i' xd5
7 .i_b3 0-0 'il'i' xd5 22 � xd5 � xc3 with an
8 c3 d5 excellent game for Black.
40 The Spanish Game

More interesting is the imme­

39
diate I4 <E:�d2 and after I4 • . .
w
-E:� f4! IS <2}e4 ( 1 5 d4 -E) xg2!) IS • . •

<2} xd3 I6 � gS '$'d7 1 7 � xd6

'ilir'xd6 1 8 �c2 the position is
equal. Here is one recent example,
Georgiev-Nunn Dubai 0/., 1986:
I7 � eJ � xe4 IS X xe4 K ae8 I9
'ilir'g4 '$'xg4 20 � xg4 �eS 2I K bi
h5 22 1;1 h4 <2l xb2 23 �eJ �dJ 24
K d i �b2 25 K bi �dJ. Draw.
As a rule, Black has refused to Timman-Hiibner (Tilburg,
play 1 3 . . . � f5 in recent games. 1 985): 1 7 <2J fl (This move was
Possibly he intends to exploit played for the first time in this
White's unambitious play and to game.) 1 7 . . . f5. (Better than 1 7
make a serious attempt to gain the . . . � f5 1 8 il_xd5 cd 19 .�;i; d4-the
initiative . . . d4 square has turned out to be
In Popovic-Velimirovic (Bor, useful- 1 9 . . . il_c5 - 1 9 . . . iJ.. e6
1 985) Black played 1 3 . . . K a7!? 20 �e3 il_c5 21 �xd5 and if 2 1
After 1 4 '2)d2 � f4 1 5 �e4 � xd3 . . . �xd4 there is the Zwischen-
1 6 il.g5! �e7 ( 1 6 . . . 'i1ir'd7 1 7 zug 22 <2l f4-20 � xd5 _ig4 21
� e3) 1 7 �xe7 x xe7 1 8 � e3 ( 1 8 .d2 �c6 22 "Wt'g5 ii. h3 2 3 <E)e3
� f6 + \li h8!) 1 8 . . . � d7 'i1ir'h5! il_xe3 24 �xe3 h6 25 'llf' h 5 g6 26
White launched a winning attack. l{te5 and White has a clear advan-
I4 gJ 'ilir'hJ tage. Hubner recommended 1 7 . . .
This position has also occurred � f6 1 8 � e l �g4 1 9 f3 il.f5 20 d4
after a different order of moves, c5 2 1 cd A,xc5 + 22 .ie3 � adS 23
for instance: 1 2 g3 �d6 1 3 K e l .-e2. But even here White's
i!!t d7 14 d3 trh3. chances are better. Another possi-
IS 1. e4 'ltrf5 bility is 1 7 . . . h5!? 1 8 a4 1l_g4 1 9
I 6 �d2 't)l;'e l � f6 20 � e3 � ae8 21 ab
If 16 K h4 �b7 1 7 �c2 .-e6 18 ab 22 d4 h4 with sharp play,
c4 � b4 1 9 c5 � xc2 20 "t�t xc2 f5! Kuzmin-Shulman, Spartakiad,
or 1 6 �c2 "Wtg6 1 7 'ilir' fl f5 1 8 K e l USSR, 1 986). 1 8 � d4. (White
f4 Black has a clear advantage. again takes advantage of the fact
16 • • . 'ilir'g6 ( 39) that the d4 square is vacant, but it
Not 1 6 . . . '2) f6 1 7 X e 1 'ilir'xd3 \ was the only move as well, as after
1 8 �e4! Tht: position in the dia- 1 1 8 � e l f4 1 9 c4 fg 20 fg j_g4
gram has become a sort of tabia 1 Black would have had an ad van­
for the 1 2 d3 variation. White has tage.) 1 8 . . . f4 (if 1 8 . . . .i b7 or
a choice between 1 7 � fl and 1 7 1 8 . . . .ie6 then 1 9 c4) 1 9 g xd5
� e l with the aim of putting the cd 20 jj_xd5 + �e6 21 �xa8
knight on e4. J;l xa8 22 Wr'f3 .I;!; ffi 23 "Wte4 � f5 24

'i!td5 + * h8 25 a4 b4? (Hiibner
gives 25 . . . � xd3 26 ab �e4 27
'i!td4 fg 28 �xg3 �xg3 29 fg ab
30 J.f4 h5 3 1 ttd6 which led to
equality) 26 cb fg 27 hg � xb4 28
�d2 � xd2 29 �xd2 J.xd3 30
K a3 and Black didn't have any
compensation for the pawn in the
ending, though Hubner managed
to save the game thanks to inac­
curate play by his opponent.
Now we return to the main
game which illustrates the second
option, although it is unlikely to
be the best.
17 J� el f5
Malanyuk played another in­
novation in the Marshall Gambit
against Smagin (Kiev, 1 986) but it
did not turn out well: 17 . . . J.c7.
After 18 �f3! �g4 19 �h4 'i!th5
20 f3 �h3 2 I lli'e2 I;� adS (more
accurate was 2 1 . . . * h8 with the
intention of advancing the f pawn
although Black's activity
wouldn't quite compensate for the
absence of a pawn in this case
either) 22 d4 f5 23 f4 and Black
couldn't find anything better than
the exchange of queens by 23 . . .
ttxe2 24 g xe2 g fe8, and after 25
II xeS + g xeS 26 *f2 J.g4 27
.1,e3! and the rest, as they say, is a
matter of technique.
1 7 . . .1. b7 hasn't justified
. game finished as follows: 23 J.d2
itself, after 1 8 �f3! � feB (instead
of 1 8 . . . ,ig4) 1 9 � xeS + g xe8
20 �h4 'i!IJ'f6 21 j_d2 b4 22 ilrf3
ilrxf3 23 �xf3 White's chances
were better (Hjartarson-Hebden,
London, 1 986).
18 e4
After 1 8 �e4 fe 19 de 11.. g4 20
The Spanish Game 41
'i!td4 lli'h5 21 ed c5! Black has an
advantage.
18 f4
19 �e4 fg
20 fg �g4
21 'i!tcl be
22 de
22 'i!txc4 was worth consider­
ing, but now Black develops a
decisive attack.
22 J;t ae8!
(40)
Two years later in Smagin-­
Hebden, Moscow, 1986 Black
didn't risk giving up the piece but
played 22 . . ..tb4. Bearing in
.
mind that Black won quickly and
efficiently after 22 . . . � ae8 in the
main game, it remains a puzzle
why White repeated the position
allowing 22 . . . 11 ae8 and Black
rejected this opportunity. The
�f4 24 .1,xf4 J.xe1 25 .!!l xel J.f5
26 'ii(( g2 � fe8. 26 . . . � ae8 was
more stubborn . . 27 .tel � e6
28 g4! J.xg4 29 :w, n J.e2 (or 29
.
. . . .1,h3 30 'l!lrxg6 � xg6 + 3 1
�g3 .1, xfl 3 2 .1,xg6 33 .1, b l ) 30
�g5! it,xfl 31 .1, xg6 k xg2 32
J.n + • m n J.xe6 •� M
42 The Spanish Game

'itr xg2 h6 35 j_g4 36 hg j_ xg5 + 12 d4 ..id6

�d6 37 b4. Black resigned. 13 � e2 (41)
23 cd il_ f3!
It becomes clear that Black 41
wins the piece back while main­ B
taining all the advantages of his
position.
24 de +
24 it_ f4 .2_xf4 25 'i!J!r'xc6 � h5 26
d6 + �h8 27 d7 iWh3! (27 . . .
;;;; xe4 28 'i!ii' c 8!) 28 \!if'C2 \!if' xd7 29
gf ;;;:; xe4! was equally hopeless for
White.
24 It has already been pointed out
25 kd2 � xe4 that the rook retreat to e2 is tak­
26 � xe4 j_ xe4 ing over in popularity from the
27 'ii.f c3 '1Wf5. traditional 1 3 g e 1 and is almost
White lost on time, but he could considered to be a refutation of
just as well have resigned. the Marshall Gambit. At the same
In the annotations to this game time, only a few years ago, the
the reader can find a number of opening books gave just two lines
games played very recently. How­ to this move. The aim of the
ever, I chose Ivanov-Agapov rook's manoeuvre is clear­
which was played back in 1 984. White, if the opportunity arises,
As you can see, the importance of will save a tempo in transferring
this game hasn't diminished since his queen to fl (in one go: i!td 1 -
then. fl , not in two: 'i1i' d l -d3-fl ). We
shall cover the modern state of the
Belyavsky-Malanyuk variation that starts with 1 3 lit e2
Minsk. 1987 (54th USSR in the notes to this game.
championship) 13 �h4
13
• • . j_g4 has also been
1 e4 e5 played, with the continuation 14
2 �fJ � c6 f3 ..i h5 1 5 j_xd5 (15 �d2 .t:J f4 16
3 j_ b5 a6 g.fl �dJ 17 g, e2 �f4. Draw.
4 j_ a4 -2!f6 Ady-Littlewood, London, 1983)
5 0-0 j_e7 1 5 . . . cd 1 6 .t:J d2 �c7 (or 1 6 . . .
6 � e1 b5 '1\!lt'h4 1 7 .t:J fl � ae8 1 8 ..ie3, and
7 j_ b3 0-0 Black's initiative runs out of
8 c3 d5 steam) 1 7 -2! fl g, fe8 1 8 ..ie3 tlll' c4
9 ed -2! xd5 1 9 a4. This position occurred in
10 .t:J xe5 -2:1 xe5 Van Der Sterren-Pein (Brussels,
11 � xe5 c6 1 984). After 1 9 . . . j_g6 Black has

retained some compensation for
the pawn, but White should be
content, because nothing threa­
tens his kingside.
14 g3 '*hS
In connection with 14 • . *h3
. 'ltxf3 I S i!tfl 'Wtg4 19 i.d2 K feS
we should look at Sokolov-Kha­
lifman (Sochi, 1 9S2): 15 -2Jd2 ( 1 5
'i!!t f l 'i!!t h5 1 6 f3 i.h3 1 7 "i!r f2 f5 l S
<2!d2 ;g aeS 1 9 .Iii xeS i! xeS 20
-'2l fl led to an unclear position) 15
• . 1l_f5 16 Jt.c2 j_ xc2 17 \\lj'xc2 f5
. 1 9 . . . � aeS 20 i.d2 �f6 2 1 M, e l
1 S f4 �g4 1 9 � fl .txf4 20 K f2
i.d6 2 1 c4 be 22 ll{xc4 f4 23 li!(xc6
"ilfe6 24 �xf4 and White had an
advantage.
18 c4 has also been played in­
stead of 18 f4. Ljubojevic-Nunn,
Szirak, 1987 ended in a crushing
defeat: 18 . . . 'fii' g4 19 � e6 �f4!?
20 � xd6? � ae8 21 cb � e2! 22
·ttt c4 + 'lrh8 23 'lll't' xe2 (23 . . .
-2)h3 + was threatened, and if
23 � f7 then 23 . . . � el + 24 � fl
� xfl + is decisive) 23 . . . �xe2 +
24 \lig2 f4! 25 be fg 26 hg �f4 +
White resigned. In Hubner-Tim­
man, Tilburg, 1987, White played
20 f3, and after 20 . . . �h3 + 21
iltg2 �f4 + 22 fjtjl g1 �h3 + 23
*g2 �f4 + the game was drawn
by repetition. The attempt to
avoid this by 22!1f h 1 is danger­
ous: 22 . . . 'iii' h3 23gf j_xf4 24
� e2 � ae8! However, 19 f4!?
intending to open up the kingside
was worth considering.
15 �d2
Better than 1 5 � e4 ti' g6 1 6
j_ c2 f5 (or 1 6 . . . 1l. f5 1 7 ;g e2
». ae8 18 i. e3 1L xc2 19 'iit xc2
'l\ii xc2 20 g xc2 <2Jxe3) 1 7 ». e2
i/fh5 1 8 �d2 f4 with a compli-
The Spanish Game 43
cated position (Eijmes-Roma­
nenko, correspondence tourna­
ment, 1 98 1 ).
15 it.h3
1 5 . . . i.g4 16 f3 i.xf3 17 �xf3
led to equality (Balashov-Tsesh­
kovsky, USSR championship,
1 980-- 1 9S l ). However, stronger
was I S K f2 'i!fe4 19 'l!l'ff3 ( 1 9 _ic2
'lte6 20 'i!lrd3 g6 is not quite clear)
and White had the better chances
(Sax-Nikolic, Plovdiv, 1 983). As
we know, White is interested in
simplifying the position and
maintaining the advantage of the
two bishops at the same time,
even at the cost of a pawn.
16 f3 (42)
42
B
In Psakhis-Geller, Sochi, 1982
White played 16 ).t el? If White
wanted to retreat to the first rank,
it was better to do so after the
preliminary 16 � e4 'i'ii' g6. 16 • • .
I!l ae8! Now Black's pressure is
growing before one's eyes. 17 f3
f5! 18 c4 <2J e3! (not hesitating to
sacrifice) 19 cS + !lfh8 20 � xe3
� xe3 21 cd f4! 22 �e4 (22 *f2
.i_g4! 24 gf � fe8) 2 2 . l;l xf3 23
. .
il.xf4 )l: 8xf4 24 d7 (24 <dd2 iite8!
44 The Spanish Game
25 �xf3 i(te3 + 26 * h l 'l!li'f2 27
-EJ h4 K xh4!) 24 . . . _i xd7 25 -EJd2
j_ h3! 26 -EJ xfJ � xfJ 27 'iij d2 � f8
28 '(11 e3 'i!!i' fS 29 'llt e2 cS 30 _ic2
'llt dS and Black won. 30 . . . 'ii!J' f6!
3 1 de "ii(( d4 + 32 * h 1 j_g4! would
have been quicker.
16 j_c7
Apart from 16 . . . j_c7, Black
has also played 1 6 . . . f5, 16 . . .
g ae8, and 1 6 . . . .!l ad8. Let's
look at recent examples of each of
these replies.
16 . . . f5 1 7 c4 ( 1 7 a4 .!l ae8 1 8
ab ab 1 9 .!l a6 was also not bad)
1 7 . . . be 1 8 -EJ xc4 i.c7 1 9 -EJe5
with an advantage for White
(Rodriguez-Malanyuk, Erevan,
1 984); 1 6 . . . .!l ae8 1 7 .!l xe8 � xe8
1 8 � e4 JJ... c7 1 9 j_d2 'i!IJ'g6 20 _ic2
and Black had no compensation
for a pawn (Garcia-Pinter,
Lucerne Olympiad 1 982); 1 6 . . .
J;� ad8 1 7 -2Je4 ( 1 7 "Wt e l intending
'i!itf2, -2J e4 and .id2 was also
playable) 1 7 . . . 'iilf x f3 1 8 -2Jg5
'i!li' h5 19 -2J xh3 'ii!J xh3 20 "tirf' fl �g4
2 1 .l bd2 .!l fe8 22 .!l ael .!l xe2 23
-w-xe2 'ii!J' xe2 24 .!l xe2 h6 25 * g2
* f8 26 'iti f3 and White's chances
were clearly better (Aseev­
Taborov, Kostroma, 1 985).
16 . . . JLc7 was played for the
first time in Kuporosov-Mala­
nyuk, USSR, 1985. Let's see what
happened. 17 a4 b4 18 c4 �f6 19
K el (better than 1 9 'l!lte l .!l ae8 20
'l!lt f2 .!l xe2 2 1 'i!it xe2 .!l e8 22 ,'i!ll- f2
'i!li' f5 with an initiative for Black)
19 • . . .!l ad8 20 �e4 o21 xe4 21
K xe4 Ji.. fS 22 .!l el ;ir h3 (not 22
. . . g fe8 j_e3 iih3 24 J.. f2 h5 25
II xe8 Iii xe8 26 "(i!;'d2 h4 27 Iii e 1 ,
but 2 2 . . . j_b6 2 3 j_e3 j_xd4 24
j_ xd4 c5 25 j_xc5 ,!l xd l 26 j_xd l
x d8 27 j_xb4 was possible) 23
,. e2 � fe8 24 iL g5 f6 25 iL e3 h5
(25 . . . j_xg3! 26 hg � xe3 27
� xe3 jfxg3 + 28 * fl 'i!i'h3 + led
to a draw) 26 c5 + *f8 27 kc4
it.. xg3 (27 . . . h4 28 "ifi'fl !) 28 hg
(now after 28 'ii( fl ii_xh2 + 29
� xh2 'i¥f xfl + 30 'lrxfl � xe3 3 1
w f2! � ee8 32 ,!l xh5 JJ... c8 33
� h8 + 'lre7 34 � e l + 'lrd7 35
� exeS � xe8 36 � xe8 wxe8 37
a5 White maintains the better
chances) 28 • . . � xe3 29 � xe3
'i!li'xg3 + 30 r;trfl �h3 + 31 'lrg1
jfg3 + 32 w fl �h3 + Draw.
In Short-Nunn (Brussels, 1 986)
White played the improvement 1 9
o21e4! (instead o f 1 9 .!l e i ) after
1 9 . . . ..-g6. (Nunn, commenting
on this game in lnformator, gives
variations which show that 19 . . .
-2) xe4 20 .!l xe4 and 'i!li'xf3 20 -2)g5
are favourable for White.) 20 -2) £2
J.. f5 21 .,ic2 .!l fe8 2 2 .§lxf5 jl{xf5
23 .!l xe8 + � xe8 24 * g2. Black
had no compensation for a pawn.
Short, though, made several mis­
takes at the end of this game and
lost.
17 �e4!
This move was played a bit
later in the games covered before.
But, possibly, the immediate cen­
tralisation of White's knight is the
most unpleasant for Black. In any
case, Malanyuk, although an
expert on the Marshall Gambit,
managed to hold out for only ten
more moves .
17 .!l ae8
A similar position arose in Gar-
 The Spanish Game 45

cia-Pinter mentioned above. The 28 j_xe3 � g6 29 "ilole2 �xe3, mat­

presence of the rooks on the e file ing) 27 _t xd5 cd l8 �xd2 "i!llt g5 29
(they were exchanged in that f4! "i!llt h4 30iitf5 l;o( e8 31 ;;;. n h6
game) favours White. Black resigned.
18 "l!IJ'd3 J! e6
1 9 �d2 I;� g6
White has successfully com­ Karpov-8passky
pleted his development, and his Bugojno, 1986
opponent makes a desperate at­
tempt to create complications on 1 e4 e5
the kingside. 2 �13 <E:lc6
20 g4 � xg4 3 .tb5 g6
21 fg I;� xg4 + Talking about rare continua­
22 � g3 f5 tions, there has been a growth of
23 � g2 "i!r;' h3 interest in Bird's Defence, 3 . . .
l4 J! fl �d4 and this will be considered
White has brought all his pieces later. At the same time the Janisch
to the rescue and Black has no Gambit (also known as the Greco
compensation for the piece. Counter-gambit)- 3 . . . f5, the
l4 j, f6 Berlin System-3 . . . � f6 4 0--0
l5 gO h5 �xe4, and also 3 . . . Jl.c5 have
l6 'i!it'e2 g fg6 almost disappeared. All that is left
27 �c2. to mention is the unfashionable
Black resigned exchange variation 3 . . . a6 4
The theoretical dispute about j_xc6 de. It seems to me there
this game continued a few months have been no substantial opening
later in Ehlvest-Geller, Vrsac, discoveries here and so we are not
1987: 18 . . . f5!? (a new move, as going to consider these variations
opposed to 1 8 � e6) 19 -dg5 (not in this book.
1 9 �f2- 1 9 . . . � xe2 20'i!i¥ xe2 4 c3 a6
� e8) 19 . . . f4 20 � xh3 fg 21 5 1t. a4 d6
� xe8! (not 21 ltrg2 gh 22 it_d2 6 d4 j_d7
'iit' g6 + ) 21 . . . � xe8 (2 1 . . . gh 7 0--0 (43)
loses immediately to 22 r;tr h I
� xeS 23 .itf4) 22 \ll g2 gh 23 it_d2 43
ltih8! l4 -dg5? (this allows Black B
to take over the initiative; White
should have played 24 ;.. xd5) l4
. . . .it.f4! 25 -de4 � e6 26 ltih1
]Lxd2?? (Black, in time trouble,
forgot about the intended 26 . . .
._ae3! which wins immediately: 27
l;;;. fl (27i/lf'e2 � xe4) 27 . . . "ilolh3
46 The Spanish Game
Probably only Smyslov and
Spassky of the top grandmasters
play the fianchetto system. Soko­
lov played 7 J,.gS against both of
them in the Candidates' Tourna­
ment (Montpel/ier, 1985). In Sok­
olov-8passky after 7 . . . f6 8 A e3
<21 h6 9 0-0 J,.g7 1 0 h3 �f7 I I
� bd2 0-0 1 2 J;� e l 'f!ke7 1 3 b4
* h8 14 j_c2 �cd8 1 5 � fl �e6
16 'f!kd2 �fg5 1 7 �3h2 f5 1 8 ef gf
1 9 f3 f4 20 Af2 �xh3 + ! 2 1 gh
�g5 22 * h i � xh3 Black de­
veloped a dangerous attack.
Black was successful in
Belyavsky-Smyslov from the
same tournament: 1 2 j_c2 (in­
stead of 1 2 K e l ) 1 2 . . . 'f!ke7 1 3
g e l * h8 1 4 "i!!i' b l �cd8 1 5 b4
�e6 1 6 a4 j_ h6 1 7 a5 j_ xe3 1 8
J;l xe3 � fg5 1 9 �xg5 fg 20 de de
2 1 "i!!i' b2 �f4 with the initiative for
Black, which brought him victory
in the end.
In Sokolov-8myslov (also from
the Candidates' Tournament,
Montpel/ier, 1985) after 9 h3 J.. g7
10 �bd2 "i!!i' e7 White played the
immediate 1 1 b4. Instead of the
solid I I . . . � f7 Black replied 1 1
• • • ed and the game unexpectedly
finished with a crushing defeat for
Black: 13 cd 0-0 13 0-0 fS 14 J.. gS
'f!ke8 IS J,.b3 + *h8 16 g e l "i!!i' c8
17 eS de 18 de J.. e6 19 J.. xe6 'flfxe6
20 <2l b3 �t7 21 �cs 'flfc4 22 g el
"i!!i' xb4 23 J.. d2 "i!!i' a3 24 I;;. e3 il!r xa2
25 K c2. Black resigned.
An unsuccessful innovation
was played in Popovic-Kovacevic
(Yugoslavia, 1 986): 9 h4 (instead
of 9 0-0 or 9 h3) 9 . . . � g4 1 0 h5
�xe3 I I fe J.. h6! 12 hg hg 1 3 d5
�e7 14 � bd2 * f7 15 * f2 *g7
16 j_xd7 'flfxd7 17 '/!It b3 f5! with
an advantage for Black.
Anyway, I decided not to waste
time on the long bishop's move
but to put it on e3 immediately.
7 J.. g7
8 I;;. el
The immediate exchange on e5
is worth attention and will be con­
sidered in the next game.
8 �ge7
Developing the knight on f6 is
no less popular. Here is a recent
example, Gonzalez-Gavrilakis
(Dubai 01., 1 986): 8 J,.e3 �f6 9
de de 1 0 � bd2 'fke7 1 1 b4 b6 1 2
h3 0-0 1 3 ..t b3 �h5 1 4 a4 a 5 1 5
b5 �d8 1 6 �c4 ..te6 1 7 .t e l
<EJb7 1 8 �e3 �c5 1 9 � d 5 J.. xd5
20 J.. xd5 g adS and Black with­
stood the pressure. All the same, 8
. . . �ge7 looks more flexible.
9 ..te3
I am still not in a hurry to
exchange on e5. In Short-Sprag­
gett (Montpellier, 1 985-you see
how popular this variation was at
this Candidates' Tournament)
after 9 de � xe5 1 0 � xe5 ..txe5 1 1
.,tb3 �c6 1 2 �d2 J.. g7 1 3 � f3
"i!!i' e7 1 4 .,tg5 f6 1 5 ..te3 �e5 1 6
�d4 0-0-0 1 7 f4 White had the
initiative.
9 0-0
10 �bd2 "i!!i' e8
As far as I know, this was a new
move. 1 0 . . . h6 had been played
before. Black's plan involves the
advance . . . f7-f5, but it was
impossible to play it right away:
I 0 . . . f5 I I A b3 + * h8 1 2 de de
1 3 ..tc5! b6 1 4 ..ta3 and White has
 The Spanish Game 47

a clear positional advantage. The have opened up to White's advan­

immediate 1 0 . . . b6, taking con­ tage.
trol of the critical cS square was 17 �fJ *b7
worth considering. 18 a4 fS!?
I 0 . . . tte8 in the game set a The positional threat 1 9 aS
trap by threatening I I . . . �xd4! compels Black to play actively.
However Black's queen finds itself But 1 8 . . . aS was better when the
on the same file as White's rook, concession of the square bS
which will play an important role wouldn't have been so dangerous
later. (if 1 9 � a3 � a7!). Spassky proba­
1 1 �bJ bly didn't expect the sudden
After I I dS Black intended not explosion which now happens in
I I . . . �d4 1 2 cd J.. xa4 1 3 "ttcl ! the centre.
x es 1 4 de de I S �c4 f6 1 6 b3 19 ef gf (44)
J.. d 7 1 3 'll!t a 3 with a clear advan­ Not 1 9 . . . �xfS 20 � f4! e4 2 1
tage for White, but I I . . . �d8! �c2, and i f 1 9 . . . .K xfS threaten­
and after . . . f7-fS, the knight ing 20 . . . X xn, I would reply 2 1
would have comfortably settled � fd2 (21 � h4 X hS) with a better
on f7. position.
11 b6
If I I . . . * h8 then the familiar
12 de de 1 3 �cS.
12 de de
After 1 2 � xeS 1 3 ttc2
White would have gained a lot of
space.
13 � c4 *h8
14 tt cl!
Preventing 1 4 . . . fS as after I S
J.. h6 Black would have serious
problems with his e pawn. 20 � fxe5! � xe5
14 J.. g4 Nothing was changed by 20 . . .
15 �g5! j_xeS 2 1 �xeS �xeS 22 � xh6
Preventing I S . . . fS again � 7g6 (22 . . . �d3 23 "ilf'gS; 22 . . .
because of 1 6 �e6. If I S J.. d l �n + 23 gf ..-g6 + 24 "!itgS!) 23
Black would transfer the bishop J.. x f8.
to e6. 21 <E) xe5 � xe5
15 h6 22 � xh6
16 hJ! J.. d7 The piece has been sacrificed
Black's bishop is compelled to for two pawns, but Black wasn't
retreat since, after 1 6 . . . � hS 1 7 concerned with arithmetic,
� xf7 + ! a xf7 ( 1 7 . . . 'll!t x f7 1 8 because his position is falling
� xb6) 1 8 g4, the position would apart.
 48 The Spanish Game

