Sunteți pe pagina 1din 162

Winning With the Catalan

Angus Dunnington

B. T. Batsford Ltd, London

First published 1997 © Angus Dunnington 1997

ISBN 0 7 134 8021 1

British Library Cataloguing-in-Publication Data.

A catalogue record for this book is

available from the British Library.

All rights reserved. No part of this book may be reproduced, by any means, without prior permission

of the publisher.

Edited by Graharn Burgess and typeset by Petra Nunn for Gambit Publications Ltd, London

Printed in Great Britain by Redwood Books, Trowbridge, Wilts for the publishers, B. T. Batsford Ltd, 583 Fulharn Road, London SW6 5BY

To Gary Lane, our Best Man


Editorial Panel: Mark Dvoretsky, Jon Speelman

Commissioning Editor: Paul Lamford General Manager: David Cummings







Part 1: Open Catalan: 4


S liJf3


Open Catalan: S



Game 1: Piket-Van der Ste rren, Dutch Ch 1984 18

Game 2: Dunnington-Richardson,

England 1997


Game 3: Hovde-Groiss, European Co rr. Ch 1984-90



Open Catalan: S




Game 4: Kengis-Meister, Togliatti 1985



Open Catalan: S.•.a6



Game 5: Krasenkov -Kaidanov, Gaus dal 1991


Game 6: Vladimirov-Thorhallsson, Gausdal 1991



Open Catalan: S




Game 7: Flear-Marciano, Toulouse




Open Catalan: S.• .id7



Game 8: Petursson -Zso.Polgar, Arhus 1993



Open Catalan: S




Game 9: Monin-Vul, Kecskemet 1992



S ••



6 0-0 0-0 7 'iic2 a6: Introduction and 8 a4






Khalifman-Lautier, Biel Z 1993



Marin-Gome z Esteban, Seville 1992



Kr arnnik -Piket, Dortmund 1995



13: H tibner -Siegel, Germany 1994





6 0-0 0-0 7 'iic2 a6: 8 'ifxc4







Ribli-Karpov, Amsterdam 1980 Permiakov-Berzin §, Latvian Ch 1994

1 10

Game 16: Ribli-Speelman, Moscow OL 1994

1 1 2



Game 17: Heine Nielsen-J.Kristensen, Ars 1995 Game 1 8: Ca.Hansen-S.Petersen, Denmark 1990

1 17

1 19


1 9:

Illescas-Epishin, Madrid 1995


Game 20: Andersson-Petursson, Reggio Emilia 1989


Part 2: Closed Catalan: 4


5 � 0-0 6 0-0


Closed Catalan: Introduction and Unes with e4xd5



2 1:

Cifuentes-Sosonko, Dutch Ch 1992


Game 22: Umanskaya-llinsky, Russia 1995



Closed Catalan: Black plays .••d5xe4


Game 23: Salov-Spassky, France 1994



Closed Catalan: White plays e4-e5


Game 24: Orlov-Tal, New York 19 90




Closed Catalan: Black plays an early •••b7-b5


Game 25 : Rajkovic-Colovic, Cetinje 1993


Index of Variations







Double Check



Good move

Good move


Excellent move


Bad move




Interesting move


Dubious move


White wins


Black wins






Team tournament








Candidates event


World championship


Team championship


European team championship


World team championship



Junior event


Women's event


Rapid game


Postal game




nth match game


Diagram follows


The Catalan Opening begins 1 d4 iCJf6 2 c4 e6 3 g3 d5 4 .ig2 (D).

Opening begins 1 d4 iCJf6 2 c4 e6 3 g3 d5 4 .ig2 (D). Throughout this

Throughout this book the diagram position will be used as a starting point at which Black chooses one of two major options:

1. - the Open Catalan;




2. - the Closed Catalan.


Of course the order of the opening moves is quite flexible - the se­ quence above, for example, is not the only route to the diagram position. It

is significant that 1 iCJf3 iCJf6 2 g3 d5

3 .ig2 c5 4 0-0 e6 5 d4 tClc6 6 c4

dxc4 leads (eventually) to an early position from Games 1-3, while 1 c4 e6 2 d4 d5 3 iCJf3 c6 4 'ifc2 iCJf6 5 g3

is a popular route to the Closed Cata­

lan. In fact, the reader will notice that

I have not standardised the initial

moves of the games (unless the in­ vestigation of alternatives or of simi­

lar lines dictates otherwise).' This is to accentuate the versatility of the

Catalan, and to cater for those play­ ers who may not necessarily open the game with 1 d4 (the Catalan has overlaps with the Reti, Queen's Gambit, Slav and Queen's Indian, for instance). With such a flexible move-order it . is hardly surprising that the Cata­ lan is rich in possibilities. There is something for everybody - White can play in true gambit style, hoping to turn a development lead into an in­ itiative that can become decisive, maintain the tension and operate in several sectors of the board in a com­ plex middlegame, or endeavour to steer the game to a favourable ending which can be very uncomfortable for


A key piece is White's light­ squared bishop - the 'Catalan' bishop. The logic behind the fianchetto is clear - White wants to put his oppo­ nent's queenside under pressure at

the earliest opportunity in order to frustrate his development. How Black addresses this problem con­ tributes in some way to the sub­ sequent nature of the game. Black

often tries to restrict the participa­ tion of the g2-bishop, a strategy that

often backfires because the time and resources could have been used more constructively. Moreover, un­ like those 'hypermodern' openings that combine the fianchetto of the



light-squared bishop with holding back the centre pawns, the Catalan sees White staking a claim for the centre with d2-d4 and c2-c4, thus guaranteeing some kind of influence in the most important part of the board. Consequently White enjoys enough space - and the harmony which this brings - to develop effec­ tively and fluidly, without having to

worry about achieving instant activ­ ity for his bishop. In the Open Catalan the scope of this piece is increased by the opening of the long hl-a8 diagonal after d5xc4. In the Closed Catalan, char­ acterized by Black's refusal to cap­ ture the c4-pawn and by the erection of a defensive barrier in the centre, White develops his forces in such a way as to facilitate another challenge on the fortified d5-pawn with an

eventual e2-e4 (e.g. 4

0-0 6 0-0 lbbd7 7 1Wc2 c6 8 b3 b6 9 :d l followed by lbb l-d2, e2-e4, etc.), when the Catalan bishop is ready to come to life. This book focuses on a selection of variations and recommendations for White which are designed to pro­ vide the reader with a working un­ derstanding of the Catalan. I have concentrated on specific variations rather than making an ultimately fu­ tile attempt to cover every aspect of the opening, but in advocating this or that particular line I have tried to cater for all styles. Hopefully those of you who are attracted to the Cata­ lan by the prospect of grinding out masterful victories in long endings will also be converted to the more

5 lbf3

complex variations, and vice versa. This is a necessity anyway, as one must always be prepared for tactics, slow positional struggles, drawish variations, etc. In the Open Catalan we deal ex­ clusively with 5 lbf3 because I be­ lieve the rather simplistic 5 1Wa4+ to be insufficient for an advantage if Black replies 5 Most of the 25 main games in the rest of the book are annotated in con­ siderable depth, and I do not want the Introduction to snowball, so I will limit myself to offering just a few practical examples which fe a­ ture typical Catalan characteristics:

White's space advantage

In most openings White tends to use . the advantage of having the first move to win more territory than his opponent. Usually Black can live with this, and sometimes he even in­ vites White to grab more than a fair share of the board (e.g. Alekhine's Defence, 1 e4 lbf6, or the Modern Defence, 1 e4 g6) with the intention of a timely counter against White's (hopefully) over-extended forces. But there is a thin line between what is and what is not acceptable for the second player and, in the case of the Catalan, the balance between the use of pieces and pawns is such that White can fight for an advantage without real fear of creating weak­ nesses. Here are a couple of examples of how 'normal ' play from Bl ack can lead to White's ostensibly harmless


In trodu ction

space advantage being transformed to something far more troublesome:

The position below arises from the Closed Catalan, Black choosing

to post his light-squared bishop on b7 (instead of the more active a6) and

meet the eventual e2-e4 by

Despite being a somewhat predict­ able, passive approach, it has been seen in many games and is very popu­ lar at club level. For more about this line see Salov-Spassky, Game 23. After the moves 1 d4 tbf6 2 c4 e6 3 tbf3 d5 4 g3 i.e7 5 i.g2 0-0 6 0-0 tbbd7 7 ifc2 c6 8 b3 b6 9 l:d1 i.b7 10 tbc3 l:c8 11 e4 dxe4 tbxe4, Por­

tisch-Radulov, Moscow Echt 1977

is one of many games from past and present which continue 12•••tbxe4 13

Wxe4 'ikc7 14 i.f4! i.d6 15 i.xd6 'ikxd6 16 c5! (D).

there is the added problem that he is unable to generate some sort of counterplay by expanding in another area of the board - White has the e5-

square in his grasp so

ruled out, and the fianchetto means that White, not Black, has the poten­ tial to advance on the kingside. These factors combine to giveWhite a space advantage on the queenside, in the centre and - at some point in the future - on the kingside. Waiting for White to march his army forward on all fronts is not a pleasant pros­ pect but, for some reason, this posi­ tion is by no means rare (even at master level). In this particular game there fol­

lowed 17 b4 l:fd8 18 'ikc2 tbf6 19




tbe5 tbd5 (it soon becomes evident that the knight is merely sitting pretty on d5) 20 a3 b5 (accentuating the sorry plight of his bishop, but Black cannot allow the enemy knight to land on d6 via c4) 21 i.e4 g6 22

h4 (D). This thematic thrust is designed to deprive Black's bishop of any free­ dom
h4 (D).
This thematic thrust is designed to
deprive Black's bishop of any free­
dom (given the time Black would de­
fend the bishop and push his own

