2 Secrets of Opening Preparation

Dvoretsky

I

Yusupov

·

Secrets of Opening Preparation

PrOgress inCfiess

Volume 23 of the ongoing series

Editorial board GM Victor Korchnoi GM Helmut Pfleger GM Nigel Short GM Rudolf Teschner

2007 EDITION OLMS

m

Mark Dvoretsky and Artur Yusupov

Secrets of Opening Preparation School of Future Champions 2 Edited and translated by Ken Neat

2007 EDITION OLMS

m

4

Books by the same authors:

Mark Dvorefsky, Artur Yusupov, School of Future Champions Vol. 1: Secrets of Chess Training

ISBN 978-3-283-00515-3

Available

Vol. 2: Secrets of Opening Preparation

ISBN 978-3-283-00516-0

Available

Vol. 3: Secrets of Endgame Technique

ISBN 978-3-283-00517-7

In Preparation

Vol. 4: Secrets of Positional Play

ISBN 978-3-283-00518-4

In Preparation

Vol. 5: Secrets of Creative Thinking

ISBN 978-3-283-00519-1

In Preparation

Vol. 1: Endgame Analysis

ISBN 978-3-283-00416-3

Available

Vol. 2: Tactical Play

ISBN 978-3-283-00417-0

Available

Vol. 3: Strategic Play

ISBN 978-3-283-00418-7

Available

Vol. 4: Opening Developments

ISBN 978-3-283-00419-4

Available

Mark Dvorefsky, School of Chess Excellence

Bibliographic information published by Die Deutsche Bibliothek Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the internet at http://dnb.ddb.de.

© 2007 Edition Olms AG Willikonerstr. 1 0

·

CH-8618 Oetwil a. S./Zurich

E-mail: [email protected] Internet: www.edition-olms.com All rights reserved. This book is sold subject to the condition that it shall not. by way of trade or otherwise. be lent. re-sold. hired out or otherwise circulated in any form of b inding or cover other than that in which it is published and without a similar condition including this condition b e ing imposed on the subsequent purchaser. Printed in Germany Editor and translator: Ken Neat Typeset: Arno Nickel

·

Edition Marco. D-1 0551 B erlin

Printed by: Druckerei Fri edr. Schmucker GmbH. D-49624 Lbningen Cover: Eva Konig, D-22769 Hamburg ISBN 978-3-283-00516-0

s

ctJ

Contents I ntroductio n (Mark Dvoretsky) .... ..... . . .

.

.

. . . . .

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

.

.

. . .

.

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

.... .......... .... ... . . 6 .

.

. .

. .

. .

.

PARTI General P rinciples of Opening Play (Artur Yusupov) . . .

.

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

........ .. .

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

.

. . . .

.9

Logic in the Open ing (Mark Dvoretsky) ..................... . . . . . . . . . .... . ....... ............ . . .. . . . . . . . . . . . . . ...... 30 Su rprises i n the Opening (Artur Yusupov) ... . . ...... .

.

.

. . . . . . . .

.

. . . .

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

. .

. . . . . .

.. . ... ...45 .

. . .

.

The creative Solvi ng of i rrational Problems i n the Open ing (Sergey Dolmatov) .............. 62 .

A practical Exercise (Mark Dvoretsky) ......... ...... .. .

.

. . . . . . .

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

. .

.

.

. .

. . . .

. . . ... ........... 84 . . .

.

. .

.

PARTII The Development of a n Opening Repertoi re (Mark Dvoretsky)

. .

. . . .

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

. . .... 95 .

. .

. .

King's I ndian Attack, from Wh ite's Point of View (Mark Dvoretsky) . ...... ................ . .. 1 26 .

Preparation for a Game (Artur Yusupov) ....... .

. . . . . . . .

.

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

.

.

You were rig ht, Monsieu r La Bourdon nais! (Yuri Razuvaev)

. . .

...

.

.

. . . . .

. .

. . . . . .

. . .

. .... ................ 1 56 . .

.

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

.

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

. ... 1 70 .

PARTI l l How do Opening N ovelties orig i nate? (Boris Zlotnik) .. .. .

. . . . . .

.

The M ove . . . g7-g5 in the F rench Defence (Aiexey Kosikov) Opening Research (Vladimir Vulfson)

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

... . .. .

.

. . . . . .

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

.

. . .

. . . . . . .

... 1 8 1

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

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

.

. . .

.. . .. . .

.

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

1 92 202

PAR T IV Middlegame Problems (Mark Dvoretsky) ................ . . . . . . .

.

.

. . .

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

The Con nection of the Opening with the Endgame (A/exey Kosikov) In the Footsteps of one Game (Mark Dvoretsky)

. . . . . . . . . .

.

. .

.

.

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

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

.................. .. 2 1 1

.

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

224 234

PARTV Games by Pupils of the School (Artur Yusupov)

I ndex of Playe rs

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

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

.

.

.

.

.

. .

. . .

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

. 255

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

274

I ndex of Openings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277

6

Mark Dvoretsky

Introduction

W terprise, one always endeavours to

hen embarking on some serious en­

plan it as well and as accu rately as possi ble. And if the project proves to be a good one, things usually develop successfu lly. Grandmaster Yusu pov and I d iscussed i n detail the directions a n d principles o f the worki ng of the school for g ifted young chess players , which we were i ntending to organ­ ise. And now it is pleasant to report that our main ideas have withstood the test of time. An indication of this is provided by the successes of pupils from the schoo l . Many of them have already become q u ite strong players, victors and medal-wi nners i n the junior championships of the cou ntry, E u rope and the worl d . In 1 992 alone, fou r of o u r pupils became champions o f t h e world o r Europe - l lakha Kadymova , l n na Ga ponen­ ko , Alexey Alexandrov and Vadim Zviagin­ tsev. Our main principle was a rel iable one, si nce earl ier it had been verified on our own experience . We realised perfectly well that we should not simply convey specific chess knowledge to the pupils - over two ten-day sessions in a year you can't show much , and besides, this is by no mea ns the most important thing when teach ing chess . It is fa r more important: a) to fam i l ia rise p u­ pils w ith the genera l ideas, methods and procedu res of p lay- they have un iversa l importance; b) to revea l rationa l ways of working on c hess and means of master­ ing genera l ideas, as well as obtaining necessa ry spec ific information; c) to analyse defic ienc ies in the p up i l s' p lay and help to elim inate them .

Another idea wh ich justified itself was the conducting of thematic sessions. Each of o u r meeti ngs was devoted to some d i rection of work on chess. The 'massed ' attack i n this d i rection - lectu res, practica l exercises, and add itional material which we provided for the pupils - gave a strong i mpetus for improvement in the g iven field . We would l i ke to be able to help any player who wishes to play more strongly than before , and who for this aim is prepared to work seriously on self-improvement. B ut the n u mber of pupils at the school is l i m ited . Therefore rig ht from the start we plan ned to prepa re instructional books, com b i n i n g lec­ tures read at the sessions with the most i nteresting articles on the given topic. The fi rst such book, based on the materials of the fi rst session of our schoo l , was publ ished in 1 99 1 . Secrets of Chess Train­ ing (Oims 2006) was a new and g reatly expa nded editio n . In it the most general questions of studying chess are analysed the disclosure and e l i m i n ation of a player's weaknesses, the tech nique of a n a lysing your own and other players' games and the extraction of u sefu l i nformation from the m , t h e role o f t h e classical heritage, and s o o n . Now you have before you our second book. It is devoted to the open i n g . What new ideas ca n we offer here to the reader? It is worth speaking about this in more deta i l . It is probable that a good 50% o f all chess books are monog raphs, analysing a specific opening variation or a col lection of them . Essentially these are reference books, and someti mes you have to consult the m , but to master opening theory with them alone is

Introduction

not easy. They offer too much information , the greater part of which is completely superfluous, and there i s too l ittle explana­ tion of the general ideas typical of the variations i n q uestio n . I n additio n , opening monographs very q u ickly become outdated . For chess a mate u rs who want to learn quickly how to play a particular ope n i n g , there are appropriate books, w h i c h analyse only the necessa ry m i n i m u m n u m ber of variations. They a re indeed very u sefu l . But to ach ieve genuine m astery i n playing the open ing stage of the game, using only ready-made prescriptions, i s i m possible. You need to study the methods of opening preparation i n genera l , reflect on the typical problems which other players encou nter, and i ndependently analyse open ing sys­ tems that appeal to you . Our book will help you i n this work. I t i s i ntended for players ( i n particular, you ng players) wishing to deepen their u nder­ standing of chess in general and the open ing stage in particu l a r, and to learn to work i ndependently on the ope n i n g . T h e first part o f t h e book describes the problems you will encou nter i n playing the open i n g , and what is needed to successfully solve them at the board . Here the central place is given over to lectu res by World Champio n s h i p Candidates Artu r Yusupov and Sergey Dolmatov (Sergey is an active collaborator at our school ). In my view, it i s exceptionally interesting to follow t h e thought train of these outstanding grandmasters, who openly describe what they th i n k a bout during a game, how they find the best m oves , and why they someti mes go wrong . But i n order to learn to take correct decisions , an acquaintance with 'theory' alone is insufficient. P ractical tra i n ing is also needed . At each session of the school we i nvariably a rra nge various events , competi­ tions, and so o n . One such tra i n i ng session is described i n the fi rst part of the book.

7

The second part is devoted to the develop­ ment of a n opening repertoire , and prepara­ tion for an event or for a specific opponent. This topic is contin u ed i n the th i rd pa rt, wh ich talks a bout the i ndependent a n alysis of opening positions and the tech nology of devising novelties. The fou rth part traces the con nection of the opening with other stages of the game, and demonstrates the conti nu ity of chess idea s . I n it t h e central idea, permeating t h e entire book, is especially emphasised : the key to you r succes s l ies not in the mecha n ical memorising of opening i nformatio n , but in assimilating the wealth of chess ideas and improving you r chess culture . However, a lthough it was n o t our m a i n objective, t h e reader will also fi nd a consid­ erable amount of u sefu l specific information - open ing novelties (some of which have not yet been employed in practice ), recommen­ d ations on the playing of the most d iverse positions, and reviews of a nu mber of opening system s (King's I ndian Attack, Closed Va riation of the Ruy Lopez, Queen's Gam bit Accepted, and others ) . F i n a l ly, b y trad ition w e conclude t h e book with extracts from the games of pupils at the school , a n n otated by g randmaster Yus upov. Here there are models of h ig h-leve l , fu ll­ blooded opening struggles, a s wel l a s examples o f i nstructive opening m i stakes . I n h i s comments on the pupils' games , Yusu pov focuses on those general prob­ lems of opening play, which he described in the lectu re which beg i n s our book. When studying his lectu re , it probably makes sense to i m mediately refer to the concl uding chapter, and examine them i n para l l e l . This book has been created b y a team of a uthors . Apa rt from the a uthor of these l i nes and grandmasters Yusu pov and Dol matov, it also includes a rticles by g randmaster Yuri Razuvaev and national masters Boris Zlotnik,

8

Introduction

Alexey Kosikov and Vlad i m i r Vulfson . I a m sincerely grateful t o all o f them . I hope that the compiled views of different experts on one and the same problems will prove interesting to the readers , and impart a certa in diversity to the book. If our work should force the reader to th i n k about the difficult b u t fasci nating problems

of open ing preparatio n , and suggest new ideas to h i m in this fiel d , the authors will consider their objective fulfi l led . For this new edition the text has been checked anew, many analytical corrections have been included , and the chapter on the King's I ndian Attack has been considerably expanded .

9

PART I Artur Yusupov

General Principles of Opening Play

L does the strategy of opening play com­

et us ask ourselves the questi o n : what

prise? If you look at the games of strong masters, you will see that the players a i m above all for t h e rapid mobil isation o f their forces. And th is is understandable: the greater the n u m ber of pieces i n play, the greater the attacking possibilities. Rapid development is the basis of opening pl ay. There is a second im portant factor: from the very fi rst moves a struggle for the centre develops. The centre is a kind of dominant heig ht in a chess battle : whoever seizes it will su bseq uently have the better prospects . Natu rally, from the very sta rt both sides focus on the central sq uares e4 , e5, d4 and d5. As a rule, players aim either to occupy the centre with pawns, or to develop piece pressure on it. Along with this, the two sides try to forestal l each other's i ntentions. It makes sense with some move to delay the opponent 's development or hinder his acti ons - it is probable that the 'loss of a tempo ' will s ubsequently be justified . Y ou sh ould not begrudge a tempo spent on prevent­ ing him from castling with this you will consolidate you r lead in d evelopment. Thus the th ird principle of opening play is to fight against the opponent's ideas, with the aim of h i ndering his development and preventing h i m from gaining control of the centre . -

You a s k , which is more i mportant: to develop you r own pieces or h i nder the

opponent's development? Of cou rse, the ideal is if you can combine the one with the other. B ut if there is a choice , in every specific case you must act in accordance with the situation- here there is no u n iversal prescription. B u t even so, it is better not to forget about you r own development. What else concerns a player in the opening? Of cou rse, the form ing of the pawn structure. It is possible that at an early stage you will be able to provoke a weakening in the opponent's pawn formatio n , or, as they say, spoil his pawns. Remember: much de­ pends on what sort of pawn stru cture you obtain - favourable or unfavourable . And , fi nal ly, from the very first moves a struggle for the initiative develops, and this is perhaps the chief essence of opening play. Can you imagine a game nowadays, where for some time the two players simply bring out their pieces, and then beg i n looking to see what has hap­ pened and what do to next? Of cou rse not. It is natural that Wh ite, who has the right of the fi rst move , should u s u a l ly aim i n the opening to gain a lead i n development, occupy the centre and be the fi rst to create threats . Before turning to specific exa mples, I should d raw you r attention to a nother importa nt factor. The modern handling of the open­ ing is inseparably linked with a plan of play in the middlegame (and sometimes even the contours of a future endgame have to be taken into account!). And today it is hard to d raw a clea r line between the

10

�

General Principles of Opening Play

opening and the middlegame, especially since, although i n somewhat different form, all the listed principles of opening strategy are also appl icable to the m iddlegame.

We see how with each move new forces come i nto play. Wh ite occu pies the centre with pawns, while B lack prepares to exert piece pressure on it.

And so, let us examine in more detail the fi rst open ing principle (rapid mobil isation of the forces). We should remember several simple rules:

Not a frequently-occu rri ng move, but a perfectly possible one. B l ack develops h i s bishop and establ ishes control over the central e4-sq uare, with the intention of playing 8 . tt'ie4 .

7 tt'ic3

1 ) do not move one and the same piece twice (there should be a serious motivation for such moves); 2) don't waste time on prophylactic moves with the rooks' pawns - it is more importa nt to develop the pieces q u ickly; 3) don't bring out the queen prematu rely: the choice of a place for it is an exceptionally important problem, si nce the character of the subseq uent play can largely depend on the position of the quee n ; 4) don't launch a premature, u n p repared attack;

i.fS

. .

8 dS

Here there are also other conti nuations .l::i. e 1 , b2-b3 and tt'ie 1 . The move made is also q u ite logica l : with gain of tempo W h ite i ncreases his spatial gains in the centre . 8 . . .

tt'ias

For the second time in the opening B l ack moves the same piece, but i n the g iven instance this is j u stified : firstly, he was forced to do this, and secondly, 8 . tt'ia5 creates a counter-th reat to the c4-pawn . . .

5) don't go pawn-g rabbing, especially in open positions, where a lead in develop­ ment has enormous sign ificance; remember that a tempo in the opening is sometimes more importa nt than a pawn . We will now analyse a game of m i n e with grandmaster Boris Gu lko. I th i n k that from its example the principles of mobilising the forces and opening play i n general will become more understandable.

Yusupov - G u l ko Reykjavi k 1 990

King's Indian Defence 1 d4

tt'if6

2 c4 3 tt'if3

g6 i.g7

4 g3 5 i.g2

0-0

6 0-0

tt'ic6

d6

Now the q uestion arises: what should Wh ite play? 9 tt'id2?!

The move made i n the game has certa i n d rawbacks. A t d2 t h e knight blocks t h e path of the bishop on c1 . I was hoping that the th reat of 1 0 b4 would force the opponent to block the position on the q ueenside by

General Principles of Opening Play

. . . c7-c5. Then Wh ite gains a tempo by advancing his pawn to e4 , and later com­ pletes his development by 'ii'c2 , b2-b3, � b2 and so o n . Alas, events took a d ifferent turn .

9 l2Jd4! looks more natural and logica l . The bishop at f5 is attacked , and the c4-pawn is indirectly defended : if 9 . . . lt:Jxc4? there fol­ lows 10 tt::lxf5 gxf5 1 1 ifd 3 , and Wh ite regains his pawn , obta i n ing the better pawn structu re. After 9 . . . i.. d 7 1 0 ifd3 Wh ite has a promising position . 9. . .

c6 !

After g a i n i ng some lead in development, my opponent resolutely opens the position . Now I was faced with a new problem : how to extinguish Black's i ntended i n itiative? I was unable to solve this task and I got i nto difficulties . The logical follow-u p to 9 lt:J d2 was the energetic 1 0 b4 ! ? . The main idea , which I did not see d u ring the game, is after 10 . . lbxd5 1 1 cxd5 i.. x c3 to play 1 2 e4 ! ( I only considered 1 2 i.. a 3? i.. x d2 ). It i s probable that Wh ite w i l l g a i n two pieces for a rook, but in return B lack will have several pawns. It is not easy to evaluate such a situation ; it is q u ite possible that we have here a position of dynamic equilibri u m . .

I n the g a m e Wh ite played routinely. i.. g 4! 1 0 e4? ! With this cu nning move B lack provokes new weaknesses: either the advance f2-f3 , which al lows B l ack tactical possi bil ities involving . . . ifb6+ , or the move of the q ueen to c2 , where after the opening of the c-file it will come u nder the u n pleasant press u re of the rook on c8 . 1 1 'Yi'c2

cxd5

12 cxd5

Captu ring with the e-pawn would have been even worse: i n this case the black bishop would have gai ned the excellent f5-square . 12 . . .

l:tc8

11

Note how consistently G u l ko brings new forces i nto play, g radually building u p the pressu re . Exploiting Wh ite's routine play, which has hampered his development, B lack has already seized the i n itiative. Now he is obl iged to follow a very i mporta nt principle, which was formulated long ago by Steinitz: th e player with an advantage is obliged to attack, as otherwis e h e risks losing his advantage! 1 3 .l:!.e1

Wh ite tries somehow to establish coord i na­ tion between his pieces. He prepares to conti nue his development with 1 4 tt::lf1 and at the same time he prevents the manoeu­ vre of the black bishop via e2 to a6, where it would exert dangerous pressu re . 13 . . .

b5

Here there were a l s o other possibilities, 1 3 . . . if b6, for example. 1 3 . . . i.. d 7 also came i nto consideration, vacating the g4-sq uare for the ma noeuvre of the knight to e5 and planning to meet 14 lt:J f1 with 14 ...lt:Jc4. 1 4 a3

Here too it was possible to play 14 . . . i.. d 7!? 1 5 lt:Jf1 lt:J c4 16 tt::l e 3 tt::lg 4. But Gulko found a much more i nteresting continuation , based on the same general evaluation of the positio n : Black has a lead in d evelopment, and therefor e it is advantag eous for him t o open up th e gam e.

12

� 14 . . .

General Principles of Opening Play

e6!

After the exchange of pawns on d 5 the bishop will obtain the f5-square , which is extremely unpleasant for White . If he replies 1 5 h3? ! , there follows 1 5 . . . exd5 1 6 hxg4 d4, and my pawn structu re will be hopelessly spoiled , i . e . B lack tra nsforms one form of advantage into another: a lead i n develop­ ment into a su perior pawn structu re. I had to try another method of defence, which can be expressed roughly as follows: when everything is bad , it is already too late to fear anyth ing! 1 5 'ii'd 3

What to do? I was already resigned to the fact that I would be unable to complete my development by normal means, and I try everything possible to complicate the play. Such tactics ca n sometimes g ive q u ite good practical results, although with correct play on the part of the opponent they should probably be pun ished . Objectively the q u iet 1 5 t'Llf1 ( 1 5 . . . exd5 1 6 exd5) was neverthe­ less the lesser evi l . 15 . . .

exdS

1 6 t'Llxb5

Whereas for Wh ite the open ing has not yet ended , Black, of cou rse, is a l ready deep in the middlegame. Now he could have played 1 6 . . . 'tlfb6! ? , secu ring a retreat for his bishop to d7. The adva nced position of the bishop at g4 is perhaps the only defect of his game (there is a possibil ity that the bishop may be cut off from the main forces). But from the sta ndpoint of fig hting for the in itiative , Gulko found perhaps an even better move . 16 . . .

Ue8 ! ?

not yet possible to predict how it will all e n d . 17 h3

i.fS

A tempting reply. Also possible was 1 7 . . . i. e6, i n reply to which there would have followed 1 8 exd5 i.f5 1 9 .l:r.xe8+ 'it'xe8 20 'iff1 . Wh ite is a pawn u p , althoug h B l ack, of cou rse, has strong cou nterplay. 1 8 g4

The only move. Wh ite, a l beit at the cost of a pawn , nevertheless ma nages to complete his development. 18 . . .

i.xe4

1 9 t'Llxe4

The knight has finally left the d2-sq u a re! 19 . . .

t'Llxe4

In view of Wh ite's reta rded development, reg a i n i ng the pawn by 20 i. xe4?! would have been equ ivalent to capitulation : 20 . . . dxe4 2 1 'it'xd6 'ii' h4 ! . 2 0 'ii'xd5? i s also bad on account of the very strong reply 20 . . . .l:!. e5! . 2 0 t'Llxa 7 ! ?

I n t h i s g a m e White acts agai nst all t h e ru les, and with accu rate play by the opponent he should have been pun ished . But I was aware that after the ' normal' development of events I had noth ing to hope for. 20 . . .

.l:!. b8 'ir' I n the event of 2 1 xd5? ! the tactical stroke 21 . . . t'Ll xf2 ! ? suggests itself, for example: 22 .Uxe8+ �xe8 23 'it>xf2 t'Ll b3 24 t'Ll c6 with a n u nclear game. But the inclusion o f 2 1 . . . Ue5! would appear to set Wh ite altogether i n solu­ ble problems: his pieces are uncoord i n ated , and the same terrible blow on f2 is th reat­ ened (22 'it'd3 t'Ll xf2 23 � xf2 'it' h4+ ! ) .

Naturally, he must exploit t h e opportun ity to bring his h itherto inactive bishop into play.

Yet another black piece comes i nto play . . .

21 i.e3

From this moment on, balancing on the edge of the abyss , I time after time found resou rces for conti nuing the struggle. I was simply fortunate that the position proved qu ite complicated and for the moment it is

22 .Uab1

i.xb2

If 22 Uad 1 the reply 22 . . . t'Ll c3 is u n pleasant, for example: 23 Ud2 d4 24 Uxb2 1:1.xb2 25 'ii'xd4 t'Ll e2+ 26 Uxe2 Uxe2 , and if 27 i. h 6 , t h e n simply 27 . . . U8e5, blocking t h e danger-

'Ll

General Principles of Opening Play

ous diagonal and remaining with a great material advantage.

22 . . .

l:tb3

Gulko conducts the game purposefully and finds a concrete way of increa�ing his advantage.

23 'ii'x d5 White would have lost quickly after 23 'ii'c2 'i'b8 24 lt::lc6 lt::lxc6 25 'ii'xc6 .l:.c8.

23 . . .

lt::l c 3

13

The correct reaction was 24...'ii'd7!. But Gulko did not notice that in the variation 25 'ii'xa5 lt:Jxb1 26 l:txb1 there is the deadly 26 ... �c3!. White would have had to con­ tinue trying to stir up trouble with 26 �c6 (26 .lbe8+ 'ii'xe8 27 �c6 'ii'b8 is no better), but not for long: 26...l:txe1+ 27 'ii'xe1 'ii'xa7 28 'ii'e8+ c:i;g7 29 �e7 .:!.b8, and Black wins. Fortunately for me, however, Gulko was tempted by a queen sacrifice.

24 . . .

lt:Jxb1 ?

2 5 �d8

l:f.xe1 +

26 �f1

A completely unclear position has arisen, in which the threats of the two sides would appear to be mutually compensating. 26 . . .

lt:Jc3

Here there is an interesting calculating problem - find the continuation which promises White the best practical chances. In essence, there are three possibilities:

1) 24 'ii'xb3 lt::lxb3 25 .l:!.xb2 lt::lc5. In my view, Black has the advantage (although it can happen that such a surrender of material will help to repair the position); 2) 24 lt::lc6 - objectively, perhaps, this counter-stroke is the strongest. However, after 24 ..lt::lxd5 25 lt:Jxd8 lt:Jxe3 Black, of course, has an obvious advantage (for example, 26 .l:.xe3 l:tbxe3 27 fxe3 �xa3); .

3) The move which I made in the game, and on which I was pinning my hopes. 24 �g5 ! ? Here Black needed to solve the last difficult problem, after which he would most prob­ ably have gained a deserved win.

27 'ii'd 2 ! 2 7 'ii'xa5? was bad: 2 7. . .lt:Je2+ 2 8 c:i;g2 li:Jf4+ 29 c:i;g1 lt:Jxh3+ 30 c:i;g2 li:Jf4+ 31 c:i;g1 �c3, and therefore White takes control of the f4-square. However, 27 'ii'a8!? was also possible. If 27 ...lt:Je2+? there follows not 28 c:i;g2? li:Jf4+, but 28 c:i;h2! c:i;g7 29 �e7 �e5+ 30 f4!, in the event of 27 ....l:.e8 there is the pretty reply 28 lt:Jc8!, while 27 ...c:i;g7 28 c:i;g2! (but not 28 �xa5? lt:Je2+) enables White to escape in good time from the unpleasant pin on the back rank.

14



General Principles of Opening Play

27 . . .

tt:'!e2+

Dvoretsky suggested the interesting varia­ tion 27 ...lld1!? 28 'ili'e3 tt:'!c4 29 'ili'e8+ c;i;>g7, and now either 30 !il..e7 tt:'!e2+ 31 c;i;>g2 tt:'!f4+ 32 c;i;>h2 tt:'!e6 33 !il..f8+! (33 !il..xc4? !il..e5+ leads to mate) 33...tt:'!xf8 34 !il..xc4 l:.b7, or 30 c;i;>h2!? l:txf1 31 !il..e7l::txf2+ 32 c;i;>g1 h6 33 ..tf8+! (not immediately 33 c;i;>xf2? tt:'le4+) 33...c;i;>h7 34 c;i;>xf2 - in each case with chances for both sides.

28 c;i;>g2

iLc3

29 'ili'h6 29 'ili'xd6 is also possible.

29 . . .

- position after 34

tt:'! c4

29...iLg7 30 'ili'd2 !il..c3 would have led to a draw.

30 tt:'!c6

!il.. g 7

3 1 tt:'!e7+

c;i;>f8?!

A mistake. Correct was 31...c;i;>h8! 32 tt:'!xg6+! fxg6 33 !il..f6! ltb7 34 !il..xe2l:lxe2 35 !il..xg7+ lhg7 36 'iff4! with equal chances.

32 'ifxh7

tt:'!f4+

33 c;i;>h2

c;i;>e8?

In the opinion of grandmaster Igor Zaitsev, Black would still have retained excellent drawing chances by continuing 33...l:r.xe7! 34 iLxe7+ c;i;>xe7 35 'ili'xg7 tt:'!e5 (or 35...tt:'!d2) 36 !il..g2 .l:txa3.

34 'ifg8+! Not 34 'ii'xg7? l:txf1 and then 35....:Xh3 mate.

34 . . .

iLf8

Black would also not have saved the game with 34...c;i;>d7 35 'ifxf7 l:txf1 36 tt:'!c6+! c;i;>xc6 37 'ifxc4+ c;i;>d7 38 'ili'xb3.

(see diagram) 35tt:'!xg6! The distance between a win and a loss proved to be very short. After this blow Black's position is no longer defensible (if 35...tt:'!xg6 there simply follows 36 !il..xc4).

. . .

ii.f8 -

35 . . .

fxg6

36 'ii'x c4

l:txf1

37 'ili'xf4 ! ? 37 'ii'xb3 c;i;>xd8 38 'ikc4 (38 h4!?) 38...l:txf2+ 39 c;i;>g3 tt:'!xh3! 40 a4 was also advanta­ geous to White.

37 . . .

.U.xa3

38 iL h4? 38 !il..f6!, controlling the a1-square, was more accurate.

38 . . .

!taa1

39 'ii'e 4+ Also quite good was 39 'ii'f6 l:th1+ 40 c;i;>g3! (after 40 c;i;>g2 White has to reckon with 40...l:.ag1+ 41 c;i;>t3 l:.xh3+ 42 c;i;>e2, and now not 42... c;i;>d7? 43 !il..g5, but 42...!txh4! 43 Wxh4 iLe7 44 'ikh8+ c;t>t? 45 'ifh7+ c;i;>f6) 40...l:.hg1+ 41 c;i;>t3.

39 . . .

c;i;>f7

40 "ii'f 3+

c;i;> g 8

41 'ii'd 5+

c;i;>g7

42 'ii'b 7+

c;i;>g8

43 !il.. g 3 White prepares h3-h4. Afraid of coming under an attack, Black decides to transpose into a difficult bishop ending.

43 . . .

l:t h 1 +

44 'ii'x h 1

l:txh 1 +

General Principles of Opening Play

45 '>t> xh 1

'it>f7

46'1t>g 2

� f6

46 ...'>t>e6 was better.

47 f4

d5

48�f2

i. d6

49 '>t> f3

'it> e6

50� d4

i.e7 !

51 '>t>g 3

i.b4!

52 h4

i. e1 +

53Wh 3

i.d2

54Wg 3

i.e1 +

55 i.f2

i.c3

56Wf3

i.g 7

More tenacious was 56 ... i.f6!? 57 h5 gxh5 58 f5+ Wf7 59 gxh5, but even then it is probable that the position cannot be held White plays i.e3, �g4, i.g5, then 'it>f3-e2d3-c2-b3-a4, and the d5-pawn is lost.

57 i.e 1 ! ? 57 i.b6!?, preparing We3 and i.d4, was also good.

57 . . .

i..f8? !

58 i.c3

i.h6?!

59 f5+ !

gxf5

60 g 5 Black resigned in view of 60...i.f8 61 h5 'it>f7 62 'it>f4.

I should like once again to draw your attention to Gulko's exceptionally consistent play in this game, right up to his fatal mistake on the 24th move. Already in the opening he began fighting for the initiative, and after gaining a lead in development he continued purposefully building on it, bring­ ing more and more new pieces into the battle, not shunning temporary sacrifices. In this connection one can single out his moves 14...e6! and 16....l:!.e8!. As for White, after violating one of the main opening principles (the rapid mobilisation of the forces) and conceding the initiative, he should have lost. But this game is also

15

noteworthy in that it demonstrates the importance of continuing to fight in any circumstances. Yes, White ended up in an inferior position. But he did not lose heart, and tried to initiate counterplay and seize the initiative. In the end he was able to set his opponent difficult practical problems, and the latter deviated from the correct course... The following is a somewhat simpler exam­ ple. In contrast to the game with Gulko, for my meeting with Boris Spassky I was well prepared.

S passky - Yus u pov Linares 1990

Ruy Lopez 1 e4

e5

2 lLlf3

lbc6

3 i.b5

a6 tb f6

4 i.a4 5 lbc3

Spassky often employs this ancient continu­ ation, which at one time was considered virtually the main variation in the Ruy Lopez, but then almost went out of use. From the common sense point of view, there is nothing wrong with the move 5 lbc3: White brings out a piece towards the centre. However, those who have studied the Ruy Lopez will know that it is nevertheless better to play 5 0-0, in order subsequently to create a pawn centre with c2-c3 and d2-d4 (which the knight on c3 hinders).

5 . . .

b5

5...i.c5!? is also possible.

6 i.b3

i.e7

7 d3

d6

8 lLld5 It is tempting, of course, to strengthen the control of an important diagonal and in particular of the central d5-square. But 8

16

w

General Principles of Opening Play

tLldS leads to simplification and effectively to an equal game.

8 . . .

tLla5

9 tLlxe7

'ifxe7

1 0 0-0

0-0

Strangely enough, a novelty. 10...c5 is usually played.

1 1 �d2

12 . . .

This rather inactive move was also em­ ployed earlier by Spassky.

11 . . . .

from the structure of the position. If Black were able to make the moves ... �b7, . ..tLld7 and ...f7-f5, the play would be to his advantage. But he must act so as not to allow the unpleasant sortie of the white knight to h4. Therefore in my preparations for the game I intended beginning with the retreat of the knight.

tLlxb3

12 axb3

tLld7 1

Black has successfully solved his opening problems, without spending time on ... c7c5 - a move which, generally speaking, is useful, but not immediately necessary. Here White should have displayed caution and prepared for the opponent's operations in the centre. But Spassky did not sense in time the danger of his position. Indeed, at this moment it is not easy to imagine that the situation may radically change literally within a few moves.

1 3 l:i e 1 ? ! 1 3 �aS c S 1 4 tLld2, a s suggested later by the ex-world champion, would have led to equality.

13 . . .

Let's think a little about the resulting position. How would you go about solving Black's opening problems? 12 ...c5 ? Well, a perfectly possible move, strengthening Black's control over the cen­ tre. Have you any other suggestions? 12...4Jd7, preparing ...f7-f5 ? A good idea. Develop the bishop at b7? Yes, this comes into Black's plans. But I did not want to play this immediately on account of the strong reply 13 tLlh4. The pieces should be brought out in the most accurate way, taking account of the opponent's resources. Let us sum up what has been said. Indeed, here the undermining of the central e4-pawn by ...f7-f5 suggests itself. It simply stems

�b7

Everything is ready for ...f7-f5. It was high time for White to think about defence, but to his misfortune he decided to pour further fuel onto the fire.

1 4 d4? 14 �gS f6 15 �h4 with the idea of 16 tLld2 was better, when Black has only a slight advantage.

14 . . .

ffi!

Here White could have captured on eS, but the variations would have developed in my favour, for example: 15 dxeS fxe4 16 exd6 �f7! 17 �c3 l::tae8 18 tLlgS 'ii'xf2+ 19 'it>h 1 e3 20 'it'g4 hS 2 1 'ii'g3 cxd6 with advantage. After lengthy reflection differently.

1 5 exf5 1 6 dxe5

Spassky

l:.xf5

played

General Principles of Opening Plav

17

20 .l:txe1

.lli. xf3

2 1 .l:te7

.l:tc8

At the cost of enormous efforts I neverthe­ less managed to win this endgame, but only thanks to a serious mistake by my opponent. This game shows that aiming for control of the centre and paying careful attention to similar actions by the opponent enables many opening problems to be solved. There is another important factor: the specific plans of the two sides are determined by the pawn structure at a fairly early stage of the game.

16 . . .

.U xf3 !

This sacrifice suggests itself. Of course, for the sake of activating the bishop at b7 Black does not begrudge giving up the exchange. The consequences of the primitive 16....lli.xf3 were less clear: 17 gxf3 .l:txe5 (not 17 ... l2lxe5? 18 f4 'ir'h4 19 fxe5 .l:txf2 20 .lli.f4!) 18 l:!.xe5 'ir'xe5 19 'ir'e 1.

1 7 gxf3

tt:Jxe5

18 .lli. f4 18 f4 would have lost quickly to 18 ...'ii'f7 or 18...'i'h4.

1 8. . . .

tt:Jxf3+?!

Here I was rather faint-hearted and decided to take play into an endgame with an extra pawn, which I might not have won. Of course, I should have continued the attack, but as bad luck would have it I made a mistake in my calculations. I thought that after 18....Uf8 19 .lli.xe5 dxe5 20 'ii'd4 a not altogether clear position would arise. How­ ever, it was sufficient to continue this variation slightly, to realise that the opposite was true: 20...ii'f6 21 ii'g4 .lli.xf3, and Black has a powerful attack. [After 22 "ikg3 e4 23

'iixcl its strength would still have had to be demonstrated- Dvoretsky.] In the game, alas, it all turned out far more prosaically.

1 9 'ir'xf3

ii' xe1 +

The next point. When developing the pieces, tried to think what you will be doing a few moves later, what direction the play will take. Only, this should not be taken literally, that supposedly already in the opening you must firmly devise and develop a plan, which will lead directly to a win somewhere in the endgame. No, of course not. I have in mind fairly short operations, of three or four moves, aimed at improving your position and worsening the position of the opponent. In essence, an entire game is a sum of mini-operations, united by a g en eral strat egic id ea, inc orp orated in the open­ ing ch os en by y ou . How such short operations are planned will be seen in the following game.

Y us u pov- Lj u boj evic Tilburg 1987

Queen's Gambit 1 d4

lt:J f6

2 c4

e6

3 lt:Jf3

d5

4 tt:Jc3

.lli. e7

5 .lli. f4

0-0

6 e3 A well-known theoretical position. Ljubomir Ljubojevic chooses a continuation which is

18



General Principles of Opening Play

less well-studied, compared with the usual 6 ...c5.

6 . . .

b6

A normal move, aiding Black's development and conforming to the idea of fighting for the centre. The entire character of the subse­ quent play will largely depend on White's next decision. Let's list them in order: what candidate moves are there? 7 cxd5, 7 ..te2 and 7 .l:tc1.

7 ltc1 I didn't want to capture on d5, as long as the opponent can recapture with his knight. When you hold the initiative, it is better to avoid simplification. Any exchange should be motivated, i.e. bring some positional or tactical dividends. Thus after 7 ... ..tb7 8 cxd5 the capture with the knight is now dubious­ the c7-pawn is lost. The combination of the moves 7 .l:.c1 and 8 cxd5 is in fact the first mini-operation, planned in this game by White.

7 . . .

c5

Pawn tension, typical of many modern openings, has arisen. In such cases it is important to decide for yourself what pawn structure you want to obtain.

chess theory, although according to lnfor­ mator it was a novelty by White (the usual continuation was 8 cxd5). The capture on c5 soon leads to Black acquiring hanging pawns. If he recaptures on c5 with his bishop, he obtains a position with an isolated pawn, to which White also does not object (then the move ... b7-b6 may prove not the most advisable).

8 . . .

bxc5

9 i.e2

..tb7

1 0 0-0

lD bd7

In principle, the opening can be considered complete. It is here, according to the rules given in certain old books, that the two players should start forming plans. In fact, I made my choice much earlier, when I gave a direction to the play with the move 7 .:Ic1. It is clear that the main factor in the battle will soon become the attack and defence of the hanging pawns.

1 1 cxd5

exd5

11...lDxd5!? came into consideration, when White has only a slight advantage. How to intensify the pressure on the opponent's pawn centre? For a start it would not be bad to attack the d5-pawn, by carrying out a small concrete operation: lDe5 and ..tf3.

1 2 lDe5 In the given instance I did not avoid a possible exchange, since there was a definite point to it. Firstly, the knight has vacated the f3-square, from which the bishop will be able to attack the d5-pawn, and secondly, in the event of 12...tDxe5 13 ..txe5 the other bishop will be attacking the f6-knight, one of the defenders of the d5pawn - and in a certain sense there is also a gain of a tempo. Black then has to retreat his knight to d7 and, possibly, even play it to b6. The initiative remains on White's side.

8 dxc5! Of course, this move is not a revelation in

12 . . .

lD b6

What would you have played here?

General Principles of Opening Play

1 3 .U.c2 and then .Ud2 ? Wel l , th is manoeu­ vre is not bad , but I have someth ing else in mind: how would you react to the opponent's last move? Remember an old procedure in such positions: by advancing y our a­ pawn, you emphasise the instability of the knight on b6.

19

and d rive my pieces away from the centre. 15 . . . .Ua6 .U If 1 5 . . . c8 there could have followed 1 6 lt:J a7! with the idea of exploiting the weak­ ness of the c6-sq uare - incidentally, a fairly typical operation with a knight on e5.

1 3 a4!

Now Black faces a difficult d ilemma: on the one hand , he does not want to allow his knight to be d riven away, but on the other hand , after . . . a7-a5 Wh ite gains the b5square. What then should he do? You suggest playing 1 3 . . . d4 ? Let's see: 1 4 exd4 cxd4 1 5 lb b5 lt:Jfd 5 1 6 � g3 and it is hard to defend the d4-pawn. But i n itself the idea is interesting . With th is pawn structu re White must constantly reckon with the possible breakthrough . . . d5-d4. In some cases it may prove rather u n pleasant. 13 . . .

aS

White's first big achievement. Now it was possible to immed iately i nvade with the knight on b5, aiming to use the lb b5 + �f4 construction for operations on the d6- and c7-squares . But after some thought I de­ cided for the moment not to deviate from my plan and to retai n the pressure on the d5pawn . 1 4� f3

�e8

If Black had repl ied 1 4 . . . � c8 , I would indeed have considered the manoeuvre 1 5 .U c2 and 16 .U. d2, in order to strengthen the pressu re on the d5-pawn . But after the move made I could no longer resist the temptation to exploit the b5-sq uare. 1 5 lbb5!

The play takes on a tactical cha racter. The unpleasant th reat of 1 6 lbxf7 and 1 7 i.. c7 is created , and to parry it Black is forced to put his rook in a rather awkward position . At the same time lb b5 is not only an attacking move, but also a prophylactic one: Black was now i ntending to play 1 5 . . . .il.d6

Think now: what wou l d you have played here? Should Wh ite immed iately play ac­ tively, or should he delay this a n d for the moment strengthen his position, denying the opponent cou nterplay? This problem is one of the most d ifficult when trying to convert a positional advantage . I n t h i s g a m e I solved it incorrectly, by h u rrying to beg i n concrete action . Wh ite had available the excellent move 1 6 . b 3 ! , se­ cu rely fixing the opponent's pawn structu re a n d allowing the pressure to be built u p u n h indered . 1 6 lbd3?!

Wh ite's choice involved an oversight i n one of the variations. It is q u ite obvious that in reply Black will advance his c-pawn , wh ich 1 6 b3! would have prevented . 16 . . .

c4

1 7 i.. c 7

In the event of 1 7 tt:Jc 7 cxd 3 1 8 lt:Jxa6 i.. x a6 1 9 i.. c7 ii'd7 20 i.. x b6 i.. b 4 ! Black would have gai ned good compensation for the exchange.

20

w

General Principles of Opening Play

'ii'd 7 !

17 . . .

The wh ite pieces - the knight on b5 and the bishop on c7 - are u nexpectedly hanging. Thus if 18 lt:lf4 there follows the tactical stroke 1 8 . . . lt:lxa4 ! . 'ii'c 8

1 8lt:le5 1 9 b3 ! ?

1 9 ..txb6 .U.X b6 2 0 lt:lg4 would have given a n equal game. 19 . . .

lt:la8!

20 bxc4

lt:l xc7

20 . . . d4?! was d ubious on account of 2 1 i. d5! �f8 2 2 'ii'f3 . ..td6!

21 cxdS 22lt:lc4!

Black would have gained a n obvious advan­ tage after 22 lt:l c6 'ii'd7! (with the idea of 23 . . . lt:l cxd5) 23 e4 lt:l xe4 24 i. xe4 .l:!. xe4 25 lt:l b8 'ike? 26 lt:lxa6 lt:lxa6. Here Black had various possibil ities. One of them was an attempt to simpl ify the position by the cou nter-sacrifice of a piece : 22 . . . i.xh2+?! 2 3 'i!?xh2 lt:l cxd5 . However, after 24 'i!?g 1 with the idea of 25 lt:l cd6 or 25 'ir'd4 White has the advantage. It was probably best to play 22 . . . ..t b4! , placing the bishop on a defended sq uare . In reply there could have followed 23 d6!? or 23 e4!? (23 . . . lt:lxe4 24 d6 with the th reat of 25 d7). I n the game Lju bojevic committed a tactical oversight, leading to defeat.

25lt:l d 6 ! ! l If 25 .. J xc5, then 26 lt:l xb7. 26 .U.xaS

ir'xd6 ..txdS

{Neither player noticed the clever reply 26 ...i.a 6! with the idea of arresting the white rook on a5. If 27 r!e 1 there follows 27...fib4!, while if 27 ..te2- 27 ...'ikb6!. White is obliged to play 27 .U.xa6lbxa6 28 'it'd4 with roughly equal chances. But this means that the move 22. ..i.c5!? should not be condemned - Dvoretsky.] 27 'ir'd4!

The decisive move. Now the game is decided . 27 . . . lt:l 27 . . . e6? 28 .U. xd5 28 ..txdS

'ir'e6 lt:lxdS

2 9 l:t d1

.l:tc8

22 . . .

..tcS?

30 :cs !

.U. xcS

23lt:lxa5!

l:txas

31 iixcS

hS

32 aS

'ii'e4

33 h 3

g6

34 'ir'c6 35 a6

'it'b4 'it'as

36 'li' b7

'ii'a 4

'it'd8 24lhc5 Here 24 .. . 'it' b8 25 'it'd2 lt:l a6 26 llc4 would not have saved Black. Lju bojevic was hoping to disentangle h imself with the move in the game, but he overlooked a simple rejoinder.

(see diagram)

37l:!.b1

Black resigned. Let u s now single out for what reason this game is of i nterest to us. In particular we see

CLJ

General Principles of Opening Play

that the struggle for the centre led t o the creation of a specific pawn structu re ( h a ng­ ing pawns), which i nfluenced to a sign ificant degree the su bsequent plans of the two sides. Such pawn structures a re a separate theme, which should be studied seriously. Here there a re rules and exceptions to the rules, there a re featu res wh ich a re common to all structures of the g iven type and those which are typical only of some specific positions; a l so known a re the standard plans, one of wh ich Wh ite tried to follow. The second th i n g , to which I consider it important to d raw you r attenti o n , is this: it wou l d seem that, by exchanging his central pawns, already in the opening Wh ite vol un­ tarily conceded the centre . But i n return he set up strong piece pressu re on the oppo­ nent's central pawns and transformed them into rea l wea knesses . Thus the game forces us to th i n k a bout the problem of the transformation of a pawn centre : it may be replaced by pieces , but sometimes it even makes sense to g ive it u p a ltogether. Genera lly speaking , a pawn centre is good not in itself, but for how it can be used for concrete aims. Say, for d riving back the opponent's pieces (remember how we care­ fu l ly looked out for the move . . . d 5-d4 ), to delay his d evelopment, or to obtai n conven­ ient squ a res for our own pieces u nder the cover of the pawns. But if, as i n the game just exa mined , the centre does not fu lfil these objectives, it ca n easily become vul nerable and be transformed i nto a target for the opponent to attack. Of cou rse, all that has been said by no means exhausts the problem of the pawn centre . Very important, for example, is the difficult problem of pawn tensions (between pawns on c4 and d5, d4 and c5 , and so o n ) . Here a whole a rray of questions a rises. Wh ich tension should be retained, a n d which not? H o w should these dynamic tensions be m a i ntained? At what moments should they be transformed i nto more static

21

structures? To some extent the game a lso g ives a n swers to some of these questions. The following example will certa inly gladden supporters of the Dutch Defence . This is a game of m i n e with g ra n d master Alexa nder Bel iavsky, i n which the plans of the two sides were determi ned by a complex pawn structure, so typical of the ' Dutch' . B e l iavsky - Y us u pov

54th U S S R Cha mpionsh i p , M i n s k 1 987

Dutch Defence 1 d4

fS

2 c4

tt:l f6

3 g3 4 ii. g2

e6 dS

Black is aiming for a 'stonewall' set-up. H i s a i m is clear: t o establish as close control as possible over the e4-square - probably it is simply hard to imagine closer control. However, this is ach ieved at the cost of a substantial wea kening of a whole complex of dark sq u a res. The e5-squ a re is deprived of pawn protectio n , a n d the fate of the game will largely be determined by how Black can cou nter the opponent's i ntention to occu py this point. 5 tt:lf3

c6

6 0-0

ii.d6

Earlier Black more often played the 'stone­ wal l ' with his bishop on e7. It seems to me that at d 6 the bishop is more logically placed . Since the d a rk sq uares h ave been weakened , it is desirable that the pieces should defend them as fa r as possible. Why were they afraid to play 6 . . . ii. d6 ? They thought that after 7 ii. f4 , by excha nging bishops , Wh ite would strengthen his control of e5 and gain an appreciable advantage. However, it tra n s p i red that i n this case too Black can put up a perfectly good fight. I n reply to 7 i.f4 I recommend the immediate excha nge of bishops - 7 . . . ii. xf4 , somewhat

22

�

General Principles of Opening Play

wea kening the opponent's kingside. Other­ wise e2-e3 will be played , after which the exchange on f4 will become extremely dangerous for Black: Wh ite will reply exf4 and develop pressu re on the e-file. This procedu re is worth bearing i n mind. 7 b3

Transparently h i nting at the possible ex­ change of the da rk-sq uare bishops on a3 an operation that is rather adva ntageous for Wh ite. 7 . . .

'ike7

I am agreeable to the exchange, but only with the i nclusion of the move a2-a4 . Why? Because in this case the unequal value of the two sides' last moves will tel l : the developing move with the q u een is more usefu l than the advance of the rook's paw n . During t h e t i m e that Wh ite spends on playing his knight from a3 to a more active positio n , Black has time to prepare . . . e6-e5. 8 i.b2 9 lt:lc3

0-0

prepared for t h i s . Therefore I chose the other p l a n , i nvolving the manoeuvre of the bishop to h5, where it will also be taking part in the play. 9 . . .

i.d7 ! ?

1 0lt:le5 1 1 lt:l d3

i.e8

Not having any g reat experience in playing the D utch Defence, Bel iavsky fa ils to fi nd a n effective p l a n . The t i m e spent on playing the knight from f3 to d3 could probably have been used more su itably. I n stead of the move i n the game Wh ite should have strengthened his position in the centre by 1 1 e3 and subseq uently perhaps even played f2-f4. 11 . . .

lt:l bd7

1 2 e3

Not a very good decision . It would appear that Wh ite is operati ng without a defi n ite p l a n . He should have accepted the fact that he has no adva ntage and played 1 2 f4 ! ? , secu ring h i mself a n eq ual game.

A natural developing move . However, I don't l i ke it. The set-up i ntrod uced i n his time by Tig ran Petrosian is much better: develop the knight on d2 and su bseq uently aim for control of the e5-sq uare with both knig hts (lt:lf3-e5-d3 and lt:ld2-f3 ). I ncidentally, i n the Dutch Defence Wh ite should be very cautious about occupying the e5-square. If, after an exchange of pieces, a pawn ends u p on this squ a re , Black w i l l no longer have a weakness on e 5 a n d usually he escapes from his ope n i ng difficulties. Now I have to try and solve the problem of my queenside development, and a bove all the question of my l ig ht-square bishop's 'employment' - virtually the main cause of headaches i n the Dutch Defence. Two fu ndamentally d ifferent plans are possible. The fi rst is to fianchetto the bishop on b7, hoping su bsequently to play . . . c6-c5. H ow­ ever, with his knight on c3 Wh ite is wel l

What can be said about this position? Black is very sol idly placed i n the centre and h i s pawn structu re is sou n d . Therefore h e a l ready has t h e rig ht t o consider active play. 12 . . .

g5!

If y o u do n o t control t h e situation i n t h e

General Principles of Opening Play

centre, it is better to refra i n from such actions. Because i n reply you risk receiving a blow in the centre , dashing you r hopes. But in the given instance there is no danger of this, since Black is closely controlling a whole complex of central sq uares. And the fact that he has set his sig hts on the kingside is quite understandable - nearly a l l of his pieces are ai med there . It follows that Black's plan stems intri nsically both from his pawn structu re , and the a rrangement of his pieces. I th i n k that my position is a l ready somewhat more pleasant.

23

on f4 (after the natu ral 1 6 . . . "ir'xg5), but. . .

Apparently Beliavsky's evaluation of the situation was the same, otherwise he would not have ai med for simpl ificatio n . 1 3 a4

White reverts to the idea of exchanging the dark-square bishops, hoping i n th is way to extinguish the opponent's i n itiative . 13 . . . ii.g6 'ii lt:i .t 13 . . . h5 14 c1 e4 also came i nto consideration, with somewhat the better chances . The point of the move i n the game is to forestall Wh ite's obvious plan: 'iic1 and �a3. Now if 14 'iic 1 there fol lows , of cou rse, 14 .. .f4. 1 4 f4

Beliavsky insists on carrying out his p l a n . 14 . . .

i. h 5

There w a s no longer anything for t h e bishop to do on g6. 15 "ir'c1

Evidently Wh ite should have decided on 1 5 .if3!?. This looks a l ittle strange, but, on the other hand , it is a l ready time for him to th i n k about defence . Besides, i t m a y wel l turn out that soon the bishop on g2 will be a worse piece that its opposite nu mber on h5 - after all, it is ru nning up against the securely d efended d5-pawn . 15 . . .

lt:i e4

1 6 fxg5

Played in the hope of establishing the knight

16 . . .

lt:ixc3 !

1 7 �xc3

i.e2

The knight on f4 could indeed have become a defensive bu lwark, and therefore I happily g ive u p my bishop for it. But for the sake of reta i n i ng th is knig ht, it perhaps made sense for my opponent to pa rt with the exchange: 1 8 lt:if4 ! ? i. xf1 1 9 .l:!.xf1 . 1 8 �fe 1

i.xd3

1 9 'ii'x d3

'it'xg5

Black has ach ieved a ma rked adva ntage. What does it consist of? In pa rticu lar, the structu re of the position . He is threatening the extremely u npleasant plan of attacking the h2-g 3 pawn chain with his h-pawn , after which appreciable weaknesses will arise in Wh ite's positio n . I n order t o neutral ise t h i s p l a n , Beliavsky carried out the fol lowing exchanging opera­ tion : 2 0 i.a3

i.xa3

21 .l:.xa3

lt:if6

22 �f1

h5!

23 'iif4

�xf4

24 gxf4

Despite the simpl ificatio n , the resu lting endgame is stil l d ifficult for Wh ite . I was the fi rst to occupy the g-file and i n add ition the

24

�

General Principles of Opening Play

wh ite bishop proved to be much weaker than my knight. This is a fai rly sta ndard 'stonewall' situation : the bishop ru n s u p against a sol id rock o f black pawns, whereas the agile knight creates threats without hindrance. Later Black was able to win in i nstructive fashion , by taking play i nto a rook endgame. Moreover, agai n a position with a structural adva ntage was created : after the exchange of minor pieces on the e4-sq uare , the black pawn which moved there seriously cra m ped the opponent. Of the instructive featu res i n this game I would single out Black's attitude to the centre: he paid consta nt attention to it and endeavoured to control the importa nt e5sq uare with his pieces . And only after cr eating a secur e position in the c entr e did Black risk b eginning active play on the kingsid e. I will show you two more games from that same 54th USSR Championship i n M i n s k . T h e first, with g randmaster Vitaly Tsesh­ kovsky, is i nteresting for the way that Wh ite made use of his pawn centre . In principle, this is already a midd legame problem, but studying the opening in isolation from the midd legame is not a very advisable occupa­ tion . The typical pawn structu re arising in the opening largely determines the entire su bseq uent play. The modern approach to the opening consists precisely in studying such typical structures, and i n a deeper penetration into their characteristic regu l a ri­ ties and typical playing methods. Yusu pov - Tsesh kovsky

54th USSR Championship, Minsk 1 987

Grunfeld Defence 1 d4

tt:Jf6

2 c4

g6

3 tt:Jc3

d5

4 cxd5

tt:Jxd5

5 e4

tt:Jxc3

6 bxc3

A fashionable variation of the Grunfeld Defence. Wh ite has a pawn centre , wh ich Black tries to undermine and attack with h i s pieces . 6. . .

Ji.g7

7 Ji.c4

0-0

8 tt:Je2

c5

9 0-0

tt:Jc6

1 0 .1i.e3

Here my opponent chose not the most critical pla n . 10 . . .

tt:Ja5 The usual conti n u ation is 1 O . . . Ji. g4. By removing the attack on the centra l d4-paw n , he g rants m e g reater scope for m a n oeuvri n g . 1 1 Ji.d3

b6

12 .l:.c1

"ilic7

The immed iate capture on d 4 ca me i nto considerati o n , since one of the problems of such positions is the d4-d 5 advance. I n many cases i t ca n b e adva ntageous for Wh ite to advance h i s d-pawn , alth o u g h i n reply Black usually in itiates cou nterplay i nvolving . . . c5-c4 and the u n d e r m i n i n g move . . . e7-e6 . For the moment I decided strengthen my positio n . 1 3 'i!Vd2

s i m p l y to

.1i. b7

Strictly speaki ng, the open ing is at a n end ­ both sides h ave developed their pieces. But for a better understa n d ing of this type of position it is usefu l to see how Wh ite uses his advantage i n the centre to develop an in itiative. 14 .1i.h6 On e of th e advantages of contr olling th e c entr e is that it is easier to switch play t o th e flanks. Wh ite considers that u n d e r the cover of his strong centre he a l ready has the right to beg i n active play a g a i n st the

ltJ

General Principles of Opening Play

25

opponent's king. The exchange of the dark­ square bishops comes i nto h i s strategic plan, since the g7-bishop is very active and in addition it is virtually the only defender of the king . .l:.ad8

14 . . . 1 5 h4 1 ?

A theoretical novelty. 1 5 d 5 also came i nto consideratio n . After 1 5 . . . c4 1 6 ..t c2 e6 1 7 . bg7 � xg7 Wh ite plays 1 8 f4! , and if 1 8 . . . exd5 - 1 9 e5 with an attack. In the game I was able to ca rry out th is idea, and i n a n even more favou rable vers i o n . 15 . . .

'it'd6?!

A rather abstract move . The centre could have been attacked more effectively by 1 5 ... ttJ c6 , and if 16 d 5 , then 1 6 . . . ltJ e5, aiming to use the central squares i n order to disrupt the coord i n ation of the wh ite pieces . 1 6 d5

The start of a n i nstructive operati o n , with which Wh ite strengthens stil l fu rther his position i n the centre . Black's reply is forced , si nce 1 6 . . . e6? is bad because of 1 7 c4. 16 . . .

c4

1 7 .i.c2

If Black could have a nticipated the develop­ ment of events , he would n ow have pre­ ferred . . . e7--e5. But Tsesh kovsky eva l uated the position incorrectly and advan ced h i s pawn o n l y one sq u a re . 17 . . .

e6?!

1 8 ..txg 7

'iii> xg 7

And here Wh ite fi nally ca rried out h i s p l a n , involving a positional pawn sacrifice.

(see diagram) 1 9 f4!

The idea is u nderstandable: after 1 9 . . . exd5 20 e5! Wh ite will occupy the excellent central sq uare d4 with his knight, after which there will follow f4-f5 and so on. That is, the

-

position after 1 8 . . . 'iii>x g7 -

pawn centre is effectively replaced by a pawn-piece centre , under the cover of which Wh ite conducts a n attack on the king with great effect. Tsesh kovsky is an experienced player, of cou rse he fu lly appreciated the danger of captu ring the d 5-pawn , and he pin ned his hopes o n attacking the centre . 19 . . .

f5

H owever, here too the d rawbacks of Black's position a re patently obvious: his king is ope n , his knight at a5 is shut out of the game, and in the centre Wh ite has more pawns - all these factors a re very sign ifi­ cant. It is not surprising that a concrete decision is fou n d . 20 ltJd4!

The knight will help the pawns to advance fu rther. 20 . . .

fxe4

21 dxe6 Of cou rse , not 21 ltJ xe6+?? o n account of 21 . . . 'ifxe6 . After making a pawn break­ through i n the centre , Wh ite has g a i ned a powerfu l passed pawn at e6 and created a mass of problems for his opponent. 21 . . .

ltJc6

22 f5 !

22 ..t xe4?! would have been significantly

26

�

General Principles of Opening Play

weaker in view of 22 . . . lt:J xd4 23 'ii'xd4 'it'xd4+ 24 cxd4 i. xe4 25 e7 .l:. xd4 26 exf8'ti'+ �xf8 , and Black has sufficient compensation for the exchange. 22 . . .

lt:Jxd4

23 cxd4

The queens must be retained . N ow Black loses by force after 23 . . . "i¥xd4+? 24 "it'xd4 .l:.xd4 25 e7 .tle8 26 i. a4 . It is easy to see that the captu re on f5 is also hopeless (23 . . . .l:.xf5? 24 � xf5 gxf5 25 "iVg5+ ). As Boris Gelfand commented , he could have de­ fended by 23 . . . .l:. de8 ! ? , after which Wh ite would have contin ued the offensive with 24 i. a4 ! .l:.e7 25 i. d 7 . 23 . . .

'ti'e7

24 i.a4!

[After 2 4 ... .U.d5!? 25 fxg6 Black should play 25. . . n df5! 26 1lxf5 � xf5, and it is not easy for White to demonstrate his advantage. In view of this, 25 :Xc4!? 'J:.dxf5 26 .U.xf5 i:.xf5 27 d5! comes into consideration, for exam­ ple: 27 . . . i.xd5 28 'ti'c3+ � h 6 29 .U.c 7 'it'xh4 30 'iie3+ 'iig 5 31 'iih 3+ 'Viih5 32 fixh5+ 1ixh5 33 e 7 i.f7 34 eafk i.xeB 35 i.xeB with quite good winning chances in the endgame - Dvoretsky.] 24 . . .

l:!.xf5

25 l:txf5

gxf5

26 "ii'f4 !

Now the e6-pawn is indirectly defended (26 .. .'it' xe6 27 'ii'c7+ ), and impending over Black is 27 .l:. xc4 with the terrible th reat of .l:. c7 . 26 . . .

i.d5

26 . . . .l:!.f8 would also not have saved Black i n view o f 27 .l:.xc4 'iifxe6 28 .l:. c7+ .l:!. f7 29 i. b3 i. d 5 30 i. xd5 'ifxd5 31 'ifg 5 + , and W h ite wins by a di rect attack. 'lt>g 6 'ii' 27 . . . �g8 28 l:t c3; 27 . . . f6 28 e7. 27 'ir'e5+ 28 l:tc3

f4

29 h 5+

Black resigned.

The triumph of Wh ite's central strategy! I n the centre h e effectively has th ree con­ nected passed pawns , and the e6-pawn is especially dangerous. What can Black do against this? If 24 . . . gxf5, then Wh ite simply goes into an endgame by 25 'ti'g 5 + , after which the e-pawn ca n be stopped only by giving up the exchange. And in the event of 24 . . . .l:l d5 the goal is ach ieved by 25 fxg6 hxg6 26 .l:l xf8 "ii'xf8 27 l:tf1 l;If5 28 .l:Ixf5 ! , and again the e-pawn forces its way through to the queening square . Therefore Black has l ittle choice .

The result of the game was justified . The play went badly for Black. F i rst he m i ssed a moment when he should have i n itiated a fight in the centre , then he delayed slig htly, and Wh ite got in fi rst with h i s a ctive offensive play. What else can be said? N ote how Wh ite combi ned the th reat of a b reak­ through in the centre with th reats to the k i n g . This is also one o f t h e advantages of a strong centre - at any moment u n d e r its cover one can beg i n an attack on the k i n g ! I n t h e fol lowing g a m e a popular variation of the Modern Benoni was played .

General Principles of Opening Play

Y u s u pov- Dolmatov

54th USSR Championsh i p , M i nsk 1 987

Modern Benoni 1 d4

lt:Jf6

2 c4

e6

3 lLlf3

c5

4 d5

exd5

5 cxd5

d6

6 lt:Jc3

g6

7 i..f4

a6

8 e4

b5

If one approaches th ings formally, such a flank attack by Black should be refuted by play in the centre by Wh ite , and in general such pawn moves a re i n co rrect, prematu re and so on. But in the Modern Benoni the advance of the pawn to b5 has a serious point: it is not so much a matter of play directly on the queenside, but rather a fight against the enemy centre ! Su bseq uently Black wi ll normally d rive away the strong knight from c3 , and i n many cases, if he should succeed i n advancing . . . c5-c4 , his knight will aim for c5 and th reaten the e4pawn . It ca n be said that the m a i n idea of the Modern Benon i is a n attempt to by-pass the enemy centre from the fla n k and exert pressure on it. White, of cou rse , should play i n accordance with the rules, i . e . p repare a breakth rough i n the centre . T h i s is t h e a i m o f his next move . 9 'i!Ve2

lt:J h 5

Black i s a l ready forced to defend against e4-e5. i.. e 7 1 0 ..lt g5 1 1 i.. h 6

The th i rd move i n the opening by one and the same piece! This would appear to be a gross violation of the laws of development. But, after a l l , d u ring th is time Black too has not made any u sefu l moves : he has wasted time on moving his knight away from the centre with . . . lt:J h 5 , and although he has

27

developed his bishop, e7 is by no means its best square , and th is has g iven me a n opportun ity t o p revent Black's castl i n g , wh ich is so necessary for h i m . 11 . . .

i.f8 i.. The best proof that the moves g 5 and ..lt h6 were not a poi ntless waste of time. 12 ii'e3

i.. x h6

1 3 ii'xh6

Now Black should have made use of the respite to develop his pieces. The theoreti­ cal conti n u ation is 1 3 . . . lt:Jd 7 . But Sergey Dolmatov employed a new move. 13 . . .

b4? !

This decision would appear to be well­ motivated : the knight is d riven away from the centre to a less active position . But in so doing Black creates a not very favou rable pawn structu re on the queenside, and the c4-square is tra nsformed i nto a desirable target for a wh ite knight. F rom here it will not only exert pressu re o n the d6-pawn , but also support the e4-e5 breakth rough . 1 4 lt:Jd 1

"it'f6

Black unequ ivocal ly demonstrates that he is also th i n king about active play. For example, Wh ite has to reckon with the i nvasion of a piece on f4 (say, if 1 5 ..lt e2 there follows 1 5 . . . lt:Jf4 ). What should Wh ite do? No way of q u ickly com pleti ng the development of the pieces is appa rent. H e is obl iged to show a l ittle cu n n i n g . 1 5 lt:Jd2!

Aga i n seemingly a move against the rules, but i n fact it is made i n the i nterests of development - g2-g3 is prepa red. In this way Wh ite simu lta neously solves several problems: he shuts the enemy knight at h5 out of the game, the bishop will support the centre from g2, and the knight may su bse­ quently go from d2 to c4 . Black stands sign ificantly worse . Now he should have conti nued his development by

28

�

General Principles of Opening Play

playing 1 5 . . . ltJd7. True, in this case too after 1 6 g3 ltJ e5 1 7 ii. e2! ( it is important to control the d3-sq uare ! ) I have an appreciable advantage. Thus if 1 7 . . . ltJg4, then Wh ite simply exchanges on g4, after which he can play both f2-f3 , and ltJ e 3 . At o n e stroke Dolmatov tries t o solve the problem of his queenside and the c4square , but it is probable that the move made by him is a l ready the decisive m is­ take. 15 . . .

ii.d7?!

than this king move , but, natu rally, this does not lead to anyth ing good . 17 g3

i.g4

1 8 ltJe3

i.f3

Here it was possible to play simply 1 9 ltJxf3 'ifxf3 20 i. g2 'iif6 2 1 ltJc4 with an obvious advantage. But this is no longer enough for Wh ite. He acts more energetical ly, remem­ bering that, when you h ave a lead i n development (although for the m o m e n t it is not so obvious) you must attack! 19 e5!

The decisive opening of the position ! 19 . . .

dxe5

20 ltJxf3 21 i.g 2

11Vxf3 'i!Vf6

22 d6

�a7

Black can stil l put u p a resista nce . Thus if 23 ltJd5 he would reply 23 . . . 'ife6! a n d then 24 . . . l:i. d7, attacking the d6-paw n . Therefore Wh ite does not h u rry. 23 0-0

What should I do now? It is possible, of cou rse , to conti nue the i ntended plan with 1 6 g3, but then Black repl ies 1 6 . . . i. b5, after which Wh ite's main tru mp - the c4-sq u a re ­ will be knocked out: in the event of the exchange of bishops on b5, a black pawn will end up on this square . B u t there is another way: realise w h a t the opponent wants, and try to prevent h i m . Fortu nately there is such a possibility. 1 6 a4!

Now the bishop can no longer go to b5. It transpires that with 1 5 . . . i. d7 Black has merely deprived his b8-kn ight of its lawful square, and now it is not apparent h ow he can complete his development. 16 . . .

'it;d8

My opponent could not fi nd anything better

ltJg 7

Black would have lost q u ickly after 23 . . . ltJ d7 24 ltJd5!? 'iig7 25 'it'e3 followed by 26 ltJ e 7 , w h e n i t transpires that t h e rook a t a 7 is extremely badly placed and loss of material is u navoidable. N ow I need to bring my heavy pieces i nto play, as otherwise the enemy defences ca n not be breached . How to place the rooks is always a difficult problem. I th i n k that h e re Wh ite coped with it successfu lly. 24 l:i.ac1 !

ltJd7

ltJf5 25 l:i.fd 1 ltJ 'i!V ltJ If 25 . . . e6 , then after 26 d5 g 5 27 'i!Vxg5 ltJxg5 the wh ite knight again i nvades at e7. 26 ltJxf5

'i!Vxf5

27 l!Ve3 !

T h i s is why the rook went t o c 1 ! The opponent has no satisfactory defence against the exchange sacrifice on c5. He did not i n fact manage t o complete his development and con nect his rooks.

General Principles of Opening Play

27 . . .

e4

28 �xe4

ife5

29 %1d5 Or 29 . 'ikxb2 30 : dxc5 .

'iife 6

. .

30 l:tdxc5

ltJxc5

31 'ikxc5

Black resigned . Both these games, with Tsesh kovsky and Dolmatov, can be called opening games, since in both cases the opponents essen-

lZJ

29

tially did not manage to get past the openi n g . This is why it is u seful to examine them as a whole. Black seemingly did not make any obvious mistakes, but from these examples we have seen that sometimes it is sufficient to take one or two i nexact deci­ sions - say, i n correctly eva l uate the situa­ tio n , fail to u nderstand the pawn structu re, not beg i n a fight for the centre at the right time, or commit a n inaccu racy i n defence in order to suffer a rapid defeat.

30

� Mark Dvoretsky

Logic in the Opening Dolmatov - Rash kovsky

H helps a player, at the board o r i n home

ow does open ing theory develop? What

47th U S S R Championsh i p , M i nsk 1 979

analysis, to fi nd the correct solution to an opening problem facing him? There is no doubt that here one cannot get by without a n abil ity for improvisation, sharp combinative vision , and accurate calculation of varia­ tions. But in our opening i nvestigations another component is nearly always present and plays a very important role - logic! I should l i ke to draw you r attention to several examples of the logical solving of open ing problems. It is clear that logic does not operate i n an empty space . It is based on specific ope n i ng knowledge, and also on the typical methods and evaluations which we have mastered , and it helps to l i n k all this with the g iven position and as a result to work out the correct decision . The more ideas that we know, the wider the scope for logic, and the deeper and more accu rate our reason i ng becomes. I will remind you of one of the standard methods which occu rs in the Sici l i a n De­ fence, in 'Scheven ingen-type' position s .

(see diagram) It is clear that Wh ite's last move was 1 3 g4. A widely-known strategic pri nciple states : against a flank attack it is desirable t o reply with a counter-blow i n the centre . 13

0

.

.

d5!

Black gained an excellent position . Note the situation which possibly existed before th is. Let us put the pawn back on e6. If here . . . d6-d5, Wh ite repl ies e4-e5, obtain ing the strong point d4 for his knight. Therefore often Black fi rst plays . . . e6-e5,

fixing the e4-pawn , and only then strikes with . . . d6-d5. Any player who plays the Scheve n i ngen Variation or the N ajdorf Vari­ ation must be wel l familiar with this idea: . e6-e5! followed by . d 6-d5 ! 0 0

0 0

I n t h e following examples w e will s e e the influence of this idea on the ta king of decisions both by Wh ite , and by Black. Dolmatov - Lerner

47th U S S R Champions h i p , M i n s k 1 979

Cb

Logic in the Opening

What should Wh ite play? He is intending to complete his development by Si. d2, l:t a e 1 and 'it> h 1 , obta i n i ng an active position . But for choosing the best move these considera­ tions are not enough - it is also necessary to employ the idea of 'prophylactic th i n k i n g ' , which we have a l ready encou ntered m a n y times.

8 f4

0-0

9 Si.e3

"ikc7

1 0 a4

b6

11 Si.f3

Si.b7

1 2 'ife1

lbbd7

31

Let us ask ourselves: 'What does Black want; what ways does he have of continu­ ing?' The answer is now clear to u s : 1 3 . . . e5 and then . . . d6-d 5 . I s this the only possibil­ ity? Hardly - it is probable that 1 3 . . . d5 14 e5 tt:Je4 also has to be taken i nto accou nt. If Sergey Dolmatov had reasoned i n this way, he would s u rely h ave made the move which he h i mself recommends in h i s notes , namely 1 3 lbg3! . Then after 1 3 . . . d5 1 4 e5 the black knight ca n no longer i nvade on e4 , while if 1 3 . . . e5 there would be the excellent reply 1 4 lLlf5 . Unfortunately, in the game Dolmatov did not think about his opponent's i ntentions. 13 'it>h 1 ?