22 �d6
After 22 . . . j_g7 23 JLxg7
* xg7 24 -;ii' g5 + Black loses at
once due to the queen's position
on e8. If it was on d8 he could
play 24 . . . -E)g6. 22 . . . j_ f6 was
more stubborn, although after 23
j_xf8 "l!ifxf8 24 "iit f4 and � ad l
White would have developed a
menacing initiative.
23 ..-gs!
The exchange on f8 would lead
to unclear play.
23 \'!lr" g6
No better was 23 . . . � f7 in
view of 24 'i!if' h4 (24 iiJ h5 � f6!) 24
. . . * g6 (24 . . . *g8 25 "i!IJ'h5 with
a deadly pin) 25 g4 fg 26 hg Jtxg4
27 "lii' xg4 + *xh6 28 � e6 + *h7
29 j_c2 + mating.
24 � xe7 + j_ xe7
25 'it Xf:7 + * xh6
26 'itxd7
The tactical battle has clearly
ended in White's favour. Since his
king doesn't have pawn cover and
can end up in a mating net, Black
should have exchanged queens
now by 26 . . . "i!!J' d6 (26 . . . 'ft!lg7 27
'ft!lc6 + ) 27 'itxd6 cd. However, a
bishop and two pawns would be
stronger than a rook in this situa-
tion.
26 f4
Now White wins in a few
moves.
27 "i!IJ'xc7 � ae8
28 � d1 � f6
If 27 . . . f3 then 28 g4 threaten­
ing � d l -d5-h5.
29 *h2
A quiet move which stresses
Black's defencelessness.
29 aS
29 . . . f3 30 g4 � e2 3 1 * g3
� xb2 32 i;i d8 "1ft g7 33 'll!i' e 5 and
29 . . . � e2 30 � d8 � xf2 3 1
� h 8 + were equally hopeless for
Black.
30 � d4 � ff8
31 � d7 g c6
32 iWeS .-r6
33 .. dS � cS
34 �e4.
Black resigned.
If 34 . . . l!l;' f5(g6) then 35 flte7
and if 34 . . . � f5 then 35 j_c2
'!\li"e5 36 � h7 + ! *g5 (36 . . .
* xh7 37 -.. xe5) 37 h4 + ! * f6 (37
. . . *g4 38 �f3 + ) 38 fltc6 +
"l\li'e6 39 i;i h6 + wins beautifully.
Geller-8alov
Moscow, 1987
1 e4 eS
2 -2!f3 -2!c6
3 j_bS g6
4 c3
4 d4 is seldom played. The crea­
tion of a strong pawn centre is
more in keeping with the Spanish
Game. This game was played at
the recent Match-tournament of
the "three generations". In the
same tournament, in Balashov­
Smyslov White immediately ad­
vanced the d pawn: 4 d4 ed 5 il.. g5
�e7 6 j_xe7 "ilifxe7 (theory
recommends 6 . . . -E)gxe7 7 -2! xd4
d5 8 -2!c3 de -2!xe4 0--0 with equa­
lity, but the capture with the
queen is quite safe too) 7 0--0 -2! f6
8 .fl. xc6 de 9 "iit xd4 c5 10 '!'i'e3
.iLg4 I I -2!c3 JLxf3 12 '!\li"xf3 0--0--0

1 3 I!l ad l I!l d4 1 4 �d5 � xd5 1 5
ed � xd I 1 6 � xd I � e8 and Black
was out of danger.
4 a6
5 il._a4
In Chandler-8passky, Vienna,
1986 White captured on c6. It is
interesting that this exchange
hasn't occurred in this book
before. So: 5 j, xe6 de 6 0-0 J.g7
7 d4 ed 8 cd �e7 9 -de3 J. g4 10
J.e3 0-0 1 1 h3 J. xfJ 12 '{lt xfJ f5
(it was possible to take the d pawn
without hesitation) 13 j_ g5 'Wt'd7
14 J. xe7 'W/ xe7 15 e5 I!l ad8 16
� adl e5! 17 � d5 'li!t'f7 18 de
j, xeS 19 � fel � deS. Draw.
5 d6
6 d4 j,d7
7 0-0 j,g7
8 de
In my game with Spassky,
which was covered before, I
rejected this exchange for several
moves. The opening reference
books don't regard the immediate
capture on e5 as dangerous, but
this game proves them wrong. In
fact, 8 d5 was possible as well.
Here is a recent example: 8 d5
�ce7 9 J.xd7 + '\!\i'xd7 1 0 c4 h6
I I �c3 f5 12 ef �xf5 ( 1 2 . . . gf 1 3
� h4 � f6 1 4 f4 e4 1 5 _ie3 0--0 1 6
h3 c5 1 7 de be 1 8 '\!ilte2 with an
advantage for White, Fischer­
Filip, Cura�ao Candidates 1 962)
1 3 �e4 -2:l f6 14 -dxf6 + ..txf6 1 5
-2:1d2 0--0 1 6 �e4 with a better
game for White (Short-Spassky,
London, 1 986).
8 -2) xe5
Apparently it was more accur­
ate to capture on e5 with the
The Spanish Game 49
pawn, not allowing f2- f4, but
there is still no reason to criticise
the other option.
9 o2) xe5 de
10 f4 tjj e7 (45)
This position has been known
ever since a famous Fine-Alek­
hine game (AVRO-tournament,
1 938). There I I fe k_ xa4 1 2
'l!lj'xa4 + �d7 1 3 'l!lj'xd7 + 'i1Pxd7
14 il._ f4 *e6 1 5 �d2 brought a
quick victory for White. The right
move was I I . . . -2:1c6 1 2 A f4 'l!lj'e7
1 3 �d2 0---0--0 with complications
(Taborov-Bronstein, USSR,
1 978).
11 f5 gf
12 ef -2) d5
This is the ingenious reply dis­
covered by grandmaster Wcsteri­
nen, but here it will be refuted.
13 \i!i'Xd5
In Bronstein-Westerinen (Yur­
mala, 1 978) after 1 3 J. b3 -21 f6 1 4
_tg5 "Wte7 1 5 -2)d2 0---0--0 1 6 \i!i'e2
� hg8 Black achieved counterplay
along the g file.
13 Jixa4
14 \i!i' Xb7 .iLb5
Black doesn't seem to have any
special compensation for the
50 The Spanish Game

pawn after 1 5 X el either, but Romanishin-Balasbov

Geller won by force. Erevan, 1986
15 f6! J.xf6 (46)
1 e4 e5
2 �fJ �c6
3 i.b5 �d4
There are a lot of surprising
metamorphoses in chess theory.
One variation or another comple­
tely disappears over the horizon
for many years, or on the con­
trary, it attracts growing interest.
A recent example of this is Bird's
Defence, which has gained con­
siderable popularity in grandmas­
No better was 1 5 . . . ..txfl 1 6 fg ter tournaments. As usual, we
K g8 1 7 -.c6 + *e7 1 8 J.. g5 + . shall look at the most important
16 c4! recent developments in this varia­
Certainly not the immediate 1 6 tion in the annotations to the
X xf6- 1 6 . . . -. d i + 1 7 * f2 game.
-.e2 + 1 8 *g3 K g8 + and 4 <B xd4
White's king is mated. The bishop retreats to a4 or c4
16 X b8 are less logical, although they
After 1 6 . . . ..t xc4 1 7 -.c6 + or allow White to avoid sharp lines.
1 6 . . . J.. d 7 1 7 X xf6 Black loses a That is what happened in Roma­
piece. Further play was actually nishin-Malanyuk (Tbilisi, 1 986):
unnecessary, but this was a short 4 _tc4 _tc5 5 o21xd4 ..txd4 6 c3
time control tournament . . . J.. b6 7 d4 -.e7 8 0--0 <B f6 9 a4 a6
11 -.o -.d4 + 1 0 j_e3 �xe4 (after 10 . . . d6 I I
18 _te3 .-xc4 de de 1 2 ..t xb6 Black's pawn
19 �a3 .e6 structure would have lost flexibi­
20 � xb5 X xb5 lity) I I K e l 0--0 1 2 J.d5 ( 1 2 de?
21 -. xr6 1!f'xf6 � xf2! 1 3 -.e2 _txe3 1 4 f(( xe3
22 X xf6 K xb2 <Bg4 with an advantage for Black)
23 x xa6 0-0 1 2 . . . �f6 1 3 de <B xd5 14 llt�d5
24 x n x ds _txe3 1 5 X xe3 X b8 1 6 �d2 b5!
25 i.h6 X b6 and Black had no difficulties.
26 g xb6 cb In Georgiev-Tseshkovsky
27 K f6 Jtdl + (Dubai 01., 1 986) after 4 j_a4
28 *fl K a1 _tc5 White played a new move-5
29 11 xb6 X xa2 + b4 (instead o f the usual 5 0--0),
fJ . and after 5 . . . ..1 b6 6 0--0 � xf3 +
30
.
Black resigned. 7 "W"xf3 t1tf6 8 11t'g3 �e7 9 ..1 b2

�g6 1 0 c4 � f4 1 1 X e l c6 1 2 d4
he achieved an advantage. Black
could play 5 . . . � xb4 and after 6
� xd4 ed 7 0--0 White would have
the prospect of an initiative but
this would only just compensate
for the pawn.
4 ed
So Black has managed to keep
the bS bishop temporarily out of
play and his d4 pawn restrains the
development of White's queen­
side. At the same time White has
gained time and intends to create
a dynamic pawn centre. Thus,
each side has its own plusses and
minuses.
5 0-0
In Sokolov-Tukmakov (Len­
ingrad, 1 987) White retreated by 5
_tc4 at once. This time 5 . . . ..tcS
(the main answer to 5 0--0) was
not possible, and after 5 . . . � f6 6
0--0 �xe4 7 ..t xn + 'll x n 8
'i!\thS + g6 9 i16'd5 + 'lfg7 1 0 il6' xe4
"'* f6 I I d3 ..t cS White achieved a
small advantage but didn't man­
age to exploit it.
5 ..tc5 (47)
The once popular 5 . . . c6 is
more passive and the theoretical
assessment that White has an ad-
The Spanish Game 51
vantage hasn't changed for a long
time. The revival of Bird's
Defence is precisely connected
with the immediate development
of the bishop.
6 ..tc4
Another popular possibility
was 6 d3 c6 7 ..tc4 (7 _ta4 �e7 8
illt h S dS 9 �d2 0--0 10 �f3 aS I I
c3 de 1 2 be f6 1 3 � e l ..td6 with a
complicated struggle, Ivanov­
Kupreichik, Minsk, 1 985) 7 . . . dS
8 ed cd 9 ..tb3 (or 9 ..tbS + J.. d 7
10 ..txd7 + 'tltxd7 I I �d2 �e7 1 2
�b3 j_ b6 1 3 ..tgS and then either
1 3 . . . 0--0 14 j_ xe7 il6'xe7 1 5 �el
'i!\tf6 1 6 * h S � ac8, Chandler­
Wolf, FRG, 1 985: or 1 3 . . . f6 1 4
..td2 a5 1 5 a4 0--0 1 6 � e l �g6 1 7
h3 � fc8, Lunch-Tseshkovsky,
Trnava, 1 986, with an equal posi­
tion in both cases) 9 . . . �e7 (48)
48
w
This position has often oc­
curred in recent tournaments.
Here we give several typical ex­
amples which show that Black's
doubled pawns are not a weak­
ness but a strength and allow
Black to seriously restrain White's
position.
Dvoirys-Balashov (53rd USSR
championship, 1 986): 10 c4 0--0
52 The Spanish Game

( 1 0 . . . dc3 1 1 � xc3 0--0 1 2 � el d5 1 0 ed ( 1 0 R.b3 de 1 1 R.g5 is

h6 1 3 � f4 .i!_e6 14 d4 � b4 1 5 unclear) 1 0 . . . �xd5 1 1 R.g5 f6
g e3 � c8 1 6 � c t �f5 1 7 � d3 1 2 R.d2 �e6 with an equal posi-
� e8 1 8 h3 with a slightly better tion.
game for White, Nikolenko-Gor­ 6 • • • d6
elov, Moscow, 1 986) 1 1 cd � xd5 7 c3
1 2 �d2 -E)e3 13 fe de 14 'i/li'h5 ed This is a new plan. White threa­
1 5 tfxc5 dc (tf) 1 6 � axel �e6 1 7 tens 8 cd � xd4 9 'iij a4 + and if 9
� xe6 fe 1 8 � xf8 + 'i/fi' xf8 1 9 'illi' e3 . . . c6 or 9 . . . �d7, the bishop on
'iit f6 with equality. d4 is under attack after 10
Dvoirys-Kupreichik (Volgo- � xf7 + and 9 . . . * f8, of course,
grad, 1 985): 10 � e l 0--0 1 1 *h5 can't be acceptable for Black.
(Black also stood better after 1 1 Note that 7 b4 j_b6 8 a4 a5 is not
-E)d2 a5 1 2 a4 � b4, Anand­ dangerous for him, but 7 d3 also
Tseshkovsky, Calcutta, 1 986) 1 1 looks quite solid here. Let's look
. . . �e6 1 2 -E)d2 a5 1 3 a4 � b4 1 4 at one recent example. Ghinda­
� e2 �d7 1 5 � f3 J.,g4 1 6 \itg5 Komponias (Athens, 1 986): 7 d3
R.xf3 17 gf � a6 and Black had an -£) f6 8 J.,g5 h6 9 1L h4 g5 (this is a
advantage. risky move, better was 9 . . . �e6)
Markland-Davies (correspon­ 10 J.. g3 �g4 I I f3 .i e6 1 2 �d2
dence tournament, 1 985): I 0 'iii' h5 �d7 1 3 R.xe6 (the winner thinks
0--0 1 1 �d2 a5 1 2 a3 a4 1 3 R.a2 that 1 3 .a b3 � b6 1 4 j_ f2 c5 1 5 c3
� a6 14 -E) f3 � g6 with a sharp d5 1 6 ed � xd5 1 7 K e l + was
position. even stronger) 1 3 . . . fe 1 4 f4 h5 1 5
Belyavsky-Tseshkovsky (53rd fg �g4 1 6 'i!l;' f3 0--0--0 1 7 -E)b3
USSR championship, 1 986): 10 -E)e3 1 8 -E) xc5 de 1 9 �e5! and
�g5 f6 1 1 .i!_f4 ( 1 1 � h4 0--0 12 White won.
-2!d2 '1Ph8 1 3 �el a5 with equal 7 -E)f6
chances) 1 1 . . . 0--0 ( I I . . . .ie6 1 2 Black has a lot of other options
� e l J.. f7 1 3 c3 0--0 14 cd R. xd4 and here are some of them. Ehl­
1 5 -E)c3 .ag6 with a complicated vest-Kupreichik (Kuibishev,
struggle, Thipsay-Kupreichik, 1 986): 7 . . . 'i!i'f6 (7 . . . c6 8 d3
Frunze, 1 985) 12 -E)d2 * h8 1 3 -£)e7 9 -2Jd2 0--0 10 1L b3 is about
� e l a5 14 a4 j_ b4 1 5 h 3 g5 1 6 level; 7 . . . -2Je7 8 cd JL xd4 9
1L h2 � a6 with good chances for ·i!ii' a4 + -2}c6 10 j_d5 it. b6 1 1
Black. ]Lxc6 + be 1 2 d4 and White
Before going any further, I also , stands better) 8 -2}a3 it_xa3 9
want to mention 6 i!l!r'h5. Kuzmin- \ 'i!ifa4 + .iL.d7 1 0 'i!itxa3 .ae7 1 1
Malanyuk (Moscow, 1 986) con- . 'i!ii' b3 ( 1 1 1Le2 0--0 1 2 d3 � fe8 1 3
tinued 6 . . . i!l!r'e7 (6 . . . � b6 7 cd 'i!!j· xd4 1 4 .il.e3, Khalifman­
'i!ite5 + !) 7 d3 � f6 (7 . . . c6 8 R.g5 Kupreichik, Minsk, 1 986 also
with an advantage) 8 'i!ith4 c6 9 favoured White) I I . . 0--0 1 2 .

1Lc4 (9 1La4 is worth a try) 9 . . . jfxb7 .il,c6 1 3 i!ltxc7 � fc8 1 4

 The Spanish Game 53

'i!ii' a 5 _:txe4 1 5 d3 � c5 1 6 'l'ta6 13 .;tf3 � g4

.;tc6 (Ehlvest and Truss give a If 1 3 . . . .;tg4 then 1 4 �e2
beautiful variation: 1 6 . . . .;t xg2 winning a tempo.
1 7 * xg2 -2l f5 I S *b7 � eS 1 9 14 .;t xg4 .;t xg4
* h i -2l h4 20 �e3!! -2l f3 2 1 �d7 15 h3 .;te6
� dS 22 "{ith3 with a winning posi­ If 15 . . . .;th5 then 16 g4 .;tg6
tion) 1 7 cd � h5 I S d5! with a 17 -2le2 and White's initiative on
clear advantage for White. the kingside becomes menacing.
8 cd .;txd4 16 -2lel .;tc5
9 * a4 + *f8 17 d4
10 -2lc3 White has accomplished the
White would have stood some­ standard advance at last. Black's
what better after 1 0 .txf7 c5 (not problem is that he doesn't have a
10 . . . .;te5 I I .;t b3 .;txh2 + 1 2 useful plan and his king is in
* xh2 -2lg4 + 1 3 *g3 -wtg5 1 4 f3! limbo. At the same time White
or 1 0 . . . .;txf2 + I I 1. xf2 * xf7 can increase his pressure on the
1 2 d4 x es 1 3 -2lc3 *gS 14 .tg5) kingside.
I I J.b3 ( I I .;td5 -2l xd5) I I . . . 17 ttb4
� xe4 1 2 -2lc3. 18 'l'td1 .;tb6
10 c6 (49) 19 b3 f6
After 1 9 . . . d5 20 a4 *gS 2 1
.;ta3 'l't a5 22 e5 transferring the
bishop to d6 would be quite un­
pleasant for Black.
20 .tdl 'l't b5
21 .te3 x ae8
22 K cl -wthS
23 'tt dl d5
24 e5 *g8
It wasn't possible to withstand
the pressure on the kingside by
playing 24 . . . f5 because of 25 a4
Black keeps his bishop in the AdS 26 � f4 'tt f7 27 b4 h6 2S b5
centre for the time being, prevent­ g5 29 �xe6 K xe6 30 f4.
ing d2-d4. However, the bishop 25 ef gf
will be compelled to waste time 26 a4 jlc7
soon and White's pawn mass will 27 .tf4 .txf4
start advancing. That is why it 28 � xf4 ty g5
was worth considering I 0 . . . 29 � c3 'll f7
.;t xc3 I I de � xe4 1 2 .;txf7 -2l xc3 30 K g3 'l'th6
1 3 'tt b3 with complications. 31 Ji fe1 !! e7?
1 1 .;t el tll' b6 32 "Wrcl!
12 d3 .;te6 The X-ray effect along the c l -
54 The Spanish Game

h6 diagonal and the g file control allowed 33 �xe6 l?fxcl 34 K g7

allows White to finish the game mate. 32 . . . :1!1 g8 was also too late
forcefully. The immediate crisis in view of 33 K xg8 * xg8 34
could have been prevented by 3 1 �xd5!
. . K hg8.
. 33 � xe6 l?fxcl
32 aS 34 K g7 + .
Losing a piece, but 32 . . . g he8 Black resigned.
2 The Russian Game
Karpov-Portisch
Turin, 1982
In 1 982 I had a theoretical duel
with Portisch in the Russian
Game at Turin, Tilburg and
Lucerne. In spite of the small ad­
vantage in the actual score, 2: I in
my favour, the opening argument
can be considered to have been
won by White 3:0 . . .
I include the game which
opened our discussion as the main
game. However, the other two are
included in the commentaries on
the main game to put them all
together. Since this variation
doesn't occur any longer at the
highest level it is possible to con­
clude that it has been abandoned,
for the time being at least, and
certain conclusions about it have
been made. I believe that this
theoretical dispute with Portisch
prepared the ground for a new
dispute which occurred in my first
match with Kasparov and con­
tinued in the second and the third
ones. The appearance of this
opening in the world champion­
ship matches initiated, as usually
happens, an explosion of interest
in the Russian Game and the in­
formation boom virtually flooded
the theory of the open openings!
All the important games from my
matches with Kasparov and the
games which developed the ideas
born in these matches are covered
below. But first we shall look at
the dispute which was a harbinger
of the later turbulent events . . .
I e4 eS
2 -E:�O er6
3 �xeS
In recent years this capture has
been almost the only move
played. But not long ago 3 d4 was
considered to be the main move.
In fact the assessment of 3 . . . ed 4
eS �e4 S i!t xd4 as favourable for
White still hasn't changed. But
after 3 . � xe4 4 j_dJ dS S � xeS
. .
and now either S . . . �d7 or S . . .
�d6 Black has good chances. In­
cidentally, I have played this posi­
tion with White and with Black
and can say that it is full of
finesse.
S• . • �d6 6 i!tf3!? is an amus­
ing innovation that I prepared
with my permanent trainer Zait­
sev at some training sessions. This
idea was tried in Zaitsev-Yusupov,
Erevan, 1982. After 6 . . i!te7 •
7 0-0 0-0 8 <2:lc3 �xc3 9 be �xeS
10 de i!txeS I I J.f4 i!tf6 12 i!tg3
<2:lc6 White forced a draw by repe­
tition of moves -13 j_gS i!td6 14
J.f4 i!tf6 because he didn't have
anything better.
I remember my game with
Hort, where I played White
(Amsterdam, 1 980): 5 . �d7 6
. .
i!te2 i!te7 7 j_ xe4 de 8 �f4 exeS
9 � xeS J.fS 10 �c3 0--0--0 1 1
0--0--0 i!te6 1 2 "t{te3 hS 1 3 h3 f6 1 4
�h2 g6 1 S * b 1 j_h6 1 6 i!tg3
56 The Russian Game
K h7 1 7 J;� he l iitb6 with compli­
cations. In Karpov-Larsen (Til­
burg, 1 980) I had a setback: 6 . . .
�xe5 7 � xe4 de 8 jtxe4 �e6 9
jtxe5 jtd7 1 0 0--0 0--0-{) I I �e3
J,. b4 12 �c3 f6 1 3 iitg3 �xc3 1 4
be h5! 1 5 h4 g5! 1 6 f3 K dg8 with a
very strong attack for Black.
So both 5 . . . � d6 and 5 . . .
�d7 solve Black's problems and
that is why White more often
chooses 3 � xe5.
3 d6
4 �f3 �xe4
5 d4 (50)
50
B ( 1 ,48): 6 . . . �c6 7 0--0 J,.e7 8
Some time ago Spassky deter­ is wholly devoted to the develop­
minedly played the ancient 5 jte2 ment of the bishop on f5.
and managed to achieve a small
advantage in a number of games.
However, in my game with him in
the same tournament in Turin I
managed to equalise after 5 . . .
jte7 6 d3 � f6 7 �g5 1!!f xe2 + 8
J,. xe2 �e7 9 �c3 c6 1 0 0--0 � a6
I I .l. fe l �c7 1 2 � fl �e6 1 3 J,.e3
0--0 14 d4 K e8 1 5 d5 �xd5 1 6
� xd5 cd 1 7 A b 5 K d 8 1 8 X a� I
� f6 1 9 c3 �c7 20 �e2 x es l l
�d4 J,.d7 22 � f3 � xd4.
5 d5
6 �dJ �c6
7 (H) J,.e7
8 X el �13 (51)
Perhaps we should jump ahead
a little and mention a few moves
from each of the games from the
matches between Kasparov and
myself where the Russian Game
was played. Then the reader will
be able to understand better how
the later events developed.
( 1 ,28): 6 . . . �c6 7 0--0 �g4 8
X e l �e7 9 c4 � f6 1 0 cd;
(2, 1 5) and (3,6): 6 . . . �c6 7 0--0
�b4 8 c4 � f6;
( 1 ,30): 6 . . . �e7 7 0--0 �c6 8
X e l � g4 9 c4 �f6 1 0 �c3;
( 1 ,4 1 ): 6 . . . �e7 7 0--0 �c6 8
c4 � b4;
c4 � f6 9 �c3 0--0 .
So, in all these games Black's
light-squared bishop either
jumped out to g4 or stayed on its
original square. Of course, all
these positions will be scrutinised
in detail later. As far as my dis­
pute with Portisch is concerned it
8 . . . � f5 was first played in
Adorjan-Hiibner, Candidates '
quarter-final 1980. In fact, this
move was known to me after my