17 dxc5

tbxc5 runs into 18 ifb4 'ike7 19

pawn to c5). Now 16


Lc l, so Black played 16

Apart from the fact that Black is

rather cramped on the queenside


Note the difference between the two bishops. White's stands majestic on e4, the perfect Catalan bishop! Teaming up with the queen to exert

pressure on the bl-h7 diagonal has induced Black to put yet another pawn on a light square (2l .h6 would have given White free access to the diagonal). Black must still keep an eye on his c6-pawn as well as con­ sider the implications of Moreover, g2 is now free for White's king in preparation for a rook to come to the h-file to begin a danger­ ous kingside attack with h4-h5, etc. In other words, after a patient build­ up of forces from his opponent, Black now faces the prospect of coming under fire on both flanks - hence the coming necessary but ulti­ mately futile queenside counter:

22 •••a5 23 bxa5 lla8 24 a4! (White's

domination affords him the luxury of aggression in any area of the board)

(being able to sur­

render the wonderful bishop is in­ dicative of White's control - the black bishop is no match for the

24•.•b4 25

knight) 25 ••Jlxd5 26 ltab1 f6 27

l2Jg4 l:tad8 28 l:txb4 .:Sd7 30 l:tb8 (D)

29 liJe3

R N'ii·?f� �A� "' � -1x•.· i<fi{/ • .I %?�< :uut .::.·,:: 8 •
R N'ii·?f�
• .I
• � ��.; • ?.,{ t� • ��0 if_;;:::.
f%ff :
• '.i{ti;



which has deprived Black of the use of d6 - a key square which is right in the heart of his half of the board. The next move is one final attempt to undermine White's hold over this critical square but, ironically, the c5- pawn has even more to contribute:


.:t'8 34 .:Xd5 f5 35 1Vc5 l:te6 36 1Vxe7 l:txe7 37 l:txc8 ! l:txc8 38 l:td8+ l:te8 39 l:txe8+ l:txe8 40 lbc:l5 1-0. A fit­

ting finish, with the powerful knight and the annoying c-pawn earning the full point. The whole game went according to plan for White. He prevented the


in the most uncompromising fashion - by occupying the c5-square him­ self. At first glance this seems like a positional error, for not only does White voluntarily create a backward

d4-pawn (on a semi-open ftle), he also presents Black with an ostensi­ bly perfect outpost for the knight on d5. Perhaps it is this factor which is responsible for seducing so many players into this kind of position.

However, unfortunately for Black,

all he can look forward to is passiv­ ity, while the extra spaceWhite gains on the queenside with c4-c5 has a kind of mushroom effect, spreading

to other areas of the board.

traditional 'freeing' advance

31 d5! cxd5 32 c6 l:td6 33 c7


Open lines

Now we see a more brutal use of White's extra space, this time on the otherflank. Sometimes in the Closed Catalan both sides postpone any pawn captures or advances in the



From the first diagram (after 16 c5) Black's chief problem has been the crippling effect of the cS-pawn,



centre until development is com­ pleted. The game Marin-J.Horvath,

Odorheiu Secuiesc 1993 illustrates

that this can be a risky approach from Black's point of view. Inci­ dentally, the opening moves of this game feature a little 'shadow-box­ ing', which is not unusual in some

1 d4 lbf6 2 c4 e6 3 g3 d5 4


J.g2 J.b4+ (this attempt to disor­ ganize White's pieces is discussed later in the Introduction) 5 J.d2 J.e7

17 e5! lbe8 18 h4 h6 (19lbg5 was

threatened) 19 lbfl cS!? (Black seeks immediate counterplay in the centre rather than waiting for the

build-up to grow out of control, e.g.

19 lbc7

22lbh2!, when White continues ex­

pansion with f2-f4-f5)20

20lbe3 :res 21 'ife2 'ifa8



21 d.S!! (D)

6 lbf3 0-0 7 0-0 c6 8 c2 lbbd7 9 J.f4 %5 10 lbhf6
6 lbf3 0-0 7 0-0 c6 8
lbbd7 9
J.f4 %5 10 lbhf6 11 b3 b6 12
:d1 J.a6 13 lbbd2 :cS 14 e4 J.b7
15 J.b2
16 :act

The repositioning of the black queen from d8 to b8 serves to re­

move the queen from the d-file and defend the b7 -bishop. However, these are preparations for a future opening up of the position after

therefore an obvious way

for White to cut across his oppo­ nent's plan is to avoid this central capture by pushing the e-pawn. Moreover, the subsequent territorial superiority offers White an opportu­ nity to concentrate on a kingside of­





Much stronger than the automatic 'ikxc4, which is only good for

equality. White aims to bring yet an­ other piece within striking range of Black'skingside by clearing the long diagonal for the dark-squared bishop. The game went 21 J.xd5 22 :xd5! (material is just one of many factors that form the chess equation

- here activity and the initiative are

far more important) 22 23

J.h3 (23 e6?! lbdf6 24 exf7+ :xf7 25 lbe5 1i'd6! is fine for Black, whereas Marin's choice simply relo­ cates the bishop on a promising new

diagonal while maintaining the ten­ sion and keeping Black under con­

:c7 24lbe3) 24

lbe3 (again White improves a piece

instead of pushing the e-pawn; after

trol) 23



no longer clear) 24 l:.fe8 26 ltJxg7! (D) lbc7 25 ltJfS
no longer clear) 24
l:.fe8 26 ltJxg7! (D)
25 ltJfS

Another entirely logical sacrifice which, after a brief examination of the layout of the pieces, is not really a surprise. White's queen and minor pieces have immediate access to the kingside. Black's army, on the other hand, is huddled together on the first two ranks, and the queen is way out on the wrong side of the board. Even if White does not have a decisive at­ tack he should be able to force a healthy return on his 26••.'it>xg7 27 i.xd7 d4 (Black judges correctly the importance of the al- h8 diagonal -the fo llowing two variations support this opinion and highlight the justification of 21

.t:.xd7 28 e6+

i.f6 29 i.xf6+ 'it>xf6 30 \i'e5+ 'it>e7

d5 !! and 22 .t:.xd5: 27


exd7+'it>xd7 32 'i'f5+, or 27



e6+ f6 29 ltJe5 !? { 29 ltJh2 !?}

30 i.xe5+ i.f6 31 e7 .t:.xd7

32 'i'g4+ 'it>f7 33 i.xf6) 28 i.xe8

:Xe8 29 bxc4 1i'b7 30 lllli2 fS (D). Now White played 31 exf6 +?,

.i.xf6 32 'ii'g4+ 'it>h8 33

'i'h5 i.g7 34 :n 'ife4 35ltJg4 .t:.e6 !

when 3 1

29 fxe5

However, the diagram position is still very good for White, and with correct play he can maintain the momentum of the offensive which started with White's exploitation of a space advantage. Once again the dark-squared bishop is a key piece, and Black's problems on the king­ side are by no means over after 31 .t:.el!, when the threat of i.b2-c l, menacingly lining up on another di­ agonal, is difficult to meet. White is clearly better thanks to his oppo­ nent's terrible weaknesses on f6, f5, g6, h5 and h6. Of course, in this example the white attack was not a sudden Tal­ like sacrificial masterpiece culmi­ nating in a deadly king-hunt. Such games are seen less frequently than chess columnists and authors (my­ self included) would have you be­ lieve, and they usually require one or two poor moves fr om a helpful vic­ tim. Instead Black (a Hungarian GM) was guilty only of being slightly pas­ sive and of losing a little time with his light-squared bishop and his queen, yet this resulted in sustained pressure from White in the form of



White's d-pawn

The position in the diagram below

arose in the game Bogdanovski­ Raicevic, Pula 1990, after the moves 1 d4 tLlf6 2 c4 e6 3 g3 d5 4 i.g2


is the main line - see Games 1-3) 7

dxc4 5 ltlf3 cS 6 0-0 tLlbd7 (6

tLla3 tLlb6 (7


8 ltlxc4 i.c5 9

ltlxd4 tLlb6 { 9


10 tLlb3 i.e7 11

i.e3 a5 12 a4 puts Black's queenside under fire, Bogdanovski-Karaklajic,

the board. The knight sits safely on e5, contributing to the power of the Catalan bishop by covering the c6- square, and the rook bolsters the d­ pawn. In fact, with a 2-1 majority in the centre, White's most natural plan should involve these pawns. The game continued : 16 e4 0-0 17 dS exdS 18 exdS i.d6 19 i.e3 "ilc7 20 tLlxd7 tLlxd7 21 axbS axbS 22 l:ta7 'ir'd8 23 i.h3 (D).

Yu goslav Ch 19 91} 10 ltlxb6 i.xb6 11 tLlb5 ! worked out well for
Yu goslav Ch 19 91} 10 ltlxb6 i.xb6
11 tLlb5 ! worked out well for White
in the game Tal- Danov, Moscow
1972: 1 1
12 �xf2 ii'b6 +
lbd4 e5 14 e3 exd4 15 'ii'xd4 'ii'e6 16
i.d2 i.d7 17 i.b4 i.c6 18 l:tad1,
and Black's king was stranded) 8
ltlxc4 tLlxc4 9 'ir'a4+ i.d7 10 'ir'xc4
l:tc8 l l lbeS bS 12 'ir'd3 c4 13 'ir'c2
'ir'b6 14 l:td1 i.e7 15 a4 a6 (D).

A second black pawn has arrived on c4, and Black has succeeded in providing it with maximum support. Meanwhile, apart from the useful a2-a4, which adds to White's ar­

moury the possibility of opening the

a-file (

meets with lbe5xd7),

White has been busy in the centre of


.'�f8 is better, when

31 'ili'h5 !? keeps White on top) 31 'ir'eS! tLlf8 32 'ir'e7 'ir'f6 33 l:tel






lill!d7 (one final try -otherwise the white rook will come to e8 -hoping for 34 .txd7? 'ihe7 35 l:.xe7 'it>f8) 34 'ii'xd7! 1-0.