In such positions this move i n itself is q u ite good and usefu l , but here it is u ntimely and it leaves Black with freedom to act. 13 . . .

e5!

1 4 lbg 3

d5!

Black has succeeded i n striking at the centre and he has seized the i n itiative . Smyslov - Hort

I nterzonal Tou rnament, Petropolis 1 973

Sicilian Defence 1 e4

c5

2 lLlf3

e6

3 d4 4 lbxd4

cxd4 tb f6

5 lbc3

d6

6 ii.e2

ii.e7

7 0-0

a6

What do you th i n k should be Wh ite's main plan over the next few moves? Most probably g2-g4-g5 (noth ing is g iven by 1 3 ii'g 3 lb c5 1 4 e5 dxe5 1 5 fxe5 lLlfe4 ! 1 6 lbxe4 lb xe4 ) . After t h e i mmed iate 1 3 g 4 ?! i t is not good to reply 1 3 . . . e5? in view of 1 4 lLlf5 (with gain of tempo) and then 1 5 g5 - there is no time for the cou nter-blow . . . d6-d 5 . It is tempti ng fi rst to attack with e4-pawn with 1 3 . . . lb c5, and only after 14 Si.f2 to conti nue 14 . . . d5 or 14 . . . e5 with an excellent game. However, Wh ite has an u nexpected tactical resou rce: 1 4 g5! lbfxe4 1 5 b4 ! , w i n n i ng a piece , the compensation for which after 1 5 . . . d 5 1 6 bxc5 bxc5 1 7 lb b3 is insufficient. However, Black can simply play 1 3 . . . d5 1 4 e5 lb e4 with chances for both sides. Vasily Vasil ievich - a n experienced , cau­ tious player - preferred to avoid u n n eces­ sary complications. 13 Si.f2 ! ?

Now, when t h e e4-pawn (and the e4-sq uare) are safely g u a rded , g2-g4-g5 is th reatened .

32

�

Logic in the Opening

However, the opponent has various ways of combati ng this th reat. For example, the non-routine move 1 3 . . . g6!? is strong. By depriving the knight of the f5square , Black prepa res 1 4 . . . e5. 1 3 . . . tt'l c5 also comes i nto considerati o n , a g a i n parrying g2-g4. T h e aggressive 1 4 b4? ! tt'l cd7 1 5 g4? seriously weakens the queenside - as was shown by Sergey Shipov, Black exploits this factor with the central cou nter 1 4 . . . d5! . I n 1 979 I was analysing th is position with grandmaster Igor Platonov, and he sug­ gested the thematic, typically Sicilian move 1 3 . . . l:fe8 ! , which is now reco m mended i n opening books. If 1 4 g 4 , then 1 4 . . . e5! is very strong, and after 1 5 tt'lf5 - either 1 5 . . . d 5 ! (the bishop on e7 is defended ) , or 1 5 . . . exf4 1 6 g5 tt'l e5! (Abramov-Akopov, corr. 1 98 1 ) . I n itially I liked t h e central advance 1 4 e 5 , exploiting t h e fact that t h e rook is occupying the e8-sq uare , to wh ich the knight might retreat. For example, 1 4 . . . dxe5 1 5 fxe5 tt'ld5 16 tt'lxd5 i.xd5 17 i.xd5 exd5 1 8 e6 i. f6 1 9 exf7+ 'iitxf7 20 'ii'd 1 , and Black has a weak d5-pawn . Alas, after 1 4 . . . dxe5 1 5 fxe5 tt'lxe5! 1 6 i. xb7 I overlooked the strong intermed iate move 16 . . . tt'l eg4 ! , which g ives Black the advantage.

14 . . .

tt'lc5?

1 5 g5

tt'lfd 7

1 6 l:.d1

Th reatening 1 7 b 4 tt'l b3 ( 1 7 . . . tt'l d 3 1 8 .l:. xd3) 1 8 tt'lde2 . The best chances of a defen ce were stil l g iven by the modest move 1 6 . . JHe8! , with the idea of meeti ng 1 7 b4 with the central cou nter 1 7 . . . e5! ( i n the event of 1 8 tt'lf5 the bishop on e7 is defended ) . 16 . . .

tt'lbB?!

Black has lost the th read of the game. J u st a few moves ago he had normal Sicilian knig hts - now where have they ended up! Using the solid ity of his centre , S myslov calmly strengthens his position and pre­ pares his pieces for an attack. 17 i.e3

tt'lc6

1 8 'i1t'g3

llfeB

1 9 i.g2

i.fB

20 l:tf2

Also a typical Sicilian move - the c2-pawn is supported . 20 . . .

'iid 7?

It was essential to exchange knig hts . H ow­ ever, Wh ite would also h ave stood better after 20 . . . tt'lxd4 2 1 i. xd4 e5 22 i. e 3 .

You see that a logical a nalysis, i n which the calculation of concrete variations based on taking account of typical motifs , has enabled us to penetrate more deeply i nto the essence of the position and even find some new ideas (naturally, they requ i re additional checking). Vlastimil Hort played superficially and soon came u nder a strong attack. 13 . . .

l:.acB?

1 4 g4 1

Now the knight i s d riven back from f6 . Hort vacates the d7 -square for it, but as a result he needlessly merely loses time. He should have resigned h imself to retreating it to e8.

2 1 tt'lf3 !

Black is in a cra m ped positi o n , and so Smyslov avoids the exchange. Besides, he

l2J

Logic in the Opening

has a concrete idea : h2-h4 and then the manoeuvre of the knight to g4 via h2. H ig h class play! 21 . . .

ltJ b4

After th is the knight immediately goes to g4. 22 ttJes

'ikc7

23 ltJg4

ltJd7

24 ..ll d4

Threatening 25 ltJ h6+ . 24 . . .

e5

ttJxeS 25 fxe5 If 25 . . . dxe5 26 ..ll e 3 with an obvious advan­ tage. 26 .l:!.df1

It was essential to exchange the terrible knight on g4. There now fol lows the conclud­ ing combinatio n . 2 7 ..llx e5

dxe5

28 tt:lf6+

�h 8

29 tt:lxh 7 !

l:.e6

If 29 . . . �xh7, then 30 g6+ is decisive 30 .l:txf7

..ll c 5+

31 �h 1

.l:!.e7

2 ltJc3

e5 ltJf6

3 ltJf3

ttJc6

4 g3

..ll b4

1 c4

5 ..llg 2

0-0

6 0-0

e4

33

7 ltJe1

For a long time it was thought that the s h a rper move 7 ltJ g 5 did n ot prom ise Wh ite a nyth i n g . H owever, the o p i n i o n changed after the world champion Ga rry Kasparov, in h i s match with Anatoly Karpov (Seville 1 987) and then agai nst Vasily lvanchuk in the 1 988 U S S R Championship, played h i s knight t o g 5 . 7 . . .

.txc3

8 dxc3

h6

Black usually prevents the exchange of the c 1 -bishop for his knig ht, which is adva nta­ geous to the o pponent. 9 ltJc2

.U.e8

1 0 ltJe3

d6

11 'ii'c 2

aS

32 .l:!.f8+

B lack resigned . An excel lent win by Vasily Vasil ievich . I ts basis was l a id in the ope n i n g , when Black did not manage to meet Wh ite's fla n k attack g2-g4 with the standard cou nter-blow in the centre. To come to the correct decision in the above examples it was necessary to make skilful use of a well-known opening idea . But often far more subtle, less obvious deta i l s of the position have to be taken i nto accou nt. New games played in a particular variation add their arguments to the u nceasing ope n i ng debate, and it is hard to get by without studying them . At one time the fol lowing q u iet variatio n of the Engl ish Open ing was very popular.

Black's last move is n ot essential (theory recommends 1 1 . . . b6), but it is q u ite popular. It was chosen by Victor Korch n o i aga i n st Ka rpov in the 6th game of their world championship match (Bag u i o 1 978). From

34 �

Logic in the Opening

the opening Karpov did not ach ieve a ny­ th ing: 1 2 a4 �e7 1 3 lt'l d5 lt'lxd5 1 4 cxd5 lt'l b8 1 5 ii. e3 ii. f5 1 6 h3 lt'l d7 1 7 c4 b6 1 8 'ii'c3 lt'l c5 1 9 b3 'it'd? 20 '>t h2 �e 7 2 1 ii. d4 f6 22 � ac1 �e8 23 ir'e3 D raw. Games from world championship matches, even those as colourless as the one j ust examined , usually provide an i mpetus for the further development of opening theory. Grandmaster Wolfgang U h l m a n n developed a promising plan of play for Wh ite and successfully employed it. Let us first try to decide for ourselves what should be a i med for here. Wh ite has two bishops. But for the moment this is not yet the 'advantage of the two bishops', si nce their mobil ity is restricted . To open the position and g ive the bishops freedom is Wh ite's main objective. If a n exchange o f knights takes place o n d5, it will then be possible to adva nce the pawn to c4 and occupy the long d iagonal with the da rk­ sq uare bishop. But what about the lig ht­ sq uare bishop? The e4-pawn , which it is ru n n i ng up agai nst, must be removed . Karpov did not even try to solve the p roblem of his lig ht-square bishop. U h l m a n n acted far more energetically. U h l ma n n - Osmanovic

Deci n 1 979

certainly be lost. Wh ite has an obvious advantage . 1 4 exf3 1 5 .l:!.ae 1

ii.d7 �f8

1 6 f4

This is the set-up for which U h l m a n n was aiming.

T h e b i s h o p is p ressing strongly along the h 1 -a8 diagona l . Sooner or later the knight will go to d5, and after the excha nge on d 5 t h e second b i s h o p w i l l come i nto play. I n the futu re Wh ite will adva nce his kingside pawns. The opponent has n o cou nterplay, as the black knig hts lack strong poi nts in the centre , and so Wh ite can play for a win without any risk.

12 ii. d2

�e7

16 . . .

lia b8

1 3 f4!

exf3

1 7 'ii'd 3

lt'le7

I n such positions it is dangerous not to capture on f3 - Wh ite plays f4-f5 , cramps the opponent's light-square bishop, and prepares a pawn offensive on the kingside. I n lnformator the following variation is give n : 1 3 . . . ii. d7 1 4 lt'l d5 lt'lxd5 1 5 cxd5 lt'l b8, a n d now, when t h e knight is a long way from the e5-square - 1 6 f5! . If 1 6 . . . f6 , then 1 7 ii. e3 followed by .l:!.f4 (or i mmed iately 1 7 l:tf4) is strong , while if Black plays 1 6 . . . e3, then 1 7 ii. c1 and l:tf3 , when the pawn will al most

1 8 h3

In such cases Aaron N i mzowitsch used to say that Wh ite had a 'qual itative pawn majority' on the kingside. It resides i n the fact that Wh ite's pawns ca n advance, whereas Black's can n ot. 18 . . .

ii.c6

1 9 lt'ld5

Of cou rse, Wh ite is not agreeable to the exchange of his lig ht-square bishop.

lZJ

Logic in the Opening

19 . . .

ttJexd5

20 cxd5

J::t x e1

21 l:txe 1

.tea

35

22 b4 !

A slight deviatio n from the plan which we have been d iscuss i n g . I n passing Wh ite also harasses the queenside. He wants to create active possibil ities there : after the exchange on b4 to create a passed pawn and press o n the backward c7-pawn . 22 . . .

axb4

23 cxb4

b5

Now a passed pawn does not result, but chronic weaknesses have been created on the c-file. 24 .tc3

.id7

25 .id4

'iii'dB

26 'i!i'c3

.:taB

27 a3

.l:tc8

28 'it>h2

There is noth ing Black can move , and in the meantime U h l m a n n consistently strength­ ens his positio n .

Wh ite has an overwhelming advantage. He can simply play 40 h4! , not fearing 40 . . . tbg6? (40 . . . c6 or 40 . . . c5 is more tenacious) 4 1 .l:txd7 tbxh4+ 4 2 � g3 tLlxf3 4 3 � xf3 'it>f8 o n account o f 4 4 g6 ! . Also very strong was 40 .t h 5 ! ? g6 41 .tf3 ..l:!. e8 42 .if6 (or 42 .l:!.xe8 .t xe8 43 .tf6) . U hl m a n n preferred a spectacular, combina­ tive way to the goa l . 40 g6!

tbxg6 ! ? 4 0 . . .f6 4 1 h 4 o r 4 0 . . . .t e8 4 1 gxf7+ .txf7 42 .i h5! g6 43 .t g4 (43 i. f3 ) was hopeless.

28 . . .

l:taB

29 l:te3

l:tcB

42 'it>g3

tLlxf3

30 .tf3

lLleB

43 'it>xf3

�8

31 'it>g2

.l:t a8

32 'ii'e 1

The battery on the e-file has been correctly set up - rook i n front, q ueen beh i n d . 32 . . .

tbf6

33 g4

lLlh7

34 .ib2 35 'ii'c 3

:tea 'ii'f6

36 'ili'xf6

tLlxf6

37 g5

hxg5

38 fxg5

lLlh7

39 .l:!.e7

lLlfB

41 ..l:!.xd7

Has Wh ite trapped .

miscalcu l ated?

lLl h4+

H i s rook is

44 h4! !

N o , he has not. This modest move is the idea of the combinati o n : the h-pawn be­ comes a queen . 44 . . .

'it>eB

45 .txg7

'iti>xd7

46 h5

llaB

And without waiti ng for his opponent's reply, Black resig ned . N ow let us suppose that you have to play this variation with Black. You already know

36

�

Logic in the Opening

U h lmann's game, and you appreciate the threat of Wh ite's two bishops. Of cou rse , you have no i ntention of losing ignomini­ ously, l ike Osmanovic. What to do? Change variation? Well , if you approach th ings this way, you will never have any decent openings - i n each one problems will a rise. No, let's devise someth ing here - it is unlikely that Wh ite has a forced way to gain an advantage . O n l y b y clearly appreciati ng t h e danger w i l l you b e a b l e t o accustom you rself t o the position, be i mbued with its spirit, and as a result fi nd an idea which will help you . This is what Black d id in the fol lowi ng game. U h l ma n n - Popov

Berl i n 1 979 1 2 �d2

�e7

1 3 f4

exf3

1 4 exf3

It appears that nothing can prevent Wh ite from carrying out his pla n . It develops very natu rally: It ae1 , f3-f4 , and at some point lt:Jd5.

home. I n any case this u n usual m a noeuvre solves his defensive problems - the active position of the q ueen ensures B lack coun­ terplay. He gains time: i n order to prepare lt:Jd5, Wh ite has to remove h i s king to h 1 and play b2-b3. But what is more important is the fact that the queen controls the d 5sq uare , and it is no longer easy for the wh ite knight simply to be placed there. If Wh ite plays f3-f4 , there is a possibil ity of excha ng­ ing knights by . . . lt:Jg4, and lt:J d 5 will not be possible because of . . . lt:Jf2+ . 1 6 'it> h 1

ii. d 7

17 b3

Ite7

.l::i. a e8 1 8 'iNd3 In order to prepa re lt:Jd 5 , Wh ite has had to place his queen on d 3 , where it comes under attack by . . . tt:Je5. Black has suc­ ceeded in doubling his rooks . If 1 9 f4 there ca n follow either 1 9 . . . lt:J g4, or 1 9 . . . lt'l e4 . 1 9 tbd5

lt'lxd5

20 cxd5

tbe5

21 'ii'd 4! The only move . 21 'i!Yc2? 'iNxd5 22 f4 'iNd 3 is bad for Wh ite .

14 . . .

'ii'e 5!

21 . . .

"iVxd4

1 5 �ae1

'ii'c 5

22 cxd4

lt'ld3

23 .l::i.x e7

.l::i. x e7

Here Wh ite already has to display accu racy: the i nvasion on e2 is th reatened . Thus in the event of 24 ..t xa5? b6 and 2 5 . . . It e2 Black has active play. How ca n Black be pre­ vented from activating his rook? Wh ite m u st be able to meet . . . l:!. e2 with .l::t d 1 , but without al lowing a check on f2 . 24 �g1 ! If now 24 . . . .l::i. e 2, then 25 .l::t d 1 , when the th reat of SL f1 is very u n pleasant, and in add ition the a5-pawn is sti l l h a ng i n g .

I don't know whether Luben Popov found this idea at the board , or prepared it at

24 . . .

b6

25 Itd1

lt'l b4

Two pawns are attacked - Wh ite is forced to exchange.

ltJ

Logic in the Opening

26 i.xb4

axb4

27 �c1

i.c8

A roughly equal ending has been reached . i.a 6 29 i.f3 '>t>f8 30 '>t>f2 WeB 3 1 a3 bxa3 32 .Ua 1 'it>d8 33 �xa3 i.b7 34 h4 f5 35 h5 l:!.e8 36 i.g2 .l':.e7 37 b4 .l::t e 8 38 .l::!.a 7 '.t>c8 39 i.f3 l:!.e7 40 .Ua1 W d8 D raw

28 f4

Now let us th i n k for U h l m a n n . 'I worked out such a good pla n , I ' m not going to g ive it u p now. I have t o fi nd a n i mprovement. ' When you know the ideas, it is easier to devise a novelty. Here it is con ceived in a purely logical way. The pattern of the position is appea l i n g , but it wou ld be good to prevent Black from playing his queen to c5. How? He can be deprived of the e5-square , by placing the pawn on f4 . This means that instead of 1 2 � d2 we should look at 1 2 f4 . It looks rather dangerous for Wh ite to play this, with his development i ncomplete and his rooks unconnected . But it should be checked . . .

37

from his main cou rse of play, and at the same time h e prevents the q ueen manoeu­ vre to c5. 14 . . .

i.d7

1 5 i.d2

If now 1 5 . . . 'i!kf8 , then after 1 6 .l::!. a e1 the same position is reached as i n the Uhlmann­ Osmanovic game, where Wh ite was able to g a i n a clear advantage . 15 . . .

'it'd8

Black nevertheless wants to switch his q ueen to the g 1 -a7 d iagonal via the b8square . 16 h3

'ii b 8

If n ow 1 7 .l:!. a e 1 , then 1 7 . . . 'iia7 1 8 � h2 'iVc5. Wh ite has a couple of extra tempi compared with the U h l m a n n-Popov game, but never­ theless Black reta i n s cou nter-chances . 1 7 ltJd5!

lLlxd5

1 8 cxd5

'ii'a 7+ lb e7

1 9 'it>h2

What would you advise Wh ite to do now?

U h l m a n n - Plachetka

Trencianske Teplice 1 979 12 f4

exf3

1 3 exf3

'ii'e 7

14 f4!

Now Wh ite only needs to play i. d2 and Mae 1 , and he reaches the very position for which he is a i m i n g . Of cou rse, he has to reckon with . . . lbe4 , but on the other hand he has the reply lLld 5 . I will take t h i s opportu n ity t o once again emphasise an idea which we have a l ready encou ntered many times . Th e d eep est moves, the best positi onal d ecisi ons, are those which c om bin e the implem enta­ tion of your own plan with pr ophylaxis against th e opponent 's id eas. This is the strength of the move order chosen by Uhlmann: the g randmaster does n ot deviate

20 f5 ! was very stro n g . It cramps the enemy pieces and g ives Wh ite the adva ntage. Unfo rtu nately, U h l m a n n played slig htly rou­ tinely. 20 .l:f.ae1 ? ! i.f5 If 20 . . .b1

b6

Wh ite's opening strategy has been justified - he has a slig ht, but enduring advantage . H e can g rad ually cramp h i s opponent b y 1 7 ..i c1 (vacati ng the f4-sq u a re for his knight), then b2-b3, depriving the black knig hts of the c4-sq uare , and so o n . (U nfortunately, here I hastened to seize control of the f5point, and played 1 7 h5?! g5 1 8 ..i xd6 ..i xd6 1 9 ltJf5 , but did not achieve a lot . ) Let u s retu rn t o t h e moment when the ' Exchange Variation' a rose after Wh ite's 8th move . I ncidentally, here S h mit had a serious th i n k , but even so he did not sense the strategic danger th reatening h i m . D u ring t h e g a m e I was concerned about the reply 8 . . . c5 ! , changing the pawn structu re . After this it is no longer an Exchange Variation of the French Defence. But what does occu r? Most probably a position with a n isolated d5-pawn for Black, and th is means the Tarrasch Variation of the French Defence. In whose favou r a re the changes in the position compared with the Tarrasch Varia­ tion? Anatoly Karpov, who at one time won a n u mber of games in this variation , wrote : 'The main idea of White 's play is control of the d4-point. He must constantly keep an eye on it. ' If Wh ite does not h ave secu re control of this point, he ca n not cou nt on any adva ntage.

42

�

Logic in the Opening

Usually in the Tarrasch Variation the knig hts stand on f3 and d2, and then , after a n exchange o f pawns on c 5 , t h e knight goes with gain of tempo to b3, from where it controls the d4-point. But here the wh ite knight has moved to the side and is not taking part in the battle for the centre . This factor improves Black's chances i n compari­ son with the normal Ta rrasch Variation . I t i s more d ifficult to evaluate the f6 , lbf7 construction . Does it not weaken Black's position? The answer to th is question is not obvious. It is easier to fig u re out if you know one of the strategic ideas which Black sometimes employs i n the Tarrasch Va ria­ tion . It was demonstrated i n the following game. G i ps l i s - Korchnoi

Amsterdam 1 976

French Defence 1 e4 e6 2 d4 d5 3 lLid2 c5 4 lLigf3 lLic6 5 exd5 exd5 6 i.b5 i.d6 7 dxc5 i.xc5 8 0-0 lbe7 9 lZ'i b3 �d6 1 0 �g5 0-0 1 1 l::i. e 1

1 3 �e2

i.e6

1 4 �g3

It is advantageous for Wh ite to exchange the dark-sq uare bishops . But Black is ready to cou nter th is attempt. 14 . . .

lLie5!

The defence is based on the strong point at e5. 1 5 lLifd4

Si.d7

I would have preferred the retreat to f7 , to secu rely defend the d5-pawn . 1 6 a4 a6 1 7 a5 'ikc7 1 8 c3 �ad8 1 9 lLic1 i.. c 8 20 tZ'id3 lLi7g6 2 1 lLif4 lLixf4 22 �xf4 l::i.fe8 23 i.. g 3 ..tc5! 24 �f1 "iVf7 25 l!Va4 'ikg6 26 l::i. e 3 lLig4 27 �d3 "iVf7 28 �xe8+ .U.xe8 29 "iVc2 g6 30 b4 i.. a 7 31 b5 lLie5 32 i.. f1 axb5 3 3 lLixb5 i.. c 5 34 �d1 �g4 35 �a 1 �a8 D raw.

Let us return to the Alekh ine Defence. Knowi ng Korch noi's game, we can conclude that the knight at f7 and pawn at f6 are not so stu pidly placed , and they may well come i n usefu l for Black i n his set-up. This factor merely strengthens our i m pression that Black has a good version of the Tarrasch Variation . Chess is not l i ke study compositi o n , where there is always just one solution . I have shown one of the sensible approaches to the positio n , but a completely d ifferent logic is also possible. When I invited Nana Alexandria to consider the positi o n , she found her own solutio n , and also a good one - the i nteresting queen check 8 ... 'ike7+ ! .

An open ing tabiya. Here many moves have ..t g4 , . a7-a6. Also been tried - . 'Wic7 , possible is the plan chosen by Victor Korch noi . 0 0

o o •

0 0

11 . . .

f6 ! ?

1 2 ..th4

'ii'b 6

How should Wh ite respond to it? The exchange of queens is not dangerous for Black. If 9 i.. e 3?, then 9 . . . 'ik b4+ is u n pleas­ ant. 9 lZ'i 1 e2 hampers Wh ite's development. 9 i.. e 2 looks natura l , but then 9 . . . i.. g 4 ! . The exchange on e2 favou rs Black, but after 1 0 f3 � d7 (followed by . . . lb c6 and . . . 0-0-0) the knight has been deprived of the f3-sq uare and it is not easy for Wh ite to complete his development.

Logic in the Opening

How does one arrive at the move 8 . . .'ir' e7+ ? After all, it is not usual to play with the q ueen, when there is only one knight developed . You can arrive at this check only after experiencing a certai n alarm , after realising that if the game develops natu rally it will take a favourable cou rse for Wh ite . Not wish ing to allow th is, you look for d rastic measures and someti mes find them. Absolutely orig inal ideas occu r extremely rarely i n chess. All this had a l ready oc­ curred , including the queen check. It is much easier to fi nd it, if you know the following game, or more correctly Larsen's comments to it on the position arising i n the opening. Larsen - Portisch

I nterzonal Tou rnament, Amsterdam 1 964

French Defence 1 e4

e6

2 d4

dS

3 t'Dc3

..ll. b4

4 exd5

exdS

5 'it'f3 ! ?

CtJ

43

..ll. d3lbc6 6 t'Dge2 t'D ge 7 followed by . . . il. f5, and there are no problems for Black. The text move, which I had played in some blitz games against Palle Ravn, is directed against the very manoeuvre . . . t'Dge 7 and . . . il. f5; after 5. . . t'D e 7 6 il. d3 t'D bc6 7 t'D e2 White's position is quite attractive. 'Because of this game 5 'fif3 became almost popular for a short while, but it disappeared again because of the reply 5. . .'ifi e 7+!, for instance 6 t'D e2 t'D c6 7 'iixd5 t'D f6 with more than enough for the pawn. 'During the game I thought of the possibility 5. . .'ifi e 7+ and toyed with the idea 6 ..ll. e 3 il. xc3+ 7 bxc3 'ii a3 8 W d2, which may look strange, but is very good for White. How­ ever, a Yugoslav game Mestrovic-Maric, Kraljevo 1 96 7, seems to prove that 6 il. e3 is of dubious value because of 6 . . . t'D f6 7 il. d3 c5!. After this I tend to believe that 5. . . 'iVe 7+ is Black's strongest move. 'Immediately after the game O'Kelly stated that the easiest solution for Black was 5. . . il. e6, but I don't agree; after 6 il. d3 'iif6 White ought to play 7 il. f4!. 'Also 5. . . c5 has been recommended, but 6 dxc5 d4 7 a3 'it'a5 8 :I:i b 1 1ooks very good for White. 'Portisch had enough to think about!'

Let us hand over to Bent Larse n :

'This set Portisch thinking! The exchange variation has had a reputation as a dull drawish line for many years; for instance, 5

It remains to add that in the game there followed 5 . . . t'Dc6 6 .i. bS t'De7 7 .i.f4 0-0 8 0-0-0 t'DaS? (8 . . . .i. e6 was better) 9 t'Llge2 c6 1 0 .i.d3 b5 1 1 h4! t'Llc4 1 2 h5 f6 1 3 g4 �aS?! 1 4 .i.xc4 dxc4 1 5 a3! ..ll. x c3 ( 1 5 . . . .i. xa3 1 6 bxa3 'it'xa3+ 1 7 W d2 b4 1 8 l::t a 1 ! bxc3+ 1 9 t'Llxc3 'it' b4 20 � h b 1 ) 1 6 t'Llxc3 'ikd8 1 7 l:the1 ( 1 7 t'D xb5?! 'it'd 5 ! ) 1 7 . . . a5 18 'it'g3 l:ta7 ( 1 8 . . . b4 1 9 .i. d 6 ! ) 19 h6! g6 20 .i.d6 l:te8 2 1 'it'f4! Wf7 22 il.eS f5 23 .i.bB l:tb7 24 'ir'eS ! :gs 25 g 5 b4 26 'ir'f6+ WeB 27 'it'xc6+ Wf7 28 'ir'f6+ 'lt>e8 29 dS l:tf8 30 'ir'c6+ 'it'd7 3 1 .i.d6 �f7 32 .i.xe7 bxc3 33 .i. b4+ Black resigned . The concl usion from what h a s been said is

Logic in the Opening

obvious. Chess eruditi o n , a knowledge of typical ideas, serves as a reliable basis i n t h e logical perception o f opening problems, and helps correct solutions i n the opening to be fou n d . However, not only in the open i n g . T h e boundary between t h e ope n i ng and the midd legame is very arbitrary. Many of the

positions i n q uestion , although they a re considered by open ing theory, ca n well be assigned to the middlegame. At one time I wrote an article about the study of typical midd legame positions and i n it I recom­ mended roughly the same approach . You can fi nd it in the fou rth pa rt of this book.

45

Artur Yusupov

Surprises in the Opening

W

hat chess player has not stumbled

on the contrary, it may awake our imagina­

into a cunning opening trap, been

tion and force the brain to work at full power.

caught in a variation that appeared to be

It often happens that the player who wins is

condemned by theory but in reality proved

not the one who prepares the variation, but

fully viable, or fallen victim to his own

the one for whom it was prepared! Relying

opening p reparation, which turned out on

entirely on the strength of his home prepara­

verification to have a 'hole'? In other words,

tion, a player is sometimes unable to force

all of us have had occasion to encounter

himself to play with full intensity. In this case,

surprises in the opening and, of course, we

any surprise in the opponent's actions, even

are well familiar with the unpleasant feelings

the most trivial, may prove fatal for him- he

associated with this.

simply does not manage to retune for a

The search for unexpected (for this read -

fierce and genuine struggle.

new for the opponent!) ideas serves as the

I should like to disclose the theme of today's

main source of opening theory develop­

lesson using examples from my own games.

ment. And if you think about it, all our efforts

I will begin with one played in the World Cup

in preparing for a game are aimed precisely

with the Hungarian grandmaster Gyula Sax.

at finding a way to perplex, surprise, stun

Of course, for such important encounters

the opponent, to upset his normal routine. To

the players make especially thorough prepa­

surprise means to win! But, after all, our

rations. I was hoping to catch Sax in one of

opponents also aim for the same thing.

the variations of the Queen's Indian De­

Of course, thorough opening preparation

fence, of which I had made a fairly detailed

significantly reduces the probability of some

analysis.

opening move or variation proving unex­ pected for you. However, it is not possible to

Yusupov

exclude unpleasant surprises completely,

-

Sax

World Cup, Rotterdam 1989

and you should be inwardly prepared for

Queen's Indian Defence

them. There are different sorts of surprises. In

1 d4

lt'Jf6

other words, a new move by the opponent

2 c4

e6

may be objectively strong, but it may be

3lt'lf3

b6

aimed at you in particular, at your reaction,

4 g3

i..a6

since any surprise is in the first instance an

5 b3

attack on your nerves. And very much depends on how quickly you are able to take yourself in hand and tune up for a full­ blooded struggle. If you become flustered, this is likely to lead to a rapid collapse. Incidentally, meeting something unexpected or unexplored is not bound to demoralise;

Apart from this, the most popular continua­ tion, I have also employed 5 lt'lbd2 several times.

5

.

.

.

6 i..d2

�b4+ i..e7

What is the idea of this loss of a tempo,

46

�

Surprises in the Opening

which Black goes i n for? The point is that after b2-b3 the natu ral place for the bishop is on b2. Wh ite may sti ll want later to switch his bishop to the long diagona l , but at c3 it is less secu rely placed than at b2 , and in add ition it takes away the natural sq uare for the development of the knight. If Wh ite does develop his knight on c3 , all the same he will su bseq uently have to move his bishop from d2. Therefore with his manoeuvre Black effectively does not lose a tempo. c6 7 � g2

1 3 e4

Preparing . . . d7-d 5 . 8 0-0

d5

9 tt:Je5

Exploiting the fact that for the moment it is not possible to capture on c4 because of the vulnerabil ity of the c6-point, Wh ite tries to occupy the centre . The knight on e5 is very un pleasant for Black, and he must ex­ change it. 9 . . .

tt:Jfd7

Again su pposedly against the rules (the second move i n the opening by the same piece), but in closed positions this is sometimes admissible. In the g iven case the battle for the centre is more important than the rapid development of the pieces. And in view of the fact that the knight at b8 is literally cri ppled by the knight on e5, the move made is i n some sense also a developing one. 1 0 tt:Jxd7

When the game was played , this move was considered al most obligatory, but at the present time theoreticians a re increasingly considering other conti n u ations. One of the new and good alternatives is 1 2 . . . tt:Jf6 . This was fi rst played by Lajos Portisch agai nst Anatoly Ka rpov ( Rotterda m 1 989), and later the same idea was used by Karpov himself in his match with me. The position is q u ite i ntricate , despite its apparent simpl icity. But we will n ot now go i nto its subtleties - for this there are reference books .

tt:Jxd7

11 �c3

Here we have that insta nce about which I told you : Wh ite is forced to waste a tempo, i n order t o bring his bishop onto t h e long diagonal . I n passing I should mention that the captu re on c4 is dangerous for Black, since after 1 1 . . . dxc4 1 2 d5! cxd5 1 3 � xg7 Wh ite prevents his opponent from castl i n g , obtaining an enduring i n itiative for t h e pawn. 11 . . .

0-0

1 2 tt:Jd2

.l:tc8

Play in the centre ! I n reply, if you remember, in the matches between Kaspa rov and Karpov ( 1 984-85 and 1 986) the adva nce . . . b6-b5 was tested , after which Wh ite played l:t e 1 . But the plan with . . . c6-c5 is more usual . 13 . . .

c5

1 4 exd5

exd5

1 5 dxc5

Wh ite is not able to win a pawn : if 1 5 � xd5 there is the reply 1 5 . . . tt:Jf6 . 15 . . .

dxc4

After 1 3 . . . c5 the g iven position arises more or less by force . If it is unfa m i l i a r to you , it is n ot so easy to work out its n u a n ces at the board . In principle, such position s , critical for the opening variation , should be very thoroug hly stud ied in home preparations and subjected to detai led analysis. 16 c6

N ow i n particu lar Wh ite has to reckon with the reply 1 6 . . . cxb3, since the captu re o n d7 does not work, as the bishop at c3 is h a n g i n g . There now beg i n s a sharp tactical skirmish , which seemed to me to be not u nfavourable for Wh ite . 16 . . .

cxb3 !

1 7 l:te1

�b5

Perhaps the most critical reply (although later 1 7 . . . b2!? 1 8 � xb2 tt:J c5 came i nto fas h i o n ) . Sax attacks the c6-pawn and is stil l

Surprises in the Opening

not afraid of the captu re on d 7 . If Wh ite wants to sell his bishop more dearly - 1 8 .. bg7 'it> xg7 1 9 cxd7 , t h e n simply 1 9 . . .'if xd7. Black has the two bishops , and it is not apparent how the weake n i n g of h i s position can be exploited .

ttJ

47

Sax, I repeat, had repl ied too confidently and q u ickly. I n such a situation the main thing is not to become flu stered , not to pa nic. You should try to delve more deeply i nto the position and try to find what exactly the opponent has in m i n d . In addition , it is i mportant to brace yourself psychologically for the fact that someth ing u nexpected awaits you , some u n pleasant su rprise . . . Realising all this, I nevertheless did not see any possibil ity of deviating from the plan ned captu re o n g7. Althou g h , I have to admit, I made the move without my former optimism. 19 i.xg7

xg7

20 ttJd4!

Now try to look ahead a l ittle , to fi nd out h ow the game may develop, and what resou rces are available to Wh ite . 1 8 l:t c1 ? A n ew move , which is worth considerin g . . . Capture on b3 with the pawn? Yes , this i s not a b a d reply, perhaps t h e best. I advise you to look carefu lly at 1 8 axb3 at home the resulti n g variations a re very i nteresti n g .

I s that a l l ? I n fact i t is a l s o possible to capture on b3 with the knig ht, in con nectio n with an i nteresti n g tactical i d e a : t h e sacrifice .ixg7 ! . It is this possibil ity that I analysed i n m y h o m e preparation - it is a pity if y o u d i d not notice it. 1 8 ttJxb3 ! ?

..ixc6

At this moment the situatio n suddenly ceased to appeal to me: Sax all too willingly went i n for the complications, which accord­ ing to all my prel i m i n a ry estimations should have been in my favou r. H e was not so na"ive as to go i n for such play, without having someth ing specific i n m i n d ! I real ised that there might be a flaw in my calculations.

Here Black d oes not have much choice . Thus 20 . . . i. f6 is b a d - Wh ite gains the advantage after 21 ttJxc6 :Xc6 22 i. xc6 ..i xa 1 23 'ii'xa 1 + . The main conti nuatio n , which I exam i ned i n my h o m e analysis, was , naturally 20 . . . i. xg2 . Let us together try a n d figu re o u t what w a s Sax's d efensive idea. Let's immediately look at the critical line - 21 ttJf5+ (other moves hardly deserve serious consideration). The reply 21 . . . d 1 (not 1 1 li:Jed2?? � c3 o r 1 1 . . . � xd2+ ) , and now Black is not obl iged to exchange queens, but can play 11 . . . 1i'a 3 . Shall we look any fu rther? To be honest, with h i s king on d 1 Wh ite's position does not inspire confi­ dence. If it appears to you that Wh ite can defend h imself here , try checking this position at home. In genera l , try to support a l l your feelings with concrete a n a lysis.

Now consider: what should Wh ite do? What 'course of behaviour' should he choose , after encou ntering such an u n usual devel­ opment of events i n the open i n g , when Black somehow over-sharply wants to seize the i n itiative and win? The value of each move is now very h i g h . T h e b2-pawn is 'hanging ' , b u t t o defend i t with t h e rook? ! . . . T h e faint-hea rted 9 li:J d6+, trying somehow to subdue the opponent's attacking mood , also does not look good : after 9 . . . � xd6 1 0 exd6 iYxb2 I th i n k that Black has a minimal advantage. 9 li:Jf3 !

The critical decision. Of cou rse, Wh ite cannot know for sure how the compl ications will end, but he acts i n accordance with the rules of opening strategy and brings new pieces i nto the battle, not bothering about the defence of the b2-pawn. 9 . . . 1 0 l::!. b 1

'it'xb2

But remember: i n a practical game you are not able to calcu late everything to the e n d , and t h e main thing is t o eva l u ate correctly the consequences of this or that conti nua­ tio n , and decide for you rself whether it is worth going i n for the variati o n , or at least seriously exa m i n i ng it. I n the g iven i n sta nce , I repeat, I considered the move 1 0 'ir'xd4 to be very dangerous for Wh ite , even on psychological g rounds, since it hands the i n itiative to the opponent. But I wanted to be the attacking side, I myself wa nted to fight for the i n itiative with every move ! 10 . . .

Si. b4+ !

The strongest reply - Black d rives back one of the opponent's already developed pieces. The captu re on a2 would be extremely risky for h i m , since it would g ive Wh ite the opportun ity to beg in an attack with 1 1 e3, as wel l as with the simple captu re 1 1 li:Jxd4 . 1 1 li:Jed2

A purely practical decision : by blocking with this knight, Wh ite red uces the opponent's choice of replies. After a l l , i n both cases Black ca n captu re on d2, but the move 1 1 li:Jed2 forces th is exchange, since i n the event of 1 1 . . . 'tlf a3 Wh ite is now free to captu re on d4. 11 . . .

Si.xd2+

1 2 li:Jxd2

'it'xa2

It is possible to make an i n itial assessment of the operation beg u n with 1 0 .:l. b 1 . Wh ite

ttJ

S u rprises in the Opening

has lost a paw n , but in return he has g a ined the opportun ity to q u ickly complete the development of his pieces. I n cidentally, which move best conforms with the g iven aim? 13 e3

Correct! I ncidental ly, if I had not foreseen this resou rce, I possibly would n ot have played 1 0 .l:t b 1 , but would have looked for some alternative . It seemed to me that White was fi rmly seizi ng the i n itiative. 13 . . .

dxe3

What to d o n ow? There a re two candidate moves: 1 4 i. b5+ and 1 4 fxe 3 , and the question is which of them is more accu rate. You think 1 4 fxe3 ? But why, can you explain? Because of the possibil ity of plac­ ing the bishop o n c4? Yes , this is the fi rst thing that strikes one. If Wh ite plays 1 4 .1b5+ , Black would manage to take control of the c4-point: 1 4 . . .

Surprises in the Opening

basis. Such tactics - t o divert the oppo­ nent fr om th e well-trodd en paths, espe­ cially if he is inferi or to y ou in class often bear fruit. The following example develops the theme of the French Defence. The game with Andrey Sokolov, which I want to show you , was played in our Candidates Match a n d , I remember, it afforded me g reat creative satisfaction .

11 dxc5 In practice 1 1 liJg5 and 1 1 i.. a 3 have also been tried .

11 . . .

bxc5

1 2 c4!

White u ndermi nes the centre .

A. Sokolov - Yus u pov

Final Candidates Match , 3rd game Riga 1 986

French Defence 1 e4

e6

2 d4

d5

3 liJc3

i.. b4

4 e5

liJe7

5 a3

i.. x c3+

6 bxc3

c5

7 liJf3

b6

Match play has its specific featu res , and in it a particular place is g iven over to open ing duels. The aim of the last move is obviou s : Black wa nts t o exchange his b a d ' F rench' bishop. I n the 1 st game Sokolov preferred to avoid the exchange, by playing 8 i.. b 5+ i.. d 7 9 i.. d 3. He obtai ned a reasonable position , but later he lost in a complicated struggle. It was clear that for the 3rd game Sokolov would th ink up someth ing new. But what? I n my preparations with m y trainer, to be honest, I did not guess rig ht, and the opening of this game came as someth ing of a surprise to me.

8 a4 A typical move in such positions.

8 . . .

i.. a 6

9 i.. x a6 9 i.. b 5+ is more often played .

9 . . .

liJxa6

1 0 0-0

liJb8

Here I had to th i n k seriously. I had only some vague memories of the g ames from the Geller-Spassky Candidates q u a rter­ fi nal match (Sukh u m i 1 968), where this variation occu rred severa l times. I n such situations you have to be g u ided by common sense, and if the position al lows it, avoid the fam i l i a r paths as fa r as possible, choosing positionally j u stified but less well­ stud ied conti n u ations. I sensed that the fu rther cou rse of the struggle would largely depend o n what sq u a res the black knig hts would occu py, and in particu lar where the b8-kn ight would be developed .

12 . . .

0-0

This move did not ca use any g reat doubts .

1 3 cxd5 But here it was time to avoid the fa m i l i a r paths , and n o t w a i t for Sokolov t o spring some surprise. As far as I remembered , Boris Spassky captu red on d5 with h i s queen .

Surprises in the Opening

13 . . .

tt:'lxd 5 ! ?

Not a b a d sq uare for t h e knight. For the moment Black is not pa rticularly afra id of c2-c4 , si nce h i s knight can occupy the b4square.

1 4 'it'd3 What would you suggest now for Black? To answer th is questio n correctly, you m u st see the aim of Wh ite's last move . Well , of cou rse, he wants to play 1 4 tt:'l g 5 , forcing the reply . . . g7-g6, which seriously weakens the dark sq uares. This weakening is very sign ificant, i n view of the absence of an opponent to Wh ite's dark-sq uare bishop.

14 . . .

h6

The position is very complicated . To 1 5 ll d 1 I was intending to reply 1 5 . . . tt:'l c6 , and i f 1 6 ia3, then 1 6 . . . tt:'l cb4 , with unclear play. Sokolov played more positively and . . . less successfu lly.

1 5 c4

57

1 7 . . . 'ike7 Wh ite strengthens h i s position by 1 8 l:i.ad 1 ! (it has to be this rook) 1 8 . . . l:i.fd8 1 9 .l:td6, and if 1 9 . . . l:Ixd6 , then 20 � xc5 , winning a pawn . After some thought, I decided that the place for the knight was at e7. Here it may come in useful for the defence of the kingside and i n general i t has good prospects : it can possibly move to f5 and then d4, o r else go to g6, attacki ng the e5-paw n . Of cou rse, when choosing the move I also had to analyse concrete variations. For example, the fol lowing o n e , wh ich looks rather dan­ gerous for Black: 1 5 . . . tt:'l e7 1 6 .l::td 1 'iix d3 1 7 l:txd3 tt:'l bc6 1 8 � e 3 , and it appears that the c5-pawn ca n not be defended . But. . . it only appears to be so! I had prepa red the reply 1 8 .. .l:Ifc8 ! and if 1 9 � xc5 - 1 9 . . . tt:'l xe5 ! . This, so to speak, is the tactical j ustification of 1 5 . . . tt:'le7. In the g iven i n sta nce the general eva l uation of the position is com b i ned with concrete calculati o n , which is usually a necessary condition for taking the correct decision. You can g ive as many general considerations as you l i ke in defence of a move , but if it then turns out that i n one of the variations a pawn is simply lost - all your a rguments will lose their point. You r eval uations must without fail be su pported by accu rate ca lculation!

15 . . .

tt:'le7 !

1 6 'i!Ve4

What possib i l ities does Black have here? With a pawn on a4 , 1 5 . . . tt:'l b6 obviously looks bad , and the knight i s also badly placed on c? . H e has to choose between 1 5 tt:'l b4 and 1 5 . . . tt:'l e7. I rejected 1 5 . . . tt:'l b4 , since I reckoned that after 16 'l!lt'e4 tt:'l 8c6 1 7 � e3 my knights . . .

would be badly placed . At b4 the knight is outwardly active, but that is all. I n reply to

The next i mportant q uestion is where to develop the b8-kn ight: at c6 or d7? After some thought, I chose 1 6 . . . tt:'ld 7 , since after 1 6 . . . tt:'l bc6 it would have been much more d ifficult to defend the c5-pawn . General spea k i n g , when you encou nter an u nfamiliar syste m , you should pay extra attention to your opponent's threats .

16 . . .

tt:'ld7

1 7 kib1 ? The best contin uation was 1 7 lt d 1 . But I saw that after 1 7 . . . 'iic 7 I did not have to fea r 1 8 .f!d6?! i n view of 1 8 . . . tt:'lf5! , when it is bad to

58

�

Surprises in the Opening

play 1 9 .llc 6? 'ii' b7 20 I!. b 1 tLl b6 with the deadly th reat of 21 . . . Itac8. However, Sokolov played su perficially. He real ised that I would answer 1 7 l::t d 1 with 1 7 . . . 'ii'c7 , and he decided to prevent th is, by preparing 1 8 l:t b7 in reply. However, 1 7 . . . 'ii'c7 is by no means obligatory. Black can easily equal ise by 1 7 . . . l:t b8. But at this moment I sensed that my position was a l ready prefer­ able and that it was possible to play for a win.

17 . . . 'iia 5! If there is a possibil ity of making a n active move , it should be made! Black seizes the in itiative . For the moment the a4-pawn is hanging. Wh ite should probably have admit­ ted his mistake and retu rned his rook to a 1 . 1 8 l::t d 1 W h y n o t simply capture t h e pawn now? After a l l , at fi rst sight Black is not threatened with anyth ing. It turns out that if 1 8 . . . 'ii'xa4? there is the idea 19 � xh6! gxh6 20 l::t a 1 'ii'c6 2 1 'ii'x c6 ttJxc6 22 l:txd7, and Wh ite is a l right. But Black is hoping for more . 18 . . . .l:I.ad8! 18 . . . tLl b6 was also not bad , but 1 8 .. J i ad8 (only not 1 8 . . . .Ufd8? 1 9 llxd7 ! ) nevertheless seemed to me to be the most energetic. As long as Wh ite's bishop l i ngers on the back rank, his king may not feel safe : it has no escape square , and the opponent is a l ready creati ng tactical threats . 1 9 'ii'c2 Sokolov decided to defend the a4-paw n , overlooking t h e spectacular reply. It is amusing that, as i n the preced ing game, it is a knight sortie from d7 to e5 that proves fatal for Wh ite! (see diagram)

19 . . .

tLlxe5 ! !

2 0 ttJxe5 20 .Uxd8 ttJ xf3+. 20 . . .

'iic 3 1

- position after 1 9 'ii'c2 -

It goes without saying that it is pleasant to make such moves .

2 1 'ii'e 2 22 �e3

'ii'x e5 tLl f5

As a result of his combi nation Black has won a pawn and is close to winning the game. H owever, he still has to overcome some tech n ical d ifficulties.

l:txd 1 + tLl The immed iate 23 . . . d4 was possibly pref­ erable, but I decided to play more simply. tLl d4 24 .Uxd 1 23 'iVf3

Here Sokolov committed a serious m istake. He should have chosen 25 'ii b7. Then Black would probably have had noth ing better than 25 . . . tLlc2 26 'iif3 ttJ xe3 , and in the resulting endgame Wh ite would retai n d raw­ ing chances .

25 �xd4?

cxd4

26 itd3

l::t d 8

27 g3

it'c5

The q ueen makes way for the e-paw n .

28 f4? ! This merely leads t o a weakening o f Wh ite's king positi o n . But all the same things are bad for h i m , and so th is move should not be strongly criticised . The rest is easy. 'ii' b4 28 . . .

29 .Ua1

aS

Surprises in the Opening

A move that suggests itself, wou l d n 't you agree? The enemy pawn is fixed and Black's own pawn advances closer to the queening sq uare . While it is possible, you should aim to strengthen you r position to the maximum, and only then seek a concrete plan for converting the advantage . I n the given instance , since Wh ite has no cou n ­ terplay a t a l l , Black has no reason t o h u rry.

30 h4

h5

31 .l:tb1

'ii'x a4

32 .l:tb5

g6

33 �2

'i!Va2+

34 �3

a4

35 .l:!.b6 Wit h the idea of 36 .l:!.xe6. Here Black repeated moves to gain time on the clock. 35 . . . 36 .l:l.b1

�g7 'it> g B

37 .l:.b6

'ii'a 1 ! Turn ing to resol ute acti o n . If 38 .U.xe6 I was inte n d i ng 38 . . . .l:!. b8! 39 .l:.!. xg6+ 'it>f8 , and Black h imself lau nches a n attack. 38 �e2

a3

39 .l:l.a6

'ifb2+

40 "i¥d2

d3+

On seeing that the opponent had managed to make the time contro l , Sokolov resigned . I n what way is this game i nstructive? I t shows how i mporta nt it is t o p a y attention to the opponent's th reats - both to those that are obvious - tactical , and to the more cam ouflaged - position a l . I n th is respect I wou l d single out the move 1 5 . . . tt:J e7, which in some way provided the basis for Black's success. At any event, it hel ped him to solve his open ing problems. Don 't forget ab out another v ery cunning

ope n ing w eapon - m ov e transp ositi on ! Sometimes it can be sufficient for the opponent simply to change the places of two moves, for all you r preparatory analytical work to come to noth i n g . If, contrary to

59

expectati o n , such a thing happens, the main thing is not to engage i n self-reproach (this won 't help matters ! ) , but to remember an a ncient truth : to reg ret a mistake that has been made is to make a second mistake! Let u s analyse a game of mine with the English g ra nd master Tony M i les. It was played at the start of the I nterzonal Tou rna­ ment i n Tu n i s and was of g reat importance for me. At that time M iles was considered to be one of the strongest players in the West, and he had ach ieved good tou rnament resu lts . I had Black, and therefore I decided to choose a safe ope n i n g . But here is what ca me of this . . .

M i les - Yusu pov I nterzonal Tou rnament, Tu nis 1 985

Caro-Kann Defence 1 c4

c6

At that time the Slav Defence was part of my ope n i ng repertoi re , and it was this that I had plan ned for my meeting with M i les. And suddenly - to my horror! - M i les replied with something q u ite u n expected .

2 e4! Of cou rse, the exclamation mark is not for the strength of the move , but for the correct psychological choice. The point is that after 1 e4 I do not play the Caro-Kan n Defence , and i n my preparations for t h e g a m e I h a d absent-mindedly overlooked t h i s simple tran­ sition i nto it. Yes , such a transposition of moves can be more u nsettl ing than any prepared variation!

2 . . .

d5

3 exd5

cxd5

4 d4

tt:Jf6

5 tt:Jc3

tt:Jc6

When you encou nter a l ittle-known system, you natu rally want to red uce it to someth ing that is more or less familiar to you . I had some idea about the variations arising after 6 tt:Jf3 il. g4 , but, alas, M iles played d ifferently.

60

�

Surprises in the Opening

6 j,gS At this my opening knowledge came to an end. What to do? Now 6 . . . e6 would have led to well-known theoretical set-ups - well­ known , but not to me. Therefore I decided to look for some other sensible path .

6 . . .

�e6 ! ?

than 7 . . . exf6 . The pawn captu res towards the centre , thereby strengthening Black's control of it. But, more i m porta nt, it does not lead to a strateg ically u n promising situation, where o n the queenside Wh ite effectively creates a n extra pawn . 8 �d2

M iles delays the development of his knight at f3 , in order later to have the possibil ity of f2-f4 , i n h i biting the advance of the black e­ pawn . N atural ly, such a decision also has its d rawbacks - since neglecti ng the rules of development may tel l . 8 . . .

It was o n l y after t h e g a m e that I learned that 6 . . . � e6 had been i ntrod uced by Sergey Belavenets , and later taken u p by Salo Flohr. At fi rst sight Black's decision may seem eccentric, but i n essence it is not without point, since it does not g reatly violate the principles of opening play. I n­ deed , I develop a piece, establish control over the centre, strengthen the d5-point, and create the possible threat of captu ring on c4 . It is hard to demand more of one move!

liaS

N ow it is rather more d ifficult for Black to bring h i s dark-sq uare bishop i nto play (he does not have . . . � f8-h6 ) , and so he completes his q ueenside development, aim­ ing for castl ing o n that side of the board . The queen will be more actively placed at a5 than at d7. The concrete j u stification of the move is that if 9 ltJxd5?! there fol l ows 9 . . . �xd2+ 1 0 'it>xd2 0-0-0, and then Black reg a i n s the pawn o n d 5 .

9 c5

0-0-0

1 0 j,bS In my view, it would have been better for Wh ite to complete h i s development by 1 0 ltJge2 and then g2-g 3 .

7 j,xf6

This exchange could have been delayed , by playing, for example, 7 l2lf3 . But, as it tu rned out, Miles was followi ng the recommenda­ tions of theory, based on a game Botv i n n i k­ Fiohr (Moscow 1 965). I n it after 7 � xf6 exf6 8 c5 Wh ite gai ned a positional advantage , defi ned b y his su perior pawn structure . 7 . . .

gxf6!

This move also has certai n d rawbacks , but on the whole, i n m y view, i t is more logical

What move would you now suggest for

Surprises in the Opening

Black? 1 0 . . . h5 and then . . . Si. h6 ? There is such an idea . But I decided to play more directly. 10 . . .

l:tg8 !

Black aims t o create a weake n i n g i n the opponent's position . In the event of 1 1 g 3 he could now have conti n ued 1 1 . . . h5, but he could also have considered 1 1 . . . Sl. g4, h i n dering the development o f the k n i g h t at e 2 a nd prepari ng t h e thematic advance . . . e7e5.

11 f4 Miles prepares i n advance to combat . . . e7But, by making many pawn moves, he delays the development of h i s pieces, which I try to exploit. e5.

11 . . .

i.h6

Black's play is simple and natu ra l . It some­ how happens of its own accord , that all h i s moves a i d t h e implementatio n o f his p l a n . From h 6 t h e b i s h o p h e l p s . . . e7-e5 t o b e carried out, b u t it is q u ite possible t h a t it w i l l also come in usefu l for t h e landing of some tactical blow.

Here it was a l ready essential for M iles to remember about development. Tru e , if 1 2 g7 28 �c5 there is no satisfactory defence agai nst the mate.

23 il.d4 Wh ite has completely implemented his idea . I have not even needed to advance my kingside pawns - the game should be decided by active play in the centre. But from this moment my opponent began to defend resourcefu lly. Georg iev played u n ­ expected ly, trying t o change t h e pattern of the play.

A completely standard move with an enemy knight on b6. 26 a5 is th reatened , and if the opponent repl ies 25 . . . a 5 , my knight obtains the b5-point, the su pport of the black knight is removed , and I can attack it, for example, with 26 'ii'f2 .

25 . . . 26 "it'f2

.Uc5

I saw that I could tra nsfer my q u een to the central square d4, and th is satisfied me perfectly well . After the game Kirill sug­ gested an i nteresti ng possibil ity : 26 .l:!. g3 'i*'c8 27 �d4+ � g8 28 f5 with a n attack. I was a i m i ng for play in the centre and did not even th i n k about moving my rook to the side. I find it hard to condemn myself for this move, since it leads to a fu rther improve­ ment in the placing of the pieces and retains a winning positi o n .

26 . . .

.Uec8

Black is forced to concede the open e-fi le.

27 'i!Vd4+ 28 .U e1

�g8 "it'd7

The creative Solving of irrational Problems in the Opening

Here I made what was probably a serious mistake . Wh ite's forces a re fully mobilised and it was now time for h i m to beg i n a concrete exploitation of h i s advantage and to calculate 29 tt'l b5! . But, to be honest, I did not look at this move at a l l , because the c2pawn is hanging. But after 29 . . . .l:!. xc2 I should have seen the simple 30 tt'l xa? , win n i n g . And i f 29 . . . a6, then all t h e same there follows 30 tt'l a?! 1:. 8c7 31 l:.de3, when there is no defence against the check on e8. The game would h ave come to its logical conclusi o n . But I was evidently not destined to win it. I contin ued playing in the same u n h u rried manner as befo re . Incidentally, players often h ave t o resolve the followi ng problem: whether to play positionally or switch to concrete actio n . I n the g iven insta nce I th i n k that u p till now I had done everything correctly, although other solutions may have been possible. But this moment is a very important one - here I delayed , violating the famous ru le of Wilhelm Steinitz: 'The player with an advantage is obliged to attack, as otherwise he risks losing this advantage '. True , the move made by me does not th row away the win , but merely complicates it. The trouble is that, as Siegbert Tarrasch said , 'mistakes never happen singly', and one m i stake often gives rise to another one.

29 �b2?

'ir'c6

And here someth ing i n expl icable happened . W hite's position is, of cou rse, totally won . The simplest was 3 0 f5! gxf5 3 1 g 6 , when 31 fxg6 fails to 32 .U e7. But 30 .t!. e? followed by .t!.de3 is also not bad . At this point I clearly saw that the only danger for me was the tactical stroke . . . tt'l c4+ . . . .

3 0 �de3?? I thought that I was conti n u i n g to strengthen my positio n and, a bove all, defending against . . . tt'l c4+ , but I overlooked that i n fact I was simply provoking th is move . . .

30 . . .

tt'lc4+ !

71

I thought that a fter 3 1 bxc4 l:lxc4 32 'i*'f6 d4 33 l!Vxc6 I would rema i n a piece u p , forgetting that t h e knight is captu red with check. After com m itting someth ing i rrepara­ ble, fortu nately I did not lose my head, and I real i sed that it was now dangerous to play for a wi n .

3 1 bxc4

l:lxc4

32 'ii'x d5

l:!.xc3

33 'ifxc6

l:13xc6

34 lie8+

�xeS

35 .Uxe8+

'it>g7

36 Wb3 And with i n a few moves we ag reed a d raw. Thus in this game too one move - 1 0 .l:!. g 1 exerted a very serious influence on the entire cou rse of the play, and simply predetermined it. The opponent's possibili­ ties were restricted and the rook success­ fu lly, without loss of time, came i nto play. The fol lowing example is a game with Alexander Bel iavsky. It concluded success­ fu lly for me.

Dolmatov - Beliavsky 56th U S S R C h a m pionsh i p , Odessa 1 989

Ruy Lopez 1 e4

e5

2 tt'lf3

tt'lc6

3 ..tb5

a6

4 ..ta4

tt'lf6

5 0-0

..te7

6 .l::!.e 1

b5

7 ..tb3

d6

8 c3

0-0

9 h3

tt'lbB

1 0 d4

tt'lbd7

11 tt'lbd2

..tb7

1 2 ..tc2

�eB

Black has played the B reyer Variation . I only

72

�

The creative Solving of irrational Problems in the Opening

had a slight knowledge of it - after a l l , one can't remember everything . Theory consid­ ers the main continuation to be 1 3 ti:Jf1 ..tfB 1 4 tt:Jg3 g6, after which it exami nes varia­ tions as far as the 20th or even 30th move . This leads to complicated positions, over which it is easy to lose control. I don't l i ke playing this way, especially with Wh ite . But there is also another syste m , which is less popular, but qu ite dangerous. I n my pre­ parations I came across the game Sokolov­ Beliavsky, played in the previous U S S R Championsh ip. Andrey Sokolov gai ned an advantage and I decided to act i n the same way. i.f8 1 3 a4

1 4 ..td3 My impressions of this position change from game to game. I thought that Beliavsky's reply 14 . . . c6 was obligatory. But then at the tournament in Reykjavik i n 1 990 Helgi Olafsson played 14 . . . exd4 1 5 cxd4 c5 agai nst me, and Yefi m Geller explai ned after the game that all this was known long ago. As they say, well known, but i n na rrow circles . . . Although the plan employed by Olafsson was new to me, this did not prevent me from gaining an opening advan­ tage by 1 6 axb5 axb5 1 7 .l:!.xa8 ..t xa8 1 8 dxc5! {th is last move , as it tu rned out, was a novelty by me) 1 8 . . . tt:J xc5 1 9 ..t xb5.

14 . . .

c6

1 5 b3

g6

Beliavsky played d ifferently against Sokolov: 1 5 . . . l:!. b8 1 6 i. a3 ti:J h 5 .

1 6 i.a3

W/c7

1 7 'it'c2

.l:i.ad8 (see diagram)

Here I stopped to th i n k and I real ised that on this occasion Beliavsky was not i ntend ing to play his knight to f4, as i n the game with Sokolov, but was planning . . . d6-d 5 . And , as it i n itially seemed to me, if he were to make

this advance in the centre he would ex­ change many pieces and equalise. And so I thought and thought, seeing as there was ample time. On all of the preced ing moves I had spent about five m i n utes, but here I thought for 40 or 50. I found the correct move earlier, but it was hard for me to decide on it. For p u rely psychological reasons - I l i ked the idea, but it was contradictory to the normal routi ne, and I could not bring myself to make it.

1 8 l:!.ab1 ! ! P rophylaxis! O n one occasion Tarrasch called such a move by Aaron N imzowitsch 'mysterious' , and in revenge N i mzowitsch gave the head ing " ' M ysterious' rook moves" to a whole section of his book. He wrote: ' We have h er e th er efor e t o do with a preventive acti on. Henc e it is only the outer form of the m ov e which is mysteri­ ous (a rook t o s eize a file which is s till closed), its strat egic end is n ot . . . The pr evention of fr eeing m ov es by the opponent is far m or e imp ortant than c onsid erations about wh eth er th e rook is effective at th e given m om ent or is occupying a passive p ositi on . ' But even so, why does this move make it d ifficult for Black to carry out his plan ned . . . d6-d5 ? It turns out that the i m mediate . . . d6-d5 is not th reatened at a l l . Say, after the natural

ttJ

The creative Solving of irrational Problems in the Opening 1 8 M ad 1 if 1 8 . . . d5? there follows 1 9 Jl. xf8 lt>xf8 20 b4 ! exd4 2 1 cxd4 , and Black is left with a bad lig ht-sq uare bishop and a n unfavourable pawn structure . Correct is 1 8 . bxa4 ! 19 bxa4 , and only now 1 9 . . . . d 5 with equality. .

.

After 1 8 .l:!.a b 1 ! ! Beliavsky ca n no longer reply 18 . . . bxa4 1 9 bxa4 d 5 , since the rook on the newly-opened file is attacki ng the bishop on b7, tying the q u een to its defence. There follows 1 9 Jl. xf8 'it> xf8 20 exd 5 , a n d then Wh ite captu res on e5, w i n n i n g a pawn . And the reply to the immed iate 1 8 . . . d5 we already know. The reader is j u stified in aski n g : what happens if Black avoids . . . d 6-d 5 , and plays, for example, 1 8 . . . Jl. g 7 . The d6-po i nt is weakened , but is this so i mportant? Without going into deta i l s , I will merely comment that if Wh ite carries out his thematic move c3c4, an exchange of pawns, opening the b­ file, is qu ite probable, a n d the rook at b 1 may also come i n usefu l .

Sasha Beliavsky is a rather straightforward player. He usually copes excel lently i n tactical variations, b u t he is sign ificantly weaker in positional subtleties. Thus here he does not want to g ive u p h i s plan . H e sees that t h e exchange o f pawns on a 4 i s unfavourable for h i m , b u t that t h e immediate . . . d6-d5 can not be d i rectly refuted . And the fact that it leads to a strateg ically d ifficult position for Black is someth ing that he does not notice . 18 . .

.

1 9 Jl.xf8

d5? '>t>xf8

20 b4! Of cou rse, Black's position is not yet lost, but at the least Wh ite has gai ned a stable opening adva ntage.

20 . . .

dxe4

21 ii.xe4! I did not want to complicate the play unnecessarily, and after 21 tt:J xe4 tt:J xe4 22

73

Jl. xe4 f5 I would have had to sacrifice my bishop on f5. I th i n k that now Black should h ave played the restra ined 21 . . .'ir' d6. Beliavsky incor­ rectly opens l i nes.

21 . . .

exd4? !

22 cxd4

'ii'd 6

Wh ite's positional advantage is obvious, but here , u nfortu nately, I made a serious m is­ take. I should have played 23 a5! .U e7 24 'iV b2 - then I wou l d have had a n extra tempo compared with the game. Without th i n king about what Black now wants (and he wants to double rooks on the open e-file}, Wh ite h u rried ly, without any particu lar hesitation , made h i s next move .

23 'i:Vb3? I was inte n d i n g to attack the f7 -point - 24 tt:Jg 5 .

23 . . . .

.l:te7!

Black hopes to seize the i n itiative , since my pawns h ave suddenly become vul nerable. 24 . . . .l:!. de8 is threatened , a n d the n , after the exchange of rooks, . . . tt:Jd 5 . H e re I h a d to com pose myself a n d fi n d what appears to be the only way of mainta i n ing Wh ite's entire constructio n .

2 4 i.c2 ! The bishop retreats here , in order to defe n d

74

�

The creative Solving of i rrational Problems in the Opening

the a4-pawn . When making my move, it was essential for me to calculate the variation 24 . . . l:!.xe1 + 25 l:!.xe 1 tt:J d5 26 l:!. b 1 bxa4 27 �xa4 tt:J c3, and now 28 tt:Je4 ! . Without th is nuance Wh ite's position would be unpleas­ ant, but, fortunately, it is there .

24 . . .

l:tde8

25 'ii' b 2!

.::t x e1 +

26 �xe1

l:txe1 +

Typical Beliavsky play. If there is a possibil­ ity, he immediately goes onto the attack; now he is threatening both a check on e2, and 31 .. ."it' d 5 .

27 tt:Jxe1

3 1 .i.d1 ! Perhaps my opponent did not notice this bishop manoeuvre? Wh ite defends agai nst everything ( 3 1 . . . 'it'd 5 32 .i.f3 ). If 31 . . . 'iie7, then 32 tt:Jf3 is stro n g . Here 31 . . . h5!? should

27 . . .

4Jb6?

A serious mistake. Black had a choice : to allow the blocking of the position by a4-a 5 , or h imself exchange on a 4 . Beliavsky incorrectly eval uated the position . H e should have chosen 27 . . . bxa4! 28 .i. xa4 tt:J b6 with an acceptable game. I should mention that if 27 .. ."fi e7 I would have replied not 28 tt:J d3?! 'i¥e2 and not 28 'it>f1 ?! bxa4 29 .i. xa4 a5, but 28 tt:Jdf3 ! �e2 29 a5! tt:J e4 30 tt:Jd3 followed by "ii'c1 .

28 a5

tt:Jbd5

29 tt:Jd3 Again Wh ite has the advantage , since the opponent has been left with a bad bishop.

29 . . .

.i.c8

An attempt to bring out the bishop to f5 ; I , of cou rse, prevent this.

30 tt:Jc5

tt:Jf4

probably have been played , but Beliavsky, as on the 1 8th move , does not wish to abandon h i s plan, and he conti nues acting i n the same manner, merely worsening his position .

31 . . . 32 .i.f3

'ifd5 'it'g 5

33 'it11 ! Black's hopes were pin ned on 33 � h2? ifh4, but now he has to sound the retreat. I n add itio n , Beliavsky was i n time-trouble.

33 . . .

tt:J6d5

34 tt:Jde4

'iie 7

35 tt:Jc3 For a player in time-trouble, an u n h u rried manner of play is very u n pleasant. I strengthen my positi o n , I exchange the opponent's active pieces, and he does not know at what moment to expect decisive action .

The creative Solving of irrational Problems in the Opening

35 . . .

'i*'d6

The correct reply. After it I i n itially thought that Black was intend ing to capture o n d 5 with a piece, and I w a s all ready t o p l a y 36 l2lxd5. But suddenly it occu rred to me that he would most probably captu re with the pawn . When trying t o c onvert an advan­ tage there are usually m or e winning chances when ther e ar e n ot id entical, but different pi ec es r emaining on the b oard . It is better to have a bishop against a knight, than a bishop ending. I sensed that i n the event of the captu re on d5 with the pawn it would be more advantageous for me to retai n my knight on c3 , rather than the bishop .

36 i.xd5!

75

h ave a lig ht-square bishop. The correct move, reta i n i ng chances of a successfu l defence for Black, was 37 . . . g5!

37 . . .

h5?

Now I th i n k that Black's position is hopeless.

38 h4 39 'ife 3

'.t>g7 tt:J e6

U nfortunately, here it was my turn to make a m i stake.

cxd5

I was happy to have g uessed my opponent's plans i n time. 37 "ir'd2

40 g3

Here is a l ittle positional problem . H ow, do you th ink, should Black conduct the de­ fence? I n time-trouble Beliavsky did not fi nd the correct solution. In Black's position the weakness of the dark squares is felt. Noth ing can be done about the queenside, but the weaknesses on the kingside could have been covered , by arranging the pawns on d a rk squares: f6g5-h6, as you a re su pposed to do when you

I n principle, I do not recommend taking rad ical decisions on the 40th move . But i n the g iven i n stance there w a s a wi n : two precise moves, after wh ich the opponent cou l d have resigned . 40 tt:J xe6+! i. xe6 4 1 'i!Ve5+ ! 'ii'xe5 4 2 dxe5. T h e i nvasion o f the wh ite king is th reatened , and if 42 .. .f6 there fol lows 43 f4 . I saw that the immediate 40 'ife5+ �xe5 41 dxe5 tt:J xc5 was u n clear, but I simply did not h ave time to th i n k about the prel iminary exchange on e6 . Such mistakes usually prove costly. Black's position is bad , but he is hoping to construct and hold some kind of fortress. The process of trying to breach the opponent's defences may p rove long and d ifficult (i ndeed , the game d ragged out for more than a nother 40 moves). If a possibil ity p resents itself of immediately and advantageously changing

76

�

The creative Solving of irrational Problems in the Opening

the cou rse of the play, it should be checked and exploited . Subseq uently such a conven­ ient way may no longer occu r.

[I think, nevertheless, that Dolmatov cor­ rectly avoided forcing events. After 40 tt:Jxe6+ Ji. xe6 41 'iVe5+ 'iVxe5 42 dxe5 f6 43 f4 Black would have succeeded in creating counterplay by 43 . . . fxe5 (43 . . .g5!? 44 exf6+ c;!;xf6 also comes into consideration) 44 fxe5 g5! 45 tt:J e2 (45 g3? gxh4 46 gxh4 c;!;g6 will not do, and 45 hxg5 c;!; g6 is also dubious) 45. . . gxh4 (but not 45.. . c;!; g6? 46 tt:J d4) 46 tt:J f4 rtJf7, and the vulnerability of the g2pawn seriously restricts White 's possibilities - Dvoretsky.] 40 . . .

f6

Here the game was adjou rned . U nfortu­ nately, in my analysis I was unable to fi nd the strongest plan for converting my advan­ tage , which was later demonstrated by Beliavsky.

41 r;!;e1 The sealed move . When I adjou rned the game I had in mind a sensible plan. The black knight will soon retreat to c?. Wh ite can play his knight to f4 , but no d i rect win is apparent. And since this is so, I decided , I need to play f2-f3 , g3-g4 , and then advance the pawn to g5 to wrest control of the dark sq uares. But before advancing the pawns , the king must b e moved away from th is wing . I nstead of 41 'it> e 1 Beliavsky suggested playing 41 tt:Je2 tt:Jc? 42 tt:Jf4 'it>f? 43 "ii'c1 ! . I did not see the retreat to c1 , and only looked at 43 'iVc3 . The main th reat is tt:J cd3 and then 'it'c5 . With the q ueen on c3 there is the defence 43 . . . 'i!Vc6 (the king stands at f? , so that tt:J ce6+ is not possible). With the queen on c1 in reply to 43 ... 'i!Vc6 he did not l i ke 44 tt:Jfd3 (with the threat of 45 'i!V h6) 44 . . . r;!;g? 45 'it'f4. Sasha admitted that because of this plan he even did not want to resume the game.