own game with Kasparov (the
first of all the 1 00 games!), played
three years before the first match
(Moscow, 1 98 1 , the Match-tour­
nament of the "three genera­
tions"). I played Black and this is
what happened: 9 � bd2 �xd2 10
�xd2 J.. xd3 1 1 ..- xd3 0-0 12 c3
�d7 (Hubner played 12 . . . •d6
and after 13 t;'f5 K ad8 14 J.. f4 a
draw was agreed) 1 3 J.. f4 a6 1 4
� e3 � ae8 15 K ae l J.. d 8 1 6 b3
� xe3 1 7 K xe3 f6 1 8 � e2 � f7 1 9
�d2 j_e7 20 �fl J.. f8 2 1 'i!!t' f3
X e7 with a small advantage for
White.
8 . . . J.. f5 was also played in the
fourth game of the Merano match
where after 9 J.. b5 Korchnoi
played an important improve­
ment 9 . . . J.. f6! Remember that
in Timman-Portisch (Moscow,
198 1 ) Black chose 9 . . . 0--0 I 0
j_xc6 be I I �e5 J.. h4 1 2 J,. e3
'{!lf'd6? (better was 1 2 . . . g e8) 1 3
..-h5! and got into a difficult posi­
tion. The idea of 9 . . . J.. f6 is to
prevent White from strengthening
the knight on e5, because after 1 0
j_xc6 + be I I �e5 j_xe5 1 2 de
0--0 Black's pieces are developing
freely and White has to be careful
to try to stop the c and d pawns'
advance. As far as the Merano
game is concerned after 10 � bd2
0--0 I I �fl the position was
equal. Admittedly, my opponent
made several inaccuracies soon
and even lost in the end. The
opening was not responsible for
the result of this game, but this
fact seriously affected Korchnoi
and he gave up the Russian Game
The Russian Game 57
altogether. It was a pity because
the next move was prepared for
him.
9 c4!
I prepared this opening surprise
with Zaitsev for the Merano
match. The innovation in fact hit
Portisch instead three times.
9 �b4
10 J.. fl
Of course not 10 cd � xf2! Pre­
cisely because of this move, 9 c4
was previously regarded as
dubious. However, White doesn't
have to hurry to clarify the situa-
tion.
10 0-0
The later game Karpov-Por­
tisch, Tilburg, 1982 continued 10
. • • de, but after 11 �c3! �f6 12
J.. xc4 0-0 13 a3 �c6 14 d5 I
managed to gain a significant ad­
vantage. The game continued: 14
. . . �aS 15 J.. a2 c5 16 J.. g5 .l;l e8
17 il\'(a4 i.d7 18 'ijc2 h6 19 J.. h4
�xd5 20 � xd5 J.. xh4 21 Ji, xe8 +
J.. xe8. Here I could have
increased my advantage by 22
Ji, e l or 22 X d l but I saw a forc­
ing variation which, unfortu­
nately, had a ftaw. After Black's
accurate reply I could, probably,
still have had a draw but I got
carried away, sacrificed a piece
and quickly lost. Here is the sad
end: 22 ..-e4? J.. f6 23 J.. b 1 *fB 24
'ith7 t;'xd5 25 j_ e4 tt'd6 26
'ijh8 + *e7 27 J.. c2 �c6 28 J.. a4
�d4 29 � xd4 j_ xd4 30 X el +
* f6 31 J.. xe8 ..-r4 32 X fl t;'e5.
White resigned.
Frankly speaking, when Por­
tisch again played the Russian
58 The Russian Game
Game at the 25th Olympiad in
Lucerne I was surprised by his
stubbornness. Apparently, the re­
sult of the previous game had
encouraged him . . .
1 1 a3 � c6 (52)
Now we have a structure simi­
lar to the one I had against Kava­
lek (I was Black) from the pre­
vious round: 9 a3 0--0 t o c4 � f6
1 1 �c3 � xc3 1 2 be �xd3 1 3
ttxd3 de 1 4 ttxc4 �aS 1 5 tta4
b6 with equality, but now White
wins time to a certain extent.
12 �cJ
Despite the fact that this move
brought me a beautiful victory in
Turin, I decided not to check
what Portisch had prepared when
I played him again later in the
year at the Lucerne Olympiad.
Since I promised to show all three
games let's see first how our last
game on the subject of The Rus­
sian Game finished. This was
quite an interesting game as well.
So Karpov-Portisch (Lucerne
01., 1 982): I longed the game. 21
12 cd ttxd5 13 � cJ <£J xt3 1 4 be
J.. g6. Black's pieces are awk­
wardly placed, for instance: 1 4 . . .
K e8 1 5 �eS (threatening 1 6 J.. c4)
1 5 . . . � xeS 1 6 Jt xeS ttd7 1 7
tt f3 with a double attack; 1 4 . . .
�d6? 1 5 c4 'ltt a 5 1 6 �d2 with a
win; 14 . . . I;� adS 1 5 �f4 with a
clear advantage. It is difficult to
blame the Hungarian Grandmas­
ter for playing 14 . . . �g6, which
prevented the direct threats but
didn't solve the basic problems.
15 c4 itt d7. After 1 5 . . . itt d6 1 6
d 5 � f6 the tactical 1 7 c5! was
decisive. 1 6 d5 �f6 17 K a2. The
tempting exchange sacrifice would
not work: 1 7 �g5 j_xal 1 8 de
'ltt xd l 1 9 K xd l .lf6 20 cb K ab8
and the b7 pawn cannot be
defended. 17 � a5 18 .lf4. It
• . •
was possible to take advantage of
the awkward position of the
knight on the edge of the board:
1 8 j_d2 b6 1 9 j_xa5 ba 20 <£Je5
ttd6 21 �c6 or 1 8 �e5 J.xe5 1 9
K xe5 b6 20 .lf4 K fe8 2 1 K ae2.
But the text move was more ener­
getic. 18 • . . K fe8 19 K ae2!
X ac8? A major concession. Black
didn't like 1 9 . . . K xe2 because of
20 ttxe2 with the threat of 2 1
<£Je5 �xeS 2 2 ttxe5, but still this
continuation wasn't so dangerous
for him. 20 �e5 itt f5? This is a
logical consequence of the pre­
vious manoeuvre. Only 20 . . .
J.. xe5 offered any resistance. 21
.ld2! Creating the unstoppable
threat of g2-g4. If 2 1 itt d2 with
the same idea then 2 1 . . . .lh5 22
f3 g5 23 .lg3 b6 would have pro­
� xc4
• • .
(53) . After 2 1 . . . b6 22 g4 ttc2
23 'W'xc2 J.. xc2 24 � xa5 J.xe5 25
A xc2 Black would Jose a piece.

22 g4! � xeS. No better were 22
. . . ftc2 23 ftxc2 � xc2 24 � xc4,
or 22 . . . f!j'xeS 23 K xeS � xeS 24
gS. 23 gf -2\ 13 + 24 * gl �hS lS
-.a4 � h4 + 26 *g3 _t xel 27
_t xel. Black resigned.
Now we return to our game in
Turin.
12 �xc3
13 be de
It is difficult for Black to do
without this exchange.
14 _t xc4 �d6
Taking stock, White's pieces
are better placed and exert clear
pressure on the centre.
In Ehlvest-Halifman, USSR
championship (Minsk, 1 987), the
latest game at the time of writing,
Black played 1 4 . . . �aS but
didn't manage to overcome the
opening problems. The game con­
tinued: l S i,a2 cS 1 6 �eS � f6 1 7
g4 �d7 1 8 .,tf4 � xeS 1 9 K xeS cd
20 cd �c6 2 1 li dS -.c8 22 h3
i,e6 23 X eS i, xa2 24 K xa2 'Wt'd7
2S dS �e7 26 K d2. Black
managed to liquidate White's ad­
vantage of the two bishops, but
there was nothing he could do
about the d pawn. In the end, he
The Russian Game 59
had to give up the exchange for
this pawn and White won.
IS �gS 'Wt'd7
16 <21h4! (54)
This manoeuvre allows White
to gain firm control of the initia-
tive.
16 <21 aS
In reply to 1 6 . . . .,tg4 there
was a choice between the simple
1 7 f3 � hS 1 8 g4 and 1 7 ft b l <2)aS
18 � d3 gaining control of the fS
square.
17 _tal bS
Black's pieces are placed in­
securely, so Portisch strives to
create strong points for them;
White has to act energetically.
18 a4 a6
After 1 8 . . . ba, I had the
pleasant choice between 1 9 �dS
.K ae8 20 x xe8 x xe8 21 �xfS
ftxfS 22 ftxa4 .K b8 23 K e l and
1 9 c4 cS 20 de (or 20 �xfS ftxfS
2 1 h4 <21 b3 22 � b l ftd7 23 de
�xeS 24 K a2 with excellent at­
tacking opportunities) 20 . . .
�xeS 2 1 X eS ftxd l + 22 K xd l
�g4 2 3 X xeS � xd l 24 x xaS.
19 ab ab
20 <21 xfS fjxfS
21 �e7
· K fb8 (55)
60 The Russian Game

nation! The bishop cannot be

taken-23 . . . • xn 24 x xaS!
x xaS 2S -.b3 + * g6 26 K e6 +
forcing mate.
23 * h8
24 � xd6 -.xn
25 K e7 -.m
26 �c5
White has a material and pos­
itional advantage. The result of
the game is clear.
After 2 1 . . . � xe7 22 X xe7 c6 26 -.r4
23 -.e2 �c4 24 K e l 'i!tdS 2S X eS 27 -.e2 h6
(2S � b3 is also strong) 2S . . . 28 K e4 -.n
X xa2 26 -.xa2 � xeS 27 -.xdS cd 29 K e5 €)c4
2S �t xeS K dS 29 X e7 X eS 30 30 x xa8 K xa8
X b7 White would win a pawn. If 31 K f5 .. g6
21 . . . K feS 22 �xd6 cd 23 � b l 32 .. e4 * h7
-.hs the weakness o f the eighth 33 h3 K al +
rank would tell: 24 X xeS + X xeS 34 * g2 K cl
2S g4 -.h3 26 X xaS -.xc3 27 35 �b4 €)d6
X a2! 2 1 . . . K fcS was also bad 36 � xd6 cd
after 22 � b l -.d7 23 -.n g6 24 37 .. d3 d5
i,a2! However, the transfer of the 38 fJ!
rook to bS is also met by an Black resigned.
unexpected and strong reaction.
22 g4!! Karpov-Kasparov
Of course it was difficult to Moscow, 1984
anticipate such an extravagant Game 41, World Championship I
move beforehand. Its objective is
shown in the following variations: I played six games with the
22 . . . -.f4 23 � xd6 -.xd6 (23 . . . Russian Game in my first three
cd 24 �ds) 24 -.n -.d7 (24 . . . world championship matches
K ffi 2S � xn + ! flhS 26 'i!tg3!) 2S with Kasparov and these games
g e2 and it is difficult to prevent had a considerable affect on the
the threat of 26 X ae I and 27 development of theory. Three
X e7. For instance: 2S . . . �c6 26 games are presented as the main
x ae l x es 27 -.xn + mating or games and the other three are
2S . . . X a6 26 X ae l X f6 ·27 'il\i'g3 covered in commentaries on the
with irresistible threats. main three. Of course, the reader
22 fitd7 will find many other important
23 � xf7 + ! games here which elaborate our
The whole point of the combi- ideas on the opening.

1 e4 e5
2 �fJ �f6
3 �xeS d6
4 .:21 fJ �xe4
5 d4 d5
6 �d3 J.. e7
7 0-0 12!c6 (56)
This is a turning point. The
defence by 6 . . . -E)c6 and 7 . . .
Ji... e7 (or in reverse order as in this
game) was researched by a Rus­
sian Master, K. Janisch as far
back as the last century. 6 . . . �c6
7 0--0 Ji... g4, with immediate pres­
sure on the centre, looks more
aggressive. However, this idea is
not new. As long ago as 1 922,
Schlechter in his "Handbook"
gave the following variation (with
a transposition): 6 . . . Ji... g4 7 0--0
�c6 8 J;l e l f5 9 c4! Ji... d6 1 0 cd
Ji... xO 'Wt'xO -2J xd4 1 2 jj-e3 'i!li'f6 1 3
Ji... xe4 fe 1 4 jj'xe4 + * f7 1 5 �g5!
jj'xg5 16 'i!li'xd4 with an advantage
for White (Capablanca-Marshall,
match, 1 909). Schlechter recom­
mended in his book, instead of the
unsuccessful 8 . . . f5, playing 8 . . .
J.. e 7! and then 9 Ji... xe4 de 1 0
J;l xe4 Ji... xO! I I 'i!li'xO �xd4 1 2
'iii' d3 �e6 with a n equal position.
Certainly, the theory has de-
The Russian Game 61
veloped a lot since then but, per­
haps, only now has it become very
clear how important it is to have
one extra tempo in this opening
system. One could say that two
different moves by the bishops­
the dark-squared to e7 and the
light-squared to g4-define the
two different routes of the theory
of the Russian Game. We will
cover the Ji... g4 move in detail
later.
The two Russian games of the
first match, where Kasparov
played White, convinced him that
this opening is not dangerous for
Black, so he chooses to play it
with Black this time. But I had a
small surprise . . .
8 c4 (57)
57
B
White attacks the centre im­
mediately. In the two previous
games in this opening 8 g e I was
played and White achieved
nothing. Here are these games.
Kasparov-Karpov, game 28 of
the first match: (the move order
was 6 • • • �c6 7 0-0 Ji... g4); 8 g e1
Ji... e7 9 c4 �f6 10 cd .,txfJ. This
move was first played by Smyslov
against Hiibner in the candidates
match (Velden, 1 983). 10 . . .
62 The Russian Game
€J xd5 was played before. I t "W' xfJ
"W'xd5. We will cover this position
in more detail later. ll "W'h3 €l xd4
13 �e3 "W'd7 14 "W'xd7 + * xd7 15
_te3 �e6 16 X adl _td6 17 ..tf5
*e7 1S €lb5 X bdS I9 � xd6 ed l0
b3 b6 21 g4 h6 ll ..td4 X ac8 23
..t c3 g6 24 ..t cl h5 25 fJ. Draw.
Kasparov-Karpov, game 30 of
the first match: (the move order
was 6 . • . ..te7 7 0--0 �c6 S X el
..tg4 9 e4 �f6): 10 €le3 (instead
of 1 0 cd) 10 . . . de I I ..t xe4 0--0 12
..te3 ..t xfJ 13 ijxfJ €l xd4 14
..txd4 \'txd4 IS � xe7 "ijxe4 16
il(fxb7 e6 17 "W'b3 "W'xb3 IS ab
� ab8 19 � a3 � feS 20 � xeS +
� xeS. Draw.
I want to remind you that the
. . . ..t f5 move-6 . . . €lc6 7 0--0
..te7 8 .l e i ..t f5-was covered in
some depth in the commentaries
on Karpov-Portisch (Turin,
1 982).
s ... €lb4
Kasparov took advantage of
my idea in the 48th game by play­
ing the Russian Game for White
again. Since the knight ma­
noeuvre didn't justify itself in the
4 1 st game, I was compelled to
move the other knight.
Kasparov-Karpov, game 48 of
the first match: S . . . €l f6 9 �e3
0--0 10 h3. Once and for all pre-
venting . . . ..tg4. 10 . . . de. After
1 0 . . . �b4 (if 1 0 . . . _te6 1 1
c5 was good) I I ..te2 c5 1 2 a3
�c6 1 3 de de 14 J.. e3 White had
some advantage (Ve1imiTovic­
Schiiss1er, Smederevska Palanka,
1 979). I I ..txe4 � a5 12 J.. d3 J.. e6
13 .1. e1 €le6. Let's consider the
possibility 1 3 . . . c5!? After 1 4
..t e 3 x es 1 5 "W'e2 cd 1 6 €l xd4
..tc4 1 7 X. ad 1 White is slightly
better but 14 . . . c4 1 5 J.. c2 €ld5
equalised for Black (Fedorowicz­
Kogan, USA, 1 985). Perhaps
White had better chances by play­
ing 14 _tg5 h6 1 5 ..th4. Now if 1 5
. . . c4 1 6 ..t xf6 ..txf6 1 7 ..te4 he
gains control of the centre. 14 a3
a6. In Lobron-Handoko (Yugo­
slavia, 1 985) Black played 14 . . .
K e8, but after 1 5 ..t b5 "W'd6
(White would have also been
better after 1 5 . . . a6 1 6 ..txc6 be
1 7 �e5) 1 6 ..tg5! K ed8 1 7 ..t xf6
..txf6 1 8 �e4 White achieved a
clear advantage and won the
game. 15 J.. f4 "W'd7. This is a
serious mistake. Correct was 15
. . • �d5, for instance: 16 ..tg3
..tf6 17 .tel €lee7 1S €le4 ..tf5 1S
�xf6 + � xf6 20 ..tb3 c6 ll €le5
�fd5 22 "W'fJ ..te6 23 I. ad I €lf5
24 j_xd5. Draw. (Gufeld­
Schiissler, Havana, 1985). 16 €le5!
� xeS 17 de €ld5 1S � xd5 ..t xd5
19 "W'cl g6. If 1 9 . . . h6 White
could play 20 x ac 1 c6 2 1 x e3
with a very strong attack. 20
.1. ad I c6. This leads by force to a
very difficult position for Black.
Black should have accepted a
worse ending by 20 . . . "W'c6 2 1
\1' xc6 ..t xc6 22 ..t c4 . 2 1 ..th6
X fdS (58) .
ll e6! fe. No better were 22 . . .
..t xe6 23 ..t xg6!, or 22 . . . ije8 23
"W'c3 f6 24 f4! 23 ..t xg6 ..tf8 24
..txf8 K xf8 25 J.. e4 � f7 26 I. e3
K g7 27 K dd3 � f8 28 K g3 * h8
29 i!te3 K f7 30 K de3. Black's
position is strategically lost

although in fact it took another
forty moves for White to win.
Before returning to the main
4 1 st game, I will allow myself one
more recollection, which is more
pleasant. I want to direct your
attention to the fact that the wave
of interest in the c2-c4 disrup­
tion of the centre started compar­
atively recently in the 80s. But in
the 70s the restrained c2-c3 was
preferred. In connection with this,
the Karpov-Korchnoi game
played in the candidates' final in
1 974 comes to mind. Let's look at
this game. We start from the posi­
tion in diagram 56.
8 j! el � g4 9 cJ fS 10 t;-bJ.
White gained nothing by playing
10 -21 bd2 t;-d6 I I t;-b3 0--0--0 1 2
i, bS J.. f6 1 3 'l!lta4 e, xd2 1 4 � xd2
,g he8 (Mestei-Wolf, London,
1 985) or by 10 c4 de I I �xc4
t;-d6! 1 2 �c3 0--0--0 ! (Kolmov­
Mikhalchishin, Minsk, 1 985). 10
. • . 0-0 II e,bdl. If I I . . . � fd2
then Black plays I I . . . -21 xf2! 1 2
*xf2 J.. h4 + 1 3 g 3 f4 1 4 * g2 fg
I S J.. e4 J.. h 3 + ! 1 6 * g l gh + 1 7
* xh2 ,.d6 + 1 8 * h i J., xe l ,
gaining a decisive advantage
(Ljubojevic-Makarichev, Amster-
The Russian Game 63
dam, 1 975). I I • • • * b8 Botvin­
.
nik suggested here I I . . . �aS 1 2
i!tbS c6 1 3 i!t a4 b S 1 4 'tltc2 �c4
or 1 2 'tlta4 �c6 1 3 � bS K f6 1 4
�xc6 K xc6 I S �eS K a6 with
equal chances. 12 hJ. 1 2 � fl . 1 2
J.,e2 and 1 2 i!txb7 were pre­
viously played, and in every case,
Black achieved a favourable posi­
tion. However, this move is not
too dangerous for Black either. 12
. • . � hS 1 3 'tlt xb7 K f6. In Ligter­
ink-Dvoretzky (Wijk aan Zee,
1 975) after 1 3 . . . �aS 14 'tlta6 cS
I S j_ e2 K b8 the chances were
equal.
14 'l!ltb3 K g6 Botvinnik anno­
.
tated the games of this match and
he regarded this move as the
primary cause of Black's defeat
and suggested the sharp 1 4 . . . gS.
However 14 . . . K g6 shouldn't
have led Black to catastrophe. IS
�e2 J.. h4? This move is Black's
undoing. No better was I S . . .
�xf2 1 6 * xf2 j_ h4 + 1 7 * fl
j_xe l 1 8 � xe l j_ xe2 + 1 9 * xe2
t;-e7 + 20 * f) x es 21 'tlt d l win­
ning. However, after I S . . . J.. d6 a
complicated struggle would lie
ahead. One possibility, suggested
by O'Kelly is 1 6 �eS � xeS 1 7
J.. x hS K xg2 + 1 8 * xg2 'tltg5 +
1 9 * fl 'i/lt h4 20 -21 xe4 i!txh3 + 21
*gl de 22 de J.. xeS 23 f4 *g3 +
with a draw. 16 x n � xf3 17
-21 xf3 J.. xfl + (a mistake-it is
either a sacrifice or an exchange,
but Black had less than five mi­
nutes for 24 moves) 18 K xfl 19
�xfl l9 * xfl *d6 20 e,gS! §!:. f8
21 ·�uJ -.ds 22 J.. f4 h6 23 �f3
K e8 24 J.. dJ K e4 2S gJ l!l f6 26
 64 The Russian Game
�c5 g5 27 -2:J xg5 hg 28 � xg5
� ee6 29 K f l 'iit g8 30 h4 I;� g6 31
� xe6. Black lost on time, but the
position was resignable anyway.
It is time to end the nostalgic
mood and go back to the Kar­
pov-Kasparov game, which was
played ten years after the one we
have just looked at . . .
9 � e2! (59)
Actually only this move is a real
innovation. Its obvious aim is to
avoid the exchange of the light­
squared bishop. In Hiibner­
Smyslov from the before-men­
tioned match, after 9 cd -2:Jxd3 1 0
� xd3 'i!lf'xd5 I I X e l � f5 1 2 ec3
( 1 2 �e5 which had been played
before, is not dangerous for
Black) 1 2 . . . e xc3 1 3 'ii!i' XC3 �e6!
14 .. xc7 j,d6 1 5 .. c2 0--0 Black
gained quite sufficient compensa­
tion for the pawn. The modem
state of 9 cd will be covered be­
Io N.
9 . •. de
We shall consider the interest­
ing possibilities 9 . . . _te6 I Q ec3
0--0 (or 9 . . . 0--0 10 ec3 1J.. e6J ip
the notes to several later games.
10 ..t xc4 0-0
11 -dc3
I I -2:Je5 -2:Jc6 1 2 -2:J xc6 be 1 3
-2:Jc3 -2!d6 1 4 �b3 �f5 1 5 d 5 c5
16 l;l e l �d4 1 7 �e3 led to an
advantage for White in Shindik­
Jannson (Italy, 1 983). The correct
reply for Black was the immediate
I I . . . -2:Jd6! After 1 2 � b3 �f5 1 3
a3 �d5 1 4 �c3 � e6 1 5 g e l c6
1 6 �c2 !Qc7 the position is equal
(Shindik-Zitz, Baden-Baden,
1 985).
11 ed6
12 j,b3 i.f6
After 1 2 . . . �g4 1 3 h3 �h5 1 4
g4! � g6 1 5 -2:J e5 followed b y f2-
f4 White i s clearly better, but 1 2
. . . * h8! was worth considering.
In Mnatsakanian-Diaz (Varna,
1 985) after 1 3 -2:Je5 f6 14 .:an �g4
1 5 h3 j,h5 16 j, e6 � xf3 1 7 gf
'i!ite8 1 8 � f4 f5 1 9 *h2 t/ei'g6 20
j, b3 �g5 Black obtained a pow­
erful initiative on the kingside.
13 h3
In a later game, Sokolov-Agza­
mov (52nd USSR championship,
1 985), White played the even
stronger- 1 3 -de5! and after 1 3
. . . -dc6 (more accurate was 1 3 . . .
c5 1 4 � f4 c4 1 5 ..£)xc4 <E}xc4 1 6
�xc4 j, xd4 1 7 <E}b5; but not 1 3
. . . � f5 1 4 �xf7! g xf7 1 5
..t xf7 + ff Xf7 1 6 'i!!'t b3 + or 1 3 . . .
_axe5 1 4 de -2! f5 1 5 �xf7 + !) 1 4
..tf4 <E} f5 1 5 -2:J xc6 be 1 6 d 5 c5 1 7
<E} a4 j,a6 1 8 gel j,e7 (better
was 1 8 . . . c4 1 9 �c2 �d6) 1 9
!4 c l k_d6 20 j_g3 White had a
big advantage.
13 j,f5
14 �e3 K fe8
15 a3 �d3 (60)
 The Russian Game 65