Typical ending

In Portisch-Radulov White's tor­ ture treatment left him with a good knight against a terrible bishop. In the next example we see the great Kar­ pov struggling on the wrong side of a Catalan ending, and again White's remaining minor piece is a knight and Black is left with a bishop. But this time the (dark-squared) bishop controls a lot of squares and could only be described as 'good'. The diagram position was reached after thirteen moves in the game Ribli-Karpov, Amsterdam 1980 (D).

is the traditional

which Black hopes to reduce his op­ ponent's influence in the centre and subsequently steer the game to

equality. White has a pawn on f4 be­ cause he parked his bishop there to attack the c7-pawn, prompting Black


d5xf4. As Black is in no position to exploit the potential weaknesses cre­ ated by the recapture on f4 , White is free to turn the front f-pawn to his advantage - extra cover of the eS­

to solve the problem with

c7-c5 break, with

square. The game continued 14 lbe4!, aiming to clamp down on the posi­

tion and inviting Black to react with

14 c5

in this specific situation), in turn in­

ducing the fo llowing sequence of ex­ changes: 15 dxc5 �xc5 16 �xcS 1Wxc5 17 'ii'xcS .txcS 18 l:.ac1.:Cc8 19 lDe5! .txg2 20 'it>xg2 (D).

(thematic but maybe not best

.� .•• B. i� � 1;«/,/,,.,� [�-� - -·� , •• , •• . �
[�-� -
, •• , ••
X "' ''
w e;;

The opening twenty moves are discussed in detail in Game 14, while here we concentrate on the rest of the game, 'rewinding' to the fourteenth move in order to see just how quick the transition from opening to end­ ing can be. A briefexamination of the pawns tells us that a natural plan for Black

Only seven more moves have been played since the previous dia­ gram, yet the nature of the game has changed completely, a transforma­ tion which is seen frequently in the Catalan. Many players (especially if playing against Karpov!) would shake hands and split the point in this



position. Moreover, Black's bishop could hardly be better placed, and one could be forgiven for assuming that this leaves Blackwith a comfort­ able game. In factWhite has the edge

on account of the main weakness of bishops - confined to either light or dark squares, versus the versatility of knights. Karpov played 20 ••.f6 (in Gavrikov-Azmaiparashvili, USSR 1981 Black tried to improve with


.Z:.ac8 22 .Z:.d7 a5 23 b3 f6 24 .:.Xc7 .Z:.xc7 25 lDf3 <;tn 26 lbd4 b4 27 lbb5 .Z:.c6 28 <;tf3 <;te7 29 f5 ! gave White excellent chances of exploit­ ing the vulnerable light squares in the

enemy camp), and after 21 00 �f8


.Z:.xc8 .Z:.xc8 25 .Z:.d7 .Z:.c2 26 lbd4 is also good for White) 24 h3 �f8 25

lbd4 <;tr7 26 a4! bxa4 27 bxa4 the

World Champion should have con­


solidated the queenside with 27

according to Ugrinovic, but 28 lbb5 .Z:.xc 1 29 .Z:.xc 1 l:te8 30 �f3 is still


when 21 e3 {21 lbd3 !?}

22 e3 g6 23 b3 �b4 (23


not so comfortable for Black. Instead the inaccurate 27

put him in serious difficulties after


28 .Z:.c4!, as the planned 28

29 .Z:.cxd4 .Z:.e8 runs into trouble in


view of 30 l:.b4 followed by Ab4-b6

and .Z:.dl-d6, etc. Consequently Black

opted for 28 when White's

advantage soon took shape: 29 .Z:.xc8

l:txc8 30 .:lc4 31 llb7+ �e7 32 l:ta7 e5 33 fx e5 fxe5 34 lbf3 l:txa4 35 lbxe5+ <;tr6 36 lbc6 �cs 37 l:txh7 l:ta2 38 <;tr3 aS 39 h4! a4 40 <;te4! (D).


work well together. The game ended:






together. The game ended: �a3. The white pieces continue 40 �f8 ( 40 l:txf2 loses to





loses to the reply

4) lbd8 !) 41 l:ta7 �d6 42 f4 l:th2

43 lbd4 - threatening 44

�e7 44 lbf3

l:tb2 45 lbe5 a2 46 lla6+ <;tg7 47 l:txg6+ <;th7 48 l:ta6 �xh4 49 l:ta7+ <;tg8 50 lbg4 !, etc.) 43 l:ta6 �7 44

lbes+ �xeS 45 <;txeS �g7 46 l:[a7+ �h6 47 l:txa4 l:txh4 48 �6 l:thS 49 e4 .:lh4 50 eS l:th5 51 e6 l:tfS+ 52 e7 �g7 53 <;td6 .:.CS 54 l:ta7+ �6 55 lld7 1-0.

l:ta6 �e7 45 f5 ! - 43








and 4


Before turning to the main lines we should have a brief look at a couple

of early deviations for Black. One

which is not uncommon is 4

This can also be classified as a form

of the Bogo-Indian Defence (1 d4

lbf6 2 c4 e6 3 lbf3 �b4+), but in this particular context we are concerned primarily with the idea of meeting 5


�d2 with the retreat S Inci­

dentally both 5 lbc3 (leading to the

Romanishin Variation of the Nimzo­ lndian Defence) and 5 lbd2 are per­

fe ctly playable, but the former is

outside the scope of this book and




the lattertends to give Black interest­

d5xc4, e.g.


'ifa4+) 6 lDf3 dxc4 7 0-0 J.xd2+ 8

'ltxd2 l:.b8 followed by

An examination ofthe position af­

ter 4

help us see why Black has 'wasted' a tempo.

5 J.d2 J.e7 (D) should

(5 lDd2) 5

ing options involving

some cases, White may even profit from the gift by bringing a rook to

the c-file or by relocating the bishop on c3 in favourable circumstances. Occasionally the bishop is fine on d2

(after 4

dxc4 7 0-0 0-0 8 'ifc2 a6 9 'lfxc4 b5 10 'ii'c2, for example, White's extra move is useful), but for us the bishop is most inconveniently placed here in

5 J.d2 i.e7 6 lDf3


(not 5





0-0 tbbd7 8 c6 (D): ••• - . rutt . . w�.t.••�.t.�i - ��·- -
0-0 tbbd7 8
c6 (D):
••• -
•.• ,.
,J, ••

Time is only one of many factors in chess. Another - equally impor­ tant - part of the game is the har­ mony of the pieces, and it is the coordination of the white pieces which Black is hoping to disrupt by tempting the bishop to d2. This square is often used by the queen's knight to support the e2-e4 advance or as a stepping-stone to b3 or c4, for example. The bishop also obstructs the defence of the d4-pawn, and in certain positions (e.g. the Closed Catalan) the appropriate post for the bishop is on b2. Nevertheless, apart fr om confus­ ing some opponents, there is little else for Black to gain from this idea. If necessary White can simply return the tempo when the time comes to put the bishop on f4 or g5 and, in

the Closed Catalan, which is exam­ ined in Part 2 ofthis book. For exam­ ple, here is the position after the moves (1 d4 lDf6 2 c4 e6 3 g3 d5 4 J.g2 i.b4+ 5 J.d2 i.e7) 6 lDf3 0-0 7

The bishop hinders White's usual method of development, so we have to find an alternative. One way to try to avoid 'correcting' the play thus far

is 9 b3 b6 10 i.c3, e.g.


J.b7 11

lDbd2 c5 12 dxcS lDxcS, when White can line up on the al-h8 diagonal with 13 'ti'b2!, an option which is not available to White with the normal move-order. Moreover, an interesting idea to spoil Black's fu n was seen in the

game Beliavsky-Azmaiparashvili, Amsterdam OHRA 1990: 1 d4 lDf6 2



c4 e6 3 g3 dS 4 i.g2 i.b4+ S i.d2

i.e7 (5


is not a good idea



has been played, for

the absence of the dark-squared bishops clearly helps White, who can be happy to continue along nor­ mal lines as though nothing has hap­ pened), and now White played 6 i.c3!?, preparing for fluid develop­ ment with extra support of the centre and influence on the al -h8 diagonal. Black's 6•••b6 then met with the new

7 �b3 !?, and after 7 •.•i.b7 8 0-0

'IVeS (8

also good for White) 9 cxdS i.xdS

0-0 9 �f4 �bd7 10 �2 is

10 �f4 0-0 11 b4 and White

clamps down on the queenside) 10



11 b4 and White clamps down on the queenside) 10 (9 exd5 with the Closed Catalan),

with the Closed Catalan), defending

the c4-pawn.

After 5


6 �f3 i.d6 7 0-0

0-0 8 bd2 Black is ready to carry

out the planned expansion in the cen­

tre: 8.••e5 (8



9 b3

{ 9 e4 }

�f4 i.xg2 11 �xg2 0-0 12 �2

then 9


10 i.b2 e5 11 cxd5 lDxd5

'iVb7+ 13 f3 �4 14 �4 'ti'c6 15 h3!

12 l !, or 9


10 cxd5 e4


(15 b3? b5) 1S•••\Wxc4 16 hxg4 White

llJh4; similarly 8


9 b3 e5


emerged on the more comfortable

cxd5 cxd5 { 10


11 llJc4 }


side of a complicated position. The

dxe5 fue5 12 i.b2). White can then

changes to the pawn structure on the

gain the advantage with 9 cxdS cxdS

kingside provide cover for White's


llJxd5 10 �4 'fke7 11 e4 llJ5b6

king (meanwhile the h-file is open


�xd6 leaves Black weak on the

for the use of the rooks - if so de­ sired) and White's pieces are more

dark squares) 10 dxeS llJxe5 11 llJxe5 i.xeS 12 llJf3, and Black has

harmoniously placed. Of course this is by no means a

nothing to compensate for his iso­ lated pawn.

fo ol-proof recipe for White, but it is

a promising (and probably unex­

pected) way to unsettle Black after

the bishop check and retreat. Another fourth move for Black

which can have independent signifi­ cance is 4•••c6 (D).


leading to


the Closed Catalan, but putting the

bishop on d6 in order to prepare

is also possible. The simplest

is 5 'ti'c2 (which fits in

reply to 4

Black can follow up with









1984, Black sought to exploit the specific order of moves with S .••bS 6

c5 (6 cxb5 cxb5 merely invites Black

to bring his took to the c-file) 6 7


dxe5 llJfd7 8 .!Df3 .!DxcS 9 0-0, but

he was already falling behind in de­ velopment. The game went 9••.aS 10

.!Dc3 .!Llba6 11 .!Dd4 �b4 12 •d1 .!De6 13 e3 �xd4 14 exd4 with a

clear advantage to White. Finally there is 4c5. As the vari­ ation 5 llJf3 cxd4 6 .!Dxd4 e5 is very



complicated and tends to give Black good attacking chances (e.g. 7 lbf3

d4 8 0-0 lbc6 9 e3 d3 ! ?), I recom­

exd5 6 lbf3

mend 5 cxd5. Then 5

transposes directly to the Tarrasch Defence. This leaves 5.•.lbxd5 6

lbf3 lbc6 (

here or on the

next move should transpose to 'b' in the note to Black's sixth move in Ci­

fuentes-Sosonko, Game 21, while



good for White) 7 0-0 (D).