In fact, his idea is u nconvi nci ng. And the point is not even that if 45 'i*'f4 there is the reasonable defence 45 . . . Ji. f5 (Wh ite's last move is not the strongest - 45 tt:J e6+! 'i!Vxe6 46 'ii'xc?+ wins). It is more important that the bishop can be played to f5 a move earlier ­ instead of moving the king. After 44 . . . Ji.f5! 45 'i*' h6 Ji. xd3+ 46 tt:Jxd3 tt:Je6 Wh ite's advantage is red uced . tt:J c7 41 . . .

�g4

42 rtJd2

Wel l , I thought, he is merely helping me to adva nce my pawns. But my opponent, as it transpired , was trying to prevent the ma­ noeuvre tt:J c3-e2-f4 . It is amusing how d ifferently we approached the position .

Ji.e6

43 f3 44 tt:Je2

-

�f7 hxg4

45 g4 46 fxg4

Ji.e6

47 g5

�f5

48 gxf6+ r;!;xf6 If 48 . . . ifxf6 , then 49 ife5 is decisive. But after the move in the game Wh ite could have won by 49 if h6! with the same idea - 'ii h8+ and it'e5. 49 'ilkg5+?! 50 "iVe3 I had already noticed my mistake and I tried to correct it. I retreated my queen in the hope that Beliavsky, not wishing to allow 5 1 'ii'e5, would return with h i s king to f6 . But he was on the alert, despite mutual ti me­ trouble, and he found the only correct defence.

tt:Je6!

50 . . . (see diagram)

Suddenly it has become altogether unclear how to proceed fu rther. The open position of Wh ite's king ties his hands. I had very little time left on the clock. I real ised that only a tiny bit more , and I would have to give u p

ttJ

The creative Solving of irrational Problems in the Opening

- position after 50 . . . ltJe6 any dreams of w i n n i n g . And here I managed to find a solutio n , perhaps one of the best in the game.

51 'ii'g 3 ! A committing move , since severa l pawns a re immed iately exchanged , and Wh ite remains with only two . I saw that I would obtain a superior knight endgame, but I didn't know whether it was won . On the other hand, even such a possibil ity might subsequently not have presented itself. 51 . . .

'iVxgJ

If 5 1 . . . lLlxc5 52 dxc5 with a g reat positional advantage, since after 52 . . . "it'f6 there fol­ lows 52 'iif4, controlling the very i mportant d4-point. 52 lLlxg3

tt:Jxd4

53 lLlxf5! gxf5 tt:J l2J If 53 . . . xf5 54 xa6 , then 54 . . . tt:J xh4 55 CiJc? is bad for Black, while if 54 . . . W e7 Wh ite decides matters with 55 tt:J c5 tLld4 56 a6 CiJc6 57 'it>e3, intending tLl b3-d4 or W f4-g5 . 54 l2Jxa6

tLlc6

55 lLlc7

tt:Jxb4

56 lLlxb5

'it>g6

57 tt:Jd4 (see diagram)

77

Wh ite had a imed for th is position , rightly assu ming that his wing pawns would be stronger than the opponent's central pawns . I nteresting p l a y com mences. I reckoned that Black had two moves : . . . W h 5 or . . . f5f4 . If 57 . . .f4 I was intending 58 'it> e2 'it> h 5 59 'it>f3 W xh4 60 'it>xf4 . Each side has only one pawn left, but Black cannot save the game. The wh ite king crosses to the queenside, d riving away the knig ht. The d5-pawn merely h i nders Black. The second variation was 57 . . . W h 5 58 W c3 tt:J a6 59 tt:Jxf5 'it> g4 60 tt:J e3 'it> xh4 6 1 tt:Jxd5 W g5 62 W c4 'it> f5 63 'it> b5 tLl b8 64 tLl b4 'it> e6 65 tt:Jc6 tLl d7 66 a6 and wins. Beliavsky fou n d a th i rd possibil ity - he made a cu n n i ng waiting move . But he was no longer able to save the game.

57 . . . 58 We3

tLla6 tt:Jc5

59 'it>f4

tLldJ+ tt:J b4

60 We3 61 �4 62 Wf3 63 tLle2 64 tLlf4

tLldJ+ tLlb4 'it>f6 d4

This is someth ing of a n achievement Wh ite has forced this pawn to advance , to where it will be more easily attacked . But 62 . . . 'it> e5 would have lost immediately to 63 tt:Jd 3 + ! .

78

� 65 'it>e2

The creative Solving of irrational Problems in the Opening

f7

66 'it>d1 ! 67 'it>d2 It is always pleasant to put the opponent i n zugzwang .

67 . . .

'it>f7

There are various ways of w i n n i n g . decided t o captu re t h e pawn .

68 tt:le2

'it>g6

69 tt:lxd4

f4

70 We2

'it>h5

71 Wf3

'it>xh4

time, if the opponent is resisti ng with all his might - i n such cases m i stakes a re prob­ able. Here is a n example from a nother opening: the Caro-Kan n Defence. The game is from the tou rn ament in Hasti ngs, one of my few wins there . I scored only three, but this was good enough to g ive me fi rst place, and even on my own .

Dol matov - Speelman Hasti ngs 1 989/90

72 'it>xf4 The win here is very simple, because it is always hard for a knight to combat a rook's pawn.

72 . . .

'it>h5

73 We5

'it>g6

74 'it>d6

'it>f7

75 'it>c5

tt:la6+

76 Wb6

tt:lb4

77 tt:lc6

tt:ld5+

78 b7

We6

Caro-Kann Defence 1 e4

c6

2 d4

d5

3 exd5

cxd5

4 c4

tt:lf6

5 tt:lc3

e6

6 tt:lf3

i.. b4

nearly always employ the Panov Attack, from where , as in the g iven i n sta nce , play frequently transposes i nto the N i mzo-l ndian Defence.

79 a6

'lt>d7

80 a7

tt:lc7

8 ii.xc4

0-0

81 tt:le5+

'it>d8

9 0-0

tt:lbd7

82 tt:lc4

tt:la8

83 tt:lb6

tt:lc7

7 i.. d 3

dxc4

84 'it>c6 Black resigned . The game could have concluded sooner, if I had played more accu rately. As it was , we saw an interesting a n d , u nfortunately, rather typical picture . Wh ite solved the problems of the position and gained a n advantage. Then he relaxed , committed some i naccuracies and squandered all his advantage or a sign ificant pa rt of it. Then he again com­ posed h imself, again outplayed his oppo­ nent, and again began acti ng carelessly . . . True, it is hard to play the best moves all the

A theoretical positio n . Here there hasn't

The creative Solving of irrational Problems in the Opening

been anyth ing new for a long time. I know the moves 1 0 l:!.e 1 and 1 0 � d 3 . Black replies 10 . . . � xc3 1 1 bxc3 b6 followed by 12 .t b7, and Wh ite develops h i s bishop at g5 with a complicated battle . In the opinion of theory Wh ite does not have any particu lar advantage. . . .

That day I was i n the mood to play someth ing un usual . Especially against Jona­ than Speelman, an out-of-the-ordinary player, who h imself seeks complications. He is a very active player, and it is pleasant to play against him - the games turn out to be interesti ng. I decided to make a com mitti ng move - to sacrifice the c3-pawn by 1 0 ig5!?.

A similar sacrifice had a l ready occu rred i n m y g a m e with J a nos Flesch (Bucharest 1 98 1 ) . However, i n it my opponent did not play 9 . . . tt:lbd7 , but i m med iately 9 . . . � xc3?! 10 bxc3 'i!Vc7 . There followed 11 � d 3 ! ? 1Ll b d 7 ( i f 1 1 . . . 'i!V xc3 , t h e n 1 2 � f4 is strong) 12 �a3! l:!. e8 1 3 tt:l d2! � d8 (totally bad is 13 'i'xc3 1 4 tt:lc4 and wins) 14 'i!Vf3 4J f8 1 5 I/Je4 with an obvious advantage for Wh ite . As you see, in not one of the l i nes did the bishop come out to g5 - at the time I thought that there was noth ing for it to do there . But now I decided to try it. The idea occu rred to me at the boa rd - I hadn't analysed it at home.

79

as follows: 'Si mply physically h ave not managed to make a mistake , since as yet the only independent move I have made is the sensible � g 5 . What could be more natural than this move? The opponent may win a pawn , but he g ives me severa l temp i , which m ust compensate for his s m a l l mate­ rial advantag e . '

12 . . .

ifxc3

1 3 .l:tc1

'it'as

...

Speelman is a bold player a n d , of cou rse , he accepts the sacrifice .

1 0 �g5 ! ?

�xc3

1 1 bxc3

�c7

12 i..d 3! I made th is last move q u ickly, since the pawn sacrifice is the idea behind the move 10 � g5. It is a l ready too late to avoid the critical variations - the cautious 1 2 'ifd3 would allow Black easy equal ity after 1 2 . . . b6 13 � b3 i.. b 7 14 .:tfe 1 .:t ac8. In genera l , this is a creative , problematic decision, involving a considerable risk. During the game I tried to convince myself

When I sacrificed the pawn , I needed to see this position and without fai l fi nd the follow­ ing move . If Wh ite plays someth ing neutra l , l e t u s su ppose 1 4 .l:t e 1 , preparing 4J e5, then after 14 . . . b6 1 5 4Je5 � b7 he remains without any real compensation . Black suc­ cessfu lly completes h i s development, and i n add ition I h ave a n isolated pawn i n the centre . Here even with material equal ity (give Wh ite back h i s pawn on b2) Black reta ins an excellent positi o n . Therefore , I repeat, it was essentia l to see beforehand the following move .

1 4 tt:Je5 ! If n o w Black plays 1 4 . . . b6, t h e n 1 5 'i!Vf3 'i!Vd5 1 6 'ilt' h 3 is stro n g , threatening 1 7 tt:l xd7 . Wh ite develops a dangerous i n itiative. Such a tu rn of events did not satisfy Spee l m a n . 14 . . .

tt:Jxe5

1 5 .l::tc 5

'i!Va3

80

�

The creative Solving of irrational Problems in the Opening

16 dxeS! It is unfavourable to captu re on e5 with the rook in view of 1 6 . . . tt:Jd5 and then . . . f7-f5 - it is not clear how Wh ite can develop his in itiative .

16 . . .

'ii'x cS

After the game Speelman suggested 1 6 . . . tt:Je4 . A q u ite unexpected possibil ity, which would not occu r to everyone. It is a thematic move, which is justified i n the event of 1 7 �xe4 'ii'xc5 . But after 1 7 l:t xc8 Black ends up in a bad position: 1 7 . . . tt:Jxg5 18 .l:f.xa8 .l::i.x a8 1 9 h4. Now 1 9 . . . .l::i.d 8 is not possible because of 20 � xh7+, while the cu n n i ng 1 9 . . . g6 is calmly answered by 20 � b 1 , and the knight is nevertheless lost.

17 �xf6

(now 21 exf6 �g8 22 'ii' h4+ Wg6 no longer works) . Wh ite ca n also g ive perpetua l check by 1 8 'it'g4+ c.t> h 8 1 9 'ii' h4 f5 20 'ii'f6+ 'it>g8 21 'iig 5 + . I s a t while my opponent w a s analysing variations, and suddenly Speel man made a different move. After the game I asked him: 'Were you playing for a win?' I t tra n spired that he had not been playing for a w i n . He was afraid of captu ring 1 7 . . .gxf6 on account of 1 8 'i!Vg4+ 'it> h8 1 9 'i!V h4 f5 20 'i¥f6+ W g8 2 1 Iie 1 !? � d7 22 l:t e3. T h e o n l y defence is 22 . . . ii'c 1 + 23 � f1 ikxe3. Fortu nately for me, Speelman reckoned that Wh ite stil l had winning cha nces. Alas, after the game we did not fi nd a win here . In add ition, Reiner Knaak found another defence: 2 1 . . .'it' c7 ! ? 22 .l:i.e3 .l:t d 8 , w h e n Wh ite is obl iged t o give perpetual check by 23 l::!. g 3+ 'it>f8 24 'ii'h8+ rt; e 7 25 'ii'f6+ 'it>f8 .

17 . . .

During the game I reckoned that this position was a draw. That is what I sti ll think. I played well , the opponent also played worth ily - what could be done, the game should end in a d raw. I thought that 1 7 . . . gxf6 would a utomatically be played . The plausible attacking attempt 1 8 'ii'g4+ W h8 1 9 exf6? is refuted by 1 9 . . . .l::i. g 8 20 'it'h4 h 5 , and Wh ite loses. A draw occu rs in the variation 1 8 � xh7+ c.t>xh7 1 9 ii'h5+ c.t> g8 (of cou rse, not 19 . . . c.t> g7?? 20 exf6+ and wins) 20 'ii'g4+ c.t> h7 (but not 20 . . . c.t> h8? 21 exf6) 21 'ir' h4+

.l::i. e 8?!

When Jonath a n played this, i n my su rprise I became slig htly fl ustered . I had been ex­ pecting a d raw, and now I looked and couldn't see a perpetual check. And I was the exchange and a pawn down . 'We l l , I 've been u n l u cky! ' , I thought. For some five m i nutes I cou l d n 't see a nyth ing sensible. I was diverted by all sorts of attacks l i ke 1 8 � b5, and I was very su rprised that I couldn't fi nd a d raw. And suddenly I realised that it was in vai n that I was seeking a d raw - I should be playing for mate!

Wxh7 1 8 �xh7+ ! ! 'ii W 1 8 . . . f8? 1 9 g4 gxf6 20 exf6 was bad for Black. 1 9 ikhS+ 20 ikg S

'it> g 8

'i¥f8

2 1 l::i. d 1 ! I th i n k that Speelman missed the sacrifice on h7 with the subsequent inclusion of the wh ite rook in the attack (although the idea is roughly the same as i n the variation he calculated after 1 7 .. .fxg6). Otherwise he

ctJ

The creative Solving of i rrational Problems in the Opening

would not have gone in for this positio n . How can h e now defend against the switching of the rook to the g- or h-file? However, my opponent again su rprised me he found a way of prolonging the resista nce for nearly a fu rther fifty moves .

-

[As was later established, the game would not h ave been prolonged if White had carried out more accurately the idea of switching the rook to the g-file: 19 'ikd3 + ! (instead of 1 9 'ikh5 +) 1 9 . . c6

41 . . .

'it>e2

42 h 5

l:lecB

4 3 iif1 +

'it>d2

44

iib5

45 g3

'it>e3 'it>d4

46 'it>g2 Everything seems to be a l right: Wh ite has not allowed . . . b6-b5, he has strengthened h i s positi o n , and he has beg u n advancing his pawns. And I a l most stopped payi ng any attention to Speelman's moves.

46. . . .

'it>c3

47 h6?! 4 7 g4! was far more accu rate.

47 . . .

The h-pawn can not promote on its own - it needs the su pport of the g-pawn . But Black is endeavouring to obtai n a passed pawn on the queenside. To participate i n a race , to see who is quicker, is unpleasant. Therefore I decided first to halt the opponent's counterplay, even if only for a time, and then to advance my pawns.

31 'ii'f3+

'it>c5

32 'ii'e 3+

'it>c6

33 'ir'f3+

'it>c5

34 'ika3+

'it>c4

35 'ii' b 3+

'it>c5

.l:.bB 'ik 37 b5+ was threatened . Speelman de­ fends his pawn and prepares . . . b6-b5. 36 a4!

37 'i¥c3+

'it>d5

38 f4

'it>e4

39 'ii'f3+ Also strong was 39 'i¥c4+ ! ? 'it>e3 40 g3 with a win ning position .

39 . . .

'it>d4

40 'ifc6

'it>e3

41 iVc1 + 41 g3! was simpler.

.l:!.hB

There is no longer a simple win. Here I decided to complicate matters, by finally al lowing the opponent to play . . . b6-b5 .

4 8 'ir'd7

b5

49 axb5

a4

50 b6

a3

51 'ii'a 4

'it>b2

52 'ii' b4+

'it>a2

53 h 7 ! By d riving t h e k i n g i n front o f t h e pawn , so that it can not advance, Wh ite has i n d i rectly defended h i s h-pawn . It ca nnot be taken, since the rook is lost. But even so, the position is not easy - I have allowed Black to advance h i s passed pawn too far. The win is achieved by j ust one tempo.

53 . . .

:bcB

54 b7

.l:tc2+

55 'it>f3

l:tb2

56 'ii'c 4+

l:!.b3+

57 'it>g4

'it>b2

.l:!.xh7 'ik 'ir' After 58 . . . a2 59 xh8 a 1 (59 . . . .l:!. xb7 60 'ii'g8 a 1 'ii' 61 h8 'ir' ) 60 b8'ii' 'ikd 1 + 61 'it>g5 Wh ite avoids the checks , since the rook on b3 is pinned . 58 'ii'c B

5 9 bB'ik

.l:tg7+

The creative Solving of irrational Problems in the Opening

60 'it>h5

l:r.gxg3

61 'ti'd6

.l::!. h 3+

62 'it>g6

l:!.bg3+

63 'it>f7

.Uh7+

64 'it>xe6

l:th6+

65 'it>f5

llxd6

66 exd6

a2

67 d7

a 1 'ii'

68 'ii'b 7+ The next move will be 69 d8'ii' , and there is not a single check. Black resigned . So, I have acq u a i nted you with one of my approaches to worki ng on chess. I n each of the ga mes exa m i ned we encou ntered -

83

most ofte n , i m m ediately after the opening ­ the problem of a key move , a move which g ives the g a me a defi n ite d i rection and exerts a n enormous infl u ence on the fu rther cou rse of the play. For me this is a deeply creative problem , associated not only with p u rely chess laws, but also with intuition , emotion , and psychological cond ition. The d iscoveries which I h ave shown could be conceived only d i rectly at the boa rd , i n the cou rse of a tense battle. Therefore I advise you : don't forget to strengthen you r m i n d , develop you r i ntuiti o n , and learn t o control you r emotions. Don't restrict you rself only to the acq u isition of knowledge, but try to i m p rove in the most varied fields.

84

� Mark Dvoretsky

A practical Exercise

I pendently seeking the replies to q uestions

invite you to test you r strength by i nde­

which the players faced in the open ing stage of one qu ite old game.

I found this game i n a splendid book publ ished in 1 979 Het groot analysebook Jan Timman, which in translation from the Dutch means 'Jan Ti mman's Big Book of Analyses ' . I studied Ti mman's analyses with great interest and benefit to myself, a n d , of course, I found many mistakes i n them. When a commentator does not restrict hi mself to general remarks , but tries to analyse a game in real depth , mistakes are inevitable in view of the complexity of the problems facing the analyst. Many of the mistakes poi nted out by readers of the book were corrected i n its English edition The Art of Chess Analysis. And i n 1 989 Ti mman presented me with a new, French edition L'art de L'analyse. Ti mman's book comprises commenta ries to games, written by h i m in various yea rs and publ ished in the Dutch chess magazi ne. The game which you will see is one of the fi rst in the book, and is perhaps annotated less wel l than t h e others . T h i s is a vivid illustration of the fact that in his youth Timman was a n ind ifferent analyst - su bseq uently, after gaining experience , he began analysing much better. The time you have for the solving of the problems (with some of which even the commentator did not cope) will not be long: from 5 to 15 min utes. But don't be afraid you will be helped by lead ing questions, outlining the problem more specifical ly. Exercises offered in competitions usually have completely clear-cut solutions - for

example, a forcing combination or an end­ game that can be accu rately calculated . Our competition is a not altogether sta ndard one. Many of the problems a re open to discussion . Sometimes it will be hard to demonstrate this 'th is' is better than 'that' . You will have to trust in you r general perception of the positi o n , you r intuition. Calcu lation of variations is also req u i red , of cou rse, but it will be more importa nt to see all the resou rces both for you rself, and for the opponent, and to eva l u ate correctly the resulting situations. The questions will be purely practical and they should be answered from the position not of an analyst, but of a practical player. You r objective is to find in a restricted time the most i m portant, most sign ificant varia­ tions for the taking of a decision , and to g uess the opti mal way of combining calcula­ tion of variations with eval u ation of position.

Polugayevsky - Mecki ng Mar del Plata 1 97 1

Semi-Slav Defence 1 c4

c6

2 lt:lf3 3 e3

d5 lt:l f6

4 lt:lc3

e6

5 b3

lt:lbd7

6 �b2

�d6

7 d4

0-0

Had Wh ite now played 8 �c2 , this would have transposed i nto a well-known position from the Mera n Variation (1 d4 d5 2 c4 c6 3 lt:lf3 lt:l f6 4 lt:l c3 e6 5 e3 lt:l bd7 6 'i'ic2 � d6 7 b3 0-0 8 ..t b2), in which the light-sq uare

A Practical Exercise

bishop is usually developed on e2. But Lev Polugayevsky deviates slig htly from the usual set-up. 8 �d3

�e8

9 Vi'c2 And now - the fi rst q uestio n .

85

Now rega rd i ng preference. 9 . . . 'it'e7 is the move that I l i ke most. Here , of cou rse, I ca n not demonstrate anything - I ca n only explai n . What is the idea? Let's investigate . Why d id n 't Wh ite castle j ust now? Because of the reply 9 . . e5 ! . He does not have time to captu re on d5 in view of the fork 1 O . . e4 , and so he has to exchange on e5 in a situation advantageous to Black. 9 'ii'c2 was played in order to take control of the e4-square , and i n the event of 9 . . e5 to be able to exchange on d5. And now comes a rather refined reply - 9 . . . 'it'e7. In itself it is usefu l in such positions, but i n add ition Black renews the th reat of a fork after . . . e6-e5. .

.

.

Only Vasya Emel i n chose this move , for which I have also g iven h i m an extra point.

1 ) Suggest for Black t h e m a i n candidate moves ( 1 0 m i n utes). So, nearly all of you correctly determ i ned the main ideas for Black. You even sug­ gested some which I did not have i n m i n d , but also deserve consideration . I thought that here there are th ree main possibil ities:

1) 9 . dxc4 1 0 bxc4 e5 - the standard plan in such positions, although when the bishop is on e2 - with the bishop on d3 it looks weaker; . .

2) the i mmediate 9 . . . e5 - each time in such cases you must consider whether you should allow the exchange on d 5 ; 3) the preparatory move 9 . . . 'it'e7; t h e i d e a of it will be explai ned a l ittle later. Vadim Zviagintsev suggests a completely different pla n : the completion of Black's development by . . . b7-b6 and . . . � b7. I am not prepared to comment on this sugges­ tion, since I have not analysed it. But it looks sensible, sometimes Black plays this way i n similar positions, and therefore i t receives an extra point.

9 . . . 'it'e7 was not made i n the game or indicated i n the notes. In effect this is an open ing novelty, and not a bad one. This is how they a re devised : one only needs to carefu lly analyse a game or o pen i n g varia­ tio n , to delve i nto its latent ideas . . . As candidates , Yan Tepl itsky suggested nearly a l l possible moves up to an including 9 . . c5 - this is rather excessive. With such a n abundance of possibil ities it is hard to hit the target, and it demands too much time. Endeavo u r with the help of evaluation to somehow restrict the list of ca ndidate moves . .

9 . . .

eS

1 0 cxd5

cxdS

In such positions there is sometimes the typical move tLl b5, but, of cou rse, not here ­ because of the check on b4 .

1 1 dxeS

ttJxeS

1 2 ttJxeS

�xeS

1 2 . . . .l:i. xe5 !? is more active .

1 3 tLle2 Polugayevsky aims to simpl ify the position , to exchange the da rk-sq uare bishops. Black again faces a com m itting decision .

86

�

A Practical Exercise

the check on a5 also has no particular point. Another active try is more interesting 1 3 . . . d4! ? . If 1 4 f4? or 1 4 .l:td 1 ? , then 1 4 . . .'ii' a5+ is strong.

2) How should Black conti nue? ( 1 0 m i n utes)

In reply to 1 4 exd4 one of you suggested 1 4 . . . j_ d6!? with the idea after 1 5 h3 'ii'a5+ 1 6 Si. c3 'ili'g5 of developing pressu re on Wh ite's kingside. Clever! But Wh ite can simply castle , and the bishop sacrifice, u nfortunately, does not work: 1 5 0-0? j_xh2+ 1 6 � xh2 tbg4+ 1 7 'it> g 1 'ii'd6 ( 1 7 .. .'it' h4 1 8 'ii'c7! ) 1 8 tb g 3 'ii' h6 1 9 l!fe 1 'if h2+ 2 0 'iitf1 . H owever, the simple 1 4 . . . j_ xd4 is qu ite sufficient for equal ity. The critical reply is 1 4 e4 ! , and it is on this that the eval u ation of 1 3 . . . d4 depends .

With qu iet play Wh ite gains a slight but enduring positional advantage. Therefore in the fi rst instance one should look at moves which disru pt the ' normal' cou rse of play. The first try is 1 3 . . .'ir' a5+ 1 4 j_ c3 j_ xc3+ 1 5 'i¥xc3 'i¥xc3+ 1 6 tbxc3 , and now 1 6 . . . d4 , because otherwise Wh ite has the more pleasant endgame. After 1 7 lb b5! B lack has to sacrifice the exchange: 1 7 . . . dxe3 1 8 tb c7 exf2 + . Nearly all o f y o u reached t h i s position i n you r calculations, b u t not everyone w a s able to accu rately calculate the variation to the end. However, in the fi rst edition of his book Timman also went wrong , attach ing an exclamation mark to 1 9 �d 2 ? , which is in fact refuted by 1 9 . . . l:i. d8 20 tbxa8 j_f5 . Of cou rse , 1 9 'it> xf2 tb g4+ 20 � g 1 .l:!. d8 2 1 tbxa8 .l:!. xd3 2 2 h 3 i s correct. This position could have been u nclear i n view of the bad placing of the wh ite king (imagine if the black rook were on d 2 ! ) , but here Wh ite is alright: 22 . . . tbf6 23 � h 2 or 22 . . . tb e3 2 3 'it>f2 . For the exchange there is no real compen­ sation. This enti re variation subsequently occu rred in the game Makarychev-Chekhov ( Mos­ cow 1 98 1 ). Since the exchange sacrifice is i ncorrect,

3 ) What does 1 4 . . . t2J xe4 lead to? (5 m i nutes) Absol utely all of you gave the correct evaluatio n . After 1 5 j_ xe4 d3 1 6 ..t xd3 Si. xb2 White should play not as suggested by Tim man - 1 7 .l:i. d 1 ? 'ii'a5+ , but simply 1 7 Si. xh7+! 'it> h8 1 8 'ii'xb2 'it>xh7 1 9 0-0, and Black has no compensation for the sacri­ ficed paw n . W h a t t h e n s h o u l d Black d o ? Wh ite's pawn structu re is better, and he is th reatening to play f2-f4 . The endgame after 1 4 . . .'iV a5+ 1 5 'ii'd 2 i s clearly i n Wh ite's favou r. Black must play as energetically as he can , and I think

ctJ

A Practical Exercise

that the only serious option is 1 4 . . . tt:J g4 ! . But how to reply to 1 5 h 3 ? It is possible to prevent the opponent from castl i n g , but after 1 5 . . .'i' a5+ 1 6 Wf1 the advantage is with White. 1 5 . . . ii' h4 is stronger. Now if 1 6 0-0 there is the g uerri l l a raid 1 6 . . . lt:J h 2 ! , when all kinds of sacrifices a re i n the a i r: . . . ..t xh3 or . . ..�'l f3+ . However, by playing 1 6 g 3 , Wh ite sudden ly reminds Black that for a long time his d4-pawn has been en prise. But Black too does not lose heart, since after 1 6 .. .'i!t' h6 17 tt:Jxd4 'ir'b6 or 1 7 . . . .l:. d8 the turmoil on the board does not e n d . You don't know, by chance, how this fi nal position should be evaluated? In analysis I found this entire series of moves, by looking at and rejecting other continuations. It is clear that B lack has the right to play this way. It is risky for h i m , for h e remains a pawn down , b u t for Wh ite too there is a risk, since his kingside is weakened and h i s king is stuck in the centre . Thus we have established that Black had a not very successfu l queen check at a5 and a tempting attempt to i n itiate complications by 1 3 . . . d4!?. But Mecking chose another move. 'ir'd6 13 . . . I n order to g ive it a n objective eva luatio n , another question must fi rst b e a n swered .

4) Comment on the series of moves made in the game ( 1 5 m i n utes) 'ir'xe5 14 ..txe5 ..td7

15 0-0 16 lt:Jd4 (see diagram)

First let's evaluate the resu lting position . After the exchange o f the dark-sq uare bishops White has retained a slight positional advantage , that is obvious. It is determi ned by the isolated d 5-pawn and the opponent's passive l i g ht-squ a re bishop. It is not clear whether Wh ite can win , but at any event he will be able to press for the entire game. Could Black have avoided such a fate?

87

- position after 1 6 tt::l d 4 -

Some of you suggested captu ring on e5 not with the q u een , but the rook. I will not g ive any poi nts for this move , because I don't see any pa rticu lar virtues in it. Wh ite repl ies 1 5 lt:J d4, not fea ring 1 5 . . . 'ii'b4+ 1 6 'ir'd2 'i!Vxd4 ?? 1 7 ..t x h 7 + , and planning .i:!.c1 and then pos­ sibly 'ir'c7. H e ca n also consider 1 5 llc1 d4 1 6 e4 . After 1 4 . . .'�' xe5 1 5 0-0 the recommendation 1 5 . . . tt:J e4 is u n convinci n g . Wh ite places his rook on c 1 and his knight on d4, and at an appropriate moment he can even exchange on e4 , leaving h i mself with a strong knight against a passive bishop. Black's position retains the same d rawbacks as i n the game. O n the other hand, 1 5 . . . tt:Jg4! g ives Black reasonable counter-chances, and enables him to sharpen the position . Of cou rse, d u ring the short time available you would have been u nable to calculate the variations exactly, but you r positional feeling correctly suggested to many of you that Black should have decided on this move . Log ic, typical of such situations, operates: before accepti ng the need to spend the entire game defend­ ing passively, you should fi rst seek active resou rces, which may change the u nfavou r­ able cou rse of the play. And if a move such as 1 5 . . . tt:J g4 is not refuted d i rectly, and it leads to u n clear situations, it should be made.

88

�

A Practical Exercise

However, Wh ite in tu rn could have played more strongly, and not allowed a n u n neces­ sary sharpening of the play. 1 5 �c3! was a good prescri ption, instead of 1 5 0-0? ! . Many of you poi nted this out. It is i mporta nt to dislodge the black q ueen from its excellent central square . The position after 1 5 . . . 'i&'xc3+ 1 6 lt:J xc3 is already known to us from the 1 3 . . . 'ii'a5+ variation - it was not i n va i n that we calculated it. As you will remember, 1 6 . . . d4 1 7 lLl b5 g ives White the advantage . And i f t h e queen moves, Wh ite can castle or first play 16 'i!fd4 . You suggest that 1 5 'it'c3 should be an­ swered by 1 5 . . . �g5 ? Wel l , I can castle , si nce 1 6 . . . � h3 is not dangerous i n view of 1 7 lt:Jf4 . And meanwh ile Wh ite is planning �d4 followed by 'jVf4 , or, even more active ­ 'it'c7. At any event, the cou nterplay which flared up after 1 5 0-0?! tt:lg4! is not obta ined

fi rst exercise , he again suggested a mass of possibilities, including some that were cor­ rect, but he did not indicate any clear preference. You should not be afra id to express you r opinion . Of cou rse, it may be mistake n , but you learn from you r m istakes. After a l l , the a i m is not to score the most points in the com petiti o n . We a re train ing you r approach to the taking of decisions, which you will be able to use i n practical games. There , in the e n d , you will be obl iged to d raw a clear concl usion , as to which you move you l i ke less, and which you l i ke more. But now we will carefu lly analyse the varia­ tion 1 5 0-0 lt:J g4 ! . H ow to defend agai nst the mate? The fi rst try is 1 6 g 3 'i¥ h 5 1 7 h4.

here. Volodya Balkan suggested the variation 1 5 "iic3 "it'h5 1 6 0-0 lt:J g4 1 7 h3 tt:l e5. You know, after 1 8 lLlf4 Wh ite is sti l l a l ittle better, si nce the d5-pawn is weak. Also I am not obl iged to castle, but can play 1 6 ir'd4 or 1 6 ir'c7 . But in genera l , this is the correct approach ­ Black must somehow try to sharpen the position . I have g iven him an extra point for his attempt to analyse 1 5 'it'c3 . Now we ca n g ive an objective eval u ation to the move 1 3 . . ."it' d6. It leads to a rather inferior, passive position . The more dynamic 13 . . . d4!? should have been preferred . I have given the maximum score - 1 0 points - to those who indicated the idea of 1 5 . . . lt:J g4! , giving Black cou nterplay, and 1 5 'i!fc3 ! for Wh ite instead of castl i n g . I n two cases I gave a score between five and ten , when the reply was based on the correct premise, that Wh ite needs to devise some­ th ing, since after 1 5 0-0 there is the strong rejoinder 1 5 . . . lt:Jg4! , but 1 5 ii'c3 was not suggested . I gave the fewest ma rks to Yan Tepl itsky (I warned h i m ! ) , si nce, as in the

5) What should Black play? (5 m i n utes) Here Black has the powerfu l move 1 7 . . . g5! , which g ives h i m an excellent position with a strong attack. The th reats a re . . . gxh4 and in some cases . . . lt:J e5. You analysed 1 8 'it> g2. 1 had in mind 1 8 . . . gxh4 1 9 .l:!. h 1 h 3 + . I n one of the replies it is written that 1 8 . . . gxh4 is bad because of 1 9 lt:J f4 . I don't th i n k so - after 1 9 . . . h3+ 20 'it> h 1 "it' h6 Black is wel l placed . H owever, it is possible that 1 8 . . . l1xe3 is even stronger. The one who suggested this receives a n additional point. The only d ifficulty of the exercise is that there is another tempting move 1 7 . . . lt:J e5, which is not so easy to refute . This was the

A Practical Exercise

one recommended by Tim ma n . But if the variation is conti nued : 1 8 tt:)f4 tt:) f3+ 1 9 'it>g2 'i'g4 (the piece sacrifice 1 9 . . . tt:) xh4+ is also insufficient: 20 gxh4 'ii'g 4+ 21 'it> h2 'ii'x h4+ 22 'it>g 1 'ii'g 5+ 23 tt:) g2 i. h 3 24 f4 'ii'x g2+ 25 'fxg2 i. xg2 26 'it>xg2 .l:t xe3 27 l:tf3 or 27 l:i ad 1 with a difficult endgame for Black) 20 l::t h 1 with the extremely u npleasant th reat of 21 i.e2 (or even 20 i. e2 tt:)xh4+ 21 'it> h 1 ), it becomes clear that the position should be evaluated i n favour of Wh ite . Let us conti nue looking at the defences against the mate th reat after 1 5 0-0 tt:) g4. The second try is 1 6 tt:)g 3 . Here , there is really not even anything to ask - it is clear that Black plays 1 6 . . . h 5 ! . And then , say, 1 7 .!:Ife 1 h4 (there is also 1 7 . . . tt:)xh2!?) 1 8 tt:)f1 h3 1 9 g3 'ii'f6 followed by . . . tt:) e5. Black begins an attack on the l ight squares; h i s position is better. Timman recommends 1 6 tt:)f4 and comes to an amusing conclusion: after 1 6 . . . tt:)f6 (in­ tending 17 . . . d4) 1 7 tt:)e2 tt:) g4 1 8 tt:)f4 the result is a draw. But does the knight have to retrace its steps? Black has two active moves : 1 6 . . . g5 and 1 6 . . . d4.

In t he fi rst, Dutch edition of h i s book, Ti m­ man considered both moves to be bad, i n the English edition - o n l y o n e o f them , a n d only in the French edition (taking account of an article of m i n e in the magazine New in Chess, poi nting out the mistakes in his book) was the correct eva l u ation g iven both moves secu re Black an excel lent game. I w i l l g ive one variation now, and you can try to find the second you rselves. 16 ... g5 17 h3 gxf4 ! 18 exf4 Wxf4 1 9 hxg4 ixg4 with advantage to Black. If Wh ite regains the pawn by captu ring on h 7 , Black will attack along the h-fi l e . Where could Timman have gone wrong here? H e was carried away by a n attempt to win a paw n : 1 7. . . tt.J xe3? 1 8 fxe3 'ifxe3+ 1 9 'it> h2 gxf4 20 l:if3, after which it is Wh ite who builds u p a powerful attack.

ctJ

89

The second variation is 1 6 . . . d4 1 7 i. xh7+ 'it> h8 1 8 h 3 .

6 ) What position s h o u l d Black go i n for? ( 1 0 m i n utes) Black has n u merous tempting conti nua­ tions. There a re variations that a re rather sharp and complicated , and i n 1 0 m i nutes of cou rse, you will not calcu late them all : The a i m is not so much to calcu late , as to assess and sense where you will sta nd better, where you r position will be more secu re. Let's try. Timman exa m i nes th ree variations. The fi rst: 1 8 . . . tt:) xf2 1 9 'ii'xf2 dxe3 20 1i' h4 g5 21 'if h6 'ii'g7 22 'ii'xg7+ 'it> xg7 23 tt:) h5+ 'it> xh7 24 tt:)f6+ , and Wh ite wins. The seco n d : 18 . . . tt:)f6 1 9 i. g 6 ! ! (the f7-pawn is awkward to defend) 1 9 . . . dxe3 20 i.xf7 'ifxf4 2 1 i. xe8 tt:)xe8 . Black has won two p ieces for a rook, but with such a king he ca nnot survive : after 22 fxe3 ! 'iVxe3+ 23 'it> h 1 Wh ite's th reats a re i rresistible. I nciden­ tal ly, i n stead of 21 . . . tt:)xe8? it is fa r stronger to play 21 . . . i. f5 ! 22 'ifc5 e2 23 1i'f8+ tt:)g8 24 l:.fe 1 1\Ve s . And , finally, the th i rd : 1 8 . . . dxe3 1 9 hxg4 exf2+ 20 Wxf2 ! 'it> xh7 2 1 J:t a e 1 . But why does Black commit hara-kiri , by h imself opening l i nes for the wh ite rooks? I n

90

�

A Practical Exercise

this last variation he has the simple 1 9 . . .'�i'xf4! (instead of 1 9 . . . exf2+?) with bri l l iant pros­ pects. Thus we see that in the event of 1 6 lt:Jf4 Black is by no means bound to ag ree a d raw - he straight away has two tempting possibilities. What then should Wh ite do? Did the inaccu racy 1 5 0-0?! really lead to a n inferior position for him? I don't th i n k so. We must consider one other conti nuatio n : 1 6 .t xh7+ 'lt> h8, and only now 1 7 lt:Jg3. Evidently Black should win the bishop for th ree pawns: 17 . . . g6 18 ii. xg6 fxg6 19 'ili'xg6, and the n , most probably, 1 9 . . J 1 g8 20 � h5+ �xh5 2 1 lt:Jxh5.

b y t h e method o f e l i m i n atio n , choose the piece sacrifice 1 6 ii. xh 7 +, after sensing (it is not possible to calculate to the end), that everyth ing else is dangerous for Wh ite, whereas here an u nclea r endgame is reached . We are now fi nished with the open ing stage. Play has gone i nto a q u iet midd legame. It would be wrong to th i n k that Wh ite has a big advantage, and certainly not a winning position . In such situations experienced and cool defenders are usually able to save the game. But what told su bsequently was the differ­ ence in class between the two players. Polugayevsky was a matu re , positional competitor. As for Henrique Mecking . . A year after this game Tigran Petrosian wrote a bout h i m : 'He is indeed not a bad player. He will possibly play better, but I am sure that he will never become world champion. And mainly because of the narrowness of his chess thinking. Mecking does not under­ stand, for example, the significance of weak and strong squares. I have played him three times. In 1 968 he lost to me because of the weakness of the light squares. A year later he readily conceded all the dark squares to me, and again suffered a defeat. And at the tournament in San Antonio ( 1 9 72) grand­ master Mecking again gave me control of the dark squares, and with them also victory. Mecking is distinguished by his active piece play, but he does not have a proper understanding of the deep features of a position, and this forces me to have doubts about his chess future. ' A severe but instructive 'diag nosis' . .

When solving studies, if you are fortunate enough to find a series of best moves, the evaluation of the final position usually does not present any difficu lty - it is either a wi n , a draw, or a loss . A practical game is far more complicated . Here forcing variations very often end in completely u nclear positions. I don't know how to eval uate this position ; if anyone knows, please tell me! But at any event I th ink this is the best that Wh ite has after 1 5 . . . lt:J g4 ! . I magine how difficult h i s task would have been ! He would have to examine the variations 1 6 lt:J g 3 , 1 6 lt:Jf4 and 1 6 g3. The n ,

I should mention that a weakness on squares of a particular colour usually arises when pawns a re arranged on squares of the same colour as a bishop. As we will now see, it is this elementary positional m istake that is com mitted by Mecki n g . Let us s e e h o w t h e g a m e developed .

ltJ

A Practical Exercise

16 . . .

ltac8

1 7 'ii'e 2

'ii'd 6

1 8 'iVb2 Note that if you have a l ig ht-sq u a re bishop, your queen should usually be positioned on squares of the opposite colour. Wh ite covers the vulnerable poi nts on the queenside and prepares a possible adva nce of h i s pawns there. 18 . . .

a6?!

Petrosian is right - Mecki ng does not know on which sq uares he should keep h i s pawns. lb g 4 1 9 l::ta c1 Too late! Now this no longer has any g reat point.

20 lDf3

'fi'b6

21 .l:i.xc8

l::t x c8

22 .l::i. c 1 Polugayevsky operates in accordance with a well-known principle: 'Ag a inst youth - go into the endgame!' It is easy to see that 22 ... ttJ xe3? does not work: 23 .l:txc8+ � xc8 23 'i'c1 or 23 'ii'e5. lbt6 22 . . .

An old Russian proverb states : 'When two do one and the same - it is not the same . ' Polugayevsky, l i ke Mecking before h i m , has placed a pawn on a sq u a re of the same colour as his bishop. But he wants to adva nce it fu rther, fixing the queenside. And if the opponent forestal l s Wh ite's plan by 26 . . . a5, the n , as shown by Tim m a n , there follows 27 � b5! � xb5 28 lbxb5. This device , a rather d ifficult one, i nciden­ tal ly, demanding a subtle eval u ation of the position , is called 'transformation of an advantag e ' . Wh ite g ives u p one of his pluses: he exchanges the opponent's 'bad' bishop, but i n return he hopes to gain another one: superiority i n the placing of his pieces. It is hard for B lack defend the i nvasion squares. For example, he loses a pawn after 28 . . . 'fi'c6 29 'iixc6 bxc6 30 lLJ d4 c5 3 1 tt:J c6 . But in the event of 28 . . . 'fi'f6 !? chasing after the pawn by 29 'iixf6 tt:Jxf6 30 lLJd6 b6 31 tt:J c8 is not justified , since in return Black activates his king : 31 . . . 'it>f8! 32 tt:Jxb6 'it> e 7 33 tt:Jc8+ (33 f3 'it> d6) 33 . . . 'it>d7 34 lb a7 'it> d6 35 tt:J b5+ 'it> c5. Wh ite should simply reply 29 tt:J d4 ! , reta i n i n g an appreci­ able advantage.

23 .l:txc8+

� x eS

26 . . .

"f/c7

24 'it'c3

� d7

27 'ii'x c7

lbxc7

25 lLJd4

tt:J e a

26 a4!

91

28 a5 In Ti mman's opinion , here Black was obl iged to try 28 . . . tt:J e6 ! ? . If 29 tt:Jxe6 , then 29 .. .fxe6 30 f4 'it>f7 , followed by . . . h7-h6, . . . 'it>f6 and . . . e6-e5. This was his best chance, promis­ ing real hopes of a d raw. I n stead there follows a s u perficial move .

28 . . .

'it>f8?!

29 'it>f1 Now this idea does not work - after 29 . . . tt:Je6 30 tt:Jxe6 fxe6? the h 7-pawn is hanging . But Black cou l d have played 29 . . . h6 and then . . . tt:J e6. 'it>e7 29 . . .

30 'it>e2

92

�

A Practical Exercise

his bishop to f3 : 36 ii. e2 ii. d 7 (36 . . . d4+ 37 exd4 exd4+ 38 'it> xd4 i. xb3 39 ii. f3 'it> c7 40 'it> c5 is hopeless for B lack) 37 ii. f3 i. c6 38 'it> b4 (with the th reat of 39 fxe5+ 'it>xe5 40 'it> c5), and the pawn endgame arising after 38 . . . d4 39 ii. xc6 bxc6 40 exd4 exd4 4 1 'iiic4 c5 42 b4 cxb4 43 �xd4 is won . And now the last exercise .

30 . . .

g6

It is laughable that a g randmaster should play this! He sees that if 30 . . . lt:J e6 there is 3 1 lt:Jf5+ , and without hesitation h e places another pawn on a sq uare of the colour of his bishop. 31 'it> d2 tt:Je6

32 tt:Jxe6?! I fi nd this move hard to understa n d . 32 � c3 suggests itself.

32 . . .

fxe6

33 f4

e5 'it> d6

34 g3

Here Ti mman demonstrates with a serious piece of analysis that Black could have held the position by 34 . . . ii. b 5 ! . It is probable that this was indeed his best chance. At any event, Wh ite could not have gone i nto the pawn endgame, si nce after 35 ii. xb5 axb5 36 �c3 'it> e6! if 37 'it> b4? there is the reply 37 . . . d4 ! .

3 5 'it>c3 Now 35 . . . i. b5 is no longer possible - in the variation 36 i.xb5 axb5 37 'it> b4 d4 Wh ite captu res on e5 with check.

35 . . .

i.e6

36 'it>b4?! I n the later editions of his book Ti mman rightly poi nts out that, before embarking on resolute action , Wh ite should have played

7) What would you have played? (5 m i n utes) Mecki ng has a l ready spoiled his position so much , that it is u nclear whether he can now save it. But one should fight in any situation. The th reat is 37 fxe5+ W xe5 38 'it>c5, brea king through on the dark sq u a res. I n the event of the exchange on f4 , the wh ite king goes to d4. It is improbable that Wh ite would fa il to convert such a n enormous positional advantage - as Black's pawns are on squares of the same colour as his bishop. 36 . . . d4! is the most natural move (at least one pawn moves onto a dark square), and for this reason I gave you only five m i n utes. There is no need to calculate it exactly; it is sufficient merely to slig htly weigh up the resou rces of both sides. It may also lose, but it may not, and in any case it will be not so simple. Wh ite must decide : whether to captu re on d4, or on e5, or perhaps to play 37 e4 . And who knows which reply is correct?

C2J

A Practical Exercise

I n itially Ti mman d id not consider this de­ fence at a l l . And when I mentioned it in my article, i n the French edition of h i s book he wrote that Wh ite wins with 37 e4 fol lowed by 38 fxe5+ 'it> xe5 39 'it> c5 � xb3 40 'it> b6. But after 40 . . . � d 1 41 'it> xb7 � f3 B l a ck has noth ing to fear. I n add ition , if desired Wh ite's 'threat' can easily be parried by 37 . . . � d 7 . The strongest conti n uation is 37 exd4! exd4 38 h4 followed by h4-h5 . It is hopeless to go into the pawn endgame: 38 . . . � f5 39 � xf5 gxf5 40 '>t>c4 d3 4 1 'it>xd3 'it> c5(d5 ) 42 h 5 ! . I n a new English ed ition of h i s book, published in 1 997, Timman tried to show that Black could save the g a m e by 38 . . . '1t> c6! . At the end of the variation s u g gested by h i m , 39 h5 gxh5 40 � e4+ ! � d 5 41 i. xh7 b5 White has the winning manoeuvre 42 � f5! followed by 43 � c8, but Black's last u nfortunate move can be replaced by 41 . . . b6! .

36 . . .

93

or h2-h4) would prove most opportu ne. Pol u gayevsky had probably a l ready seen how he would break throug h , and he knew that he would not req u i re any reserve tempi to g ive h i s opponent the move .

41 . . .

�h1

42 i.e2

�g2

43 �g4

ii.e4

44 .tea

rt;e7

45 i.e6

'>t>d6

46 ii.g8

h6

47 ii.f7

h5

Forced .

48 �e8

�e2

49 ii.f7

i.e4

exf4?

Yet another confirmation of Petrosian's opinion , that Mecking has a n ind ifferent understanding of position . The conclusion of the game shows how such endings a re won .

37 gxf4

i.g4

38 'it>e3

i.f3

38. . .'it>c5 39 b4+ .

39 '>t>d4

�g2

If White's bishop were to end u p on the h 1 a8 diagonal , then , by playing e3-e4 , he would attack the b7-pawn.

40 h4

i.f3

41 b4 Before embarki ng on decisive acti o n , i n accordance with t h e wel l-known endgame principle 'do not h u rry ! ' , Wh ite strengthens his position to the maxi m u m , making all the useful moves . However, the eval u ation expressed i n this last sentence is not i n fact as obvious as it appears. Are Wh ite's pawn moves really useful? After all, a zugzwa n g situation might arise, in which a reserve pawn move (b3-b4

50 f5! This breakth ro u g h cracks Black's defences. lf 50 . . . g xf5 5 1 � xh5, and the passed h-pawn decides matters.

50 . . .

ii.xf5

51 ii.xd5

.tea

52 e4

'it>e7

53 'it>e5

g5

54 hxg5

h4

55 g6

h3

56 g 7

h2

57 g8�

h 1 'iV

94

� 58 'it'f7+ 59 'ii'f8+

A Practical Exercise

'it> d8

Black resigned . Let us thi n k about the causes of Black's failure . As regards the middlegame and the endgame - here everything is clear. We have already said enough about Mecki ng's lack of u nderstanding of a simple positional problem - on which squares he should keep his pawns . But, after a l l , t h e roots o f Black's defeat were laid back in the open ing or i m mediately on emerging from it, when he ended u p i n an inferior position . Why did this happen? Mecking played timid ly, and conceded the i nitiative to his opponent. He did not exploit the active possibil ities that were available to him in the ope n i n g : 1 3 . . . d4! , 1 5 . . . tt:lg4! (and , incidental ly, also in the endgame - 36 . . . d4 ! ) . Possibly because he did not sense the strategic danger of his position . Perhaps Mecking thought that i n a q uiet situation it would be easier to defend against an experienced grandmaster. But th is is radi­ cally incorrect. For an experienced g rand­ master there is noth ing better - g ive him a position in wh ich he is not risking anyth i n g , a n d t h e o n l y q uestion is: w i l l he win or not. Double-edged play, when there is a risk of losi ng, is far more u n pleasant for h i m . It cannot b e said that Wh ite played the open ing ideally. By not finding 1 5 'it'c3 ! , he thereby allowed 1 5 . . . tt:lg4! , sharply chang­ ing the character of the play. But otherwise Polugayevsky acted very sensibly.

Passive tactics when playing the opening are u n promisi n g . O n the contra ry, here the maxim u m accu racy and the maximum en­ ergy a re requ i red . After all , the outcome of the opening battle often determines the entire futu re pattern of the game. The main theme of our exercise was i n fact the struggle for the i n itiative in the open i n g . You have had some tra i n i ng i n the concrete solving of open ing problems. And now for the results of o u r competition. There was q u ite a compact g roup of leaders. The d ifference of one point between first place and those who shared 2nd-4th is not very sign ificant. All the leaders performed excellently, and successfully coped with most of the exercises. But nevertheless the com petition had a winner, and he was Maxim Bog uslavsky, with 38 points . * Con­ g ratulations! Second to fou rth places were shared by D ragiev, Emelin and Georg iev, who each scored 37 points. On 35 points were Makariev and Zviagi ntsev. Maka riev had a serious set­ back at the start, i n the calculation of the 1 3 . . . 'it'a5+ variation . And Vad i m did not cope, when asked to comment on a series of moves, and he did not suggest 1 5 . . . tt:l g4! - a very serious omissio n . This was his main loss , as he solved all the remaining exer­ cises well . Volodya Baklan scored 33 points. Wel l , and the one who performed least successfuly was Ya n Tepl itsky. See how dangerous it is to arrive at a competition at the last moment. After his journey he obviously did not have time to accl imatise.

* The author does not consider it necessary to g ive exact details of how the poi nts were awarded . The main th ing to note is that several pupils performed al most equally successfully (translator's note).

ctJ

95

PART I I Mark Dvoretsky

The Development of an Opening Repertoire

Tworking on you r open ing repertoire . This here a re many d ifferent approaches to

is an individual matter, and every player has his own principles. But I hope that what I have to say will p rove u sefu l .

W h i c h ope n i n g s t o i n c l u d e i n yo u r re perto i re Your choice of openings should be made

primarily in accordance with your own tastes and style of play. This is a seem­ ingly obvious principle, but nevertheless it is quite often violated . When I was teach ing in the I n stitute of Physical Culture , one of my students, a candidate master, who wasn't having much success i n tou rna ments , showed me his games. What su rprised me was that, al­ though he was a q u iet and sensible l a d , he played sharp open ings: the Sicilian , King's Indian . . . Why was this? I t a l l tu rned out to be very simple. He had been studying i n a group at the Moscow Pionee rs Palace, the trainer of which was fasci nated by the theory of fash ionable opening variations. That is, lhe player's choice of openings depended not on his own tastes, but on those of his trainer. I advised h i m to change his reper­ toire, and in particu lar to switch to 1 d4 with White. Soon the student's resu lts i m p roved , since he began playing h is own sort of chess. This was a n exa mple involving a candidate master. But it seems to me that a q u ite

s i m i l a r mistake was made by g randmaster M i khail Tal i n h i s prepa rations for the World Championship Retu rn M atch with M i khail Botv i n n i k . In the fi rst match Tal had prob­ lems with Wh ite i n the Caro-Kan n Defence, although the score i n th is opening was nevertheless i n his favou r. In the Return Match he decided to ' d u m bfound' Botv i n n i k with the t h e n rare 3 e5 syste m . From the p u rely chess point of view it was perhaps not badly p repared , and Tal did indeed have some i nteresti ng ideas there . But he ended u p with a m i n u s score i n this system: he won one game, but lost two , with several d raws . The explanation was simple: Tal had a very good understa n d i n g of and a feeling for open positions with active piece play, whereas e4-e5 leads to strateg ic, closed positions. Botvin n i k easily found his bear­ ings i n such play, whereas for Tal it was by no means h i s strongest side. He went i n advance onto Botvinn ik's territory, where the latter felt more confident. And it was no longer so i mportant how the opening had been p repa red . One ca n obta i n promising positions, but it you have a poor feeling for such play, m i stakes a re q u ite probable. And that is what happened : Tal often fa iled to exploit the adva ntages of his position . This is a banal consideration , that the open ing should be stud ied i n accordance with you r own tastes. There is another, slig htly less banal consideration - you r opening repertoi re should b e constructed i n rel ation t o you r own memory. For players

96

�

The Development of an Opening Repertoire

with a brill iant memory (such as, for exam­ ple, Victor Gavri kov or Yu ri Balashov) it makes sense to i nclude i n their opening repertoire complicated modern opening sys­ tems, where there is a g reat deal of theory, you have to know an enormous nu mber of games, and remember various subtleties. There are nu merous such variations: for example, the ultra-sharp Najdorf Variation in the Sicilian Defence. I n openings such as the Gru nfeld Defence or the King's I ndian Defence Wh ite has an enormous choice , a n d it is he w h o determines t h e opening set­ up - whereas Black has to be p repared for everything. Only if you have a good memory can you play them with Black. There is another factor, which it is usefu l for players with a good memory to exploit. They ca n permit themselves to vary their opening repertoi re and employ d ifferent openings, since they are capable of masteri n g , re­ membering and subsequently choosing those openings which are the most uncomfortable for their opponent. For players with a less good memory it is dangerous to embark on such a cou rse. I know for myself what agonizing work it is to repeat 'theory' before a game. Everything is recorded i n you r notebooks, you have already looked through it ten times, and all the same you don't remember it. It is better to aim for 'opening schemes' - logical systems with less theory, i n which what is more important is an u nderstanding of position and a knowledge of typical ideas and methods , rather than specific details or precise move orders . Openings in general c a n b e a rbitrarily divided into 'opening variations' and 'open­ ing schemes' . Of cou rse, th is is a com pa ra­ tive division , since in the theory of any opening there are both exact, specific variations, and logica l , systematic elements - it is merely a question of their correlatio n . T h u s , with a good memory you can boldly

adopt 'opening variations ', but with an indifferent one you should aim for 'open­ ing schemes '. A n exa m p l e of a n ' o pe n i n g s c h e m e ' I n my t i m e , w h e n I w a s stil l a fi rst category p layer, I became i nterested i n the q uestion of how to play with Black against the Closed Variation of the Sici l i a n Defence. We are sometimes troubled by stra nge p roblems! How to play against the normal Sicilian was someth ing that, a p parently, I knew. At that time my trainer was Alexa nder Rosha l , and it has to be said that he was a good trainer. At one tra i n i ng session he showed me a system of play agai nst the C losed Variation of the Sicilian Defence, which a p pealed to me - it seemed logica l . I saw that it was su itable not only in this open i n g , but also agai nst a n u mber of similar set-ups by Wh ite , such as the King's I ndian Attack; i . e . the g iven scheme was fai rly flexible and u n iversal . I also recom­ mend it to you - you will not reg ret it. I i m mediately began employing the new plan.

Gorod i lov - Dvoretsky Len ingrad 1 964

French Defence 1 e4

e6

2 'it'e2

c5

Black's plan is su itable for many eventuali­ ties, including against the C h igorin Va riation of the French Defence.

3 g3

tt:'lc6

4 tt:'lf3

g6

5 �g2

�g7

6 0-0

tt:'lge7

7 d3

0-0

After the game I learned that i n such situations one has to reckon with the

The Development of an Opening Repertoire

positional th reat of e4-e5. A fine game played on this theme was Petrosian­ Pachman (Bled 1 96 1 )*. Well , when you a re only just beg i n n i ng to employ a new system , there are many su btleties that you do not yet k n ow. A deep understanding is developed by practice . 8 c3

d6

9 tt:Je1 ? At that time I myself played this i n similar positions: I retreated my knig ht, advanced my pawns by f2-f4 and g 3-g4, and thought t ha t I wou l d soon g ive mate . Strictly speak­ ing, I was also interested in how to parry such attacks as Black.

9. . .

Ilb8

The advance of the b-pawn is B lack's m a i n plan. He creates cou nterplay o n t h e q u een­ side.

1 0 f4

b5

1 1 lt:Jf3

b4

97

pressing on a weak, vul nerable pawn on c3. The other bishop has also occu pied a n excellent d iagonal and is attacking t h e d 3pawn . Black controls the open b-file and he will i ntensify the pressu re on the q ueenside by . . . "ifa5. His well-devised scheme of d evelopment has al lowed him q u i ckly to d evelop an i n itiative on the queenside. And what ca n Wh ite do? See how flexibly the black knig hts a re placed . They are defending each other, and at the same time the knight on e7 is contro l l i ng the f5-sq uare . Black m u st keep a careful look-out for f4-f5. If this breakth roug h becomes a th reat (for example, after g 3-g4 ) , he forestalls it by playing . . . f7-f5 . In so doing he retains control of all the central squares, a n d his position rema i n s solid and flexible.

14 'ii'c 2? My opponent wants to d evelop his q ueen's knight, but he ru ns i nto a tactical stroke typical of this set-up.

1 2 �e3

bxc3

14 . . .

lt:Jb4!

1 3 bxc3

�a6

1 5 cxb4

�xa 1

There is no need to show any more of the game - Black is a sou n d exchange to the good and he won easily. My fi rst experience of employing this set-up proved successfu l . Things went equally wel l i n this next game.

Tu rovs ky - Dvoretsky Moscow 1 964

French Defence 1 e4 think that strateg ically the position is almost won . Not long ago the g7 -bishop was running up against a secu rely d efended white pawn , but after . . . b7-b5-b4xc3 it is I

•

e6

2 'ii'e 2

c5

3 tt'lf3

tt:Jc6

4 g3

g6

5 �g2

�g7

This game, as wel l as some others which will be mentioned , will be found in the appendix to the lecture.

98

� 6 0-0

The Development of an Opening Repertoire

d6

You will see that on this occasion I a l ready understa nd the position better and I do not allow e4-e5. Cicero once said ' Everyone can make a mistake, but to persist i n you r delusions i s merely senseless . '

7 tt:Jc3

lt:Jge 7

8 d3

0-0

Now the position is similar to a Closed Variation of the Sicilian Defence (the knight has come out to c3) . But it is more comfortable for Black than in the Closed Variation, since there Wh ite does not usually block the f-pawn with his knight, and it is also not clear why he has played 1We2 - he does not need th is move .

9 ..ll. e 3

bishop, which becomes 'bad ' . Both strate· g ies a re possible, and each time a concrete decision must be made as to which of them to follow.

10 . . .

�e8? !

W h y is t h i s move d u bious? A t that time I did not yet real ise that Black has to reckon not only with 1 1 ..ll. h6, but also 1 1 d4. After the open ing of the d-file the d6-pawn becomes vul nerable. Usually in such cases d3-d4 should be prevented by . . . ltJd4.

1 1 lt:Jd1 ? Wh ite wants to occu py the centre by c2-c3 and d 3-d4 . This plan is too slow - it would make sense only if Black were u nable to 'latch on' to the c3-pawn with his b-pawn.

.l:i.b8

Black's plan is the same - to advance his b­ pawn , lengthening the diagonal of his bishop on g7. 10 iVd2

11 . . .

bS

1 2 c3

b4

1 3 d4

bxc3

1 4 bxc3

..ll. a 6

1 5 l::!. e 1

cxd4

1 6 cxd4

1Wa5!

Black's pieces are far more harmon iously placed than the opponent's. His bishops, as i n the p revious game, a re raking the entire board , h i s rooks control the open b- and C· files, and the d4-pawn is weak. I n the event of 1 7 iVxa5 tt:Jxa5 the black knight goes to c4.

1 7 lt:Jc3

�ec8

1 8 �ec1

it'a3

Th reatening the i nvasion of the rook at b2.

1 9 .l::!. a b1 Here there is an interesting problem, a n important o n e for t h e entire variation . Wh ite probably wants to play ..ll. h 6. Black can preserve his strong bishop from exchange by 1 0 .. J:!. e8, and if 11 ..ll. h 6, then 1 1 . . ..\l. h B . B u t another reaction is also possible: allow the exchange, rearra nge the pawns on dark squares ( . . . e6-e5, . . . f7-f6 ), and play for the restriction of the opponent's lig ht-sq uare .

tt:Jas

When you know a p l a n , you don't even h ave to th i n k long , as all the moves are natural. The play proceeds of its own accord : Black could have acted l i ke this even in a blitz game.

20 .l:i.xb8

l::!. x b8

21 .l::!. b 1

l:i. c8!

22 lt:Jd 1 ? lt:J 22 b5 really was better.

ltJ

The Development of an Opening Repertoire

22 . . . The gathering of the harvest beg i n s . H ow­ ever, 22 . . . .t d3 was even stronger.

23 �b4

'ifxa2

24 ltJc3 25 .tc1 ? !

'11i'a 3

26 l:lxb4

ltJb6!

27 .td2

ltJc6

'ii'x b4

White resigned, since he loses a second pawn. I n demonstrating these games, I h ave not delved into variations. Fi rstly, because my opponents were only fi rst category players . N uances a n d specifi c deta i l s a re better studied in games by stronger players . And secondly, because we are exa m i n i ng not an 'opening variation ' , but a n 'ope n i ng scheme' . In such cases what is more important to you is not a detai led variational a nalysis, but the pattern of the play, the plans of the two sides , and typical methods.

Such comparatively easy play succeeded not only in j u n ior events . I also successfully employed th is set-up later, against strong opponents. Bronstein - Dvoretsky Moscow 1 976

99

follows 7 tt:l ce2 ! followed by c2-c3 and d3d4. The variation 7 . . . ttJxe2 8 ttJ xe2 .t xb2 9 l:t b 1 is advantageous to Wh ite, since 9 . . . 1i'a5+? 1 0 .t d2 1i'xa2 1 1 :Xb2! 'ii'x b2 1 2 .t c3 i s bad for Black.

tt:lge7

7 f4 8 lDf3

Which do you th i n k is more accu rate , 8 . . . 0-0 or 8 . . . ltJ d4 ? We have a l ready mentioned that Black has to reckon with d 3-d4 . It is not always dangero u s , but it seemed to me that it was better to prevent it. Now I a m not so s u re : after 8 . . . 0-0 9 d4 there is 9 . . . 'ii'b6 or 9 . . . cxd4 1 0 tt:l xd4 'ii' b6. The same reaction follows after 9 0-0 l:I b8 1 0 d4. On the other hand, when the knig ht is on c6 , the pawn sacrifice 1 0 e5!? (which is not bad with the knight o n d4) loses its strength .

8 . . .

ltJd4

9 0-0

0-0

One of the basic positions of the Closed Sicilian has been reached . Wh ite has several conti n u ations: 1 0 'ili'd 2 , 1 0 .tf2 , 1 0 .l:l b 1 and the afore-mentioned pawn sacrifice 1 0 e5. It is precisely here that general considerations are i nsufficient - one can not get by without a fam i l i a rity with the specific theory of the variati o n . But if you desire you ca n ca rry out this work you rself, whereas here we have other objectives .

1 0 g4

Sicilian Defence 1 e4

c5

2 tt:lc3

ltJc6

3 g3

g6

4 .tg2

.i.g7

5 d3

e6

d6 6 .te3 But why not 6 . . . ltJ d4 ? This is what Arnold Denker played against Vasily Smyslov i n the 1 946 USSR-USA match . Remember: the d4-square should be occupied by the knight only after the white knight has appeared on f3 or e2. If 6 . . . ltJ d4?! there

What should Black play? It is very danger­ ous to al low Wh ite to break up the black king's pawn screen with the pawn sacrifice f4-f5 . As we have a l ready said , the standard reaction to this th reat is the cou nter . . . f7-f5.

10 . . .

f5 ! (see diagram)

But with what should Black recaptu re on f5? For such positions there is a ' rule of th u m b ' : captu re with the opposite pawn t o the opponent. If g4xf5 reply . . . e6xf5 , and if e4xf5 , then . . . g6xf5 . I don't know how to

1 00

�

The Development of an Opening Repertoire

b2-pawn, he wants to play c2-c4 . I guessed that this was his plan and I prepared an antidote. 17...

lia eB

1 7 . . . �ad8 , preparing . . . d6-d5, was also good. 18�c1

'it'b7

Black's position is preferable: his pieces are exerting unpleasant pressure on the oppo­ nent's centre.

- position after

1 O f5

-

. . .

explain this logically, but my experience of playing this set-up suggests that usually that is the way things are. 11 gxf 5

exf 5

1 2 'ii'd 2 The standard . . Jib8 and . . . b7-b5-b4 is now rather slow. Black must complete his development and actively fight for the centre. Where is his light-square bishop best placed? Often in such cases it is developed at e6 (after first playing . . . '>t>h8 , so as not to have to fear lbg5). But I decided to place my bishop on c6, in order to oppose the white bishop on the long diagonal. 1 2...

�d7

1 3 'ii'f 2 White finally dislodges the knight from d4 . 1 3...

lLlxf 3+

14 �xf 3

�c6

I want to prepare (by playing . . . b7-b6, and perhaps . . . it'd?) the advance . . . d6-d5, after which Black wins the battle for the centre. 1 5 'ir'g 2

b6

16liad1

'it'c7

Black intends . . . 'ir'b7, . . . liad8 and . . . d6-d5. 17lLl e2 Of course, the b2-pawn cannot be taken White replies c2-c3 and then begins trap­ ping the bishop . Now, after defending the

19 c4?! Positionally this move is justified: it prevents . . . d6-d5 and prepares the manoeuvre of the knight via c3 to d5. However, in carrying out his plan, Bronstein underestimated the flank diversion I had prepared. Evidently he should have chosen 1 9 lbg3, when Black would have replied 1 9 . . . d5 20 e5 d4 with somewhat the better chances. 19... 'ii'a 6! I have a special term for such moves: 'strategic double attack' . The capture on a2 is threatened, but in addition Black wants to exchange pawns in the centre. If the opponent recaptures on e4 with a piece, my knight will obtain the colossal f5-point. And if he recaptures with the pawn, I will pick up the c4-pawn. 20 lbc 3

fx e4

The Development of an Opening Repertoire

21 dx e4

21...

.1i.d4+

22'iti>h1

'ii'xc4 'iti>h8

24 l2J e2

.1i.g7

25 b 3

'ifa6

26 l2Jg 3

'iib7

Black has again set up a battery along the long diagonal, which he had a few moves ago. After the natural move 27 l:txd6 he had prepared the counter 27 . . . lhf5! (28 :Xc6 tt:lh4!).

27l:tg1? White should have supported his bishop: 27.l:!.f1 . Now I am able to force events to my advantage.

27... 28 l2Jxf 5

1 01

S ome remarks o n t h e tech n i q u e of

2 1 ttJxe4 was nevertheless better. In my opponent's place I would have begrudged giving up the pawn.

23 .l:.f e1?!

ttJ

l2Jf 5! l:txf 5

28 . gxf5 29 .1i.b2 l:tf7 was also good. . .

29l:txd6

.1i.x e4

30 .1i.x e4

'ii'x e4

31 'ii'x e4

l:tx e4

32l:d8+

l:.f8

33 l:r.xfB+

.1i.xf8

34f 5

.1i.g7!

35fxg6

.1i.d4

36g7+

'iti>g8

37 .l:tf1

'iti>xg7

Black has a decisive advantage.

We will return again to this type of position, but for the moment we will continue our discussion about the development of an opening repertoire.

work i n g o n t h e o pe n i n g

The most unsuccessful method, which I know that many of you employ, is to copy opening information into a notebook. One can't think of anything worse! You fill up the pages of the notebook with games and variations, then new games, fresh ideas and additional variations appear, and it is not clear where you should put them . Some pages turn out to be unsuccessful, and have to be altered or even discarded, and in a notebook you can't insert new pages. Gradually you develop almost an aversion to your opening notebook, as you sense how out of date it is, and how awkward it is to record novelties in it. All information,

and opening information in part icular, should be recorded in a card index . These can be either small cards, or large sheets. If necessary, you write a new sheet, add it to any other, or discard a sheet that is no longer appropri­ ate; i.e. you can do what you want with them. Yusupov, Dolmatov - all the players who have worked with me, have a set of folders with opening analyses, devoted to different openings and even individual varia­ tions. And another piece of advice: leave large margins - there will almost certainly be things to add. Where you feel that some­ thing new will appear, leave space. Write only on one side of a sheet. Of course, today such a method of working has become out of date. It is clearly far more convenient to maintain your card index on a computer. There it is always as though brand-new: you can easily adjust, amplify or correct things, and there is a system of opening keys, with the help of which everything can be neatly arranged. The computer processing of opening information is a topic demanding a special discussion; here we will not dwell on this question.

1 02

�

The Development of an Opening Repertoire

'Yo ur own theory'

Let us suppose that you have successfully chosen your openings, and your index is maintained irreproachably: in it are both the latest games, and extracts from opening articles. You understand everything per­ fectly well and you remember it. 'An opening advantage is guaranteed', you think and . . . you will be wrong. Because to achieve great successes it is not enough to know 'official' theory. It is essential (as Botvinnik com­ mented in this time) to have 'your own theory of the openings'. It is very import ant to introduce into your repert oire some opening schemes or variations where your views diff er from the theoretical ones, even if only very slightly. This may be a novelty, as a result of which an entire variation is reassessed or a scheme, which was considered bad, is rehabilitated. Or it may be the non-tradi­ tional assessment of a known position. Let us suppose that the position is considered not too favourable, but it appeals to you. You work out a plan and reckon that this position, which has a dubious reputation, is one that you can go in for. In general, you should have something special of your own, your own systems, for which you have a feeling and have ana­ lysed. For a player who knows only that which has already been played, it is hard to count on success. Against an experienced opponent he will never gain an advantage: the latter also knows everything. But thanks to 'your own theory' you may be able to outplay the opponent in the opening, put him in an uncomfortable position, or lure him onto ground where he will not understand around what the struggle revolves. How to expand y o u r o p e n i n g re perto i re

It rarely happens that a player scratches his head, and then decides: 'Shouldn't I be

studying, say, the Nimzo-lndian Defence?' He picks up the Encyclopaedia and studies it. This happens, but not often. Usually the introduction into the repertoire of a new scheme or variation is preceded by some impulse. For many young players this is the help of a trainer. He says: 'I have some good analysis of a certain opening system; I will show it to you and you will beat everyone.' This often proves useful. Only, don't be­ come accustomed to working in this way. After all, sooner or later the trainer's supply of ideas will dry up, and you yourself will reach a level where no trainer can help you any more. Then you yourself have to devise things. But for the time being a trainer's help is indeed your 'magic wand' . After Valery Chekhov won the qualifying tournament for the 1 975 Junior World Championship, it transpired that those open­ ings which he had employed earlier were not suitable for the world championship. There was no active opening for Black, and with White he played all kinds of rubbish, although before coming to me he had studied in the Pioneers Palace with an openings trainer. Realising where our weak points were, and what problems needed to be solved, I invited grandmaster Evgeny Sveshnikov to a train­ ing session. The range of openings which he was able to show is well known - he has been playing them all his life. For Black - the Chelyabinsk Variation, and for White - the Sicilian with 2 c3. This was just what we needed: to obtain a system with White in the Sicilian and an active system with Black against 1 e4. At that time the theory of the Chelyabinsk Variation was not yet devel­ oped, and it was constantly employed only by Sveshnikov and Gennady Timoshchenko. Sveshnikov helped us to master these two openings and at the world championship Chekhov successfully employed them. They became part of his repertoire.