Black makes a losing move in

time trouble. 30 . . . * f7 or 30 . . .
�c2 were necessary.
31 a4! g d4
Not 3 1 . . . �eS 32 aS �d6 33
a6 R.xcS 34 J.. xcS ;g, xeS 3S
� e8 + 36 a7.
32 aS g xb4 ( 61)

An ambitious move. I S . . . �c6

was safer.
16 � b1 eS
17 de �e4
Not 17 . . . � xb2 18 � xb2
� xc3 1 9 cd! .txb2 20 _txf7 + !
*xf7 2 1 'i!i'fdS + and White has an
irresistible attack.
18 .tel! �xbl
Not 1 8 . . . �g3 1 9 fg K xe3 20 A historic position. White
'ifi'd2 k_d4 2 1 � xd4 22 *h2 �g6 could have won by advancing his
23 �dS! a pawn for just one more square:
19 "flt xd8 K axd8 33 a6! .tb3 (33 . . . �a4 34 a7
20 � xbl � xe3 _tc6 3S � e6 J.. d S 36 K d6; 33 . . .
21 g xb7 �xeS � b8 34 l;l xd l �a3 3S � b7!) 34
22 � xeS � xel � xb3 � a4 (34 . . . � xb3 3S
23 i;l xa7 l;l e8 + and 36 a7) 3S �cS K aS 36
White has won a pawn, but it l;l e4! 'lff7 37 i;� a4! l;#. xa4 38
was probably better to do so by 23 �xa4 .td4 39 �c3! and Black
i;� c l i;� d l + (23 . . . _te4 24 would have lost his bishop.
i;� xa7) 24 i;� xd l � xd l 2S � xa7. By winning this game I would
23 �d1 have won the match with a score
24 � e7 � xe7 of 6: I . After this chess history
2S .txe7 g d3 would have taken an entirely dif­
26 � gS �b2 ferent turn . . .
27 .tb4 33 K xd1 .td4
The threat was 27 . . . .tc2 win­ 34 �e6 .ta7
ning back the a3 pawn. Now if 27 3S � d7
. . . J.. c2 then 28 � e I ! 3S �xg7 K b2 36 � fS _t xf2 +
27 h6 37 * fl also offered White few
28 �e4 fS winning chances.
29 �eS g dS 3S K b1 +
30 g el f4'! 36 * hl .t xfl
66 The Russian Game

37 �xf4 K al 54 K b4 K e7
38 �e6 �;t xa5 55 l;!. c4 K e8
56
Black makes it harder for him­ g3 j_b2
self. As was pointed out by I. 57 * f3 K e6
Dorfman, he could draw by 38 . . . 58 K c5 j_ d4
g5! 39 K g7 + (39 K d5 � e l ) 39 59 J;� d5 ,i e5
. . . *h8 40 X g6 ffh7 41 � f8 + 60 .W. b5 J... c7
*h8 42 a6 (42 X xh6 + ffg7 43 61 g c5 J... b6
J;t g6 + * xf8 44 X f6 + ffg7 45 62 K c8 j, d4
J;t xf2 K xa5) 42 . . . ,i b l + ! 43 63 ll g8 + J... g7
* g3 .W. a3 + 44 * g4 K a4 + ! 64 h4 K a6
39 � xg7 + ff h8 65 wr4
40 K f7 ,ie3 Alas, after 65 h5 + ff xh5 66
41 ff g3 ll xg7 K a3 + 67 fff4 ll f3 +
The game was adjourned in this Black had a perpetual rook sacri­
position and after analysing it I fice.
convinced myself that it was 65 li aS
almost impossible to exploit the 66 K e8 ll f5 +
extra pawn. I tried to achieve this 67 ff e3 K e5
for the next 30 moves, but without 68 K g8 K e7
success. 69 ff f4 ll f7 +
41 i,d2 70 ff g4 h5 +
42 ll d7 j_c3 71 *h3. Draw.
43 fff3 ff g8
44 .£Jf4 g f5
45 ffe4 x f7! Khalifman-Arkhipov
46 J;t d8 + Moscow, 1985
White had more chances of vic­
tory in a minor pieces endgame: 1 e4 e5
46 � xf7 *xf7 47 'llf f5. However, 2 �f3 22} f6
Kasparov's trainers I. Dorfman 3 � xeS d6
and E. Vladimirov published an 4 �f3 �xe4
interesting analysis soon after the 5 d4 d5
match, which showed the method 6 �d3 i,e7
of Black's defence in this ending. 7 0--0 � c6
46 ff h7 8 c4
47 ll d3 g e7 + Having seen the previous game,
48 fff3 j_b2 the unsophisticated reader might
49 ll b3 .}cl think that it was possible to dis­
50 -2!d5 K e5 courage Black from playing 8 . . .
51 -2!f6 + ff g6 � b4 here for a long time, due to
52 -21 e4 II t'5 + the 9 j_e2 retreat. But this was
53 *e2 K e5 not the case! Soon after that game

an improvement was found on
Kasparov's 9 . . . de.
8 Black now has an unpleasant
9 J.. e2
This is the innovation. Who
was the inventor? The Inter­
national Master Arkhipov played
this move twice in the Moscow
International tournament in l 98S.
He put the N sign by this move in
his commentaries on his game
with Tseshkovsky in lnformator.
However a month before this
game Christiansen played this
move against Ljubojevic at
Linares. So apparently the "scien­
tific" discovery was made by the
two players independently of each
other.
10 � c3
1 0 cS!?, with the aim of gaining
space on the queenside ( l l a3
�c6 1 2 b4) hasn't been tried yet.
After 1 0 . . . �c6 ( 1 0 . . . aS l l
�eS) I I -.!ta4 aS 1 2 <21eS White
has a small advantage.
10 0--0
1 1 i.. eJ
I I cd will be considered later.
11 J.. f5 (62)
In all three of S. Arkhipov's
games, which are mentioned here,
The Russian Game 67
he played this move. We will
cover the other possibilities later.
threat- 1 2 . . . �xc3 1 3 be �c2.
12 "f!rbJ
Ivanov against Arkhipov (Mos­
cow, l 98S) played 1 2 a3 and after
1 2 . . . �xc3 1 3 be �c2 1 4 K a2
�xa3 ( 1 4 . . . � xe3 l S fe with a
small advantage for White) I S
K xa3 J.. xa3 1 6 cS J.. b2 1 7 -.!tb3
J,. xc3 1 8 -.!txc3 c6 1 9 l U I K e8 20
». a3 *c7 21 �d2 bS 22 ll, a6
White stood better. However, Iva­
nov pointed out that the position
would have been uncl�ar after 1 6
. . . bS 1 7 *b3 J.. xcS 1 8 de c6 1 9
�d4 J.. d 7 20 J.. f4 a S 2 1 J.. d6
». e8.
12 c6
In the same tournament at
Moscow 1985 against Tsesh­
kovsky, Arkhipov continued 12 . • .
de 13 J.. xc4 a5 (more accurate
than 1 3 . . . � xc3 1 4 be �c2 I S
X ad l � xe3 1 6 fe with pressure
on the n square) 14 aJ � xcJ 15
be (after I S ab bS! 16 be be 1 7
-.!txc4 J.. e6 1 8 *bS ab 1 9 K xa8
-.!txa8 20 cb J.. d S Black is better)
15 a4 16 1!rb2 �cl 17 ll, a2
• • .
� xeJ 18 fe J.. d6 19 -.!txb7 (correct
was 1 9 �eS! i._ e6 20 i._xe6 fe) 19
• • • II. b8 20 *d5? (the losing
move, but Black also had an ad­
vantage after 20 *a6 *d7) 20 . • •
J.. dJ! 21 "t!rc6 (2 1 J_xd3 i.. xh2 + ;
2 1 K d l c6! 22 -.!txc6 K c8) 21 • . •
.txn 22 • xn K b1 + and Black
won.
Apparently, after the game
Arkhipov was convinced that, de­
spite his successful result, the
68 The Russian Game

exchange on c4 didn't solve all the 17 e xe4

opening problems. Some rounds After 1 7 exc6 €)exc3! 1 8
later he tried a refinement and exd8 ee2 + 1 9 fr h l <2!xcl
delayed the exchange by one Black gains a material advantage.
move. 17 � xe4
13 .M acl 18 x xe6
1 3 cd cd 14 x ac l aS IS a3 Better was 1 8 <21 xf7 * xf7 and
-E) xc3 16 x xc3 a4 1 7 'it d l �c6 only now 1 9 g xc6.
offered White no advantage but a 18 'lte8!
more promising continuation was 19 K eel
1 3 cS -E)xc3 14 be �c2 I S ll, ac l Not 19 K fc l owing to 19 . . .
�xe3 1 6 fe. The waiting move by <2!xe3 20 �xf7 j_dS!
White's rook now allows Black to 19 �d6
capture the initiative. 20 e xti "Wtxti
13 • . • de 21 K c6 .M d8
This capture is now much There is nominal equality on
stronger because Black has the board, but Black's pieces are
already prepared the b7- bS ad- very well coordinated.
vance. 22 j_g5 �f4!
14 � xe4 b5! (63) This leads by force to a number
of exchanges and a winning
bishop endgame.
63
w 23 'lifxf7 + fr xti
24 j_ xd8 j_xe6
25 � g5 � xg2!
26 d5 J.. xd5
27 g dl fr e6
28 g xd5 fr xd5
29 fr xg2 fr e4
The difference in actiVIty is
great and this proves to be deci-
sive.
15 �xti + !? 30 J.. e3 a6
The modest I S �e2 led to the 31 h3 ,i e5
loss of a pawn: I S . . . J.. e6 1 6 32 b3 J.. d4
lltf"d I <2!xc3 1 7 g xc3 j_ xa2 1 8 b3 33 J.. g5 fr d3
aS and aS-a4, or 1 7 be �xa2 1 8 34 *0 fr e2
� c2 and bS-b4. 35 �e3 J.. f6
15 .M xti 36 fr e4 *b2
16 €)e5 <2!d5. 37 fr d3 fr xa2
This is the only reply but it is 38 fr e2 aS
sufficient. If 16 . . . tt"e8 then 1 7 39 j_b6 a4
g4! � xc3 1 8 be etc. 40 b4
 The Russian Game 69

White would lose immediately end, I decided to try this variation

after 40 ba b4. with Black myself.
40 *•3 9 0--0
41 �cS h5 10 �cJ � e6
42 �f8 g6 11 J.eJ �f6 (64)
43 �cS g5
44 f3
If 44 � d6 then 44 . . . g4! 45 hg
h4 followed by h4-h3 and J.e7
winning the b pawn.
44 �e5
45 �e7 _ftgJ
46 � xg5 * xb4
47 f4 *c4
48 fS *ciS!
49 J.d8 J.d6
50 J.h4 *e4
51 f6 * fS This was played for the first
52 *cJ * e6 time in another game by the
53 �g5 b4 + Yugoslav grandmaster: Ljuboje­
54 *c4 bJ. vic-Christiansen (Linares, 1 985).
White resigned. After 12 K c l c5 1 3 a3 cd 14 � xd4
J. xd4 1 5 � xd4 €) xc3 1 6 l;l xc3
€)c6 1 7 cd €)xd4 1 8 de fe ( 1 8 . . .
Ljubojevic-Karpov � xe6 was safer, when White
Bugojno, 1986 would have had only a marginal
advantage) 1 9 �d3 l!l f7 20 1ltd2
1 e4 e5 itf6 21 l;l fc l White achieved a
2 �f3 €)f6 considerable advantage.
3 � xe5 d6 I captured on e4 in my game
4 �f3 � xe4 with Ljubojevic. But before going
5 d4 dS any further, note that White has
6 �dJ another two options, which are, in
7 0--0 fact, absolutely safe for Black: 1 2
8 c4 cd � xc3 1 3 be � xd5, and Black is
9 J.el simply better; 1 2 a3 €) xc3 1 3 be
After the 4 1 st game of the first €)c6 1 4 cd §txd5, and the position
Karpov-Kasparov match, it is equal.
seemed nobody would want to 12 de
play this position with Black. But, 13 c6
as we shall see, the theory is deve­ 14
loping rapidly and Black has 15 aJ
found various antidotes. In the 16 €) c2
70 The Russian Game

1 6 . . . ttd7 1 7 K fd 1 �c7 threa­ Perhaps Black should have

tening b7-b5 was perhaps better. played 26 . . . *d7 and if 27 h3
17 K fd1 x ae8 then 27 . . . f6 28 't!i'c5 �e6 fol-
18 't!i'a4 c5! lowed by . . . l;t e8 and . . . 't!i'f7.
Not allowing 1 9 b4 which 27 h3 f6
would have cramped Black's 28 't!i'c5
queenside. Another option was 28 � b6
19 K acl! 't!i'dl + 29 *h2 11td6 + 30 K g3
If 1 9 d5 �d7 20 't!i'b3 't!i'd6 K b8.
followed by . . . �e5, I would 28 't!i'XO
have been content with my posi­ Not 28 . . . *xd4 29 X eS +
tion. But now White renewed the *f7 30 *e7 + .
b2-b4 threat, having overpro­ 29 � xeS b5
tected the c4 pawn. 30 �b4
19 ed 30 b3 � f7 3 1 a4 K d l + 32
20 � xd4 �c5 *h2 ba 33 ba .M a l 34 �a3 .l. a2
Black gains some counterplay also led to a draw.
by sacrificing a pawn. Despite 30 h5
what happens afterwards, the re­ 31 K e7?
sult of the opening favours White. White has been exerting pres­
21 tt-xa7 � d3 sure on Black's position through­
2 1 . . . �xd4 22 X xd4 K xd4 23 out the game and he could still
�xd4 �d3 24 X c2 led to an develop an initiative by 3 1 *h2
advantage for White. .l. d7 32 h4 threatening *g3, f2-
22 �xd3 f3, * f4. Now I can breathe at
The other exchange, 22 �xe6, last.
looked dangerous. 22 . . . �xc l 23 31 K d1 +
� xd8 � xe2 + 24 * fl K xd8 25 32 * hl K cl
'ilrxe2 Jl xd l (25 . . . � xb2 26 But not 32 . . . X fl because of
tt-xb7 and so on) 26 *xd l tt-d7 + 33 � e l !
27 *cl would be dubious for 33 �c3 �;t el
Black. However, after 22 . . . 34 �d4
tt- xe6 23 x c2 (23 � xd3 ed 24 If 34 *g3 then 34 . . . h5-
11txb7 X b8 with an equal posi­ h4 + .
tion) 23 . . . b5! Black has suffi­ 34 Jt dl
cient counterchances. 35 K d7 K d3
22 � xd4 35 . . . h4 36 �;t d8 + *h7 37
23 � xd4 ed �xf6 X xf2 was also possible.
24 X xd3 l;t xe4 36 h4 *h7
Not 24 . . . K a8 owing to 25 37 f3 *g6
't!i'b6 K a6 26 't!i'b3 X ad6 27 't!i'c3. 38 K d8 *"
25 x xe4 � xe4 39 �b6 Jl xd8
26 x e3 *d6 40 � xd8 *g6
 The Russian Game 71

Another possibility was 40 . . .

65
f5 4 1 *g3 g6 42 * f4 * e6. B
41 *g3 * rs
42 * fl *eS
43 *e3 J.. b3
44 j_ c7 + * dS.
Draw.
Despite the peaceful result, it
should be admitted that Black
didn't completely solve his open­
ing problems.
getic advance of the f pawn Black
Karpov-8eirawan acquires opportunities of active
Brussels, 1986 counterplay on the kingside. I I
. . . J.. f5 and I I . . . J.. f6 have
1 e4 eS already been covered in detail.
2 �fJ �(6 Only recently, in Sokolov-Smys­
3 � xeS d6 lov (Moscow, 1 987), Black tried
4 �fJ � xe4 another move- I I . . . �f6. The
s d4 dS game continued 1 2 a3 �c6 1 3 b3
6 j_d3 J.. e7 �e4. The exchange on c3 when
7 0-0 � c6 White has a pawn on b2 is not
8 c4 �b4 very good for Black because
9 J.. e2 j_e6 White strengthens his centre. In
10 �c3 0-0 fact, this was proved by the main
1 1 J.. e3 game. However, now that the b
In the same tournament in pawn has moved one square and
Brussels Ljubojevic played I I cd b2xc3 is impossible, Black's
against Seirawan. After I I . . . knight returns to e4 to exchange
� xc3 1 2 be �xd5 1 3 "i/ri'c2 ( 1 3 on c3. Unfortunately, White cap­
�d2 �b6 is equal) 1 3 . . . c5 1 4 c4 tures the knight himself and Black
� b4 1 5 'i!IIJ' e4 "iiri' d 7 1 6 dc? J.. f5 1 7 has lost two tempi which in the
ilrd4 ilre6 1 8 J.. b2 J.. f6 1 9 "iiri' d2 end is decisive. 14 �xe4 de 1 5 d5
J.. xb2 20 "i/ri'xb2 �c2 Black gained ef 16 J.. x f3 j_d7 1 7 de J.. xc6 1 8
an advantage, which he converted �xc6 be 1 9 jf'f3 iitd3 20 g ab l
into a point. After the game it was jf'g6 21 a4 a 5 2 2 J.. f4 j_d6 23
suggested that White should have j_ xd6 'i!\f'xd6 24 � bd l jf'b4 25
played 1 6 j_ b2. Note one more � d7 � ab8 26 g xc7 jf'xb3 27
variation: I I �xe4 de 1 2 � e l jf' xc6. Despite the fact that Smys­
with chances for both sides. Iov managed to save the game in
11 fS!? (65) the end, it is unlikely that he
An interesting idea from the would like to repeat his opening
American champion. By the ener- experiment.
72 The Russian Game

12 a3 <E:� xc3 pionship, 1986). In the position in

13 be �c6 diagram 62 Black played 19 • • .

14 'ii!J' a4 a6! It was not a difficult move but

A tense struggle would have now it becomes clear that White's
ensued after 14 cd j_ xd5 1 5 c4 rook is deprived of the b5 square
J.. x f3 16 J.. xf3 f4 1 7 J.. d 5 + * h8 which is an important fact. 20 h3
1 8 j_cl .;t f6. j_hS 21 <E:� gS? White got flustered
14 f4 by being faced with an unexpected
IS j_d2 .h8 move. After 2 1 � bd l (2 1 �h2
16 � abl � b8 was also interesting) he would still
17 � ret de have had the initiative. 21 • . •

18 .,ixc4 j_g4 j_ xe2! This is much stronger than

19 �e2 (66)
This was an important mo­
ment. Jumping a little ahead, I
want to point out that here, after
1 9 . . . Ad6 20 h3 j_ h5, White's
rook made an unusual ma­
noeuvre: 21 � b5! Ae8 22 '/!!f c2 a6
23 .!;;. f5! The rooks were
exchanged, the light squares in
Black's camp became weak and I
took advantage of this fact in the
end.
Soon after this, Seirawan found
a way to improve Black's play to
some extent. The game where it
happened should be looked at in
full.
Rohde-Seirawan ( USA cham-
21 . . . _ixg5 22 j_xh5 f3 23 j_xg5
il1i'xg5 24 j_xf3! � xf3 (24 . . .
� xd4 25 \'!\lixd4 � xf3 26 � xb7
� bf8 27 � xc7 .!;1 xh3 28 "(lfd6
A! g8 29 illf' e6 with the huge and
terrible threat of '/!!f xg8 + ) 25
'ilifxc6! and White has a clear ad­
vantage.
22 -2)e6 "i!\rdS 23 � xe2 f3! Black
creates dangerous threats by sac­
rificing the exchange. 24 � eel fg!
25 -2) xf8 � xf8 26 i11i' d l "i!\rfS!
Black has a winning position but
it is interesting to see how the
game finished. 27 * xg2 '/!!f xfl +
28 *h i �f3. White is forced to
give up a piece in order to avoid
mate. 29 � e3 � xe3 30 j_ xe3
ili"xe3 31 i!l';'g4 g6! 32 � fl . No
better was 32 i!i"c8 + *g7 33
� xb7 _id6 34 � xc7 + j_xc7 35
'/!!( XC7 + �e7. 32 • . . *g7 33 � f3
'/!!f e l + 34 * g2 j_ xa3 35 '/!!f f4
\'!\lie7 36 � e3 •f7 37 '/!!f e4 j_d6 38
� e2 -2)e7! 39 '/!!f xb7 �dS 40 � fl
-2)e3 + White resigned in view of
the line 41 * g l Ah2 + ! 42 * xh2
\'!\lixf2 + 43 * h l "i!\rg2 + !
It is time to return !O my game
with Seirawan.
19 j_d6
 The Russian Game 73

I want to remind you again that hopeless. However he managed to

1 9 . . . a6 was correct here. organise a stubborn resistance
lO h3 _th5 and the game went on for another
Wasn't it better to play 20 . . . 40 moves-just like another
..tf5 here? game!
11 .�;. b5! _te8 37 ..tb3 b5
If 2 1 . . . _tg6 then 22 c4 was 38 *" 1lt f8
unpleasant. 39 .tel tli'g8
ll 'itel a6 40 j_b3 + tli' h8
This is too late, but White also 41 h4 �b7
stood better after 22 . . . _tg6 23 4l e4 'iiir g8 (67)
.,td3 ..t xd3 24 lltxd3 threatening
g h5 and �g5.
J;. xfS
l3 J;. fS! 67
24 llt xfS ..tg6 w
15 'itt"g4 •r6
16 A.e4 g f8
The queen exchange 26 . . . ..tf5
1 7 -.. gs 'iiir xg5 28 �xg5 wouldn't
help Black.
l7 a4 ..tel
28 *h5 h6
Not 28 . . . jl xa4 29 � g5 h6 30
� f7 + *h7 3 1 J;. e6!
19 J;. e8! ..t fS 43 'i!1J'd3
The a4 pawn is taboo again This retreat didn't affect the re­
because of 30 �g5! sult of the game but it did cause
30 ..tdS! ..td7 the whole evening to be wasted on
31 Iii xf8 + 'itxf8 completing the adjourned game
lf 3 1 . . . ..txf8 then 32 �e5. . . . After 43 J.c2! threatening
jl �h4 Ae8 1Lc3, d5 and 'i!IJ'xh6 mate the game
33 'l!lrel would have finished much sooner.
It is difficult for Black to defend 43 iit h7
the light squares even without 44 *el g5
rooks. But White shouldn't hurry. 45 e5
Now after 33 �g6 ..1xg6 34 "iif' xg6 It was better to exchange the
J?je7 White would let go some of queens at once.
his advantage. 45 j_e7
33 J?jd8 46 ab ab
34 "iir e4 flt e7 47 h5 _ta5
35 �g6 + ..1 xg6 48 ..tel
36 "iif' xg6 e6 Here again 48 "Wrxh7 + with an
Black's position is strategically easy winning endgame was cor-
74 The Russian Game

rect. Now to achieve a win I have

68
to create a small endgame study. w
48 . • . "\!lr e7 +
49 *dl
After 49 *f3 g4 + ! only Black
would have winning chances: 50
*xg4 '\tg5 + 5 1 * f3 (5 1 *h3
lt;'xh5 mate) 51 . . . "i!Jxh5 + 52 g4
(52 *e4 "i!Jg6 + , 52 *xf4 "i!fg5 + )
52 . . . "i!f h l + .
49 'iii' e l +
so lt;'xfl + king off from the pawn. After 66
51 �g3 .tc7 .tc5! Black could still hold
52 fg out.
53 66 g4
The queens are exchanged at 67 * c3 ,.td6
last. Black admittedly has an 68 .lg7 .tf4
extra pawn but he doesn't have 69 *b4 *dS
any hope in the minor piece end- 70 * xb5 * c7
game. 71 * c5 .te3 +
53 *g7 72 *d5 .tf4
54 d5 -2! xc5 73 .tf8
55 .tbl + *18 Black is in a total zugzwang.
56 de � a6 73 *b6
57 .t a3 + *eS 74 .td6 .t g5
58 .te6! 75 .t xg3.
This is a pretty method to cut Black resigned.
the king off from the pawn.
58 ... .tb4 Sax-Yusupov
Of course not 58 . . . <2) b4 + 59 Thessaloniki Olympiad, 1984
.txb4 .txb4 60 c7.
59 .tbl ..U8 I e4 e5
60 .td7 + *d8 2 � f3 -2!f6
61 .te5 <E) b4 + 3 <2) xe5 d6
62 *dl -2!d5 4 <2)f3 � xe4
63 .te6 -2!c7 5 d4 d5
64 .tf7 -2! e8 (68) 6 .t d3 j)_ e7
Black's last chance was 64 . . . 7 ()....{) -2! c6
�a6. But now comes the pro­ s J;� el .tg4
mised endgame study . . . 9 c4 <2)f6
65 .1. xeS! *xeS 10 cd .t xf3
66 .tf6! II �xf3 �xd5
Now another bishop cuts the (69)