8 lbxd4 0-0 9 lbc6 1 l lba3 ! is


7 0-0


With 7 Black prevents e2-

e4 and puts the d-pawn under pres­

sure. Now White can head for a

favourable queenless middlegame

with 8 dxc5 'ii'xd 1 9 l:txd1


lbg5-e4, but 8 lbe5! gives us a posi­ tion almost identical to that reached

11 lbg5 ! followed by



identical to that reached 11 lbg5 ! followed by 10 lbf6 after 7 lbe5 in the

after 7 lbe5 in the game Piket-Van der Sterren, Game 1, the only differ­ ence being the disappearance of the black pawn on c4. This obviously fa­ vours White, who can choose a con­ tinuation in which he no longer has to take the trouble to recapture Black's extra pawn, for example


bxc6 11 1i'a4.

.i.d7 9 lbxc6

10 .i.xc6+


Open Catalan : 5


This move is quite logical. After re­ moving the c4-pawn Black chal­ lenges the d-pawn. We areconcerned with 6 0-0 lbc6 (others are insuffi­ cient because they give White a free hand in the centre). Then White has 7 lbe5 and 7 'i!fa4. The fo llowing game deals with the active 7 lbe5.

Game 1

rather than surrender these pluses just to level the material.




loses material to 8 e3


lbxf7+), while 7••.lbd5 and 7 .••'fkc7 invite White to damage the queen­

side pawns with 8 lbxc6. The only feasible alternative is to deal with the audacious knight by re­

9 .txc6+, 10 'i!fxd8+ and 11

7 lbxd4?


Piket - Van der Sterren Netherlands Ch 1994 moving it -7••. tt:lxe5 8 dxe5 (D).
Piket - Van der Sterren
Netherlands Ch 1994
moving it -7••. tt:lxe5 8 dxe5 (D).
1 d4 lbf6 2 c4 e6 3 lbf3 d5 4 g3
lbe5 (D)
And now:
a) 8
9 .l:.xd1:

After this aggressive invasion into his opponent's half of the board White must be prepared to continue the game a pawn down if necessary. In some lines White should be content to rely on his positional superiority

.i.d7 (10





a1) 9•.•lDd5 should not be met with 10 .i.xd5? exd5 11 .l:.xd5 .i.e6 (White's Catalan bishop is a valuable piece). Instead 10 lba3! is dangerous

for Black, e.g. 10

e4! cxb2 12 .txb2 lbc7 13 lbc4

gives White more than enough for the pawn) 11 lbxc4 .tb5 12 b3 (Nei­ shtadt gives 12 lbe3! lbxe3 13 .txe3

.l:.b8 14 l:.d2.


for White) 12

.i.e7 15 a4 ! as good


Open Catalan: 5



i.a3) 13 i.b2 i.e7 14lL!d6+ i.xd6 15 exd6 0-0 16 e4, and the soon to be protected passed d6-pawn was a problem for Black in Marosi-Kral, Hungary 1994. a2) 9.••lbd7 is another possibil­ ity, when 10 f4 l:[b8 (freeing the b­ pawn and avoiding something like

11 lL!a3 0-0 12 i.xb7) 11

a4 guarantees excellent chances for

White in the queenless middlegame. Either White will regain the pawn with a bind or else attempts to keep it will just make matters worse for Black. Black has tried several moves

f6 12

exf6 gxf6 13 lba3 h5 14lL!xc4 h4 15

a5 hxg3 16 hxg3 l:[g8 17 .:.a3, or

here without success, e.g. ll

10 i.e7



12lbc3 i.e7 13 a5lL!a8 14

lL!b5, etc.


b1) In Andersson-Guyot, France



9 lba3!:

1993, Black played the greedy 9


and Ulf 'the Wolf' turned down the potentially wild 10 lbc4 b5 in favour of the simple 10 bxc3lL!xc3 11 'it'c2 lbd5 12 l:[d1 i.d7 13 lbc4, with con­ siderable pressure for the pawn.

White's judgement proved correct:

14 i.b2 i.e7 15 lL!d6+

i.xd6 16 exd6 0-0 17 i.xd5 exd5 18

.:.xd5 b6 19 i.xg7 ! �xg7 20 l:[g5+ �h8 21 l:[h5, etc.

10lL!xc4 i.c6 is less

risky. Petrosian-Panno, Palma de Mallorca 1969, was slightly favour­

able for White after the continuation

.:.c8 12 e4 lL!b6 13 lL!d6+

i.xd6 14 exd6 lL!c4 15 'it'g4 'it'xd6


16 1We2!) 16 1Wxg7 'ifd4

13 'it'c8

b2) 9


11 a3


17 i.h6.



lbd7 9 f4 :

cl) Vukic-Hau sner, Banja Luka

1981 went 9

lL!b5 'ifb8 12 i.e3 .i.d7 13 lL!d6+ i.xd6 14 exd6 'ifc8 15 .i.f2 0-0 16 e4 .i.c6, and now 17 a4 ! would have

maintained a healthy initiative for the pawn.


11 lL!xc4 'fic7 12 i.e3 lL!b6 13

lbd6 !? l:[d8 14 1Wc2 with pressure on the c5-pawn) 11 'fixd8+ .i.xd8 12 i.e3 i.e7 13 l:[fc1 and White had the upperhand in Sveshnikov-Arseniev,

.:.b8 14lL!xc4


17 .i.c6+Wf8

18 b5 fo llowed by 19 .:.a4) 17 bxc5


White homed in on the c-pawn.

Let us return to the position after


bxc5 18 .:.a4 a6 19 .:.c 1 l:[b2 20


10 lba3lL!b6 11

c2) 9-.i.e7 10 lba3lL!b6 (10

USSR 1986. After 13

lL!xc4 15 .:.xc4 b6 16 b4 0-0 (16

17 .:.c7, or 16





.:.xc4 b6 16 b4 0-0 (16 17 .:.c7, or 16 .i.a6? cx i.d7 (D). 8 lba3



White sends his other knight into the game, not worrying about re­ storing material equality. Variations

which involve the sacrifice of a pawn or two need to be treated appropri­ ately by the player making the in­ vestment (which is exactly how one

should think of sacrific;es). Usually


Open Catalan: 5 . c5

White receives immediate compen­ sation in the form ofsmooth develop­ ment, space and the greater activity which these factors provide. Rather than allow the fire to burn out prema­ turely the trick is to nurture these cu­ mulative advantages and generate a lasting initiative which will enable White to keep up the pressure. You may notice that some of the players on the white side of the game extracts which follow are well­ known fortheirfondness ofcompli­ cated positions, and it will come as no surpri se to learn that Kasparov has taken an interest in this variation.

8 cxd4

Not 8.•.ltlxd4? 9 -txb7, though two other moves are occasionally seen:

a) In R.Garcia-Esain, Zaragoza

1992, Black thought he could simply ignore the tension in the centre with 8••. -te7, but after 9 lLlxd7 'fi'xd7 10

dxc5 'ii'xd1 11 l:.xd1 -txc5 12 lLlxc4 �e7 13 -tf4 l:.hd8 14 -txc6 bxc6 15 lLle5 the c6-pawn was a liability.

is aimed at reducing

the influence of White's Catalan bishop. The game Kakhiani-A.Sok­ olov, Helsinki 1992, went 9 lLlaxc4

10 lLlxc6 -txc6 11

1i'xd4 leaves White rather active) 10



12 'ii'xd5 -te6 13 'ii'xd8+ lLlxd8 14 1 b6 White should have played

15 lLld3 -te7 16 lLlf4 with an initia­




lLlxd4 (9


-te3 !. After 10




fxe3 lLlc6 12 lLlxf7) 11

popular alternatives, both tried by Andersson in a match with Kasparov in Belgrade 1985. a) Kasparov met 9•••l:.c8 by 10 'ii' b3, but three years later in Lj ubo­ jevic-Yusupov, Belfort 1 988, 10 was played, and it does seem to offer White even better chances of an ad­ vantage:

a1) The following position was reached after the natural lO•••lLldS


was reached after the natural lO ••• lLldS (D). In return for the sacrificed pawn White

In return for the sacrificed pawn White has a considerable lead in de­ velopment and active pieces, and

Black's king is still in the centre. Consequently White should not be afraid to enter into any complicated sequences because his 'extra' pieces will rule, and Black will be too busy addressing the problem of his re­ tarded development to preventWhite from recovering the pawn with inter­


est. Thus: 1 1 lLlxd7 lLlxf4 (11

12 lLle5 is better for White thanks to

his slightly more harmonious pieces)


-txc6! bxc6 (12

12 l:.xc6






lLlxf8 'ii'd5 !? (13

13 lLlh3+

14 �g2

The text forces White to work for an advantage. There are two less

'ft'd5+ 15 e4! 'ifxc4 16 1i'b4 17 lLlxe6 ! fxe6 18 'ii'h5+ clearly favours

Open Catalan: 5



White) 14 gxf4 'ii'xc4 15 l:tc1 'ii'd5


'ii'a4 ! (16 lLlxe6 fxe6 is equal)