The Development of an Opening Repertoire

Moreover, I also used the notes made at the training session for myself, and I expanded my repertoire. Later I showed the Chelya­ binsk Variation to Artur Yusupov and Sergey Dolmatov, and for a time they also played it. So that a few hours of study with Sveshnikov helped for a time to form the opening repertoire of a whole group of players. Another example was Dolmatov's prepara­ tion for the 1 978 World Junior Champion­ ship. At that time a similar situation arose: Sergey did not have a serious variation with White against the Sicilian. I myself could not help him , since I did not play anything worthwhile, and at the first convenient opportunity I used to play � b5. To a training session before the world championship we invited grandmaster Vladimir Tukmakov, an expert on the Sicilian Defence for Black. For such a specialist, showing the main ideas for White was not such a difficult task. The consultation with Tukmakov proved excep­ tionally useful for Dolmatov. At the world championship he played normal lines against the Sicilian, successfully combating the Scheveningen Variation, and since then throughout his career he has successfully played the main variations of the Sicilian Defence with White Information often reaches us by accident. Once, when I was still a university student, I called in at a lecture by grandmaster Yuri Razuvaev about the Exchange Variation of the Ruy Lopez. He showed some recent games by Robert Fischer, and explained the main ideas. I so liked this 90-minute lecture that, after also looking at the Exchange Variation myself, I later won several good games with it. Thus a suggestion by a trainer or expert may prove very useful, and provide the impetus for including an opening in your repertoire. And it is understandable why: when you begin studying an opening, you are faced with an enormous body of material, a large

�

1 03

number of games, and several pages of minute text in the Encyclopaedia of Chess Openings. You don't know what you espe­ cially need, which systems are the main ones, and which are secondary. You see the variations, but you don't understand what stands behind them. But if a trainer explains the main ideas and assists you in making a choice, this, of course, is a serious help. Generally speaking, you can also expect help not only from a trainer. It is very productive to work as a pair, with one of your friends. Each of you has his ideas and his opening researches, and it is useful to exchange them and analyse them together. The drawbacks in such cases are always less than the benefits. Of course, you will no longer be able to play an opening variation against a colleague, which whom you have studied it together; and if he is the first to employ a novelty, you will no longer be able to make use of its surprise effect - these are the minuses. But at the same time, firstly, you will obtain information which previously you did not have; secondly, your opening variations will be better developed . In the end you are competing not with your colleague, but with the remaining players in the world. You will be disarmed against your friend, but better armed against all remain­ ing players, and this is more important. Here the many years' work of Yusupov and Dolmatov is a very notable example. Many variations were developed in their joint analyses. Yusupov borrowed some ideas from Dolmatov, and vice versa; as a result they both improved their opening reper­ toires. Thus a second way of expanding your opening repertoire is to exchange informa­ tion with a colleague. A third source is the analysis of games. It is in this way that strong players find for

1 04

�

The Development of an Opening Repertoire

themselves the most important ideas. At the first session of our school, Yusupov showed a game of his with Anatoly Karpov, and described how the Open Variation of the Ruy Lopez appeared in his repertoire. I should remind you: he analysed the Karpov­ Savon game ( 1 971 ) and found an improve­ ment for Black. The novelty provided the impetus for a study of the Open Variation in general.

N ezhm etdin ov -N.N. Kazan 195 1 Grilnfeld Defence

1 d4 2 c4 3 lbc 3 4cxd 5 5 e4 6 bxc 3 7�b 5+

lLlf6 g6 d5 ltJxd 5 ltJxc 3 c5 ltJc6?!

There is another way of improving: choose for yourself a chess leader, whose ideas and style of play you like. The opening repertoire of this player can be copied, by studying the systems employed by your idol.

This move was recommended in the first edition of the Encyclopaedia (in the second edition the mistake was corrected, and the main move for Black became 7 . . . �d7).

Wo rking with l i te ratu re

Apart from this stroke, 8 . . . a6 must also be considered. Now a forcing exchange of blows begins.

Chess literature is a very important source of new information. You should be con­ stantly looking through magazines, chess books and lnformator. You never know where you will run across an idea which will later come in useful. Even in old publica­ tions. Here is an example: many years ago I studied a games collection of Rashid Nezhmetdinov. This was a remarkable book: very vivid games, and wonderful combina­ tive play. There I noticed a combination which Nezhmetdinov had carried out in a simultaneous display against an amateur. It appealed to me, and I included it in my card index of exercises. A few years passed. Something interested me in the GrOnfeld Defence, and I opened the Encyclopaedia of Chess Openings. And suddenly I discovered that the opening variation from Nezhmetdinov's game, in which he found a forced win with White, had a diametrically opposite evaluation in the Encyclopaedia in favour of Black. I immediately realised that this was a trap, in which it would be possible to catch some zealous reader of opening books. -

8d 5

'ii'a 5

9 'ii'a 4! What should Black do? Nezhmetdinov has defended his bishop, and after the ex­ change on a4 Black loses a piece. 9...

�xc 3+

10 'it> e2

�d7

1 0 . . .'iVxa 1 1 1 dxc6 is completely bad for Black. In the event of 1 0 . . . �g7 1 1 dxc6 0-0 1 2 .l::tb 1 , in my view, he gains insufficient com­ pensation for the sacrificed (or lost? ) piece. 11 dxc6

bxc6

1 2 �xc6

l::i.d8

The Development of an Opening Repertoire

In the Encyclopaedia this position is evalu­ ated in favour of Black. The white rook is hanging. Do you see what happens after the natural move 1 3 l::tb 1 ? Quite correct: 1 3 . .'i'd3+!! 1 4 �xd3 �xc6+ , and Black wins a pawn. This happened in a game lsakov-Nikitin ( 1 947). .

You suggest 13 lDf3. An interesting move. It is not considered in the Encyclopaedia, so let's have a look at it. What happens after the capture of the rook? 1 3 . . . 'ii'xa 1 1 4 .l:!.d 1 , or more accurately 1 4 �xd7+ .l:.xd7 1 5 l:td 1 . Excellent - White develops and prevents the capture of the rook. What can Black play? Probably again 1 3 . . . 'ifd3+!. But now the capture of the queen is not obligatory. After 1 4 �e1 'ii'c3+ White can repeat moves -already not bad - but he can also play for a win: 1 5 �d2 'ifxa 1 + 1 6 �e2 . It is probable that the rook cannot be taken, since there follows 1 7 ltJeS . This means 1 6 . . . 'ifb2 , guarding the e5-square. And White, we assume, will play 1 7 l:ld 1 or 1 7 l:tc1 . Does he have sufficient attack for the sacrificed exchange? It is interesting to give some thought to this position. But before delving into a complicated analy­ sis, you should always ask yourself: 'Did I not miss something earlier, at the very start of my calculations?' It is stupid to study lengthy variations, which in fact are not obligatory. And thus here after 1 3 . . . 'ii'd3+ 1 4 �e 1 Black has the excellent reply 1 4 . . . �g7!, which simply refutes 1 3 tiJf3. I suggest that you try to find for yourself the solution to the problem: h ow did N ezhm et­ dinov win th e gam e? I showed this opening variation to Yusupov and Dolmatov. It was of more interest to Yusupov, since he constantly plays 1 d4, whereas Dolmatov plays it only occasion­ ally. Thus we have an opening trap. Can it be used? First two questions must be answered.

The first: what if Black chooses a different

ltJ

1 05

move order, and plays not 6 . . . c5, but 6 . . . �g7 ? Then one either has to prepare for the main variations of the Grunfeld Defence, or find a way of avoiding them. We began analysing 7 �a3. Theory considers this move, and later Dolmatov won an excellent game against Vladimir Bagirov. Neverthe­ less, there is something 'non-Grunfeld' about the bishop on a3. In the end we realised that in this way it is hard to count on an opening advantage. This means that one cannot get by without a mastery of the 'normal' Grunfeld. The second question: what happens if after 6 . . . c5 7 i.. b5+ Black avoids 7 . . . tDc6 ? According to theory, 7 . . . tiJd7 promises White the better chances after 8 lDf3. After all, what is one of the problems for White in this variation? Black attacks the central d4-pawn with his c5-pawn, his bishop and his knight on c6. White defends it with his bishop and knight. In principle it is advantageous for him to develop his knight on f3, but he has to reckon with the pin . . . �g4. Therefore White usually places his knight on e2 , or if on f3, then after first making the not very neces­ sary move .l:tb 1 , in order to remove the rook from the a1 -h8 diagonal. But after 7 �b5+ tiJd7 8 lDf3 White does not have to worry about his centre - the opponent has neither the pin . . . �g4, nor the development of his knight on c6. The most natural move is 7 . . . �d7. After 8 �xd7+ 'ii'xd7 9 lDf3 White has achieved something: he does not have to fear the pin on his knight and he succeeds in securely defending his centre. But does he have anything real? On the basis of some old game, theory gave equality. We began studying the resulting positions and found some ideas for White, and it even appeared that he could hope for an advantage. Then we came across on article in a foreign magazine, where it was shown that with accurate play Black could nevertheless

1 06

�

The Development of an Opening Repertoire

equalise. We were unable to refute this conclusion, and so in the end we lost interest in the idea of the trap. Even so, at the 1 98 1 Student Team Champi­ onship in Graz, Yusupov managed to lure one opponent into the prepared variation, and he won against Robert Morenz exactly as in the Nezhmetdinov game. So that the study of this old book brought some fruit, even if only slight. More often, of course, novelties are to be found in more recent games and articles. I will tell the story of one successful discov­ ery. For the 1 984 World Junior Championship Alyosha Dreev and I prepared in Estonia at a joint training session with another partici­ pant in the tournament, Lembit 011. Interna­ tional master Iva Nei, Oil's trainer, brought to the session a whole suitcase-full of chess literature, including many foreign maga­ zines. I had not seen them before and in my spare time I began looking through them. In a Bulgarian magazine I came across an article devoted to one of the lines of the Carlsbad Variation in the Queen's Gambit. (It is probably more correct to call this the Exchange Variation, as prescribed in official theory, but the resulting pawn structure is usually called the 'Carlsbad', and I am accustomed to using this name for the entire system). 1 d4

d5

2 c4

e6

3 lbc3

lbf6

4 cxd5

exd5

5 ..ig5

..ie7

6 e3

0-0

7 ..id3

l:te8

8 lbf3

lbb d7

9 0-0

c6

1 0 'ir'c2

lbt8

1 1 l::ta e1

One of the possible continuations (nowa­ days 1 1 h3 is more popular), and a very old one. It was first played by Frank Marshall against Akiba Rubinstein at the Moscow International Tournament of 1 925. Theory considers that this variation leads to equality. lbe4 1 1 ... 1 2 ..ixe7

'ii'x e7

1 3 ..ixe4

dxe4

1 4 lbd2

f5

1 5 f3

exf3

1 6 lbxf3

..ie6

1 7 e4

fxe4

1 8 :xe4 This move is stronger than 1 8 ltJxe4 ..if5.

1 8 ...

.Uad8

1 9 l::tfe1 In the afore-mentioned game Marshall won quickly by 1 9 .l:!.e5 h6 20 lbe4 ir'b4?! 2 1 a3! 'ii'c4 22 'iff2 ..if7? (22 . . . b6!) 23 b3! 'ir'xb3 24 ltJfd2 ir'a2 25 lbc3. However, Black could have defended better: 20 . . . ir'c7!? (with the threat of 2 1 . . . l::txd4) or 20 . . . J::f.d 5. 1 9...

h6

The game Tai-Vaganian (Moscow 1 975) subsequently developed as follows: 20 .l:!.e5 �f7 21 'it'e4 .l:td6? 22 'ife3 ltJd7 23 l::t a5, and Black has no time to defend his a7-pawn in view of the terrible threat of 24 ltJeS ltJxe5

The Development of an Opening Repertoire

25 .Uxe5. But this game too does not demonstrate any advantage for White - by playing for simplification by 21 . . . ..id7! Black could have hoped to equalise.

The author of the article suggested an interesting set-up for White: 20 .l:!.1 e3 ! ?

ikf7

21 'it'e2 White concentrates his forces closer to the kingside. If necessary, he can avoid ex­ changes on the e-file by occupying the e5point with his knight and can then switch his rooks to the neighbouring files on the kingside. This idea seemed promising to me, from the purely chess and from the practical point of view. After all, it is not possible to keep track of every periodical; an article in a Bulgarian magazine would be known in Bulgaria, but in other countries it might not be noticed. This meant we would have ideas with which our opponents were unfamiliar. However, this variation was of little use to Dreev, since at that time he only played 1 e4. I simply transcribed the analysis, realis­ ing that sooner or later it would come in useful. When Yusupov and I were preparing for the 1 985 Candidates Tournament I suggested to Artur that we should investigate this system. I showed him the variations given in the article, and it appealed to him. First we analysed the resulting positions, and then we played a training match with a time control of 1 5 minutes each per game. Incidentally, to consolidate opening informa­ tion which is being studied, I strongly recommend you to play games with a shortened time control (of course, after preparatory analysis and with additional analysis after the games). They do not take much time, but in the opening, even if a player has done preparatory work on it, new problems are usually discovered. In the opinion of Yusupov, you should play in turn

ctJ

1 07

for both sides, which aids a more objective view of the position. In our series of games a mass of novelties was generated, and we gained a far better feeling for the opening than before the training. Yusupov included the Carlsbad Variation for White in his opening repertoire. At the Candidates Tournament he won a fighting game against Boris Spassky and crushed Jesus Nogueiras, and later too he successfully employed this system. It stands to reason that Yusupov's suc­ cesses were not a result of some particular strength of the given variation. On the contrary, we came to the conclusion (which was not hard to predict beforehand) that, as in any sound opening, with accurate play Black can equalise. Simply we were some­ what ahead of our opponents, we under­ stood the positions more deeply, and we had in reserve some ideas which were unfamiliar to them. Since Artur plays the Queen's Gambit not only with White, our analyses also came in useful for reinforcing his defence with Black, In particular, in the 8th game of his Candi­ dates Match against Jan Timman, when the opponent, playing White, desperately needed to win, Yusupov employed the novelty 1 8 h 6 ! ? (instead 1 8 . . . .Uad8 ) prepared at our training session, and easily equalised. ...

With White, as with B l a c k

We have spoken briefly about the basic principles of building up your opening repertoire. Now I will dwell in more detail on one specific method. It is appropriate for those who in the opening do not aim with White for the maximum, for a definite advantage, but seek their 'own game', their type of position. Sometimes with White is makes sense to employ a system which you like for Black; i.e. with White you play like Black with an extra tempo.

1 08

�

The Development of an Opening Repertoire

Soon after I included in my repertoire a successful method of play against the Closed Variation of the Sicilian Defence, with which I won the majority of my games, the idea naturally occurred to me of employ­ ing the same set-up with White. What was needed for this? Obviously to begin playing the English Opening. Of course, in this Black has many different systems, for which one has to be prepared. But if he aims for a King's Indian set-up, there is a chance of catching the opponent in the favourite scheme. Grandmaster Razuvaev once told me about a specific method, which he employs in his (very productive) work on opening theory. He looks for the most important, representa­ tive examples of the opening in question. A game in which both players (or at least one of them) acts logically and consistently; where valuable ideas and typical methods occur. Such games help one to understand the opening set-up more deeply and to remember it more easily. The game which I will now show you may serve as a key example for the system which we are currently discussing. Dvoretsky - Timoshchenko

7 0-0

d6

8 d3

i.e6

How to reply, do you remember? 9 t'Lld5! Of course, . . . d6-d5 must be prevented . 9 . . .

'ii'd 7

1 0 l:tb1 I can now explain why the knight is worse placed at e7 than at f6 or h6. Here Black cannot play 1 O . . . i.h3, because after the exchange I will capture the c7-pawn. With the knight on f6 or h6, 1 O . . . i.h3 is possible, since after the exchange on h3 White cannot capture the c7-pawn because of . . . t'Llg4 with mate. A little tactical detail, which is very significant. If Black were free to exchange the light-square bishops, he would not stand badly. Incidentally, the position with the knight on f6 is examined in my article 'The superfluous piece' in the book Secrets of Chess Training. 10 . . .

t'Lld8

The game Dvoretsky-Veselovsky (Mos­ cow 1 967) went 1 O . . . a5 1 1 a3 l::ta e8? (the queenside should not have been aban­ doned to its fate) 1 2 b4 axb4 1 3 axb4 lLldB 1 4 b5 c6 1 5 bxc6 bxc6 1 6 l'Llxe7+ �xe7.

USSR Team Championship, Moscow 1 966 English Opening

1 c4

e5

2 l'Llc3

t'Llc6

3 g3

g6

4 i.g2

i.g7

Black is playing the Closed Variation of the Sicilian with reversed colours. 5 e3

t'Llge7

An interesting moment. I consider this move to be rather weak: in this variation the knight is best placed at f6 or even h6. The basis of this evaluation is purely tactical and not at all obvious. You will see it within a few moves. 6 t'Llge2

0-0

A little exercise for you to solve yourself: what is White's most exact course of action?

The Development of an Opening Repertoire

1 1 b4

tt::l x d5

If 1 1 . . . c6 there follows 1 2 tt::lxe?+ 'ir'xe? 1 3 b5. After the exchange of pawns on c6 the bishop comes out to a3, the queen to a4, and so on - White has an easy, pleasant game. The move made by Black leads to a closed type of game.

ii.h3 1 2 cxd5 What should White play in such cases? 1 3 e4! The standard plan: after the light-square bishops have been exchanged, the pawns are moved onto light squares. 13 . . . 14 'iitx g2 1 5 f3

ii.xg2 f5

I thi n k that White's position is preferable. He has more space, and the black bishop is running up against its own pawns. If Black undermines the centre with . . . c7-c6, there follows tt::lc3. At some point subsequently White exchanges on c6, and in the event of ... b7xc6 he advances b4-b5, obtaining the central d5-point for his knight. Black in turn will acquire the possibility of placing his knight on d4 , but these points are not equivalent. The knight at d4 is attacked by the white bishop, whereas Black is unable to exchange his bishop for the white knight. Here it is, the advantage of a 'good' bishop over a 'bad' one.

15 . . .

tt::lf7

l2J

1 09

My next move is one that, to be honest, I am proud of. At that time I had only just become a master, but the move, in my opinion, is one worthy of a grandmaster. It was found in a purely logical way, and we will now follow this logic. What does Black want? It is always useful to ask yourself this question. Most probably, to exchange his bad bishop: 1 6 . . . ii.h6. Should I agree to the exchange? If the bishop is moved to b2 or g 1 , the black bishop will take control of the c 1 -square and it will be hard for me to put into effect my natural plan - pressure on the backward c?­ pawn with the heavy pieces along the c-file. In addition, the black knight will obtain the excellent square g5 in the vicinity of my king. It is very dangerous to 'stick to your principles' and reason routinely: since the opponent's bishop is 'bad', it means that it should not be exchanged. But for any rule there are numerous exceptions. After some thought I decided: there was nothing to be done, I would have to exchange. But on which square? I can allow him to capture on c 1 , when I occupy the c-file with gain of tempo. I can capture on h6 - there the black knight will be badly placed. The second option seemed more reliable. But how should I arrange my pieces on the c-file? The most natural set-up is the queen on c2 , the rook from f1 to c1 , and the b1 -rook can go to c3. This means we have a choice between 1 6 'ii'c2 and 1 6 .l::!.b3. The rook move is apparently more accurate. After the exchange on h6, the fork 'it'c 1 may be a possibility - with a simultaneous attack on c? and h6. This last consideration fully justified itself (true, in a slightly different form) in a game which I played two years later against national master Anatoly Kremenetsky (35th USSR Championship, Kharkov 1 968 ). After 1 6 l:tb3 he did not reply 1 6 . . . i.h6, but 1 6 . . . h6. I acted in accordance with my plan:

110

�

The Development of an Opening Repertoire

1 7 l:tc3, and after 1 7 . . . lt:Jg5? 1 8 St.xg5 hxg5 1 9 "it'c1 ! two black pawns- c7 and g5 - were under attack. 1 6 .l:r.b3 ! !

it.h6

1 7 it.xh6

lt:Jxh6

1 8 .l:r.c3

.l:r.f7

What should I play? The natural plan is 'ii'c2 and l:!.c 1 . Black defends his pawn with . . . l:!.c8 . What to do next? Obviously, advance the queenside pawns. However, in this case a well-known positional principle applies: before attack­ ing, you should create a target to ' latch on' to. 1 9 'iic 1 !

'it>g7

20 �a3

a6

Forced - Black has to free his rook. But now the a6-pawn makes it easier for White to open lines on the queenside. He plays a2a4, b4-b5 and after . . . a6xb5 he recaptures on b5 with his queen. 21 l:!.fc1

l:!.c8

22 'ii b 3 The more aggressive move 22 "it'a5 came into consideration.

fence, then . . . lt:Je8 . There is also another tempting idea: to play 22 . . . "it'b5 and then 23 . . . "it'b6, from where the queen can embarrass White with the threat of an invasion on e3 or f2 . With correct defence I think that Black would have had an inferior but tenable position. However, my opponent chose a faulty plan. 22 . . .

g5?

Timoshchenko mounts an attack on the king, or more correctly, he thinks that he does. The standard reaction to flank activity by the opponent is a counterblow in the centre. But if 23 d4? there follows 23 . . . fxe4, then ... lt:Jg4 or ... 'ii'g4 and . . . l:!.cf8 - and all the black pieces join the attack. This means that for the moment the routine recommen­ dation is inapplicable. What does the opponent want? Does he intend to play 23 . . . g4 ? But then White has the excellent reply 24 f4!. On g4 the pawn takes away this square from the queen and the knight. Now let us consider 23 . . .f4. The reply 24 g4? is not possible in view of 24 . . . lt:Jxg4. What is the drawback to Black's move? It removes the attack on the e4-pawn and I can finally strike at the enemy centre: 24 d4!, simulta­ neously including my rook and queen in the defence of the kingside. It follows that at the moment I have nothing to fear - replies are prepared against each of the opponent's attacking moves. This means that I can coolly make a useful move on the queen side. A good example of clear, logical reasoning, based on 'prophylactic thinking' . 23 a4!

f4

24 d4! What should Black do now? He has a terrible knight on h6. His natural strategy is to switch it to better squares: . . . lt:Jg8-f6, and if the c7-pawn requires more reliable de-

All according to the rules: against a flank attack - a counterblow in the centre, and a timely one. Black has two attacking possi­ bilities: he can play . . . g5-g4 either immedi­ ately, or first exchange on g3. But after

ctJ

The Development of an Opening Repertoire

24 . .fxg3 25 hxg3 g4 there is the very strong reply 26 f4!. .

24. ..

g4

25 dxe5

dxe5

How should White proceed further? I saw that there was an excellent square at e6 for the white knight. 26 gxf4!

exf4

27 t2Jd4

'it>h8

28 t2Je6

c6

A methodical move. Try more often to pay attention to such 'trifles' . I force the oppo­ nent to place his king on g8 , after which I don't have to fear any counterplay on the g­ file. Probably I would also have won without this, but it is always useful first to restrict the opponent's possibilities.

29 . . .

very seriously, so as not to allow him any counter-chances. The knight can be taken, of course, but why allow even the slightest sharpening of the play? 33 .l:tg 1 +

t2Jg7

34 bxc6

bxc6

35 l:tc2 ! The game is won with quiet moves. Black cannot take on d5, and 36 .l:!.cg2 is threat­ ened. 35 ...

29 'i¥b2 !

'it>g8

30 b5!

111

l:l.e8

The last hope: 36 .Ucg2? .Uxe6. 36 .Uxg7+ 36 .Uxc6 was also strong. 36...

llxg7

37 'i!Vxg7+

'iVxg7

38 t2Jxg7

cxd5?

Of course, 38 .. .'it>xg7 was also completely hopeless. 39 t2Jxe8 Black resigned. Take note: I showed some of the previous games from my notebook, but I remember the game with Timoshchenko without a crib, even though it was played a quarter of a century ago! Why? Because at one time I did some serious work on it, I thought about the meaning of every move, and this game became for me a defining, fundamental one in the given variation. From the Ki ng's Indi an Defence to the Ki ng's Indi an Attack

The triumph of White's opening plan - his offensive on the queenside. 30 ...

axb5

31 axb5

gxf3+

32 'it>xf3

t2Jf5

In desperate situations a player often re­ sorts to 'kamikaze' tactics. Such final bursts of activity by the opponent should be treated

Inveterate King's Indian players sometimes try to obtain their customary positions with reversed colours and an extra tempo. One would imagine that it is not so difficult, knowing the ideas of an opening, to make advantageous use of the extra tempo. In fact, easy problems do not exist in chess and an extra tempo does not always prove to be a blessing. Why? Engraved on my

1 12

g2 c3! 1 8 . . . l:!.ae8 1 9 dxc4 e4 is not dangerous in view of 20 fxe4 l:!.xe4 2 1 it'd3. b5

1 9 bxc3 20 a5

Here Lev Alburt offered a draw, and I, displaying optimism unwarranted by the position, declined. But after a tense struggle the game nevertheless came to a peaceful end . In the following example White used his extra tempo more successfully. Dvoretsky - Tataev Beltsy 1 972 King 's Indian Attack

1 e4

c5

2 lDf3

g6

3 d3

�g7

4 g3

ttJc6

5 �g2

ttJf6

6 0-0

d5

1 14

� 7 lbc3

The Development of an Opening Repertoire

0-0

I reached this position several times with Black after 1 d4 lbf6 2 c4 g6 3 lbf3 i.. g 7 4 g3 0-0 5 i.. g 2 d6 6 0-0 lbc6 7 lbc3 e5 (sometimes 7 . . . i..f5 or 7 . . . i..g4 is played here, but the most usual move is 7 . . . a6).

White has a choice between 8 dxe5 and 8 d5. The pawn exchange is not as harmless as it appears. Initially I replied to 8 dxe5 with 8 . . . dxe5. However, in the variation 9 'ji'xd8 .l:r.xd8 1 0 i..g5 i..e6 1 1 lbd2 it is not so easy to equalise, for example 1 1 . . . h6 1 2 i.. xf6 �d2 1 3 i..xg7 'it>xg7 1 4 i.. xc6 bxc6 1 5 b3, and White's chances are somewhat better thanks to his superior pawn structure (Koyfman-Dvoretsky, Moscow Champion­ ship 1 966). The game Vaganian-Dvoretsky ( USSR Championship First League, Tbilisi 1 973) went 8 dxe5 lbxe5 9 lbxe5 dxe5 1 0 'ji'xd8 .l:r.xd8 1 1 i..g5 .l:td4! 1 2 e3?! .l:i.xc4 1 3 .l:!.ac 1 c6! 1 4 i..xf6 i.. xf6 1 5 lbe4 .l:i.xc 1 1 6 lbxf6+ Wg7 1 7 lbe8+ 'it>f8 1 8 .U.xc 1 'i.t>xe8 1 9 .l:i.xc6 i.e6! 20 I1c5 .U.c8 Y.,-Y... White also achieves nothing with 1 2 b3 c6, but he retains pressure by 1 2 lbd5! lbxd5 1 3 cxd5 e4 1 4 .l::!.fd 1 - here the theory books give inaccu­ rate information. After 8 d5 lbe7 in the 1 966 Moscow Championship Vladimir Yurkov chose 9 c5

against me. I had only just become a master, I had no serious knowledge of the theory of this variation, and I had never seen the move 9 c5. Since Aaron Nimzowitsch's book My System, which I had previously read with great pleasure, was still fresh in my mind, at the board I was able to find an idea in the spirit of Nimzowitsch, with a blockading knight on d6: 9 . . . lbe8 1 0 cxd6 lbxd6. The continuation was 1 1 e4 c5?! 1 2 a4 ?! i.. d7 1 3 .l:i.e 1 h6 1 4 b3 f5 with a good game for Black. White could have seized the initiative by 1 2 dxc6 lbxc6 1 3 i.. g5! f6 1 4 i.. e3 followed by 1 5 i.. c5 (lvkov-Uitumen, Palma de Mallorca 1 970) . Therefore Black does better not to hurry with 1 1 . . .c5, but to first play 1 1 . . . h6!, and then, according to circumstances, 1 2 . . . c5, 1 2 . . . c6 or 1 2 . . .f5. The main continuation is 9 e4 lbd7. The game Doda-Dvoretsky (Polanica Zdroj 1 973) went 1 0 lbe1 f5 1 1 lLld3 lbf6 1 2 f3?! ( 1 2 f4 is better, as Etruk played against me a year earlier; 1 2 exf5, 1 2 i.. d2 and 1 2 i.. g5!? have also occurred) 1 2 . . . h6! 1 3 i.. d2 g5 (threaten­ ing 1 4 . . . f4 ) 1 4 exf5 lbxf5. The initiative is with Black, whose subsequent plan is . . . 'ii'e8-g6, oo.lbd4 , oo.i.f5, o o. .l::!. f7, o o. .l:i.afB, .h6-h5 and .g5-g4 . 0 0

0 0

The manoeuvre of the knight to d3 is a little slow. White has some worthy alternatives: 1 0 i.. e3, 1 0 i.. d2 and 1 0 b4 . He can also consider the non-routine 1 0 lbg5!? h6 1 1 lbh3 followed by 1 2 f4 . After this excursion into the theory of the King's Indian Defence, let us return to the Dvoretsky-Tataev game . (see diagram)

So, I have an extra tempo. Which moves for White are useful, and which are not? What would you say, for example, about 8 .l:i.e1 ? Yes, this would be a move in the style of my game with Alburt, a move which does not improve, but worsens the position. Black replies 8 . . . d4 9 lbe2 e5. Now White needs to

l2J

The Development of an Opening Repertoire

10 . . .

b6

1 1 'it?h2

..ia6

1 15

1 2 e5

-position after 7

. . .

0-0-

play f2-f4, and in this case his rook is clearly worse placed at f1 than at e 1 . 8 h3! In every eventuality it is useful to defend the g4-square - both in an endgame arising after 8 . . . dxe4, and in a blocked position after 8 . . . d4. Knowing the ideas of the corresponding variation of the King's Indian Defence, about which we have only just spoken, it was not difficult to find the correct solution with White.

My opponent did not want to play a theoretical position a tempo down, and he tried to devise something new. 8...

h6

9 tt::l d 2 ! ? I f now 9 . . . d4, then it is clear that I have made a useful move. After 9 . . . ..ie6 or 9 . . . e6 the planned advance f2-f4 gains in strength. Black should probably have played 9 . . . dxe4 10 dxe4 (nothing is given by 1 0 tt::lcxe4 1Llxe4 1 1 tt::lxe4 b6! 1 2 tt::lxc5 bxc5 1 3 ..ixc6 �xh3) 1 0 . . . "�c7 1 1 f4 .l:!.d8 1 2 e5 tt::l e8 .

9. . .

e6

10 f4

Now it is unfavourable for the opponent to open the centre: after 1 O . . . dxe4 1 1 dxe4 and 12 e5 'holes' appear in his position at d6 and f6, and the white knight will occupy a powerful position on e4.

An interesting positional problem: to where should the knight retreat, d7 or e8? (We will not even consider the h7-square - there the knight is out of play. ) What is Black intending to do next? Of course, he won't want to weaken his pawn chain with . . . f7-f6. He intends to seize space on the queenside by . . . b6-b5-b4. How should White react? He will probably try to prevent this plan by 1 3 a4. At the same time the knight thrust to b5 becomes a possibility. After 1 2 . . . tt::le8 Black retreats his bishop to b7, and then plays . . . a7-a6, . . . tt::l c7 and . . . b6-b5. Now tt::lb5 is no longer dangerous, since the knight on e8 is defending the d6-square. But with the knight on d7 none of this is possible, and White gains the better prospects, since the oppo­ nent does not have any normal plan. 12 . . .

tt::l d 7?

1 3 a4!

l:.c8

14 tt::l f3

'ir'c7

1 5 tt::l b 5

..ixb5

1 6 axb5

tt::l d4

1 7 c4! A good positional move: the diagonal of the

1 16

�

The Development of an Opening Repertoire

g2-bishop is lengthened, and the mobility of the opponent's pieces is restricted. 17 . . .

dxc4

1 8 dxc4

'ikb8

19 'iVa4 was threatened.

24 h4!

1 9 i.e3 19 l2Jxd4!? cxd4 20 b3 also came into consideration. 19 . . .

l2Jxf3+?!

After this exchange Black remains without any counterplay at all. He should have sacrificed a pawn: 1 9 . . . .l:!.fd8 . 20 'iixf3

advancing on the queenside. Black has parried the immediate threats, but nearly all of his forces are stuck there. In such cases the switching of the attack to the opposite wing proves very effective.

.U.fd8

21 l:!.fd 1 There is now the unpleasant threat of 22 'ii'b 7, winning a pawn. 21 . . .

l:.c7

22 �d6!

i.f8

23 .l:!.c6

.l:!.dc8

How should White proceed further?

Threatening 25 h5. 24 . . .

h5

25 g4

hxg4

26 'ikxg4

i.g7

27 h5

tZ:lfS

White must play i.e4 and .U.g1 . But with which move should he begin? When trying to convert an advantage, one should care­ fully watch for the opponent's counter­ chances. The natural move 28 i.e4? is refuted tactically: 28 . . . l:.xc6 29 bxc6 i.xe5! 30 fxe5 'ikxe5+ with the threats of 31 . . . 'iVxb2+ and 31 . . .f5 . 28 .U.g1 ! White has a decisive advantage, which he converted into a win. When playing with reversed colours those set-ups which we like for Black, it is hardly ever possible to directly transfer from there any specific variations. On other hand, as I hope you have seen, a wide use can be made of the typical plans, methods and evaluations of the corresponding source openings. And in general, for a deep understanding of any opening, a study of general ideas is work that is no less and possibly more important that the memoris­ ing of specific variations.

There is nothing that Black can move, and his pieces are running up against his own pawn on c5. Here it is useful to remember the 'principle of two weaknesses' . It usually operates in the endgame, but sometimes also in the middlegame, when the opponent is completely tied down and deprived of counterplay. Up till now White has been

To solve opening problems, one sometimes has to use ideas typical of quite different schemes. This means that the practical player should not restrict himself to a study of games played only with 'his' openings. Study well-annotated games, even if 'other' openings are employed in them. You will not expand your opening repertoire straight

The Development of an Opening Repertoire

away, but it is possible that some game will provide the impetus for this. But more important, you will expand your arsenal of positional ideas, methods and evaluations, which, as I have already said, may possibly

ttJ

1 17

be used in the most varied openings. Record ideas that appeal to you in the form of 'positional sketches', as described in the first book of this series (Secrets of Chess Training).

Solutions Nezhmetdi nov -N.N. (Kazan 1 951 ) . 1 3 'ili'b3 ! !

1fxa1

Bad is 1 3 . . . 'it'xb3 1 4 �xd7+ and 1 5 axb3, when White wins a piece.

1 4 �b2

�b1

1 5lt:lf3 !

'ii'x h 1

1 6lt:le5 Threatening mate in one move. 16 . . .

e6

1 7 �xd7+

.l:i.xd7 .l:i.d8

18 'ii' b 8+ 18 :;t>e7 1 9 lt:Jc6 mate. ..

1 9 'iVb5+ White has built up a decisive attack. The game concluded 1 9 .. :lt>e7 20 �b7+ f6 2 1 'i'xf7+ 'it>g5 22lt:lf3+ 'it> h 5 23 g4+ ! 'it>xg4 24 'i'xe6+ 'it>f4 25 �e5+ 'it>xe4 26 lt:Jg5 mate. Dvoretsky - Veselovsky (Moscow 1 967) . White's plan is clear: play o n the queenside. I n one order or another he intends 'i!Va4 , ia3, l:tb6(b8) and so on. When carrying out your plan, you are obliged to reckon with your opponent's intentions. What does Black want here? Obviously it would be advantageous for him

to weaken the pressure on his queenside, by exchanging the light-square bishops: 1 7 . . . �h3. For this reason the natural move 1 7 'i!Va4? would be a serious inaccuracy. It is easy to forestall the exchange of bishops by 1 7 l:i.e 1 . But in itself this move is not needed by White, and it does not come into his plan. 1 7 �a3 ! The strongest continuation. While intensify­ ing the pressure on the queenside, at the same time White preserves his light-square bishop from exchange. In Nimzowitsch's opinion, it is such moves, combining the implementation of your own plan and prophy­ laxis against the opponent's ideas, that comprise the essence of genuine positional play. Black managed to hold out for only a few more moves: 1 7 . . J:tfe8 1 8 'it'a4 'it'c7 1 9 'ii'a 8 f5? ! ( 1 9 . . J::td 7 with the idea of 20 . . . d5 was more consistent, against which I would probably have played 20 l:tb8 d5 2 1 l:tfb 1 dxc4 22 dxc4) 20 llb8 �f8 (20 . . . e4!?) 21 �b4 'iia 7? 22 .l:ta1 (22 �xd6) 22 . . . 'it'd7 23 .l:ta6 �f7?! 24 �a5 Black resigned . In the final position the domination of the white pieces on the queenside is striking.

118

�

The Development of an Opening Repertoire

Appendix (games, mentioned in the lecture, which will supplement your impressions of the opening variations analysed)

Petrosian - Pachman Bled 1 96 1 King 's Indian Attack

1 tt:Jf3

c5

2 g3

tt:Jc6

3 i.g2

g6

4 0-0

i.g7

5 d3

e6

6 e4

tt:Jge7

7 IZ.e1

0-0?!

7 . . .d6 is safer. 8 e5!

d6

9 exd6

'ii'x d6

10 tt:Jbd2

'ii'c7

Either here or on the following move . . . b7b6 should have been played. 1 1 tt:Jb3!

tt:Jd4?

12 i.f4

ii'b6

1 3 tt:Je5

tt:Jxb3

14 tt:Jc4! 14 axb3 tt:Jd5 1 5 tt:Jc4 'ii'c6 was less accurate. 14 . . .

ii'b5

In the event of 1 4 . . . 'ii'd8 1 5 axb3 the threats of 1 6 i.d6 and 1 6 .l:ta5 are unpleasant. 1 5 axb3

a5

Black is forced to defend against 16 l:!.a5. 1 6 i.d6

i.f6

1 7 ii'f3

g7

1 8 .l:!.e4?! The combination which White carried out on his next move was already possible here. 18 . . .

lidS

1 9 'ii'xf6+! 20 i.e5+ 21 i.g7!

xf6 g5

Black resigned. Smyslov - Denker USSR-USA Match, Moscow 1 946 Sicilian Defence c5 1 e4 tt:Jc6 2 tt:Jc3 g6 3 g3 4 i.g2 i.g7 e6 5 d3 tt:Jd4? ! 6 i.e3 7 tt:Jce2 ! d6 8 c3 tt:Jc6

8 . . . tt:Jxe2 looks more natural. 9 d4 cxd4 1 0 tt:Jxd4! White must capture on d4 with a piece, in order to subsequently exploit the weakness of the d6-pawn.

ctJ

The Development of an Opening Repertoire

10 . . .

ltJxd4

1 1 ii.xd4

e5

1 1 . . . tt:Jf6 was better. 1 2 ii.e3

lbe7

1 3 lt:'le2

0-0

1 4 0-0

ii.e6

1 5 'iid 2

'fic7

Not 15 . . . d5 in view of 1 6 ii.c5.

21 .l::t d 1

119

.i:!.fd8

2 1 . . . ii.xc4 22 l:tac 1 . 22 l:tac1

11ac8

23 b3

b6

24 lt:'lc3! White is planning the advantageous ex­ change of the light-square bishops. And if 24 . . . �h7 there follows 25 ..te4!, intending 26 h4 , or 26 lt:'lb5 and 27 'iid3. 24 . . .

'iie 7

25 ii.d5

�h7

26 ii.xe6

'iixe6

27 l:td3

.Uc7

28 .Ucd 1

l:tf7

29 lt:'le4

SiLtS

30 .U.d5

'it'g4

3 1 l:t1 d3 The loss of the pawn is now unavoidable. 31 ltJxd6?! ii.xd6 32 .Uxd6 would have been premature: 32 . . . 'ii'xd 1 +!. 1 6 .l::tfc1 ! In order to gain control of the d5-point, White must prepare c3-c4 . 1 6 b3, with the same aim, is weaker in view of 1 6 . . . b5 ( 1 7 a4 bxa4 1 8 J:txa4 ii.xb3) . But now if 1 6 . . . b5 there follows 1 7 a4! a6 ( 1 7 . . . bxa4 1 8 11xa4 a5 1 9 �ca 1 with the threat of 2 0 b4) 1 8 l:td1 , for example: 1 8 . . J:tad8? 1 9 axb5 axb5 20 lla7, or 1 8 . . Jlfd8 1 9 axb5 axb5 20 :Xa8 l:.xa8 2 1 'lxd6, or 1 8 . . . ii.b3 1 9 'iixd6 'iixd6 2 0 .l::txd6 .ba4 2 1 lt:'lc1 , preparing 22 b3.

31 . . .

ii.e7

32 lt:'lxd6

ii.xd6

33 l:txd6

l:tdf8

34 'fixeS At first sight, a rather risky decision. By giving up his f2-pawn, White exposes his king. But Smyslov has accurately calculated that he will be the first to begin an attack. 34 . . .

l:txf2

35 .:td7+

l:tf7

36 l:txf7+

l:txf7

16 . . .

f5

37 l:td8 !

l:tg7

1 7 c4

fxe4

38 'ilea

g5

1 8 lt:'lc3

lt:'lf5

39 iVh8+

�g6

40 .Ud6+

w

I n the event of 1 8 . . . ii.xc4 White has a pleasant choice of 1 9 lt:'ld 1 , 1 9 ltJxe4 or 1 9.b3.

19 ltJxe4

ltJxe3

1 9 . . . tt:Jd4 20 c5! d5 2 1 lt:'lg5 ii.f7 22 f4 with an attack for White. 20 'iix e3

h6

41 'iix h6 It has all become clear. Black has no counterplay - his rook is passive, his king is vulnerable, and in addition, being two pawns down, he cannot agree to the exchange of queens.

1 20

�

The Development of an Opening Repertoire

41 . . .

it'f5

42 l:!.d1 !

lli'c5+

43 'it>g2

�e7

44

.l:!.f1 +

'it>g8

45 �f6

'ii'e8

46 �f5

g4

47 .Uf2

'ike7

48 �d3

.l:tg5

49 Ite2

'ii'f8

50 �e4

:tg7

51 'ii'd 5+

'ii'f7

after 1 2 i.b5 b6 1 3 l:!.ad1 a6 1 4 i.d3 b5 1 5 i.b1 i.b7 he obtained an excellent position. 1 1 .i:td1 11 0-0?! was inaccurate in view of 1 1 . . . lbe5! with equality. 11 . . .

e6

1 2 0-0

�c7

52 l:!.e6! Black resigned. Dolmatov - Bagi rov Frunze 1 983 GrOnfeld Defence

1 c4

tt::lf6

2 tt::lc 3

d5

3 cxd5

tt::l xd5

4 d4

g6

5 e4

ttJxc3

6 bxc3

i.g7

7 i.a3

tt::l d 7

Black can get by without this move, by playing 7 . . . b6!? followed by . . . i.b7, . . . 0-0 and . . . c7-c5 , after which his knight will be able to occupy the more active c6-square. 8 tt::lf3 9 'i!Vb3

c5 0-0

1 0 i.d3 Obviously weaker is 1 0 i.e2 cxd4 1 1 cxd4 tt::lf6. If 1 0 l:td 1 the Encyclopaedia of Chess Openings recommends 1 0 . . . cxd4 1 1 cxd4 tt::lf6 1 2 i.d3 i.g4 1 3 �xb7 i.xf3 1 4 gxf3 'iVxd4 , evaluating the resulting position as roughly equal. 10 . . . b6?! Black played more actively in the game Evans-Korchnoi (Buenos Aires 1 960): 1 0 . . . 'i!ic7 1 1 0-0 .l::!.b8 (intending . . . b7-b5), and

1 3 e5! The start of a sharp plan of attack on the kingside, typical of such positions. 13 . . .

i.b7 .l:.fc8 14 tt::l g 5 1 4 . . . h6? loses to 1 5 tt::l xe6! 'ilic6 1 6 d5. 15 'i!ib1 1 5 i.e4 came into consideration, with the aim of occupying the central e4-point with the knight (if 1 5 . . . h6, then 1 6 tt::lxe6 is still strong). But Dolmatov avoids the exchange, hoping to use his bishop to break up the enemy king's defences. 15 . . .

l:!.ab8

1 6 h4

b5

This pawn offensive on the queenside is too late. 1 6 . . . 'ir'c6! was better, threatening mate and intending . . . 'ii'a4 . 1 7 h5

b4

1 8 hxg6!

bxa3

If 1 8 . . .hxg6, then 1 9 i.xg6 fxg6 20 ttJxe6 it'c6 2 1 d5 is decisive.

lLJ

The Development of an Opening Repertoire

1 9 gxf7+

'it>h8

20 i.xh7 Threatening 'iVb 1 -g6-h5. 20 . . .

lbf8 i.h6

21 1i'd3 If 21 cxd4 , then 22 'iVh3 i.xe5 23 i.g8+ lt>g7 24 'iig4! 'it>h6 25 f4 . . . .

22 'ii'h 3 !

i.xg5

2 3 i.g8+

'it>g7

24 l::i. d 3

30 �xf5!

exf5

3 1 'ilt'xg6

.l:!.b1 +

32 'it>h2

:b6

33 e6

'>t>d6

34 1i'xf5

cxd4

35 cxd4

9-;c7

121

And Black resigned. Timman

-

Yusu pov

Candidates Match, 8th Game, Tilburg 1 986 Queen 's Gambit

1 d4

24 . . .

.Jte4

24 . .1Lf4! would have created more difficul­ ties for the opponent, but even in this case, according to analysis by Dolmatov, White wou l d have retained a decisive advantage: 25 'i'h4! tt::lg6 26 'ii'f6+ �f8 27 'ifxg6 cxd4 28 cxd4 .i.a6 29 .l:!.g3! (but not 29 l:!.f3? �c 1 !) 29 .1Lxg3 30 'ifh6+ �e7 3 1 f8'i¥+! (31 't!Yxe6+? �xe6 32 f8'if+ �d7 is much weaker) 31 . . . .l:!.xf8 32 'ifg7+ 'it>d8 33 'ifxf8+ 'it>d7 34 i.xe6+! �xe6 35 'ii'f6+ �d7 (35 'it>d5 36 'ii'f3+ �e6 37 d5+ �d7 38 't!Yxg3) 36 "ii'xa6 i.h4 37 ..Wxa3, and White has four pawns for the bishop. . .

. . .

. . .

d5

2 c4

e6

3 tt::lc 3

tt::lf6

4 cxd5

exd5

5 i.g5

i.e7

6 e3

tt::l b d7

7 'ilt'c2

0-0

8 i.d3

.Ue8

9 tt::lf3

tt::lf8

1 0 0-0

c6

1 1 .i:.ae1

tt::l e4

1 2 .i.xe7

'iVxe7

1 3 i.xe4

dxe4

1 4 tt::l d 2

f5

In the World Team Championship (Luzern 1 985) against Zoltan Ribli, Yusupov em­ ployed a riskier plan: 1 4 . . . b6!? 1 5 'ilt'a4 b5! and gained a quick draw. (Incidentally, in that same event in the very same variation he beat Li Zunian with White). Timman was evidently hoping for a repetition of this variation and had certainly found a way of improving on Ribli's play. But Yusupov had made a deep study of this variation, which in case of necessity gave him the possibility of varying his plans, thus avoiding prepared lines by the opponent.

25 llg3

'iVd8

26 f4

i.f5

27 "iVh2

tt::l g 6

1 5 f3

28 fxg5

'it>f8

1 6 tt::lxf3

i.e6

29 'iVh6+

'it>e7

1 7 e4

fxe4

exf3

1 22

The Development of an Opening Repertoire

1 8 l:txe4

h6!?

1 9 tt'le2?! An unsuccessful reaction to the novelty. 1 9 .l:tfe1 was preferable, or 1 9 tt'le5, as Yusupov had played against me a year earlier in some training games with a 1 5-minute time control.

Taking account of the situation in the m atch, with an exchange sacrifice Yusupov forces a draw. 28 . . . ll.f8 29 tt'lf5 'iife6 or 29 . . . 'lt>h7 was also not bad . 29 'ii'f7 +

'lt>h7

30 'ii'x e8

'ii'x b2 !

3 1 l:!.f2

'iifa 1 +

32 l:tf1

'ii'a 2

Natural tactics in such a situation - without changing the position, to reach the adjourn­ ment and to check at home whether or not there are any winning chances. 33 l:!.f2

19 . . .

'i!fb4!

Black forces favourable simplification.

34 l:tf1

"i'Va2

35 l:tf2

'ii' b 1 +

3 6 l:tf1

'ii' b 2

37 l:tf2

'ii' b 1 +

38 l:!.f1

'ii'c 2

39 l:tf2

'ii'c 1 +

40 l:tf1

'iifd 2

41 llf2

'iifd 1 +

Draw.

20 a3 In the event of 20 tt'le5 Black can simply capture the pawn: 20 . . . il.xa2 . 'it'b3 20 . . . 20 . . . 'ii'c4 21 tt'lc3 was less good. 21 'it'd2

il.d5

22 l:!.xe8

.l:!.xe8

23 tt'le5

tt'le6

24 tt'lc3

'ii'b 6

25 tt'lxd5

cxd5

26 tt'lf3

tt'lf8

Black wants to tie down his opponent by 27 . . . l:!.e4 and 28 . . . tt'le6, but with the same idea 26 . . . tt'ld8 was more accurate. In some cases the knight could have gone to c6, for example, in reply to 26 l:!.c 1 . 27 'ii'f2 !

tt'lg6

Intending . . . .l:!.e4 or . . . tt'lf4 . 28 tt'lh4!

tt'lxh4

'ii'a 1 +

M u k h i n - Dvoretsky Moscow 1 969 King 's Indian Defence

1 c4

tt'lf6

2 tt'lf3

g6

3 tt'lc3

il.g7

4 e4

d6

5 d4

0-0

6 il.e2

e5

7 d5

tt'la6

8 0-0

tt'lc5

9 tt'ld2?!

a5

1 0 b3

il.h6!

11 'ii'c2

tt'le8

1 2 tt'lf3

il.xc1

1 3 'ii'x c1

f5

1 4 exf5

il.xf5

1 5 'ii'e 3

'it'f6

lLJ

The Development of an Opening Repertoire

1 5 . . b6!? is also good. .

1 6 ti:Jd2

tt:Jg7

24 . . .

tt:Jxe2+

25 'ii'x e2

�d3

1 23

But not immediately 25 . . . 'ii'b 2? 26 tt:Jxe4 .

17 a3? 1 7 f3 is correct, with roughly equal chances.

26 'ii'e 1

'ili'b2 !

26 . . . 'ifd4!? was also tempting, creating the terrible threat of 27 ... l:txf2! 28 1i'xf2 e3. Then 27 l:f.c3 tt:Jxb4! (but not 27 . . . l:txf2? 28 �d3!) 28 tt:Jb3 'ifb6 is hopeless for White. The only defence is 27 g3!, when the immediate 27 . . . l:Ixf2?! 28 'it'xf2 e3 allows White to save himself by 29 'ii'f4!. Therefore the following combination suggests itself: 27 . . . tt:Jxb4!? (27 . . .'ii b 2 is stronger) 28 tt:Jxb4 (28 tt:Jb3 'ii'b2 29 tt:Jxb4 'ii'xb3 with advantage to Black) 28 .. J�xf2 29 'i1Vxf2 e3.

17 . . .

�c2 !

1 8 b4 After 1 8 tt:Jb5 White was concerned about 1 8 c6, although perhaps he should have played this and followed up with 1 9 b4!? . . . .

18 . . .

axb4

19 axb4

.l::!. xa 1

20 .U.xa 1

ti:JfS

21 'ili'h3 21 'i'f3? e4 or 21 tt:Jde4? �xe4 22 tt:Jxe4 'fh4 was bad for White.

21 . . .

tt:Jd4!

22 1i'e3

tt:Ja6

23 tt:Ja2 Hardly any better was 23 b5, after which I had in mind the following variation: 23 . . . tt:Jc5 24 .tf1 e4 25 1Ia2 tt:Jf5 26 'ii'h3 e3 27 l:lxc2 tt'ld4 28 ti:Jde4 tt:Jxe4 29 tt:Jxe4 'ii'f4 30 .l:.a2 e2! .

23 . . .

e4!

24 .l:.c1 In the event of 24 .l:i.e 1 !? Black would have replied as in the game: 24 . . . tt:Jxe2+ 25 'ii'xe2 .td3 26 'ti'e3 ir'b2 .

Now 30 'ii'f4 exd2! no longer helps White, since his rook is not defended by the knight. However, he finds the fantastic reply 30 'ii'f7+!! (it has to be this square!) 30 . . .'�xf7 3 1 ti:Jf3, when the queen can not go to e4 because of the knight fork, and otherwise White plays 32 tt:Jxd3 and 33 tt:Jfe1 , consoli­ dating his forces. 27 tt:Jc3

tt:Jxb4

28 tt:Jdxe4

�xe4

29 tt:Jxe4

ti:Jd3

30 l::t b 1

it'eS! ?

This move, winning a pawn, was prepared beforehand and therefore it was made

1 24

�

The Development of an Opening Repertoire

quickly. But meanwhile, here Black should have stopped and carefully calculated the variations.

6 d4

During the game both I and my opponent thought that 30 . . . 'iVd4?! was refuted by 31 'ii'e2 'ii'xc4 32 lbg5 ii'xd5 33 'ii'e7. We did not notice the only, but sufficient defence against the mate: 33 . . . �f2! 34 'ii'xh7+ 'it>f8 . However, after 34 lDf3 or 34 'iVh6+ 'it>e7 35 lbf3 the outcome remains unclear. Instead of 31 . . . 'ii'xc4? Black has the deci­ sive 31 . . . l:tf4! 32 lbg5 l:!.xf2 . However, White's play can also be improved: 31 ii'e3!, and Black is nevertheless obliged to go in for the complicated variation given above (31 . . . ii'xc4 32 lbg5 etc . ) . After 3 0 . . . 'iVc2! 31 ii'd 1 nothing i s given by 31 . . J:!xf2? 32 lbxf2 'iixf2+ 33 'it>h 1 'ii'e3 34 �f3(f1 ), but the simple 31 . . . 'it'xc4 is ad­ equate. Here the white queen is less active than after 30 . . . 'iVd4 . 31 'it'e2

lbxf2

32 l:!.e1

lbxe4

33 'ifxe4

l:!.e8!

34 'it'xe5

l:!.xe5

lbc6

7 0-0

e5

8 d5

lbe7

9 e4

lbd7

1 0 lbe1

f5

1 1 lbd3

lbf6

1 2 f4! ?

fxe4

1 2 . . . lbxe4 1 3 lbxe4 fxe4 was more accu­ rate, because after the move in the game my opponent could have replied 1 3 fxe5!? . 1 3 lbxe4

lbxe4

1 4 �xe4

exf4

1 4 . . . ..th3 was slightly weaker, since after 1 5 l:!.e1 Black has to reckon with 1 6 fxe5 dxe5 1 7 lDf2 . 1 5 lbxf4

lDf5

35 .l:!.xe5?! It is probable that the rook endgame is also lost. 35 . . .

dxe5

36 'it>f2

'it>f7

37 'it>e3

'it>e7 !

White resigned, since after 38 c5 b6 39 'it>e4 bxc5 40 'it>xe5 he soon ends up in zugzwang. Etruk - Dvoretsky Viljandi 1 972 King 's Indian Defence

1 c4 2 g3

lbf6

3 �g2

..tg7

4 lbc3

0-0

5 lbf3

d6

g6

An important position for the evaluation of the variation. If 1 6 'i¥d3 Black replies 1 6 . . . ii'f6 followed by . . . �d7 and . . . .:tae8, when he is excellently placed. White should play 1 6 lbe6 �xe6 1 7 dxe6. In this case his chances are evidently preferable, although a draw is still the most probable outcome. 1 6 g4? A nervy move, handing the initiative to the opponent. 16 . . . 1 7 h3

lbd4

ttJ

The Development of an Opening Repertoire

If 1 7 .lte3 there follows 17 .. . ii'h4! 18 �xd4 'i'xg4+ 19 'it'xg4 �xd4+ 20 'ittg2 �xg4 or 19 ig2 .l:!.xf4 with advantage to Black. 17 . . .

1Vh4

1 8 'ife1 Comparatively best. 1 8 1Vd3? is bad be­ cause of 1 8 . . . �xg4! 19 hxg4 1Vxg4+ 20 'itth1 ie5 21 �d2 �xf4 22 �xf4 .l:!.xf4 23 .U.xf4 'i'xf4 24 'it'xd4 l:te8 25 .U.e1 1Vh4+. In the event of 18 �e3 1Vg3+ 1 9 tt::lg 2 .l:!.xf1 + 20 'ixf1 the simplest is 20 . . . it'xh3, but 20 . . . �e5 is also decisive, or even 20 . . . �xg4 21 hxg4 l'xg4 22 1Vd3 tt:Je2+ 23 'ittf2 .U.f8+ 24 'itte 1 1Llg3. 18 . . .

'iVxe1

1 9 .l:txe1

�d7?

Indecisive! I wrongly avoided the previously intended 1 9 . . . .lte5 20 tt::l e2 tt::lf3+ 21 �xf3 :Xf3 22 'ittg2 .l:tf7 (22 . . . �d3!? ). Now if 23 ih6 it is possible to capture the b2-pawn, and otherwise . . . h7-h5 is unpleasant, open­ ing up the game with the two bishops.

1 25

Black loses after 25 . . .l:txe4? 26 lixe4 l:If2+ 27 'ittg3 nxd2 28 lle7. 26 'ittg 3 !

:as

In the variation 26 ... �c8 27 .l:!.b8 �xg4 28 .l:!.xe8 �xe2 29 .l:te7 �xc4 30 �h6 the activity of the white pieces compensates for the two lost pawns. 27 c5!

�cB

28 cxb6

�xb7

29 bxc7

l:tacB

30 �b4?? The decisive mistake! After 30 �a5 White should be able to gain a draw, although 30 . . . .l:tfe8 would still have set him some problems. 30 . . .

l:!.xc7

31 �xd6

l:tc3+

32 'itt h 4

:es

33 .U.b2

:txe4

34 l:!.xb7

h6!

35 �b4

l:ld3

20 'ittg 2!

.l:!.aeB

36 �c5

l:!.f4

21 �d2

�e5

37 .l:!.xa7

.l:!.ff3

22 tt:Je2

tt:Jxe2

38 .l:!.aB+

'ittg 7

23 .l:txe2

�xb2

39 .l:!.a7+

'ittf6

24 .l:tb1

�d4

40 .l:!.a6+

'itte 5

25 l:!.xb7

�b6

White resigned.

1 26

� M a rk Dvoretsky

Ki n g 's I n d i a n Attack (from Wh ite's Poi nt of View) A. player who opens 1 e4 can, if he wishes, /"'\ include in his opening repertoire the system 1 e4 e6 2 d3!? or 1 e4 c5 2 tt:Jf3 e6 3 d3!? . Avoiding the French or Sicilian Defence, White develops his pieces in the same way as Black does in the King's Indian Defence. In so doing White is counting not only on the extra tempo compared with the King's Indian Defence, which he has be­ cause of the reversed colours, but mainly on Black's not very favourable move . . . e7-e6 (in the King's Indian Defence it is extremely rare for White to play e2-e3). The same positions can also be reached from the Reti Opening - 1 tt:Jf3 d5 2 g3 c5 3 �g2 tt:Jc6 4 0-0 - if now or a move later Black plays . . . e7-e6. (However, the King's Indian Attack also includes various other set-ups by Black - provided only that White has developed his pieces in King's Indian fashion. ) But since the given review is based on personal experience in the King's Indian Attack and largely contains my own games, and I usually played 1 e4 , it is to this move order that the main attention will be paid. Some of the games given below are not of theoretical importance, since at least one of the players did not act in the best way. However, they may prove useful for becom­ ing familiar with the pattern of play charac­ teristic of the given opening, and the typical strategic and tactical ideas and methods employed in it. I. B l ack deve l o ps his b i sho p on e7

1 e4

e6

2 d3

dS

3 lt:Jd2

c5

4 tt:Jgf3

tt:Jc6

5 g3

tt:Jf6

6 �g2

�e7

7 0-0

1) Basic ideas for White (examples with poor play by the opponent) Knight sacrifice on g5 Dvoretsky - Damsky Moscow 1 969 7 . . .

0-0

8 .l:te1

b6?!

9 e5

lt:Jd7

1 0 lLlf1

�b7

11 h4

'flc7?

If Black had attacked the pawn a little earlier, White would have had to defend it with the not very useful move \!Ve2 . But now the attack on the pawn is pointless, since it can be defended by the bishop.

tLJ

King's Indian Attack (from White's Point of View)

1 2 ..ltf4

b5

Compared with the normal handling of the variation (8 . . . b5 etc . ) Black has wasted several tempi. 1 3 lt:J1 h2

d5-d4?

This makes it easier for White to launch a decisive attack.

1 27

Knight sacrifice on d5 Dvoretsky - Yusupov Blitz game, Moscow 1 987 7 . . .

jt'c7

8 l::!.e 1

0-0

9 e5

lt:Jd7

1 0 jt'e2

b5

1 1 h4

a5

1 2 lt:Jf1

a4

1 3 a3

b4

1 4 i.f4

bxa3

1 5 bxa3

i.a6

1 6 lt:Je3 With the black queen on c7, it is advanta­ geous for White to play his knight to g4 via the e3-square (rather than h2), with the strong threat of lt:Jxd5!.

1 4 lt:Jg5!

h6

15 jt'h5!

hxg5

16 hxg5

.l:!.fb8

16 . . .

l:tfe8?

1 7 lt:Jxd5!

exd5

1 8 e6

'i*'c8

If 1 6 . . . .l:i.fd8 1 7 lt:Jg4 lt:Jf8 , then 1 8 ..lte4 followed by 'lt>g2 and .Uh1 is strong. 1 7 lt:Jg4

lt:Jf8

1 8 lt:Jf6+! Stronger then 18 i.e4 lt:Jg6. 18 . . .

i.xf6

If 1 8 . .gxf6 1 9 exf6 e5 20 i.d5 (or 20 ifh6 lbe6 21 i.e4) 20 . . . i.d6 21 i¥h6 lt:Je6 22 g6 with mate. .

1 9 exf6

e5

20 i.d5!

.l:te8

20 exf4 21 l:te7; 20 .. J:td8 21 'it>g2; 20 . . . it'd7 21 fxg7 lt:Jg6 22 jt'xg6 'ii'xd5 23 jt'h6.

1 9 exf7+!

�xf7

20 'ilr'e6+

'lt>f8

2 1 lt:Jg5

..ltxg5

22 i.d6+

Ji.. e 7

. . .

21 g6

lt:Jd8

22 gxf7+ Black resigned in view of 22 . . .lt:Jxf7 23 'ifg5 g6 24 'ifh6.

23 i.xd5 Black resigned.

1 28

�

King's Indian Attack (from White's Point of View)

Undermining of the centre by c2-c4 Dvoretsky - Gorchakov Moscow 1 973 7. . .

0-0

8 �e1

b5

9 e5

t'Lld7

10 l'Llf1

f5?!

1 1 exf6

i.xf6

If 1 1 . . . t'Llxf6, then 1 2 i.f4 , aiming to occupy the e5-point. 1 2 h4 1 2 .l:!.xe6? was a mistake in view of 1 2 . . . t'Llde5, but 1 2 t'Lle3!? with the idea of c2-c4 was possible. 12 . . .

t'Llb6

1 2 . . . h6!? 1 3 t'Lle3 or 1 3 l'Ll1 h2 . 1 3 t'Llg5!

"it'd6

Dvoretsky-Damsky game) with the idea of t'Llg4 and t'Llxh6+ . If 1 5 . . . t'Lle5, then 1 6 f4! hxg5 1 7 hxg5 g6 1 8 fxe5 gxh5 1 9 exd6 .ig7 ( 1 9 . . . i.xg5 20 t'Llxd5!) , and now either 20 i.h3 followed by t'Lle3-g2-f4 , or the same strike at the centre 20 c4!? . Dolmatov - Meyer Philadelphia 1 99 1 7 . . .

0-0

8 .l:!.e1

b5

9 e5

t'Lle8

1 0 l'Llf1

f6?!

11 exf6

i.xf6

1 2 t'Lle3

it'd6

If 1 2 . . . e5 Dolmatov was planning 1 3 c4! bxc4 1 4 dxc4 e4 1 5 l'Lld2 with advantage to White.

If 1 3 . . . e5, then 1 4 c4! is strong, demolishing Black's centre.

But here too 1 4 c4!? came into consideration: 1 4 . . . bxc4 1 5 dxc4 i.xg5 ( 1 5 . . . t'Llxc4? 1 6 t'Llxe6!) 1 6 i.xg5 (or 1 6 hxg5) 1 6 . . . t'Llxc4 , although the correctness of the positional pawn sacrifice would still have had to be demonstrated. Another interesting possibility is 1 4 "ii'h5!? h6 1 5 t'Lle3 (the same idea as in the

1 3 c4!

t'Llc7

14 t'Llg4

e5

1 5 l'Llxf6+

gxf6

In the event of 1 5 . . . 'i1Vxf6 White has a pleasant choice between 1 6 cxd5 and 1 6 i.g5!?. 1 6 cxd5 1 7 l'Lld2 ! 1 7 . . .f5 1 8 t'Llc4!. 1 8 t'Lle4

'ilie7

ltJ

King's Indian Attack (from White's Point of View)

1 9 i.. h 6

.Ufd8

20 l:tc1

c4

21 'ifh5

Itac8

22 dxc4

bxc4

23 f4! White has an undisputed advantage. Some players aim to play c2-c4 as early as possible, and therefore after 1 e4 c5 2 tt:'lf3 e6 3 d3 d5 they refrain from 4 tt:'lbd2 in favour of 4 'ir'e2 . After completing their kingside development, they plan to play c2c4 and tt:'lc3. An exchange of knights on d4 is normally advantageous to White Fischer - U .Geller Netanya 1 968 7. . .

0-0

8 �e1

'ikc7

9 e5

tt:'ld7

1 0 'ike2

b5

1 29

have proved premature because of the counter 1 5 . . . i.. b4!. The following variations are possible: a) 1 6 tt:'ld2?! exd5 1 7 e6 'ikxf4!? 1 8 gxf4 .l::!.xe6 1 9 'ii'd 1 .l:tg6+ and 20 . . . tt:'lf6 with a dangerous attack for Black; b) 1 6 .l:!.ec 1 ? exd5 1 7 e6 l:.xe6! 1 8 'ifxe6 'ifxf4! 1 9 'ifxd7 'it'xc 1 !; c) 16 .l:ted 1 ? exd5 17 e6 .l::!.xe6! 18 'iVxe6 'ikxf4! 1 9 'ifxd7 'iff3! 20 tt:'lh2 'ir'xd1 +!; d) 1 6 l:.eb 1 !!. The only move, which can be found by the method of elimination. Now after 1 6 . . . exd5 1 7 e6 the move 1 7 . . . l:txe6?! no longer works: 1 8 'ii'xe6 'ir'xf4 1 9 'ii'xd7 ii'f3 20 tt:'lh2!. The move in the game is stronger than 1 5 'ifg4?! 'it>h8 1 6 .l:!.e2 (intending tt:'lh2-f3) in view of 1 6 . . . .l:.c6 1 7 .:!.c 1 b4 1 8 tt:'lh2 i..a6 with the unpleasant threat of 1 9 . . . i.. xd3. 15 . . .

l!c6

1 6 l:!.ac1

i.. a 6?

Now the thematic stroke on d5 secures White the advantage. Black should have chosen between 1 6 . . . 'ifb6 and 1 6 . . . i.. b 4 .

1 1 h4

a5

12 tt:'lf1

tt:'ld4? !

1 7 i.. x d5!

1 3 tt:'lxd4

cxd4

14 i..f4

l:ta6 ! ?

1 8 e6 1 9 exd7 20 "ili'g4!

exd5 'ifd8 .l:t e6 f5

If 20 . . . 'ii'xd7, then 2 1 i.. e5 is decisive. 21 'ii' h 5

'i!fxd7

22 tt:'lf3

g6

23 'ii' h 6

i..f6

24 lhe6

'i!Vxe6

25 i.. e 5! ! An elegant stroke, after which White's positional advantage becomes decisive. The obvious 25 .U.e1 ? would have unexpect­ edly led to a draw after 25 . . . 'iixe1 +!! 26 tt:'lxe1 i.. g7 27 'it'g5 i.. f 6. 1 5 tt:'lh2! The standard sacrifice 15 i.. xd5?! would

25 . . .

i.. xe5

26 l:!.e1

f4! ?

27 .U.xe5

'iid 7

1 30

�

King's Indian Attack (from White's Point of View)

28 h5!

fxg3

29 hxg6!

gxf2+

If 29 . . . l:txf3, then 30 l:te8+! 'ii'xe8 31 'ii'xh7+ xf7 1 9 'ii' h5+ g6!) 1 8 . . . hxg5 1 9 hxg5 g6 2 0 'i!Vh4 (20 'ii' h 3? �xg5) 2 0 . . . lbd4 2 1 c.t>h2 (2 1 lbg4 lbf5 22 'ii' h 3 c.t>g7!) 2 1 . . . lLlf5 22 lLlxf5 exf5 (22 . . . gxf5 23 .l:th 1 �f8 24 c.t>g1 �g7 25 'ilt'h7+ c.t>f8 26 g6 and 27 �h6) 23 .l:th 1 �f8 24 c.t>g1 �g7. Alas , I overlooked the powerful stroke 23 e6! (instead of 23 .l::. h 1 ? ) . 1 8 lLlh3?

�f8

White's play has come to a standstill. Black, who has the possibilities . . . d5-d4 or . . . lLld4 available, stands clearly better. 3) Black delays castling Dvoretsky - Cook Saint John 1 988 7 . . .

b6

8 l:!.e1

�b7 'i!Vc7

9 c3 9 . . . 0-0 .

1 6 .l:ta2 After the immediate 1 6 c4 I did not like 1 6 . dxc4 1 7 dxc4 lbb6 1 8 tLld2 it'd4?! - and I was wrong: 1 9 .l:tb 1 or 1 9 �e4 .l:ted8 20 'i'b1 would have promised White an advan­ tage. . .

I decided to advance c2-c4 in a more favourable situation. For example, 1 6 . . . a4 1 7 c4 lbb6 1 8 l:td2! 'it'c7 1 9 "ifc2 . And in the event of 1 6 . . . lbb6, as it seemed to me, the attack 17 tLlg5 gains in strength .

16 . . .

lbb6

17 lbg5 Quickly played, without proper calculation. However, Black was already threatening 1 7 . . .d4.

1 0 it'e2 In such situations White should not be in a hurry to advance e4-e5. After 1 0 e5 lbd7 it is bad to play 1 1 d4? cxd4 1 2 cxd4 lbb4

1 32

�

King's Indian Attack (from White's Point of View)

1 3 .l:!.e3 'ir'c2! 1 4 'ir'e 1 'iig6 ( Hort), while if 1 2 tiJf1 !? there follows 1 2 . . . 0-0-0 with chances for both sides (but not 1 2 . . . ttJcxe5? 1 3 ttJxe5 ttJxe5 1 4 .lii.f4 .lii.d6 1 5 ..txe5 ..txe5 1 6 'iih5 with advantage to White, as in the game Nadyrkhanov-Saltaev, Tashkent 1 993) . 10 . . .