69
w 1 7 M xe4 � xc3 1 8 be �a5 with
My first match with Kasparov
lasted five whole months, and
many ideas which were born there
were refined elsewhere during the
actual marathon, in particular at
the olympiad in Thessaloniki.
This game was played there. The
reader should remember that this
position occurred in the 28th
game of this match, where after 1 2
i!th3 the game finished in a quick
draw. But there were other
queen's moves and one of them
was tried in this game.
12 i!t g3
The exchange on d5 has been
played many times but it doesn't
appear to offer White much: 1 2
"i!fxd5 � xd5 1 3 �c3 0--0--0 ! (after
1 3 . . . �db4 White manages to
achieve an advantage: 14 �e4
<E:� xd4 1 5 �e3 c5 1 6 � xd4! cd 1 7
-2!b5 * f8 1 8 a3! �c6 1 9 X ac l d3
20 � xd3 J,.f6 21 b4 g6 22 J,.e4,
Popovic-Kurajica, Yugoslavia,
1 984; or 1 5 . . . K d8 1 6 J,.xd4
K xd4 1 7 a3 �c6 1 8 <E:�b5 K d2 1 9
o2lxc7 + * d8 20 � b5, Abramo­
vic-Rukavina, Yugoslavia, 1 985)
14 �e4 � b4 1 5 J.. d2 � f6 ( 1 5 . . .
o2lxd4 1 6 �xd5 ..txd2 1 7 .l, ed l c6
1 8 X xd2 cd 1 9 K xd4 was also
The Russian Game 75
sufficient for a draw) 1 6 a3 � xe4
complete equality (Ehlvest-Mik­
halchishin, Lvov, 1 984). In
Lobron-Mikhalchishin, White
played 1 6 � xc6 (instead of 1 6 a3),
and after 16 . . be 1 7 a3 .1. xd4! 1 8
.
ab M xd2 1 9 K xa7 M e8! Black's
position became even better.
Before going any further, I
want to remind you that 1 2 i!th3,
which Kasparov chose in the 28th
game, occurred for the first time
in Velimirovic-Kurajica (Bela
Crkva, 1984), and Black captured
on d4 with the queen there, 1 2 . . .
'5'xd4, when 1 3 �c3 X d8 1 4 �f5
h5 1 5 i!tg3 * f8 1 6 �e3 i!tb4 1 7
a 3 -.as led to approximate equa­
lity. 1 4 � b5!? was worth a try.
One way or the other, I decided to
avoid surprises and deviated my­
self by capturing on d4 with the
knight which solved Black's prob­
lems.
12 i!txd4
13 �c3 � ( 70)
70
w
Up to now the game has fol­
lowed Hiibner-Smyslov, which is
regarded as the original of the
exchange on f3. Black played 1 3
76 The Russian Game

. . . II d8 there and had to work 2 1 <21 xa6 �e7 22 �c5! <2lxf5 23

hard to draw. By castling at once � xf8.
Black doesn't just equalise but 21 K c4 l2) xf5
actually manages to take over the 22 K xb4 K d7
initiative. 23 K cl
14 �b5 l\f'g4 23 � f4 1ost after 23 . . . J;�. c8 24
14 . . . l!f'b4 was also worth con­ J;�. xb7 <2ld5 25 �e5 �xc7 26 K c l
sidering; in Abramovic-Kurajica �e8!
(Yugoslavia, 1 984) the players 23 l;l c8
soon agreed to a draw after this 24 K bc4 J;�. cd8
move. 25 h3 �xe3
15 l\f' xg4 26 fe \ti'f8
1 5 l!f'xc7 was dubious in view of 27 e4 *e7
1 5 . . . �c5! with an attack for 28 K b4 K dl +
Black, but after 1 5 �xc7 Ji adS 1 6 29 J� xdl J� xdl +
'{itxg4 � xg4 1 7 �e2 the position 30 \ti'fl *d6
would be equal. 31 e5 +
15 � xg4 White makes a decisive mistake
16 �f5 in time trouble. The endgame was
Here again 1 7 �e2 was better, unpleasant for him, but after 3 1
but 1 7 �xc7 �c5! was rather *e2 it would have been possible
risky. to put up a stubborn resistance.
16 l2)f6 31 \ti' xe5
17 �xc7 � adS 32 -E)a8 b5
18 �e3 33 a4 �d5
1 8 �b5 was bad owing to 1 8 34 .W. b3 ba
. . . �i dS. 35 .� b7 .w. bl
18 a6 36 \ti' f3 a3.
An accurate move; after 1 8 . . . White resigned.
�d6 1 9 � b5 �xh2 + 20 * xh2
l;l d5 21 �xa7 White would have
stood better. Belyavsky-Smyslov
19 � acl Reggio Emilia, 1986/1987
A mistake, after which Black
obtains an initiative. Correct was 1 e4 e5
1 9 � ed t �d6 20 �b6 �e5 21 2 �f3 � f6
J4 xd8 � xd8 2 2 � xa6 retaining 3 �xe5 d6
equal chances. 4 �f3 -E) xe4
19 �b4! 5 d4 d5
20 � n 6 �d3 �c6
I f 20 � ed t then 20 . . . �a5! 7 0--0 �e7
20 �d4! 8 c4 �b4
I f 20 . . . �a5 White would play It seems appropriate to men-

tion the 48th game from the first
Kasparov-Karpov match where 8
. . . � f6 was played, and present
more recent material on this sub­
ject.
Ljubojevic-Shiissler (New
York, 1 985): 9 �c3 0-0 10 .l, e l
de I I � xc4 J.. g4 1 2 J.. e3 � xf3 1 3
iltxf3 � xd4 1 4 j_xd4 iltxd4 1 5
.l, xe7 -t�rxc4 1 6 iltxb7 c6 1 7 iitb3
iltxb3 1 8 ab K fe8 1 9 l, a3 .l!l fe8
20 X xe8 + l, xe8 2 1 * fl with a
better endgame for White.
Kudrin-Wolff, USA champion­
ship, 1985: 9 �c3 0-4) 10 cd �xd5
1 1 X el j_e6 12 a3 J.. f6 13 � e4
� de7? (the wrong knight goes to
e7; 1 3 . . . �ce7! equalised) 14
� g5! � xg5 15 �xg5 �f5 (but
now 1 5 . . . h6 was correct) 16 d5
�xe4 17 it xe4 -2! b8 18 -.h5 h6
19 X ael � xd5 20 �xf7 �f6 21
� xh6 + *h7 22 iith3. Black
resigned
Fedorowicz-Kogan (USA
championship, 1 985): 9 h3 0-0 1 0
�c3 de I I j_xc4 � a 5 1 2 j_d3
j_e6 13 K e l c5!? (a new move
compared to the 48th game) 1 4
J.. e3 c4 1 5 �c2 �d5 with equa­
lity.
In the last game, Christiansen
suggested 1 4 J,.g5! h6 1 5 j_h4, in
order to meet 1 5 . . . c4 by 1 6
J.. xf6 � xf6 1 7 j_e4 with some
initiative in the centre. One way
or the other, we can see that 8 . . .
-2! f6 remains viable, and certain
failures at the "highest level"
don't mean that the adherents of
the Russian Game have lost faith
in this opening.
Now we return to the dispute
The Russian Game 77
started by the 4 1 st game of the
world championship match. It
serves as a guideline for our dis­
cussion on the Russian Game.
9 cd
This capture, which allows the
exchange of the light-squared bis­
hop, was known even before the
4 1 st game, as was pointed out
before, and it also occurred more
than once after this first game
with Kasparov. Therefore this or­
der of moves still remains topical.
At the same time the reputation of
this variation as safe for Black
remains intact.
9 � xd3
10 iltxd3 -.xd5
( 71)
This position has been known
for at least twenty years, so it is
impossible to cover it in full. As
usual, we are interested in the
newest material.
11 K el �f5
Before presenting Belyavsky's
move here, I want to remind you
that 12 �c3 � xc3 13 ttxc3 once
used to be White's only continua­
tion. Let's look at two important
78 The Russian Game

examples of the way to play this soon become quite dangerous. In

position. First let us return to the principle, it would be possible to
1 983 candidates' match. put a full stop here, but the epilo­
Hiibner-Smyslov, Velden, 1983: gue of this game was so fascinat­
1 3 . . . ..t e6. (The tempting 1 3 . . . ing that it is a pleasure looking
c6 was definitely refuted a long through it to the end. 23 -.aS
time ago: 14 ..t h6!! X g8 - 1 4 . . . *b8 24 � d3 -.b3 2S K cl b6 26
gh 1 S x es 't!td7 1 6 x ae l _te6 1 7 't!teS .M c8 27 <E� f4 tt b2 28 K c6
dS! cd 1 8 ll xe6 fe 1 9 ttxh8 + .,t f8 *b7 29 *h2 'it'xf2 30 �d3 iiit dl
20 't!tf6 etc; 1 4 . . . .,te4 l S .,t xg7 31 tre4 *b8 32 -BeS *b8 32 .tieS
X g8 1 6 X xe4! 't!txe4 1 7 ,M e l K d8 33 .t�d7 + ! *c8 ( 72)
't!txe l + 1 8 i!txe l l;l xg7 1 9 'tlteS­
l S X eS 't!td7 16 X ae l .,te6 1 7
�gS! 0-Q-0 1 8 � xf7!, and White 72
w
won, Browne-Bisguier, USA
championship, 1 974). 1 4 ftkxc7
.,td6 1 S 't!tc2 ()....{) 1 6 .,td2 .tfS! 1 7
't!tb3 't!txb3 1 8 ab f6 1 9 .,tc3 * f7.
White's extra pawn is neutralised
by Black's two strong bishops.
The game ended in a draw.
Van der Wiei-Short, Biel. 1985:
13 .,te6 14 X eS. Perhaps tak­
• . •

ing on c7 was better but it was still

not too dangerous for Black. 14 34 X d6!! Black resigned. An
. • • 't!tc6 (Black also has a solid instructive fin ish: 34 . . . K xd7 3S
position after 14 . . . -.d7 l S .,tgS ..-as mate: 34 . . . cd 3S t�rc6 mate.
f6 16 X e3 fg 1 7 X ae 1 ()....{) 1 8 A beautiful victory, but in no way
X xe6 .,t f6) I S *el. After 1 S due to the opening . . .
ttxc6 be the slight weakness in Now we must return to the
Black's pawn chain would be main game.
compensated by the presence of 12 �eS!! ( 73)
two bishops. IS . . . 0-6-0 16 .tgS
.txgS 17 X xgS .,tdS 18 �eS lltb6
(Black would have had a fine posi­ 73
B
tion after 1 8 . . . 'tlth6 1 9 l;l g3 f6)
19 Iii xg7 � hg8 20 Iii g3 '{lt xb2
(wasn't it better to pick up a pawn
on d4?) 21 K dl � xg3 22 hg
.txa2? After 22 . . . l;l e8 Black
shouldn't lose. In fact, he cuts his
bishop out of the game by taking
the a2 pawn, and White's threats

Although this move has also
been known for quite a while,
theory (I mean the Encyclopedia
of Chess Opening) doesn't devote
much attention to it. If 1 2 . . . f6
then it gives 1 3 jl(f3, recom­
mended by P. Keres, and if 1 2 . . .
�h4? 1 3 g3 <E) xg3 then 1 4 f;f3!
even decides the game (Zuidema­
Barendreght, Amsterdam, 1 966).
However, 1 2 . . . f6 should be con­
sidered in more detail. The point
is that after 1 2 . . . f6 1 3 f;f3 Black
can reply 1 3 . . . g6! and after the
knight leaves the e5 square Black
has no problems. And if 14 g4 fe
1 5 gf gf White is even in more
trouble ( 1 6 f;xf5 K g8 + 1 7 tlf fl
f;c4 + ).
Makropoulos-Toth (Budva
1 9 8 1 ) countiuned 1 3 �c3 <E) xc3
14 -.xfS �bs 1 5 lll' h 5 + g6 1 6
ii(( h3 fe 1 7 � xeS "tlt'xd4 1 8 'il{(e6
-.dl + 19 g el "1Y!i'd7 20 j_g5 and
the chances were equal.
A new move was played in De
Firmian-Plaskett (Copenhagen,
1 985): 1 5 "WI' g4. After 1 5 . . . � xd4
16 �d3 Black was lured by the
exchange- 1 6 . . . �c2? and was
subjected to a mating attack: 1 7
� b4! -2) xb4 1 8 '$'xb4 c5 1 9 'iltg4
lti fl 20 J.h6! gh (20 . . . g hg8 2 1
� ad l gh 22 '$'xg8 + ) 2 1 � ad l h5
(21 . . . j{(c6 22 )4. xe7 + ! * xe7 23
'i-tt' g7 + *e6 24 � el + with a
crushing position) 22 i(( e2 'iltf5 23
'i'ii' xe7 + *g6 24 � d7 h4 25 h3
;;;;. ag8 26 � e4 *h5 27 ilt fl +
� g6 2 8 g d5. Black resigned.
resigned.
De Firmian, in his annotations
on this game in lnformator, sug-
The Russian Game 79
gested 1 6 . . . * fl 1 7 � xe7 +
* xe7 1 8 llt' xg7 + with an unclear
position. But it turns out to be, in
fact, very clear! A game between
two young masters-Ulibin­
Serprer (Sochi, 1 986) continued
for only seven moves: 1 8 . . . � fl
1 9 "i/!r'g4 � adS 20 b3 "\li'g6 2 1
iWfxg6 hg 2 2 j_a3 + * fl 2 3 � e 1
� h5 24 ii. b2 �e6 25 -E:�f3 � hd5.
White resigned.
So, 1 7 X xe7 + was unsound,
and therefore the whole variation
is not dangerous for Black. Per­
haps one can draw the conclusion
that, after 1 2 <E)e5 Black's reply 1 2
. . . f6 is sufficient to equalise. But
what was Belyavsky intending
when he played 1 2 �e5 and what
was Smyslov afraid of when he
rejected 1 2 . . . f6? We shall proba­
bly learn the answers to these
questions in the future.
12 g6
It should be noted that Black
does not have much choice here.
According to Arkhipov 1 2 . . .
0--{)-0 was insufficient owing to 1 3
ilff3 ( 1 3 �c3? 'iltxe5) 1 3 . . . g6 1 4
g4 j_h4 1 5 �c3 ( 1 5 �d3 �x£2 1 6
f;xd5 �h3 + 1 7 tlfg2 K xd5 1 8 gf
j_ xe 1 1 9 �xe 1 l, e8 favours
Black) 1 5 . . . �xc3 1 6 be �e6 1 7
f;xd5 K xd5 1 8 g5 with an advan­
tage for White. And after 1 2 . . .
<E)d6 1 3 �c3! f;a5 ( 1 3 . . . f;xe5
14 de �xd3 1 5 ed cd 1 6 �d5) 1 4
il(re3 (or 1 4 f;f3) Black i s under
strong pressure.
Judging by the quick draw
which it led to, it is possible to
come to the conclusion that Smys­
lov's new move 1 2 . . . g6 was
80 The Russian Game
successful. However, Black had to
overcome a lot of obstacles to
achieve the draw and so 1 2 . . . f6
is still more solid.
13 jtf3
13 g4 was bad because of 13 . . .
-dxf2! But why not 1 3 -dc3? The
first impression is that the inser­
tion of the moves 1 2 � e5 g6, in
comparison with the games
covered above, favours White. In­
deed, after 1 3 -dc3 -2l xc3 14 'i!(( xc3
1 4 . . . Ae6 is bad owing to 1 5
�h6! 0-0--0 1 6 � xf7! �xf7 1 7
� xe7 and White is a pawn up
with the better position. 1 4 . . . c6
lose because of 1 5 �c4! when
there is no satisfactory defence
against 1 6 -d e3 and 1 7 d5. The
correct move is 1 4 . . 0-0--0 with
. A Xe4 1 9 �xe4 + 'i!li'e6 20 �d4
a good position for Black because
1 5 � xg6 is impossible owing to 1 5
. . . hg 1 6 � xe7 'i!!i' d6! attacking
the rook and the h2 pawn.
It seems that White had a dan­
gerous initiative after the queen
move chosen by Belyavsky. But
Smyslov coolly took the pawn . . .
13 'i!li'xd4
14 -dc3 jj'xeS
Black goes on picking up every­
thing possible. The material will
be won back soon but there will
be nothing left on the board . . .
IS � r4
After 1 5 -d xe4 Black would slip
away just in time by 1 5 . . . 0---0 .
IS 'i!li'aS
16 b4 ( 74)
After 16 -2l xe4 again there was
enough time for 16 . . . 0---0 ( 1 7
�g3 �e6). Belyavsky probably
counted on this b pawn advance
74
B
when playing 1 3 jj-f3. If now 1 6
. . . 'i!(( xb4 ( 1 6 . . . 'i!ii' b6), then 1 7
-dd5! decides immediately. The
same move is played after 1 6 . . .
'i!li'a6. The ensuing variations are
worth attention.
16 . . . 'i!li'a6 1 7 �d5! AdS (there
is nothing else) 1 8 g4! ( 1 8 � xe4 +
0---0 2 1 �h6 f6 22 AXfS * xf8 23
'i!!i' d2 *g7 24 � e l �d7 is good
for Black). Black has three op­
tions now, but none of them offer
much.
1 8 . . . c6. This move blocks the
queen's path to the e6 square, and
White takes advantage of this fact
immediately. 1 9 � xe4 + Axe4 20
'i!(( xe4 + *d7 (20 . . . * f8 2 1
�h6 + * g8 22 'i!li'e8 mate) 2 1
� d I and White has a crushing
position.
1 8 . . . �e6 1 9 "tl!i'xe4 0---0 20
Ah6 � e8 (20 . . . �xd5 2 1 "ti!i'xd5
winning the exchange) 2 1 'i!li'e5;
18 . . . 0---0 19 gf gf ( 19 . . . -d f6
20 A h6) 20 * h l with an irresist­
ible attack.
But Black still has one defence
which is sufficient.
16 'i!li'a3!
17 �dS
 The Russian Game 81

Alas, this time this move allows "Wtxd4 i!l'e7 + 1 0 A.e2 �e7 (Chi­
the exchange of queens. But 1 7 burdanidze-Agzamov, Frunze,
�xe4 i!l'xf3 would even have led 1 986). Also after 6 A.e2 j_e7 7
to Black's advantage in the end­ 0--0 0--0 8 d4 j_f6! Black's pieces
game, because Black could hardly successfully interact in the centre.
be expected to fall for 1 8 � f6 + Here are two examples also taken
tlr f8 . . . 1 9 � h6 mate . . . from the international tourna­
17 'l!ltxf3 ment at Frunze, 1 986.
18 gf �d8! 9 h3 � e8 10 j_f4 (if 10 �c3
19 fe Makarichev had prepared l 0 . . .
White can only be proud of the -21 xd4! I I -21 xd4 j_ xd4 1 2 \iri' xd4
fact that he has been attacking for -2J xc3 1 3 i!l'xc3 � xe2) 10 . . . j_f5
five moves being a piece down . . . I I � e l 'I!Vtd7 1 2 �a3 j_ xh3 1 3 gh
19 �e6 ))lfxh3 1 4 j_e3 �g3 with an attack
20 � xc7 + j_xc7 for Black (Chiburdanidze-Mak­
21 � xc7 f6 arichev):
22 a3 � c8 9 A.e3 (9 d5 -2Je7 1 0 j_d3 j_f5
23 x acl .,.,. I I � e l -E:�c5 and 9 it.d3 j_f5 10
Draw. � e l � e8 also lead to good posi­
This was a short game but you tions for Black) 9 . . . � e8 l 0
can't call it a grandmaster draw. � bd2 il_f5! I I �b3 d5! 1 2 � e l
de 1 3 JL xc4 �d6 1 4 �e2 -2J b4,
Kasparov-Karpov and Black had an advantage
Moscow, /985 (Kupreichik-Makarichev).
Game 15, World Championship II 5 dS
6 ..1d3 �c6
1 e4 eS Interest has revived in the
2 � f3 �f6 ancient continuations 6 . . . j_d6
3 �xeS d6 and 6 . . . -E:�d7. Since these moves
4 �f3 �xe4 haven't reached "world stan­
5 d4 dards" yet, we will omit them in
Another move order adopted this book. In order to present the
quite often is 5 c4. The logical material in a more compact way,
reply 5 . . . �c6! was suggested by we are compelled to reject some
Makarichev. Now if 6 d4, then 6 games and variations.
. . . d5 is good, e.g. 7 �c3 � b4 8 7 0-0 j_g4 ( 75)
'i/lf'c2 i�We7 + 9 �e3 �g4 10 f(;'cl We now start considering
.axc3 l l be �a3 1 2 'i!\i'd2 �b4! another popular plan in the Rus­
with a decisive advantage for sian Game, involving the imme­
Black (Kupreichik-Mikhalchi­ diate development of the light­
shin, Kuibishev, 1 986). After 6 squared bishop on g4 and leaving
�c3, Black equalises easily by 6 the dark-squared bishop on its
. . . �xc3 7 de �f5 8 �d4 �xd4 9 original square. It seems that
82 The Russian Game

not hurry with a check but should

save it for the future.
9 J.. xf3
Other continuations could lead
to serious problems for Black: 9
. . . j_e7 1 0 cd �xd5 I I J.. e4; 9 . . .
�xd4 1 0 iitel + (the check now
comes in handy!) 10 . . . j_e7 ( 1 0
. . . �e6 I I �e5) I I �xd4 de 1 2
�f5 cd 1 3 �xg7 + * f8 1 4 J.. h6
* g8 1 5 f3 with a strong initiative
there is not much difference, but for White.
nevertheless the play sometimes 10 iitxf3 �xd4
takes a totally different direction. II K el +
Black creates serious pressure on The game took the usual direc­
the d4 pawn by saving a tempo on tion after this check, but only for
the move . . . j_e7, but at the same one move! Later we shall consider
time his king is delayed in the the interesting continuations 1 1
centre and becomes the object of '.tfe3 + and I I '({(( h3 in detail.
White's attack. This variation has II .J.. e7 ( 76)
its own plusses and minuses. It is
interesting that the position in
diagram 75 wasn't even con­
sidered in ECO ( 1 9 8 1 ).
8 c4 �f6
9 �cJ
Black has created pressure on
the d4 pawn and Kasparov de­
cided to sacrifice it immediately.
This idea, though not new, is
quite dangerous for Black. We
shall cover 9 cd below. 9 � e l +
J.. e7 would transpose to the posi­ 12 "t�tdl!?
tion from the 28th game of the Lobron-Karpov (Hannover,
first match. As you will remem­ 1 983) continued 1 2 .-g3 de 1 3
ber, 1 0 cd J.. x f3 I I "t�t xf3 "t�txd5 1 2 j_xc4 ( 1 3 -. xg7 lost to 1 3 . . .
"t�t h3 �xd4 1 3 �c3 "t�td7 1 4 � f3 + 1 4 * h I K g8 1 5 "tltxf6
illl' xd7 + •xd7 led to a quick � xe l ) 1 3 . . 0--0 14 J.. g5 j_d6 ( 1 4
.

draw. Of course, this result is . . . -dc2 1 5 J! xe7!) 1 5 ilfh4 h6! 1 6

acceptable to Black, and so Kas­ J.. xf6 -.xf6 1 7 -.xf6 gf l 8 X e4 c5
parov took another path after 46 1 9 i;i h4 * g7 20 �e4 j_e7 2 1 �g3
minutes' thought. He came to the f5 22 g h3 j_d6 23 f4 b5 24 j_d3
sensible decision that he should c4 25 j_xf5 g fe8. The pawn has
 The Russian Game 83

been won back at last, but Black the centre; if 1 6 . . . � c8 then 1 7

has a big positional advantage ..te5! is unpleasant) 1 7 <2) xe7 +
and he easily transformed it into f/ xe7 1 8 J.. d6. However, Black
victory. could avoid all his difficulties by
1 2 llt d l was undoubtedly an playing 1 6 . . . <2)xf4 1 7 l, xe7 +
improvement, although, as will be 'll f8 1 8 .K e5 f/d6!, e.g. 1 9 K f5
seen later, it was not too danger­ .!';;. d8 20 � xf4 ftxd5 etc.
ous for Black either. 16 0-0
12 �e6! There was no time to defend the
Black manages to simplify the pawn: 1 6 . . . a6 1 7 _te3 0--0 1 8
position by this move. After 1 2 . . . � ad I with strong pressure.
de 1 3 _txc4 0--0 ( 1 3 . . . c5 1 4 17 �xe7 +
•a4 + ) 1 4 � xe7 flxe7 1 5 'i!(f xd4 Not 1 7 "«fxb5 _tc5!
Black would have been on the 17 -. xe7
verge of defeat. White also had an 18 -.xbS a6
advantage after 1 2 . . . 0--0 1 3 cd. This is more accurate than
13 cd <2)xd5 immediate 1 8 . . . X fd8 1 9 ..t e3
14 _tbS + c6 "«fd7 20 "«fxd7 x xd7 2 1 � ad l .
IS � xdS cb ( 77) 19 "«fb3 K fd8
20 ..te3 K ac8
20 . . . b5 also led to equality.
77 21 .K acl b6
w
22 b3 <2!d4.
Draw.
After 23 J.. xd4 l, xc 1 24 .K xc 1
.K xd4 both sides have run out of
steam.