!? (16


17 llfd 1 and


.'�xf8 17 l:.fd1 c5 18 llxd4 are

not enough for Black) 17 f5 ! exf5 18 llfd1 �xf8 19 llxd4 'ii'e6 20 l:.c5 (20

lldc4 !? and 20 'ii'xa7 !? are also

good) 20

21 'ifa5 and White

emerged with a useful initiative.

a2) In Lalit-Sonntag, Bad Woris­

hofen 1989 Black tried 10 but



this rash thrust merely adds to Black's

problems on the queenside and does

nothing todiminishWhite's pressure

queenside and does nothing todiminishWhite's pressure defend the potential weakness on b7. We may safely conclude

defend the potential weakness on b7. We may safely conclude that Black runs the risk ofcastling into an attack


the centre. After 11 lLlxc6 �xc6

with 12


Play continued: 13


�xc6+ l::txc6 13 lLle5 llc5 14 l:.c1

lld1 �e7?! (13


invites 14 .:.d3

'ii'd5 15 'ii'd3 a6 16 b4 the German IM must have been regretting his carelessness already. In fact there was no keeping White out of the self­

inflicted weakness on c6, and after

17 llxc3 dxc3 18 'ii'xc3

i.e7 19 'ii'c8+ �d8 20 lLlc6 0-0 21 'ii'xa6 White had safely won a pawn.

b) More recently the game Topa­

lov-Morozevich, Madrid 1996 saw

Black produce a dubious novelty af­



e5 allows 14

'iic4+ fo llowed by 15 'ii xf7, so best

14 'ii'c4 'iia6 15 llxd4

'ifxc4 16lhc4 l:.d 1+ 17 �fl �d7 !


which limits White to a slight advan­

tage) 14 l:.xd4 i.c5 (14

l:.xd8+ :txd8 16 'iic2, threatening

�c1-e3) 15 llc4 �b8 16 'ifi>fl !? (also worthy of investigation is 16 l:c3,

when Morozevich gives 16




and 15 llb3, while 13



18 l:.c2 { 18 b4 lLla4 } 18


ter 9


10 lLlxe5 1i'b6 1 1 lLlxd7


<li>h 1 { 17 �fl lLlc5 18 'iif4+ �a8

lLlxd7 12 'ifa4. The young Russian


e3 lLld3

is unclear} 17



decided against the 'dull' but accu­

'ii'f4 + e5 19 'ii'xf2! .:td1+ 20 �fl lLle4

rate 12 13 'ifxb4 �xb4 14

i.xb7 l:b8 - which is only margin­

ally better for White - in favour of


the provocative 12(D).

The continuation provides us with

a good illustration of the power of

the Catalan bishop on the h1-a8 di­ agonal. To make matters worse for Black the exchange of his light­ squared bishop means that he cannot even make a challenge on the long diagonal, leaving him less able to


21 'ii'xb6 axb6 22 .:.f3 llc8 23 �f4!

:xa1 24 �xe5+ l:.c7 25 .:txf7 with a


and now, instead of 17 �gS? 'ii'xb2 !, White could have concentrated on

very good ending for White) 16

the b7-pawn with 17 �e3!, for ex­

ample 17

llb4 ! 'ii' f6 19

llxb7+ 'iii>c8 20 l:tc1+!, or 17


18 llccl �xe3 19 l:tcb1, etc. This

leaves 17

19 l:tb4

lLlxc5 19 'ii'c2) 18

18 l:td 1 (18 :txc5 !?





lLlc5 20 l:txb6 lLlxa4 21 llxd8+ llxd8

22 Open Catalan: 5


22 l:xb7+ �c8 23 fxe3, when Black is struggling to hold on to his pawns.

10 .,3

Hitting the b7-pawn, though there seems to be some confusion regard­ ing the validity of this 'threat' - see note to White's 1 1 th move.

10 •.•


avoids Kasparov's im­

provement (next note) and should transpose to the main game, but many players do not bother to defend the b-pawn.

10 1Wc8

11 i.f4

A maj or alternative is 11 1Wxb7,

which Kasparov brought back to life in the Human vs Computer chal­ lenge match with Deep Blue in Philadelphia 1996. The move first appeared in Vaganian-S erper, Lu­ cerne Wcht 1993, but failed to ftnd much popularity - probably some­ thing to do with a few strong annota­ tors mysteriously giving the line "1 1 1i'xb7 ? �xe5 12 �xe5 l:b8 13

1i'a6?? i.b5". Strange. Of course, af­

ter 11

is legal, forced and therefore best.

Then 13

was dealt a blow in

the game Sulava-Dizdarevic, Cat­

tolica 1 9 93, when White found 14

i.g5 ! f6 (14

15 �xd7) 15


1i'g4!! , earning a clear advantage

16 1i'h3 fxg5 17 i.xd5

g4 18 1Vg2. More to the point is

putting the question to the



12 li:)xe5 l:b8 , 13 1Wf3

after 15

13 i.d6,



a) 14 lDc4 i.b5 ! is roughly equal.

b) 14 i.f4 �5 ! (14

�4) 15 �g6?? loses to 15

Jlxb2? 15


c) Vagani an-Serper continued 14

lbxd7 1i'xd7 15 i.g5 i.e5 ! 16 l:ab1

h6 and Black had a comfortable po­

sition ( 1 6

Black). d) However, in Kasparov-Deep Blue, Philadelphia (2) 1 996, the

also looks fme for


PCA World Champion chose 14

lDc6!. After 14

i.xc6 15 1i'xc6 e5



l:[b1 ! Black's best is 16

when White's bishop pair (grip on the light squares) is enough for an

edge. Instead the computer played

! 17 1Wa4 1i'b8, and the

clever 18 i.g5 ! i.e7 19 b4! i.xb4 20

i.xf6 gxf6 21 1Wd7! left White domi­

nating the light squares. This contin­ ued throughout the game, which Kasparov eventually won on the 73rd move to level the score at 1-1

(he went on to take the match 4-2). Kasparov's 14 �6! gives White an extra option in the main line, but perhaps the next turn in chess fash­ ion will prompt more players to eliminate this possibility by moving

over to 10


so 11 i.f4 remains the most impor­ tant continuation anyway. White maintains the tension and prepares to

bring his rooks to the centre, still eyeing the b7-pawn.

16 l:tb6?

.'ti'c8 instead of 10

11 1Wc8 (D)

.1 ?�\Wr- �-- w�.l-.i.Ni•i -·­ - . - '<'Y0i ii'!t@ - ••••• • ��
.1 ?�\Wr- �--
-·­ -
, "
u w
�� -;:;?Jrj

Op en Cata lan: 5



12 l:r.acl

according to Morozevich and Yur­

12 l:r.fd1 is also played. With the

kov) 15



l:r.ac8 16 l:r.cd1

text White simply forgets about the

b6 17 .i.xd4 ii'e7 18 .i.xc5 l:r.xc5 19

d4-pawn and worries Black's queen

i.xd5 exd5 20 e3 iVe4 2 1 l:r.d4 ii'c2

on the c-file, confident that the build­


Wb4 l:r.fc8 23 l:r.fd1 is good for

up of pressure compensates for the sacrificed pawn. Indeed a brief look at the diagram position should be

White, Ribli -A.Sokolov, Reykjavik 1988) 16 l:r.fd1 (Glek's suggested improvement is 16 lied1 l£le7 17

enough to convince us that Black is rather cramped, while White enjoys

i.xd4 i.xd4 18 e3, which he evalu­ ates as slightly better for White)

free and easy movement for his



17 l:r.xc8 l:r.xc8 18 .i.xd5

pieces, a useful outpost on the e5-

( 18 i.xd4 .i.xd4 19 llxd4 l:r.c 1 + 20

square and significant influence on

l:ld 1? l£lf4 ! shows that White, too,

the b-file, c-file and the h1-a8 diago­

must be careful) 18


! (better

na l.

than 18


19 i.xd4) 19 Wxd5

12 ••.


Obstructing the Catalan bishop is

practically the only plan available to Black since there is not much in the way of choice here. For example, the

is not to

be recommended, as was demon­ strated in Zia Rahman-Norri, Erevan

OL 1996: 13 l£lxf7 ! �xf7 14 i.xc6

i.xc6 15 l£le5+ �g8 16 l:r.xc5 with much the better position for White. Black went from bad to worse, the

game ending 16

17 l:r.fc 1 l£lxf4

18 gxf4 l:r.d5 19 l£lxc6 bxc6 20 llxc6

ostensibly natural 12



exd5 20 .i.xd4 i.xd4 21 llxd4 l:r.d8 !

(2 1

lowing White to keep his winning

chances alive with 22 b4) 22 e4 �f8

23 exd5 �e7 24 �g2 �d6 25 �f3


26 l:r.e4+ �d6 27 l:r.e2 l:r.d7 ! 28 l:r.d2

!?) 26 l:r.d2. Now

Black played 26

tant improvement on Tukmakov­

A.Sokolov, Biel 1992, which saw

fail to negate White's ad­

26 .l:tc5?

�e5, and 25

(an impor­

l:r.c8 (Tukmakov proposes 25

.l:r.c5 has also been played, al­



vantage after 27 �e4 .:tc4+ 28 �f5,

.l:tc5 29 h4 llxd5+

for example 28

'il'e8 21 f5 l:r.ad8 22 l:r.xe6 ii'h5 23


.:txd5+ �xd5 31 h5 h6 32 g4 �d4

l:r.c5 iVxf5 24 .l:txd5 1-0 (24



f4 �e3 34 g5 �f3 35 �e5!! �g4


.:.e8+ 'l;f7 26 l:r.xd8).


g6 ! f5 37 �e6 �xf4 38 �f7). GM

13 l£ld3 !?