0-0-0

23 ..td2 White has a decisive advantage. 23 . . . tbd7 24 hxg6 e5 25 tba5 lt:Jb6 26 lZ'lc6 fxg6 27 .l:!.xe5 .lii. d 6 28 .Ue6 :d7 29 ii.a5 Wb7 30 ttJxa7+ Wxa7 31 ..txb6+ "jgxb6 32 llxb6 Wxb6 33 'ii'a 4 .l:!.hd8 34 "jgc6+ Wa7 35 Uxd6 Black resigned.

1 1 a3 Logical: White prepares an offensive on the queenside, where the enemy king has just taken shelter. In reply 1 1 . . . h6!? came into consideration, and if 1 2 b4 , then either 1 2 . . . g5 or 1 2 . . . tbe5. 11 . . .

..ta6

1 2 e5

tiJd7

1 3 h4! It is important to hinder the thematic advance . . . g7-g5. 13 . . .

h6

14 h5

tiJf8

If 1 4 . . . g5 White was intending 1 5 hxg6 fxg6 1 6 ..th3 tiJf8 1 7 tiJf1 or 1 7.b4!? . 1 5 b4

g6

1 5 . . .f6 1 6 exf6 ..txf6 1 7 �b 1 . 1 6 .l:!.b1

..tb7

17 bxc5

bxc5

1 8 lbb3?! 1 8 hxg6 ttJxg6 1 9 lbb3 was probably stronger, and if 1 9 . . . h5 20 i.g5, aiming at the c5-pawn. After the move in the game Black should have replied 1 8 . . . gxh5 1 9 i.e3 tiJd7. 18 . . .

..ta6?

1 9 ..te3 !

ttJxe5

If the black king has remained in the centre, the plan involving the exchange on d5 also makes sense. Dvoretsky - Bogomolov Moscow 1 967 7 . . .

b5?!

8 .U.e1 Even stronger was 8 exd5 exd5 9 d4! (9 c4!?) 9 . ..c4 1 0 a4! b4 1 1 ttJe5 with the threat of 1 2 tbxc4 (suggested by V iorel Bologan). 8 . . . i.b7 9 exd5

exd5

1 0 c4! A typical blow at the centre. 10 . . .

bxc4

1 1 dxc4

0-0

1 2 cxd5

ttJxdS

1 3 ttJc4

19 . . . c4 20 tbc5!? ..txc5 21 ..txc5 with an attack. 20 ttJxe5 Also possible was 20 .lii.f4 ttJxf3+ 21 'i!t'xf3 .lii.d6 22 ..txd6 �d6 23 c4 or 23 d4 . 20 . . .

'ir'xe5

21 'ir'c2

"jgc7

22 c4

d4? !

White's position is somewhat preferable. In

lLJ

King's Indian Attack (from White's Point of View)

1 33

the subsequent battle, which was not with­ out its mistakes, he managed to outplay his opponent. 1 3 . . . �f6 1 4 �g5 ! ? �xg5 1 5 ltJxg5 ltJd4 1 6 'ld2 h6 1 7 ltJe4 ltJb4 1 8 liad 1 ltJbc2 1 9 .U.f1 f5 20 ltJf6+ 'it'xf6 21 �xb7 l:i.ad8 22 Wh1 'lf7 23 ltJa5 .U.b8 (23. . . ltJb4!?) 24 �g2 l:l:xb2 25 'ii'c 3 .l:!.b5? (25 . . . .J:txa2 26 'ii'xc5 ti:Je6) 26 a4 l:.bb8 27 'ii'x c5 (27 ltJc6 ! ) 27.. .l:.bc8 (27 . . . 'ii'f6) 28 ltJc6 l:txc6 29 �xc6 l:l:c8 30 'ii'x c2 ! .i:txc6 31 'ili'd3 l:td6 32 f3 'ii'd 5 33 l:.c1 Wh7 34 g4 'ii'f7 35 l:i.c4 Black lost on time. -

If White plays e4-e5, Black should aim for . . . g7-g5 - it is on this advance that the evaluation of the position depends.

C hekhov - A. lvanov Qualifying Tournament for the World Junior Championship, Sochi 1 975 1 g3

ltJf6

2 �g2

d5

3 ltJf3

e6

4 0-0

i.e7 b6

5 d3 6 liJbd2

�b7

7 e4

c5

8 e5

ltJfd7

9 .U.e1

ltJc6

1 0 ltJf1

'ii'c 7

Already here 1 0 . . . g5!? came into considera­ tion.

pos i t i o n after

1 2 . . . d4 -

hxg5 1 6 �xg5 �xg5 1 7 ltJxg5 ltJcxe5. 1 5 d4

liJf8

1 6 ltJ1 d2

ltJg6

1 7 �e3 White has an advantage in space and good prospects of mounting an attack on the queenside, whereas the opponent has no real counterplay. 1 7 . . . cxd4 1 8 cxd4 �b4 1 9 'ii'a 4 ltJge7 20 a3 �xd2 2 1 ltJxd2 Wb8 22 .l:!.ac1 ltJd5 23 ltJe4 .l::i. d 7 24 .l:i.c2 .l:i.c8 25 .l:tec1 .l:i.dd8 26 �f1 'ii'e 7 27 �d2 'iff8 28 l:txc6 l:.xc6 29 l:.xc6 'ii'e 8 30 �b5 a6 31 .l:tc8+ Black resigned. I I . B l a c k d e ve l o ps h is b i sho p on g 7

1 e4

e6

1 1 ii.f4

h6

2 d3

d5

1 2 h4

d4?

3 ltJd2

c5

(see diagram)

12 . . .g5! or 1 2 . . . 0-0-0 13 ltJe3 g5 ( 1 3 . . . ltJf8) was better.

1 3 c3

dxc3

14 bxc3

0-0-0

It was still not too late to play 1 4 . . . g5 1 5 hxg5

4 ltJgf3

ltJc6

5 g3

g6

6 �g2

�g7

7 0-0

ltJge7

7 . .. 4Jf6 is evidently weaker in view of 8 exd5 ltJxd5 (8 . . . exd5 9 lle 1 +) 9 ltJb3 b6 1 0 c4 ltJc7 1 1 d4 .

1 34

�

King's I ndian Attack (from White's Point of View)

8 l:le1

12 . . .

a4

1 3 a3

b5

It was possible to exchange a pair of knights by 1 3 . . . tt:Jd4!?, but after 1 4 tt:Jxd4 cxd4 1 5 tt:Jh2 White would have retained the better chances. 1 4 lLl1 h2

b4

1 5 .tf4

�h7

1 6 lLlg4

lLlg8

1) It is unfavourable for Black to castle - White's attack is very dangerous Dvoretsky - Ubi lava Tbilisi 1 979 8. . .

0-0

9 e5 9 h4!? h6 1 0 e5 has also been tried. 9. . .

'ir'c7

1 0 'ife2 White carries out a standard plan of attack: lLlf1 , h2-h4, .tf4 , tt:Jh2-g4 and so on. Compared with the variation with the bishop on e7, it is harder for the opponent to defend, since the dark squares on his kingside have been weakened by the move . . . g7-g6. In this connection, serious consid­ eration should be given to the attempt to sharply change the character of the play by 1 0 . . . g5!? (a recommendation of Viktor Ciocaltea). 10 . . . 11 h4

aS h6

1 2 lLlf1 It is usually advantageous for White to block the play on the queenside - therefore it made sense to play 1 2 a4!? , then c2-c3 and lLlf 1 . I decided not to deviate from my plan.

1 7 c4! After 17 h5 g5 the sacrifice on g5 does not work. In order to strengthen it, White wants to win control of the e4-square by attacking the d5-pawn with his c-pawn. 17 . . .

bxc3

1 8 bxc3

.ta6

1 9 c4!

dxc4

20 dxc4 Now 2 1 h5 (and if 2 1 . . . g5 22 .txg5) cannot be prevented. 20 . . .

.l:tab8

21 h5! 2 1 lLlf6+? was less good: 2 1 . . . �h8 22 h5 g5, or 22 1i'e3 tt:Jd4! 23 tt:Jxd4 cxd4 24 'ir'xd4 l:tfd8. 21 . . .

�h8

21 . . . g5 22 .txg5! was hopeless.

1 35

King's I ndian Attack (from White's Point of View)

22 hxg6

fxg6

23 lLlf6!

lt:Jge7

2 3 lt:Jd4! was a tougher defence. . . .

24 .Uad1 24 tt:lh4!? was also strong.

24 . . .

.Ubd8

25 lld6!

�b7

26 .l:!xe6

lt:Jd4

26 . . lLlf5 27 lt:Jd5 'it'f7 28 l:1xc6 �xc6 29 e6 and wins. .

27 lLlxd4

cxd4

28 'ii'd 3

�xg2

29 'it>xg2

�f7

30 .l:!. h 1 !

lLlf5

31 g4

'ii' b 7+

32 'it>g1

lt:Je3

33 �xe3

'ii'f3

34 'ii'x g6

�xf6

35 .l:i.xh6+ Black resigned. Dvoretsky - Khal ifman Sverdlovsk 1 987 8 . . .

0-0

9 e5

'it'c7

1 0 1Ve2

b6

1 1 h4

�a6

1 2 lLlf1 1 2 c3!? also does not look bad, but then White has to reckon with 1 2 . . .f6 1 3 exf6 .U.xf6 1 4 lLlf 1 e5.

12 . . .

lt:Jd4

1 3 lLlxd4

cxd4

1 4 �f4

lt:Jc6? !

1 4 . . l:tac8 1 5 .l:i.ac1 1Vc5 was better, with the intention of attacking the queenside pawns with the queen. However, 1 6 lLlh2 followed by lLlg4 or h4-h5 would have given White a dangerous attack. .

1 5 a3

15 . . .

'ii'd 7

If 1 5 . . . l:tac8!? I would have simply replied 1 6 l:tac 1 ! , since 1 6 lLlh2 would allow Black to complicate matters by 1 6 . . . �xe5!? 1 7 �xe5 lt:Jxe5 1 8 'ii'xe5 'ii'xe5 1 9 .l:i.xe5 .l:i.xc2 . 1 6 lLlh2

l:tae8

1 7 lt:lg4

f6

After 1 7 . . . h5 1 8 lLlf6+ �xf6 1 9 exf6 (threat­ ening 20 �h6) 1 9 . . . 'it>h7 20 �e5 and 2 1 f4 White would have gained an overwhelming advantage. 1 8 exf6

�xf6

1 9 lt:Jxf6+ 20 �h3

.Uxf6 .l:i.xf4

20 . . . �c8 2 1 �e5 was also hopeless. 21 gxf4

�c8

22 'ii'f3 22 f5 was premature in view of 22 . . . e5. 22 . . . 'iii'f7 23 'it'g3 'it>h8 If 23 .. JU8 there follows 24 f5. 24 .Ue2

l:tf8

25 It.ae1

lt:Jd8

26 f5!

gxf5

27 1Ve5+

'it'g7+

28 1Vxg7+

'it>xg7

29 f4 And White won the endgame.

1 36

�

King's I ndian Attack (from White's Point of View)

2) The pawn should not be advanced to e5 before Black castles kingside because of counterplay by . . . h 7-h6 and . . . g�g5

better chances for White. In the event of 1 8 . ..�c6 he has the promising piece sacri· fice 1 9 tt:lxe6! fxe6 20 .l:txe6, although it does not lead to a direct win: 20 . . . .l:tf7 21 �g5 �e8 22 �xe7? �d7.

Dvoretsky - Ani kaev USSR Championship Semi-Final, Odessa 1 972

18 . . .

�xd5

1 9 �xd5

tt:lxd5

20 tt:le4

l:fd8

2 1 �g5

f6 11d7

8. . .

b6

9 e5? !

'jic7

22 �d2

1 0 'iie2

h6!

23 .l:te2

1 1 h4

g5!

12 hxg5

hxg5

1 3 tt:lxg5

ii'xe5!

14 ifxe5

tt:lxe5?!

After 14 . . . �xe5! Black would have stood better. 1 5 tt:lc4!?

Draw. 3) The tension in the centre is retained Dvoretsky - Averki n USSR Championship, First League, Odessa 1 974 8 . . .

b6

9 c3

�b7

If 9 ...�a6?! White has 1 0 exd5 tt:lxd5 1 1 'it'a4 . 1 0 tt:lf1

h6

1 1 h4

d4! ?

The endgame after 1 1 ... dxe4 1 2 dxe4 'iix d 1 1 3 .l::!. xd 1 is evidently more pleasant for White. 1 2 c4

e5

1 3 h5

15 . . .

tt:lxc4?!

He should have decided on the positional exchange sacrifice 1 5...dxc4! 1 6 �xa8 cxd3. 1 6 dxc4

�b7

1 7 c3 1 7 a4!? . 17 . . .

0-0

1 8 cxd5?! 18 a4! was stronger, with somewhat the

White obviously needs to advance f2-f4. With this aim he could have played 1 3 tt:l3h2 immediately. 13 . . .

�c8

1 3 ...1i'd7!? looks more natural. 1 4 tt:lh4 White is consistent, although it is unclear where the knight stands better: here or on h2 . 14 . . .

�f6

1 5 hxg6

fxg6

1 6 f4

tZJ

King's I ndian Attack (from White's Point of View)

1 37

1 0 . . . d4 or 1 0.. . a4 came into consideration.

16 . . .

1 1 dxe4

it'xd 1

1 2 �xd 1

a4

1 3 l:!.b1 !

0-0

h5!

Not 1 6 . . . exf4? because o f 1 7 l2Jxg6! l2Jxg6 1 8 e5. 17 f5? A strategic mistake. It was necessary to maintain the tension: 1 7 l2Jh2.

17 . . .

g5

Black's position is better, since White has no active possibilities. Dol matov - A.Sokolov Interzonal Tournament, Manila 1 990 8 .. .

b6

9 c3 In the opinion of Sergey Dolmatov, 9 l2Jf1 is more accurate, with the intention of playing 1 0 e5, for example, in reply to 9 . . . d4. And 9 dxe4 1 0 dxe4 'ii'xd 1 1 1 �xd 1 leads to roughly the same favourable endgame for White that occurred in the game. . . .

9...

a5!?

1 0 lLlf1 White intends e4-e5. In the game Ljubojevic­ Kasparov, Niksic 1 983 (where the moves 9 h4 h6 were included) White played 1 1 a4 , to which Kasparov replied 1 1 . . . �a 7!? . But after 1 0 a4 Dolmatov was concerned about the reply 1 0 . . . d4 .

10 . . .

dxe4

If 1 3 . . . a3, then 1 4 bxa3, attacking the b6pawn. 1 4 .ltf4

e5?!

For the sake of a few active moves it was hardly worth Black shutting in his own bishop on g7 and weakening the light squares. However, in the variation 1 4 . . . a3!? 1 5 �dc 1 ! axb2 1 6 l:!.xb2 the b6-pawn is vulnerable, and at the board it was not easy to judge how the complications would conclude after 1 6 . . . l2Jb4 1 7 l:!.d2! l2Jxa2 1 8 l:!.a 1 e5!. 1 5 .lte3

.lte6

1 6 b3

axb3

1 7 axb3 White's position is preferable. 17 . . .

.l:.fd8

1 8 lLl1 d2

f6

1 8 . . . l:!.d3 1 9 l:tdc 1 and 20 .ltf1 . 1 9 .ltf1

.ltf8

1 9 . . . l2Jd4!? came into consideration, after which Dolmatov was intending 20 cxd4 cxd4 2 1 .ltxd4 exd4 22 .ltc4 . 20 b4!

�

1 38

King's I ndian Attack (from White's Point of View)

20 ..tc4 was premature on account of 20 . . .'�f7. 20 . . .

!Ia3?

9 c3

'i!ic7?

An inaccurate move order. 9 . . . ..tb7 is cor­ rect.

It was essential to exchange on b4 , in order to open the diagonal for the dark-square bishop and possibly make use of the d4point. 21 bxc5

bxc5

22 �dc1

tt:lc8

23 ..tc4

..txc4

23 . . . �f7? is not possible in view of 24 .l:.b7+ . 2 4 tt:lxc4 25 tt:lfd2

�a4

26 tt:lb6

l::t a 2 l::t x a1

27 .Ua1 28 �xa 1

tt:ld6?!

f5 fxe4?

29 tt:ld5

29 . . �f7 was a tougher defence. .

�f7

30 tt:lf6+ 31 tt:lfxe4

White now has a decisive positional advan­ tage, which Dolmatov accurately converted into a win. 31 . . . tt:lf5 32 .l:ta6 tt:lxe3 33 fxe3 (33 .Uxc6!? ) 33 ... tt:lb8 34 .l:If6+ �e7 35 l:tb6 tt:ld7 36 J:tc6 �as 37 �2 �a2 38 �e2 �dB 39 �d3 ..te7 40 h4 tt:lb8?! 41 .l:te6! tt:ld7 (4 1 . . . �d7 42 �xe5 tt:lc6 43 tt:lxc5+ ) 42 tt:lc4 l:tg2 43 tt:la5 ..tf8 44 tt:lc6+ �c7 45 tt:lxe5 tt:lxe5+ 46 l:txe5 �dB 47 h5 ..te7 48 hxg6 hxg6 49 .l::te6 g5 50 .l:te5 'it>d7 51 �c4 .Ug1 52 tt:lxg5 l:txg3 53 tt:le4 .l::t h 3 54 tt:lxc5+ �dB 55 tt:le6+ �d7 56 tt:ld4 ..tf6 57 .l::tf5 �e7 58 �d3 .Uh1 59 c4 l:ta1 60 .l::t b 5 .Ua3+ 61 �e4 �d7 62 l:!.b7+ �c8 63 l:tb3 l:ta1 64 �d5 l:t.d 1 65 c5 Black resigned. 4) Exchange of pawns on d5 Yu rtaev - Dvoretsky Frunze 1 983 8...

b6

1 0 exd5!

exd5

If 1 O . . . tt:lxd5, then 1 1 d4! cxd4 1 2 tt:lxd4 tt:lxd4 1 3 i.. xd5 is strong. 1 1 tt:lf1

0-0

1 2 ..tf4

'it'd7

1 3 d4 1 3.h4!? was also good. 13 . . .

cxd4

1 4 cxd4

..tb7?

14 . . .tt:lf5 was better, not fearing 15 tt'le5 tt:lxe5 1 6 dxe5 ..tb7 followed by 1 7 . . . d4. But now after 1 5 !i.c 1 (as in the game) or 1 5 h4 White has a significant advantage . Kaiszau ri - Tu kmakov Vilnius 1 978 8 . . .

b6

9 exd5!?

exd5

9 . . . tt:lxd5!? comes into consideration, and if 1 0 d4 cxd4 1 1 tt:lb3, then not 1 1 . . . i..b 7?! 1 2 tt:lfxd4 tt:lxd4 1 3 tt:lxd4 (threatening 1 4 c4 ) 1 3 . . . :tc8 1 4 :txe6+!! fxe6 1 5 tt:lxe6 'ii'd7 1 6 tt:lxg7+ 1i'xg7 1 7 i.. xd5 with a powerful attack (Komlyakov-Moskalenko, Noyabrsk 1 995), but 1 1 . . . d3!? 1 2 'i!Vxd3 0-0 1 3 tt'lfd4

ttJ

King's I ndian Attack (from White's Point of View)

tt:Jxd4 1 4 lt:Jxd4 .i.b 7 1 5 .i.d2 l:tc8, and 1 6 c4? does not work in view of 1 6 . . ..i.a6 1 7 b3 tt:Jc7 (a recommendation by llya Odessky) .

0-0

10 d4

If 1 0 . . . lt:Jxd4, then 1 1 lt:Jxd4 .i.xd4 1 2 lt:Jb3 with advantage to White. And if 1 0 . . . c4 he has the strong reply 1 1 lt:Jes lt:Jxd4 1 2 tt:Jdxc4. 1 1 dxc5

bxc5

1 2 lt:Jb3

c4

1 3 lt:Jbd4

.i.g4

14 .i.e3

'it'd?

1 5 'ii'd 2 From the opening White has gained some­ what the better position.

1 39

dubious for Black, but 1 O c4 !? comes into consideration. . . .

1 0 lt:Jb3

10 . . .

'ir'b6

1 O . . . .i.g4!? is preferable, although after 1 1 h3 .i.xf3 1 2 'ii'xf3 0-0 1 3 Si.f4 , intending .U.fe1 and .U.ad 1 , White retains positional compen­ sation for the sacrificed pawn. Now he could have simply played 1 1 .i.f4 ! 0-0 1 2 .i.d6 and 1 3 ..tcs with the better chances. I preferred 1 1 Si.g5, and after 1 1 . . .lt:Jf5 ( 1 1 . . . 0-0!? 1 2 lt:Jfxd4 lt:Jf5 ! ) 1 2 .U.e 1 + ii.e6 I went in for a double-edged combina­ tion: 1 3 g4 lt:Jd6 1 4 lt:Jfxd4! ( 1 4 c3 lt:Je4 ! ) 1 4 . . . .i.xd4 1 5 lt:Jxd4 'ir'xd4 1 6 .i.xd5!?. After fascinating complications the game ended in a draw.

Dvoretsky - Vulfson I l l . B l a c k p l ays . . . .id6 a n d

Moscow 1 986

. .

8 exd5!? (Instead of 8 .:te1 ) .

1 e4

e6

8 . . .

2 d3

d5

exd5

If 8 . . ti:Jxd5!? I was intending 9 lt:Jb3 b6 1 0 c4 followed by 1 1 d4 , but, as Odessky has pointed out, Black retains a good position by continuing 1 1 . . .lt:Jde7! 1 2 d4 .i.a6 ! .

3 li:Jd2

c5

4 lt:Jgf3

lt:Jc6

.

9 d4

cxd4

1 0 . . ti:Jxd4 ? ! 1 1 lt:Jxd4 .i.xd4 1 2 lt:Jb3 is .

5 g3

.i.d6

6 .i.g2

lt:Jge7

7 0-0

0-0

8 lt:Jh4!?

. tbge7

1 40

�

King's I ndian Attack (from White's Point of View)

- position after

1) Black meets the opponent's pawns with . f7-f5 . .

Ciocaltea - Li berzon Netanya 1 983 8...

f5

9 f4

j.c7

1 0 c3

'it>h8

11 exf5

exf5

1 2 tt:'ldf3 This arrangement of the white knights is typical of the given variation. 12 . . .

j.e6

1 3 J:le1 Another development scheme is 1 3 j.e3!? , followed by j.f2 , 'iid2 , l!ae1 and a possible d3-d4.

16

. . .

J:!.ab8 -

his pieces and fully neutralising the pres· sure on the b-file. 17 . . . tt:'lc8?! This knight retreat allows White to carry out a typical diversion on the kingside. 1 7 . . . .U.ae8 was better, and if 1 8 l:tc2 , then 1 8 . . . tt:'ld 8. 1 8 tt:'lg5 Threatening 19 'ii'h 5. 18 . . . 18 .. J:if6 was preferable.

g6?!

1 9 c4! d4 Not 1 9 . . . dxc3? 20 j.xc6 followed by 21 j.c3+ . By blocking the queenside White has freed his hands for action on the kingside, where he is stronger. tt:'lb6 20 'ii'e 2 nbe8 21 b3

13 . . .

j.g8

1 4 j.d2

�d7

22 �f2 23 j_f3

tt:'lc8 nxe1 +

1 5 a3

a5

24 Ilxe 1

16 a4

l:tab8

25 llxe8 26 g4!

.l:Le8 'iixe8

(see diagram)

Black intends 17 . . . b5 1 8 axb5 .l:txb5 with pressure on b2 . 1 7 l:tc1 1 An excellent prophylactic move. If 1 7 . . . b5 White replies 1 8 axb5 l:txb5 1 9 .l::t c2 followed by j.c1 and l:tce2 , harmoniously deploying

tt:'ld6 In the event of 26 . . . fxg4 27 j.xg4 White would have followed up with f4-f5. 27 gxf5 tt:'lxf5 27 . . . gxf5 came into consideration, intending 28 �g2 tt:'lb4! 29 'ii'h3 tt:'lxd3 30 tt:'lxf5 tt:'lxf5 31 'iixf5 'ii'g6 32 'it'g4 "iff6 with chances for both sides.

ctJ

King's I ndian Attack (from White's Point of View)

2 8 .ie4

t:Lle3?

28 . . t:Llce7 was better. .

29 .ixc6!

bxc6

29 . 'ii'xc6 30 'ii'xe3 dxe3 31 Si.c3+ or 30 .be3 dxe3 31 'ii'b 2+ . . .

30 t:Lle4

t:Llg4? !

31 'ii'g 2

t:Llh6

32 'ifg5

'i*'f8

33 ii'xc5

'ifxc5

14 . . .

h6

1 5 d4

l:td8

1 6 Si.f2

cxd4? !

1 7 t:Llxd4

t:Llxd4

141

1 8 Si.xd4 White has an overwhelming positional ad­ vantage. 2) Black does not play

. . .

f7-f5

34 t:Llxc5 Dvoretsky - Dieks

Black resigned.

Wijk aan Zee 1 975 Dvoretsky - C hekhov

8 . . .

b6

Sverdlovsk 1 987

9 f4

Si.c7 ! ?

1 0 f5

exf5

8 .. .

b6

9 f4

f5

1 0 exf5

exf5

1 1 lLldf3

'it'c7?!

1 2 c3 It is also possible to place the bishop on e3 immediately, in order after 1 2 . . . d4 1 3 Si.f2 to undermine Black's centre by c2-c3.

12 . . .

Si.a6

1 3 .l:r.e1

l:tae8

If 1 O . . . dxe4?! there is the strong positional pawn sacrifice 1 1 f6! gxf6 1 2 t:Llxe4. 1 1 exf5 A quieter alternative is 1 1 t:Llxf5 dxe4 1 2 t:Llxe7+ with roughly equal chances. 11 . . .

f6

1 2 c3

Si.a6

1 2 . . . t:Lle5!? 13 d4 cxd4 1 4 cxd4 t:Ll5c6 looks tempting, but White has the strong manoeu­ vre 1 5 t:Llb 1 ! and 1 6 t:Llc3. 1 3 lLldf3

14 .\tel ! 1 4 tZ:lg5 suggests itself, but after 1 4 . . . �d7! there does not appear to be any favourable combination.

'ifd7

1 4 g4

'ifd6?!

1 5 I::!.f2

.l:i.ae8

1 42

�

King's I ndian Attack (from White's Point of View)

1 6 ..ih3!

1 1 . . . e5 comes into consideration.

Now Black has to reckon with g4-g5, but my main idea was to gain control of the f4-point after lt:Jg2 . 16 . . .

h6

1 7 lt:Jg2

d4

1 8 c4

..ib7

19 lt:Jf4 White's position is preferable. However, his opening strategy was rather risky and it probably requires refinement. Fischer

-

lvkov

Santa Monica 1 966

1 2 c3

lt:Ja5?!

White also stands better in the event of 1 2 . . . ..tc5+ 1 3 'iti>h 1 e5 1 4 f5 lt:Jc8 ( Dvoretsky­ Mikhalchishin, Tbilisi 1 980). Possibly he should play 1 5 'ii'h5, intending g3-g4-g5 (but, of course, not immediately - because of the reply . . . ..tf2 ). 13 e5

..tc5+

14 'iti>h1

lt:Jd5

If 1 4 .. Jlc8 White has the strong reply 15 b4! cxb3 1 6 axb3. 1 5 lt:Je4

ii.b7

1 6 'ii' h 5

tt:Je7 ..ixe4?!

8. . .

b6

9 f4

1 7 g4

dxe4

1 8 ..txe4

g6

1 0 dxe4

..ia6

1 9 'ii' h 6

lt:Jd5

1 1 l:!.e1

Black also has a bad position after 19 . . 'iti'h8 20 lt:Jf3. .

20 f5

l:!.e8

21 fxg6

fxg6

22 lt:Jxg6!

'ii'd 7

23 lt:Jf4

.:tad8

24 tt:'lh5

'iti>h8

25 lt:Jf6

lt:Jxf6

26 exf6

l:!.g8

27 ..tf4

l:!.xg4

28 l:!.ad1

l:!.dg8

29 f7 ! Black resigned. 11 . . .

c4

1 1 . . . ii.c7 has also occurred: 1 2 c3 ii.d3 ( 1 2 . . . 'iii'd7?! 1 3 "ii'h5 .l:i.ad8 1 4 e5 f5? 1 5 exf6 .:Xf6 1 6 lt:Je4 .l:!.h6 1 7 'ifxh6! and White won, Lerner-Dolmatov, Kharkov 1 975) 1 3 e5. The game Dolmatov-Lautier (Polanica Zdroj 1 991 ) continued 1 3 . . . 'i!i'd7?! ( 1 3 . . . b5 fol­ lowed by . . . c5-c4, . . . ii.b6+ and . . . lt:Jd5 was stronger) 1 4 lt:Je4 .Uad8 1 5 'ilig4 ..txe4 ?! 1 6 ..ixe4 lt:Jg6 1 7 tt:'lf3 lt:Jce7 1 8 ii.c2! with advantage to White.

I V. B l a c k l e aves h i s pawn on c7

1 e4

e6

2 d3

d5

3 lt:Jd2

lt:Jf6

4 lt:Jgf3

lt:Jc6

1 43

King's I ndian Attack (from White's Point of View)

-

Dvoretsky - E k Wijk aan Zee 1 975 5 c3 Black intends to play . . . e6-e5 and it is quite probable that at some point he will ex­ change pawns on e4 . In this case the bishop will be not too well placed on g2 . Therefore it makes sense to develop it on the f1-a6 diagonal. What essentially results is a Philidor Defence with reversed colours and two extra tempi for White. 5...

e5

position afte r

11 . . . 1 2 h3 13 a3 14 .i.b2 1 5 exd5 1 6 c4 1 7 g3 1 8 lt:Jxe5 1 9 l:txe5

10 . . . l:.e8 -

..ig4 ..ih5 ..td6?1 li:Jb8? lt:Jxd5 li:Jf4 lt:Je6 .i.xe5

And White won.

5 . . . a5 is more accurate.

Dvoretsky - Orlov

6 ..ie2 In such positions it is worth making the useful move b2-b4!. 6. . .

a5

7 0-0

il.. e 7

8 l:te1

0-0

9 'ii'c 2

h6

1 0 ..if1

l:.e8 (see diagram)

1 1 b3 One of the typical plans in such positions is b2-b3, a2-a3, ..ib2 and b3-b4 , as a result of which White gains space on the queenside and creates potential threats to the e5pawn. In reply Black did not manage to find a good arrangement of his forces.

Moscow 1 984 (time control: 30 minutes for the game) 5 c3

dxe4

6 dxe4

.i.c5

1 44

cJif

King's I ndian Attack (from White's Point of View)

7 ii.b5!

ii.d7

4 0-0

lbc6

8 0-0

0-0

5 d3

e5

6 lZ'lbd2

ii.e7

9 'ife2 9 b4 !? ii.b6 1 0 'ife2.

7 e4

0-0

8 c3

dxe4?

9 . . .

a6

1 0 ii.d3

e5

11 b4

ii.a7

1 2 lZ'lc4

.l:te8

1 3 ii.g5

h6

9 dxe4

'iic 7

14 ii.h4

ii.g4

1 0 'ii'c 2

.l:te8

1 5 .l:f.ad 1

'iVe7

1 1 .Ue1

1 6 h3

ii.h5

17 a4

'ili'e6

1 8 lZ'le3

g5

1 9 ii.g3

g4

1 9...ii.xe3 was preferable. 20 hxg4

lbxg4

21 lZ'ld5

.l:tac8

22 ii.c4

'ii'g 6

An unfortunate exchange. Now White ac­ quires a clear plan of play against the weaknesses at c4 , d5 and f5.

I did not play 1 1 lbc4 , to avoid suggesting to the opponent the correct arrangement of his forces: 1 1 ...ii.e6 1 2 lbe3 h6. 11 . . .

ii.f8

1 2 lZ'lf1

g6?

White's next move should have been pre­ vented by 1 2...h6. 1 3 ii.g5!

i.g7

14 ii.xf6

23 .l:td3 Not 23 lZ'lh4? lZ'lxf2. 23 . . .

lbe7

24 lZ'lh4

lbxf2?

The decisive error. But the advantage is also with White after 24...'ili'g5 25 lbxe7+ ltxe7 26 'ili'c2. 25 lZ'lxe7+

.U.xe7

26 ii.xf2

ii.xe2

White wants to seize control of the d5-point and so he exchanges one of its defenders ­ the knight on f6. Of course, he could have first played his knight to e3. 14 . . .

ii.xf6

1 5 lZ'le3

i.e6

27 lZ'lxg6 Black resigned. V. E x p l o i t i n g the wea k n e s s of t h e l i g ht s q u a res

Dvoretsky - Rogozhnikov USSR Schoolboys Championship, Moscow 1 965 1 g3

lbf6

2 ii.g2

d5

3 lZ'lf3

c5

1 6 ii.f1 !

1 45

King's I ndian Attack (from White's Point of View)

And now he exchanges another defender of the d5-point - Black's light-square bishop.

29 h4

�e7

30 it'd3 !

b4

16 . . .

a6

1 7 i.c4

l1ad8

31 tt:Jg5

bxc3

1 8 l::t a d1

b5

32 bxc3

tLld8?

19 .idS!

tt:Je7

33 "ii b 5?

20 i.xe6! Transformation of an advantage: White gives up his central point, but spoils the opponent's pawn structure. With the inter­ mediate move 1 9 ii.d5! he lured the enemy knight to the unfortunate square e7, from where it will have to return with loss of tempo to c6. Weaker was 20 c4?! �xd5 21 tt:Jxd5 tt:Jxd5 22 exd5 �g7 followed by . . . f7-f5 and . . . e5e4, or 22 cxd5 c4, intending . . . �f6-e7c5(d6). 20 . . .

fxe6

21 l:.xd8

l:.xd8

22 �d1 The exchange of rooks makes it harder for Black to defend the doubled pawns and widens the scope of the white queen. 22 . . .

tt:Jc6

23 a4! Now the queenside pawns also become vulnerable. 23 . . .

l:txd 1 +

24 11Vxd 1

�f7

25 tt:Jg4

h5

26 tt:Jxf6! Another transformation of advantage. White exchanges the opponent's bad bishop, but in return he further weakens Black's central pawns and also the dark squares on the kingside, where an invasion will be threat­ ened .

If 30 . c4 there follows 31 'ir'e3 and 32 'ii'b 6. .

.

Carried away by his plan of "iib5 and 'ii'xc5!, White overlooked the possibility of 33 'i\Vxd8!. 33 . . .

tt:Jf7

34 'i\Vxc5!

'ii'd 7

35 "iic 7! And soon Black resigned.

Dvoretsky - Kupreich i k USSR Championship, First League, Odessa 1 974 1 e4

c5

2 tt:Jf3

e6

3 d3

d5

4 tt:Jbd2

tt:Jc6

5 g3

dxe4?!

6 dxe4

b6

If 7 �g2 there is the strong reply 7 . . . �a6, and therefore I develop my bishop on the f1a6 diagonal. 7 i.b5!

i.d7

The pawn structure arising after 7 . . . i.b7 8 tt:Je5!? 'i/c7 9 tt:Jxc6 �xc6 1 0 'i/e2 seems to me to be favourable for White, although, of course, here Black is very close to equality. 8 'ii'e 2! It is important to prevent . . . a7-a6. Now if 8 . . . a6?! there follows 9 ..txa6 tt:Jb4 1 0 ii.d3 tt:Jxa2 1 1 tt:Je5! with advantage to White, only not 1 1 0-0? in view of 1 1 . . . tt:Jc3!. 8. . .

tt:Jf6

9 c3

�e7

26 . . .

�xf6

27 axb5

axb5

1 0 0-0

0-0

28 'ir'd2

'ito>g7

1 1 a4

'ii'c 7

1 46

�

King's I ndian Attack (from White's Point of View)

by 2 1 . . . tt:lxc4 22 'it'xc4 l:tfd8 23 l:ted 1 i.f6, intending . . . 'iic6. 22 axb5

axb5

23 .1i.xb5

c4?!

24 l:ta4! Black loses material, without gaining any compensation. 24 . . . .l:!.fb8 25 .1i.xc4 tt:Jxc4 26 .l:!.xc4 'i'b6?! 27 tt:lxe5 l:tb7 (27 . . . 'i!Yxb2 28 l:tc8+!) 28 b4 .1i.f8 29 .l:tc6 'it'd8 28 b4 .1i.f8 29 l:tc6 'i!i'd8 30 'it'c4 'it'd2 31 tt:lf3 'ii'c 2 32 e5 l:ta2 33 .U.f1 l:.b2 34 l1c8 Black resigned. 1 2 .l::t e 1 12 tt:lc4? a6 13 i.f4 'it'b7 is weaker, but 1 2 e5 tt:ld5 1 3 i.d3 came into consideration. 12 . . .

e5

Black is afraid of e4-e5, and so he weakens his d5-point. White's subsequent plan is roughly the same as in the previous game. 1 3 tt:lf1

a6!?

14 i.d3

V I . B l a c k c h ooses a S i c i l i a n set-up - h e d o e s n ot play . . . d 7-d 5

1 e4

c5

2 tt:lf3

e6

3 d3

tt:lc6

4 g3

g6

14 .1i.xa6 tt:lxe4 is unclear. 14 . . .

tt:la5

If 14 . . . h6, then 1 5 tt:le3 followed by 1 6 tt:lf5. 1 5 .1i.g5

tt:lb3?

16 i.xf6!

i.xf6

1 7 l:ta3

.1i.e6

After 17 . . . tt:la5 1 8 tt:le3 .1i.e6 1 9 i.xa6 c4 20 .1i.b5 Black has no compensation for the lost pawn. 1 8 tt:le3 18 .11Lxa6?! c4 1 9 .I1Lb5 .1i.e7 was less clear. 18 . . .

.1i.e7

1 9 tt:lc4! Stronger than 19 .11Lc4 tt:la5.

Black intends . . . i.g7, . . . tt:lge7, . . . 0-0 , . d7d6, . . . �b8 and . . . b7-b5. In my view, this is one of the best systems of defence. If now White develops his knight on c3, it trans­ poses into the Closed Variation of the Sicilian Defence. Here we will examine other plans for White. . .

19 . . .

.1i.xc4

20 .11L xc4

tt:la5

21 �aa1

b5?

An unsuccessful attempt to confuse mat­ ters. Black should have patiently defended

ctJ

King's I ndian Attack (from White's Point of View)

1) 5 d4 variation Dvoretsky - F i l i powicz Varna 1 980 5 d4!?

cxd4

6 tt:Jxd4 The loss of a tempo (d2-d3-d4) is not as pointless as it may appear at first sight. Black has to reckon with tt:Jb5, and, in addition, after the development of his bishop at g7 the d6-square will come weak. Here the d-file is not blocked for White by his bishop on d3, as in a line of the Paulsen Variation, in which a similar situation arises (1 e4 c5 2 lLlf3 e6 3 d4 cxd4 4 tt:Jxd4 a6 5 1d3 g6!? ). 6...

a6

7 i.. g 2 7 c4!? i.. g7 8 i..e3 also comes into consideration.

7. . .

i.. g 7

8 tt:Jxc6

bxc6

The endgame after 8 . . . dxc6 9 ii'xd8+ Wxd8 1 0 lt:Jd2 is obviously in favour of White, since the opponent's dark squares on the queen­ side are weak. 9 0-0

tt:Je7

If 9 . . . d5, then 1 0 exd5 cxd5 1 1 c4 and 1 2 t/Jc3.

1 47

After 1 0.. . 0-0 White consolidates his advan­ tage by 1 1 tt:Jc3! (less accurate is 1 1 c4? ! a5 followed by . . . i.. a6) . The best chances of a successful defence are promised by Mikhail Tal's recommendation of 1 0 . . . i.. b7, intend­ ing . . . tt:Jc8 . 1 1 i.. d 2 ! ? Nothing i s given b y 1 1 tt:Jd2? ! i.e5 1 2 'it'd3 d5 1 3 tt:Jb3 'it'b5, but it was possible to play either 1 1 tt:Ja3!? i.. e5 1 2 'ifd 1 !? ( 1 2 'ii'd3? ! d5 1 3 i.d2 'it'c7 is weaker) 1 2 . . . d5 1 3 exd5 exd5 ( 1 3 . . . cxd5? 1 4 tLlc4 ! ) 1 4 .l:!.e1 , or 1 2 i.d2 ! i.xd6 1 3 i.. xa5 with advantage in the endgame. 11 . . .

'it'e5

1 2 'iix e5

.txe5

1 3 i.. c 3

.txc3

If 1 3 .. .f6, then 1 4 f4 is not bad, but 1 4 .txe5 fxe5 1 5 tt:Jd2 , with the extremely unpleasant threat of 1 6 tt:Jc4 , is even better. 1 4 tt:Jxc3

d5

1 5 tt:Ja4

.l::. b 8

1 6 b3 1 7 f4

0�

1 7 .l::tfd 1 , intending 1 8 .U.ac1 and 1 9 c4 with an advantage, was evidently even stronger. 17 . . .

.l:!.d8?

This makes it easier for White to play c2-c4 . 1 8 .l:!.fd 1

h6

1 9 c4

g5

19 . . . d4 20 .l:!.d3 f6 21 e5 or 20 . . . g5 2 1 .l:!.ad 1 gxf4 22 gxf4 tt:Jg6 23 .l:!.xd4 .l::t xd4 24 �xd4 tt:Jxf4 25 .l:!.d8+ was bad for Black. 20 .l:!.d4!

gxf4

21 gxf4

tt:Jg6 exd5

22 exd5 23 cxd5 24 tt:Jc3 ! 25 .l:!.ad 1

cxd5 ii.e6 'it>g7

26 tt:Jxd5

i.. x d5

27 i.. x d5 1 0 'i!Vd6!

'it'a5

White is a healthy pawn to the good.

1 48

�

King's I ndian Attack (from White's Point of View)

Dvoretsky - Chubinsky Philadelphia 1 990 5 d4 6 lt:'lxd4 7 �g2 8 0-0

cxd4 a6 'it'c7! ? �g7

Now if 9 lt:'lxc6 Black replies 9 . . . dxc6!. This means that White must aim to set up a strong pawn-piece centre. 9 �e3 1 0 c4! 11 lt:'lc3

lt:'lge7 0-0

18 'ii' b 6

:ac8

1 9 �h3

l::t x c1

20 l:!.xc1

llc8

21 b3!

l:!.xc1 +

22 �xc1 Black's position is strategically hopeless. 22 . . . lt:'lc8 23 'ife3 lt:'le7 24 'it'c3 c.t>f7 25 a4 'ii'd 8? (25 . . . �f6 was a tougher defence) 26 'ii'f3 'it'd7 27 g4! 'ili'c7 28 i.a3 f4 29 g5 h5 30 gxh6 �xh6 3 1 'i¥g4 WeB 32 'ii'e 6 'ii'c 2 33 i.xd6 Black resigned. Dvoretsky - F i l i powicz Polanica Zdroj 1 973 5 d4 6 dxc5

�g7?! b6

An interesting idea. After 6 .. .'�i'a5+ 7 c3 'ii'xc5 8 �e3 Black would stand worse. 7 cxb6

'ii'x b6

8 lt:'lbd2

d5

9 i.d3

lt:'lge 7

1 0 c3

0-0 a5

1 1 0-0 1 2 a4 11 . . .

d6

In the event of 1 1 . . . lt:'lxd4 1 2 �xd4 'ii'xc4 1 3 �xg7 Wxg7 1 4 �c 1 (or 1 4 'ii'd6) White has more than sufficient compensation for the sacrificed pawn. 1 2 �c1 13 i.xd4

lt:'lxd4?! e5?!

13 . . . i..xd4 14 'it'xd4 e5 really was better. 1 4 �e3 1 5 lt:'ld5 16 cxd5

�e6 i.xd5

White has gained a great positional advan­ tage. If 1 6 . . . 'i¥a5 he was intending 1 7 'ii'd2! 'it'xa2 1 8 'ii'b4 or 1 7 . . . 'ii'xd2 1 8 i.xd2 followed by 1 9 �b4 . 16 . . .

'it'd7

1 7 'ii' b 3

f5?!

The opponent has gained definite although hardly sufficient compensation for the sacri­ ficed pawn. 12 . . .

l:.b8

1 2 . . . e5 came into consideration.

ctJ

King's I ndian Attack (from White's Point of View)

1 3 l:ta3 !

.ib7

1 4 .l::t b 3

it'a7

1 5 'iie 2

.l:tfd8

1 49

It made sense to place this rook on e8 , in order to prepare . . . e6-e5. 1 6 l:te1

ita8

1 7 h4! The active possibilities for both sides are restricted, and so White makes a semi­ waiting pawn move, which in some cases may nevertheless come in useful. 17 . . .

d4?!

1 8 lbc4!

.ia6

1 9 l:txb8

l:ixb8

20 itc2

'ii b 7

2) Bishop sortie to g5 Veselovsky - Dvoretsky

21 ..if1 !

dxc3

Moscow Championship 1 973

22 bxc3

.ixc4

5 .ig5 ! ?

22 . .'ii b3? is unsuitable because of 23 'ilt'xb3 .U.xb3 24 l2Jxa5 (or 24 l2Jcd2 ) . .

23 .ixc4

.l:Ic8

24 .id2

l2Je5!

25 l2Jxe5

.ixe5

26 .ib5

'ii'c 7 ! ?

27 l:te3 27 'iitg2!? is also strong.

27 . . .

'ikc7

6 .ig2

.ig7

7 lbc3

lbge 7

8 'ifd2

h6

Black avoids the exchange of bishops: 8 0-0 9 .ih6 . 9 ..ie3

l2Jd4

1 0 0-0

d6

1 1 lbe1

.id7

. . .

lbc6

28 'ii'a 2 White has retained the advantage. I think that quite a good reaction to d3-d4 is a positional pawn sacrifice.

5 d4

cxd4

6 l2Jxd4

.ig7!

7 lbb5

d5

8 exd5

exd5

9 'iix d5

-.e7+!

10 ..ie2

.ig4!

The right way - it is important to prevent the simplification of the position by 1 1 'ii'd6.

Draw. A position that is hard to evaluate has arisen.

1 50

�

King's I ndian Attack (from White's Point of View)

3) Plan with c2-c3 and d3-d4

1 2 lDc3 1 3 ..tf4

Fischer - Pan no

14 .l:!.c1

Buenos Aires 1 970

1 5 b3! 1 6 lDe2

5 .ig2

.ig7

6 0-0

lDge7

7 .l:!.e1

d6

If 7 . . . 0-0, then 8 e5 is unpleasant. 8 c3

0-0

8 . . . e5!? is a sound continuation - see the following game. 9 d4

cxd4

1 0 cxd4

l:tc8 lDa5 b5 b4 .ib5

16 . . . l:txc 1 was more accurate. 1 7 'ii'd 2 1 8 g4

lDac6 aS?

In a cramped position one should aim for exchanges. 1 8 . . . .ixe2! was essential. How­ ever, after 1 9 'ii'xe2 'ii'b6 20 .ie3 lDb8 21 .if1 White would have stood better. 1 9 lDg3 20 h4! 2 1 i.h6 22 'ti'g5 1

'ilfb6 lDb8 liJd7

Threatening 22 .ixg7 �xg7 23 'ii'xe7 or 23 lDh5+ . 22 . . .

.l:!.xc1

23 llxc1

.ixh6

Black also has a difficult position after 23 . . .f6!? 24 exf6 .ixh6 25 'ti'xh6 ltJxf6 26 .ih3. 25 llxc8+

.l:.c8 lDxc8

26 h5!

'ifd8?

24 'ifxh6

10 . . .

d5?!

10 . . . 'ii'b6 11 d5 .ixb2 12 .ixb2 'ii'xb2 is stronger, leading to rather sharp variations, not unfavourable for Black. For example: 1 3 dxc6 'ii'xa 1 1 4 'ii'b3 ( 1 4 cxb7 .ixb7 1 5 'ii'b3 .id5! 1 6 exd5 .:!.ab8 1 7 'ii'd3 liJxd5) 1 4 . . . lDxc6 1 5 lDc3 lDd4 1 6 .l:!.xa1 lDxb3 1 7 axb3 .id7 (Ljubojevic-Hubner, Buenos Aires 1 998) , or 1 3 lDbd2 lDa5 1 4 'ii'a4 'ii'b6 1 5 e5 ltJxd5 1 6 exd6 'ii'd8 1 7 lDe5 b6 1 8 d7 .ib7. 1 1 e5 Now White's chances are preferable. It is not easy to attack his centre: 1 1 . . . lDf5 1 2 lDc3 'ii'b6 1 3 lDa4 , or 1 1 . . .f6 1 2 exf6 .ixf6 1 3 i.h6. 11 . . .

i.d7

It was necessary to play 26 . . . lDf8 27 'ii'f4 (27 'ii'c 1 !? ) 27 . . . Wc7 (28 h6 or 28 lDg5 was threatened), although after 28 .if 1 !? .ixf1 29 �xf1 White would have retained a n obvious advantage. 27 lDg5 28 .ie41 !

lDf8

A spectacular breakthrough of the enemy defences. However, 28 lDxh7!? lDxh7 29 hxg6 fxg6 30 Wxg6+ �h8 3 1 Wxe6 was almost equally strong, or 28 hxg6!? fxg6 (28 . . . hxg6 29 'ifh8+!) 29 lDxh7!. 28 . . .

'ife7

28 . . . dxe4 29 lD3xe4 is totally bad, while if 28 . . . .tea, then 29 hxg6 hxg6 30 liJh5! is decisive.

lZJ

King's Indian Attack (from White's Point of View)

29 lt:'lxh7!

lt:'lxh7

30 hxg6

fxg6

1 2 � b2

1 51

'i!Vd7

30 . . . lt:'lf8 31 g7 leads to a quick mate.

3 1 il.xg6

lt:'Jg5

31 . . . 'it'g7 32 �xh7+ 'it'xh7 33 'ii'xe6+ is also completely hopeless. 32 lt:'lh5

lt:'lf3+

33 '>t>g2

lt:'l h4+

34 '>t>g3

lt:'Jxg6

35 lt:'lf6+!

�f7

36 'i!Vh7+ Black resigned. Fang - Dvoretsky Philadelphia 1 991 (time control: 45 minutes for the game) 5 �g2

�g7

6 0-0

lt:'Jge7

7 l:te1

d6

8 c3

e5! ?

9 lt:'lbd2 The attempt to break through immediately in the centre by 9 il.e3 0-0 1 0 d4 is unsuccess­ ful. Black replies 1 0 . . exd4 1 1 cxd4 , and now either 1 1 . . . il.g4 (although then he has to reckon with the exchange sacrifice 1 2 dxc5), or 11 ... d5!? . .

After 9 a3 0-0 1 0 b4 h6 it is wrong to play 1 1 bxc5? dxc5 1 2 c4 in view of 1 2 . . . f5! 1 3 lt:'lc3 ( 1 3 exf5? e4 ) 1 3 . . .f4 1 4 lt:'ld5 g5 with advantage to Black (Djindjihashvili-Dvo­ retsky, Philadelphia 1 99 1 ), while 1 1 lt:'lbd2 transposes into the game which we are now examining. 9 . . .

0-0

1 0 a3

h6

I n the event of 1 0 . . . a5?! 11 a4 White's position is preferable.

1 1 b4

�e6

A different development scheme was tried in the game Lau-HObner, Munich 1 988: 1 1 . . . b6 12 ii.b2 �b7 1 3 'it'b3 'it'd?.

1 3 d4?! A clever, but dubious pawn sacrifice. How­ ever, also after other continuations Black would not have stood badly. For example: 1 3 'iie2 liae8 1 4 "ilr'f1 f5 ( 1 4 . . . a6!? ) 1 5 b5 lt:'ld8 1 6 d4 exd4 1 7 cxd4 fxe4 1 8 lt:'lxe4 �g4! 1 9 lt:'Jed2 lt:'lf5 (Dominguez-Dvoretsky, Ter­ rassa 1 996) . 1 3 lt:'lb3 b6 1 4 d4 comes into consideration. 13 . . .

exd4

1 4 cxd4

cxb4

1 5 "ikb1 After 1 5 axb4 lt:'lxb4 White would not have time to prevent . . . d6-d5 in view of the threat of 1 6 . . . lt:'ld3. 15 . . .

bxa3

1 6 �xa3

d5

1 7 lt:'le5 !

lt:'Jxe5

1 8 dxe5

d4

1 9 f4

lt:'lc6

This and the following moves were made with the aim of safeguarding the d4-pawn. Probably it should have been disregarded for the sake of rapidly advancing the queenside pawns: 1 9 . . . b5!? 20 lt:'lf3 (not 20 f5? gxf5 21 exf5 �xf5) 20 . . . a5 21 lt:'Jxd4 b4 . 20 lt:'lf3 20 .l:i.f 1 (trying to frighten the opponent with

1 52

�

King's I ndian Attack (from White's Point of View)

the advance f4-f5) 20 . . . ii.h3 21 lLlc4 did not work in view of 2 1 . . . ii.xg2 22 c.t>xg2 b5! 23 lt:ld6 lLlxe5! 24 fxe5 ii.xe5, and the white knight is trapped. 20 . . .

ii.c4

21 .l::t d 1

ii.e2

22 .l:td2

ii.xf3

Of course, not 22 . . . d3?! 23 lLle 1 lLld4? 24 .l:!.axd3. 23 ii.xf3 24 ii.e2

.l:!.fd8 b5?!

Playing on the opponent's time-trouble. After 24 . . . .l:!.ac8 25 'ii'd 1 White would have retained compensation for the sacrificed pawn. 25 ii.xb5

.Uab8

26 ltxd4? Now Black's trappy tactics prove justified. But meanwhile, after 26 'ii'f 1 ! the advantage would have passed to his opponent. 26 . . .

ii'e7?!

Obviously White was hoping for 26 ... ifxd4+? 27 ii.xd4 lt:lxd4 28 .l:!.d3!. I quickly made a move planned beforehand, not noticing the winning continuation 26 . . . 1\Vc7! 27 .Uxd8+ lt:lxd8. 27 l:txd8+

lt:lxd8

28 �d3

'ii'b4

29 lla2

lLlc6

Threatening 30 . . . ii'b6+ followed by . . . lLlb4 . After 30 �h 1 'ii'b3 Black would have retained the initiative, which in time-trouble would be not easy to extinguish. But now came a final oversight. 30 ifc2??

ii'b6+

31 �g2

lLlb4

32 'iic4

lLlxa2

33 ii.d4

'ii'b4

34 ii.xa7

'it'd2+

35 �h3

.l:tb2

White resigned.

Fischer - Durao Olympiad, Havana 1 966 5 ii.g2

ii.g7

6 0-0

lLlge7

7 c3 An attempt to create a strong pawn centre, while saving a tempo on the non-essential move .l:.e1 . But here too with accurate play Black can successfully solve his opening problems. 7 . . .

0-0

In this situation too 7 . . .e5!? is not a bad reply. After 8 ii.e3 it is risky to reply 8 . . . b6 9 d4 exd4 1 0 cxd4 d5 in view of 1 1 exd5 tt:lxd5 1 2 ii.g5. The other standard reaction 8 . . . d6 9 d4 exd4 1 0 cxd4 ii.g4 is stronger: a) 1 1 d5 lLle5 1 2 lt:lbd2 ii'd7? 1 3 lLlxe5! �xe5 1 4 f3 ii.h3 1 5 lLlc4 with advantage (Bologan-Rogozenko, Riga 1 995). I think that Black was obliged to accept the exchange sacrifice: 1 1 . . . ii.xf3! 1 2 �xf3 �xb2 1 3 lLld2 �xa 1 1 4 'ii'xa 1 lt:ld4!. b) 1 1 dxc5 dxc5 ( 1 1 . . . ii.xb2? 1 2 lLlbd2 is much weaker) 1 2 'ii'c 1 0-0 1 3 lLlbd2 ( 1 3 ii.xc5 .Uc8 1 4 lLlc3 ii.xf3 1 5 ii.xe7 ifxe7 1 6 ii.xf3 lt:ld4 with excellent compensation for the sacrificed pawn) 1 3 . . . b6 1 4 h3 ii.e6 with a roughly equal game ( Hug-Hart , Interzonal Tournament, Petropolis 1 973). 8 d4

t2J

King's I ndian Attack (from Wh ite's Point of View)

After 8 . . . cxd4 9 cxd4 it is unfavourable for Black to play 9 . . .'ifb6? (which, I should remind you, worked with the inclusion of the moves .l:!.e 1 and . . . d7-d6) 1 0 d5 �xb2 1 1 ixb2 'i!Vxb2 1 2 dxc6 'ifxa 1 1 3 cxd7 .Ud8 1 4 'lb3 (Bologan) . He continues 9 . . . d5 1 0 e5, and a position from the previously analysed game Fischer-Panna is reached with an extra tempo for Black (there the rook already stood at e 1 ), resulting from the fact that the pawn has advanced to d5 in one move - from d7. Now 1 0 . . .b5? is incorrect in view of the weakening of the c5-square: 1 1 lbbd2 a5 1 2 1Llb3 a4 1 3 ltJc5 'ikb6 1 4 �f4 f6 1 5 .l:!.e 1 with advantage to White (Bologan-Kurz, Biel 1 995). Black's best counter-plan involves pressure on the enemy centre. But if it is implemented inaccurately White still retains the better chances, as shown by two games by Ljubomir Ljubojevic. 10 .. .f6 1 1 .Ue1 fxe5 1 2 dxe5 .ii. d7 1 3 lbc3 .U.c8 1 4 .ii.f4 lbf5 1 5 'ii'd2 lLla5 1 6 b3 b5?? ( 1 6 . . h6 was essential) 1 7 .1l.g5 'ifb6 1 8 g4 and the knight has no move (Ljubojevic­ Tatai, Manila 1 973). .

1 0 ... lt'lf5 11 lLlc3 f6 12 .Ue1 ( 1 2 �f4?! fxe5 1 3 dxe5 allows counterplay that is thematic for this variation: 1 3 . . . h6 1 4 h4 lbxh4! 1 5 1Llxh4 .l:!.xf4! 1 6 gxf4 �xh4 with excellent compensation for the sacrificed exchange)

1 53

1 2 . . . fxe5 (if 1 2 . . . �d7 there is the unpleasant reply 1 3 g4 ) 1 3 dxe5 �d7 1 4 .ii. f4 h6 1 5 h4 .i.e8 (here the same idea 1 5 . . . lt'Jxh4 1 6 lbxh4 �xf4 - if 1 6 . . . g5 White has both 1 7 lbg6, and 1 7 ltJxd5!? - 1 7 gxf4 'iixh4 does not work on account of 1 8 ltJxd5! .l:!.d8 1 9 lbf6+ .i.xf6 2 0 exf6 with a great advantage for White) 1 6 'ii'd2 'iib6?! 1 7 .l:lad 1 ! .l:!.d8 1 8 lba4 ifb5? 1 9 b3 �h7 2 0 .i.f 1 'it'b4 2 1 'it'xb4 lbxb4 22 lLlc5 .i.f7 23 a3!, and White won ( Ljubojevic-Timman, Hilversum 1 973) . Now you will b e able t o properly appreciate the subtle prophylactic move 1 2 . . . �h8!, made in the game Dvoretsky-Kalinin, Wijk aan Zee 1 999. In the tactical variations involving . . . lbxh4 White will not have the counter-stroke ltJxd5!. If 1 3 .i.f4 there follows 1 3 . . . g5 1 4 exf6 .1Lxf6 1 5 .i.e5 ltJxe5 ( 1 5 . . . .i.xe5!?) 1 6 ltJxe5 'ii'b 6. 1 3 h4 .1Ld7 1 4 .1Lf4 (in reply to 1 4 g4 Black can sacrifice a piece: 1 4 . . . lbxh4! 1 5 lbxh4 fxe5) 1 4 . . .fxe5 1 5 dxe5? ltJxh4! 1 6 ltJxh4 g5, and the advantage is already with Black . White should have played 1 5 ..txe5 ltJxe5 (or 1 5 . . . .1l.xe5 1 6 dxe5 iib6) 1 6 ltJxe5 with chances for both sides. After this lengthy theoretical excursion, let us to return to the Fischer-Durao game. 8 . . .

d6?!

Black hopes to exchange on d4 a little later, at a more appropriate moment, but his hopes are not destined to be realised. 9 dxc5!

dxc5

1 0 'ii'e 2

b6

1 1 e5 1 1 .Ud 1 ? is inaccurate in view of 1 1 . . . i.a6. 11 . . .

a5

1 2 �e1

i.a6

1 3 'it'e4

.l:!.a7

14 lLlbd2

.i.d3

1 5 'ii h 4

ltJd5

1 6 'it'xd8

.l:!.xd8

1 7 a4!

.l:!.ad7

1 54

�

King's I ndian Attack (from White's Point of View)

Black's pieces are actively placed, and therefore it may seem that he stands quite well. But in fact in his position there are rather many weak squares, and this factor gives White a great positional advantage. Fischer begins the processing of these weaknesses with the exchange of the light­ square bishops. 1 8 ..tf1 !

..txf1

1 9 'it>xf1

ttJde7

20 ttJc4

ttJc8

21 ..tg5

ttJ6e7

22 tiJfd2

h6

23 ..txe7

l:txe7

24 :a3!

�c7

25 l:.b3

l:tc6

26 ttJe4

..tf8

27 'it>e2

..te7

28 f4

'it>f8

29 g4

'it>e8

30 l:!.f1

l:!.d5

31 l:!.f3

.l:td8

32 l:!.h3

..tf8

6. . .

ttJge7

It is possible, of course, to give the play a quite different character, by choosing 6 . . d5!?, but this is sometimes not to the taste of those who aim with Black for Sicilian positions. .

7 ..te3 7 d4? ! is premature: 7 . . . cxd4 8 cxd4 'i'b6. 7 . . .

d6

7 . . . b6!? was worth trying, and if 8 d4, then not 8 . . . d5? ! 9 dxc5, but either 8 . d6 or 8 . . . cxd4!?. After 8 0-0 0-0 9 d4 in the game Radulov-Taimanov, Interzonal Tournament, Leningrad 1 973, White gained a promising position after 9 . . . cxd4 1 0 ttJxd4 ..tb7 11 ttJxc6 ..txc6 12 'iid6 tLlc8 13 'iid2 'ike? 14 ttJa3, but 9 . . . ..ta6 1 0 l:!.e1 cxd4 11 4:lxd4 ttJe5 came into consideration. . .

33 ttJxa5! Fischer's play creates a highly artistic impression. It is instructive to follow how he exchanged the pieces that he did not need, strengthened to the maximum the placing of his remaining pieces, and finally landed a decisive blow. 33 . . . bxa5 loses to 34 lLlf6+ 'it>e7 35 .l:tb7+ . 33 . . J::t c 7 34 ltJc4 l:!.a7 35 ttJxb6 ttJxb6 36 l:!.xb6 l:!.da8 37 lLlf6+ 'it>d8 38 l:!.c6 .l:tc7 39 .l::!.d 3+ 'it>c8 40 l:!.xc7+ 'it>xc7 41 .l:r.d7+ 'it>c6 42 l:!.xf7 Black resigned. Bukhtin - Dvoretsky Moscow Championship 1 972 5 ..tg2

..tg7

6 c3!? An attempt to begin active play in the centre, by saving another tempo: on castling.

8 0-0 The logical consequence of the move order chosen by White was the immediate 8 d4. If 8 . . . cxd4 there follows 9 ttJxd4 ! 0-0 1 0 0-0, while if 8 . . .'ii' b6 - simply 9 'iid2 . Possibly Black should maintain the tension in the centre by 8 . . . b6!?, but this is already something of a concession compared with the variations analysed earlier. 8 . . . 9 tiJbd2

0-0

Cb

King's Indian Attack (from White's Point of View)

The consequences of 9 d4 have been seen in our analysis of the previous game. White employed a different arrangement of his forces in the game Udov-Dvoretsky, Mos­ cow Championship 1 967: 9 'if'c 1 .U.e8 (avoid­ ing the exchange of the dark-square bish­ ops after 1 0 ..lth6) 1 0 ttJa3 .l:tb8 1 1 .l:td 1 ?! ( 1 1 .th6 ..lth8 1 2 ttJc2 ; 1 1 d4!? ) 1 1 . . . b5 12 ttJc2 b4 1 3 d4 bxc3 1 4 bxc3 'if'a5 ( 1 4 . . . cxd4 15 tt'lcxd4) 1 5 'if'a3 'ii'xa3 1 6 ttJxa3, and now Black should have played 1 6 . . . d5!. 9 . . .

b6

The pawn does not manage to advance to b4, since the variation 9 . . . b5?! 1 0 d4 b4 1 1 dxc5 bxc3 1 2 bxc3 is advantageous to White. However, it made sense to take control of the d4-square by 9 . . . e5!? . The arrangement of Black's forces is the optimal one both with his pawn on e6, and on e5. 1 0 d4

aS!?

11 a4 Changing the character of the play by 11 dxc5 bxc5 1 2 e5!? came into consideration. 11 . . .

..lta6

1 2 l:te1

cxd4

This exchange could have been delayed, by

1 55

choosing 1 2 . . . �c8!? , and if 1 3 dxc5, then not 1 3 . . . dxc5?! 1 4 'if'b3 followed by ..ltf1 or l2Jc4 , but 1 3 . . . bxc5. ttJxd4 1 3 ttJxd4 In the event of 13 . . . 4Je5 White would have neutralised the bishop at a6 by 1 4 tiJb5. 1 4 cxd4 14 ..ltxd4 was preferable, when White's position would have remained more pleas­ ant. 14 . . .

ttJc6

Another possibility was 1 4 . . . l:!.c8 , and if 1 5 'ifb3- either 15 . . . d5, or 1 5 . . . ttJc6 1 6 d5 tiJd4 1 7 ..ltxd4 ..ltxd4 1 8 dxe6 'if'f6!? . 1 5 tiJb1 !

tiJb4

1 6 ..ltf1

..ltxf1

1 7 .l:.xf1

d5

1 8 e5

l:tc8

1 9 ttJa3 1 9 ttJc3!? , intending 20 h4 , looks more natural. Black would have replied 1 9 . . .f6. 19 . . .

'if'd7

20 'if'b3

f6

Draw. The two sides' chances are roughly equal.

1 56

\t> Artu r Yusu pov

P re pa ration for a Game '

D rov liked to repeat, tirelessly urging ifficult in study, easy in battle', Suvo-

logic, such an approach to the solving of opening problems, has a right to exist.

his generals to make practice attacks. This ageless precept of the great commander is also fully applicable to chess. The more difficult and painstaking your preparatory work, the easier and more inspired the game itself will be!

As a rule, players nevertheless do not stick too rigidly to such tactics, preferring occa­ sionally to vary their opening set-ups. Because otherwise it becomes just too easy for their opponents to prepare.

Preparation for a game is a highly individual process. And here much depends on what kind of opening arsenal you have at your disposal. Some players solve this problem fairly simply: they do not prepare for every specific opponent, but make preparations for an entire tournament, deciding before­ hand the range of openings that they are intending to employ. They hope that, if they encounter any surprise in the opening, they will be able to deal with the situation directly at the board. Usually these are players with a narrow, well-developed opening reper­ toire. An example is provided by grandmaster Andrey Sokolov. To predict what opening you will have in a game with him is not at all difficult: with White he exclusively chooses e2-e4, and with Black, say, the Queen's Indian Defence. And that's all! Andrey sticks accurately to his opening repertoire, but he knows it thoroughly, of course, since he has great experience of playing his systems. Sokolov reckons that if an opponent suc­ ceeds in casting doubts on some variation of his, he will quickly be able to repair it; for his next opponent it will be significantly more difficult to find a defect in this variation, and in the end he will have a completely 'fire­ proof' repertoire, which is especially impor­ tant when playing Black. In general, such

As an example I will show a preparation which I made with Mark Dvoretsky at the Candidates Tournament in Montpellier be­ fore my game with the Cuban player Jesus Nogueiras. At that time he had a rather narrow repertoire, and we noticed that in the Queen's Gambit he often employed one and same variation, a very risky one, in my view. Delving into the position, we found a new plan, which I in fact employed in the game. The novelty proved uncommonly effective! It need hardly be said how much our task was eased by the narrowness of the opponent's opening repertoire.

Yusu pov - Nogueiras Candidates Tournament, Montpellier 1 985 Queen 's Gambit

1 d4

d5

2 c4

e6

3 tDc3

c6

4 tDf3

tt)f6

5 i.. g 5

tDbd7

6 cxd5

exd5

7 e3

i.. d 6

It was this set-up that Nogueiras used to employ. 8 i.. d 3

tDf8

Preparation for a Game

1 0 0-0

tD

1 57

.ll x e5

11 dxe5 In our preparations we considered 1 1 . . . tt:J6d7, after which we were intending 1 2 .lii. f4 , and if 1 2 . . . ii'xb2?! 1 3 l:!.c 1 lt:Jg6, then 1 4 .lii. xg6 hxg6 1 5 e4!, opening lines in the centre. Instead of capturing the pawn, Black does better to complete his development with 1 2 . . . tt:Jc5. This is what happened in the game Gulko-Smagin (Moscow Champion­ ship 1 984) . 11 . . . Black wants to play . . . lt:Jg6, . . . h7-h6 and force the exchange on f6. An obvious drawback to his plan is the fact that he has already moved the same piece twice in the opening, and he intends to move it a third time, which clearly violates the principles of development. In addition, 8 . . . tt:Jf8 delays castling. All this encourages White to play actively. 9 tt:Je5 This move suggests itself. Black has lifted his control of e5, and the white knight promptly heads there. In reply to 9 . . . lt:Jg6 there follows 1 0 f4 . The plan involving the occupation of the central e5-point was introduced back in the 1 9th century by the great American player Harry Nelson Pillsbury. Here is a more recent example: the game Chernin-Cvetkovic (Belgrade 1 988) went 1 0 . . . 0-0 1 1 'ilt'c2 ( 1 1 0-0!? ) 1 1 . . . 'ti'e8 1 2 0-0 ib4?! 1 3 'it>h 1 , and soon White launched a decisive offensive. 9. . .

'it'b6

Black wants to exchange the powerful knight on e5 before it is supported by the f­ pawn. With this aim he violates yet another opening principle, by bringing his queen out too early. If in addition he decides to capture with his queen on b2 , his pawn-grabbing may cost him dearly: after 1 0 0-0 'ir'xb2 1 1 lk1 White gains a significant lead in development.

lt:Jg4?!

An unsuccessful reply. An attempt to play actively while insufficiently well-developed cannot be recommended. Nevertheless, in our preparations we considered this move, although, of course, our analysis was rather cursory. At home it had seemed to me that by playing 1 2 .ll f4 lt:Jg6 1 3 il.xg6 hxg6 1 4 h3 lt:Jh6 1 5 e4 White would gain an advantage. But at the board I realised that Black can defend by 1 5 . . . 'ti'xb2 1 6 .Uc 1 0-0 , returning the extra pawn. Delving into the position, I found a stronger continuation. Can you see what is was? No? What if I make the leading suggestion that the position of the knight on g4 must be indirectly exploited? 1 2 'ii'a 4! The extremely unpleasant threat of 13 tt:Jxd5 has been created. If 1 2 . . . .lii. d7, then White can choose between 1 3 'ii'a3!? f6 1 4 exf6 gxf6 1 5 il.h4 and the more camouflaged idea of 1 3 e6!? .lii. xe6 1 4 tt:Jxd5 il.xd5 1 5 'ti'xg4 . In both cases, with a lead in development and the two bishops, his position is strategically won. 12 . . .

ii'xb2

Black, as they say, throws caution to the winds. 1 3 .Uac1 Of course, not 1 3 tt:Jxd5? �xe5. But now it is not possible to capture with the knight on e5, if only because of 1 3 . . . tt:Jxe5 1 4 .l:tc2 'ir'b6 1 5

1 58

�

Preparation for a Game

ltJxd5 ( 1 4 ltJxd5 ltJxd3 1 5 .l:r.xc6 is also strong). 13 . . .

..td7

1 8 ..txb5

ltJe6

1 9 'ili'b2

cxb5

20 ..th4 Black resigned. It goes without saying that this was a severe punishment for his violation of opening principles!

Here I thought for a long time, realising that in such a position there was simply bound to be a forced win. 1 4 .l:r.c2 'ili'b6 1 5 l:.b1 'ili'c7 1 6 ltJxd5 seems to suggest itself, but after 1 6 . . . 'ili'xe5 nothing definite is apparent, while if 1 6 ..tf4 Black can continue the fight with 1 6 . . . ltJe6. And here it suddenly dawned on me! 1 4 iVd4 ! ! A move of murderous strength. With the simple centralisation of his queen White simultaneously creates five (!) threats: 1 5 ltJxd5, 1 5 ltJb5, 1 5 ltJe4 , 1 5 .l:r.b 1 and 1 5 e6 ( 1 5 . . . ..txe6 1 6 'ili'xg7). It is not possible to defend simultaneously against all of these. For example, 1 4 . . . 'ili'b6 parries four of the threats, but the fifth proves decisive - 1 5 e6!. 14 . . .

f6

1 5 exf6

gxf6

1 6 ..txf6

llg8

If Nogueiras had played 1 6 . . . ltJxf6 1 7 'ili'xf6 .l:r.g8 , the opposition of the queens could have been exploited by 1 8 ltJxd5. 1 7 ltJb5 17 .. .'it'xd4 1 8 ltJd6 mate.