Kasparov-Karpov
London, 1986
Game 6 World Championship 1/1
The tension has gone. White
has the mighty knight in the 1 e4 eS
centre and Black has doubled 2 � f3 �f6
pawns, but one of them is an extra 3 <21 xeS d6
pawn. One thing balances the 4 .:E) f3 <2! xe4
other and a draw is soon agreed. s d4 d5
16 "«fb3 6 _td3 <2)c6
1 6 J.. f4 preventing Black cas­ 7 0-0 ..t g4
tling looked tempting e.g. 1 6 . . . 8 c4 �f6
()...?.{) ( 1 6 . . . _td6 1 7 J.. xd6 "«fxd6 9 <2!c3 ..t xf3
1 8 � f6 + 'll e7 1 9 .:E)d5 + .:E)f8 20 10 -.xo � xd4
lltf3 and Black's king is delayed in 11 tt"e3 + ( 78)
84 The Russian Game
78
B two bishops and an initiative for a
As you will remember, we have
been following the 1 5th game of
our previous match. In that game
Kasparov had played a home pre­
pared innovation here, but his I I
tte3 + in the present game had
already been suggested a year ago
and so was no surprise to me.
11 �e6
12 cd � xdS
After 1 2 . . . �c5 White could
play 1 3 ttf3 �d4 1 4 g e l + * ffi
1 5 tt f4 �xd5 1 6 �xd5 i!fxd5 1 7
'Wtxc7 �e6 1 8 i!fg3 �;i dS 1 9 .!!!: d l
h 5 20 h4 with advantage.
13 �xd5 'Wfxd5
14 �e4 -¥t b5
15 a4
If 1 5 i!ff3? Black would have
replied 1 5 . . . �d4.
15 i!fa6 ( 79)
White has the advantage of the
pawn. In order to avoid losing the
b7 pawn Black has been forced
to put his queen on the edge of
the board. In lvanchuk-Serper
(Sochi, 1 986), which should be
regarded as the initial game in this
line, Black tried 1 5 . . . lltc5 but
after 16 � xb7 K b8 1 7 b4! lltb6
1 8 11txb6 cb he had the worse
endgame and after 1 9 �c6 +
*d8 20 §l d l + *c7 2 1 b5 §l d8
22 §l xd8 <2! xd8 23 �d5 �e6 24
�xe6 fe 25 .,t f4 + White
exploited his advantage.
Evidently, it is 1 5 . . . 'f(a6
rather than I I 'f(e3 + that should
be considered the innovation in
this game.
16 � dl
The immediate 1 6 b4 achieved
nothing after 1 6 . . . �xb4 1 7 � b I
�c5 1 8 ill!' -f3 c6 1 9 i;;;. xb7 0--0 . If
1 6 jff3 �d4! 1 7 �e3, Kasparov
advises Black not to win the
exchange ( 1 7 . . . �e2 + 1 8 * h i
�g3 + 1 9 hg ill!' Xfl + 20 * h2
with an unclear position), but
recommends 1 7 . . . 0--0--0 ! with an
extra pawn and excellent chances.
The variations occurring after 1 6
illt f3 � b8 ( 1 6 . . . �c5? 1 7 b4
.axe4 1 8 -¥t xe4 + �e7 1 9 � e l
i!i'e6 20 �g5), will be mentioned
in the annotations to the other
games. The continuation 1 6 �d3
'l!l!t b6 was not dangerous for
Black.
16 .,te7
This game has been studied by
many theoreticians (Dlugy, Mak­
arichev, Nunn, Timman and

others), but I believe that it is
impossible to achieve an exhaus­
tive analysis without a computer.
Interesting variations occur
after 1 6 . . . �c5 1 7 'l'tf3 c6 I S
X d7!? (80)
80
B
It is bad to accept the sacrificed
rook: I S . . . 'llr xd7? 1 9 'l'txf7 +
�e7 20 J, f5 c5 (also inferior is 20
. . . "t!te2 2 1 J,e3 or 20 . . . ..-c4 2 1
�g5 K heS 22 K d l + 'llr c7 23
J.. xe7) 21 J.. g5 'itd6 22 J,xe6 +
'l't xe6 23 K d l + .
Black needs coolly to complete
his development: l S . . 0--0 1 9
. the Zwischenzug l S . . . X adS.
J.. d 3 ( 1 9 b4 J.. d4) 1 9 . . . 11tb6 (but
not 1 9 . . . "t!ta5 20 �d2! 11tb6 2 1
b4 J.. d4 2 2 a 5 or 1 9 . . . b 5 20 b4
J.. xb4 2 1 "t!te4) 20 a5 1!tb4 2 1 J.. d2
1!th4 (bad is 2 1 . . . "t!txb2 22 J,c3
1!tb3 23 1!th5 g6 24 J,xg6) 22 g3
1!tf6 (22 . . . 1!th3 23 J.. f5) 23
11txf6 gf and Black has no prob­
lems. 24 X xb7 K fdS 25 X b3
�d4 26 K c3 J.. b4 etc is bad for
White.
17 b4 (81)
White offers another pawn sac­
rifice, but it would have been too
dangerous to accept it: 1 7 . . .
.i,xb4 l S 1!tf3 c6 ( I S . . . �c5 1 9
The Russian Game 85
X d4) 1 9 K d7 0--0 20 1!th3 g6 (20
. . . h6 2 1 J,xh6) 2 1 J.. xg6! hg 22
J.. b2 �g7 23 11th6 forcing mate.
Another way is l S J.. b2 0--0 1 9
1!th3 g6 ( 1 9 . . . �g5 20 "t!tg4 1!th6
21 J,cl ! J,e7 22 h4 winning a
piece; 1 9 . . . f5 20 J.. x f5 �g5 2 1
1!tb3 + \ll h S 2 2 K d7 and so on.)
20 X d3! h5 (20 . . . �g5 21 ..-h6
f6 22 J.. d 5 'llr hS 23 "t!txg5!) 2 1
"t!txh5! gh 22 K g3 + .
17 . . . 0-0
18 1!th3
After l S b5 or l S ..-n Black
would take over the initiative by
18 . • . g6
Not l S . . . h6 19 ..-fs g6 20
"t!te5 or 20 1!tf3 threatening J.. xb7
and J.. xh6.
19 J.. b2
No better were: 1 9 b5 .1. adS! 20
l, e l (20 J.. h6 X xd l + 2 1 X xd l
"t!txa4) 20 . . . 11tb6 o r 1 9 "t!tc3
�g5! 20 J,xg5 (20 J.. b2 J.. f6!) 20
. . . J.. xg5 21 "t!txc7 (2 1 a5 11te6) 2 1
. . . X abS and White would only
have a symbolic positional advan­
tage. 1 9 . . . �g5 in the second
variation was Black's only
defence. In Aseev-lvanchuk
(Irkutsk, 1 9S6) Black tried to play
86 The Russian Game

more sharply by 1 9 'ft'c3 c5!? but 11 .idS

after 20 .,tb2 �d4 2 1 .td3 'ft'b6
22 a5 'ft'c7 23 be _txc5 24 ..t fl
X fdS 25 K a4 he was subjected to
a deadly pin: if 25 . . . .1. acS then
26 X axd4 etc.
19 'ft'c4!
The queen emerges at last, and
Black's torments are over. Black
could not play 1 9 . . . � f4 20 'ft'e3!
'ft'e2 (20 . . . ..tg5 21 \lrh l ) 21 'ft'd4
f6 22 .,tf3! The pawn was still
taboo: 1 9 . . . _txb4 20 X d3! with
the deadly threat of 2 1 'ft'xh7 +
\lr xh7 22 X h3 + .
10 K d7
White could win back a pawn
by 20 ..t xb7, but after 20 . . .
J;� adS he would have had
nothing. Of course, Black should
have also considered 20 .td5
when playing 19 . . . 'ltc4. But he
had an important tempo move 20
. . . 'ltc2 here, which gives him at
least equal play: 2 1 _te5 (2 1
� ab l g adS) 2 1 . . . � adS 22
� ac t (22 J;� d3 .,tg5 23 K ad l c6
24 -.xe6 J;l xd5!; 22 .l;l dc l i!td2!
23 _txe6 fe 24 ;ltxe6 + II f7) 22
. . . •xc l ! 23 � xc l l;l xd5.
lO M ae8!
A forced move but good
enough to dampen White's offen­
sive. As usual, the b4 pawn was
taboo: 20 . . . i/txb4 (20 . . . •xe4
2 1 *c3 f6 22 l;l xe7 or 2 1 . . . �d4
22 l;l e l ! 11tf4 23 X xd4 11tf6 24
K xe7 11txe7 25 X e4! with a win­
ning position) 2 1 _txg6 � g5 (2 1
. . . fg 22 ttxe6 + X f7 23 .t f6!
.Iil eS 24 J;� e l ) 22 .txh7 + � xh7
23 _ta3! tth4 24 .txe7 ttxh3 25
gh J;� fcS 26 J;� a3.
Now it is White who has to be
accurate. After 2 1 .t xb7 'ft'xb4 22
.ta3 ttxa4 23 X xe7 X xe7 24
'ft'c3 c5! 25 .t b2 �d4 26 X xa4
�e2 + 27 * fl �xc3 2S .t xc3
.11 xb7 he is left with nothing.
11 'ft'xb4
11 .tc3
If 22 .t a3 then 22 . . . 'ft'd4, and
after 22 .te5 Black can play 22 . . .
.tf6! 23 .t xf6 � f4.
ll �f4! (82)
82
w
Now the roles have been
reversed due to Black's threats of
. . . 11txc3 and . . . �e2 + . White's
attack has run out of steam and
after numerous exchanges Black
achieves a better endgame. How­
ever, a draw is inevitable.
l3 .t xb4 � xh3 +
l4 gh .t xb4
15 .l;l xc7 b6
Perhaps Black should have
tried 25 . . . K e5 transferring the
rook to f5 with pressure on f2.
16 K xa7 \lr g7
17 X d7 K d8
If 27 . . . g e5 White could
defend by 2S g a2! K f5 29 K b2
J.. c5 30 K c2! (intending a4-a5)
30 . . . K cS 31 _te6!
 The Russian Game 87

28 K xd8 1t xd8 posed into the 28th game of my

29 X d1 J,.d6 first match with Kasparov which
30 K d3 bS continued 1 2 '$'h3 �xd4 1 3 e,c3
31 •n X d7 '1td7 14 '$' xd7 + * xd7, with a
32 .gl j_c5 quick draw.
33 *n h4 11 J,.e7
34 J.. c4 X e7 1l i.b5 '$'d6!
35 K fJ i.d6 The pawn capture 1 2 . . . '$'xd4
36 'IJgl .K c7 was too risky, because of 1 3 �c3
37 J,.b3 f5 with a strong initiative
38 l. d3 i.c5 13 �c3
39 K c3 '1J f6 Of course, 1 3 X e l 0-Q 14 i_ xc6
40 A c4 g5 be 1 5 '$'xe7 was bad because of 1 5
41 K c2 '1Je5 . . . Jl fe8 winning.
42 J,.c4. Draw. 13 H
14 J,. xc6 be (83)

Timman-Yusupov
Tilburg, 1986

1 e4 e5
2 �fJ �f6
3 �xe5 d6
4 �fJ e, xe4
5 d4 ciS
6 J,.d3 �c6
7 H J,.g4
8 c4 �f6
9 cd 15 J,.e3
9 e,c3 J,. xf3 1 0 ttxf3 � xd4 White supports the d4 pawn
led, as we saw, to a sharp tactical hoping to take advantage of the
skirmish. The exchange on d5 can doubled pawns on the c file. We
be regarded as a positional solu­ shall return to this position after
tion. I will look at some more the game.
examples on 9 �c3 in the notes to 15 �ciS
the next game. 16 X acl .K fe8
9 Plaskett twice occupied the
10 tt xfJ other semi-open file. 1 6 . . . l, ab8
1 1 tt e2 + 1 7 b3 � xe3 1 8 '$'xe3 Jl fd8 1 9
This is an innovation, specially K fd l c5 ( 1 9 . . . J,. fll 20 g3 with
prepared by Yusupov for the strong pressure on Black's posi­
quarter-final candidates' match. tion, Mestel-Piaskett, London,
I I X e l + i_e7 would have trans- 1 986; 19 . . . i. f6 20 �e4 i_ xd4 2 1
88 The Russian Game
-E}xd6 �xe3 22 fe X xd6 23 II xd6
cd 24 X xc6 l:l d8 25 .1 c7 with
good winning chances in the rook
ending) 20 d5 �ffi 21 -E}e4 and
White had a clear advantage
(Short-Piaskett, London, 1 986).
17 <2l xd5!
This is an unusual solution.
Black is given an opportunity to
undouble the pawns, but after 1 7
. . . cd 1 8 l';c2 followed by 1 9 l';c6
he would have no counterplay.
But White's unexpected move
is met by Black's unexpected
reply . . .
17 litxd5!
18 l';c4 �d6
19 b3 (84)
If 1 9 l';xc6 then 1 9 . . . tf' xa2.
84
B
19 K e6
20 g3
White could not yet take the
pawn-20 'Wf'xc6? �xh2 + .
20 'Wt'f5
The International Master M.
Dvoretzky compared this purely
positional pawn sacrifice with the
pawn sacrifice in the Marshall
Gambit. A small material gain for
White is well compensated by the
activity of Black's pieces. After 20
. . . X ae8 White would probably
not have taken the pawn but
simply played 2 1 -.xd5 cd 22
X c6. 20 . . . "Wth5 was also worth
considering.
21 itc2
After 2 1 'iit xc6 K ae8 22 d5 X e4
23 �c5 h5 Black would have had
sufficient compensation for the
pawn. Timman drives the hostile
queen off the convenient f5 square
with a Zwischenzug.
21 tlt h5
22 -.xc6 X ae8
23 -.g2 X e4
24 X fel i!tf5
24 . . . � b4 was useless because
of 25 .,td2! taking advantage of
the weakness of the 8th rank.
However, Black should have pre­
vented the rook exchange planned
by Timman by 24 . . . i!fa5! 25 a4
h5 or 25 X e2 i!tf5 followed by . . .
h7-h5. Then it would have been
White who needed to play accu­
rately to avoid trouble.
25 J.. d2 h5
26 f3 X xel +
27 X xel K xel +
28 J.. xel i!td3
29 -.n J.. aJ!
The extra pawn is almost irrele-
vant here as well.
30 *g2
31 'Wt'fl
32 i!fe2
More dangerous for Black was
32 �f2 J.e3 (32 . . . -.b2 33
"Wtb5!) 33 i!fe2 J.. xf2 34 'iit x f2
forcing transposition into a pawn
ending- 34 . . . -.b6 35 'iit xb6 ab.
But, as was established by Tim­
man's second, grandmaster U.

Andersson, White could not win
this ending either. 36 * f2 *f8 37
*e3 *e7 38 *f4 *d6!
32 t�tb2
23 l!;xb2 � xb2
34 g4 g6
But not 34 . . . hg 35 fg.
35 gh gh
The weakness of Black's pawns
on the queenside guarantees
White some advantage in the
bishop ending, although it should
not be enough for a win. Timman
made it very easy for his opponent
by allowing the exchange of
another pair of pawns.
36 *g3 �e5 +
37 *h4 i_ xh2
38 * xh5 �d6
39 *g5 *f8
40 *f5 *e8
41 �c3 c6
42 �e5 �e7
43 �d4. Draw.
Thus, White's interesting posi­
tional plan, involving the dou­
bling of his opponent's pawns on
the c file and the exchange on d5,
only led to a draw in this game.
But an improvement was pre­
pared by the young English
grandmaster Short. Let's return
to the position in diagram 83.
Short-Oiafsson, Reykjavik,
1987. 15 K dl! (this seems to be a
better way to defend the d pawn
than 1 5 J.. e3) 15 . . . K fe8 16 �f3
<EJd5 17 �xd5 l!;xd5 18 •xd5.
White is happy to exchange
queens this time. 18 . . . cd 19 J.. f4
c6? (85) . Strange though it may
seem, this is almost a decisive mis­
take. This passive move puts
The Russian Game 89
Black into a critical situation.
Black should have tried the active
1 9 . . . J.. f6 20 JLxc7 Ji e4 or 20 . . .
X ac8 2 1 K ac l x e4 or even 1 9 . . .
.,td6. Now Short, despite his
youth, demonstrated a high level
of technique in exploiting his
positional advantage.
20 J;� acl J;� ac8 21 * " f6 22
.1;1 d3 * f7 23 f3 �f8 24 I. b3 h5 25
h4 K e6 26 a4 *e8 27 K el
K xel + 28 fi xe1 c5 29 de J.. xc5
30 X b7 i.d4 31 fl d2 �b6 32 b4!
K c4 33 �d6 g5 34 a5 i.g1 35 hg
fg 36 b5 X a4 37 a6 l, a2 + 38
fid3 *d8 39 i. b8 X a3 + 40 *cl
X aS? 41 � c7 + . Black resigned.
Kupreichik-Yusupov
Minsk, 1987
54th USSR Championship
1 e4 e5
2 .ao �r6
3 �xeS d6
4 �f3 €l xe4
5 d4 d5
6 i.d3 �c6
7 0-0 i. g4
90 The Russian Game
8 c4 �f6
9 �c3
So we return to the sharp and
exciting variation introducing the
pawn sacrifice. It could be said
that this game, played in the 54th
USSR championship, is the latest
contribution to the theory and
practice of this variation today.
9 � xf3
10 ttxf3 � xd4
Before demonstrating Kuprei­
chik's innovation I want to look
at the information which had
been known before this game.
After l l � e l + �e7 1 2 -. d l
we again transpose into the 1 5th
game of my second match with
Kasparov, which ended in a draw,
and after 1 2 • g3 de 1 3 il. xc4 0--0
1 4 �g5 Ad6 into Lobron­
Karpov, where Black didn't have
any difficulties either. The immed­
iate I I iii' g3 was dubious because
of I I . . . �e6 and Black could
win time by . . . Ad6.
The surprise 11 *e3 + as you
, 20 �xb7 (if 20 � xe7 then 20 . . .
remember, was given to me by
Kasparov in the 6th game of our
third match and, after 1 1 . . .�e6
12 cd � xdS 13 � xdS iitxdS 14
�e4 't!tbS 1S a4 iita6 16 g d1 �e7
1 7 b4!? 0--0 1 8 'tit h3 g6 Black
repelled White's attack and
"demanded" a draw from a posi­
tion of strength.
An attempt to improve this var­
iation for White was made in Tim­
man-Yusupov, Hilversum, match,
1986. White played 17 •f3 in­
stead of 1 7 b4. The game didn't
last long. 17 • . . K d8 18 �dJ (the
exchange on d8 was unpromising)
18 . • . -.as! 1 9 tt'xb7 now lost to
1 9 . . . .a xd3!, and so the one
square queen advance has relieved
Black from any problems. 19 �d2
�b4 20 �e3 (20 �b5 + c6 2 1
�xc6 + be 22 'Wfxc6 + t1P f8 23
�xb4 + tt'xb4 24 K xd8 + <21 xd8
25 tt'd5 ttb8! or 25 ttd7 ttb7 26
K d l ttxd7 27 K xd7 t�Pe8 28
X xa7 h5 and Black has an advan­
tage) 20 0-0 21 ..-xb7 �cS 22
• • •
�xeS �xeS 23 tt bS. Draw.
Black managed to obtain a
draw in that game, but it was not
the end of the story. In Howell­
lvanchuk, Groningen, 1986-87
Black found a way to gain the
initiative: 17 g b8 (instead of
. • •
1 7 . . . � d8). The pawn is
defended and White has to start
thinking how to win back his
material. 18 b4. This is Yusupov's
recommendation, but Ivanchuk
proved that this pawn advance in
this situation gains nothing for
White. 18 0-0 19 � d7 � bd8!
• . .
*d6 2 1 �a3 • xe7 22 b5 �c5 23
�xh7 + * xh7 24 -. hs + *g8 25
�xeS *f6! with an advantage for
Black) 20 'ilitc4 21 � xd8 � xd8
. . •
22 �e3 •xb4 23 � e4 (after 23
�xa7 c5 the bishop would have
ended up in an unenviable posi­
tion) 23 . �cS 24 � xeS •xeS
. .
1S h4 *d4 2S � e1 •xa4. Ivan­
chuk carried out the technical side
of the game simply and convinc­
ingly. 27 hS � gS 28 -.rs �xe4 29
� xe4 *d1 + 30 * h2 g6! 31 hg hg
32 ._gs lil dS 33 ._r6 'i!lthS + 34
� h4 *xh4 + ! White resigned.
So it is possible to draw the

conclusion that neither 1 1 tte3 +
nor 1 1 .l. e l + is dangerous for
Black. Kupreichik makes one
more attempt to prove that the
pawn sacrifice is quite sound, and
he does so against one of the best
specialists of the Russian Game.
1 1 tt h3!! (81)
86
B
Black has saved time by not
playing . . . J.e7 but White still
used to play K e l + .
11 de
Better than 1 1 . . . c6 1 2 X e l +
J.e7 1 3 J.g5 �e6 14 J. xf6 J. xf6
1 5 cd cd 1 6 J. f5 or 1 3 . . . de 1 4
J.xf6 g f 1 5 J.xc4 with more than
enough compensation for a pawn.
1 1 . . . J.e7 1 2 J.g5 de 1 3 J. xc4
would have transposed to the
position in the main game, and 1 2
K e l de 1 3 J.xc4 0-0 would have
led to a sharp position in which it
is difficult to give preference to
either side. However, White can
improve by 1 2 cd � xd5 1 3 .l. el
�e6 14 J.g6! �df4 1 5 J.xf4
� xf4 1 6 J.xf7 + 'lr xf7 1 7 'ltf5 +
when White wins back a pawn
and maintains the initiative.
12 J. xc4 J.e7
Better than exchanging queens
The Russian Game 91
e.g. 1 2 . . . 'ltd7 1 3 X e l + J.e7
( 1 3 . . . 'lidS 14 'ltxd7 + 'lr xd7 1 5
X d l c5 1 6 J. xf7 winning back a
pawn and maintaining the two
bishop advantage) 1 4 'ltxd7 +
� xd7 1 5 �d5 �e6 1 6 X xe6! and
White takes the upper hand.
13 J.g5 0-0
14 x adl c5
15 X fel (87)
87
B
1 5 . . . X e8 would now be dan­
gerous for Black after 1 6 J. b5 ( 1 6
J.d3 'ltc8 1 7 'lth4 h6 1 8 J. xh6
� f5) 16 . . . � xb5 1 7 & xd8 J. xd8
1 8 X xe8 + �xeS 1 9 J.xd8 Jl xd8
20 'lr fl and White has an advan­
tage due to his mobile queen.
15 h6
16 j_ xh6!?
The bishop sacrifice is tempt­
ing, but 1 6 X xe7! may be better,
e.g. 1 6 . . . hg ( 1 6 . . . 'ltxe7 1 7
�d5 etc) 1 7 X xb7 with advan­
tage to White.
16
17 tt xh6
18 K d3
19 'lth5
After 1 9 K g3 � f5 20 ttg6 +
<2)g7 it wouldn't be easy to besiege
Black's fortress.
92 The Russian Game

19 .f6 be the decisive mistake and that

the correct one was 2 1 . . . 'II g7!
. . .