A promising alternative to a tried and tested, lengthy variation which leads to an ending in which accurate defence offers Black good drawing chances. Kharlov-A.Sokolov, StPe­ tersburg Z 1993 is typical: 13 l£Jxd7 'il'xd7 14 l£le5 l£lxe5 15 i.xe5 (15 i.xd5 !? 'ifxd5 16 'ii'xd5 exd5 17 l:r.xc5 l£lc6 18 l:.xd5 llad8 is equal

Andrei Sokolov has considerable ex­ perience with this variation - as you may have noticed from these exam­ ples - and his hard work finally

.:te8, for


brought dividends in 26 27 �f4 meets with 27

venting the advance of White's king. The game actually ended 28 h4 h5 29 f3 f6 30 g4 g5+ 31 �g3 hxg4 32

�xg4 gxh4 33 �xh4 1!z-1h. At the



Open Catalan: 5


beginning of this note I said that this

line is tried and tested - it is not un­ usual today for the merits of a par­ ticular variation to be influenced by

improvements in a

Nevertheless, I would not be surprised to see someone come along with new ideas for White, and the process will start all over again. Returning to the main game, we find ourselves back in the opening stage, thanks to Piket's preference to

rook ending !

keep the pieces on the board - at least for the time being.

13 .•.


White's dark-squared bishop cov­

ers some key squares on the h2-b8 diagonal, so it makes sense to re­

move it. The immediate 13

for example, gives White the oppor­ tunity to plant a knight on d6.


14 gxf4 (D)

oppor­ tunity to plant a knight on d6. i.e7, 14 gxf4 (D) This is quite a

This is quite a natural capture in several variations of the Catalan, and here White chooses to take with the pawn even though a knight can come to f4 . Black is not in any position to exploit the slight weakening of the pawn cover in front ofWhite's king, and the f4-pawn allows White to

remain in control of the useful eS­ square. Meanwhile White's knight is more actively posted on d3, monitor­ ing cS and eS (in some cases the b4- square may be significant). Note that the arrival of the pawn

on f4 also rules out

e6-e5, so now

the d4-pawn cannot be defended comfortably, although Piket's play thus far suggests he is less interested

in the d-pawn than in play on the c­


14 i.e7


15 lDceS

As White is ready to double on the

c-file, Black prepares to add support

to the pinned knight, connecting the

rooks in the process. White still man­

ages to create a dangerous initiative

in the game, so perhaps Korchnoi's

should be considered.

By placing a rook on the same file as White's queen, Black offers to give White a taste of his own medicine in the event of the b-file being opened after a capture on c6. White should continue as he does in the game, turning the screw a little more, rather than impatiently seeking to win back his pawn.

15 .J:lab8!?

16 :c4

On c4 the rook keeps an eye on

the d-pawn just in case White is pre­ sented with an opportunity to switch

plans. 16


Consistent. Others:

a) If Black considers c4 to be the

wrong square for the rook he can try

17 'ii'xb7 lDa5, instigating

a series of exchanges from which

White emerges with a big lead in the

16 'fkd6

Open Catalan: 5



resulting ending: 18 'ii'xd7 'ifxd7 19





21 :xc8+ .i.f8 22 axb3

ll:lxd7 ll:lxc4 20 ll:lxf8 :xf8 21 :c l! :c8 22 ll:le5 ll:lb6 23 :xc8+ ll:lxc8

24 .i.b7, etc. b) Korchnoi's suggested move,

is terrible for Black since the queen and bishop are no match for White's

21 'ii'xb8+


16 .• can be answered with

pieces. Nor is 20 to be recommended.


17 ll:lxc6 .i.xc6 18 .i.xc6 bxc6 19


21 h6

22 'ifc4

:xd4 or 17 ll:lxd7 'ifxd7 18 .i.xc6

It is safe to say that White has

bxc6 19 :xd4, when Black's c-pawn is a liability in both cases.

taken control of the c-flle!




Now all the white pieces play a role in the assault on Black's queen­ side, so White is ready to reap the rewards of his investment. The b7- pawn needs defending, the d-pawn is weak and the c6-square is attacked

Black's d-pawn is still exposed and the knight could not be better placed than it is on d3, defending the f4 -pawn and the b2-pawn as well as covering b4, c5 and e5. Black's pas­ sive bishop, on the other hand, is se­ verely restricted.

by four white pieces (White's queen

22 :Xc6

23 :xc6


and d3-knight are also available if necessary). It is clear that Black is unable to retain his extra pawn.

24 b3!



Fixing another pawn on a dark square (a5) and releasing the knight from its queenside guard duty. With the only useful file belonging to








:Xc6! (D)

White could have played 20 'ii'a4 (threatening the d-pawn) flrst, but in­ stead he exploits Black's weak back rank in order to provide the queen with a more active post.

White he simply improves his posi­ tion in readiness for the ending which will result from inevitable ex­ changes (Black is under so much pressure that trading off a couple of pieces is the lesser evil).

that trading off a couple of pieces is the lesser evil). 24 'ifb 7 25 1'3!

24 'ifb7

25 1'3!

More fine-tuning. White does not want the enemy queen to come to e4 after :c6-c7.

25 g6

26 �n

This time White nudges his king off the gl -a7 diagonal.

26 � gT

Black, too, is waiting, but in his case there is little else to do, e.g.


Open Catalan: 5


26.•.1i'b5 27 :c5 'il'xc4 28 bxc4 ! :as 29 a4 ! i.d8 30 :c6 �f8 31 :d6 i.f6 32 c5, etc.

27 :c7


27•••1i'b5 28 'il'xe6.

28 �cS!

With the king still on gl White would have to deal with the possibil­

ity of


28 :eS

Not 28


29 �xe6+.



:ds (DJ

28 1i'xc7? 29 �xe6+. 29 �e4 :ds (DJ 30 1i'c5 Consistent with White's strategy thus far.

30 1i'c5

Consistent with White's strategy

thus far. Black's weaknesses on a5 and d4, coupled with the relative strengths of the pieces, indicate that White will benefit from an exchange of queens. Black's prospects are rather cheerless, for he can only de­ fend. Interesting is 30 f5 !? gxf5 31 �g3 (threatening 32 �f5+ with the point


runs into 32 �h5+ �g6 33 lLJf4+

�g5 34 'il'c 1

32 m+ �g6 33 llJf4+. Black should


33 'il'xf7+), when 31



and 31 loses to

play 31 and after 32 �xf5

�xf5 33 'il'd 3+ �e5 ! 34 l:.xt7 i.h8 !

the onus is on White to justify his


piece sacrifice (Black's king seems quite safe on e5).

30 1i'xc5




Renewing the threat on the e6-

pawn. 31

32 lta7

33 llJdJ!



Heading for e5 to attack the f7- pawn. Black is struggling on both sides of the board and White has a firm grip on the centre . Victory for White should just be a matter of time, though Black's next allows his opponent to remove the last pair of rooks while winning the pawn.


33 l:.c6










The beginning of the end. White's

'good' knight dominates the 'bad' bishop - a fe ature not uncommon in Catalan endings. The remaining moves of the game

were: 36

37 �el �5 38 �d2

�xf4 39 �d3 e5 40 lL!e7! e4+ (an­


�5 38 �d2 �xf4 39 �d3 e5 40 lL!e7! e4+ (an­ i.b6 other way for the

other way for the game to end could

be 40

e4 44 f4 +} 44

b4 axb4 45 a5 �h3 46 a6 i.b8 47 lL!xb4 i.a7 48 lL!c6 i.b6 49 �b5,

etc.) 41 fxe4 �g5 42 e5 i.cS 43 �6

�f5 44 �c4 1-0.

An interesting game . It is not yet clear whetherWhite has insufficient winning chances in the ending dis­ cussed in the note to White's 13th move (has Sokolov had the last word?). Whatever the upshot, Piket's

i.d6 43 a4 �h4 ( 43

41 lLJd5 i.c5 42 �c4


Open Catalan: 5



tension-building approach looks ef­ fective. White has a menacing initia­ tive in the early middlegame and it is scarcely noticeable that Black has an extra pawn. This is exactly how these positions should be played.

Now we turn to 7 'il'a4, which re­ tains the tension and is seen more frequently than 7 lDe5. White aims

to tidy up in the centre and rely on his development advantage in the middlegame. Black has two ways to deal with the pressure on the c6- knight - break the pin with 7

or ignore it altogether with 7 First the more sober option:


Game 2

Dunnington - Richardson England 1997

1 d4 lDf6 2 lDr3 e6 3 c4 d5 4 g3 dxc4 5 .tg2 c5 6 0-0 lbc6

7 1i'a4 (D)


7 •••cxd4 is examined in the next main game. Others:

a) 7 .••.te7 ignores the potential pressure on the queen's knight. A

logical reply is 8 lDe5, when both

8.••0-0 9 lDxc6 bxc6 10 dxc5, Pomar­


lDxd7 'ii'xd7 10 dxc5 are clearly bet­

ter for White.

lDd7 was ftrst seen in the

game O' Kelly-Euwe, New York 1951, and after 8 dxc5 .txc5 9 'ii'xc4 0-0

10 o!Dc3 a6 11 lbe4 i.e7 12 .te3 lDb6

13 1i'b3 lDd5 14 l:[fd 1 'il'c7 Black

achieved equality. It makes more sense not to post the dark-squared bishop on e3, where it can easily be­ come a target. I.Almasi-B .Lengyel, Budapest 19 93 favoured White after

h6 13 l:[fd1 'il'b6 14 .tc3

fkc7 15 l o!Ob6 16 'il'b3 lDd5 17

fkd8 18 o!Dc3. Another possibil­

.J:[d1, when 1 l


ity is



Puig, Malaga 1964, and 8

b) 7




13 b3 o!Of6 14

15 'ii'xe4 f6 16 lDd4 accentuated

White's influence over the h1-a8 diagonal in Bischoff-Sonntag, Bun­

desliga 1987. The main problem

with 7.Ji).d7 is that it does nothing to address Black's usual develop­ ment problem.

is a natural enough of­




fer to exchange queens, which en­ joys greater popularity at club level than in international practice. Black

judges that White must lose time with his queen if he is to avoid a pre­ mature exchange, but eventually the black queen becomes exposed on a5, as was demonstrated in the game Htibner-Doghri, Erevan OL 1996: 8 •xc4 cxd4 9 o!Oxd4 o!Oxd4 10 •xd4


o!Dc3 e5 13 i.g5 i.e7 14 l:[fcl! h6 15

.te3 .J:[d8 (this makes matters worse for Black by depriving the queen of a

11 "it'c4 (11


1 1



Open Catalan: 5


retreat square) 16 a3 l:td6 17 b4 'ifd8 18 ll:Jb5 .i.e6 19 'ifc7 l:.d7 20 'ifxd8+ .i.xd8 21 .i.xa7, etc.