'ii'x b5

I thought that I had convincingly refuted the ' Nogueiras variation'. Imagine my surprise when a few years later I suddenly saw the match game Timman-Lj u bojevic (Hilver­ sum 1 987), played with the same variation. True, in it Black lasted only a little longer. Instead of 1 0 . . . ..txe5 he played 1 0 .. .'ifxb2 (Ljubomir Ljubojevic very much likes to have extra pawns) 1 1 .l:r.c1 ltJg6 1 2 f4 ( 1 2 ..txf6!? gxf6 13 ltJg4 ) 12 ... 0-0 13 l:tc2 'i'b6? ( 1 3 . . . 'ii'a3 was better) 14 ..txf6 gxf6 1 5 tt:'lg4 ..txg4 1 6 'ii'x g4 ltt h 8 1 7 l:tb1 'ili'c7 18 ttJxdS 'ii'd 8 1 9 ltJc3 ..txf4 20 .l:te2 ! .l:r.e8 21 tt:'le4. Jan Timman gained a marked advantage, and the incorrect combination made by the opponent merely hastened his demise: 21 . . . ..txe3+?! 22 l:txe3 'ii'x d4 23 .:e1 tt:'leS 24 'ili'f5 , and Black resigned in view of 24 . . . ltJxd3 25 ltJxf6 . Let us return once again to my game with Nogueiras. The question may be asked: if I had prepared the entire variation before­ hand, why in the course of the game did I have to look for a stronger continuation? Well, this is a valid question and one which is worth dwelling on in more detail. Here it all depends on the time spent on the preparation. It is one thing to study some variation at home, in the quiet of your study, so to speak, and quite another when you have to do this during the course of a tournament. It is clear that during immediate preparations for a game (usually lasting a couple of hours) it is hard to take all nuances into account. That was also the case here . We did not study in detail how to play against 1 1 . . . ltJg4 , as our objective was a different one: simply to convince ourselves that after this continuation too the position

ctJ

Preparation for a Game

was favourable for White and rather danger­ ous for Black. The move planned at home, 1 2 .i.f4 , was sufficient to confirm this evaluation. This by no means signifies that it is the strongest. You should not blindly trust your opening preparations. And not only because a mistake may creep in to your preliminary analysis . The main thing is something else: the tension of the struggle, the stressful competitive situation itself sharply accentu­ ates your intuition, strengthens your imagi­ nation, and also raises your calculating ability. This is why, however meticulous your analysis has been, during a game you should check the variations you have found, and seek stronger alternatives . At the board you may hit on some new and unexpected idea! Of course, on moves such as 9 lDe5 and 1 0 f4 (in reply to 9 . . . lDg6) it is not worth spending time, but when there ari ses a position which has not been analysed, but merely estimated, here a serious verification is simply essential. Thus, success in the opening largely de­ pends on your ability to guess what variation the opponent will choose, and successfully prepare for it. This is especially important (but also difficult! ) when meeting a player such as Timman, for example, who has a very broad opening repertoire. But there is no guarantee that, during the couple of hours spent at a tournament preparing for a specific opponent, you will definitely devise the optimal way of combat­ ing the opening set-ups employed by him. Therefore a second significant factor of success is the quality of your preparation before the competition, the depth of your opening erudition, and the breadth of your own opening repertoire. Any experienced player will no doubt easily remember instances when his old prepara­ tions came into action. For example, as in

1 59

my game with the Hungarian grandmaster Zoltan Ribli, which we will now examine. Preparing for a game with Ribli is difficult, since he is one of those players who varies his opening systems. He has quite a wide repertoire and he has a thorough knowledge of the variations he employs . However, at that tournament in Montpellier I somehow managed to guess what he would play, and, in addition, in the variation in question I had an important improvement stored up. So that my entire preparation essentially re­ duced to me simply looking through my notebook and refreshing the lines in my memory! I should mention that in making immediate preparations for a game during the course of an event you should observe a sense of measure - what is the point of spending five hours studying, if after this you arrive for a game with a headache and unable to understand anything? The optimal variation is when all your main analytical work is done beforehand, and during an event you merely remember your analyses , refresh them in your memory. Yusu pov - Ribli Candidates Tournament, Montpellier 1 985 Queen 's Gambit

1 d4

lDf6

2 c4

e6

3 lDf3

d5

4 lDc3

c5

5 cxd5

lDxd5

6 e4

lDxc3

7 bxc3

cxd4

8 cxd4

lDc6

9 .i.c4

b5

All this is well known in theory. 1 0 .i.e2

.i.b4+

11 .i.d2

'ii'a 5

1 2 d5

exd5

1 60

� 1 3 exd5

Preparation for a Game

t:De7

Why did Ribli choose this variation? I think because he was hoping to catch me unawares - after all, this was not his main opening weapon. In addition, I had never played this as White, and this position had not occurred in any of my games. But here he was simply unlucky: how could he have known that three or four years earlier I had made a serious study of the Tarrasch Defence Deferred and had found an im­ provement for White? 1 4 0-0

�xd2

1 5 t:Dxd2

0-0

16 t:Db3

1id8

An interesting pawn structure has arisen. White has a strong passed pawn in the centre, which Black is intending to blockade and possibly attack. 1 7 �f3 1 7 �xb5 has also been tried. In the game Kir. Georgiev-Ribli (Sarajevo 1 985) there followed 1 7 . . . �b7 1 8 t:Dc5 (in the event of 1 8 d6 t:Df5 Black has counter-threats) 1 8 . . . 'ii'b6! 1 9 t:Dxb7 'it'xb5 20 .l:tb 1 'ii'xd5 2 1 .l:te 1 'ii'xd 1 22 �bxd1 t:Dc6 23 .l:td7 t:Db8! 24 .l:tde7 t:Dc6, and the players agreed a draw. 17 . . .

t:Df5

guides, but after 1 8 . . . 'ii'd6!? Black would appear to equalise. However, from the very start it appeared to me that the position is rather more pleasant for White. And the analytical work that I did merely reinforced this opinion. 1 8 .U.c1 ! The fruit of my efforts at home. White develops his rook and simultaneously pre­ vents the reply 1 8 . . .'tid6 (in view of 19 �c6). 18 . . .

t:Dd6

It is well known that the knight is a good blockader, so that Ribli's decision is posi­ tionally justified. He was probably hoping that after the obvious 19 t:Dd4 �d7 20 tt:Jc6 his queen would come out to f6 and the position would become equal: indeed, the knight at d6 is excellently placed and the black rooks will now occupy the e-file. But it turns out that White can prevent the queen from coming out to f6. 1 9 'ii'd 4! An unpleasant continuation for Black. Now 1 9 . . . 1if6 leads to the spoiling of his pawn structure. Even so, this was probably the lesser evil. What Ribli played was less good. 19 . . . 'ii' b 6?! Here the queen stands worse than at f6. It is remote from the kingside and White can develop his initiative there unhindered. It is clear that now he should no longer ex­ change queens. 20 'ii'f4!

�d7

21 t:Dd4

�feB?!

A significant inaccuracy; the rook should have been kept at f8 for the defence of the royal residence. 22 t:Dc6 (see diagram)

Here 1 8 'il¥d3 is analysed in the theoretical

White has an appreciable advantage, largely the result of successful opening prepara­ tion. The position reached was not only familiar to me, but also close to me in style.

CLJ

Preparation for a Game

161

[Already here it was possible to begin the 'gathering of the harvest', by playing 24 :Xe8+!, when 24 . . . 1:1xe8 is not possible because of 25 lDe5. And if 24 . . . �xe8 White decides matters with 25 lDe7+ �f8 26 ilb4 "ikxf2+!? 27 �h 1! or 25. . . �h8 26 ltJc8! "ikd8 2 7 "ikc7! (27 "ikb4!? �d7 28 lDd6) 27 . . . fixc7 28 llxc 7 - Dvoretsky.]

24 . . .

ltJc4

25 h3

h6

26 i.d3! - position after 2 2

l2Jc6 -

Can one expect more from the opening? Perhaps the only instance closer to the 'ideal' is when you know all the moves right to the end - as Lev Polugayevsky some­ times achieved. But in my view such an approach is irrational: it demands an enor­ mous expenditure of energy, and for what ­ in order to win, as a rule, only one single game?! Of course, here we have a different case. But for a 'normal' player, say what you like, this is ideal preparation. Generally speak­ ing, if you find a move such as 1 8 .l:tc1 , you see that the resulting position is slightly better for you and also that it is to your taste - then confidently go in for it!

White has strengthened his position to the maximum, and now it is time to switch to concrete play. I exploit the fact that the double capture on c6 leads to loss of material: 26 . . . �xc6 27 dxc6 l:txe 1 + 28 l:ixe 1 'iixc6 29 �e4 'iie8 30 �h7+ . 26 . . .

ltJb2

What else? 27 �b1 The bishop has no intention of leaving the active diagonal. Black's position is practi­ cally hopeless. 27 . . .

�xc6

28 dxc6

l:txe 1 +

29 .:lxe1

'ii'xc6

I remember that Ribli was upset by this turn of events. Perhaps it was for this reason that subsequently he did not defend in the most tenacious way.

22 . . . 23 l::tfe 1

ltJc4

White probes the weakness of the e7square. 23 . . . 24 �e4

ltJb2

The threat of the knight fork on d3 has been parried, and Ribli has nothing better than to return his knight to its starting point.

Here, as they say, many roads lead to Rome. In particular, 30 "ikb4! is very strong ­ both attacking the knight, and defending the

1 62

�

Preparation for a Game

rook on e1 . But I was attracted by another idea. 30 i.e4! Now 30 . . . �c4 was the most tenacious, but after 31 �d2 l:te8 32 i.h7+ 'l!;>xh7 33 l:!.xe8 White is the exchange up with a winning position. Ribli played differently and . . . fell into a prepared trap. 30

.

.

�c3

.

31 l1c1

tt:Jd3

Ribli was pinning his hopes on this counter­ stroke. Now 32 .Uxc3? is incorrect: 32 . . . tt:Jxf4 33 l:tf3 tt:Je2+ (the same check would have followed after 33 i.xa8) 34 'i!;>f 1 l:!.e8 35 l:te3 tt:Jd4 36 i.h7+ 'i!;>f8 , and Black remains a pawn up. 32 �xf7+ ! ! Loss of material becomes inevitable. Black resigned. The following example is practically on the same theme, with the significance differ­ ence that, even with the best will in the world, I simply could not have guessed my opponent's choice of opening: Kevin Sprag­ gett employed the Tarrasch Defence for the first time in his life! The Canadian grand­ master prepared very thoroughly for our Candidates match, and the Tarrasch De­ fence was in fact one of his opening surprises. But he guessed wrongly! And in two respects: firstly, the prepared variation was one that I had analysed well, and secondly, it corresponded more with my style of play that with his own. After suffering a disaster, Spraggett did not persist and in the match he didn't employ the Tarrasch Defence again. Yusu pov - Spraggett 3rd match game, Quebec 1 989 Queen 's Gambit

1 d4

d5

2 tt:Jf3

c5

3 c4

e6

4 cxd5

exd5

5 tt:Jc3

tt:Jc6

6 g3

tt:Jf6

7 i.g2

i.e7

8 0-0

0-0

9 i.g5

i.e6

The usual 9 . . . cxd4 1 0 tt:Jxd4 h6 leads to more complicated play. 1 0 dxc5

i.xc5

11 i.xf6

'ifxf6

1 2 tt:Jxd5

'i!t'xb2

1 3 tt:Jc7

.Uad8

1 4 �c1

�xc1

1 5 .Uaxc1 This variation occurred in the 1 6th game of the Petrosian-Spassky World Champion­ ship Match (Moscow 1 969), and then for many years it went out of favour. But in 1 988 , at the tournament in Linares, the Spanish grandmaster (then still an interna­ tional master) Miguel lllescas employed it against Alexander Beliavsky and equalised . It was evidently this game that Spraggett was guided by. What oversight, you may ask, did he make in his choice of opening? Spraggett did not take account of the fact that I had also participated in that tournament and, natu­ rally, I had also paid attention to the afore­ mentioned game, since two rounds later I was due to play White against lllescas. These are the subtleties which sometimes have to be taken into account during preparations! What is important is not even whether a particular variation occurred in the opponent's games, but whether he was there, where this variation was employed! Spraggett effectively fell victim to my prepa­ ration for the game with lllescas (in which, incidentally, another variety of the Tarrasch Defence occurred). However, it so hap­ pened that I had first studied the given

ltJ

Preparation for a Game

position long before the tournament in Linares. 15 . . .

il... e 7

Spassky preferred 15 . . . b6 against Petrosian, but after 1 6 lt:Jxe6 fxe6 1 7 il... h3 or 1 7 e3 he would have encountered difficulties (in the game Petrosian played less accurately - 1 7 I!c4). 1 6 lt:Jxe6

fxe6

1 7 l::t c4

il... f6

1 63

1 8 e3! Realising that an exchange of rooks is un­ favourable for him, White does not hurry with the move l:tb 1 . Besides, it is not yet clear whether it will subsequently be required (the rook may prove useful on c1 ). The move made by me, which I had prepared for the game with lllescas, proved to be a theoreti­ cal novelty, although what, it would appear, could be more logical? - White takes control of the d4-square, on which exchanges undesirable for him could have occurred. 18 . . .

l:td6

1 9 h4

h6

20 .l:!.e4

l:tfd8

21 il... h 3

At this point 1 8 l::tb 1 was usually played. Here are a few examples, showing that Spraggett had serious grounds for thinking that he would be able to make an accurate draw: 1 8 . . . l::td 7 19 h4 li:Jd4!? , and Black advanta­ geously simplifies the position ( Ftacnik­ Minev, Bucharest 1 978 ); 1 8 .l:!.d6 (the b7-pawn is indirectly de­ fended) 1 9 h4 h6 20 .l:!.e4 b6 2 1 il... h3 Wf? 22 J:!.c1 (22 e3 l:tc8 23 g4 g5! also leads to equality, as in the afore-mentioned Beliavs­ ky-lllescas game) 22 . . . .U.e8 23 e3 l:te7 24 i.f1 lt:Ja5!, and Black does not experience any difficulties (Ornstein-Schneider, Copen­ hagen 1 98 1 ) . . . .

Incidentally, in the Encyclopaedia o f Chess Openings, none other than Garry Kasparov judged the given opening variation to be sufficient for equality.

I prevent the exchange of a pair of rooks. The attempt to insist on this by 21 . . . e5 would have had serious consequences for Black: 22 Wg2 lid1 23 l::txd1 l:txd1 , and the white bishop gains the opportunity to attack the b7-pawn from c8 (then the knight on c6 will also be attacked), whereas the black bishop on f6 is bad. Naturally, such a prospect did not appeal to Spraggett, and he prefers to defend without making any unnecessary weakening. 21 . . .

'it>f7

22 'it>g2

.l:!.e8

Here the fact that White did not hurry with .Ub 1 again came in useful. Delving into the position, I realised that there was altogether nothing for the rook to do on b1 . Moreover, I came to the conclusion that White's main objective now was the activation of his knight, for which I made a concrete plan. Let's ponder over this together and try to decide - what plan? (see diagram)

It is clear that the only square via which the knight can come into play, without fear of being exchanged, is d2 . But how can I take control of it? That's right, by defending it with

1 64

�

Preparation for a Game

25 .l:!.f4! Creating the tactical threat of 26 J:txc6 �c6 27 tt:Je5+ , while if 25 . . . 'it>g8 , then 26 4Jd2! followed by tt:Je4 is very strong. 25 . . .

'it>g6

26 g4!

- position after

22

..

J:te8 -

a rook. So that the first stage of the plan is clear - the manoeuvre of the rook from f1 to c2 . And when the knight arrives on c4, not only do concrete threats appear, but it also becomes possible to advance the kingside pawns, f2-f4 for example - a good way of improving the position. I think that already here one can draw conclusions about Spraggett's preparations for the game with me, and my preparations for the game . . . with lllescas. Black has no prospects, whereas White has obtained the sort of position that he wanted - he has a slight but enduring advantage, and a plan for strengthening his game. I have to admit that during the game I rather forgot my previous analysis. But unexpect­ edly this proved opportune: at the board the need arose to think seriously about the position, as though to evaluate with a fresh glance, and calculate the variations anew. So that even the opponent's somewhat unexpected choice of opening proved to be a factor in my favour. 23 �c1

.l::i. e7

24 .l:!.c2

b6?

An imperceptible, but significant mistake Spraggett weakens the position of his knight. White immediately exploits this fac­ tor.

It transpires that if 26 . . . tt:Je5 there is the unpleasant reply 27 g5, for example: 27 . . . hxg5 (it is better to play immediately 27 . . . lt:Jd3 28 h5+! 'it>f7 29 gxf6 tt:Jxf4+ 30 exf4 gxf6) 28 hxg5 lt:Jd3 29 gxf6 tt:Jxf4+ 30 exf4, and if 30 . . . gxf6 31 f5+! with an overwhelm­ ing advantage. The simple 27 tt:Jxe5+ �xe5 28 l:tf8 is also strong - the black king is in an anxious position. 26 . . .

�a1

The only move. 27 .Uc1

il.b2

28 .l::i. c2

il.a1

The repetition of moves has allowed me to gain time for thought. For the moment I could not see any decisive strengthening of the position: to 29 h5+ 'it>h7 30 g5 Black replies 30 . . . g6, and if 29 .l:!.fc4?!, then simply 29 . . . tt:Je5. Therefore I decided to make a neutral move (at the same time slightly strengthening my position) , inviting the opponent to guess what in fact White was contemplating. 29 a4! It is not easy for Black to defend, especially when short of time, as Spraggett was. 29 . . .

tt:Je5?!

30 tt:Jxe5+

�xeS

31 �f8 ! In this version, a position with opposite­ colour bishops suited me fine. Here it is obvious that the black king will come under a mating attack. 31 . . .

.Udd7

3 1 . . . 'it>h7 was more tenacious. 32 f4

Preparation for a Game

A strong reply. Now 32... i.e?? is bad on account of the decisive breakthrough 33 fS+ exf5 34 gxf5+ 'it>h7 35 f6!.

32 . . .

l:tc7

33 l:td2

i.c3

34 l:!d6

'it>h7

35 g5

hxg5

36 hxg5

i.b4

37 .:tdd8?! Unfortunately, I did not notice a spectacular blow, which would have concluded the game immediately: 37 g6+! 'it>h6 (37 . . . 'it>xg6 38 i.fS+ ) 38 i.fS!. 37 . . .

'it>g6

If 37 . g6 the most energetic is 38 J:Ih8+ 'it>g7 39 l:.h6 with the terrible threat of 40 .l:.dh8 . . .

38 'it>f3

l:tf7

39 l:th8

e5

40 i.g4 Of course, 40 i.e6 would have won the exchange, but White wants to weave a mating net. And he succeeds in doing this. 40 . . .

exf4

41 .l:td5!

fxe3+

42 'it>g3 Black resigned. This is how, in a rather strange way - thanks to preparation for an altogether different

lLJ

1 65

game - I managed to win this important encounter. An instance which once again reminds us that serious analytical work is never wasted - of course, if the results are recorded, comprehended, and lodged in your memory. As a rule, in chess, work done for future use sooner or later (it may even be many years after) justifies itself. I must once again emphasise that, in my view, the main preparatory work should be carried out beforehand. Not at competitions, but between them! I remember a conversa­ tion with Korchnoi at the Tilburg tournament of 1 987. Complaining about his not very successful play, Victor Lvovich said that, un­ fortunately, he had not had time to prepare properly for the tournament and had arrived at it without any fresh ideas. This had upset him. Indeed, before an important event it is very important to have something in mind, to produce a definite reserve of new ideas without this it is hard to count on success. Young, not too skilful players, sometimes try to build their preparations on the study of less explored, so to speak, side variations. In principle, this is normal for a player who has not yet acquired a broad opening repertoire, who on account of his youth has simply not had time to assimilate the avalanche of theory which currently engulfs the professional player. However, there is no point in transforming the avoidance of theoretical continuations into an end in itself, by deliberately choosing not the best lines, and basing your play on various types of 'crooked', trappy moves - such strategy is incorrect. What course, then, is the most advisable? My position is simple: in the opening you should aim to make objectively the best moves, even if in your preparations this demands a significantly greater volume of work and a more detailed analysis. And in any case, that which you play must be

1 66

�

Preparation for a Game

thoroughly studied, and you should sense all the nuances. Summing up what has been said, let us formulate the two main principles for the choice of opening when preparing for a specific opponent. The first. You should proceed from your own possibilities, i. e. aim to obtain a position which you yourself know well and which corresponds to your chess tastes and style of play. Since, if you have a leaning towards positional actions, and as a result of your preparation you obtain gambit play with wild complications, you risk seeing your hopes dashed, despite the most conscientious preparatory work. Therefore experienced grandmasters sometimes re­ ject even promising continuations, if they do not correspond with their style. It is hard, you will agree, to expect from Garry Kasparov that, even playing Black, he will go in for a passive position. It is simply not in his nature! See how he played in his matches against Anatoly Karpov: deliberately avoid­ ing passive positions, he preferred to give up a pawn and make a draw in a compli­ cated struggle, rather that, without sacrific­ ing anything, make the same draw by accurate, defensive play. And by contrast, see how Karpov prepared. You will see that in the opening he did everything possible to avoid unnecessary complications, and went in for them only when he was sure that he had prepared a genuinely powerful, promis­ ing continuation, when such an evaluation was supported by deep analysis. Thus, the main objective of your preparation is to obtain a game which is comfortable for you. The second objective is perhaps more subtle try to lure the opponent into positions which are least in keeping with his style, and do not correspond to his chess tastes. In this case the probability -

of mistakes by him is sharply increased. Remember, for example, Gata Kamsky's failure at the Linares tournament of 1 99 1 . It is largely explained by the fact that his opponents quickly discovered defects in his opening repertoire, and they easily dragged Kamsky into positions which were unfamiliar to him (simply on account of his youth and lack of experience). Right through the tournament he suffered terribly, especially as Black, with which he lost all his games! As a 'positive' example I will describe my game with the English grandmaster Jonathan Speelman from the same tournament. I will not show it on the board, but will simply inform you of the problems which I encoun ­ tered in my preparations. This was the game from the first round, and it can happen that your mood for the entire event will depend on its result! I had White, and the opponent's opening repertoire was not a secret to me. I wasn't worried about the Queen's Gambit, although, of course, just in case I had a specific variation pre­ pared. Looking at Speelman's recent games, I came to the conclusion that the probability of him employing the Slav Defence was also not very great. Therefore for it too I did not particularly prepare, especially as I had something in reserve. But I had to reckon with the fact that Speel­ man, a player with a fighting and original style, might well answer 1 d4 with 1 . . d6 - it has to be said that he has a good feeling for Pirc Defence-type positions and he quite often plays them. Even so, I will not hide the fact that initially I was tempted to play this. But then I rejected this idea, since here I had nothing prepared. I decided not to embark on a critical theoretical dispute, preferring to save effort on preparation. .

And suddenly an idea occurred to me, how even in this case I might drag Speelman into 'my' type of position. After 1 d4 d6 I should play 2 g 3 ! . You will ask, what is the

Preparation for a Game

advantage of this move order, compared, say with 2 lt:Jf3 ? The subtle point is that in reply to 2 lt:Jf3 the English grandmaster often employs the system 2 . . . lt:Jf6 3 c4 .l1.g4 . But by playing 2 g3, I sharply weaken the strength of the bishop move to g4 , since I succeed in supporting the knight with the bishop from g2 , which would hardly appeal to Speelman. Of course, this move order might lead to a normal King's Indian with the move g2-g3, but this did not bother me this variation was part of my opening repertoire. As you see, I was helped in my manoeuvrings by the fact that I had a choice of different systems against the King's Indian Defence. If I had employed , for example, only the Samisch Variation, it would have been far harder to avoid Speelman's intricate prepa­ rations. Thus after 1 d4 d6 2 c4 he could have replied 2 . . . g6, managing without . . . tt:Jf6 . I would also have had to analyse this . . . As it was , with the one move 2 g3 all the preparation was practically completed!

In the game after 1 d4 d6 2 g3 we transposed into a King's Indian Defence, a position which was well familiar to me was reached , and as a result I was able to achieve the more pleasant game. I was quite satisfied with the outcome of the opening. It goes without saying that such preparation is very economical. But, in my view, immedi­ ate preparations should as far as possible be economical and rational, because exces­ sive expenditure of energy immediately before a game may, as I have already said, boomerang during play. This is why you should be able to vary your play in the opening. Firstly, this increases your chances of luring the opponent into a position that is uncomfortable for him, and secondly, it becomes more difficult to prepare for a game with you. Another example of successful preparation

ltJ

1 67

may be provided by my game with Kasparov from the same tournament in Linares. Here what told was mainly the factor of surprise. Kasparov did not expect that I would choose the Dutch Defence (Leningrad Variation). And when I replied to 1 c4 with 1 . . . f5, for a moment he became flustered . I saw that my reply was unpleasant for him, and that he had obviously not anticipated it in his preparations. As a result, Kasparov spent more time than me on the opening, but even this was not the main thing: speaking in tennis language, his well-developed first serve did not go in, and he had to use his second, less powerful one. I soon managed to seize the initiative. And yet a win in this last round game would have been so important for him! Of course, it is no bad thing to spring a surprise in the opening. However, I don't recommend that for this you employ a variation that you are playing for the first time in your life. It is extremely dangerous to bluff, especially against a skilled opponent. In principle, it is possible to give numerous pieces of advice about preparation. There are as many opinions as there are people! Every top player has his own prescriptions, often an entire system of preparation. For example, I can mention the so-called 'Capa­ blanca rule' . It is very sensible and useful. Capablanca said that for a tournament he usually prepared one opening each for White and for Black (in the latter case, apparently, one each against 1 d4 and 1 e4 ). How did he justify such an approach? By the fact that if in his set-up there were some defects, the opponent, who in a tournament did not have a great deal of spare time, would be unlikely to discover them. And indeed, if you have been struggling with a variation, say, for a week and have not found a refutation, why, you may ask, should the opponent be able to find one in the two hours before a game?

1 68

�

Preparation for a Game

Thus preparation for a tournament begins long before it starts. This enables you to save strength during the event, and eases the choice of opening for a specific oppo­ nent. As I have already said, the ideal version is when you don't need to analyse anything at all, but merely to refresh your home preparations in your memory. Of course, like any ideal, it is unattainable, but you should aim for it! If you are nevertheless obliged to work immediately before a game, approach the preparation rationally. Don't endeavour to refute without fail the opponent's set-up sometimes it makes sense to avoid it. Don't get involved in a theoretical discussion in a line where you feel that the opponent is well prepared. Only if you sense some serious defect in a variation chosen by him (remem­ ber the 'Nogueiras variation'), can you not spare any effort in search of a concrete refutation. It is very important to be able to guess the opponent's actions: how he will prepare, what he will chose, what he will expect from you. Without this it is hardly possible to arrange your preparation correctly. Endeav­ our to understand who you are dealing with: whether it is a player who is combinative or positional, bold or cautious, dogmatic or experimental . . . For example, if you know that the opponent handles endings badly, is it worth trying to devise something in the opening? Isn't it simpler to find a variation which leads directly to the endgame? However, if your own opening arsenal is limited, even the most correct compiling of the opponent's creative portrait will not enable you to exploit his weak points. In this case you effectively have no choice. Play that which you know well, i.e. not against a specific opponent, but as though against his pieces (remember the title of the collection of best games by the Yugoslav grandmaster Svetozar Gligoric: I play against Pieces? ).

The most striking representative of this tendency, which, incidentally, to some ex­ tent resembles the afore-mentioned 'Capa­ blanca rule', was, as is well known, the great Akiba Rubinstein. Such an approach is not flexible, but it is economical. And, of course, in this case you have to know your openings thoroughly! Otherwise the opponent, after easily reckon­ ing what opening will occur, will confuse you with some prepared surprise. The creation of a brilliantly-developed, practically irre­ proachable opening repertoire is, naturally, something that is within the capabilities of only a very experienced player. But even for young players such a course is not at all appropriate: you can expand your arsenal gradually, analysing one variation after another, instead of aiming straight away to employ all the openings. I should like to mention one more 'opening device', although it relates more to the field of chess psychology. I have in mind the camouflaging of your intentions. Imagine that you have guessed right with the choice of opening variation and the opponent has fallen into your trap. Nevertheless you don't let on that you have caught him in your variation, but, on the contrary, you do everything possible to conceal this from him, so that he does not sense the danger in time. Incidentally, this is how Kasparov and Karpov act. And often, despite the fact that both the position and the entire subsequent play are well known to them, they continue to think for a long time over their moves. I should like to warn against becoming carried away by such devices. In principle they are possible, but in very moderate, so to speak, medicinal doses. In other words, camouflaging is one thing, but don't waste too much time - this is not without its dangers. Firstly, during the course of the game unforeseen problems may arise; secondly, time may be needed for the

Preparation for a Game

conversion of an advantage. Therefore I would advise you to act differently: if you know how to play in the resulting position, and it has been thoroughly studied by you at home, then make the moves quickly! In this way you will create additional psychological pressure on your opponent. After all, he will realise that he has been caught in a variation, he will no longer be so certain in his actions, and this will give you a serious psychological initiative. And it is for the seizure of the initiative - both purely chess, and psychological - that we should aim when playing the opening. At first sight some of my pieces of advice may seem to contradict one another. For

ctJ

1 69

example, as you will remember, in my game with Nogueiras I did not follow my last recommendation of quickly making the previously prepared moves. In fact here there is no contradiction. Chess is not an arithmetic game, it is by no means unam­ biguous, and the employment of a particular rule sometimes depends on the most minute nuances in the situation. Don't try to work out for yourself a strict set of instructions for all eventualities in life - it is more important simply to know various approaches to the solving of the problems arising. And the choice of a particular approach will often be purely subjective, depending on the style and tastes of the player.

1 70

� Yu ri Razuvaev

You were ri g ht, Monsieu r La Bou rdon na i s ! Everything returns to its own circles. It is only these circles that turn.

Andrey Voznesensky

Egeneration (to say nothing of those who

ven so, the chess childhood of our

are a little older) was a happy one. The era of lnformator and EGO (in this form the Encyclopaedia of Chess Openings sounds more frightening) came later. We didn't know about opening indexes, we didn't waste hours recording games, our hands weren't covered in dried glue, and our scissors did not cut up the innumerable bulletins of transient tournaments. The front of the so-called information boom had not yet arrived, and the language of essential symbols was pleasantly sparse and under­ standable to anyone. Nowadays, on open­ ing a copy of lnformator, even an experi­ enced professional initially has to anxiously skim through the incomplete (for the mo­ ment!), but furiously increasing page of the new Esperanto. At that time chess books could simply be read (hardly any reference books on the openings were published). True, far fewer books were published, but I agree with the opinion that in childhood it is more important to have only one book, but a good one and a favourite one. Therefore it was not difficult for us to satisfy our needs. And to obtain a book, you could simply go into a shop and buy it. But everything changes, including chess. The avalanche of information engulfing us today is accompanied by an intensive, total study of the opening. During the past twenty years the investigative surge has acquired

the character of an epidemic; we have laboured mightily, digging over masses of chess earth. The initial position, so frighten­ ing a hundred years ago, has lost its innocent mystery, and in all the openings (classical, or those which were once i rregu­ lar) reliable paths have been not only laid, but also thoroughly trampled down. What is meant by opening theory? It seemed that this was the method of play in the initial position. But how everything has changed now. Now we choose some position after, say, eighteen moves - and start from there!

For the uninitiated, I will describe how this tabiya arises. 1 e4 c5 2 l2lf3 d6 3 d4 cxd4 4 l2\xd4 l2lf6 5 l2\c3 a6 6 �g5 e6 7 f4 't1Vb6 8 't1Vd2 �xb2 9 .Ub1 'ii'a 3 1 0 f5 l2\c6 1 1 fxe6 fxe6 1 2 tt'lxc6 bxc6 1 3 e5 dxe5 1 4 .iLxf6 gxf6 1 5 l2\e4 il.. e 7

You were right, Monsieur La Bourdonnais!

1 6 ..lte2 h5 1 7 I1b3 'it'a4 1 8 lbxf6+ ..ltxf6 1 9 c4 But surely this isn't all forced? - the reader will exclaim, trying to suppress a mixed feeling of surprise and mild panic. Calm down, of course it isn't. But chess also has its fashions- this is one thing. Secondly, it is easier to analyse this position than the initial one. However, it has to be admitted that in some places we have delved so deeply, that we no longer remember the initial aims and directions. But we remember that 'there is no striving more natural than the striving for knowledge'. Sergey Makarychev once told me about a visit to Tunis. Shortly before his departure he dropped in at a chess club, and in one of the rooms he came across a group of juniors, absorbed in an analysis of intricate branches of the Chelyabinsk Variation (this was at the time when this paradoxical system was beginning to 'blossom' ). Calling in two days later at a chess club in Tunis, Sergey was considerably staggered to see an equally enchanting picture, moreover who would have thought it! - the positions on the boards were remarkably similar. Yes, when in the not too distant future powerful computers finally become involved, we will become witnesses to (and participants in) events far more vivid than those brilliantly depicted in the prophetic speech by Ostap Bender in Vasyuki.

ttJ

1 71

been established: 1 ) the old Italian School (a glorious time: pawns were sacrificed as though they were of no consequence); 2 ) Fran9ois-Andre Philidor (affirmation o f the basics of the positional school, the first attempt at a harmonious view on chess etc . ) . There's n o denying that these books were diligently and well written, but, as a rule, unfeeling paper has turned everything into an assertion of dogmas, and for ever excluded us from active searching. I, for example, am still tormented by the origin of the Steinitz Gambit (1 e4 e5 2 tt'lc3 tt'lc6 3 f4 exf4 4 d4). From certain authors I read with astonishment that the creator of this stag­ geringly bold and imaginative idea is re­ garded as a dogmatist. However, if the experiments of Wilhelm Steinitz so stagger us today, how they must have shocked the contemporaries of the first world champion. A worthy response to the birth of the new gambit was made by the Homer of chess Sam Loyd. See what a problem he created in connection with this.

Moreover, one cannot help noticing that the further you go into the forest, the less generalisations there are. Of course, chess ideas (like any others) have a great magical attraction, but sometimes it is useful to stop and look around. Enormous experience has helped mankind to accurately establish that 'a thousand ways lead to fallacy, but only one to the truth' . And the history of chess ideas eloquently speaks of the dialectics of chess theory.

The solution is amazingly beautiful: 1 'it>e2!! (if it were possible, I would add another couple of exclamation marks from myself) 1 . . . f 1 'iV+ 2 'it>e3.

In literature a customary scheme has long

Or the birth of the Alekhine Defence. How

1 72

�

You were right, Monsieur La Bourdonnais!

was it that the great maestro, who so valued time in chess and for a couple of tempi was capable of any sacrifice, could believe in the paradoxical 1 e4 tiJf6 ? In general, it is time to say that the develop­ ment of chess ideas more resembles a wild vine than a conifer. Before our eyes open­ ings are born, sink into oblivion, and then are revived anew.

This position is as old as chess itself ( 1 d4 d5 2 c4). At any event, it features in the Gottingen manuscript (late 1 5th century) and in Damiano's book ( 1 51 2 ) . The first books give the variation 2 . . . dxc4 3 e4 b5 4 a4 c6 5 axb5 cxb5 6 b3. Ruy Lopez (according to tradition - the first major opening theoretician) in 1 561 in his treatise Libro de Ia invenci6n liberal y arte del juego del Axedrez refined the move order: 5 b3 and if 5 . . . cxb3 6 axb3 with advantage to White. More than forty years later, Salvia, who glorified Leonardo da Cutri in his romance II Puttino, altramente detto if cava/iero e"ante, pointed out an attractive trap which is familiar to us all: 2 . . . dxc4 3 e3 b5 4 a4 c6 5 axb5 cxb5 6 1i'f3. In fact the gambit proved to be hypothetical, and Black set about seeking counterplay. The Syrian player Phillip Stamma, the inventor of algebraic notation, in the second

edition of his famous treatise The Noble Game of Chess ( 1 7 45) pointed out the pos­ sibility of 3 . . . e5!? . Highly important i n the history of the 1 d4 d5 2 c4 dxc4 3 e3 e5 variation were the La Bourdonnais-McDon­ nell matches. Sadly, all that most of us remember of this wonderful duel is the final position from the 1 6th game of the fou rth match.

In their matches the afore-mentioned varia­ tion occurred frequently, always with La Bourdonnais playing White, and McDonnell correspondingly playing Black. Unfortunately, the author of these lines does not possess any works by the participants in the matches, but he has read with great interest the comments of Mikhail lvanovich Chigorin, who called the games between La Bour­ donnais and McDonnell 'brilliant inspirations of the past' . We will pick out the most interesting of them. La Bourdonnais - McDonnell 6th game of the second match Queen 's Gambit

1 d4

d5

2 c4

dxc4

3 e3

e5

4 i.. x c4

exd4

ttJ

You were right, Monsieur La Bourdonnais!

5 exd4

lbf6

1 7 lbg6+

'iit g 8

6 lbc3

!IL.e7

1 8 !IL.b3+

lbbd5

Here Chigorin makes an interesting com­ ment: 'La Bourdonnais acknowledges 6 . . . i.. d6 to b e better than the move in the present game. ' Since the theory of the 3 . . . e5 variation was poorly developed, La Bourdonnais' evaluation, speaking in edito­ ria l language, deserves consideration. 7 lbf3

0-0

8 0-0 In the first match 8 !IL.e3 occurred, but the move played here looks more natural. 8 . . . c6 Here again there is an interesting comment by Chigorin: 'This method was suggested by Philidor, but La Bourdonnais considers it bad. ' But what does modern theory say? Unfortunately, during the intervening 1 50 years it has not proved possible to give an exact reply. It can only be mentioned that until recently they believed Philidor. Now they have doubts.

1 9 lbxd5 Very pretty - if 19 . . . lbxh5 20 lbde7 mate. 19 . . . cxd5 20 !IL.xd5+ lbxd5 2 1 'ii'x d5+

.:tf7

22 lbe5

!IL.e6

23 'ii'x e6

!IL.xe5

24 dxe5

fxe3

25 l:txe3 and White won. The following game gives food for thought ­ its authentic uniqueness makes a modern and genuinely brilliant impression. La Bourdon nais - McDonnell 1 5th game of the first match Queen 's Gambit

1 d4

d5

9 h3

lb bd7

2 c4

dxc4

10 !IL.e3

lb b6

3 e3

e5

11 !IL.b3

lbfd5

4 !IL.xc4

exd4

5 exd4

lbf6

6 lbc3

!IL.e7

7 lbf3

0-0

8 h3

c6

Modern-day players more often play as Steinitz did - 1 1 . . . lbbd5. 1 2 'ii'e 2 The 1 7th game of the first match went 1 2 a4 a5 1 3 lbe5 !IL.e6 1 4 !IL.c2 f5? 1 5 'iWe2 f4 1 6 .id2 'ii'e8 1 7 .Uae1 !IL.f7 1 8 'ike4 g 6 1 9 .ixf4! lbxf4 20 'ii'xf4 !IL.c4 2 1 'ii' h 6 !IL.xf1 22 .ixg6! hxg6 23 lbxg6 lbc8 24 'ii' h 8+ ctm 25 'i'h7+ �6 26 lbf4 !IL.d3 27 l:te6+ 'ot>g5 28 i'h6+ 'iitf5 29 g4 mate. 12 . . .

'iii' h 8

13 l:tae1

!IL.d6

1 4 !IL.c2

f5?

Constancy in delusions is customarily called obstinacy. 1 5 lbe5

f4?

1 6 'ii'h 5

lbf6

1 73

1 74

� 9 .te3

You were right, Monsieur La Bourdonnais!

.tf5

La Bourdonnais - McDonnell 7th game of the third match

1 0 g4!? Such play demands not only accuracy, but also inspiration: the centre is open, Black is guaranteed counterplay, and therefore the outcome will be decided in a direct battle.

Queen 's Gambit

1 d4

d5

2 c4

dxc4

3 e3

e5

4 .txc4

exd4

lbbd7

5 exd4

lbf6

hxg6

6 lbc3

10 . . .

.tg6

1 1 lbe5! 1 2 lbxg6 1 3 h4!

lbb6

1 4 .tb3

lbfd5

1 5 h5

lbxe3

1 6 fxe3

.th4+

1 7 �d2

gxh5

18 'ili'f3

.tg5

1 9 l:taf1 ! ! The reader wishing to read detailed com­ ments on the game can find them in Yakov Neishtadt's splendid book Nekoronovannye chempiony (The uncrowned champions) . I should add that every move of La Bour­ donnais is imbued with energy and strength. 19 . . .

ifxd4+

20 �c2

iff6

21 l:txh5

iVg6+

Even the exchange of queens would not have saved Black - 21 . . .iVxf3 22 �xf3 .te7 23 .l:!.hf5 .tf6 24 g5 .txc3 25 bxc3 lbd5 26 e4 etc. After 21 . . . g6 Chigorin gives the follow­ ing variation: 22 iVh3 gxh5 (if 22 . . . iVe5 23 .l::!.f5) 23 .l:!.xf6 .txf6 24 'it'xh5, and Black has no defence against g5--g6. 22 e4

lbd5

23 .l::!.fh 1

.th6

24 g5!

f5

25 lbxd5

cxd5

26 .txd5+

�h7

27 .l:!.xh6+

�xh6

28 gxh6 Black resigned.

6. . .

.td6

As has already been mentioned earlier, it was this move that La Bourdonnais consid­ ered best. 7 lbf3

0-0

8 h3

.l:!.e8+

9 .te3

.tf4

A rash move. ' Don't cut everything that grows', Kozma Prutkov severely warns. 1 0 'ii'd 2

'*'e7

1 1 0-0

.txe3

1 2 fxe3

it'xe3+

1 3 'it'xe3

.:txe3

1 4 lbe5! And here is the retribution. We see that even 1 50 years ago one could be caught in home preparation. 14 . . .

.te6

ltJ

You were right, Monsieur La Bourdonnais!

1 5 .txe6

fxe6

1 6 Wf2

l:txe5

1 7 dxe5 etc. The La Bourdonnais-McDonnell matches demonstrated the attacking possibilities for White after 1 d4 d5 2 c4 dxc4 3 e3 e5. The position arising after the emergence from the opening was characterised by Chigorin as follows: 'A t the given moment the position of Black's game is very similar to the position which occurred in Zukertort's and Lasker's games with Steinitz. The only different is that in the present game Black has a pawn on c6, and not one on e6. But the plan of defence is the same, both with McDonnell, and with Steinitz. ' After the La Bourdonnais-McDonnell matches the variation sank into oblivion and practi­ cally went out of use. Somewhere in the first quarter of the 20th century they came to the conclusion that it is better for White to play against a pawn on e6, rather than on c6. Why did they decide this? The author has to frankly admit that he has not dug as deeply as this; moreover, it seems to me that they simply decided, and that was it. Later, as often happens, virtually only 3 ltJf3 was played (in extreme cases 3 e4 ), and in notes for the ill-informed it was stated that 3 e3 was weaker because of the counter-stroke 3 . . . e5. Of course, there were a few doubt­ ers, but no one paid any attention to them. In 1965 a book by Neishtadt on the Queen's Gambit Accepted was published. The vener­ able theoretician wrote: '3 ltJf3 is correct, not allowing . . . e7-e5 - book authors unani­ mously assert. However, the strength of the . . . e7-e5 counter-stroke should not be over­ estimated.' As often happens, help arrived from another side. In the mid- 1 970s there began a surge in the popularity of the Petroff Defence. A serious analysis was made of the following

1 75

variation: 1 e4 e5 2 ltJf3 ltJf6 3 ltJxe5 d6 4 ltJf3 ltJxe4 5 d4 d5 6 .td3 .te7 7 0-0 ltJc6 8 .l:!.e1 .tg4 9 c4 ltJf6 1 0 cxd5 ltJxd5 1 1 ltJc3 0-0 1 2 .te4 .te6 . Several years of tests, and the evaluation arrow moved from ( ) to (± ). It transpired that, with the e-file open, it was not easy for Black to defend. An echo of these changes was provided by games, played in the Queen's Gambit Accepted. =

Ti mman - Panno Mar del Plata 1 982 Queen 's Gambit

1 d4

d5

2 c4

dxc4

3 ltJc3

e5

4 e3

exd4

5 exd4

ltJf6

6 .txc4

ii.e7

7 ltJf3

0-0

8 h3

ltJbd7

9 0-0

ltJb6

1 0 ii.. b 3

c6

1 1 lte1

ltJfd5

A familiar position. Note Jan Timman's next move: Black has moved his knight away from the kingside, and White immediately

1 76

c;i?

You were right, Monsieur La Bourdonnais!

directs his second knight closer to the black monarch. 1 2 lt:Je4

�e8

1 3 ..ltd2

..ltf5

1 4 lt:Jg3

..lte6

1 5 ..ltc2

lt:Jd7

1 6 a3

li:Jf8

1 7 ..ltd3

g6?!

In Timman's opinion, 17 .. .f6 followed by . . . .Itt? was better, but even in this case the initiative is with White. 1 8 ..lth6

lt:Jf6

1 9 'ifd2

..ltd5

20 lt:Je5

lt:Je6

21 ..ltc2

lt:Jd7

22 lt:Jg4

..ltg5

23 ..ltxg5

'ifxg5

24 'ifb4

li:Jf6 !

25 l:!.e5

'ifh4

26 lt:Jxf6+

'ii'xf6

27 lt:Je4

'it'd8

Typical of this type of position: Black defends well, but he is unable to completely extinguish White's initiative. With his last move Oscar Panno missed an opportunity to simplify the position by 27 . . . ..1txe4 28 �e4 .l:!.e7. 28 ..ltb3!

a5

29 'it'c3

..ltxb3

30 't!Vxb3

1i'xd4

31 li:Jf6+

'it>h8

32 I:lae1

.l:!.eb8?!

33 I:l1 e4

'ifd8

34 llxe6!

fxe6

35 1i'c3

'ife7

If 35 . . .'il*'f8 Timman gives the spectacular finish 36 lt:Jxh7+! 'l.t>xh7 37 I:lh4+ 'ir'h6 38 'ikf6!.

39 'it'c5+

'it>g7

40 l:!.e7

'it>xf6

41 'ife5 mate The conclusion to this excellent game is steeped in the traditions of the 1 9th century. In the following example White was not able to exploit the advantages of his position, but his method of play in the opening is undoubtedly of interest. Browne - Petrosian Las Palmas 1 982 Queen 's Gambit

1 d4

d5

2 c4

dxc4

3 lt:Jc3

e5

4 e3

exd4

5 exd4

lt:Jf6

6 ..ltxc4

11.. e7

7 lt:Jf3

0-0

8 0-0

li:Jbd7

An interesting moment. The American grand­ master does not waste time on prophylaxis (8 h3) , and Tigran Petrosian does not play 8 . . . ..1tg4 - apparently, both are right. 9 1i.. b 3

lt:Jb6

1 0 .l::!. e 1

c6

11 1i.. g 5

li:Jbd5

Petrosian judges this position to favour White. 1 2 li:Jxd5

cxd5

1 3 lt:Je5

11.. e6

1 4 li:Jd3! (see diagram)

A subtle manoeuvre. White rapidly expands his possibilities, whereas Black is left with an unpleasant choice of technically difficult endings.

36 lt:Jh5+

\t>g8

37 �xe6

'ir'f7

14 . . .

lt:Je4

38 lt:Jf6+

'it>f8

1 5 11.. x e7

'ir'xe7

1 77

You were right, Monsieur La Bourdonnais!

5 exd4

tt'lf6

6 tt'lf3

i.e7

7 0-0

0-0

8 h3

tt'lbd7

9 tt'lc3

tt'lb6

1 0 i.b3

tt'lbd5

Following P h i l idor and Stei n itz!

- position after

1 4 ltJd3 -

1 6 f3

tt'lf6

1 7 tt'lc5

l:tac8

1 8 .:te3

.:r.c6

1 9 'ilfd2

b6

20 tt'lxe6

l:txe6

1 1 :e1

c6

1 2 i.g5

i.e6

1 3 tt'le5

tt'lc7

The preca rious placing of the black pieces on the e-fi le beg i n s to tel l . For example, 1 3 . . . 'it'a 5 14 tt'lxf7 ! i.xf7 1 5 tt'lxd5 tt'lxd 5 1 6 i.xe7 or 1 4 . . .�xf7 1 5 'it'e2 ! (pointed out by Lev Psa k h i s , but 1 5 i.xf6 is also good ), and 'it is h a rd to offer Black any good advice ' , which i n t h e la nguage o f symbols looks l i ke th is: (+-).

21 .:.ae1

l:txe3

22 'ilfxe3

'ii' b4

23 'it'c3!

[ The accurate reply 1 4 . . . ltJxc3! casts doubts on the combination: 1 5 bxc3 ( 1 5 tt'lh6+ �hB 1 6 bxc3 i.xb3 1 7 'iixb3 Wxg5 1 8 tt'lf7+ 'i:.xf7 1 9 Wxf7 tt'ld5) 1 5. . . i.xf7 1 6 .l::txe 7 Wxg5 -

'iid 6

Dvoretsky.]

24 l:te5

h6

25 'itf2?! Here Walter B rowne began h i s u s u a l 'time­ trouble race ' , and on the 4 1 st move the game ended i n a d raw. As was shown by Petrosian , instead of the move in the game, 25 g4 fol l owed by �g2 , h2-h4 and g4-g5 was very strong , with adva ntage to Wh ite . Dear reader, h ave you ever seen a n article without a bri l l iant example by the a uthor? 'A ridicu lous q uestion ' , you will say. I agree.

Razuvaev - Bag i rov Ya roslavl 1 982

Queen 's Gambit

1 d4

d5

2 c4

dxc4

3 e3

e5

4 i.xc4

exd4

1 4 i.c2

.:tea

1 5 'ifd3

g6

1 6 'ii'f3

tt'lfd5

1 7 i.xe7

'ilfxe7

1 8 Wg3

.l:.ad8

1 9 .:.ad 1

tt'lf6

20 f4!

tt'lh5

21 'it'f2

f5

There is no othe r defence agai nst f4-f5 .

22 g4

tt'lg7

23 gxf5

gxf5

If 23 . . . tt'lxf5 Wh ite has the good reply 24 i.xf5 gxf5 25 �h2 �h8 26 d5 tt'lxd 5 27 tt'lxd5 .l:.xd 5 28 llxd 5 cxd5 29 'ilfd4 Wg7 30 llg 1 (+-).

24 �h2

�h8

25 l:tg1

i.d5

26 l:tg 5 a n d Wh4 was th reatened .

1 78

You were right, Monsieur La Bourdonnais!

26 l'Llxd5

cxd5

S u ch moves a re easy to fi nd and pleasant to make.

34 . . .

l:!.xe8

35 'ifxd5+ Black resigned .

27 Itg6!

l'Lle6

After 27 . . . g8 Wh ite has many good a lter­ natives , one of them being 28 l:txg 7+ xg7 29 l::t g 1 + 'it>h8 30 'ifh4 (30 . . . 'ifxh4 31 t'Llf7 mate).

28 .l:!.dg1 29 lth6! 30 it.b3

.l::.f8 Wg8 t!.d6

If 30 . . . l:tf6 , then 31 'it'h4 is decisive .

31 'iVg2 32 it.xd5 33 :xe6

.l:!.fd8 :xd5 'iif8

33 . . . 'it'c7 would a l low an amusing fi n i s h : 34 'tixd5 lixd 5 35 :ea mate .

O u r excursion has come to an end, and the u n pleasant, responsible moment for sum­ ming u p has a rrived . We often a rgue about the strongest move i n various positions but, thank God , at p resent we all play d ifferently. It seems to me that in the opening you can fa ntasise a l ittl e , a n d endeavour to find you rself (or someone else's position wh ich is the most comfortable for you ) . I n this q u estion I a m all for subjectivity. So, you were right, Monsieur La Bourdonnais!

Postscri pt Severa l yea rs h ave passed since the publi­ cation of th is a rticle. At the req u est of Mark Dvoretsky I read it through anew, but decided to refra i n from correcting it. The a rticle conta i n s not only reference material, but a l so the thoug hts and fee l i ngs wh ich excited me seven years earlier. It seems to me that ideas put down on paper are as though sepa rate from us and cease to belong only to the author. Therefore I did not want to touch material wh ich had already partly become the property of others. I m u st fra n kly admit that I reread the article with a feeling of anxiety. The point is that since ch ild hood one of my favou rite chess books has been the primer by the legendary Jose Ra u l Capablanca . The extreme clarity, when the word s acq u i re an al most math­ ematical mea n i n g , the i n nate sense of h a rmony - and a l l this fi l led with genui nely Moza rt-l i ke ch arm. One does not want to part with such a book for long , and I often retu rn to it.

34 lie8!

But on one occasio n , when I was a l ready a g randmaster, I ca me across some l i nes which took me aback.

l2J

You were right, Monsieur La Bourdonnais!

1 d4 lDf6 2 c4 e6 3 lDc3 ll.b4 4 'ii' b 3 c5 5 dxc5 lDc6 6 lDf3 lDe4 7 ll.d2 lDxc5 8 'i!kc2 0-0 9 a3 ll.xc3 1 0 ll.xc3 a5 1 1 g3 'ife7 1 2 .tg2 e5 1 3 0-0 a4

1 79

is skil led i n s u btle piece play, is h ighly su ited to position s from o u r theme .

Vaganian - H u bner Ti l b u rg 1 983

Queen 's Gambit

1 d4

Under the d iagram I rea d : 'White has the two

bishops and a solid position. Black in compensation has a strongly posted knight at c5. All things considered, the position seems to be slightly favourable to White, though probably not favourable enough to win. This variation is taken from a game Stahlberg-Nimzowitsch, won by White. Since Nimzowitsch has specialised in this kind of defence, it must be assumed tha t there is nothing better for Black in this variation . . . ' How u nfortu nate! After a l l , we a re talking about one of the most correct of modern open ings - the N i mzo-l ndi an Defence. It has to be adm itted that i n the eva l uation of open ing ideas even geni u ses ca n be mis­ taken . And since that time, when I write opening a rticles and books, I i nvariably h ave a feeling of impending doo m . But on this occasion I was l ucky, and d u ring the intervening seven years the eval u ations made i n the article h ave remained i n force . In conclusion I h ave pleasu re in offering to the reader another i nteresting example. The easy, elegant style of Rafael Vag a n ia n , who

d5

2 c4

dxc4

3 lDc3

e5

4 e3

exd4

5 exd4

ltjf6

6 ll.xc4

ll. e7

7 lDf3

0-0

8 0-0

lDbd7

The attempt to simpl ify by 8 . . . ll.g4 g ave Wh ite the adva ntage after 9 h3 ll.xf3 1 0 'ifxf3 lDc6 1 1 ll.e3 ltJxd4 1 2 'ifxb 7 c5 1 3 ll.xd4 cxd4 1 4 l:1ad 1 l:tc8 1 5 b3 l:!c 7 1 6 'iff3 l:.d7 1 7 lDe2 i n the game Zaich i k-Karpeshov (Volgodonsk 1 983).

9 l:.e1

lDb6

1 0 ll.b3

c6

1 1 ll.g5

ll.g4

Carelessness, bordering on frivolousness. Without his lig ht-sq u a re bishop it will be very d ifficult for Black to 'suppress' the bishop on b3.

1 2 'ili'd3

ll. xf3

After 1 2 . . . ll. h 5 , as recom mended by certa in commentators, 1 3 lDe5 is very u n pleasant.

1 3 'ifxf3 1 4 ll. xe7

lDfd5

The position is fu l l of tem ptations and m i rages. 1 4 ltxe7 lDxe7 1 5 .l:!.e 1 looks very a ppea l i n g , but Vaga n i a n g ives a brill iant refutation : 1 5 . . . lDbc8 1 6 'ife2 .l:;le8 1 7 ll.xf7+ c;fo>xf7 1 8 'ili'e6+ 'iti>t8 1 9 .l:te3 liJd6 20 l:!.f3+ lDef5 ! 2 1 l:txf5+ lDxf5 22 ..Wxf5+ ii'f6 ! ! , and Wh ite loses beca use of the lack of an escape square fo r h i s king .

14 . . . 1 5 .l:te5 !

lD xe7

1 80

�

You were right, Monsieur La Bourdonnais!

A subtle tech nique: Wh ite does not a l l ow the black knight to go to d5, and this sign ifica ntly strengthens the pressu re of the bishop on b3.

15 . . .

ttJg6

1 6 .U.e4

ttJd7

1 7 l:.d1

ika5

Black stops half-way. The place for the black knight is at f6 . Although even i n this case after the prepa ratory 1 8 .U.e3 Wh ite plays d4-d 5 , and the su periority of the bishop over the knight will be appreciable.

18 l:!.e3

l:.ad8?

Black misses the last opportun ity to play . . . ttJf6. Now Wh ite succeeds in incl u d i ng h i s knight i n t h e attack, after which B l a c k will b e u n a b l e t o h o l d t h e position .

1 9 ttJe4

'ir'c7

20 ttJd6 was th reatened . From this moment Wh ite beg i n s to d i ctate matters.

20 h4!

h6

The h-pawn was taboo: 20 . . . ttJxh4 2 1 'i'h5 ltJg6 22 �h3 h6 23 'ir'xg6 ! (indicated by Vag a n i a n ) .

2 1 'ii'g 4

'ifi> h8

22 h5

ltJf4

22 . . . ii'f4 23 ikxf4 ttJxf4 24 tiJd6 ! was no better.

23 l:.g3

g5

24 hxg6

fxg6

25 l:te1 The movements of the wh ite pieces a re u ncommonly h a rmonious and natural. I n s u c h situations one has t o sense the melody, and then the moves suggest them­ selves .

25 . . .

.U.de8

26 .l:!.ge3

ttJb6

27 ttJc5

'ir'c8

An oversight, which h astens the end.

28 'ir'xf4! Black resigned .

tLJ

1 81

PART I l l Boris Zlotn i k

H ow do O pe n i n g N ove lties ori g i nate? A striving for the new is the first demand of human imagination. Stendhal

D

efi n i ng in aphoristic fo rm the specific nature of the th ree stages of the chess ga m e , Rudolf Spiel m a n n once wrote : 'In the opening a chess player is a book, in the

We will beg i n with novelties from the fi rst g ro u p , which even today, despite the gen­ eral i nformation boo m , sti l l occu r q u ite ofte n .

middlegame a creator, and in the endgame machine. '

Zlotn i k - Gik

a

There is no doubt that in chess, as in l ife , the commonplace and the prosaic predominate over the u n u s u a l and the a rtistic. However, even in the ope n i n g , and especially i n the endga me, there i s , of cou rse , scope for creativity. The search for opening d iscover­ ies against the backg round of a chess player's everyday work i n mastering new systems and variations, and pe rfecting those that he a l ready em ploys, is one of the most attractive aspects of chess, i n which the investigative and a rtistic components of the game i ntersect. Si nce the fi nding of opening novelties is to a certain exte nt an i ntimate process, with the aim of giving greater cla rity to the explana­ tio n , most of the g iven exa mples have been taken from the author's games. It is customary to sepa rate open ing i n nova­ tion s i nto two g ro u ps . The fi rst co nta i n s those wh ich were generated d i rectly d u ring a tournament game; i n the second a re those fou nd at home, i n the q u iet of one's study.

D u b n a 1 968

Sicilian Defence

1 e4

c5

2 tt::lf3

d6

3 d4

cxd4

4 tt::lx d4

tt::lf6

5 tt::l c 3

g6

6 j{_eJ

j_g7

7 j(,c4 As Alexa nder Alekh ine l i ked to say in such cases: 'lapsus m a n u s ' . The reason why the natural 7 f3 was not played was a banal one - before the game the you n g , newly-fledged master had been bath ing i n the Volga and s u n n i n g h i mself, and on sitting down at the board i n a mellow, rel axed state , he auto­ matically b rought out h i s bishop to c4 , th i n king that 7 f3 0-0 had a l ready been pl ayed .

7 . . .

tt::l g 4

8 ii.b5+

'it>f8

9 0-0 ! ?

1 82

w

How do Opening Novelties origi nate?

9 0-0 . This idea a l so does not occur in modern tou rn a ments. Such a conspi racy of silence is usually explained not only by ignorance. The a n swer came only in 1 987 at the U S S R J u nior Ch a mp ion s h ip , when one of the participa nts , Sergey Tivia kov - then one of the most promising you n g players in the cou ntry - showed me a cou nter-novelty for Black. I th i n k that, if only i n the moral sense, copyrig ht exists in chess; therefore I ca n not g ive B lack's idea here, and I would l i ke to suggest to the readers that they try to fi nd it themselves . In the game Skold-Botv i n n i k (Stockholm 1 962 ) , which I knew at that time, Wh ite played 9 'ii'd 2 , avoiding the deformation of his pawn structure after the exchange on e 3 . However, as t h e su bseq uent cou rse of events showed , Black's two bishops a re a more weig hty factor than the loss of the right to castle.

9. . .

lt:ixe3

1 0 fxe3

e6

1 1 i.c4

"ii'e7

1 2 lt:icb5!

..tgB?

Annoyed by my absentm i nded ness on the one hand , and feeling the need to reassu re myself on the other, I took a purely emotional (my emphasis - B .Ziotn ik) deci­ sion to sharply change the cou rse of events . After the move i n t h e g a m e and the exchange on e3 Wh ite is saddled with doubled and isol ated e-pawns, but on the other hand the f-fi le is opened and his lead in development becomes threate n i n g . F o r about 20 years I did n o t return i n m y thoug hts t o t h e diagram position , th i n king that the accidentally devised novelty was a one-off, and was su itable only for surprising the opponent i n a single game. Therefo re , when I saw m y game with Evgeny G i k i n a book by Eduard G ufeld , devoted to the Dragon Variation ( 1 982 ) , I was g e n u i nely surprised to see a q uestion mark attached to 7 . . .'�Jg4 . In his book Botvinnik's Best Games, Volume 3: 1 957- 1 9 70 (Moravian Chess 2001 ) M i khail Botv i n n i k rega rd s 7 i.c4 as an opening slip, but he ma kes no mention of

1 2 . . . i.e5 was necessary. N ow Bl ack's posi­ tion is h it by a tornado of sacrifices. It is interesting that the su bseq uent events made such a strong impression on my opponent, that after the game h e not only cong ratu­ lated me on my w i n , but a l so thanked me for my beautifu l play. A rare i n sta nce of g entle­ manly behaviour in the u ltra-com petitive world of chess.

1 3 lt:ixd6! !

"ii'x d6

1 4 lt:ixe6!

"ii'xe6

Hoping to obta i n th ree pieces for the queen.

How do Opening Novelties orig inate?

1 83

Mate would h ave resu lted from 1 4 . . . 'i!Vxd 1 1 5 Uaxd 1 tt'lc6 1 6 tt'lc 7 �b8 1 7 .ltxf7 + '>t>f8 1 8 .ltb3+ '>t>e7 1 9 llf7 mate .

(14 . . . .lte5! was more tenacious. However, even then White would have retained an obvious advantage, by continuing 1 5 "ii'xd6 �xd6 1 6 tbdB .lte6 1 7 .ltxe6 fxe6 1 8 'fJ.ad1 �el 19 tbf7 - Dvoretsky.]

1 5 'ii'd 8+

.ltf8

1 6 l::! xf7 !

'>i?xf7

Or 1 6 . . .'ii' xf7 1 7 l:!.f1 .

1 7 "ii'x c8

"ii'xc4

1 8 "ii'x c4+

'>t>g 7

1 9 "ii'd 4+ Black resigned . Rather than as the result of an emotional fit, improvised novelties, i.e. those not pre­ pared before h a n d , occu r fa r more often as the result of studying a position and g rasp­ ing its essence i n a state of max i m u m concentratio n , d i rectly against the ticking of the clock in the tou rn a ment h a l l .

Kuzovki n - Zlotnik Moscow 1 98 1

Queen 's Indian Defence

1 d4

tt'lf6

2 c4

e6

3 tt'lf3

b6

4 g3

.ltb7

5 .ltg2

.lte7

6 0-0

0-0

7 d5!?

exd5

8 tt'lh4

c6

9 cxd5

tt'lxd5

1 0 tt'lf5

tt'lc7

1 1 tt'lc3

d5

1 2 e4

.ltf6

1 3 exd5

cxd5

1 4 .ltf4

tt'lba6

1 5 .l:!.e1

At the time when this game was played , the chess world was u nder the i nfl uence of Ga rry Kasparov's b ri l l i a nt win over Slavolj u b M a rjanovic (Olym piad, M a lta 1 980), wh ich went: 1 5 . . . \i'd ?? 1 6 .lth3 '>t>h8? ( 1 6 . . . �d8 ! ) 1 7 tt'le4 ! .ltxb2 1 8 tt'lg5 ! with a decisive attack for Wh ite . While I was th i n king about my n ext move, I recalled rather vaguely that Kaspa rov had gai ned a spectacu l a r wi n , but I had no thoug hts of my own rega rd i n g this position . Delving into the situation , I came to the conclusion that the most u n pleasant piece for Black was the knight on f5 . Therefore I beg a n studying 1 5 . . .'�d 7 and 1 5 . . . .ltc8 . I d i d n 't l i ke the fi rst of these because of 1 6 .lt h 3 . M y basis for choosing the second was the variation 1 5 . . . .ltc8 1 6 tt'ld6 .ltxc3 ! 1 7 bxc3 .lte6 1 8 c4 dxc4 1 9 .ltxa8 "ii'x a8 with excellent prospects for Black. After the game I acq u a i nted myself with Kasparov's notes to his game with Marja­ novic, and i n pa rticular, with his eva l u ation of 1 5 . . . .ltc8 as being p romising for Wh ite in view of the variation 1 6 tt'ld6 .lte6 1 7 tt'lxd 5 tt'lxd 5 1 8 .ltxd5 .ltxd5 1 9 "ii'x d5 .ltxb2 20 'fJ.ad 1 . Appa rently, afte r my present game with Kuzovkin (lnformator 3 1 /6 1 8 ) Kaspa rov changed h i s opinion about 1 5 . . . .ltc8 , and in h i s book The Test of Time h e commented

1 84

�

How do Opening Novelties originate?

that it was acceptable. It would appear that 1 5 . . . �c8 is not only a possible move , but also the strongest. At any event, this idea was employed against Kasparov by Anatoly Karpov i n the second game of their fi rst match ( 1 984/85) after a slig htly d ifferent move order: 1 3 �f4 �c8 ; there followed 1 4 g4 4:Jba6 1 5 .l:!.c1 �d7 1 6 'ii'd 2 tt:Jc5 , and Black's position proved very sou n d . After this the entire gambit with 7 d5 went out of fashion. Retu rn ing to the sou rce game, I should mention that the comparatively best move for Wh ite was nevertheless 1 6 4:Jd6.

1S . . . 1 6 tt:Jd4? !

�c8! 'ii'd 7

17 .l:!.c1

�b7

1 8 �eS

�xeS

1 9 l:!.xeS

l:tfe8

32 �2

l:!.a3

Wh ite resigned . The m a i n form of novelty i s , of course, the one p repared at home in the q u iet of one's study. One ca n d i stinguish three types of such novelties: accidenta l , on the basis of analogy, and the creation of new positions. I will g ive some exa mples of each type. One of the most impressive novelties of 1 988 came i n the following position. 1 e4 e6 2 d4 dS 3 tt:Jc3 �b4 4 eS cS S aJ �xc3+ 6 bxc3 tt:Je7 7 'ir'g4 0-0 8 4:Jf3 tL'lbc6 9 �d3 fS 1 0 exf6 .l:txf6 1 1 �gS eS!

Black has managed to consolidate h i s position, reta i n ing his extra pawn .

20 l:thS

g6

21 'ji'd2

f6

22 �h4

tt:Jcs

23 4:Jb3

l:!.ad8?

23 . . . 4:J5e6 was better. Now Wh ite could have reduced Black's adva ntage to the minimum by 24 tt:Jxc5 bxc5 25 tt:Ja4 tt:Je6 26 tt:Jxc5 tt:Jxc5 27 �xc5, although after 27 . . . d4 it would not have been easy for h i m to defend.

24 l:!.d 1 ?

lieS

2S tt:JxcS

bxcS

26 l:!.a4 26 tt:Ja4 was also no better on account of 26 . . . 'ti'e7 27 'ji'a5 tt:Je6 .

26 . . .

a6

27 J:taS

'ii'e7

28 f4

.l:te3

29 b4

cxb4

30 tt:JxdS

.i.xdS

31 .i.xdS+

�g7

B lack's l ast move had been known fo r a long time and the verd i ct of theory was u nani­ mous: it was bad to play th is i n view of 1 2 �xh 7 + 'it>xh 7 1 3 ifh5+ �g8 1 4 .i.xf6 gxf6 1 5 dxe5. Citi ng the generally-accepted opinion, the author of these l i nes also gave this variation i n his book on the French Defence, pub­ l ished i n 1 98 2 . Howeve r, d u ring the process of working on the book I rea l i sed that many existing eva l uations were based on old games, and games that were not a lways played by players of h i g h standard . Both these facto rs , the age of the eva l u ations and the low standard of the players, necessa rily suggests that the concl usions of theory in

tLJ

How do Opening Novelties originate?

such cases may be d u bious.

9 .li£.d3

f5

The h i story of the move 1 1 . . . e5! - the novelty here is mainly the change of assessment from q u estion mark to exclama­ tion mark - is as fol lows .

1 0 exf6

.l:!.xf6

1 1 .ll£. g 5

e5!

In the summer of the previous year, Yu ri Dokhoian (who was then sti l l a master a n d a student at the chess facu lty) a n d h i s tra i n e r, the teacher Sergey Kishnev, asked me h ow to ease Black's d ifficu lties in the variation 1 1 .tg5 �f7 1 2 jLxe 7 .l::!. x e 7 1 3 'ii' h 4. I knew wel l from m y own practical experience what a than kless task it was to defend as Black i n t h i s variati o n . A s t h e author o f a n open ing book, i n which , ta king acco u nt of the con­ siderations only j u st expressed , a n u m ber of assessments were d u b ious, I suggested , or more precisely expressed my certa inty, that the m ove 11 . . . e5 was perfectly possible. As the result of a brief joint analysi s , to general su rprise, including that of the per­ son what had suggested i nvestigati ng this conti nuatio n , it was establ i shed that 1 1 . . . e5 was a promising move , and it beca me clear that a new branch on the tree of open ing theory had been generated . The opportu n ity to be the fi rst to test 1 1 . . . e5 in practice , and to reap the ha rvest of the novelty, fel l to Dokhoia n . Here is the game which ap­ peared fi rst i n the a n n a l s of ope n i n g theory (Inform a tor 46/383 ) .

Psakhis - Dokhoian USSR Champions h i p , F i rst Leag u e , Klaipeda 1 988

French Defence

1 e4

e6

2 d4

d5

1 85

12 'it'h4 Accord ing to Dokhoi a n , 1 1 . . . e5 left Lev Psakhis i n a state of shock. H e thought over his reply for more than 40 m i n utes, and he sti l l d i d not risk going i n for the main conti n u atio n : 12 .li£.xh7+ 'it>xh7 1 3 'it'h 5+ 'it>g8 1 4 jLxf6 gxf6 1 5 dxe5, g iven in all the books on the F rench Defence.

Despite the i rrational nature of the position , defi ned i n particu lar b y the exposed position of the black k i n g , the situation is one not only of approxi mate material equal ity, but also positional equal ity. The secret of the position is that Black should aim for the exchange of queens by 1 5 . . . �f8 with the idea of 1 6 . . . ii'f7 . Analysis has shown that it i s not so simple to advance Wh ite's three ( ! ) co n nected passed pawn on t h e kingside. The game Abramovic-Dokhoian (Belgrade 1 988) went 1 2 'ii'g 3 .l:!.xf3 1 3 gxf3 c4 ! 1 4 .ll£. x e7? ( 1 4 .i.e2 was better) 1 4 . . . 'it'xe7 1 5 jLe2 exd4 1 6 'iiif 1 .li£.f5 with a clear advan­ tage for Black.

3 tbc3

jLb4

4 e5

c5

5 a3

jLxc3+

6 bxc3

tbe7

12 . . .

e4

7 'tlt'g4

0-0

1 3 .li£.xf6

gxf6

8 tbf3

tbbc6

1 4 'ii'xf6

exd3

1 86

�

How do Opening Novelties orig inate?