20 X g3
20 J1 h3! *'g7 2 1 f4 �xf4 22 The game was very tense and
�d5 �g5 23 �e7 + � xe7 24 mutual inaccuracies are bound to
X xe7 looks more promising for happen in such a sharp struggle.
White. 22 b4 X xe4
20 K ae8 No better were 22 . . . *'h6 (g6)
This time Black could have 23 J;l xg5 or 22 . . . � xh4 23 � xf6
played the more accurate 20 . . . x xe l + 24 * h2.
� f5. After 2 1 �e4 *'g7 22 K g4 23 X xe4 �f4
(22 X xg5 � xg5 23 � xg5 �d6!) 24 X g4 �b6
22 . . . �h6! 23 X g3 � f5 and if 24 . . . �d6 25 ,l xd4 cd 26 �d3
the opponents wished the game didn't help Black. He has many
could end with a repetition of pieces, but they are all scattered.
moves. 25 •xeS �c6
21 �e4 *b8 (88) 26 .hS .d6
27 J,.xf7 *' d1 +
28 *h2 ttd6 +
29 f4 *'c7
30 �b3 .d6
31 l;l e8 � xf4 +
32 g3 J,. xg3 +
33 *h3
Black lost on time but there was
no way to avoid mate.
I believe that there are still a lot
of secrets in this variation not
Yusupov considers this move to discovered even in this game!
3 The Scotch Game
Timman-Karpov
London, 1984
There are three main grand­
master games and more than ten
in the commentaries on them-I
didn't expect myself that the an­
tique Scotch Game, played today,
could amass so much material.
The material is quite rich and
should give the reader an oppor­
tunity to invent some original
lines which could serve as opening
surprises. But, it should be admit­
ted, that there have been no major
turns in the development of this
opening in the last few years. On
the whole, the Scotch Game has
maintained the reputation of a
reliable opening for Black.
My encounters with the Dutch
grandmaster often become
theoretical disputes. This time,
Timman decided to surprise me
from the very start by choosing a
rare opening.
I e4 e5
l �fJ �c6
3 d4 ed
4 � xd4 �f6
5 � xc6 be
6 e5
6 i_d3 is more common but the
e pawn advance is also well
known.
6 tte7
Timman is, perhaps, one of the
few modem grandmasters who
adopt this ancient opening from
time to time. The same year in
Bugojno he won an interesting
game against Smyslov in the var­
iation 4 . . . i_c5 5 �f5! and this is
covered later.
Korchnoi played 6 . . . <2ie4 in
this position instead of 6 . . . f!ie7
against Timman in the same year,
and played a new move. Timman
replied with the innovation 7 i_e3
instead of the usual 7 f!/0. The
game continued 7 . . . d5 8 ed cd 9
i_d3 � f6 1 0 0-0 i.e? I I c4 0-0
1 2 �c3 i_e6 1 3 K e l d5 1 4 cd
o2) xd5 1 5 � xd5 i_ xd5 with a level
position (Sarajevo, 1 984).
7 f!iel o2ld5
8 c4 i_a6
9 f!/e4
9 �d2 was not dangerous for
Black after 9 . . . -21 b4 (but not 9
. . . � f4 1 0 f!/e4 o21g6 I I f4 0-0--0
1 2 g3 d5 1 3 cd i_ xfl 1 4 l. xfl cd
1 5 f!ia4 with a better game for
White) 1 0 � n c5 I I a3 �c6.
9 �b6
Here I could have offered a
symbolic draw by playing 9 . . .
� f6 1 0 f!/e2 �d5 but, as I re­
member, I was in a fighting mood
that evening. 9 . . . f5 was also
worth considering.
10 � dl
This is an innovation (opening
reference books mentioned only
10 �c3 with equality), but it is
not quite successful.
10 0-6-0
(89)
94 The Scotch Game

an advantage by 1 5 . . . ttxe5 1 6
'f!txe5 fe 1 7 0--0-0 d 5 and i f 1 8
�a5 + then 1 8 'll b6 1 9 b4
J.. b 5!
15 fe
16 fe K e8!
17 J.f4 i!th4 + !
This is much stronger than 1 7
. . . it b4 + 1 8 ttxb4 J.. x b4 + 1 9
* d 1 II h f8 and Black would be
l l c5 only slightly better.
White opens up the position 18 g3
believing that Black's king will be 19 X cl 'll a8!
unsafe, but, as it turned out, his 20 h4 d5!
own king was even more insecure. Keene recommended 20
Curiously enough, only a year �d6?! 2 1 X c5 j_xe5 22 J.. xe5 (22
later the Dutch grandmaster X xe5 K xe5 23 .t xe5 X eS) 22 . . .
chose this opening against me d5 23 g4 x xe5 24 gf x xe4 + 25
again and this position arose from *d 1 . Black's position would be
another move order 9 . . . � f6 1 0 preferable here as well, but my
'f!te2 �d5 I I �d2 0--0-0 1 2 'f!te4 move was more energetic.
� b6. Timman now advanced 11 'f!te3
another pawn by 1 3 a4, and I If 2 1 'f!tc2 either 2 1 . . . c5 or 2 1
played a new move - 1 3 . . . d5! . . . .1. xe5 2 2 ..l xe5 ttxe5 + 23
After 1 4 cd �xfl 1 5 d6 J;l xd6! 1 6 .f2 (23 . d l i!th5 + !) 23
� xfl X e6 1 7 f4 g 5 White got into ..ld6 was good for Black.
difficulties again. However, hav­ 11 g5!
ing obtained a significant advan­ 11 �xg5 .t b4 +
tage, I let it slip in time trouble 13 * fl X hf8 +
and the game ended in a draw. In 14 .g1 (90)
order to give a full picture, this
game is covered below as well.
ll J.xfl
11 cb � a6
13 ba
The initiative has already
passed to Black's, but perhaps it
was better to capture another
pawn by 1 3 be!?
13
14 �b3
15 f4 14 K. xe5!
After 1 5 J.. d2 Black would gain My opponent seemed a bit sur-
 The Scotch Game 95

prised when I sacrificed the rook. h4) 1 4 -.e3 � xg2 + ( 1 4 . . . ttxg2

He thought the worst was already 1 5 ttxa7) 1 5 � xg2 ttxg2 1 4 0-0-
over. 0 c5! with a complicated struggle.
25 'i!txeS II 0-0-0
26 ., h2 In 011-Kalinin (USSR, 1 986)
White resigned the knight retreated and after I I
After 27 * h3 the sly move 1 9 . . . � b4 1 2 � f3 c5 1 3 a3 �c6 1 4
. . . fla8, which opened a path for � d2 0-0-0! 1 3 �c3 Black carried
the bishop, now proves to be use­ out the same idea as in the main
ful: 27 . . . �c8 + 28 g4 K f3 + . game- 1 3 . . . d5! The queen
exchange 1 4 ed X xd6! 1 5 'i!txe7
� xe7 1 6 �e2 K e6 led to approxi­
mate equality. I managed to gain
Timman-Karpov more.
Amsterdam, 1985 12 'lte4 �b6
13 a4?! d5!
I e4 e5 White had prepared a surprise
2 �f3 �c6 (the a pawn advanced instead of
3 d4 ed the c pawn), but stumbled into an
4 � xd4 �(6 innovation himself. This coun­
s �xc6 be terthrust in the centre is very pro­
6 e5 'i!te7 mising. The restrained 1 3 . . . fl b7
7 'i!te2 �d5 in Handoko-Kovacevic (Zagreb,
8 c4 � ·6 Rijeka, 1 985) led after 1 4 a5 �c8
9 'i!te4 � (6 1 5 b4! fi bS 1 6 �e2 f6 1 7 f4 fe 1 8
10 'i!te2 �d5 fe K e8 1 9 � f3 ._e6 20 c 5 �xe2
II �d2 2 1 'i!txe2 d6 22 J;� b l de 23 tta6
So, White refused to repeat ._g6 24 K b2 �e7 25 0-0 g hf8 26
moves and the position known to � xe5 K xfl + 27 •xfl tte6 28
us from the previous game arose � f4 to a clear advantage for
again. I I b3 has also been played White. But now White didn't have
(without the repetition ttt e2 - time to drive the knight from the
e4-e2, �d5-f6-d5). Here is the b6 square and so the a pawn ad­
most recent example: I I . . . 0--0-Q vance lost much of its point.
1 2 'ltb2 �b6 1 3 �e2 � e8 14 � f4 14 cd J_ xfl
g5 1 5 �g3 �g7 1 6 �c3 f5 1 7 f4 gf 15 d6 ». xd6!
18 �xf4 �xe5 1 9 J_xe5 -.xe5 20 16 � xfl
0-0 (20 0--0-{) d5!) 20 0 0'Wt'd4
0 Of course the rook could not be
with advantage for Black (Ljubo­ taken: 1 6 ed iltxe4 + 1 7 �xe4
jevic-Seirawan, Wijk aan Zee, J.. xg2; other captures of the
1 986). Hort suggested 1 2 J.. b2 (in­ knight were no better: 1 6 'lll' xfl
stead of 12 iitb2). 12 . . . 'Wt'g5 1 3 ». d5, 16 l;l xfl g d5 preparing . . .
�d2 � f4 (better than 1 3 . . . f6 1 4 f7-f6.
96 The Scotch Game

16 • • . K e6! (91) 23 K g8 +
24 flh1 K g4
25 <2\bJ
91
w Timman claimed after the game
that 25 �c4 would have been
better but here again 25 . . . f6
(Black would gain no advantage
by 25 . . . �xf4 26 �e3 K h4 27
<21f5 or 25 . . . K xf4 26 .l, xf4
<21 xf4 27 K fl �d3 28 K xf7) 26 f5
g e8 27 b3 fe 28 f6 e4 29 f7 X f8 30
�e5 K g5 3 1 �xc6 fld7 would
maintain Black's clear advantage.
The e pawn should be attacked 25 • . • <2\ xf4
from the front, for if 1 6 . . . K d5 After 25 . . . K xf4 26 X xf4
1 7 f4 g5 ( 1 7 . . . itd7 1 8 �e3) <21 xf4 27 X fl �d3 28 X xf7 X xe5
White would gain an important Black would have an advantage,
tempo by 1 8 �e3. but White could defend by 27
17 f4 g5 �d4 .M xe5 28 �xc6 with a dou­
18 gJ itb4 + ble attack on the rook and the a7
The queen exchange is the sim­ pawn.
plest way to consolidate Black's 26 Jil ae1
advantage. 1 8 . . . gf 19 gf f6 (not 26 �d4 X eg6 was bad for
1 9 . . . ith4 + 20 �g3 -2:ld5 2 1 White.
�d2 �h6 22 0--0 K g8 23 K a3) 20 26 • • • K eg6
� g3 fe 2 1 f5 led to sharp play. Of course it was tempting to try
19 ttxb4 to mate by 26 . . . g h6, X g2 and
Not 1 9 <21d2 gf 20 gf X g8. K hxh2 mate, but White can
19 • • • �xb4 + defend by 27 X e3 X g2 (27 . . .
20 �dl X gh4 28 X f2) 28 h3 <21d5 29
Black would also stand better X d3 .
after 20 * f2 f6 2 1 ef X xf6 22 27 gO h5
*g2 X eS! or 20 * d l gf 2 1 gf f6. 28 �d4
20 �xdl + Not 28 h3 K g3 29 K xf4 X xh3
21 �xd2 gf mate. 29 X efl �e6 was also bad.
22 gf �d5 28 • . . �d5
After 22 . . . f6 White has the Better than 28 . . . h4 29 <21 f5
interesting continuation 23 �e4 threatening 30 �e7 + or 30 �e3.
fe 24 f5! 29 X d1 X e4
23 H 30 �bJ
I think that White shouldn't If 30 <21 xc6 I intended to reply
have withdrawn the king from the 30 . . . .M xc6 3 1 X xd5 X c2, but if
centre: 23 .ll fl was safer. 30 X xf7 then 30 . . . ee3.
 The Scotch Game 97

30 X el g xa4 would have allowed 48 h6

31 K gl g e4 49 g ffi etc.
White defends by the only 48 h6 X b8
move: if 3 1 X xf7 then 3 1 49 X h5
X gg2 was decisive. This time it is White who plays
31 X xe5 inaccurately. I was more afraid of
By now I was running short of 49 h7 X h8 50 X f7 threatening
time. 3 1 . . . X xg l + 32 * xg l � f3 -e5 -g6.
X xb2 33 'E)c5 X c2! 34 e6 f6 was 49 X h8
much stronger. 50 h7 *b4
32 X xf7 .!I xgl + 51 � h6 � f6!
33 * xgl � b4 Leading to complete equality.
Possibly 33 . . . � e l + 34 *g2 5 1 . . . *xa4? was bad because of
(34 :!HI � xfl + 35 * xfl �e3 + 52 �g5 and -E:)f7.
36 *e2 �c4) 34 . . . � bl 35 :ac5 52 � xf6 J;� xh7
!;l xb2 + 36 *g3 *d8 was even 53 � f4 + * aJ
stronger. 54 t�rel � d7
34 !;t h7 �dJ 55 -E:)dl g d8
This is another missed oppor­ 56 � c4 � b8
tunity in time trouble. After 34 . . . 57 � cl � b4.
tlr b7 it is unlikely that White Draw.
could have escaped.
35 'E)d4 *b7
36 bJ a5 Timman-Smyslov
36 . . . a6! was more accurate. Bugojno, 1984
37 X h6 � b4
38 *fl K e4 I e4 e5
39 �e6 tlrb6 2 �f3 �c6
40 tlrf3 g el 3 d4 ed
41 h4 �d5 4 -E:) xd4 J,.c5
42 �d4 X eJ + Together with 4 . . . :a f6 and 4
43 *fl K c3 . . . ith4 this bishop move is one
44 X xh5 K dJ of the most popular continuations
44 . . . *c5 45 �e6 + * b4 was in this position.
worth considering. 5 �f5
45 :af3 X xbJ Now White has several options.
46 .l. f5 X b4? Apart from 5 � f5 he can play 5
This move completely loses the J,.e3, 5 -E:) b3 and 5 .;axc6. None of
initiative which could have been these moves promises White
maintained by 46 . . . \IreS. much. Here are fairly recent ex­
47 b4 'ili'c5 amples of each of them.
6:
Now Black has to play accu­ Sveshnikov-Geller, Sochi, 197
�g e7 7 g3 . Th e
rately to maintain equality: 47 . . . 5 i_ eJ 'tlt'f6 6 c3
98 The Scotch Game
other theoretical continuations 7
�c2, 7 .,tb5, 7 .,tc4 are no more
promising 7 . • . d5 8 .,tg2 de 9
�b5 ..t xe3 10 '2) xc7 + *fB 1 1
0-0 X b8 1 2 fe 'ilte5 13 'iit b3 f5 14
�b5. Draw.
5 � b3 has been more popular
in recent years, but, as before,
Black has no difficulties.
Pasman-Unzicker (Beer-
Sheva, 1 984): 5 . . . ..t b6 6 �c3
�ge7 7 a4 a6 8 *e2. This was an
innovation. 8 a5 and 8 .,tg5 had
been tried earlier. 8 . . . (}-() 9 .,tg5
d6 10 �d5 ..ta7 1 1 0--0--0 _te6 1 2
� f6 + ?! An amusing knight sacri­
fice. 1 2 . . . gf 1 3 .,txf6 h6 14 e5
* h7 1 5 ed cd 1 6 *h5 X g8 1 7
.,td3 + X g6 1 8 .,txg6 + fg 1 9
'tlre2 .,t xb3 20 cb *d7 2 1 X be l
X eS! 22 .,tc3 d5 and there was no
trace of White's attack.
Prandstetter-Hort (Czecho-
slovakia championship, 1 984): 5
• . ..,tb6 6 a4 a6 7�c3 �f6 8 g3.
This was another innovation. 8
.,te2 and 8 .,tg5 had brought
White no success. 8 • • •d6 9 .,tg2
(White should have played the
preventive 9 h3) 9 • • • .,tg4! 10
_tO ( 1 0 f3 .,td7 and White's king
would be stuck in the centre) 10
• . .�e5! 1 1 .,t xg4 �fxg4 12 0-0
b5! 13 �d5 _ta7 14 b4 c6 15 �f4
g5 16 '2) b3 gb. Black transforms
his advantage into a victory by
attacking energetically. 17 .,tg5
*d7 18 .,t xb4 '2)g6 19 'i!t'O '2)4e5
20 ..-r5 �xb4 21 'i!t'xd7 + * xd7
22 gh K ag8 + 23 *b1 X g4 24 0
X xb4 25 * h2 X g8 26 X ad1
.,tg1 + ! White resigned, because
he couldn't prevent mate.
H iibner-Spassky (Linares,
1 985): 5 . • ..,t b4 + 6 c3 ..te7 7 f4
d6 8 .,tdJ .,tb4 + . The ubiquitous
bishop prompts pawns to advance
on both sides . . . 9 g3 .,t f6 10 .,te3
h5 1 1 b3 g5 12 *d2 gf 13 gf � ge7
14 � a3 _tb4 + 15 ..tfl � g6 16
*d1. Draw.
And the last example concerns
5 � xc6- 5 . . . *f6 6 f4 ..-xc6 7
�c3 � f6 8 .,t b5 *e6 9 *e2 (}-()
1 0 e5 �d5 1 1 �e4 .,te7 1 2 (}-() f5
1 3 ef �xf6 1 4 � xf6. All this was
played in Larsen-Brinck-Claus­
sen (Copenhagen, 1 979). The
Danish grandmaster here recom­
mended 1 4 . . . ..-xf6 1 5 *c4 d5!
1 6 'i!t'xd5 + .,te6 with sufficient
compensation for the pawn .
5 g6
In a later game Timman-Bohm
(Holland, 1 985), Black played a
new move 5 . . . d6. Of course, you
can't call this a real innovation,
because this pawn sacrifice had
already been recommended by the
first world champion W. Steinitz.
He gave the following variation: 5
. . . d5 6 � xg7 + * f8 7 �h5 *h4
8 '2)g3 � f6 9 �e2 �e5 10 h3
il g8 with a better game for Black.
I don't know why Black
rejected this manoeuvre in the re­
cent games referred to. Maybe he
was always apprehensive of some
possible tricks.
So Timman-Bohm continued
as follows: 5 . . . d6 6 � xg7 + (it
was possible to play safe by
retreating the knight to e3 or g3,
but in this case White couldn't
expect much) 6 . . . *f8 7 '2)h5 (7
� f5 .,t xf5 8 ef 'i/lth4 and then . . .
 The Scotch Game 99

K e8 with the initiative for Black) 1984: 12 a6 13 _tdl -2J d4 14

• • •

7 . . . *h4 8 -2Jg3 � f6 9 _te2 _tdJ c6 15 f4 �b5. Draw.

�eS. Now compare this position 13 _tc4
with the position which would White takes aim at the r7
happen as a result of Steinitz's square, but after 1 3 . . . .te6 it is
ancient recommendation. The dif­ unlikely that he could take advan­
ference is that in the modern game tage of it.
the pawn was on dS rather than 13 aS? (92)
d6. But in both instances Black
held the trumps.
92
10 b4. Curiously enough in the w
same year Timman played the
Scotch Game against Handoko
(Zagreb, Rijeka, 1 98S), as Black
this time, and after 1 0 f3 �eg4 1 1
fg -2J xe4 1 2 *dS! fS 1 3 II fl � f6
14 I. xfS White obtained an ad­
vantage. Black should have
played 1 0 . . . .M g8 or 1 0 . . . �hS
and there would be nothing to be
afraid of. 1 0 . . . _txb4 + 1 1 c3 14 -2J g4!
_tcS 1 2 _ta3 �e4 1 3 _txcS � xeS. White secures the initiative by
Black's position was favourable this unexpected move. The point
even though the game ended in a is that the dark-squared bishop is
draw after a long endgame. in the wrong place. The impres­
6 -2Je3 -2Jf6 sion is that there is a misprint in
7 -2Jc3 0-0 the diagram and the bishop
8 .tdJ J;� e8 should be on g7 . . .
9 0-0 -2Je5 14 _te6
10 *h1 d6 Black didn't like opening the f
1 1 �el · � c6 file: 1 4 . . . �xg4 1 S fg _te6 ( I S . . .
In Handoko-Smejkal (Zagreb, �e6 1 6 *f3) 1 6 �dS.
Rijeka, 1 98S) which was played 15 _t gS
later, Black played more accu­ The ex-world champion's dark
rately: 1 1 . . . �ed7! and after 1 2 squares are gaping and so he has
f3 a6 1 3 a 3 -2J ffi 14 b4 _ta7 I S to surrender his queen.
_tc4 c6 1 6 *d3 _te6 1 7 .t xe6 15 �xg4
� xe6, the chances were equal. 16 _txd8 <2)e3
12 f3 �d4 17 *cl .txc4
1 2 . . . a6 was safer. Timman­ 18 * xe3 <2) xcl
Giigoric finished quickly in the 19 iitdl <2) xal
same tournament in Bugojno 20 .tf6!
100 The Scotch Game

Black has managed to level the Yes, it is difficult to play with-

material but the threat of '*h6 out the queen . . .
forces him to accept unfavourable 30 . . . J.,xb3
simplifications. If 30 . . . * g7 the strongest re-
20 J;l e6 ply would be 3 1 X xa4.
21 x xal K xf6 31 tt xf6 a3
22 b3 J.. e6 32 ttd4 cS
23 � a4! i. a3 33 ttxd6 K c8
24 tt c3 K f4 34 ttd2 J.. e6
25 g3 J., b4 35 f4 b6
26 'l!ltd3 i;l f6 36 g4! J.,xg4
27 a3 b5 37 f5! X e8
28 ab ab 38 ttxh6 gf
29 *g2! ba 32 b3.
30 tt d4! Black resigned.
4 The Italian Game
Karpov-Yusupov
Bugojno, 1986
A modest, but poisonous varia­
tion of the Italian Game involving
d2-d3 occurs from time to time
in most big tournaments. I had
already become interested in this
idea during my preparation for
the Merano match where it oc­
curred twice. Although both
games ended in a draw, White, in
the first game in particular, was
dominant. And then, five years
later, I managed to overcome
quite a dangerous opponent in the
Italian Game. In the notes, I shall,
as usual, cover the modern state
of the opening.
I e4 e5
2 �f3 <2lc6
3 J.. c4
3 J.. b5 against Yusupov auto­
matically leads to the Open Varia­
tion of the Spanish. But I am not
always in the mood to play it. The
main idea of 3 J.. c4 is to reach one
of the Spanish positions in the end
(because in the Spanish the light­
squared bishop most frequently
ends up on the a2-g8 diagonal
after J.. fl -b5-a4-b3), but
avoid any sharp variations such
as the Open.
3 �c5
If Black plays this move the
opening is usually called the Ita­
lian Game. But another order of
moves is possible - 3 . . . <2l f6 4 d3
(White rejects the classical 4 �g5)
4 . . . J.. c5 5 c3 d6 etc. In that case
the opening is called the Two
Knights' Defence. The 3 . . . � f6 4
d3 variation is comparatively new
which explains the confusion in
terminology.
4 cJ �f6
5 dJ
The theoretical continuation
here is d2-d4. The Italian Game
has been known for about five
hundred years and White's main
aim has always been to create a
pawn centre by means of c2-c3
and d2-d4. Even though it is
now known that Black can suc­
cessfully meet this plan with . . .
�g8 - f6, it seems that 5 d4 is still
objectively stronger than 5 d3. So
why then did the restrained d2-
d3-thanks to the efforts of the
Soviet grandmasters S. Dolma­
tov, and S. Makarichev-gain
substantial popularity in the 80s?
This could be explained by the
fact that the ancient continuation
d2-d4 leads to a very sharp sit­
uation on the one hand and to a
quick release of the tension and
simplifications on the other. And
in the end, the player who remem­
bers more variations happens to
be in a favourable position. After
d2-d3 stable pawn chains are
created (as in the Spanish Game)
and the mastery of the players in
positional manoeuvring becomes
more important. These, in my
102 The Italian Game
opinion, are the reasons explain­
ing why the move d2-d3, which
was practically absent from open­
ing books in the 70s, has forced
out the traditional d2-d4. I am
talking, of course, about the
grandmaster level.
Despite this it would be wrong
to assume that the classical lines
of the Italian Game have now
been played out. For example,
let's take the only "Italian" with
d2-d4 in Jnformator 42, the
most recent when this book was
written.
Szecsi-Heinrich, correspon-
dence tournament, 1986. S d4 eel 6
cd ,i.b4 + 7 <2lc3 (the main move;
the other ancient continuation is 7
,i.d2) 7 • . •<2l xe4 (7 . . . d5 leads
to a difficult position for Black. A
famous example is Steinitz-Bar­
deleben, Hastings, 1 895, which
ended with a beautiful combi­
nation with the penetration of
White's rook to the seventh rank)
8 0-0 ,i. xc3 9 dS .i_ f6 10 K el <2le7
1 1 K xe4 d6 12 ,i.gS ,i. xgS 13
<2)xgS h6 14 -.e2 hg IS K el .i_e6
16 de f6 (93) .
93
w
This position is assessed by
theory as better for Black. How­
ever, after 17 K e3! dS 18 X h3! it
became clear that Black's king
was not so impregnable. 18 • . •
K f8 (in the same correspondence
tournament, Tasnadi against
Szecsi played 1 8 . . X xh3, and
.
after 1 9 gh g6 20 -.n f5 2 1 -.c3
d4 22 -.b3 b6 23 -.b5 + * f8 24
-.e5 *g8 25 -.f6 White had a
clear advantage) 19 .i_d3 -.d6 20
-.hS + *d8 21 -.n X e8 22
tt xe8 + * xe8 23 K h8 + <2lg8 24
K xg8 + *e7 25 K xa8 -.b4 26
* " -. xb2 27 ,i.g6. Black
resigned.
As you see, innovations lie in
wait for us round every corner in
the Italian Game . . . But now it is
time to return to the modern 5 d3.
s d6
6 D-0 D-0
7 K el
The intermediate 7 b4 has been
played many times in recent years,
but I decided to leave this "side
variation" out of these commen­
taries to save space.
After 7 <21 bd2 a6 8 .i. b3 �a 7
the position from the 8th game of
the Merano match would occur.
This game continued as follows: 9
h3 �e6 1 0 ,i.c2 d5 I I l;l e l de 1 2
de <2) h5 1 3 <21 ft 'ii'f xd l 1 4 l;l xd l
l;l ad8 1 5 ,i.e3 f6 1 6 �xa7 <2)xa7
17 �e3. White achieved a small
positional advantage, which
could have been decisive. How­
ever, I didn't find the study-like
win and Black managed to escape.
7 a6
8 �b3 ,i.a7
And this position arose in the