I prefer this to the messy alterna­

lbe5 ! ?)

gives the game a completely di ffer­ ent character and White must tread

carefully. The simple 8 'ifxc4 threatens 9 dxc5 and consequently forces Black to make a decision regarding the

centre. The clearing of lines that re­

sults from

c5xd4 or d4xc5 leaves

tive 8 dxc5, when 8



White with pressure on both the h 1- a8 diagonal and the d-file. The other option open to Black is to keep the

position as closed as possible with a


c5-c4, giving


White the opportunity to erect a

strong centre.


8 b5

Black exploits the 'exposed' posi­

tion of his opponent's queen in order to expand on the queenside. The ma­

jor alternative is 8 ••• cxd4, when 9

ll:Jxd4 l:tc8 10 lbc3 brings us to a


14 l:tac 1 and White is dangerously

active. a2) Equally effective is 11 .i.d2,


ll:Jb3 .td6 14 l:tfd1 �e7 15 ll:Jb5

.i.b8 16 ll:Jc5, Sakharov-Borisenko,

12 11i'xb4

.i.xb4 13 ll:lb3, when White benefits

more from the queen exchange.

b) 10 •.• 'ii'b6 is intended to fru s­

trate White's development by aiming at the b2-pawn, but in Ribli-Prand­ stetter, Warsaw Z 1 979, White sim­ ply carried on regardless: 11 ll:lxc6

.txc6 12 .te3 !. Now the 'threatened'

USSR 1971, or ll

12 'ifxc5 .i.xc5 13




12is weak in view of 13

14 ll:Jb5) 14

l:tfc l fo llowed by 15 ll:Jb5, so the

game continued 12 .•• 'fi'b4 13 'ii'xb4 .i.xb4 14 .i.xa7 .txc3 15 bxc3 .txg2

16 �xg2 l:txc3 17 l:tfb1! l:c7 18 l:tb2

and White had a small but enduring advantage.

c) Perhaps the most natural con­

tinuation for Black is 10

lbxd4 11

'fi'xd4 .ic5 12 1i'b4 .ic6, challeng­

ing the Catalan bishop before cas­ tling. 13 l:td1! and now:

l:tab1 'ifc2 (13






poses White no prob­

c 1) The game Portisch-Radulov,


Buenos Aires OL 1978 went 13



'ii'xb4 .i.xb4 13 ll:Jdb5 ! �e7 14 a3

.i.xc3 15 ll:Jxc3 l:thd8 16 .i.f4 and

1 1

do not alter the assessment) 12lDb3

clearly favours White, for example

12 'ifc7

1 3 'ifxb4 .i.xb4

14 ll:Jb5 �e7 15 a3) 13 .i.g5 (13 .tf4 e5 14 .i.g5 .te6 15 'ii'a4 and 13 ll:Jb5 'ifb8 14 lbc5 are also good for an ad­

vantage) 13

0-0? 14 ll:lb5 !)

continuation 12

(Black should avoid the

12 'ifxc5 .i.xc5 13 ll:Jdb5



l:td1 .i.e7

( l l






14 .txc6+ l:txc6 15 .ih6!! (D).

.i.xc5 13 ll:Jdb5 a1) 11 l:t d1 .i.e7 ( l l 'ifb4 'ii'c5 'ifb4 a6 (13

Open Catalan: 5



A witty, prepared improvement

on 15 i.g5 i.e7 16 llJe4 �xe4 17


'ife l+! 20 Wg2 'iff2+ (which leads to a draw), 15 i.h6 must have come as

quite a shock to Radulov. As so often happens when someone has the du­ bious pleasure of being the first player to be on the receiving end of such a dangerous move, Radulov soon found himself with a terrible

position after 1S

16 Wg2 0-0 17 i.xg7 Wxg7 18 'ilg5+

Wh8 19 'ifxf6+ and 20 'ifxf2) 16 'ifxf6 0-0 17 lDe4. Black's broken kingside

was the telling factor: 17

'ife5 i.e7 19 a3 ! 11i'b6 20 l:.d7 i.g5

21 b3 ! (21 h4 f6 and 21 �xg5 :c5

are less accurate; now White threat­ ens to bring his queen back to b2 be­

fore hitting the bishop with h2-h4)

21 1Vxb3

Wh8 24 :adI! (24 11i'f6+ 'itg8 25

:d4 :c4) 24

:td4! 1Vg6 (this time 26

to 27 l:.xf7 !) 27 11i'f3 h5 28 :xb7 and

White won. Days after this game, in the last round of the Olympiad, Ribli (Por­ tisch's team-mate) was also allowed to play 15 i.h6, the victim on this occasion being Lj ubojevic. The fa­


but 16 l:.d2 eS 17 i.e3 could not have encouraged him. Black trails behind in development, the d5-square is weak, his queen and rook are poorly placed and White - whose forces en­ joy more harmony - is about to dou­ ble on the d-flle. After 17 .•.'ifa6 18

19 11i'g5 ! 0-0 20 'iixe5

:ad l

Black had nothing to show for the

mous Yu goslav GM tried 1S

11Vxf2+ 18 Wh 1 f6 19 'ifg4






22 �xg5 hxg5 23 1Vxg5+


25 'fkf6+ 'itg8 26

l:.c4 loses

pawn (20

l:.xc8 :xc8 23 1i'd4 !). Black was not so generous in Quinn-Kelly, Dublin 1 995, defending the e5-pawn with

17 •••11VaS . There followed 18

1 9 11i'g5 ! h6 20 1Vf5 i.e7 21 l 0-0 22 �d5 �xd5 23 l:.xd5 1Va4 24 1Vd7 with a clear advantage to White. As a trainer at Olympiads for the teams of Botswana and Nigeria I can

appreciate that Lj ubo was probably

too busy sightseeing and enjoying himself to study dozens of games from the previ ous rounds. However, what is surprising is the number of players who have fallen into this trap in the many years since Buenos Aires 1978! Hulak-Sabovic (another Yu goslav GM !), Yugoslavia 19 85, is

another such example at interna­

tional level. This time Black also

chose 15

.tf8, but after 16 l:.d2 he

did not present White with an out­

post on d5 (16

offering instead


16 ••.a6 with the aim of regrouping

the queen and rook. This did not af­ fe ct the result: 17 l:.adl :c8 18 e4


'ifc7 19 e5! �g8 (19

20 i.g5 �e7 21 1Wa4+! �6 22 llJe4

23 l:.d8+ mates) 23

h6 (22

i.f6 :g8 24 a3 i.e7 25 i.xe7 'itxe7

26 'fkc4 1-0. Before turning to a 1 3th move al­

ternative for Black let us briefly look

This was

tried in Quinn-Palmer, Dublin Z

1993, whenWhite demonstrated that

15 i.h6! is equally effective with the

i.f8 16 l l:.d8

(losing a pawn, but the threat of a discovered attack on the queen looks



l:.d8 l:.c8 22

a3 a6



at (14 .txc6+) 14


queen on c6: 15


Open Catalan: 5


decisive) 18 :xd8+ �xd8 19 i.g5 i.e7 20 'ii'd4+ 'ii'd7 21 'ii'xa7 �e8 22 'ii'a8+ and White won. c2) On account of the problems Black has in the examples in 'c l' the

move 13

c2 1) The idea behind putting the queen on a5 was illustrated in the game Csom-Peters, Hastings 1978/9, when White insisted on continuing the hitherto successful theme with

15 i.xc6

:xc6 16 J.xg7 Black threw a span­

nerin the works: 16

! 17 �xf2

�xg7 with no serious problems for the second player, since g5 is de­ fended and Black threatens to swing

his queen over to h5.

c22) More to the point is 14 i.d2. White gained a slightly better ending in Ftacnik-Peters, Hastings 1980/1:

15 lLld5 ! lLlxd5 16 'ii'xe7+


i.b4 :Ce8 19 e4 allows White to keep

lLlxe7 17 i.xa5 i.xg2 (17


was introduced.

14 J.h6. After 14



14 i.e7


the bishop pair) 18 �xg2 lLlc6 19 f6 20 b4 a6 21 a4 lLle7 22 i.el

and the bishop enjoys more freedom than the knight.