1 4 . . . exf3 was wea ker on account of 1 5 gxf3 tt::lg 6 1 6 'ii'x d8+ tt:Jxd8 1 7 dxc5! tt::l e 6 1 8 ..txg6! hxg5 1 9 0-0-0 tt:Jxc5 20 1hd5 with advantage to Wh ite (ind icated by Dokhoian).

1 5 cxd3

cxd4

1 6 tt:Jxd4

tt:Jxd4

1 7 'i*'xd4

..tf5

1 8 0-0

tt:Jc6

1 9 'ii'e 3

d4

20 cxd4

'iix d4

21 .l:tfd 1

.l:!.d8

22 l:i.ab1

.l:td7

23 'ii'g 5+ Draw. We will now consider a n exa mple of an open ing novelty, devised by analogy. This type is the one that occu rs most often . The degree of analogy with known exa mples ca n differ, of cou rse, from being obvious to h a rd to establish . We will consider an exa mple which is roughly eq uidista nt from the two extreme poi nts .

A very unusual type of move , especially for the ope n i n g stage - a piece is simply placed en prise. How did the idea of this move a rise? The prototype was provided by the game Fischer-Korchnoi (Candidates Tou rna­ ment, C u rac;:ao 1 962), i n which afte r the moves 1 e4 d6 2 d4 tt::l f6 3 tt::l c 3 g6 4 f4 i.g7

5 tt::lf3 0-0 6 ..te2 c5 7 dxc5 'ii'a 5 8 0-0 'ii'x c5+ 9 'it>h 1 tt:Jc6 1 0 tt::ld 2 a5 1 1 tt::l b3 'i'b6 1 2 a4 tt::l b 4 1 3 g4? the following position a rose .

Makarychev - Zlotn ik Moscow 1 978

Sicilian Defence

1 e4

c5

2 tt::lf3

tt:Jc6

3 tt:Jc3

g6

4 d4

cxd4

5 tt:Jxd4

i.g7

6 ..te3

tt:Jf6

7 ..tc4

'ii'a 5

8 0-0

0-0

9 tt::l b 3

'ii'c 7

1 0 ..te2

d6

1 1 f4

a5

1 2 a4

tt:Jb4

There fol lowed 1 3 . . . ..txg4 ! (an idea of Evgeny Vasyu kov) 1 4 ..txg4 tt::lx g4 1 5 'i'xg4 tt:Jxc2 1 6 tt::l b 5 tt:Jxa 1 1 7 tt:Jxa 1 'ii'c 6, and Black had an obvious advantage .

1 3 ..tf3

..tg4!

If t h e two diagrams a re compared , it is easy

l2J

How do Opening Novelties originate?

to establish their similarity. The next step i n the search for the i d e a was my o w n d i s m a l experience , i n a game with the Kha rkov master Alexa nder Va isman (Moscow 1 964 ) , whe n , playing Wh ite i n the last but one d iagra m , instead of 1 3 �f3 1 played 1 3 g4?, after which , of cou rse , there followed 1 3 . . . .bg4 ! . Reflecting o n the results of these two games , I came to the fol lowing conclusion: si nce after the captu re of the g4-pawn with the bishop Black gains a n advantag e , it ca n be assumed that in the similar positi o n , but without the win of the wh ite g-paw n , the bishop move to g4 is sufficient for eq u a l ity. Return ing to the game with Sergey M a ka­ rychev, I should mention that the g ra n d mas­ ter ve ry q u ickly g rasped the essence of the novelty and after l iterally a few m i n utes' thought he offe red a d raw. But I was loathe to part with the game . . .

1 4 il.xg4

tt:Jxg4

1 5 'it'xg4

�xc3 !

1 6 bxc3

tt:Jxc2

1 7 il.d4

tt:Jxa1

1 8 l::tx a1

e5

1 8 . b5 1 9 axb5 'iic4 20 tt:'ld2 'ii'x b5, creati ng a passed pawn on the a-file, a l so ca me i nto consideration . .

1 87

The critical position , reached practica lly by force after the accepta nce of the sacrifice . N ow 20 . . . l:.fe8 w a s essentia l , and i n view of the th reat of 2 1 . . .'it'c4 it would seem that Black's chances a re even prefera ble. However, I was so pleased that I had fi nally ma naged to e mploy the novelty which I had been n u rturing for two years , that I was u nable to force myself to play with fu ll i ntensity, a n d I made a series of second-rate moves .

20 . . .

f6?

2 1 �c5

l:i.f7

22 'ii'e 3!

'ii'c 6? !

22 . . . .Ua6 was better, followed by . . . .l:ta6-c6 a n d . . . b7-b6.

23 �b6

l!d7?

B l u ndering a paw n . It was sti l l not too late to mai nta i n eq ual ity by 23 . . . l:l.a6! 24 �xa5 b6 25 �b4 l:i.xa4 .

[After 26 l:!.xa4 "fixa4 27 'ikxb6! Black's position is difficult. The knight is taboo (2 7. . . "fixb3 ? 28 "Yiib 8+ cJ;;g 7 29 il.f8+), and White, after playing h2-h3 and tt:Jc5, will create threats to the enemy king - Dvoretsky.]

.

1 9 fxe5 20 �g3

dxe5

24 tt:Jxa5

'ii'e6

25 c4!

f5

[Black misses his last chance to complicate matters by 25. . .1id3! 26 'i\Yc5 (26 'itkxd3 "fixb6+ and 27 . . . l:!.xa5; 26 'ilf2 :I:!d2) 26 . . . 'ild6! ? (or 2 6. . . 'it'g4!?) 2 7 'ii'xd6 'i:J.xd6 28 �c5 l:!.d7 29 �b4 l:!.d4 - Dvoretsky.]

26 c5 27 "ii' b 3

f4 'ii'x b3

28 tt:Jxb3

g5

29 cj;;f 1

f3

30 gxf3

l:i.d3

3 1 tt:Ja5 32 cJ;;g 1

l:i.xf3+

33 tt:Jc4 And Wh ite wo n .

l:i.f7

1 88

�

How do Opening Novelties originate?

We will now examine two ga mes by well­ known grandmasters . I n the fi rst of these the novelty was apparently devised by making use of the classical heritage, while i n the second a new position was created .

d 5-pawn n u l l ified all these attem pts. Thus, for exa m ple, the game B ro n stei n-Geller (Teesside 1 97 5 ) went 1 5 a4, and after 1 5 . . . li:Jb6 1 6 .l:d 1 'ike? 1 7 axb5 axb5 1 8 lLla3 b4 ! ? 1 9 cxb4 .l:.xa 3 B lack obta ined good play.

Romanishin - Geller

1 5 li:Jd2

43rd USSR Champions h i p , Yerevan 1 975

1 6 li:Jf1 !

Ruy Lopez

1 e4

e5

2 li:Jf3

lt:Jc6

3 �b5

a6

4 �a4

li:Jf6

5 0-0

�e7

6 .l:te1

b5

7 i.b3

0-0

8 d4

d6

9 c3

i.g4

1 0 d5

lt:Ja5

11 i.c2

c6

1 2 h3

i.xf3

1 3 'it'xf3

cxd5

14 exd5

lt:Jc4

li:Jb6

An i nteresti ng g a m bit idea . It is noteworthy that a n u m be r of pa rticipants in this tou rna­ ment - the 42nd U S S R C h a mpionsh i p were 'green with envy' on seeing Wh ite's last move , stating that their attempts to find some new idea in the g iven position had been uns u ccessfu l . N ow 1 6 . . . lt:Jfxd5 i s unfavo u rable because of 1 7 i.e4 , but at fi rst sight it is u n clear what compensation Wh ite will h ave after the captu re on d5 with the other knight.

16 . . .

li:Jbxd5

So, the position , for which Oleg Roman ish in was a i m i n g when he played 1 5 li:Jd2, has been reached . What does Wh ite have for the sacrificed pawn? And h ow, fi nal ly, did Rom a n i s h i n devise his idea? Before the present game this position had occu rred many times in various tou rna­ ments . Repeated attempts had been made to demonstrate Wh ite's su periority thanks to his pa i r of bishops, but the weakness of the

I would ventu re to suggest that a prototype was provided by the fa mous game B ran­ stein-Keres (Cand idates Tou rn ament, Bu­ d apest 1 950): 1 e4 e5 2 li:Jf3 li:Jc6 3 �b5 a6

4 .ta4 li:Jf6 5 0-0 i.e7 6 .l:i.e1 b5 7 i.b3 d6 8

ttJ

How do Opening Novelties originate?

c3 0-0 9 d4 il.g4 1 0 h3(? ! ) ..txf3 1 1 'i'xf3(?!) exd4 1 2 'it'd 1 dxc3 1 3 t2Jxc3

20 b4

lZ'lfe8

21 il.e3

tZ'lbS

22 il.c2

.l:lc8

23 .U.xa6

l:txc3

24 .l::!.a 8

'ikc7

25 1i'd5

lZ'lf6?

1 89

This move a l lows Wh ite to reg a i n h i s pawn , while keeping the i n itiative. The variation 25 . . . .l::!. x c2 26 'ii'x b5 lZ'lf6 27 J::t e a 1 ! was also to h i s advantage . T h e best move w a s 25 . . . tZ'ld4 ! , for example: 26 il.xd4 lZ'lf6 27 �xf8+ il.xf8 28 il.xh7+ 'iitx h7 29 'ii'a 8 'ii'c 8 30 'ii'x c8 l:.xc8 , and B lack's chances a re by no means worse .

26 l:l.xf8+ Despite the forma l difference of the two last diagrams, one ca nnot help but notice their similarity i n content - i n both cases i n return for the sacrificed pawn Wh ite has the bishop pair and p rospects of a n attack on the king .

1 7 tZ'lg3

tZ'lc7

After 1 7 . . . l:l.e8 1 8 lZ'lf5 �f8 1 9 il.b3 tZ'lb6 20 �g5! with the th reat of 21 tZ'lh6+ Wh ite has an obvious adva ntage .

18 a4

bxa4

1 8 . . b4 came i nto considerati o n . Wh ite ca n play 1 9 cxb4 , reg a i n i n g the paw n , but after 1 9 . lZ'le6 20 lZ'lf5 g6 21 t2Jxe7+ 'iix e7 or 1 9 . g6 20 il.h6 .l::!. e 8 21 lZ'lf5 il.f8 B l ack has a good game. I n the event of 1 9 lZ'lf5 bxc3 20 �h6 tZ'le6 ! (20 . . . cxb2? 21 il.xg7 bxa 1 'it' 22 .U.xa 1 lZ'lfe8 23 'ii' h 5 o r 22 . . . t2Jce8 2 3 'iig 3 , with a win for Wh ite i n both cases ) 2 1 'ii'g 3 tL'lh5 24 'it'g4 Black has a pleasant choice between 24 . . . lZ'lf6 with a repetition of moves , and the tempti ng 24 . . . cxb2. [As was pointed .

. .

il.xf8

26 . . . 'iitxf8 was dangerous because of 27 'ir'xb5 l:l.xc2 28 J::t a 1 .

27 il.xh7+

'iit x h7

28 1i'xb5 Thus, material equality has been restored , and the passed b-pawn and the possibil ity of sharply activating h i s rook mean that Wh ite's position is to be p referre d . Without doubt, the psychological i n itiative was a l so now on h i s side - after a l l , Wh ite has managed to demonstrate the correctness ( i n practice ! ) of h i s idea. Also of sign ifica n ce was the fact that, in search of a refutati o n , Yefi m Geller had expended a g reat deal of time.

. .

by Vlado Kovacevic, after 1 9 . . . bxc3 ?! there is the simple reply 20 bxc3! ttJe6 2 1 �a3 with the dangerous threats o f 2 2 l:!.xe5 and 22 �ad 1 . In the game Kurajica­ Smejkal, Titovo Uzice 1 9 78, Black preferred 1 9 . . . tZ'le6!? 20 'iib l .U.eB - Dvoretsky.] out

1 9 ..txa4

.l::!. b 8

Therefore it is not s u rprising that, althoug h Wh ite's positional advantage is not so g reat, B lack lost rather q u ickly.

28 . . .

'iit g 8

29 l:ta1

dS

30 l:ta7

'ii'c4

31 'ii b 8

d4

31 . . . 'ii'x b4 was more accu rate , and after 32 'ii'x e5 l:tc8 B l ack wou l d h ave had better chances of a successfu l defence.

32 ..tgs

tZ'lh7?

[Here too 31 . . . 'iixb4 was necessary Dvoretsky.]

1 90

�

How do Opening Novelties originate?

d3

33 �e7

33 . . . 'it'c8 was no better, for example: 34 'it'xe5 Itc 1 + 35 �h2 d 3 36 �xf8 'ii'xf8 (or 36 . . . lbxf8 37 lbf5) 37 'iie 3 Itc8 38 'ii'x d3 'it'xb4 39 l:txf7 .

34 �xf8

lbxf8

35 Ua8

d2

36 'ii'xf8+

�h7

37 'ii' h 8+

'itt g 6 �f6

38 'it'h5+ 39 'iff5+ Black resigned .

We will now consider an exa mple showing the creation of an orig inal position with a pawn sacrifice that had no analogue, and which occu rred , incidentally, i n the same tou rnament against the same opponent.

Gulko - Geller 43rd USSR Championsh i p , Yereva n 1 975

GrOnfeld Defence

1 d4

d5

2 lbf3

tbf6

3 e3 ! ? A very ra re move for a game at g rand master leve l , but it is this that beg i n s the con struc­ tion of the i ntended position .

3. . .

g6

4 c4

�g7

5 cxd5

ttJxd5

6 �e2 Another u n usual move . 6 lbc3 would h ave led to fa miliar l i nes of the G rO nfeld Defence. Wh ite's outward ly u n p retentious play natu­ rally provokes an active reaction by his opponent. . .

6. . .

c5

7 e4

lbb6

8 d5

0-0

Thus, an u n usual position has been created , one that o n ly formally resembles known positions. However, Wh ite has lost a tempo on the advance of h i s e-paw n , as a result of which he is beh ind in development, and in add ition h i s pawn centre i s i n danger, in view of the u nderm i n i ng moves . . . f7-f5 and . . . e7e6. He is now req u i red to play resourcefully and h e m u st be prepared to take risks.

9 a4! The prelude to an interesting pawn sacrifice. The natural 9 0-0 was less good because of 9 . . . e6 1 0 d6 (after 1 0 dxe6 i.. x e6 Black has a n excellent game) 1 O . . . lbc6 with the idea of . . . lbc6-d4, and it becomes d ifficult to defend the queen's paw n . I n view of the th reat of 1 0 a 5 , Black's next move i s forced .

f5

9. . . 1 0 lbbd2

1 0 a5 did not work on account of 1 0 . . . fxe4 1 1 axb6 exf3 1 2 l:!.xa 7 fxe2 1 3 'iVa 4 i.. d 7 and 1 4 . . . lba6 .

10 .

.

fxe4

.

1 1 ttJxe4

lbxd5

1 2 0-0

b6

1 3 i.. c4

e6

1 4 a5! (see diagram)

ctJ

How do Opening Novelties originate?

191

1 9 .td3 After 1 9 .txe6+ and mass exchanges on e6, Wh ite would h ave reg a i ned h i s paw n , but ended u p i n a n i nferior positi o n .

19 . . .

.tb7

20 tt:Je5

.l:Id8?

Aga i n a natu ra l move proves to be a m i stake, and th is time, apparently, a deci­ s ive one. After 20 . . . lt:Jb4 ! 21 I:txa8 .txa8 (2 1 . . . 'ii'x a8? 22 'ir'h 5 ! ) 22 ii.b1 lt:Jc6 there would sti l l h ave been a l l to play for.

21 ii.b1 Wh ite h a s reached the position fo r which h e was a i m i n g , when he sacrificed the paw n . 1 4 .l:te 1 w a s less good on account of 1 4 . . h6! , when it is not easy for Wh ite to develop h i s i n itiative . .

14 . . .

lt:Ja6

Appa rently the strongest reply. 1 4 . . . lt:Jc6 , for example, was weaker on account of 1 5 axb6 lt:Jxb6 1 6 .tb5 lt:Jd4 1 7 lt:Jxd4 'it'xd4 1 8 .tc6 l:!.b8 1 9 .te3! with the better game for Wh ite.

1 5 axb6

axb6?

This natural move i s a poor one. By no means always , especially i n such a n u n u­ sual position , is the natural the best. 1 5 . . 'iVxb6 was stronger. Then noth ing is given by 1 6 .txd5 on account of 1 6 . . . exd5 1 7 ir'xd 5+ .te6, w hi l e after 1 6 .l::le 1 tt:Jac7 Black would h ave reta i ned adequate d efen­ sive resou rces . .

16 tt:Jeg5

tt:Jdc7

17 'it'e2?! 1 7 'ii'b 3 was stronger, when 1 7 . . . b5 is insufficient i n view of the sim ple 1 8 .txe6+ , while i f 1 7 . . . 'i!Vd6, then 1 8 l:.e 1 is good . [ 1 8 'i:!e 1 is a poor move in view o f 1 8. . . b5! preferable is 1 8 �d1 'iWc6 1 9 lt:Jh4! b5 20 �h3! 'ike8! 21 SLd3 e5 22 'ikg3 .tf5 with chances for both sides - Dvoretsky.]

17 . . .

'ife8

1 8 .Ue1

b5

l:.d4

21 . . . lld5 was com pa ratively better, after which there wou l d h ave followed 22 f4 .

22 lt:Jef3 H e re 22 lt:Jxh7 would h ave been more q u ickly decisive. After 22 . . . ii.xe5 23 lt:Jxf8 'ir'c6 24 f3 Black would h ave lost the exchange. [In fact, the capture on h7 is incorrect in view of the counter 22 . . 't:.f5! Dvo retsky.] bpk .

22 . . .

.:td7

23 lt:Je5

l:.d4

-

24 .l:ta3 All Wh ite's pieces h ave joi ned the attack and the th reatened sacrifice on h7 ca n not be prevented . After Black's reply in the game, Wh ite ' p rosaically' restricts h i mself to the win of the exchange.

24 . . .

lt:Jd5

25 tt:Jef3

lt:Jf4

26 .t xf4

.Udxf4

27 lt:Jxe6

.l:t b4

28 'ir'd 1 !

ii. xf3

29 gxf3

'iic6

30 lt:Jxf8

ii. xf8

3 1 .t e4

'iff6

32 'ifd5+

Wh8

33 �xa6! Black lost on time, but h i s position i s , of cou rse, hopeless.

1 92

� Al exey Kos i kov

The Move

.

.

.

g7-g 5 i n the F re n c h Defence

C

hess has been in existence for a millennium and a half. But despite such a venerable age, it is cu rrently experiencing a second youth . The popul arity of the game is growi ng and the n u m ber of tou rnaments is increasing. Views on chess strategy i n genera l , and on open ing theory i n particu l a r, are ra pidly chang i n g . I n recent yea rs systems which previously were considered u n prom ising , such as the Dutch Defence or the Ital i an Game, h ave become fashionable. N ew trends in open ing theory have appeared . Fifty yea rs ago the Chelyabinsk Va riation was simply consid­ ered an anathema - it was thought that Black's position consisted entirely of wea k­ nesses . And the Volga Gambit? Already i n t h e open ing Black sacrifices a paw n , and then d reams a bout the endgame. Mean­ while, both of these systems a re now very popu lar. Moreover, Wh ite sometimes seeks ways of avoiding them , not al lowi ng them. Great changes a re also occu rri ng with i n opening systems: approaches t o them , ways o f playing them , a n d evaluations a re chang ing. The French Defence is an old love of m i n e ; I have been employing it for more than quarter of a centu ry. Using its exa mple I would l i ke to show what changes a re cu rrently occu rri ng in the i nterpretation of various open ing l i nes. What a re the classical concepts of standard French positions, arising after the adva nce of the wh ite pawn to e5? Black's plans have always been associated with pressu re on the d4-pawn and the development of his i n itiative on the q ueenside. Someti mes

Black also plays . . . f7-f6 , after which a struggle beg i n s for the e5-point: White reinforces h i s centre and tries to organise an attack (with pawns or pieces) o n the kingside. Modern chess has become 'tota l ' ; the struggle i n it is conducted by all the pieces and on any part of the board . And for Black i n the French Defence, nowadays a cou nter­ attack on the kingside is an equally custom­ ary wea pon as play on the q u eenside.

I fi rst encou ntered the move . . . g7-g5 some thirty years ago, when a n a lysing the game Sakharov-Petrosian ( U S S R Champion­ ship Semi-Fi n a l , Kiev 1 957).

1 e4

e6

2 d4

dS

3 tt:Jc3

..tb4

4 e5

c5

5 ..lli. d 2

tt:Je7

6 a3

..txc3

7 ..txc3

cxd4

8 'iix d4

tt:Jts

9 'iig 4? ! 9 'it'f4 is more accu rate .

9. . .

hS

1 0 'it'f4 (see diagram)

10 . . .

gS!

1 1 ..tbS+

tt:Jc6

1 2 'ifd2

d4!

1 3 ..tb4

�dS!

1 4 'iie 2

�xg2

The Move

. . .

g7-g5

lb

in the French Defence

1 2 lle1

1 93

0-0-0

1 3 �a3

- position after

1 0 'iVf4 -

'ifxf3

1 5 'ir'f3 1 6 ltJxf3

�d7

1 7 ltJxg5

ttJxe5

Black has g a i ned a n advantage and h e went on to wi n . For those times the move . . . g7-g 5 looked rema rkable and appeared merely to be a n exceptio n , i n no w a y disproving t h e general rule. But I remembered this idea, a n d I began employing it freq uently and not unsuccessfu lly.

Tumenok - Kosi kov Kiev 1 977

French Defence

1 e4

e6

2 d4

d5

3 ltJc3

� b4

4 e5

ltJe7

5 a3

i.xc3+

6 bxc3

c5

7 a4

ltJbc6

Nowadays in this position I prefer 7 . . . 'Wic7 .

8 lDf3 9 i.d3

�d7 �c7

1 0 0-0

c4!

1 1 i.e2

f6

What possible plan does Black h ave? I n the event of the pawn exchange on f6 he will endeavo u r to play . . . e6-e5. But if the pawn tension in the centre i s m a i nta ined, it makes sense to play . . . f6-f5 at some point and then attack on the kingside. I plan ned the following piece set-up: switch the roo k from d8 to f7, where it will not only assist the attack, but also help i n the defence along the 7th ra n k ; place the knight on f5 a n d s u pport it with . . . h7-h 5 ; play the king to a8 and retreat the bishop to c8 , su pporting the b 7 -point a n d vacating the 7th ra n k for the rook. But in some cases . . . i.d7e8-g 6 ( h 5 ) is also possible. With what should Black beg i n ?

13 . . .

�df8 !

The correct move . N ow noth ing is g iven by 1 4 i.d6? ! 'ir'd8 , when after . . . l::i.f7 and . . . ltJf5 the bishop comes u nder attack. Apparently Wh ite does best to play 1 4 i.f1 .l:i.f7 1 5 g 3 ! , i ntend ing after t h e exch ange o f pawns to bring out the bishop to h3 ( D uebai i-Fichtl , Bambe rg 1 972).

1 4 a5

.l:i.f7

15 a6? A serious strateg ic mistake - the blocking of

1 94

�

The Move

. . .

g7-g5

the queenside is to Black's advantage. I n addition , the a6-pawn may su bseq uently become wea k.

15 . . .

b6

1 6 �d6

ii'd8

1 7 liJh4 Wh ite prevents 1 7 . . . l2lf5 and prepa res f2-f4 .

17 . . .

f5!

Black's fla n k attack may be su ccessfu l only if the centre is stable. The hasty 1 7 . . . g5? is a mista ke in view of 1 8 exf6 .

1 8 f4

in the French Defence

24 . . .

Wd8!

Before the sta rt of a decisive assault it is u sefu l to safeg u a rd one's own king . Now 25 l:txb6 axb6 26 a7 l2lxa7 27 l:txa7 l2lc8 is no longer d a ngero u s . Wh ite should probably have set u p a defen sive l i ne by ii'f2 , l:i.e1 , l:tad 1 and .l:i.d2 .

25 'ii'c 1 ? Wh ite has practically no chance of creating a n attack, and n evertheless h e aims for one. This is the d ifference betwee n obstinacy and ten acity! Let's outl i n e a plan of action for Black. It is i m po rta nt for him to adva nce his pawn to f4 (after fi rst playing . . . l:!.g4 ) . I n this case the pressu re o n g2 is i ntensified , and the f5sq u a re is vacated for the knight and the b 1 h7 diagonal for t h e bishop o r queen. But fi rst he must e l i m i n ate Wh ite's hopes associated with the rook sacrifice on b6. When the plan is clear, the su bseq uent moves a re easy to make .

18 .

.

.

g5!

A typical pawn sacrifice for the sake of opening l i nes - a kind of 'Volga Gambit' i n the French Defence.

19 fxg5

"i!Vg8

20 "i!Vd2

.l::!. g 7

21 l2lf3

h6

22 gxh6

�g6!

22 . . . .l::!. x g2+ would have been a blunder, since afte r 23 �h 1 followed by 24 1'1g 1 the i n itiative on the kingside is seized by Wh ite ­ he simply has more pieces there .

23 �f1

1'1hxh6

24 1:1eb1 Now Black has to reckon with the rook sacrifice on b6, say, i n reply to 24 . . . �e8 ? .

25 . . .

tt:Jc8!

26 �a3

�e8

27 "i!Ve3

l:th8

28 Si.. c 1

l2lce7

29 "i!Vf2

l:!.g4

30 �e2

.1l.h5

30 .. .f4 31 l2le 1 was p rematu re .

31 lZ'ld2 When defending it i s i n general recom­ mended to exchange pieces . H owever, the exchange of the lig ht-sq u a re bishops does not bring Wh ite any rel ief.

31 . . .

.l::!. g 7

32 i.xh5

1'1xh5

33 lZ'lf3

�g4!

34 �d2

f4

35 l:!.e1

�h7

Th reate n i n g 36 . . . i¥xc2 .

36 l:!.ac1 37 Wf1 ?!

l2lf5

The Move

. . .

g7-g5

5 i.d3

c5

6 c3 7 lDe2

ltJc6

8 cxd4 9 exf6 1 0 lDf3

f6 lDxf6

1 1 0-0

'ii'c 7

1 2 ltJc3 1 3 i.g5

a6

1 4 i.h4

Note the g reat activity of the black pieces . The accu m u lation o f positional adva ntages usually prepa res the g round for a decisive combi n ative breakthroug h . And here such an appropriate moment has i n fact a rrived !

37 . . .

.l:!xh2!

38 lDxh2

ltJg3+

39 ii'xg3 Forced : 39 'it>g 1 iixh 2 + ! l e a d s t o mate .

39 . . .

l:l.xg3

40 'it>g1

'ii'g 6

41 i.xf4

.l:i.xg2+

42 'it>f1

l:.g1 +

43 'it>f2

ii'g2+

After 44 We3 the knight, which a l l the game has been sta n d i n g i n a m b u s h , lands the concl uding blow: 44 . . . ltJe7 ! 45 l:l.xg 1 lDf5 mate. Wh ite resig ned .

ctJ

in the French Defence

cxd4

i.d6

0-0 lDh5!

There is a struggle i n progress for the centra l d4- and e5-poi nts . Wh ite wanted to exchange the dark-sq uare bishops , whereas for Black it is adva ntageous to g ive u p his knight for the enemy bishop.

15 i.g3?! Wh ite should not h ave fa llen i n with h i s opponent's plans. 1 5 l:i.e 1 w a s stronger, a n d su bseq uently even � h4-g 5-e 3 , i n o rder t o support t h e wea k d4-paw n .

15 . . .

ltJxg3

1 6 hxg3

g6!

The g 7 -sq u a re m ust be vacated fo r the queen ; from there it i ntensifies the attack o n the key d4- a n d e5-sq u a res. B lack's positio n is a l ready prefe rable.

17 .l:!.c1

'ii'g 7

18 i.b1

Today in the French Defence B l ack ad­ vances h i s pawn to g 5 i n the most va ried situations. Here is one more example.

Smag i n - Vaiser Barnaul 1 984

French Defence

1 e4

e6

2 d4

d5

3 ltJd2

lDf6

4 e5

lDfd7

1 95

18 . . .

g5!

1 96

�

The Move

. . .

g7--g5

This is not an attack on the ki ng. By th reatening the f3-kn ight, B lack fu rther in­ tensifies the pressu re on the enemy centre .

1 9 Ite1

.Jtd7

There is no need to h u rry: the complications after 1 9 . . . g4 20 lbe5 ..txe5?! 21 dxe5 lLlxe5 22 lLlxd 5! exd 5 23 �xd 5+ lbf7 24 .l:!.c7 favou r White.

20 'ii'd 2? 20 'it'd3 was better. There now follows an exchange sacrifice, typical in such posi­ tions.

20 . . .

.l:!.xf3 !

21 gxf3

lLlxd4

22 cJi>g2 If 22 l:te3 , then 22 . . . ..tf4 ! is strong.

22 . . .

.l:!.f8

23 .l:!.h1

h6

24 'ii'd 1

lLlxf3

25 lLlxd5!?

in the French Defence

26 gxh4 If 26 'lt>g 1 , then 26 .. .'i!tf7! is strong, attacking the f2-point and i ntend i n g 27 . . . lLlf3+ and 28 . . . exd 5 .

26 . . .

gxh4+

27 cJi>f1

.Jtb5+

28 c.t>e1 Here 28 . . .'it'g 2 ! would h ave won immedi­ ately. Black played less accu rately (28 Si.b4+ ) , b u t a l l t h e s a m e h e soon won . ...

And now I wou l d l i ke to reveal to y ou the story of how I conceived a new opening idea , which i n the m i d - 1 980s became very popul ar. You will see from with in the mech a n ism of how a n opening novelty emerges. When studying the Ta rrasch Va riation of the French Defence, on one occasion I was considering how B lack should play after the following moves:

1 e4 2 d4

e6

3 lLld2

lbf6

4 e5

lLlfd7

5 f4

c5

6 c3

lLlc6

7 lLldf3

Who has outwitted whom? 25 . . . exd 5? 26 it'xd5+ 'ii'f7 is incorrect in view of 27 i.e4 ! , when i t i s now Black who h a s problems. And if 25 . . ..Jtc6? there follows simply 26 �xc6 ! bxc6 27 lLle3 .

25 . . .

lLlh4+ ! !

Nevertheless the move . . . g6-g5 also comes in usefu l for the attack.

d5

The Move

. . .

g7--g5

A 'prompt' was found in a game which made an enormous impression on me.

Reshevsky - Vaganian S ko pje 1 976

7. . .

'ii'a 5

8 �2

il.. e 7

9 il.. d 3

'ii' b 6

1 0 tLle2

f6

1 1 exf6 In an earlier game Adorja n-Vaganian (Tees­ side 1 97 4) Wh ite chose 1 1 'lt>g3 ( i ntend ing h2-h3 and 'it>h2). But this did not lead to a qu iet l ife : Rafael Vaga n i a n repl ied 1 1 . . . g 5 ! ? 1 2 � e 1 cxd4 1 3 tt:Jexd4 ( 1 3 cxd4 gxf4+ 1 4 tt:Jxf4 fxe5 1 5 dxe5 tLlc5) 1 3 . . . gxf4+ 1 4 il.. xf4 fxe5 1 5 tt:Jxe5 tLldxe5 1 6 .l:!.xe5 ( B lack a l so has an excellent position after 1 6 il.xe5 tt:Jxe5 1 7 �xe5 il.d7 1 8 'iVh5+ 'it>d8) 16 . . .tt:Jxe5 1 7 il.. x e5 �g8+ 1 8 'i.t>h3 �g5, and in a sharp skirmish he fi nally won .

11 . . .

il.xf6

1 2 'it>g3

cxd4

1 3 cxd4

0-0

1 4 l:re1 ? The decisive mistake . 1 4 h3 followed by 'it>h2 was essenti a l .

ttJ

in the French Defence

1 97

In o rder to emphasise the vul nera b i l ity of the wh ite king , Vaga n i a n blows u p the central fortifications with the aid of material sacrifices .

1 5 fxe5

tLldxe5!

1 6 dxe5

il.. h 4+ ! !

1 7 'it>xh4

�xf3 ! !

N ow 1 8 gxf3 'iVf2+ leads t o a q u ick mate .

1 8 �f1

'ili'b4+

1 9 il.f4

'i!t'e7+

20 il.g5

'ii'e 6!

2 1 il.f5

l:txf5

22 tLlf4

'i¥xe5

23 'ii'g 4

l:tf7

24 ifh5

tLle7

25 g4

tLl g6+

26 'it> g3

il.d7

27 � ae1

'ikd6

28 il.. h 6

.U. af8

Wh ite resigned. This b ri l l iant rout suggested to me the main idea by which Black should be g u ided , i n o rder t o exploit h i s l e a d i n development: h e m ust deta i n the e n e m y k i n g i n t h e centre , and t h e n b l o w u p t h e centre at any cost. But in the game Panchenko-Kosi kov ( D ne­ p ropetrovsk 1 978) my opponent repl ied to 7 . . :�a5 with 8 dxc5 ! 'iVxc5 (of cou rse, 8 . . . il.. x c5? 9 b4 is bad) 9 tLlh3 followed by tLlf2-d3 . Wh ite g a i ned a n advantage and he went on to w i n . Wh ite a l s o has a nother excellent set-u p : 8 il.. e 3 ! cxd4 (8 . . . b5 9 dxc5 ! b4 does not work because of 1 0 tLld4 il.b 7 1 1 a 3 ! bxc3 1 2 b4 : Tsesh kovsky-Vaga n ia n , Vil n i u s 1 975) 9 tLlxd4 tLlxd4 1 0 il.xd4. In the search for a n i m p rovement i n Black's play, a logical thought occu rred: why place the queen on a 5 , if a l l the same it later returns to b6?

7. . . 14 . . .

e5! !

'i¥b6 ! ?

I began studying t h i s move, ru mmaging

1 98

�

The Move

. . .

g7--g5 in the French Defence

through opening books. The theory of that time stated that Wh ite g ai ns a n advantage.

8 g3 9 cxd4

cxd4

After 9 ltJxd4 ltJc5 followed by . . . ltJe4 and . . . f7-f6 Black gains cou nterplay.

9...

�b4+

1 0 'iii> f2

f6

1 1 '>t>g2 If now 1 1 . . . 0-0 , then 1 2 �d3 followed by ltJe2 and h2-h4.

u p the game, by removing the barrier of pawns . It was th i s that gave b i rth t o a n idea , which at fi rst sight seemed crazy: 1 1 . . . g 5 ! ? . Nearly all Bl ack's forces a re g rou ped together on the q u eenside, and yet h e l a u n ches tactical operations on the kingside, where the opponent has more piece s . Anti-positional? N ot a ltogether since, as we h ave already mentioned , Black is better developed and it is very i m po rtant for h i m to open lines. In add ition , the wh ite king i s s heltering on the kingside. ·

I n February 1 980 the P remier League of the U S S R Championship was held in Vilnius. I was there for several days a n d one evening I showed my idea to Gennady Kuzm i n , a g randmaster with a very u n usual way of th i n k i n g . N evertheless, h i s verd i ct was un­ ambiguous: 'Th i s ca n not be, because it can never be . ' True , he was not able to demon­ strate anything with variations. I n the summer of 1 98 1 , also i n Vi l n i u s , the All-U nion Schoolch i l d re n 's Sparta kiad was held . Appea ring for the U kra ine team was one of my p u p i l s , 1 3-yea r-old Lena Sed i n a . I focussed on this position fo r a long time. Bl ack has a stra nge cl ump of pieces on the q ueenside ('iVb6 , ltJd ? , �c8 , .Ua8 ) - they seriously h i nder their mutual development. Of cou rse, it is possible to play 1 1 . . . 'iVc7 , intending . . . ltJb6, . . . �d? and . . . 0-0-0, but during this time Wh ite will also complete his development, and his spatial su periority will leave its mark on the entire su bseq uent play. But let's approach the situation from the other side. At least Black has developed fou r pieces, whereas t h e opponent h a s devel­ oped only one knig ht, which i n add ition is depriving the other knight of its best square f3 . It is also Black to move. He has a sign ifica nt lead in development. As is wel l known , in closed positions t h i s is not too importa nt a factor. Hence Bl ack must open

I n a n i m porta nt match against t h e Moscow tea m , with the agreement of the trainers the novelty was put i nto action fo r the fi rst time. The experi ment p roved successfu l . Without going i nto deta i l s , I will show the opening stage.

Saburova - Sedina Vi l n i u s 1 98 1

11 . . .

g5!?

12 exf6

g4

1 3 f7+!

'>t>f8 !

1 4 ltJ e5 If 1 4 ltJg5 Black was i ntending 1 4 . . . ltJf6 with the threat of 1 5 . . . h 6 .

14 . . .

ii'xd4

1 5 ii'xd4

ltJxd4

The Move

. . .

g7-g5

l2J

in the French Defence

1 6 tt:Jxg4

r#;; xfl

Polyantsev - Kaplun

1 7 l'L'lf3

tt:Jxf3

U kra i n i a n Spa rta kiad 1 983

1 8 r#;;xf3

b6!

10 . . .

g5!

1 9 �b5

.ii. b 7

11 fxg5

tt:Jdxe5 tt:Jxe5

20 .l:!.d1

We7

21 �xd7

..t>xd7

1 2 tt:Jxe5 1 3 'i!tg2

22 �d2

�d6

1 4 l'L'lf3

�f8 !

Black has achieved a good position , and i n the e n d s h e went on t o w i n t h e game.

1 5 b3

.ii. g 7

16 �b2

�d7

Of cou rse, one does not h ave to be a g randmaster to rea l i se that at some point Black's position was rather dange ro u s . T h u s , for example, i n stead of 1 4 tt:J e 5 Wh ite should nevertheless h ave considered 1 4 t2lg5 l'L'lf6 , and now 1 5 h 3 ! .

1 7 .tr.c1

h6!

1 8 gxh6

llxh6

I did not want t o expose t h e k i n g too m u c h , and s o n e w searches for a n improvement to Black's play were sought. I n the end I was able to fi nd another way of i m plementing the sa me idea . It turns out that the g-pawn can also be advan ced a move earlier: 1 1 ... g 5 ! .

1 99

tt:Jc6

1 9 �e2? 1 9 tt:Je5 was better

19 . . .

0-0-0

[ 1 9 . . . e5! with the threat of 20. . �h3+ was very strong - Dvoretsky.] .

20 tt:Je5

�e8

21 'it'd2

'itb8

22 �f3

f6

23 tt:Jxc6+

�xc6

Black has ach ieved a positional advantage and l ater he successfully converted it. But it was not i n this game that my novelty was fi rst tested in practice. A few months earl ier it had been employed by Lena Sed i n a .

Voronova - Sedina Women's U S S R Championship Semi-Final 1 983

10 . . .

g5!

1 1 .ii. e 3

f6

With th is move o rder the u nderm i n i ng move gai ns in strength , although 1 1 . . g4 also came i nto con sideration . .

I showed m y new idea t o a n o l d friend o f mine, t h e Ternopolsk master Leonid Ka pl u n , and with m y agreement he employed i t i n the autu m n o f 1 983.

1 2 �h3

h5!

13 �xe6 [It is curious that the E n cyclopaed ia of Chess Ope n i ngs recommends 13 exf6 g4 14 f7+ 'itt fB 1 5 l'L'lh4 with the evaluation 'advantage for White '. Although in fact after 1 5. . . gxh3

200

�

The Move

. . .

g7-g5 in the French Defence

1 6 l?Jg6+ �g7 17 l?JxhB l?Jf6! 18 l?Jxh3 l?Jg4+ 19 �g 1 i..d7 the evaluation changes to diametrically opposite - Dvoretsky.] 13 . . .

gxf4

1 4 i..x d7+

i.. x d7

1 5 gxf4

fxe5

1 6 dxe5

i.. c 5

1 7 'ii' b 3

d4!

A sharp position has been reached , i n which Black has more than sufficient compensa­ tion for the sacrificed pawn . She went on to wi n . Si nce then nu merous games have been played with my variation , and improvements have been found both for Wh ite , and for Black, but as far as I know a refutation has not in fact been d iscovered . This is not surprising: the move 1 0 . . . g5 is positionally justified and is fu lly i n accordance with the modern 'total' approach to the struggle. While the variation was not yet widely known and was stil l i n the development stage, I natu rally took a l ively interest i n its fate. But when it fi nally began to be con­ stantly employed in tou rnaments of the most varied standard , my interest i n it g rad ually cooled . It is a ted ious busi ness processing nu merous recently-played games, taking opening refi nements virtually as far as the 40th move . A search for new paths beg a n . Generally speaking, the key to success in the open ing is largely associated with the abil ity to be ahead of opening fashion, even if only by half a step. However, new ideas do not appear as a result of the mechan ical analysis of variations - one must endeavour to penetrate i nto the essence of the events occu rring on the board . Remember the Polyantsev-Kaplun game. Black had to retreat his bishop from b4 to a better post. And in the Sed i na-Voronova game Wh ite managed to develop her bishop

on e3 - because of the bishop on b4, the q ueen could not captu re the b2-pawn . Of cou rse, it is tem pti ng to force �f2 , but is it not possible to get by without this check? Serebro - Kos i kov

Kiev 1 984 9...

i.. e 7 ! ?

1 0 i.. h 3

Bad , o f course, is 1 0 i.. d 3? l?Jxd4. 10 . . .

f6 ! ?

1 1 �f1

A complicated position, most probably fa­ vou rable for Black, results from 1 1 l?Je2 0-0! 1 2 i.. x e6+ ( 1 2 0-0 fxe5 1 3 fxe5 l?Jdxe5) 1 2 . . . �h8 1 3 i.. x d7?! i.. x d7 1 4 0-0 i.. g 4. But the conti nuation in the game also suited me - I have managed to force the king to move without giving a check on b4. 11 . . .

0-0

1 2 'iti>g2

g5!?

1 3 'ili'b3?

Fearing a n attack, Wh ite seeks the ex­ change of queens, but in so doing he takes h i s q ueen away from the kingside, where the main events a re bound to occu r. The fate of Black's idea depends on the evaluation of the sharper moves 1 3 exf6 ! ? and 1 3 fxg5!? . 13 . . .

'ii'a 6

The Move

1 4 .i.xe6+

'it>h8

15 .i.xd5

g4

1 6 .i.e3 ! ?

tt'lb6!

17 .i.e4!

gxf3+

1 8 .i.xf3

lbc4

1 9 tt'le2

fxe5

. . .

20 dxe5

The position may stil l seem unclear. In fact, Black wins by force . How? The sacrifice on e5 suggests itself. But in the variation 20 . . . lb4xe5 21 fxe5 .U.xf3 22 'it>xf3 i.g4+ 23 'it>xg4 'iix e2+ 24 'it>h3 the king escapes from the pursu it. N o , the sacrifice must be associated with another idea . 20 . . .

lb4xe5 !

2 1 fxe5

.U.xf3 !

22 'it>xf3

.i.e6 ! !

The th i rd sacrifice i n a row! It is i m po rtant to include the rook in the attack with gain of tempo. 23 'it'xe6

CtJ

g7-g5 in the French Defence

.U.f8+

But not 23 . . . lbd4+?? 24 lbxd4, and the queen is defended .

20 1

24 .i.f4

24 tt'lf4 tt'ld4+ would have led to the loss of the quee n , but even so th is was Wh ite's best chance . 24 . . .

'ii'd 3+

25 'iitf2

.i.c5+

26 'it>e1

"ir'e4!

27 .i.g5

Wh ite also loses after 27 .U.f1 tt'ld4 28 'it'g4 lbc2+ 29 'it>d2 .U.d8+. 27 . . .

.i.b4+

28 'it>d1

"ii'x h 1 +

29 'it>c2

'ti'e4+

30 'it>b3

"ii'd 3+

31 'it>a4

b5 mate

Thus, i n the French Defence, along with the classical methods of play, Black can and should make use of plans i nvolving sharp attem pts to seize the i n itiative . And the signal for the start of active play on the territory of the opponent is often provided by the cou nter-blow . . . g7-g 5 !

202

� Vladimir Vulfson

Opening Research

H variations proceed? I will share my own

ow does the study of forcing opening

experience of working on the Dragon Varia­ tion of the Sicilian Defence. Robert Fischer stated that in th is variation a grand master ca n lose to a fi rst category player, since Wh ite's strategy is so very simple and logica l . Yes, i ndeed , here seri­ ous difficu lties lie in wait for Black, but players who employ the Dragon Variation try to compensate for this with a better knowl­ edge of opening theory and typical ideas. You can't play this variation by blindly copying the games of other players . Many ideas, which are employed at the board , are first thoroughly checked at home. Success is achieved not by the player who has learned the variation better from the books , but by the one who has carried out more research work. I will descri be a few epi­ sodes from battles i n the Dragon Variation.

9 it..c4

it..d 7

1 0 h4

�c8

1 1 il.b3

lt:Je5

1 2 0-0-0

lt:Jc4

1 3 il.xc4

.l::i.xc4

1 4 h5

li:Jxh5

1 5 g4

li:Jf6

1 6li:Jde2

'i!t'a5

1 7 it..h 6

it..x h6

18 'ii'x h6

.Ufc8

I n 1 974 Anatoly Karpov's win over Victor Korch noi in the second game of Candidates Final Match made a g reat impression on everyone. Karpov - Korchnoi

Moscow 1 974

Sicilian Defence 1 e4

c5

2li:Jf3

d6

3 d4

cxd4

4li:Jxd4

li:Jf6

5li:Jc3

g6

6 it..e 3

it..g 7

7 f3

li:Jc6

8 'ii'd2

0-0

1 9 .Ud3

This move ca me as a g reat su rprise to Korch noi . He thought for a long time, but was u nable to fi nd a sensible plan of defence, and after 1 9 . . . .U4c5? 20 g5 .Uxg5 21 .Ud5 .Uxd5 22 li:Jxd 5 .Ue8 23 li:Jef4 it..c6 24 e5 i.. x d5 25 exf6 exf6 26 'iVxh7+ 'it>f8 27 'iVh8+ he suffered a crushing defeat.

I m mediately after the game Mikhail Botvinnik suggested that Black should have defended by 1 9 . . . 'iVd 8 , with the idea of switch ing the q ueen to the defence of the kingside. It was

lb

Opening Research

203

with an analysis of this conti n uation that I began a fight to restore the reputation of the Dragon Variation . First I convinced myself that the d i rect play for the win of a piece - 20 e5 does not achieve anyth i n g . Black defends with 20 . . . dxe5 2 1 .Uhd 1 tl.8c7 2 2 'ii'd 2 e6 2 3 g5 ltJd5 , or, i f Wh ite conti nues playing for mate: 2 1 tt:Jg3 .l:td4 2 2 g5 ltJh5 2 3 ltJxh 5 gxh5 24 'i'xh5 ii.f5 . But can Black gain equal ity i n the endgame after 20 g5 ltJh5 21 ltJg3 'iVf8 ? I was u nable to fi nd a n advantage for Wh ite i n the variation 22 'ii'xf8+ 'ot>xf8 (but not 22 . . . .Uxf8 23 ltJxh 5 gxh5 24 lL'ld5 } 23 ltJxh5 gxh5 24 l:txh5 �8c5 , since in the event of 25 f4 b5 the weakness of the e4-pawn is felt. H owever, things are far from bri l l iant for Black if the white pawn reaches h6: 22 ltJxh 5 'i!Vxh6 (22 . . gxh5 23 'ikxh5) 23 gxh6 gxh5 24 �d2 and he does not have fu l l eq ual ity. I s he really obl iged to defend th is i nferior end­ game? .

19 . . .

- position after 25 f4 (analysis) -

22lL'lxh5

gxh5

23 'iWxh 5

'iWg7

24 f4

Now 24 . . . b5? 25 f5 b4 is clearly too late on account of 26 Ii.d h 3 with a mating attack. But Black has a cou nterblow i n the centre. 24 . . .

d5!

ii.e6 ! ?

This move became t h e next stage o f the search . The idea of it is to switch the q ueen to the defence of the kingside via the centra l square e5. 20 g 5

lL'lh5

2 1 ltJg3

'ike5

It would appear that Wh ite wins by 22 Itxh 5 gxh5 23 ltJxh 5 , w h e n t h e th reat o f f3-f4 followed by ltJh5-f6+ looks i rresistible. But Black nevertheless has the possibil ity of gaining a d raw, by sacrificing fi rst the exchange: 23 . . . .Uxc3 24 bxc3 l:txc3 , and after 25 f4 also a rook.

see that the king can not h ide from the perpetual check. I n cidental ly, th is idea was found by the very young Leonid Yu rtaev.

I n th is way B lack succeeds i n neutralising the opponent's main th reat of f4-f5 followed by .l::i.d 3-h3 and in winning the ideal square f5 for h i s bishop. It is easy to see that variations such as 25 f5 dxe4 a re i n his favou r.

Things a re more d ifficult for Black, if Wh ite captu res on h5 with his knight.

Wh ite appears t o reta i n all the pluses o f h i s

(see diagram) 25 . . l:txc2 + ! 26 'it>xc2 �c5 + . It is not h a rd to .

25 Ii.hd 1

204

�

Opening Research

position in view of the weakness of the 8th rank. So, is the enti re variation really bad for Black, and does Karpov's move 1 9 .l:!.d3 set him insol uble problems? I had al most des­ pai red of fi nding anyth ing, when I suddenly discovered a paradoxical defence. 25 . . .

'ir'f8 !

If Wh ite does not go in for the repetition of moves 26 �h 1 'iVg7 , he is forced to concede the f5-square. Of cou rse, he remains a pawn up, but the pressu re on c2 ties down his pieces . It only remained to wait for one of my opponents to go i n for the position after 1 9 .l:!.d3 . But time passed , and the theory of the Dragon Variation proceeded along other main lines. It was establ ished that after 1 6 tt:\de2 Black obtains good play by 1 6 . . . .l:!.e8 . Then his plans began increasingly often to be associated with the imped ing . . . h7-h5 . S o that, unfortunately, t h e analysis which I have showed you is now no longer topica l . For many years I awaited an opportun ity to employ an i nteresting preparation in another branch of the Dragon Variation . Let us return to the 'tabiya' arising after 1 e4 c5 2 tt:lf3 d6 3 d4 cxd4 4 tt:\xd4 tt:lf6 5 tt:\c3 g6 6 Si.e3 �g7 7 f3 0-0 8 "ir'd2 tt:\c6 9 Si.c4 Si.d7 1 0 h4 tt:\e5 1 1 Si.b3 .Uc8 1 2 0-0-0 tt:\c4 1 3 Si.xc4 .l:txc4 1 4 h5 tt:\xh5 1 5 g4 tt:\f6.

Along with Karpov's move - 1 6 tt:\de2 , White also has these possibil ities: 1 6 Si.h6, 1 6 e5 and 1 6 tt:lb3. I will not dwell on all these conti n uations, but will j u st describe one theoretical due l. Ba c k i n 1 976, wh e n I discussed t h e move 16 e5 with Yu rtaev, we came to the conclusion that along with the book move 1 6 . . . tt:lxg4 Black could also consider 1 6 . . . dxe5. After later working out all the details, I began waiti ng for someone to employ 1 6 e5 against me. And the n , 1 0 years later, the national master Arkhipkin went in for this position . Ark h i pkin - Vu lfson

Moscow 1 986 1 6 e5

dxe5! ?

1 7 tt:lb3

"i/c7

1 8 g5

ii..f5

1 9 gxf6

exf6

20 ii.. h 6

This natural move is wrong - playing for mate does not g ive Wh ite a nyth i n g . 20 . . .

g5!

Now the main defensive fu nctions are handed over to the lig ht-squ a re bishop. 2 1 tt:\e4

.l:!.xe4!

The only way! I n the event of 21 . . . Si.xe4 22 fxe4 .l:!.xe4 Wh ite would have gai ned a very strong attack by 23 'iVh2! �h4 24 'ikxh4. 22 fxe4

ii..xe4

23 ii..x g7

'it>xg7

24 .l:!.h3

ii..g 6

Black wants to advance as q u ickly as possible his pawn mass on the kingside. Now Wh ite should on no account exchange the queens; his play must be based on the creation of tactical th reats to the black king. Possibly I should have preferred 24 . . . b6, in order to restrict the knight and prevent the following manoeuvre of the wh ite rook. 25 .l:.c3

"f/e7

ctJ

Opening Research

205

26l:!.c5

White is cou nting only on 26 . . .f5 , which is not good because of 27 'it'c3 . He should have thought about activating his knight: 26 li::lc5. 26 . . .

h5

27 a4

h4

28 a5

a6

29 1i'd5

The seemingly logical 29 .l:.d5 ru ns i nto a n unexpected activation o f Black's forces: 29 . . .l:!.c8 30 c3 'ii'b 4! . But now I prepare the advance of my f-pawn .

-

position after 44..."ifc5 -

.l:!.e8

29 . . . 30 l:!.c3

g4

Malyutin - Vu lfson

3 1 ltJc5

f5

Moscow 1 987

32 ltJxb7

g3

33 ltJd6

'ilg5+

24l:!.h2 (instead of 24 l:th3). This novelty did not pose Black any problems. After 24 ... .l:.c8 25 ltJa1 ii.g6 26 c3 'ii'c4 27 a3 h5 28 ltJc2 g4 29 f3llg8 28 �4 g6 29l:tg 3 g5+ 30 'iii>f3 ltJb6 31 hxg5 hxg5 32 lih3 l:td7 33 'it>g3 'it>e8 34 lid h 1 i.b7 3 5 e5! dxe5 3 6 ltJe4 ltJd5 3 7 ltJ6c5 .tea 38 ltJxd7 ..txd7 39 l:th7 .l:r.f8 40 l:la 1 d8 41 .l:ta8+ .tea 42 ltJc5 Black resigned .

best method of studying typical middle­ game positions is to make a selection of games and then analyse the plans and tactical and strategic methods used in them.

It is clear that you should select games conducted i n exemplary fash ion by both sides, or by at least one of the players . You are recommended to select g a mes, and not j u st fragments, since it is usefu l to

214

�

Middlegame Problems

visualise the complete picture , beg i n n i ng with the open ing su btleties and concluding with the endgame, i n which there may also be featu res typical of the system i n q uestion. The examples chosen should normally be by players, i n the games of whom positions of this type constantly occu r. Thus when studying the midd legame arising from the Exchange Variation of the Ruy Lopez, one should devote particular attention to the games of Lasker and Fischer. But, of cou rse, other games may also be valuable. For example, the great Capablanca , em­ ploying the Exchange Va riation as Wh ite for virtually the only time in his career, gave a classic example of attack in th is open ing after castling on opposite sides.

For the moment David J anowski has man­ aged to prevent the opening of l ines on the q ueenside, but at a h i g h price - the enemy knight has establ ished itself at d5. Now it only remains for Wh ite to prepare d3-d4 and c4-c5 . Black has no defence. 18 ... tLlg5 19 .l:!.f2 tLle6 20 ii'c3 .l:i.d7 21 .l:!.d1 'it>b7 22 d4 'iVd6 23 .l::!.c 2 exd4 24 exd4 lt.'lf4 25 c5 ttJ xd5 26 exd5 'i!fxd5 27 c6+ 'it>b8 28 cxd7 'iix d7 29 d5 .l:!.e8 3 0 d6 cxd6 31 'i'c6

Black resigned . I n the followi ng game Wh ite followed the plan demonstrated by Capablanca . Dvorets ky - Korya kin

Moscow 1 97 1

Ruy Lopez

Capa blanca - Janowski 1 e4

e5

RuyLopez

2 lLlf3

tLlc6

1 e4 e5 2lLlf3 ttJ c6 3 ii.b5 a6 4 ii.xc6 dxc6 5 ttJ c3 ii.c5 6 d3 ii.g4 7 ii.e3 ii.xe3 8 fxe3 'i&'e7 9 0-0 0-0-0? 1 0 �e 1 lLlh6

3 ii.b5

a6

4 ii.a4

ttJ f6

St. Petersburg 1 9 1 4

5 0-0

ii.e7

6 ii.xc6

dxc6

7 d3

Ji.g4

8 h3

ii.xf3

9 'ilfxf3

'i!Vd6

It is evidently better to fight for equal ity by 9 . . 0-0 1 0 tLld2 tLld7 1 1 ltJc4 Ji.g5 . .

1 0 tLld2

'i&'e6

1 1 ltJc4

0-0-0

(see diagram)

1 1 l:.b1 !

The pawns at a6 and c6 provide a good target for Wh ite's q ueenside pawn storm . 1 1 . . . f6 1 2 b4lLlf7 1 3 a4 Ji.xf3 1 4 l:txf3 b6 1 5 b5! cxb5 1 6 axb5 a5 1 7 lLld5 'iVc5 1 8 c4

I n open ing books you will find the recom­ mendation of Paul Keres 1 2 'iVg3 (with the idea of f2-f4 ) , but then Black can immedi­ ately equal ise by 1 2 . . . ttJxe4 ! 1 3 'ii'x e5 'it'xe5 1 4 ttJxe5 tLld6 . It is better to chose the pla n demonstrated by Capablanca i n the previ­ ous example. 12 ii.d2! Classical examples should be used, but

4J

Middlegame Problems

215

aS

1 8 axb5

If 1 8 . . . axb5, then 1 9 �a7! is good , for example: 1 9 . . .f6 20 l:1b 1 (20 lt:id5?! .l::i. x d5 ! ) , or 1 9 . . . lt:ig6 20 lt:id5 .l:.xd5 2 1 l:!.a8+! 'it> b 7 2 2 exd5 'ii'x d5 23 l:!.xh8 lt:ixh8 2 4 �xg5. 1 9 lt:id5

f6

Black is not helped by the exchange sacrifice 1 9 . . . .l:.xd5 20 exd5 'ii'x d5 , after which 2 1 c4 ! is strong .

- position after 1 1 0-0-0 . . .

not copied. The ' routine' 1 2 .l:!. b 1 is less accu rate - after a l l , i n the event of the open ing of l ines on the q ueenside, the rook will be needed on the a-file. Capablanca was able to prepare b2-b4 only by .Ub 1 ! , whereas here Wh ite has another way, with which he also con nects his rooks and completes his development. 12 . . .

lt:id7

The sharper 1 2 . . J'ldg8 was preferable. 13 b4

Also a motif from the analysis of the previous game - the advance should be begun with the b-pawn . 1 3 a4? is i ncorrect because of 1 3 . . . c5! . With the pawn on a2 the move . . . c6-c5 is not to be fea red , since Wh ite replies a2-a3 and then b2-b4 , achiev­ ing the opening of l i nes on the q ueenside.

20 c4

�cs

21 �e3!

�xe3?

21 . . . lt:id7 is more tenacious, for example: 22 d4 h4 23 'ii' h 2 �d6 24 c5 exd4 25 cxd6 dxe3 26 dxc7 exf2+ 27 .Uxf2 l:!.de8 . lt:id7

22 fxe3

The structu re of the position is the same as i n the Capablanca-Janowski game. By exploiting the weakness of the f6-pawn , Wh ite need not prepare the c4-c5 break­ throug h , but ca n carry it out straight away. 23 c5!

h4

24 'ir'e1

lt:ixc5

25 .Uxf6

'ii'e 8

26 .i:lc6

'it>b8

27l:!.xc5

Black resigned . J u st how viable Capablanca's plan has proved is also shown by the followi ng example.

13 . . .

h5

Mecking - Korchnoi

1 4 a4

g5

Cand idates Match , 1 2th game Aug usta 1 974

1 5ir'g3 !

Pro phylaxis - Wh ite prevents the opening of lines on the kingside. 1 5 . . . g4 is now poi nt­ less in view of 1 6 h4! . 15 . . .

b6

1 6 lt:ie3

lt:if8?

First 1 6 . . . 'it>b 7 is better. 1 7 b5!

cxb5

Ruy Lopez 1 e4 e5 2 lt:if3 lt:ic6 3 �b5 a6 4 �xc6 dxc6 5 0-0 'ir'd6 6 d3 f6 7 �e3 �g4 8 lt:ibd2 0-0-0

(see diagram)

216

�

Middlegame Problems

then it was al most forgotten , and only many years later did it again attract attention .

9 .:tb 1 !lile7 1 0 b4 g5 11 a4lilg6 1 2 b5 (in

the g iven insta nce the advance of the wh ite pawns leads to the opening of l i nes)

1 e4

e5

2 tt:lf3

tt:lc6

3 ..tb5

a6

4 ..ta4

tt:lf6

5 0-0

..te7

6 l:!.e1

b5

7 ..tb3

d6

8 c3

0-0

9 h3

tt:la5

1 0 ..tc2

c5

1 1 d4

fic7

1 2 tt:lbd2

tt:lc6

1 3 d5

12 ... cxb5 13 axb5 axb5 14 .:txb5 9c6 1 5 .:tb2 -*.c5 1 6 lilb3 -*.b4 1 7 lilfd4! exd4 1 8 9xg4+ 9d7 1 9 9xd7+ .:txd7 20 lilxd4! -*.c3 21 .:ta2 .:txd4 22 .:ta3 ! By his elegant

combination Henrique Mecking has ob­ tai ned a winning position. 22 ... .l::i.b4 23 .l:!.xc3 .l:i.e8 24 f3 '.t>d7 25 l:!.a 1 .l:i.b5 26 '.t>f2 '.t>d6 27 l:!.aa3 h5 28 l:!.a4 c6 29 .l:tca3 g4 30 l:!.a5 l:!.ee5 31 J:txb5 l:!.xb5 32 fxg4 hxg4 33 '.t>g3 .l:tb 1 34 ..td4l:!.c1 35 .l:.c3 b5 36 ..txf6 b4 37 .l::i.b 3 .l::i.f1 38 ..tg5 c5 39 c3 bxc3 40 .l:.xc3 J:td 1 41 ..te3 c4 Black

resig ned . Now let us see how one might attempt a systematic study of a typical middlegame position. As an example I propose to investigate the old plan with d4-d5 in the Ch igorin Variation of the Ruy Lopez. Let us select and analyse some games. But fi rst of all I should mention one principle i n t h i s type o f work. I t is useful to study methods of playing typical positions in an historic sense, in their development.

This enables the ideas of the position to be more deeply understood . The move d4-d5 in the Chigorin Va riation of the Ruy Lopez was popular in the 1 920s,

tt:ld8

13 . . .

Wh ite's plans after other knight moves are i l l u strated by two games of Yefi m Geller. Geller - Mecking

I nterzonal Tou rnament, Palma de Mallorca 1 970 1 3 . . . tt:la5 14 b3! ..td7 1 5 tt:lf1 tt:lb7 16 tt:lg3 c4? ! ( 1 6 . . . .U.fb8) 1 7 b4 .U.fc8 1 8 tt:lf5 ..tf8

(see diagram)

CtJ

Middlegame Problems

21 7

on the kingside is a lso strong : 1 5 g4 tt:Jc8 1 6 tt:Jg3 g6 1 7 cJi>h2 tt:Je8 1 8 i.. h 6 tt::\ g 7 1 9 l:lg 1 , as i n the game Klova ns-Sch neider, J u rmala 1 978) 15 ... .l:!.fb8 1 6 tt:Jg3 tt:Jc8 17 a5! c4 1 8 i.. d 2 i.. f8 1 9 tt::\h 2 tt:Je7 2 0 Sl..g 5! tt:Je8 2 1 'it'd2 'ii' b 7 22 b 4 ! cxb3 23 Sl..x b3 .l:tc8 24 l:lec 1 h6 25 i.. e 3 f5 26 exf5 tt:Jxf5 27 tt:Jxf5 i.. xf5 28 tt:Jf1 tt:Jf6 29 tt:Jg3 i.. g 6 30 c4! bxc4 31 i.. xc4 i.. e 7 32 i.. a 2 �xc1 33 nxc 1 'ii' b 5 34 �c6 'fid3 35 Sl..c4 'fixd2 36 i.. xd2 .il.f7 37 tt:Jf5 .il.d8 38 tt:Jxd6 i.. xd 5 39 .l:t.xa6l:!.xa6 40 Sl..xa6 Black resig ned .

19 tt::\h 2! a5 20 �e3! axb4 21 cxb4 i.. xf5? ! 22 exf5 c3 23 tt::\g 4! i.. e 7 2 4 tt:Jxf6+ i.. xf6 2 5 �e4! 'i*'d7 2 6 �f3 �c7 2 7 h4 f/ie7 28 g3 tt:ld8 29 a3l:!.cc8 30 �b1 �c7 3 1 �e2 �b8 32 .l::!.b 3 �d7 33 'i*'f3 i.. e 7 34 �e3 i..f6 35 �e4 i..e 7 36 g4! f6 37 �e3 tt:Jf7 38 �bxc3 .Sbc8 39 Sl..e4 il..d 8 40 i.. d 2 �c4 41 �xc4 .Sxc4 42 .l::!.c 3 i.. b 6 43 l:txc4 bxc4 44 g 5 fxg5 4 5 hxg5 il..d 8 46 'it' h 5 c3 4 7 Sl..e 3 h6 48 f6 Black resig ned .

1 4 a4

l:lb8

As the fol lowi ng game shows , it is danger­ ous to weaken the q ueenside by playing 14 . . . b4 (or 14 . . . i.. d 7 1 5 axb5 i.. x b5 16 .il.a4 ). Capablanca - Vidma r

N ew York 1 924 14 . . . b4 1 5 tt:Jc4 a5 (otherwise 1 6 a 5 ! )

Geller - He rnandez

Las Palmas 1 980 1 3 tt::\a 7 14 tt:Jf1 i.. d 7 ...

1 6 tt:Jfxe5!? ( i n t h e o p i n i o n o f Alexander Alekh ine, 1 6 Sl..e 3 tt:Jd7 1 7 tt:Jfd2 came i nto consideration) 1 6 . . . i.. a 6 1 7 i.. b3! d xe5 1 8 d6 il..x d6 1 9 'i!Vxd6 ifxd6 20 tt:Jxd6 tt:Jb7?! (20 . . . .l:!.b8 was better) 2 1 tt:Jxb7 Sl.. x b7 22 cxb4 ! cxb4? (too passive - 22 . . . axb4 23 f3 1 5 a4! (hi ndering the manoeuvre . . . tt:Ja7-

.il.a6 was necessary, followed by a possible . . . c5-c4 ) 23 f3 l:lfd8 24 Sl..e 3 h6 25 .l:!.ed 1

c8-b6 ; however, the direct plan of an attack

.il.c6 26 nac1 .il.e8 27 'it>f2 .l::!.xd 1 28 l:lxd 1

218



Middlegame Problems

�c8 29 g4 �d7? (29 . . . �8 was more tenacious) 30 �b6 �e6 31 �xe6 fxe6 32 .lld 8+! l:.xd8 33 �xd8 lZ'ld7 34 �xa5 lZ'lc5 35 b3! lZ'lxb3 36 �xb4 lZ'ld4 37 a5 Black

defence and for a possible cou nterattack on the kingside. Now his objective is to ex­ change the rooks, in order to eliminate the th reats to his king.

resigned .

2 1 'ir'f1

�d7

Incidentally, in his a n notations to this game Saviely Tartakower mentioned a recommen­ dation by Richard Teichmann - 14 . . . .lla 7 ! ? , a move which h a s not occu rred i n practice. Thus by reading old books and studying old games and the notes to the m , one can unexpectedly hit u pon interesti ng ' novelties' and ' i mprovements'.

22 �e3

lla8

23 'ir'g2

l:1xa1 !

24 .l:txa 1

'ir'b7

25 'it>h2

.Ua8

26 'ii'f1

lla6

27 lZ'ld2

'it'a8

28 l:txa6

'it'xa6

Black's position is preferable, mainly be­ cau se Wh ite has no active possibilities, whereas B l ack can prepare . . . h7-h5 and .. . f6-f5 . 29 lZ'lb3?!

lZ'lg5

3 0 '1t>g2

h5!

31 h4

lbf7

32 gxh5

gxh5

33 'it>h2

'ii'c 8

34 'it'g2

'it>f8

35 lZ'ld2

We have finally reached the main position of the variation. The followi ng game is regarded as a classic example of play for Black. Thomas - Rubinstein

Baden-Baden 1 925 15 axb5

axb5

16 lZ'lf1

lZ'leS!

17 g4

g6

1 8 lZ'lg3

lZ'lg7

19 'it>h1

f6

35 . . .

f5 !

20 l:tg1

tZ'lf7

36 exf5

�xh4

37 f6

�xf6

38 'ii'f3

�h4

White has adopted a typical attacking set-up on the kingside, while Black has demon­ strated the best regrouping of his forces for

39 �g6

Middlegame Problems

219

39 tt:Jxh5 or 39 ..th6 is bad because of 39 . . . ..ig4 .

39 . . .

..txg3+

40 fxg3

..tf5 !

41 ..ixf7?

It was nevertheless better to take the pawn : 41 ..txh5 tt:Jxh5 42 'i!Vxh 5 . However, after 42 . �g4 followed by . . . 'ii'f5 Black reta ins . .

the advantage i n the endgame i n view of the weakness of the d5-pawn . 41 . . .

�xf7

42 tt:Je4

'ii'd 7

43 ..th6

'it>g6!

44 ..txg7

'lt;lxg7

45 b4? !

c4

45 . . cxb4 46 cxb4 ..ixe4 47 'i!t'xe4 'it'g4 was .

also strong . 46 t2Jd2

'i!t'f7

Black is not only a pawn u p , but he also has good prospects of winning a second one the d5-pawn . H is position is won . 47 'i*'e3 it'xd5 4 8 'it'g 5+ ..ig6 4 9 'ti'e7+ �g8 50 'it'd8+ �f7 51 'i!Vd7+ �f6 52 'ii'd 8+ �f5 53 'it'd7+ 'lf;lf6 54 'i!Vd8+ 'it>g7 55 'ii'e 7+ 'ii' f7 56 'i'xd6 'i'f2+ 57 'it>h3 'it>h6! (zugzwang) 5 8 tt:J b 1 'i!Nf5+ 5 9 'it> g 2 'ii'x b1 60 'ii'f8+ �g5 611i'd8+ �g4 62 'ii'd 7+ 'iff5 63 'it'd 1 + �g5

Wh ite resigned . Tartakower wrote that 'The value of this game lies not in any variations, but in the general construction of that solid defensive basis, from which Black's counterattack develops.' I n the game Thomas-G runfeld, played two rounds later, Black employed the same plan of defence as Akiba Rubinste i n , but later he manoeuvred less accu rately. The game dragged o n , and on the 78th move it reached the following position :

Black fi nally carried out the familiar ad­ vance: 78 . . . f5 ! 79 •h2 fxg4 80 fxg4 •f6 81 tLlf1 •f4 ! 82 •g2 �g7 83 tLlde3 tLle7 84 •f3 g5! 85 h 5 tLlh6 86 tLlg3 tLleg8 87 tLlh1 •h2 88 �d2? (it would have been more

d ifficult for Black to break through after 88 tbf2 tbf6 89 'Ot>f1 ) 88 . . . tLlf6 89 tLlf2 •xf2 ! 90 •xf2 tLlxe4+ 9 1 �e1 tLlxf2 92 �xf2 e4! 93 .i.xc4 (desperation ! ) 93 . . . bxc4 94 tLlxc4 tLlxg4+ 95 �e 1 �f6 96 tLlxd6 �e5 97 tLlf7+ �xd5 98 tLlxg 5 h6 99 tLlf7 .i.b5 1 00 lbd8 e3

Wh ite resig ned . Rubinstei n 's game is very usefu l for an understanding of Black's methods of play, but h i s opponent manoeuvred unconvi nc­ ing ly. If he was i ntend ing an attack on the kingside, why then open the a-file? - Black promptly bega n exchanging rooks on this file. I t was more logical to block the queen­ side, i n order to free his hands for the attack on the other side of the board , or else not touch the q ueenside pawns at a l l , since it is not so easy for Black to obta i n cou nterplay there . Let u s consider some examples. Bogolju bow - Rubinstein

Baden-Baden 1 925 15 c4 ! ?

( I n stead o f 1 5 axb5). N ow i n t h e event o f an exchange of pawns Wh ite will exert pres-

220

�

Middlegame Problems

sure on the queenside. Rubinstei n coolly replied : 15 . . .

b4

The game conti nued : 1 6 b3 l2Je8 1 7 g4 g6 18 'it>h1 l2Jg7 19 l:tg 1 h5! 20 l2Jf1 hxg4 2 1 hxg4 f6 (Black h a s a defensible position) 2 2 l2Je3 l2Jf7 23 l2Jh4l2Jh 8 .