1 Oth game of the Merano match.
After 9 � bd2 J.. e6 1 0 � fl J.. xb3
1 1 'itxb3 'itc8 1 2 �g3 x es 1 3 h3
X b8 14 J.. e 3 'ite6 I S 'itxe6 fe
Black's doubled pawns were not
important and, so we soon agreed
to a draw.
Possibly 9 . . . �g4 was even
stronger, e.g. 1 0 A e2 'llr h 8 1 1 h3
� h6 1 2 �fl fS 1 3 J.. g S (not 1 3
J.. xh6 gh 1 4 ef J.. xfS I S J.. d S 'itd7
16 'llr h2 �e7 1 7 J.. xb7 K ab8 1 8
J.. e4 �g6 1 9 K d2 � f4 20 J.. xfS
K xfS 2 1 �g3 K g8 22 � g l X fgS
23 d4 e4 with an advantage for
Black, Ivanov-Glek, Borzhomi,
1 984; Black also stood well after
I S . . . �e7 1 6 J.. xb7 X b8 1 7 J.. e4
J.. xe4 1 8 K xe4 X xb2 1 9 d4 'ita8
20 X g4 e4 21 � 3d2 dS 22 'ita4
�g6 23 K bl X xb l 24 � xb l
'itb7 2 S 'itb3 'itxb3 26 ab cS
Gipslis-Podgaets, Kiev, l 98S;
however, in the second instance,
White might have tried 21 'itcl
X b6 22 'itxh6 �fS 23 'itgS ef 24
� g3 'itd8 2S 'itxd8 X xd8 26
�xfS X f8 with an unclear posi­
tion) 1 3 . . . 'ite8 14 d4 fe l S K xe4
J.. fS 1 6 X h4 � fi 1 7 J.. e3 'ite7 1 8
�g3 J.. g6 1 9 J.. c2 J.. xc2 20 'itxc2
g6 2 1 �e4 ed 22 J.. g S �xgS 23
�exgS d3 24 'itd2 hS and Black
had an advantage again (Smali­
Hawelko, Lucerne Olympiad
1 982). Thus, it is not advisable to
allow the knight to g4 . . .
9 h3 h6
After 9 . . . � hS 1 0 J.. gS White
stands better. Dzindzichashvili­
Korchnoi (Chicago, 1 982) conti­
nued 9 . . . 'f!le7 10 � bd2 J.. e6 1 1
� fl X ad8 1 2 �g3 dS 1 3 'ite2
The Italian Game 103
'ite2 'llr h8 14 ed J.. x dS l S d4 K fe8
1 6 � xeS �xeS 1 7 de J.. xb3 1 8
J.. gS Jl dS 1 9 ab X xeS and the
chances were equal.
10 � bdl �hS
n �n 'itf6
12 J.. e3 (94)
94
B
12 • • • � f4
If Black was going to put the
knight on f4 then he should have
done so only after the exchange
on e3: 1 2 . . . J.. xe3 1 3 � xe3 � f4
led to approximate equality.
Yusupov played a substantial
improvement against Hawelko
(Dubai 01., 1 986): 1 2 . . . J.. e6 1 3
J.. xe6 fe 1 4 J.. xa7 II xa7 I S �e3
X aa8 16 a4 � f4 1 7 'llr h2 K ad8 1 8
It f1 dS. A s result of White's quiet
play, Black captures the initiative.
After 1 9 ed ( 1 9 'itc2 � xd3 20
'itxd3 'itf4 + 2 1 \lr g l de) 1 9 . . . ed
20 d4 e4 2 1 � e l �e7 22 g3 � fg6
23 f4 ef 24 �xf3 �fS 2S � eS (2S
<2l xfS 'itxfS 26 'itd2 was neces­
sary) Black carried out a favour­
able series of exchanges: 2S . . .
� xe3! 26 X xf6 �xd l 27 M xf8 +
Jl xf8 28 K xd l �xeS 29 de c6.
The rook ending was clearly in
Black's favour.
104 The Italian Game

13 � xa7 K xa7 After 26 hg � xg6 27 'iit' h 5 'Wr'g5!

14 * h2 �e7 Black would already have the
IS �e3 ll a8 initiative. ·

16 a4! 26 'ft' f3
White has created the un­ 27 '(llt xf3 K xf3
pleasant threat of 1 7 a5. 28 hg �xg6
16 �e6 29 �e3 d4
17 �xe6 fe 29 . . . ll d8!? was necessary.
White stands better after this 30 '2Jc2 K 3r7
capture with the pawn and so Black also stood worse after 30
Black should have played 1 7 . . . . . . de 3 1 be K d8 32 �el (but not
�xe6. 32 �e3 K r7 33 K d2 K fd7) 32 . . .
18 '2!gl K ad8 g f] 33 f4.
19 g3 �fg6 31 cd!
20 :� n dS This leads to a clearly better
2 1 *e2 �c6 ending. However, the victory is
22 � g2 ll r7 still far away.
23 b4 ll df8 (95) 31 �xd4
Or 3 1 . . . ed 32 f4 �ge5 33 �f2
h5 3 4 *g2.
95
w 32 � xd4 ed
33 f4 cS
If 33 . . . �e5 then 34 �f2!
34 aS b6
34 . . . �e5 35 � f2 �c6 36 K c l
�xa5 3 7 K xc5 wouldn't change
matters.
3S ab K b7
36 ll cl K xb6
37 ll xcS
White also stood better after 23 Even stronger was 37 K f2 ll c8
. . . K fd7 24 K ad l (or 24 � f3 38 K fc2 K b5 39 l2l f2 .
II r7 2 5 �ge 1 with the threat of 37 ll xb2 +
h4-h5 and � f3-h4) 24 . . . de 25 38 K fl K xfl +
i!txe4 i!t f5 26 �e3. Of course not 38 . . . II fb8
24 ll adl because 39 K a5.
24 b2-b4 was also good. 39 �xfl K a8
24 �ge7 40 li aS (96)
2S hS After 40 � g4 a5 41 f5 ef 42 ef
25 f4 was, perhaps, more dan­ �e7 '2J f6 + ! the position would
gerous for Black. become sharper. This endgame is
2S g6 quite interesting and theoretically
26 '2! h3 important. Although it is not the
 The Italian Game 105

43 'lrf8
96
B 44 � e5! a5
If 44 . . . h5 then 45 f5 would be
decisive.
45 g4 a4
46 f5 ef
47 gf � xf5
Or 47 . . . a3 48 f6 a2 49 i#: xe7
a l ( 'i!lt' ) 50 .!Qg6 + *g8 5 1 x g7
mate.
theme of the book, it is worth 48 ef K a5
considering in detail. 49 X eS! x xc5
40 . . • .!Q e7 50 � d7 + 'lre7
41 .!Qg4 .!Qc6 51 � xc5 a3
41 . . . *g7 allows 42 .!Qe5. 52 'lrg3 'lrd6
42 X eS! 53 .!Qb3 'lre5
Stronger than 42 11 a4 a5 43 54 'lrg4 h5 +
.!Q xh6 + 'l/g7 44 .!Qg4 X b8 Black also lost after 54 . . . a2 55
(threatening 45 . . . X b4 46 X a3 � a t '1P f6 56 '1J f4 h5 57 .!Qc2 h4 58
a4 and � a7 -b5) 45 X c4 .!Q b4. 'lrg4 h3 59 * xh3 * xf5 60 'lrg3.
42 � e7 55 'lrg5 h4
56 f6
• • •

Other continuations were 'l/e6

totally hopeless: 42 . . . � b4 43 Black's resistance has been
X c7! a5 44 � xh6 + 'l/f8 45 g4 broken e.g. 56 . . . h3 57 f7 h2 58 f8
.!Q xd3 46 'l/g3 with a threat of ('it') h l ( iit ) 59 'i!lt'e7 + '1Jd5 60
g5-g6; 43 . . . .!Q xd3 44 'lrh3 a5 iitb7 + .
45 .!Q f6 + '1Jf8 46 'lrg4 with an 57 *g6 h3
amusing idea: 'lrg6 and X f7 58 f7 h2
mate. 59 f8('{11( ) hl( 'i!lt)
43 X c7! 60 !Q c5 + *e5
43 .!Q xh6 + '1Pf8 44 K a5 .!Qc6 61 'i!ltb8 + .
45 .1. g5 'lre7 was not so clear. Black resigned.
5 The Four Knights' Game
Polovodin-Tseitlio
Leningrad, 1985
While sorting out the material
for chapter 3-the Scotch
Game- l came across an amusing
game where a combination of this
opening with the Four Knights'
Game occurred. This game was
interestingly annotated by its win­
ner and so I decided to include it
in my book. Of course, not much
has changed in the ancient Four
Knights' Game, but one example
is worth seeing.
1 e4 e5
l �f3 �c6
3 �c3 �f6
4 d4
4 .l b5 is more characteristic of
the Four Knights' Game. For
those who wish to have a quick
draw there is a well known line: 4
. . . �d4 5 � xd4 ed 6 e5 de 7 ef
tt'xf6 (after 7 . . . cd + 8 .l xd2
tt'xf6 9 0--0 .le7 10 .l e i White
has a dangerous initiative for a
pawn) 8 de tt'e5 + 9 ti'e2 tt'xe2 +
1 0 .t xe2 and there is nothing to
fight for. Of course, I am not
going to recommend the reader to
finish the game this way, es­
pecially as, according to the old
analysis, White achieves some ad­
vantage by playing 5 .la4 (in­
stead of the exchange on d4). The
correct response to 4 . . . .l b5 is
the long-established symmetrical
4 . . . .l b4, after 5 0--0 0--0 6 d3 d6
7 .lg5 .t xc3 8 be tt'e7 the posi­
tion is equal. Incidentally, it re­
minds me of one of my old games:
8 . . . .ld7 9 d4 h6 1 0 .l h4 x es 1 1
ll e l a6 1 2 _id3 .lg4 with com­
plete equality (Urzike-Karpov,
Stockholm, 1 969).
4 ed
4 . . . .l b4, played for the first
time in Paulsen-Morphy (New
York, 1 857) leads to different
paths. Modem theory recom­
mends 5 � xe5! � xe4 (if 5 . . .
tt'e7 or 5 . . . 0--0 White plays
tt'd3, .ld2 and 0--0--0 with a pro­
longed initiative) 6 tt'g4 � xc3 7
tt'xg7 X f8 8 a3! Now after the
extravagant 8 . . . �xd4 Polovo­
din gives the following beautiful
variation: 9 ab � xc2 + 1 0 'i1rd2
�xal 1 1 'ilrxc3 a5! 1 2 .lc4! tt'e7
1 3 X e l d5 14 .l b5 + c6 1 5 �xc6!
tt'xe l + 1 6 .ld2. This is an amus­
ing situation! White's king is
stuck in the centre exposed to the
threats of the hostile pieces and
still Black is helpless: 1 6 . . . tt'e4
1 7 � b8 + ! 'i1rd8 1 8 tt'xf8 + 'ilrc7
19 tt'c5 + 'ilrxb8 20 tt'd6 + 'ilra7
2 1 _le3 + tt'xe3 22 fe. Everything
has become quiet and White has a
winning endgame.
8 . . . j_a5 is more solid. But
after 9 � xc6 de 1 0 tt'e5 + tt'e7 1 1
tt'xe7 + 'ilr xe7 1 2 j_d2 .lf5! 1 3 be
_lxc2 1 4 c4! _txd2 + 1 5 'ilrxd2
J.. f5 16 _ld3 J.. xd3 1 7 K he l +

* f6 1 8 * xd3 White has all the
chances in the double rook end­
ing.
5 �d5!? (90)
97
B
After 5 �xd4 .,lb4 the classical
position of the Scotch Game oc­
curs. But this knight move to the
centre turns the game into a gam­
bit. To be more precise, it is the
Belgrade Gambit. I want to re­
mind you that this gambit
appeared back in the 40s in
Yugoslav tournaments. Not
everyone wishes to sacrifice a
pawn (or even two) so early in the
game merely to obtain the initia­
tive. For that reason the gambit
hasn't become very popular. All
the same this gambit leads to
quite lively and exciting play. I
think those who favour stormy
complications should include the
gambit in their repertoire.
5 .,le7
Taking the second pawn by 5
. . . � xe4? is too risky e.g. 6 �e2
f5 7 �g5!? (much better than 7
_lf4 d6 8 0--0--0 �e7 9 � xe7
..t xe7 10 g4 c5 I I ..th3 fg 1 2 �xe4
d5 1 3 �e5 0--0 14 K hg l ..t f6 1 5
�h5 gf when White's attack has
The Four Knights' Game 107
evaporated and he cannot restore
material equality, Hartoch-Tim­
man, Leuvarden, 1 978) 7 . . . d3! 8
cd �d4 9 �h5 + g6 1 0 �h4 c6 1 1
de cd 1 2 ed ..tg7! (if 1 2 . . . �a5 +
White has two fairly strong rep­
lies- 1 3 *d l ! and 1 3 ..ld2 with a
complicated position) 1 3 *d l !
( 1 3 .,lc4, 1 3 �g3 and � h3 have
also been recommended and they
need careful analysis too) 1 3 . . .
h6 1 4 � f3 � xh4 1 5 � xh4 * n 1 6
.,le3 f4 1 7 .,l xd4 .,lxd4 1 8 _ld3
d6 19 � xg6 Ji gS 20 � xf4 ..txb2
2 1 11 b I ..l e5 22 1. b4! (Polovo­
din-Hazai, Helenchuv, 1 980) and
White had a much better end­
game.
In an earlier game, Po1ovoclio­
Tseitlin, Leningrad, 1979 Black
played 5 . . . � xd5. The game
continued: 6 ed J.b4 + 7 J.dl
�e7 + 8 �el J. xdl + 9 * xdl!
� xel + 10 .,l xel �b4 I I II heI
0-0 ll d6 c5 13 .,lc4 b5 14 ..t xb5
I. b8 15 .,lc4 .,la6 16 �e5! ..txc4
17 � xc4 � c6 18 l. e7! � xe7 19
de 1. fe8 l0 K el K b6 ll � xb6 ab
ll b4! d6 l3 a4! and Black
resigned.
Is there anything to add about
other lines of the Belgrade Gam­
bit? 5 . . . � b4 has been less
researched and 5 . . . ..l b4 + is
hardly ever played.
6 ..lf4 d6
It seems that the c7 pawn
should be defended because 6 . . .
� xd5 7 ed ..t b4 + 8 ..td2 leads by
transposition to a position which
occurred after 5 . . . �xd5. And
after 6 . . . 0--0 !? ..txc7 �e8 8
�xf6 + ..txf6 9 �e2 -.e6 1 0 e5
108 The Four Knights ' Game

§le7 I I a3 b6 1 2 0--0--0 JJ.. b7 1 3 A less active move would guar­

* b l .l, ac8 1 4 JJ.. d6 §l xd6 1 5 ed antee White solid pressure. Black
ttxd6 1 6 tte4 White has more tries to create counterplay on the
than enough compensation for kingside by his pawn advance.
the sacrificed pawn (Polovodin­ 14 hJ h4
Lerner, Belzi, 1 979). 15 JJ.. h2 K e8
7 <2) xd4 � 16 "tt d2
The more common continua­ White carries out his plan
tion is 7 . . . <2) xd5 8 ed e xd4 9 methodically. The exchange of the
ttxd4 0--0 1 0 0--0--0 JJ.. f6 I I ttd2. major pieces along the e file is not
The d5 pawn now gives White going to make Black's life any
reasonable prospects. easier.
8 -21 b5! <2) xd5 16 Jlg5
8 . . . <2J xe4 was dubious 17 f4 JJ.. f6
because of 9 <2) bxc7 K b8' 10 JJ.. e 2! 18 X ae1
with a big advantage. Black's pieces are restrained.
9 ed �e5 His queen impedes the coordina­
The tactical trick 9 . . . §lg5?! tion of his rooks.
failed to 10 §lxg5 ttxg5 I I ttd2 18 <2Je7
K e8 + (or I I . . . "tlte7 + 1 2 JJ.. e2 19 * hi �rs
<2Je5 1 3 0--0--0 and White is clearly 20 JJ.. dJ K xel
better) 1 2 *d l ! §lg4 + 1 3 J1 K e3 21 K xel a6
14 de! .l, ae8 1 5 cb x xn 1 6 JJ.. e2 White doesn't take any risks.
and White won (Bellon-Pomar, The weakness of Black's pawn,
Las Palmas, 1 975). which was powerful not long ago,
10 JJ.. e2 � g6 has now become a factor.
If White manages to complete 22 � ·J g6
his development and take control 23 �c2 Jl g7
of the open e file, he will undoub­ 24 JJ.. xf5 JJ.. xf5
tedly have an advantage. His 25 Jl g l
opponent should try hard to pre­ The bishop joins the struggle
vent this. After 1 0 . . . c6 I I <2)c3 with great force. The exchange 25
�g6 ( I I . . . "tltb6 1 2 X b l !) 1 2 . . . §l xc2 26 ttxc2 would doom
Jlg3 c 5 1 3 0--0 Jlh4 1 4 ttd2 Black to passive defence.
§lxg3 1 5 hg X eS 16 a4 White 25 tt d7
controls more space and his pieces 26 il fl! il f6
are better placed (Bragin-Fokin, 26 . . . tta4 would give White a
Sochi, 1 975). strong attack after 27 �d4 ttxa2
11 JJ.. gJ JJ.. f6 28 �xf5 gf 29 JJ.. xh4 followed by
I I . . f5 1 2 f3 f4 1 3 ilf2 JJ.. h4
. .1. e l -e3-g3, but this was
was possibly better. Black's last chance.
12 cJ Jld7 27 <2!d4 JUR
13 � h5!? The endgame after 27 . . . K e8
 The Four Knights ' Game 109

2S a xeS + -.xeS 29 �xf5 gf 30 32 c4

*el ! would be joyless for Black. But not 32 -.r4? ,txg2 + 33
l8 �fJ .d8 frxg2 X xe l 34 -.xh4 X e5!
29 � g5! 32 c5
The h4 pawn is surrounded, but 33 ,tc3 *d7
that isn't all. 34 fp g1 b5
29 ,txg5 35 *'4 b4
30 fg X e8 36 ,tf6 ,txgl
3 1 ,td4! 37 -.xh4 X xel +
The exposed long diagonal de­ 38 fp xgl X e2 +
cides the game. 39 fp g1
31 Black resigned.
Index of Complete Games

Page numbers in italics refer to complete games quoted in the notes to the main
games.

Adorjan-Hiibner 56 Ljubojevic-Portisch /6
Ady-Littlewood 42
Madl-Veroci 17
Belyavsky-Malanyuk 42
Belyavsky-Smyslov 76 Nunn-Tal 28
Nunn-Timman 21
Chandler-Spassky 49
Chiburdanidze-Tseshkovsky 38 Polovodin-Tseitlin 1 06, 107
Popolitov-Trushakov 38
Ehlvest--Geller 45 Popovic-Timman 29
Popovic-Yusupov 32
Geller-Salov 48 Prandstetter-Blatny 36
Geller-Tal 1 8 Prandstetter-Hort 98
Georgiev-Nunn 40 Psakhis--Geller 43
Georgiev-Tseshkovsky 33
Gufeld-Schussler 62 Rohde-Seirawan 72
Romanishin-Balashov 50
Howell-lvanchuk 90
Hiibner-Korchnoi 27, 31 Sax-Short 9, 12
Hiibner-Smyslov 78 Sax-Yusupov 74
Hiibner-Spassky 98 Short-Olafsson 89
Hiibner-Timman 43 Smagin-Hebden 41
Hiibner-Yusupov 21 Sokolov-Karpov /6
Sokolov-Smyslov 46
lvanov-Agapov 39 Sokolov-Spassky 46
Sveshnikov--Geller 97
Kapengut-Malanyuk 34 Szecsi-Heinrich 102
Karpov-Kasparov 60
Karpov-Korchnoi 58 Tal-Korchnoi 24
Karpov-Portisch 55, 57, 58 Timman--Giigoric 99
Karpov-Seirawan 7 1 Timman-Karpov 1 5, 93, 95
Karpov-Spassky 45 Timman-Smyslov 97
Karpov-Yusupov 25, 32, 1 0 1 Timman-Yusupov 87, 90
Kasparov-Karpov I , 4 , 7 , I I , 6 1 , Tseshkovsky-Agapov 3 7
62, 8 1 , 83 Tseshkovsky-Arkhipov 6 7
Khalifman-Arkhipov 66
Kudrin-Wolff 77 Ulmanis-Van Der Haiden 35
Kuporosov-Malanyuk 44
Kupreichik-Yusupov 89 Van Der Wiel-Korchnoi 27, 31
Van Der Wiei-Short 78
Ljubojevic-Karpov /6, 69
Ljubojevic-Nunn 43 Zaitsev-Yusupov 55
Index of Main Variations

As this book is not a detailed study of the whole of any particular opening, only
the major variations dealt with are listed in this index. Transpositions and other
variations are considered throughout the text, and the reader is recommended
to study all of the games in each section, rather than considering each game in
isolation.

Spanish
I e4 e5 2 0 f3 0c6 3 i.b5
(a) 3 . . . g6 45
(b) 3 . . . 0d4 50
(c) 3 . . . a6 4 _ta4 0 f6 5 0--0
(c l ) 5 . . . 0 xe4 1 8
(c2) 5 . . . ..te7 6 K e 1 b 5 7 i.b3
(c2 1 ) 7 0--0 8 c3 d5 33
(c22) 7 . . . d6 1

Russian
1 e4 e5 2 0 f3 0 f6
(a) 3 d4 55
(b) 3 0xe5 d6 4 0f3 0 xe4 5 d4 d5 6 i.d3 0c6 7 0--0
(b1) 7 . . . i.g4 8 1
(b2) 7 . . . _te7
(b2 1 ) 8 X e 1 56, 74
(b22) 8 c4 6 1

Scotch
1 e4 e5 2 0 f3 0c6 3 d 4 e d 4 0 xd4
(a) 4 . . . 0 f6 93
(b) 4 . . . _tc5 97

Italian
1 e4 e5 2 0f3 0c6 3 i.c4 i.c5 4 c3 0 f6
(a) 5 d3 1 0 1
(b) 5 d4 1 02

Four Knights '

1 e4 e5 2 0 f3 0c6 3 0c3 0 f6 4 d4 ed 5 0d5 106