9 'il'd3 (D)

9 'il'xb5? runs into 9

lowed by 10

lLlxd4 fo l­

9 'il'xb5? runs into 9 lowed by 10 lLlxd4 fo l­ 9 Inviting White to open


9 'il'xb5? runs into 9 lowed by 10 lLlxd4 fo l­ 9 Inviting White to open

Inviting White to open lines. If

Black does not feel comfortable with


practically surrendering the centre in return for chances ofcounterplay on the queenside. White then has two sensible squares for the queen:

this, the main alternative is 9

a) 10 'il'c2 leaves the d l -square

free forthe rook, and on c2 the queen supports the central thrust e2-e4. This is perfectly playable, but unless White plays the committal a2-a3 (which weakens the b3-square) he can expect to lose another tempo with his queen sooner or later when Black, after completing his devel­ opment, is able to go on the offensive

lLlc6-b4(-d3). The prospect of

an enemy knight planting itself on d3 (after e2-e4) should be taken seri­

ously. In the game Hausner- Klovans, Pardubice 1994, White failed to do this, overestimating his kingside at­

tacking chances: 10

:c8 (depriving

White of a pin on the hl -a8 diagonal)


12 'i'e2 lLld3 13 lLle5 ! lLlxc 1 14 :xc 1

merely helps White) 12 :dt 0-0 13 lLlc3 (13 'ii'e2 lLlb4 14 lLle5 !, defend­ ing d3, was seen in Vukic-Pfleger, Ybbs 1968, when White emerged


15 J.g5 i.e8 16 lLlc3 h6 17 i.e3

14 1We2 lLld3

15 J.g5 b4. Now Hausner continued

with his aggressive but faulty plan, turning an unpleasant position into disaster in only a few moves: 16 e5

cxb2 19

lLle5 fxe5 20 1Wg4+ i.g5 21 :abl f5 !

moves: 16 e5 cxb2 19 lLle5 fxe5 20 1Wg4+ i.g5 21 :abl f5 ! with 11


11 e4 i.e7 (the immediate ll

with a clear advantage after 14

lLlc6 18 a4 !) 13


bxc3 17 exf6 gxf6 18

22 1Wxg5+ 'ii'xg5 23 i.xg5 e4 0- 1.

Op en Catalan: 5



b) 10 'ii'd1 is more versatile. Black

cannot realistically hope to prevent the advance of the e-pawn, so there is no need for White to hurry. The fact

that the queen is not exposed on d1 leaves White with time to try some­ thing else if he so desires, and one such example in reply to the normal

.!DeS, which offers

White excellent chances of generat­ ing an initiative. The continuation 11.-lDxe5 12 dxe5 lDg4 13 W'd4 h5 14 h3 i.c5 15 'ilff4 lDh6 16 lDc3 is very

good for White due to the threatened 17 lDe4, so Black may as well play


1 972 continued 12 lDc3 a6 13 i.f4

15 e4

lDb3 16 l:[b1 i.e8 17 i.e3 White's

0-0 14 a3 i.e8 (after 14

when Csom-Hort, Zagreb






command of the centre is impres­ sive) 15 lDxc6 i.xc6 16 e4 (D).

of the centre is impres­ sive) 15 lDxc6 i.xc6 16 e4 (D). White's fluid pawn centre

White's fluid pawn centre and space advantage outweigh Black's chain ofpawns on the queenside, and the constant threat of White march­ ing his (effectively passed) d-pawn down the board makes life difficult for the defender. The diagram position is the kind that Black wants to avoid when he

which retains the op­

tion of pushing the c-pawn in more favorable circumstances if White reacts inaccurately.

plays 9


10 i.xc5


11 lDc3

Now Black must deal with the at­ tack on his b-pawn.



Solving the problem by moving the target, and giving White a turn to make a decision. Others:

a) The main alternative for Black

is 11 ••.lDb4, when all three squares available to the queen have been


a1) 12 'ii'b1 leads to an unclear

13 a3 lDbd5

position after 12

14 lDe5 lDxc3, e.g. 15 bxc3 i.xg2 16 'ilfxb5+ �e7 17 �xg2 'ilfd5+ 18 lDf3

lDe4 !, or 15

l:r.b6 17 lDd3 i.e7 18 i.e3. a2) 'iid1 i.c6 13 i.g5 'ilfxd 1 14 l:tfxd1 is slightly better for White ac­

cording to Inkiov. a3) My favourite is the odd-look­ ing 12 'ii'd2!?, as seen in Vladimi­ rov-de la Villa, Marchena 1990. Vladimirov has had some experience with this variation, so the choice of d2 for the queen - probably the fruits of home preparation - must be taken seriously. In the game White's idea

was revealed after 12

'ii'b6 1 3 lDe5

l:r.d8 14 a3 lDc6 15 lDd3 0-0, when

16 'ii'g5 saw the queen spring into


lDxc5 (avoiding 17 lDxe5? lDxe5

18 'it'xe5, which leaves White open

to captures on f2 combined with

.'ti'xc5 18 'ii'h4 ! .i.e6

19 i.g5 and White was well on top


i.xc6+ .:.xc6 16 bxc3

action on the kingside: 16


lDf6-g4) 17


Open Catalan: 5


(his lead was decisive after 19

20 .i.e3 !).


b) In Barbero-Zichichi, San Ber­

nardino 1 988, Black ignored the

threat to his b-pawn and castled into


the tactical justification being that

12 �bS? lDb4 13 'ii'c4 .i.xf2+ wins

for Black. Instead Barbero played 12

.i.gS, and after 12

produced the new 14 'ii'd2! , improv­ ing on the old 14 'ii'e4 and threaten­ ing to send the queen over to h6. The


15 a3 lDa6 16 l' 1, and the poor a6-

knight and the weak b5-pawn leave Black struggling) 15 'ii'h6 lieS ! 16 1:tad1 .i.f8 ! 17 'ii'h5 'ile7, when 18

a3 ! would have given White a clear

19 lDxb5 !,


1:txc6 21 .txc6 lDxd 1 22 .txe8, or

18 .• .i.xf3?


plus, for example 18

19 lbd4 lbxc3 20 lbxc6

game continued 14

lDb4 13 .i.xf6 gxf6

safety (or so he thought) with 11






19 .txf3 llc5 20 'ii'g4+,

despite being a sensi­

ble-looking move, is rather passive,




for structural weaknesses on both

sides of the board.

12 lDe4

GM Andrei Sokolov has shown

that 12 lDbS achieves no more than

approximate equality:

.i.e3 .txe3 14 'ii'xe3 lDe7 ! 15 lDd6 (an attempt to improve on Yusupov­ A.Sokolov, Riga Ct (2) 1 986, which also gave White nothing after 15

1:tfd 1 lDed5 16 'it'd3 'ilb6 17 lDbd4

h6!) 15

'ii'xd6 18 lDxe7+ 'ii'xe7 19 1:tfc 1 llfc8 20 l:Xc7 1:txc7 21 a3 'h- 112 Ehlvest-A.Sokolov, Belfort 1 988. If the knight is not dangerous once it arrives on d6, then there is little rea­ son for 12 lDb5 . The text removes the f6 -knight, which is a key defen­ sive piece.

16 lDe5 .i.c6! 17 lDxc6





12 lDxe4




Smagin's addition to the three ex­

isting moves:

a) 13 ••.'ir'b6?! 14 'ii'g4 0-0 15

.th6 .td4 16 1:tad1 ! f5 17 'ii'h4 .i.f6

allowing White to develop smoothly


.tg5 llcd8 19 .txf6 gxf6 20 1:td6

with 12 .tg5. Then the pin on the h4- d8 diagonal combined with White's

and Black is under pressure, D.Pau­ novic-Baquero, Belgrade GMA 1988.

planned build-up on the d-file practi­


In Gulko-Renet, Clichy 1 986

cally forces Black to make some sort

Black provided his light-squared

of compromise. A.Petrosian-Mar­

bishop with room to manoeuvre:

janovic, Erevan 1989 is a good ex­



14 lDe5 ! .i.b5 15 .te3 1i'b6

ample of the problems Black can


.txc5 1:txc5. After 17 1:tadl ! (17

13 'ild2 (13 'ii'd1 .tc6

14 a3 lDbd5 15 lDe5 gives White an

face: 12


initiative) 13


14 .txf6 gxf6 (or

a4 bxa3 18 bxa3 'ii'c7 19 lDd3 .txd3

20 'ii'xd3 with equality, Bronstein­

A.Sokolov, Moscow 1983) 17




15 a3 lDc6

16 lDe4 'fie?


lDd7 .txd7 19 llxd7 00 20 'it'd3


b4) 15 a3 lDc6 16 lDe4 .te7 17

lLlf6 21 llb7 'ii'a5 22 a3 ! ltJd5 23

1:tacl �f8 18 1:tfd1 lDb8 19 l:.xc8 'ii'xc8 20 1:tc l 'fid8 21 lDc5 .tc8 22

.txd5 llxd5 24 axb4 'ii'a4 25 'ii'e3 a6 White had a slight endgame advan­

'ii'c3 and Black was beginning to pay


Open Catalan: 5



c) After 13••.0-0, 14 lbg5?! f5 15

'ii'c4 lbd4 works out well for Black, therefore White should play the sim­

ple 14 lld1, when 14

f6 16 .i.e3 is enough for an advan­

tage on account of Black's loose queenside and inferior pieces. White turned his attentions to the king­ side in Fominykh-Ruban, Voronezh 1988, endeavouring to win the game with his g-pawn( !). Instead of 15 .i.g5, which forces the uncomfort­

able reply 15

16 llac 1 llfd8

White played 15

.i.f4, and after 15

he lashed out with 17 g4, which was

\We7 15 .i.g5!



a1) After 14•••.i.xe3 15 'ii'xe3 White is slightly better, though with­ out the weakness on e6 Black has much less to worry about than in the main game.

f5 (transpos­


ing the position after 15 'ii'd3 .i.xe3

16 'ii'xe3 e5 17 a3 !? as slightly in White's favour in view ofhis pull on the queenside. This does seem to of­ fer White reasonable chances of gen­ erating something from the pressure, but at least White must be on the

lookout for

f5-f4) here,

to 14 .i.g5 f6 15 .i.e3 f5 ), assess­


Stohl gives 14



then followed by the bizarre finish

whereas in the main game (with



18 g5 l:lxd 1+ 19 l:lxd 1 e5

Black's pawns on e6 and f6) he en­

20 .i.c l l:ld8 21 lle1 lbd4 22 gxh6 f5 ! 23 'ii'b1 lbxf3+ 24 exf3 .i.xf2+ 0- 1. Fominykh's oddity is a warning. This is most definitely not the way

joys more freedom. b) Stohl-Smagin, Prague 1992 went 14 lld1 e5! 15 .i.e3 f5 (Black wants to avoid drifting into a passive position, as simply completing de­

to play this variation for White, who

velopment with 15


16 'ii'xe3

should be looking to his better­

0-0 leads to a better game for White

placed pieces and superior pawn