Here Yefi m Bogolj ubow demonstrated two typical attacking ideas: the open ing of l i nes by f2-f4 and the knight sacrifice on f5 . 24 f4! exf4 25 l2Jef5! l2Jxf5 (Black did not

risk accepti ng the piece sacrifice , preferri ng a qu iet and roughly eq ual position) 26 gxf5 g5 27 .l1.xf4 llf7 28 .l1.h2 llh7 29 l2Jg2 lbf7 30 l2Je3 .l1.d7 31 'iit>g 2 'itog7 32llh 1 llbh 8 33 'iie2 'iic8 34 .l1.g3 'iig 8 35 l2Jg4 llxh 1 36 l::tx h1 .l::t x h1 37 'itoxh 1 'iih 7+ 38 'it>g2 'ii h 5 39 �d1 l2Jh6 40 'iie 1 l2Jxg4 41 .l1.f3 a5 42 it'e2 iL.e8 43 �xg4 'it'h6 44 'itog 1 �f7 Draw.

However, Wh ite's attack is not as harmless as it might appear from the games we have examined. Here is a typical example. Du binin - Suetin

Russian Federation Team Championship 1 950

RuyLopez 1 e4 e5 2 l2Jf3 l2Jc6 3 .l1.b5 a6 4 .l1.a4l2Jf6 5 0-0 iL.e7 6 'iie2 b5 7 .l1.b3 d6 8 c3 0-0 9 h 3

lb a 5 1 0 .l1.c2 c5 1 1 d 4 'ilc7 1 2 d5 iL.d7 13 'it> h 1 c4 1 4 l::tg 1 l2Je8 1 5 g4 l2Jb7 1 6 tt:lbd2 g6 1 7 l2Jf1 l2Jg7 1 8 �h6 f6 1 9 l2Je3 �hB?

1 9 . . . l2Jd8? is not possible because of 20 l2Jf5! gxf5 2 1 .l1.xg7 xh7 32 'ifh5+ l2Jh6 33 l:tg61\Vf8 34l:i.xh6+ 'i'xh6 35 f6+ e4 36 �xe4+ Black resig ned .

Here the plan of attack under consideration proved of use to Wh ite in a d ifferent opening variation - with 6 'ii'e 2 (instead of 6 .l:i.e1 ). When choosing this system , Wh ite m u st be

lLJ

Middlegame Problems

prepared after 6 . . . b5 7 .i.b3 0-0 8 c3 for the gambit 8 . . . d 5 ! ? , and if 9 exd 5 ..tg4 ! , but his opponent did not go i n for th is variation . Later too Black played passively ( 1 3 . . . tZ:ic4 came into consideration , or a move earlier 12 . . . c4 followed by . . . tZ:ia5-b7-c5 , . . . a6-a5 and . . . ..ta6, as i n the game Marjanovic­ Lengyel, Sarajevo 1 980), did not create any counterplay on the q ueenside, and merely carried out Rubinstei n 's defensive set-up on the kingside. Wh ite effectively had a n extra tempo , which he saved on the move 6 .l:te 1 (the rook went to g 1 in one go) a n d , exploiting his opponent's su bseq uent i nac­ curacies, he carried out a bri l l iant combina­ tive attack.

221

In the 1 st game of the Final Candidates Match Spassky-Korchnoi (Kiev 1 968) White developed his bishop on d2, but he did not achieve much : 1 8 .i.d2 .l:ta8 1 9 tZ:ie3l:i.fc8 20 'it>h2 .l:txa 1 21 'ifxa 1 'ii'd 8 22 iVa? �a8! 23 'ti'xb7 l:i.b8, and the players ag reed a d raw. 18 . . .

l:'!.a8

1 9 'ifd2

lUeS

1 9 . . . l:tfb8 and 20 . . . 'ifc8 is better. 20 ..td3

g6

2 1 tZ:ig3

..tf8

22l:!a2

c4

23 ..tb1

1i'd8?!

But even th is fi ne game did not cast doubts on Rubi nstein's plan of defence, which successfu lly withstood many tests. Things reached the point that the d4-d5 system disappeared for a long time from tou rna­ ment play, although several attem pts were made to revive it. And , fi nally, supporters of Wh ite found ideas which breathed new life i nto this old syste m . Let us return t o t h e main position o f the variation after 14 a4 l:i.b8 . Ka rpov - Unzicker

Olympiad , N ice 1 974 1 5 axb5

axb5

1 6 b4!

tZ:ib7

Black can also switch d irectly to Rubinstei n 's defensive set-up by 1 6 . . . c4 1 7 lLif1 tZ:ie8 . The plan chosen b y h i s opponents - g2-g4, tt:Jg3, Wh2 and l:i.g 1 - is evidently not too effective here . But Wh ite found another plan : 1 8 tZ:i3h2! f6 1 9 f4 tZ:if7 20 tZ:if3 g6 21 f5 tbg7 22 g4 with pressu re on the kingside, as in the game Karpov-Spassky (4 1 st U S S R Championsh ip, Moscow 1 973), which , how­ ever, ended in a d raw. 1 7 tZ:if1 1 8 ..te3

..td7

24 ..ta7!

From a7 the bishop completely paralyses the opponent's counterplay on the queens ide. U nder its cover Wh ite gains the opportun ity to calmly prepare a storm on the other side of the board . 2 4. . . tZ:ie8 2 5 ..tc2 tZ:ic7 2 6 l:i.ea 1 "i/e7 27 ..tb1 .tea 28 tZ:ie2 tZ:id B 29 lLih2 ..tg7 30 f4 f6 3 1 f5 g5?! ( 3 1 . . . gxf5 32 exf5 lLif7) 32 ..tc2 ! ..tf7 33 tZ:ig3 tZ:ib7 34 ..td1 h6?! 35 ..t h 5 'ife8 36 'it'd 1 tt:'ld8 37 l::ta 3 Wf8 38 l:'!.1 a2 'it>g8 39 lZ:ig4! 'iitf8 40 tZ:ie3 'it>g8 41 ..txf7+ tZ:ixf7 42 'it'h5 tZ:id8 43 �g6! 'it>f8 44 tZ:ih5 Black resigned .

However, in my view the most accu rate plan

222

�

Middlegame Problems

for Wh ite was demonstrated in the fol lowing l ittle-known game, wh ich was j udged the best at the international tou rnament in Polanica Zdroj in 1 972. In it Wh ite did not exchange pawns on b5. Zuckerman - Kostro

Polanica Zd roj 1 972 15 b4!

tt'lb7

16 tt'lf1

�d7

17 ..td2

J:tfcB

answer is a l ready known from the previous games: Wh ite m u st carry out f2-f4 . 2B tt'le1

l:tafB

29 f4

exf4

30 ..txf4

tt'le5

31 tt'lf3

WeB

I n the event of 3 1 . . . tt'lxf3+ 32 'ifxf3 Black would have been deprived of h i s only trump - h i s centra lly-placed knight. 32 tt'ld4

.l:.eB

1 B ..td3

It appears that Wh ite is preparing play on the queenside, but i n fact he has a q u ite different intention . By pressure on the b5pawn he forces the opponent to move . . . c5c4, after which he blocks the queenside by a4-a5 and then begins play on the kingside. But what is the difference with the Bogol­ jubow-Rubinstei n game, where Wh ite , by playing 1 5 c4, forced the i m mediate block­ ing of the queenside? Let us fol low how events developed . 1B . . .

.:taB

1 9 'ii'e2

c4

It perhaps made sense to change the character of the play, by sacrificing the exchange: 1 9 . . . bxa4 20 �xa6 tt'lxe4 ! ? 2 1 ..txb7 tt'lxd2. 20 �c2

tt'leB

And here 20 . . . a5 should have been consid­ ered , although after 21 axb5 �xb5 22 tt'le3 the advantage would have remained with White . 21 aS!

tt'ldB

22 g4

g6

23 tt'lg3

f6

24 �h2

tt'lf7

25l1g1

tt'lg7

26 :g2

�hB

27 .l:.ag 1

l:tgB

How to fu rther i ntensify the pressu re? The

A proper eval uation can now be made of Wh ite's strateg ic idea . The a rrangement of the q ueen side pawns is ideal for h i m , and he has gained possession of the very important d4-square. It should be mentioned that, even if Black were able to evacuate his king to the queenside, Wh ite could have dis­ tu rbed it with a foray along the g 1 -a7 diagon a l . 33 tt'lgf5!

gxf5

34 gxf5

tt'lf7

It would appear that Black has no satisfac­ tory defence. Wh ite is i ntend i n g , depending on the circumstances , to place h i s bishop on h6, triple heavy pieces on the g-fi le, ad­ vance his h-pawn , and i nvade with his knight on e6 . 35 l:.g4

tt'le5

36 .l:th4

.l:1ef8

Middlegame Problems

37 .l:!.g6 !

ti:Jf7

38 l:txh7+!

�xh7

223

Bronste i n - Wi n i warter

Krems 1 967

391i'g4

Black resigned . I should l i ke to add one more featu re , in favou r of the pawn set-up on the q ueenside chosen by Bernard Zuckerman - the idea of a positional piece sacrifice on c4.

36 �dxc4! bxc4 37 �xc4 .*.b5 38 �b6 .*.xe2 39 •xe2 .*.e7 40 �cB •xeS 41 .*.a7 �d7 42 •xa6 Black resig ned .

Thus, i n this interesting variation the scales have tipped i n favou r of Wh ite . It is now Black's turn to come u p with someth ing . . .

224

� Alexey Kosikov

The Connection of the Opening with the Endgame Pa latni k - Kosi kov

D novel by the Strugatsky brothers Mon­ o you remember the science fiction

day begins on Saturday? Its title conceals a deep philosophical thought: our tomorrow's problems are generated today (if not yester­ day! ). This is also the case in a chess game: the endgame sometimes beg ins back i n the open ing!

When studying the opening it is not enough to mechan ically learn variations by heart you must have a deep understanding of the events taking place on the board , and for this you need to grasp the ideas of the forthcoming middlegame, and even the endgame. Besides, g iven the enormous flood of information which overwhelms chess players today, theory has developed so much that open ing analyses someti mes conclude deep in the endgame. I remember a game played in the Premier League of the 1984 USSR Championsh ip between Igor Novikov and Vlad imir Tukmakov. Wh ite employed an improvement on the 36th ( ! ) move , after wh ich an ending with bishop against pawns arose and, by gaining a wi n , h e changed the evaluation o f o n e o f t h e then fashionable variations of the G rO nfeld De­ fence. I should l i ke to share with you my impres­ sions of how the open ing struggle proceeds, and to show a game i n which from the very start the forthcoming endgame had to be reckoned with. During the lesson you will be invited to carry out a few assign ments for analysis at home.

Odessa 1979

Slav Defence d5

1 d4 2 c4

c6

3 4Jf3

4Jf6

4 tt:Jc3

dxc4

5 a4

i.f5

6 tt:Je5

The conti n u ation of Alexander Alekh i ne, which even today remains topica l . The alternative is 6 e3. 6 . . .

e6

7 f3

i.b4

8 tt:Jxc4

0-0

9 i.g5

h6

1 0 i.h4

c5

11 dxc5

'ifxd 1 +

1 2l:l.xd 1

i.c2

1 3 .l:i.c 1

i.h7?!

Black embarks on a n u n promising path . To those interested i n this variation, I recom­ mend that they study the move 13 ...i.xa4!, made by Jan Ehlvest against Evgeny Bareev (Ta l l i n n 1986, lnformatorNo .41 , Game 435) . 14 e4

tt:Jc6

(see diagram) Let us weigh up the pluses and minuses of the resulting position and make an evalua­ tion of it.

1 . Wh ite is somewhat beh i n d i n develop­ ment. In add ition he has a complex of weak squares on the queenside.

t2J

The Connection of the Opening with the Endgame

225

23 �xe2

Black has the 'advantage of the two bish­ ops'. But th is is a p u rely dogmatic evalua­ tio n : what advantage can one speak of, g iven the position of the bishop at h7? 23 . . .

..txc3

This excha nge wou ld appear to be forced. There was no way that I could be satisfied with 23 ...tt:'Jc6 24 lL'lb5 a6 25 lL'lbd6, whe n , exploiting t h e absence o f my k i n g and lig ht­ squ a re bishop, the opponent launches a n attack on t h e queenside. -

position after 14 ...tt::lc6-

24 bxc3

tt:'Jc6

25 lL'ld6

b6

2. Black's bishop on h7 is extremely badly

placed. I n order to bring it i nto play, he must either advance ...f7-f5 - but then the e6point is weakened , or else clear the a2-g8 diagonal (for example: ...tt:'Jd7 , ...f7-f6 , . . . e6-e5 and ... ..tg6-f7 ) - but this demands a lot of time. Whose pluses are more sign ificant? P rac­ tice has shown that Wh ite's chances a re better. 1 5 ..tt2

lL'ld7

1 6 ..te2

tt:'Jxc5

1 7 0-0

lL'lb3

Black, naturally, tries to exploit the weak­ ness of the d4-square. 1 8 .l:tcd 1

.U.fd8

1 9 .l:txd8+!

Rook exchanges should extingu ish Black's initiative , after which the main strateg ic defect of his position - his out-of-play bishop on h7 - will be especially perceptible.

26 r;t;d3

The centralisation of the king suggests itself. H ow can Black now parry the threat of it i nvad ing h i s position? 26 ...tt:'Je5+? 27 Wd4 f6 is useless i n view of 28 tt:'Jc8. H i s king must be u rgently brought u p to the defence.

19 . . .

.U.xd8

26 . . .

'iii>f8

20 �d1

lL'lbd4

27 'it>c4

'it>e7

28 ..tg3

21 �f1

Of cou rse, not 2 1 ..tf1 ?? ..txc3 22 bxc3 tt:lxf3+ . 21 . . .

tt:'Jxe2

22 .l:i.xd8+

tt:'Jxd8

But what to do now? 28 . . .

e5!

After 29 r;t;d5 r;t;d7 ! the pawn can not be taken : 30 ..txe5? f6 (or 30 ...�g8) , and the

226

�

The Connection of the Opening with the Endgame

wh ite king will be mated in the middle of the board ! 29 tt::lf5+

Nothing is given by 29 tt::l b 5 f6 followed by . . ..Jtg8+ - the endgame is certainly not worse for Black. Bad is 29 tt::l c8+? c;t>d? 30 'it>b5 'it'c7 ! , when there is no defence against 31 ...a6+ , wi nning the knig ht. 29 . . .

.Jtxf5

30 exf5

'it>d6

bishop and pawns on b7, d 7 , f7 and h7. It is qu ite probable that, by attacking the enemy pawns, the knight will force them to advance onto squares of the colour of the bishop and the latter will be transformed i nto a 'bad' bishop. Let us return to our position. The bishop is not so strong , and there a re more pawn weaknesses in Wh ite's position . His only active possibil ity is 31 'it>b5, and this is the one which m u st be exa m i ned in the fi rst instance . Black replies 3 1... tt:Je7 ! , not fear­ ing 32 f6 gxf6 33 c;t>a6 tt::l d 5 , when he retains his extra pawn. The variation 32 .Jtf2 tt:ld5 occu rred by transposition i n the game - we will see it i n due cou rse. It remains to check 32 'it>a6 tt::l xf5. Now 33 �xa7 is hopeless for Wh ite: 33 . . . tt::l x g3 34 hxg3 (34 �xb6 tt::le2 35 c4 tt::l d 4) 34...'it>c6 35 g4 g6 followed by ...f7-f5 and e5-e4. He is forced to play 33 .Jtf2 , but then there fol lows 33 ... �c6 34 �xa7 tt::l d 6 35 g4 (35 Jl.xb6 tt::l c8+ ) 35 . g6, and it is hard for Wh ite to defend agai nst his opponent's clear pla n : ... tt::l c4 , . ..f7-f5 and . .. e5-e4-e3 . . .

The position has changed marked ly. Let's try and evaluate it: who stands better, and by how much? Some of you gave preference to Wh ite , and you consider his advantage to be sign ificant. Why? A 'good' bishop in an open position and the more active king. No, such an evaluation is dogmatic and su perficial, and I categorically disagree with it. In fact the advantage is now with Black, and it is a considerable advantage. I will express my views regarding the relationship of knight and bishop i n endings. Even in an open position a knight may prove stronger than a bishop if both sides have weaknesses. Here is a somewhat abstract example. I magine an endgame without kings: Wh ite has a knight and pawns on a2, c2 , e2 and g2; Black has a dark-square

3 1 .Jtf2

tt::l e 7

32 c;t;b5

tt::l d 5

After 32 ...tt::l xf5 Black wou l d have had to reckon with 33 a5. 3 3 c4

tt::l c 7+

34 �b4

�c6

35 a5?!

The position is now one that few of you will want to try and uphold with Wh ite. The move in the game loses by force , but 35 g4 tt::l a 6+ 36 c;t>a3 tt::l c5 also looks completely hope­ less. 35 . . .

tt::l a 6+

36 c;t>c3

bxa5

37 .Jtxa7

tt::l c 5

38 g4

(see diagram)

ltJ

The Connection of the Opening with the Endgame

227

And now, your fi rst assignment . F r o m t h e openi ng Wh ite appa rently gained an advantage. Where did he squander it, and how can his play be i m proved?

The Slav Defence is a n old, well-tested opening weapon of mine. Associated with it a re memories, both joyfu l and sad , success­ fu l d iscoveries and vexing losses. I should now l i ke to describe to you the history of one of these discoveries. Black has not only a n outside passed pawn (and, as a result, a win i n almost any pawn endgame); also im porta nt is the fact that the white bishop is shut out of play. H ow can this factor be exploited? That's rig ht - it is a l ready time to beg i n pursuing t h e bishop. 38 . . .

t'Lla4+ !

39 'it>b3

t'Llb6

Th reatening 40 . . . 'it>b7.

Magerra mov - Kos i kov

All-U nion Qual ifying Tou rnament, Daugavpils 1978

Sla v Defence 1 l'Llf3

d5

2 c4

c6

3 cxd5

cxd5

4 d4

t'Llf6

5 t'Llc3

t'Llc6

6 i.f4

i.f5

40 �b8

f6

7 e3

e6

41 h4

t'Llc8 ! !

8 � b5

l'Lld7

9 'it'a4

'i¥b6

Complete domination o f knight over bishop. The rest is elementary. 42 'it>a4

Wb7

43 �xe5

fxe5

44 'it>xa5

'it>c6

45 'it>b4

l'Lld6

46 'it>c3

'it>c5

47 'it>d3

t'Llxc4

48 'it>e4

�d6

49 g5

h5

50 f4

exf4

51 �xf4

t'Llb6

52 'it>e4

t'Lld7

53 'it>d4

l'Llf6 !

An elegant concluding stroke. Wh ite re­ signed .

9 . . . l:!.c8 is also played . 1 0 t'Llh4

There was a time when it was thought that this move virtually refuted 9 . . . 'it'b6 . The point is that the primitive 10 . . . i.g6 al lows Wh ite , by breaking through in the centre , to gain a serious advantage: 11 t'Llxg6 hxg6 12 e4 ! dxe4? 13 d5! . 10 . . .

�e4!

(see diagram) When I began studying the position after 10 t'Llh4 , I sensed that the only defect of Wh ite's plan was the u nfortunate position of his knight on the edge of the board . This was how the move 1O . . . i.e4 ! orig inated .

228

�

The Connection of the Opening with the Endgame

b) 1 3...exd5! ? 1 4 tt'lf5 tt'lc5 1 5 .i.xc6+ bxc6 1 6 .l:txc5 �xc5 17 0-0 0-0 ( 1 7. ..g6) 18 �e5 .i.f6 19 .i.xf6 gxf6 20 'Wid1 Wh8 21 'ikh5 d4! (the game Ehlvest-Sergeev, Leningrad 1979, went 2 1 ...klg8? 22 'iixfl 'iif8 23 'ii'e 6 with chances for both sides) 22 exd4 'ikd5, and Black gains the advantage - Dvoretsky.] 1 2 0-0 a6 1 3 .i.xc6

- position after 1 O . ..te4! . .

It was first made two months earlier, i n the game Buturin-Kosikov from the 1978 U krain­ ian Championsh ip. 11 0-0 .i.e? ( 11 . . . a6 or 11. . . .l:i.c8 came i nto consideration) 12 tt'lxe4! ? dxe4 13 d5! tt'lc5! 14 dxc6 0-0! 15 'iic4 bxc6! 16 .i.a4 tt'lxa4 (less good was 16 . . . .i.xh4 17 .i.d6 tt'lxa4 18 .i.xf8 tt'lxb2 19 'ifxe4) 17 'li'xa4 .i.xh4 18 1Wxe4 .i.f6 ( 18 . . . 'ifxb2 was extremely dan­ gerous) 19 .i.e5 .i.xe5 20 'it'xe5 l:tad8 2 1 l:.fd 1 'ili'b5 ! 2 2 'ii'x b5 (22 �c3 .l:i.d5 23 e4 l:tc5 with counterplay) 22 . . . cxb5 23 'it>f1 l::I xd 1+ 24 .l:.xd 1 .l:.c8 , and the players agreed a draw. This game apparently did me a bad service . At t h e qualifying tou rnament i n Daugavpils (a few rounds before the meeting with El mar Magerramov), playing against N a u m Rash­ kovsky, I underestimated a similar break­ through in the centre and ended up in a difficult position . 1 1 .l:i.c1 ! ? .l::tc8

[Some opening information: quite possible is 11 . . ..i.e 7 12 tt.Jxe4 dxe4 13 d5, and now Black has a choice: a) 13...tt.Jc5 14 dxc6 0-0 1 5 .l:.xc5 'it'xc5 1 6 0-0 bxc6 17 .i.xc6 1J.ad8! 18 .i.g3 g5 1 9 'ikxe4 ( 1 9 .i.xe4 gxh4 20 iLf4 1J.d2!) 1 9...gxh4 with equality (Azmaiparashvili­ Dvoretsky, Tbilisi 1 980);

With what should Black recaptu re on c6? 13 . . . 'it'xc6? (hoping for 14 tt'lxe4? 'iixc1 ) is clearly bad becau se of 14 'ifd 1! . How can he defend against the th reats of 15 tt'lxe4 and 15 tt'le2 ? If 14 . . . tt'lb6, then 15 tt'lxe4 'iixc1 16 ii'xc1 .l:.xc1 17 tt'lf6 + ! gxf6 18 !!xc1. The game went 1 3 . . . 1J.xc6? 1 4 tt'lxe4 dxe4 1 5 d5! exd5 1 6 tt'lf5. Wh ite's g reat lead in development decides the outcome. There followed 1 6 . . . g6 1 7 tt'ld4 l:i.xc1 1 8 .l:i.xc1 'li'xb2 19 l:r.c8+ �e7 20 tt'lb3 ! (the si mplest way to the goa l ) 20 . . . g5 21 .i.xg5+ f6 22 'i!Vb4+ Wf7 23 'Yi'xb7 ii..e 7 24 'ir'xd5+ '.t>g7 25 'ii'x d7 .l::t x c8 26 'it'xe7+, and Black

resigned in view of 26 . . . �g6 27 �xe4+ Wxg5 28 'ikf4+ Wg6 29 'it'g4+ and 30 �xc8. And now - you r second home assign­ ment. Evaluate the consequences of 1 3 . . . bx c 6

.

Let us return to the game with Magerramov.

The Connection of the Opening with the Endgame

In reply to 1O . . . ..ie4! he made a n atu ra l , but poor move . 1 1 f3? !

..i d 3 ! !

I t is th is that constitutes t h e tactical j u stifica­ tion of Black's plan. It was also possible to play this a move earlier, but then Wh ite would have gai ned an advantage by 11 liJxd5! exd5 12 ..ixd 3 . It is i m po rtant to provoke f2-f3 , to deprive the wh ite knight of the f3-squa re . 1 2 liJxdS!

Practically forced . I n the event of 12 ..ixd3 'i'xb2 13 0-0 'i'xc3 14 ..ib5 ltJb6 Black's position is close to winning. 12 . . .

..ixbS

1 3 lLlxb6

lL1

229

Black not so m u ch threatens to regain the exchange, as vacates the c6-squa re for his 'erra nt' bishop. Now a n amusing situation a rises: Wh ite is the excha nge and a pawn u p , but both h i s knig hts a re trapped . 1 5 'itd2

If 15 .Uc1 I was planning to reply 15 . . . ..ic6 . [Here i t should be examined what happens after 1 6 '&te2 .te l ( 1 6 . . . lbd5 17 Ji.. g 3 g5 1 8 e4) 1 7 lbc7 + '&td8 1 8 a3, a n d i f 1 8 . . . lba 2, then 1 9 d5!?. It is possible that the position is still in favour of White.

In the game Glyanets-Dvoretsky (Tbilisi 1 979) 15 g4 ?! (vacating the g2-square for the knight) 1 5. . . lbd3+! 1 6 '&te2 lbxf4+ 1 7 exf4 was played.

Of cou rse, not 13 'ii'x b5? exd 5 , when Black is a piece u p . 13 . . .

..ixa4

1 4 lLlxa8

14 lLlxa4 is bad : 14 . . . ..ie7 15 ..ig3 g5 (here it is, the price of the move f2-f3 ! ) .

1 7. . .ie 7!!. An important intermediate move.

The immediate 17 . . . ..id6 is weaker: 1 8 1:!ac 1 0-0 (1B. . . 'i'li;e7?! 1 9 lbc7 Ji.. c 6? 20 d5; 1 8. . . ..ic6! 1 9 d5 exd5 20 liJf5 with some complications) 19 lbc7 Ji.. c 6 20 .:!.xc6 bxc6 2 1 lba6 Ji.. xf4, and the endgame is most probably drawn. I n itially I had been planning 14 . . . ..ie7, which leads to u nclear positions. For example, 15 b3 !? ..ixh4+ 16 'itd2 0-0 17 ltJc7 . The following move, found l iterally a few hours before the game, is far stronger. 14 . . .

lLlb4!

1 B lbg2. 18 .Uac 1 0-0 is bad for White. If 1 8 g5 there follows 1 8 . . . Ji.. d6 1 9 tr.ac 1 0-0 20 lbc7 Ji.. c 6 21 .Uxc6 bxc6 22 lba6 Ji.. xf4, and White loses a pawn. 1 8... ..id6 1 9 1lac1 ..ic6! 20 lbe3 0-0. Black has gained a material advantage and later he successfully converted it - Dvoretsky.]

230

�

The Connection of the Opening with the Endgame

15 . . .

i.. e 7

1 6 lt:Jc7+ 1 7 .l::i.h c1

39 �2

tt:'ld4

'it>d8

40 ..t>e3

tt:'lb3

g5!

41 l:td1

tt:'la5

42 .l::i.e 1

i.. e 6

The u ltra-sharp open ing skirmish has con­ cluded to Black's obvious advantage.

43 .i::t e 2

18 tt:'lg6

18 �g3 gxh4 19 i..f4 tt:'lb6 was no better for White. 18 . . .

hxg6

1 9 �g3

tt:'ld5

20 tt:'lxd 5

exd5

21 �c7+

'it>e8

22 e4

The rooks need scope, so Wh ite tries to open lines. 22 . . .

�c6

23 exd5

�xd5

24 �d6

tt:'lb6

25 �xe7

'i;xe7

26 h3

tt:'lc4+

27 'it>c3

Not 27 ..t>c2 tt:'le3+ and 28 . . . tt:'lxg2. 27 . . .

'it>d6

28 .i::t e 1

.i::t c 8

29 .l:!.e2

tt:'la3+

30 ..t>d2

.l:!.c2+

31 ..t>e1

.l::i.x e2+

32 'it>xe2

tt:'lc2

33 .i::t c 1

After 33 .i::t d 1 �xa2 the win for Black is not in question. 33 . . .

tt:'lxd4+

34 'it>f2

a5

35 a3

a4

36 'it>e3

tt:'le6

37 .l::i.c2

b5

White's position is hopeless - only elemen­ tary accu racy is requ ired . But, as is wel l known , t h e hardest th ing is t o win a won position . 3 8 l:!.d2

'it>e5

Now 43 . . . tt::l c4+ suggests itself, and, which­ ever way the wh ite king moves, Black's king breaks through on the opposite wing. I n­ stead of this I com m it a n oversight, which makes things m u ch harder for Black. 43 . . .

�f5?

44 f4+ !

gxf4+

45 �3+

'it>d4

46 'it>xf4

Wh ite has managed to redu ce the nu mber of pawns on the board , which improves his su rvival chances. 46 . . .

tt:'lc4

47 g4

�d3

48 .l::i. h 2

f5

This move is forced - the th reat of creati ng a passed h-pawn was too serious. 49 gxf5

�xf5

50 h4

tt:'le3

51 .l:!.d2+

�d3

52 'it>g5

tt:'lf5

After 52 . . . tt:'lc4 Black wou ld have to reckon with 53 l:!.xd3+ 'it>xd3 54 'i;xg6.

'2J

The Connection of the Opening with the Endgame

53 'it>xg6!

Weaker was 53 .l:i.h2 'it>e3 54 �f6 lt:Jd4 55 l:!.h3+ tt:Jf3 56 l:!.h 1 �d2 , and Black wins.

231

black king m u st help i n d riving the enemy king as far away as possible. 6 1 l:!.h2

tt:Je3

53 . . .

tt:Jxh4+

62 'it>f3

tt:Jc4

54 �f6

�c4

63 .l:th4+

�e5!

64 .l:!.h5+

.ltf5

65 .l:!. h2

.lte4+

66 � g3

�d4

67 'it>f4

tt:Je5

68 l:!.e2

lt:Jd3+

69 'it>g4

The goal has been achieved - now it is time to attack the wh ite pawns . 69 . . .

.i.d5

70 �g3

Here the game was adjourned . Analysis showed that it was not easy to convert the adva ntage, but that to make a d raw was even more d ifficult. 55 .l:i.f2

The sealed move. 55 . . .

tt:Jf5 !

56 'it>e5

tt:Je3

57 .l::td 2

Wh ite would have lost after both 57 �d6 'it>b3 58 'it>c6 (58 .l:!.f3 tt:Jc4+ and 59 . . . tt:Jxb2 ) 58 . . . tt:Jd 1, and 57 .l:!.f4+ 'it>c5 58 .l::tf2 lt:Jc4+ 59 'it>e6 b4 60 axb4+ 'it>xb4 61 'it>d5 'it>b3 62 .l::!. f3 tt:lxb2 . 57 . . .

jLg6!

58 'it>f4

58 'it>d6 ! ? 'lt>b3 59 'lt>c5 tt:Jc4 60 .l:i.d4 tt:Jxb2 61 .l:i.b4+ 'it>xa3 62 .l:!.xb5 was probably more tenacious. 58 . . .

tt:Jc2

59 .l:!.g2

jLd3

60 l:l.d2

'it>d4! !

Before embarking on decisive action, the

If 70 'it>f5 , then 70 . . . ..tf3 ! 71 .U.c2 'it>e3 72 �e6 .i.d 1 73 .l::!. h 2 .i.e2 ! proves decisive. Amazingly precise coord ination by the black pieces! 70 . . .

.i.b3!

7 1 .l::!. h 2

'lt>e3!

The th reat is . . . .i.b3-d 1-e2 . And if 72 .l:!.h5, then 72 . . . ..tc4 73 .l::!. h 2 tt:Je 1! followed by 74 . . . .i.e2 and 75 . . . tt:Jd3 . Therefore Wh ite resig ned . N atural ly, after this game Wh ite began looking for new possibilities . 1 1 0-0-0 ! ?

232

�

The Connection of the Opening with the Endgame

This was played by Artu r Yusupov against Alexander Bel iavsky in the 1979 U S S R Championship. Now 11 . . . Jte 7 does not work because of 12 f3 !.

learn to solve independently the problems facing you , not relying on the opin ion of a trainer or recog n ised experts.

11 . . .

lieS

Assignment 1 ( Palatn i k-Kosi kov).

1 2 f3

Jtg6

1 3 lt'lxg6

hxg6

14 'lt>b1

a6

U p to the 23rd move Wh ite's actions did not cause either you or me any doubts. The first questionable moment a rose in the following position .

Events developed as follows: 1 5 Jtd3 (it stands to reason that after the exchange on c6 Wh ite cannot hope for an opening advantage) 15 . . . Jtb4 16 l::!. c 1 0-0 17 a3 i.xc3 18 .Uxc3 e5! 19 dxe5 lt'lcxe5 20 'iic2 l::!.xc3 21 'ii'x c3 lt'lxd3 22 ifxd3 lt'lc5 23 'ii'd4 (of cou rse, not 23 �xd5? i n view of 23 ... .Ud8 and 24 ....Ud2) 23 . . . g5! 24 Jte5 .Ud8 . Black managed to equalise, and the

game ended in a draw. A new attempt to improve Wh ite's play was made in the game Mordasov-Vekshenkov (Al ma Ata 1980). 15 ..te2

Events later developed in similar fash ion to the previous game: 15 . . . Jtb4 16 J:Ic1 0-0 17 a3 ..txc3 18 .Uxc3 e5 19 dxe5 lt'lcxe5 , but after 20 l1hc1 l1xc3 21 llxc3 Wh ite gained the advantage. And now a th i rd home assignment. In reply to 15 Jte2 find the best plan of defence for Black, one wh ich does not involve a weakening of h i s pawn struc­ ture. D i s c u s s i o n of h o m e ass i g n m e n ts

I have to say that no one was able to g ive accu rate repl ies to all th ree assig nments. Of course, they were qu ite d ifficu lt, and be­ sides, as in general in the majority of opening positions, the solutions offered may seem ambiguous and questionable. Even so , such train ing is very i mportant and usefu l. If you wish to feel confident i n the opening stage of the game, you need to

Apparently not wishing to allow a weakening of Wh ite's queenside pawns, a g roup of you (Zviagi ntsev, Bog uslavsky, Kiryakov and Makariev) suggested the variation 23 tt'la2 il.. e 7 24 'itxe2 lt'lc6 25 lt'lc3 a6 (25 ...f5 ! ? ) 26 a5, giving preference to Wh ite i n the fi nal positio n . But why? After the thematic break­ through 26 ...f5! (wh ich , u nfortunately, was not even considered ) 27 e5 (27 exf5 ..txf5) 27 ...f4 ! the problem of the l ig ht-square bish­ op is solved . There can follow 28 lt'la4 j;_c2 29 lt'lc5 ..txc5 30 ..txc5 Jtb3 with equality. N o , this recommendation is unconvincing. I th i n k that the moves made i n the game 23 'it>xe2 ..txc3 24 bxc3 lt'lc6 are correct. We have already mentioned that Black's main problems are his bishop, which is shut out of the game, and his less centralised king. Wh ite's success largely depends on whether he ca n qu ickly 'engage' the oppo­ nent, by i n itiati ng concrete play before the approach of his main forces.

CZJ

The Connection of the Opening with the Endgame

In this connectio n , I l i ked the idea of Svidler, Baklan and Eme l i n , who suggested 25 a 5 ! , and then 2 6 l2Jd6 l2Jxa5 27 il_xa7 followed by 28 il_b6. Wh ite does i ndeed have a n obvious advantage. But the move i n the game 25 l2Jd6 ! is no worse . On ly, i n reply to 25 . . . b6 i n stead of the routine 26 'it>d3? Wh ite should h ave played more concretely: 26 c4! . For example, 26 . . .f5 27 c5 bxc5 28 i;_xc5 fxe4 29 fxe4,

and only then 30 'it>d3 with excellent winning chances.

233

i nsufficient com pensation for the pawn . [ The balance can be retained by 19 'ikd4! (instead of 1 9 il_ e5) 1 9 . .cxd4 20 .UxcB+ 'ikdB 2 1 .l:txdB+ 'it>xdB 22 exd4 - Dvoretsky.] .

16 . . . ii..c 5! ? . An alternative , apparently also qu ite possible, was suggested : 16 . ..tLlc5 17 'ifd4 ii'd8 18 lLlf5 l2Je6. 17 b4 ! ? 'ikxb4 1 8 'ikxb4 ii..x b4 1 9 I::!. x d5 g6!

with chances for both sides. Ass ignment 3 (Mordasov-Vekshenkov)

And Wh ite's decisive mistake after 26 'it>d3? Wf8 27 'it>c4 'it>e7 28 i;_g3 e5 was the move 29 lLlf5+?. Ass ignment 2 ( Rash kovsky-Kosikov).

Most of you expressed the opinion , with which I agree, that by choosing 1 3 . . . bxc6! Black would have gained a satisfactory position.

Maka riev, Zviagi ntsev and Kiryakov recom­ mend 15 . ..it'a7 followed by 16... b5. An i nteresting plan, but, as Yusupov men­ tioned , not without its dangers . There can fol low 16 l:!c1 b5?! 17 tZ:lxb5! axb5 18 il_xb5 it'xa4 19 ii.. x a4 l2Ja7 20 .l::t x c8+ l2Jxc8 21 .l:!.c1 tZ:lb6 22 ..lib5, and the th reat of a2-a4-a5 is very serious .

Emel i n and Baklan found the i nteresting variation 14 .l::tfd 1 ii.. e 7 15 f3 g5 16 fxe4 gxf4 17 lLlf3 dxe4 18 l2Jxe4 fxe3 with an u nclear game. This is probably so, but i n the fi rst instance one should consider the sharp attempt by Wh ite to break through i m medi­ ately i n the centre . 14 l2Jxe4 dxe4 1 5 d5 exd5 1 6 .i::tfd 1 I n the event of 16 lLlf5 g6 17 l2Jd6+ i;_xd6 18 i;_xd6 c5! 19 ii.. e 5 f6 and 20 . .. 1!t'c6 he has

All t h e other students chose 1 5 . . .l2J a 5 16 .l::tc 1 l2Jc4 with a not altogether clear posi­ tio n , i n which after 17 b3 I would neverthe­ less prefer to play Wh ite. My suggestion is 1 5 . . . i;_e7 1 6 e4 (otherwise Black has no problems) 1 6 . . . 'ikd8 ! ! . With this u n p retentious move, vacati ng the b6square for his knight, Black succeeds in maintaining his centre. For example: 17 exd5 tZ:lb6 18 'ii' b 3? ( 18 'ikc2 tZ:lb4 19 'it'b3 is better) 18 . ..l2Ja5! 19 'it'c2 t2Jxd5 , and Black has the advantage.

234

� Mark Dvoretsky

In the Footsteps of one Game (some non-theoretical reflections)

All theory, dear friend, is grey, But the golden tree of life springs ever green. J. W. von Goethe

A open ing, but it by no means ends there .

game of chess only beg ins with the

It sometimes happens that t h e treatment of the open ing determ i nes the fi nal result, but far more often the outcome depends on the skill of the two contesta nts in the subse­ quent stages of the game. And nevertheless many young players spend all their free time only on strengthening their open ing reper­ toire, by a brief examination of cou ntless recent games, publ ished i n magazi nes, bul leti ns or lnformators, or recorded on computer discs . As a result, i mperceptibly to themselves they merely become narrow specialists. Without pu rposefu l work on the middlegame and the endgame, knowledge in these fields remains patchy, and the general understanding of chess suffers . I am convinced that a h ighly cu ltured , harmo­ niously prepared player is bound to do better than an 'open ings-expert' . On a brief examination of a game, beh ind its bare score it is very hard to see the pro­ blems faced by the opponents , the ideas behind the moves made, and pretty varia­ tions which remained off-stage. I prefer to study games by strong players , fu rn ished with detailed comments, desirably written by one of the two contestants. Such com ments not only draw the readers i nto the creative laboratory of a g randmaster, but often provoke a desire to argue with the conclu­ sions suggested , and stimulate your own analytical searches.

Chess games hardly ever repeat them­ selves. But situations a rising in them , ideas and typical methods - these are often repeated. Sometimes one insign ificant epi­ sode provokes a long cha i n of associations, and similar instances in you r own games or those of others are remembered . Such associations are very useful - they hel p the stud ied material to be repeated and con soli­ dated . Many of the tales from the cycle A Thousand and One Nights are l i n ked by common heroes or develop from one and the same i n itial situation . Without cla i m ing to be a Scheherazade , I will acquaint the readers with severa l chess stories, in the remember­ ing of wh ich I was helped by one not especially interesting game (or more pre­ cisely, its fi rst half). I hope that these stories will provide a reasonable i l l u stration of the ideas expressed i n the preamble to the lectu re . I n trod u ct i o n

Gavri kov - Dolmatov

Tal l i n n 1985

Queen 's Gambit 1 t'bf3

t'bf6

2 c4

e6

3 t'bc3

d5

4 d4

iLe7

CLJ

In the Footsteps of one Game

5 i.g5

0-0

6 e3

h6

7 i.h4

b6

8 i.d3

.1 b7

9 0-0

lL'lbd7

1 0 'fke2

c5

1 1 .l:!.fd 1

ltJe4

235

1 2 .1g3

What can be said about the system of development chosen by Wh ite? It seems to me that its strateg ic idea is based on the uncomfortable opposition for the opponent of the wh ite rook and the black quee n , which has no su itable squ a re to go to, such as e2 in the wh ite position. A similar motif also occu rs in other openings (for example, i n the Tarrasch Defence to the Queen's Gambit), but here we will not develop this topic, remembering that at our d isposal we do not have a thousand and one nig hts . 12. . .

ltJxg3

1 3 hxg3

dxc4?

Here or on the previous move it was better to play . . . cxd4. We will return to this moment, although some time later. 1 4 i.xc4

i.xf3

What else? Black has to reckon with both 15 dxc5 and 15 d 5 . Bad is 14 . . . cxd4 15 ltJxd4, when the terrible 16 ltJxe6 is th reatened . 1 5 gxf3 !

cxd4

1 6 exd4!

In the event of 16 .l::tx d4 i..f6 followed by 17 . . . 'it'e7 Black would have time to safe­ guard his queen agai nst the action of the enemy rook. But now it is hard to defend against the thematic central breakthrough d4-d5! 16. ..

1 8 .l:tac1

.l:!.c8

1 9 i.a6

.l:!.c7

20 iVf3

'it'd7

21 a3

.l:!.d8

22 i.d3

g6

23 it'e2

.1f8

Sergey Dolmatov has successfu lly deployed his forces and has completely equal ised . The game contin ued for a long time yet (and was far from fau ltless) , and i n the end it fi n i shed in a d raw. But one gains the impression that Wh ite stood better and did not exploit all this chances, wou ldn't you agree? At some point he delayed . But where? F i rst story : b i s h o p o r k n i g ht?

.1d6

1 7 f4

17 d5 was prematu re i n view of 17 . . . e5 1 8 tt:Je4 '1We 7 . 17 . . .

Only here did the play deviate from the sou rce game played thirty ( ! ) years earlier. G l igoric-Unzicker (Olympiad , Dubrovn i k 1950) developed as follows: 1 7 . . J1e8 18 i.b5 ! .l:!.c8 ( 18 . . . .l:!.e7 19 ltJe4 and 20 d 5 ! ) 19 ltJe4 'fke7 20 d5! e5 2 1 'ikg4 .l:!.ed8 22 i.xd7 .l::tx d7 23 ltJxd6 'fkxd6 24 fxe5 'fie? 25 d6, and Wh ite won .

ltJf6

Wh ite did not i n fact put i nto effect his main positional threat, d4-d5 . Let us return to the position after Black's 17th move. Why not break through i n the centre at this moment?

236

�

In the Footsteps of one Game

The answer is simple: both players saw the variation 18 d5 exd5 19 tt:Jxd5 tt:Jxd5 20 i.xd5 (20 .Uxd5 'iif6) 20 .. Jk8 . On the next move the black queen goes to f6 and White's pressu re on the d-file evaporates. If this is so, why not try captu ri ng on d 5 not with the knight, but the bishop? 18 d5!

Dvorets ky - Romanov

Moscow 1963

Nimzo-lndian Defence 1 d4 tt:Jf6 2 c4 e6 3 tt:Jc3 i. b4 4 e3 c5 5 l2lf3 d5 6 i.d3 0-0 7 0-0 cxd4 8 exd4 dxc4 9 i.xc4 b6 1 0 i.g5 i.b7 1 1 'ii'e 2 i.e7 1 2 l:l.fd1 'iic 7? 1 3 i.b3 tt:Jc6 14 .l:!.ac1 �ac8

exd5

19 i.xd5! !

If 19 . . . tt:Jxd5 , then both 20 tt:Jxd5 (the black queen has no conven ient square ) and 20 .l:txd5 are strong. 19 . . . .Uc8 is better, but here too Wh ite retains some in itiative , by conti nu­ ing (if there is noth ing better) 20 i.b7 ! ? with the th reat of 21 tt:Jb5 (the immediate 20 tt:Jb5 is weaker in view of 20 .. .'�We7 21 �xe7?! ii..x e7 22 tt:Jxa7 �c2 and 23 . . . ii.. c5) . Grandmaster lossif Dorfman once half­ jokingly, half-seriously formu l ated the princi­ ple: 'The worst bishop is always better than the best knig ht' . Apparently, somewhere i n our consciousness w e agree with h i m , because w e frequently overlook moves such as 19 i.xd 5 ! ! - it is a pity to al low the exchange of a bishop. Overcoming this psychological barrier could be helped by familiarity with situations i n which s i m i l a r non-routi ne decisions were taken . ('Similar' and 'non-routine' - at first sight, what incompatible words! But, as Isaak Lipn itsky wrote in his remarkable book Voprosy sovremennoy shakhmatnoy teorii (Problems of modern chess theory): 'Crea­ tively concrete decisions are by no means some negation of chess generalisations, since the negation of one set of often obvious rules and laws occurs when other, perhaps more latent rules and laws, are established. ' Here are two examples on the given topic from my own games.

Please don't judge the preced ing play too severely - after a l l , at that time the two players had only second category rating! But even today I have no reason to criticise my subsequent actions. H owever, at a young age non-routi ne decisions a re some­ times found more qu ickly than when one is more matu re: due to lack of experience and knowledge, stereotypes have not yet had time to form . 1 5 d5!

exd5

1 6 i.xd5!

.Ufe8

1 7 �c4

tbxd5?

1 8 tt:Jxd 5

�b8

1 9 i.f4

tt:Ja5

20 tt:Jxe7+

I::. x e7

21 �xeS+!

Black resigned . The fol lowi ng example, played i n a n event of far h igher standard , will be analysed in more deta i l .

In the Footsteps of one Game

G u l ko - Dvoretsky

43rd U S S R Championship, Yerevan 1975

Ragozin Defence 1 d4

lt:Jf6

2 c4

e6

3 lt:Jc3

� b4

4 lt:Jf3

d5

5 e3

0-0

6 ii.d3

lt:Jc6

7 0-0

a6

tLJ

237

This variation of the Ragozin Defence was handled somewhat d ifferently by M i khail Botv i n n i k. The first game of his Return Match i n 196 1 with M i khail Tal went 4 e3 0-0 5 i.. d 3 d5 6 a3 dxc4 7 i.. xc4 i.. d 6 8 li:Jf3 lt:Jc6 9 li:Jb5?! e5 (9 ...i.. e 7 ! ? ).

A purely prophylactic move , provoked by the

need , when carrying out the main plan dxc4 8 �xc4 ii.d6 ( i ntend ing 9 ...e5), to reckon with 9 lt:Jb5 and 9 ii.b5! ? .

7

...

However, accord ing to theory t h e fi rst of these is not dangerous . I n the afore­ mentioned book Lipn itsky gives the interest­ ing game Ba n n i k-Cherepkov ( 1952 ) , wh ich is of i m med iate releva nce to our theme . In it after 9 lt:Jb5 'ii'e 7 ! 1 0 i.. d 2 (also noth ing is g iven by 10 lt:Jxd6 cxd6 followed by 11.. .e5) 1 0 lt:Je4 1 1 �e1 e5 12 dxe5 . . .

10 lt:Jxd6 ifxd6 ! 11 dxe5 'it'xd 1+ 12 'it>xd 1 lt:Jg4 13 '>t>e2 lt:Jcxe5 with an active position for Black. Why d id n 't Botv i n n i k play 10 dxe5 lt:Jxe5 11 lt:Jxd6 with a n advantage? We a l ready know the a n swer - because of 10...ii.xe5!. Later Botv i n n i k improved Wh ite's play: 9 b4! e5 10 i.. b 2 i.. g 4 1 1 dxe5 ( 11 d 5 lt:Je7 , as i n the th i rd game o f t h e Retu rn Match , leads to an u nclear game).

Black did not play 12 . . . lt:Jxe5? (on account of 1 3 li:Jxd6 with the better cha n ces for White ) , but 1 2 . . . i.. x e5! . The su bsequent events too k a tense cou rse: 13 lt:Jxe5 lt:Jxe5 14 ii. e 2 .l:.d8 1 5 'ii'a 4 lt:Jd2 16 lt:Jxc7 i.. d 7 17 ifa5 lt:Jxf1 18 lt:Jxa8 lt:Jxe3 19 fxe3 .l:.xa8 20 'i¥b4 'i¥xb4 2 1 i.xb4 , and the game ended i n a d raw.

238

�

In the Footsteps of one Game

I th ink you will agree that now the move 11...1i.xe5! very much suggests itself (and then, accord ing to Konstantinopolsky - 12 'ifxd8 .l:!.axd8 13 b5 1i.xf3 14 gxf3 l2Ja5 with equal ity). We are surprised to learn that Tal chose 11...l2Jxe5? !. After 12 .fi.e2 "ike? 13 l2Jb5 lifd8 14 'ifc2 a6 15 l2Jxd6 cxd6 16 'ti'd 1 the position was markedly worse for him. 8 h3!

An important prophylactic move, the point of which becomes clear on an examination of the variation 8 a3 dxc4 9 i.. xc4 (9 i.. x h7+ ! ? ) 9 ...1i.d6 10 e4 e5. Now i t would b e adva nta­ geous for Wh ite to retai n the tension in the centre, but after 11 i.. e 3 both 11...l2Jg4 and 11...1i.g4 are strong. If he wishes, Black can conti nue 'jockeying for position' by 8 ...h6!? (it is usefu l to deprive the wh ite pieces of the g5-square), but he decides to clarify the situation. 8...

dxc4

9 1i.xc4

i.. d 6

10 e4

e5

11 i..e 3

At the moment when this game was played , I trusted Lipn itsky's eval uatio n , which as­ su med that Black gained good cou nterplay by activity on the queenside: 11...h6 12 .l:::i.e 1 b 5 13 1i.b3 1i.b7 (now theory regards this d if­ ferently - the Encyclopaedia of Chess Open-

ings eval uates the variation in favour of Wh ite). Nevertheless, I made another move. 11 . . .

exd4?!

Why? Behind this there is a rather cu rious story. I n those yea rs I was helping Botvinnik to conduct lessons with talented young players at the school d irected by him. Not long before the championship of the cou ntry I showed the j u n iors the game Taimanov­ Fischer ( Buenos Aires 1960), which was full of fasci nating events i n all its stages. I n his youth Robert Fischer used to employ the Ragozi n Defence. On reach ing this same position , he chose 11...exd4? ! 12 l2Jxd4 1i.d7 13 l:te 1 ike?? ( 13 ...l2Je5 14 i.. f 1 l2Jg6 was preferable, althoug h , as shown by the game Keres-Lipn itsky, played in the 19th USSR C h ampionsh i p , 195 1, the plan of g2-g3, i.. g 2 and f2-f4 ensures Wh ite a n enduring i nitiative) 14 1i.g5! l2Jxd4 15 l2Jd 5 ! 'ife5 16 f4, and Mark Taimanov gai ned a decisive advantage. One of the j u n iors asked why Fischer did not simpl ify the position by 12...l2Jxd4 13 1i.xd4 (if 13 'i*'xd4 ! ? c5 14 'ti'd2 there follows not 14...b5? 15 lifd 1 i.e? 16 'ifxd8 .l:I.xd8 1 7 l2Jd5 l2Jxd5 18 i.. x d5 .l:I.a7 19 b4 , but 14...1i.e5 ! , th reatening the exchange o n c3 followed by the captu re of the e4-pawn) 13 ...c5 or 13 ...b5.

ltJ

In the Footsteps of one Game

239

I was ready for th is question (since I had analysed the position beforehand), and I explai ned that if 13...c5 Wh ite does not play 14 i.e3?! b5, but 14 i.xf6 ! 'iixf6 15 f4 , intending e4-e5, when he stands better. Stronger is 13...b5! 14 i.b3?! c5 15 �xf6 'ii'xf6 16 f4 (or 16 tiJd5 'it'e5 17 f4 �xe4) 16...c4 17 e5 �c5+, and Black solves a l l his problems. But instead of 14 �b3?! he has to reckon with 14 e5!.

- position after 12 ..txd4 ! ! -

1 4 e5?

bxc4

Agai n , so not to deviate too much from the main theme, I won't mention the subsequent d ifficult struggle and the very clever trap in which G u l ko finally caught me. However, this game, l i ke the Taimanov-Fischer game, can be found i n the appendix to the lectu re.

And here the twelve-year-old Garik Kasparov suggested the brilliant cou nter-stroke 14... c5! ! . The main variation was qu ickly fou n d : 15 exf6 cxd4 16 fxg7 .l!te8 17 i.d5 dxc3 18 .ixa8 cxb2 19 .l!t b 1 iNf6 , and Black has excel lent compensation for the sacrificed exchange. Against Boris G u l ko I decided to use this novelty, found i n a joint ana lysis at the Botvinnik School. Alas, my opponent fore­ stalled me. He captu red on d4 not with the knight, l i ke Ta imanov, but with the bishop. 1 2 ii.xd4! !

(see diagram) After pondering over the position , I realised that I was u nable to combat successfully the threat of e4-e5!. 12 . . .

ttJ xd4

1 3 'iix d4

b5

Here, fortunately for me, Boris was too hasty.

Wh ite could have reta i ned the advantage by 14 �b3! c5 15 '*'e3 c4 16 �c2 ! , and only if 16 ...b4 - 17 e5! bxc3 ( 17 ...ii.xe5 18 ttJxe5 bxc3 19 'ii'x c3 i.e6 20 l2Jxc4) 18.llad 1. The mistake made by my opponent can be interpreted , if we remember a well-known psychological chess principle: 'A threat (in this case e4-e5) is often stronger than its immediate execution '. But th is is a topic for a qu ite d ifferent conversation ... S e co n d story : study t h e c o m m e nts of g ra n d m aste rs !

Let us again retu rn to the place where we bega n - to the divergence between the games Gligoric-U nzicker and Gavrikov­ Dolmatov. Let us open Svetozar Gligoric's book I Play Against Pieces. In a note to Black's 17th move i n his game against Wolfgang U nzicker he g ives the variation 17 ...l2Jf6 18 d 5 exd5 1 9 i.xd5! llc8 20 tiJb5. Why did Viktor Gavrikov not make use of this 'prompt'?

240

�

In the Footsteps of one Game

He didn't know the sou rce game? U n l i kely, since he had already employed th is varia­ tion of the Queen's Gambit many times, and Gavrikov works assiduously on opening theory. He had forgotten? Altogether i mprobable Vi ktor has a phenomenal memory, and he appears to remember absol utely everything. I th ink the point is that players who are fasci nated by opening theory try to qu ickly digest as much fresh i nformation as possi­ ble and are often not incli ned to spend time on a thorough study of the games they examine, on an analysis of the commentar­ ies on them . Apparently Gavri kov either did not notice Gl igoric's comment, or he had not read his book at all and had only seen the Gl igoric-Unzicker game (or more precisely, its opening stage) in an open ing book. Many years ago a similar story occu rred with Yu ri Balashov - a player with the same kind of absolute memory as Gavrikov, and with the same approach to the study of the open ing. Balashov - Dvoretsky

USSR Spartakiad, Moscow 1967

King 's Indian Defence 1 d4

ltJf6

2 c4

g6

3lLlc3

Ji.g7

4 e4

d6

to the certa i n conclusion that I was not very fam i l i a r with opening theory. I n the Four Pawns Va riation it is not possible to act on general g rounds - one is obl iged to play ' move-by-move' and in sharp positions fi nd the only correct conti nuations. To conduct such a struggle without a n accu rate knowledge of the extensive theory of the variation is very d ifficu lt. Balashov was highly skil led i n the art of opening preparation for a specific opponent. I remember how i n the U S S R Championsh ip F i rst League i n 1 974 he defeated Yuri Razuvaev with Black in the main variation of the G ri.i nfeld Defence , in which his opponent was a major expert. After the game Balashov stated that the plan chosen by Black had a l ready been employed not long before by Vasily S myslov. ' How did I miss this game, seeing as I carefu l ly follow the theory of this variation ? ! ' Razuvaev lamented. I n reply Balashov smiled cu n n i ngly: 'You see , it was only publ ished i n the magazine The Chess Player, and you don't receive it.' 5 . . .

0-0

6lLlf3

c5

7 d5

e6

8 i.. e 2

exd5

9 cxd5

:es

1 0 lLld2

A clever and, for me, extremely u n pleasant choice of opening variation.

d id n 't know anyth ing about this move (however, I also d id n 't remember the sharp variations a rising after 10 e5). I began looking how to deviate from theory, and I made an u nexpected move.

At a train ing session before the Spartakiad, Yu ri and I were sharing a room. It was assumed that he would be playing on the junior board for the Moscow team , and I was his 'understudy'. However, the d irectors of the Russian team managed to get Balashov to play for them and they withdrew h i m from the Moscow team's training session. Our brief contact was sufficient for Yuri to come

' Bravo , excellently devised , boozer nose!', said Bu ratino to the joiner G i u seppe. (Bura­ tino - a hugely popu lar Russian pu ppet cha­ racter, loosely based on P i nocch io and created by Alexey Tolstoy - translator). But, u nfortu nately, I was a l ittle late : a year earlier, at the Olympiad in Hava n a , th is had already been played by Robert Fischer. His

5 f4

10 . . .

c4! ?

In the Footsteps of one Game

opponent, Artu ro Pomar, made a poor response: 11 ..tf3 tLlbd7 12 0-0? b5! 13 'it> h 1 a 6 14 a 4 l:tb8, a n d Black gained the advantage. 1 1 a4

tLla6

1 2 0-0

tLlc5

1 l e5!?

The qu iet 13 ..tf3 was probably preferable. Incidentally, on this topic a fai rly extensive theory has now accu m u l ated.

241

20 tLlxe4

..txa1

21 tLled6

.tel!

22 'ii'x cl

I breathed a sigh of relief, since I had not seen a clear reply to 22 'ii'c 1, although I was hoping that I would find one in case of necessity. After the game I asked Balashov why he d id n 't play this. ' Risky' , he shrugged his shoulders. 22 . . .

'iVxd6

1l . . .

dxe5

2l tLlxd6

l:.xcl

1 4 tLlxc4

e4! ?

24 ..td4

l:tdl

25 tLlxe8

l:txd4

2 6 l1e1

'it>f8

Afraid of coming under an attack, I decided on a positional pawn sacrifice. 14...exf4 15 i.xf4 tLlce4 ! was also possible. 1 5 .tel

tLldl

1 6 ..txdl

exdl

1 7 'ifxdl

..tfS

1 8 'iVd2

.l:tc8

1 9 bl

27 tLld6

.l:.xd5

28 l:r.e8+

'it>g7

29 tLlxb7

..te6

It is clear that the game should end in a d raw. That in fact happened , although only after many adventu res. One of the instruc­ tive episodes i n the rook endgame that soon a rose is analysed i n my book School of Chess Excellence 1 - Endgame Analysis in the chapter ' Rook against pawns'. After the game Balashov took me aback, by saying that 'all this had a l ready happened ' , and referri ng t o a n a rticle b y M i khail Tal about the i nternational tou rnament in Mal­ lorca i n issue N o . 5 of the magazine Shakh­ maty in 1967. Of cou rse, I found the maga­ zine and saw there the game Pomar-Toran (Palma de Mallorca 1966 ) , which up to the 19th move took exactly the same cou rse. But what staggered me most of all was Tal's com ment on Black's 19th move (tLlg4 ) 'During the game I thought that 1 9 . . . tLle4 20 ttJxe4 ..txa 1 21 tLled6 ..tc3 was stronger. But analysis showed that by continuing 22 'ikc 1 ! White retains an advantage, both material and positional. ' .

When he sacrificed the pawn Black calcu­ lated as far as here and he was hoping that in the resulting position he would fi nd sufficient tactical resou rces . 19 . . .

tLle4

Here I calculated a variation which leads al most by force to an acceptable ending for Black.

It was not without reason that I feared this move. But why did Balashov, who remem­ bered the game, the issue of the magazine

242

�

In the Footsteps of one Game

where it was published (and , it would seem, even the page nu mber), not make use of the resou rce suggested by Tal? The answer is already known : it is probable that he had merely seen the game, but had not stud ied it, and had not taken an interest in the commentary. Meanwh ile, objectively the captu re on c3 is not a mistake, and Wh ite lost his advantage only on the next move . After 22 'i!Vxc3 'ii'x d6 he had the brilliant possibil ity 23 'ii' h 8+ ! ! �xh8 24 i.d4+ �g8, a n d only now 2 5 tt:'lxd6, when h e remains a healthy pawn to the good . Wh ite's spectacu lar zwischenzug was overlooked by both players - it was discovered more than 30 years later by grandmaster Viorel Bologan , when check­ ing my analyses on a computer. I should mention in passing that nowadays nearly all players - from ordinary amateu rs to lead ing grandmasters - make active use of computers i n the study of sharp opening variations. Playing prog rams or analytical modu les help to avoid tactical errors and suggest latent resou rces i n a position - as a result, the qual ity of opening analyses and the speed with which they are carried out have increased marked ly. Alas, there is also the other side of the coi n , which is often forgotten . You r tactical vision needs con­ stant training, as otherwise it will let you down at the board . By trusti ng the 'electronic nan ny' , we get out of the habit of checking for ou rselves the correctness of ideas fou n d . A s a result o f t h e lack o f training, the correspond ing skills, so important for any player, are also weakened . As regards the evaluation of the position , it can be concluded that 1 9 . . . tt:'le4 is insuffi­ cient for equal ity. Black should go in for the brill iant combination, found by Roman Toran after an hour and a half's thought in the afore-mentioned game. 19 . . .

tt:'lg4!

20 i.d4

..txd4+

21 'ilt'xd4

.l:txc4 ! !

Black's idea is illustrated b y t h e variation 21 . . . 'ilt'h4 22 h 3 l:!.xc4 23 bxc4 l:!.e3 24 hxg4 .l:!.h3 25 gxh3 'ilt'g3+ with perpetual check. But it is not enough to find a n idea - it must be implemented in the most accu rate way. By playing 24 li'd2 ! (instead of 24 hxg4?) 24 .. J!td3 25 'ilt'e 1 Wh ite would have parried the attack. Therefore Tora n changes the move order. 22 bxc4

l:!.e3 !

Threatening both 23 . . . 'ilt'h4 , and 23 . . J:td3 . 23 h 3

l:!.d 3 !

2 4 'it'xa7

'ii' h 4

25 .U.a2 !

The only possibil ity of conti n u i n g to play for a wi n . We a l ready know that 25 hxg4 �h3! leads to perpetual check, while if 25 tt:'le2? there follows 25 . . . l:!.xh3 ! 26 gxh3 'ilt'xh3 27 l:!f2 i.e4 . 25 . . .

tt:'le3

26 'ilt'b8+

�g7

27 'ilt'e5+

27 . . .

f6?

The losing move ( Black was most probably i n severe time-trouble). 27 . . . �g8? was also wrong , if only because of 28 l:!.b2! tt:'lxf1 29 l:txb7. But after 27 . . . Wh6 the game would apparently have ended i n a d raw: 28 tt:'le4

In the Footsteps of one Game

243

t'Llxf1 29 t'Llf6 t'Lle3 30 t'Llg8+ 'it>h5 3 1 t'Llf6 + . Wh6

28 'ii'e 7+ 29 t'Lle4!

From this point onwards it is only Wh ite who is attacki ng. The th reat is 30 'ii'f8+ 'it?h5 3 1 t'Llxf6+. 29 . . .

.txe4

30 'iix e4

l:!.b3

31 �b2 ! ?

f5

32 �xb3 !

fxe4

3 3 l:!.xe3

Wh ite has a decisive material advantage. 33 . . .'iVe7 34 'it>h2 g 5 35 fxg5+ Wxg 5 36 .l:!.fe 1 �c7+ 37 'it?h1 �xc4 38 l:!.xe4 'ii'c 3 39 Ud 1 'ii'c 2 40 llee 1 Black resig ned . T h i rd sto ry : what h a s my o p p o n e n t devised?

Let us again return to that with wh ich we beg a n . Remember that Dolmatov did not m anage to gain equal ity against Gavrikov. A sound plan of action for Black was demon­ strated i n the game which we will now examine. Gavri kov - Yus u pov

I nterzonal Tou rn a ment, Tu nis 1 985

Queen 's Gambit 1 d4

t'Llf6

2 t'Llf3

d5

3 c4

e6

4 t'Llc3

.te7

5 .tg5

0-0

6 e3

h6

7 i.h4

b6

8 i.d3

i.b7

9 0-0

t'Llbd7

1 0 "it'e2

c5

1 1 i.g3

t'Lle4

1 2 Ufd 1

12 . . .

cxd4!

1 3 exd4

If 1 3 t'Llxd4 Artur Yusu pov was i ntend ing 1 3 . . .t'Llxc3, i n order to weaken the oppo­ nent's press u re on d 5 and to obta i n a shelter for the q ueen on the c-file. 1 3 . . . t'Llxg3 1 4 hxg3 t'Llf6 i s also possible, for example: 1 5 .l:!.ac 1 i.b4! 1 6 cxd5 i.xc3 1 7 .l:.xc3 t'Llxd5 1 8 l:!.cc1 .l:.c8 1 9 i.a6 , and here i n the game Novi kov-Lputian ( 5 1 st USSR Champion­ shi p , Lvov 1 984 ) the players ag reed a d raw. 13 . . .

t'Llxg3

14 hxg3

t'Llf6

In contrast to the Gl igoric-U nzicker and Gavri kov-Dolmatov games, here Black se­ curely controls the important d5-point. Not long before the I nterzonal Tou rnament, Yusu pov headed the Moscow Pioneers Palace tea m in the competition sponsored by the newspaper Komsomolskaya Pravda (grandmaster captains g ive simu ltaneous d isplays with clocks against the jun iors from the other teams). Artur later told me that i n t h e match agai nst B a k u in o n e o f t h e games he employed this opening variation with Wh ite , but he did not gain any advantage h i s young opponent played the open ing stage extremely accu rately. As it transpired , the j u n iors of the Baku team had been helped i n their preparations by their captai n - Garry Kaspa rov. W h e n he met Gavri kov,

244

�

In the Footsteps of one Game

Artu r tried to remember th is game . . . 1 5 ltJe5

:cs

1 6 .U.ac1

16 ltJb5? is clearly premature: 1 6 . . . dxc4 (the immed iate 1 6 . . . a6 1 7 ltJa7 l:tc7 is also possible) 1 7 Ji.xc4 a6 1 8 ltJa7 l:tc7 with the th reats of 1 9 . . . b5 or 1 9 . . . 'jj'a 8. But now Black has to reckon with ltJb5 . 16 . . .

dxc4

1 7 ii.xc4

ltJd5

18 Ji.b3

ltJxc3

sharply improve in positional play, the technique of converting an advantage, and various other fields.

Of cou rse , g randmaster Yusupov developed the skill of prophylactic th i n king long ago, from the time when he was not yet a g randmaster. Let us see this ski l l in action .

19 bxc3

Among the numerous deep conceptions expounded by Aaron N i mzowitsch i n his book My System , one of the most i mporta nt seems to me to be the idea of prophylaxis. Ni mzowitsch wrote: 'Neither attack nor defence is, in my opinion, a matter properly pertaining to position play, which is rather an energetic and systematic application of prophylactic measures. ' This idea seems paradoxical and incomprehensible, but 1 hope that an analysis of the present game wil l help it to be understood . This is what N imzowitsch understands by prophylaxis: 'What it is concerned with above all else is to blunt the edge of certain possibilities which in a positional sense would be undesirable. ' He considers two forms of prophylaxis: the over-protection of strateg ically important poi nts and the pre­ vention of freeing pawn moves. But prophy­ laxis can be understood more widely, as the prevention not only of pawn moves by the opponent, but also other ideas of his. Therefore from the standpoint of the practi­ cal player it is more usefu l to tal k not even about prophylaxis, but about ' prophylactic thinking' - an inner resolve to identify the opponent's ideas. I will disclose one my professional secrets, confi rmed by my entire experience of training work: a player who masters the skill of prophylactic thinking will significantly raise his standard, and

Artu r asked h i m self: 'What does the oppo­ nent want, and what would he play if it were h i m to move?' Perhaps he was i ntending to sacrifice his knight on f7? ( However, it is not clear whether the sacrifice is correct. ) Black certainly has to reckon with activity in the centre by c3-c4 and d4-d5 . Is that all? N o , there is also another th reat: 'ii'd 3 a n d then ii.c2 . The move . . . f7-f5 weakens the posi­ tion too much , while in reply to . . . g7-g6 there follows the knight sacrifice on g6. Now, knowing the opponent's ideas, it is easier to choose you r move . For example, 1 9 . . . Ji.a3 comes i nto consideration - after 20 llc2 it is not easy for Wh ite to set up the battery on the b 1 -h 7 d iagona l . But does this move help against c3-c4 and d4-d5, and with what plan can it be l i n ked? ( Remember, N i mzowitsch demands 'an energetic and systematic application of prophylactic meas­ u res . ' ) 19 . . .

l:t c7 ! !

A g randmaster move! Black improves his

ctJ

In the Footsteps of one Game

position and prepares . . . 'ifc8 (or . . . 'ii'a 8), which is a good antidote to Wh ite's offensive in the centre . Now almost certainly the knight sacrifice on f7 will be incorrect. 20 'ii'd 3

245

a bout his own active possibilities. 24 . . .

g6

..if6

It transpires that after 2 1 ..ic2 g6 Wh ite cannot play 22 tt::l x g6? fxg6 - the black roo k participates i n t h e defence a l o n g t h e 7th rank. 2 1 tt::l g 4

h5!

Parrying the obvious threat of 22 ..ic2 g6 23 tt::lx h6+. 22 tt::le 3

22 tt::lxf6+ 'ifxf6 would have led to equal ity, but for the moment Gavri kov is hoping for more.

25 .l:Ie1 ? !

Wh ite decided t o improve the placing o f h i s rooks, b y playing .l:!.e 1 and .l:!.cd 1 . A sensible operation? U ndoubtedly. However, abso­ lutely i ll-timed . Now it was his turn to employ 'prophylactic thinking' and forestall Black's prosaic idea of activating his forces: . . . c.itg7 and . . . �h8. This problem would have been most simply solved by 25 'ii'e 3! (but not 25 'ifh3? i.g5! and 26 . . . 'it>g7), for example, 25 . . . i.g7 (25 . . . c.itg7? 26 'ifh6+; 25 . . . i.g5 26 f4) 26 lt:Je5 with roughly eq ual chances. 25 . . .

'it>g7

26 �cd 1 ?

But now what does Wh ite want? E ither, as before , c3-c4 and d4-d5, or 23 d5 i m medi­ ately, i n order to then support the d5-point by c3-c4. 22 . . .

.l:td7 !

Now 23 c4? .l:Ixd4 is not possible, while i f 2 3 d5 Yusupov was i ntending 23 . . . i.g5 ! 2 4 f4 ii.. e ? followed by . . . i.c5 . 23 ..ia4 �d6 24 lt:Jc4? i.a6 is also pointless. 23 g4

hxg4

24 tt::l x g4

Now what is threatened? P robably nothing . After strengthening his position with a series of prophylactic moves, Black can also th i n k

Consistent, but bad! Wh ite should have forestalled the threatened attack on his king with a series of exchanges: 26 'ii'e 3 (or 26 tt::l xf6 i m mediately) 26 .. J::i h 8 27 tt::l xf6 'i!Vxf6 28 ir'e5 with somewhat the better endgame for Black. 26 . . .

.l:!.h8

27 'ii'g 3

What does Wh ite want? Obviously, to simplify the position by 28 tt::l xf6 'it'xf6 29 'ii'e 5. But it is not hard to parry th is threat with a move which at the same time comes into Black's plan of attack. 27 . . .

.l:!.h 5!

246



In the Footsteps of one Game

How qu ickly Wh ite's position has become difficult! 28 'ii'f4?

28 tt::l xf6 was nevertheless better. 28 . . .

..te7 !

The bishop switches to a powerfu l attacking position - d6. A possible variation is 29 tt::l e 5 ..td6 30 'ii'e 3 'ii' h 4 31 f3 :xe5! 32 dxe5 ..tc5 33 l:txd7 'ii'xe 1 + .

Appendix The Taimanov-Fischer and Gulko-Dvoretsky games, the opening stages of wh ich we have a l ready exa m i ned , conta i ned many other i nteresting and i n structive events you can now make their acquai ntance . Ta i manov - Fischer

B uenos Aires 1 960

29 1i'c1

..td6

Ragozin Defence

30 tt::l e 5

'ii' h 4

31 f3

'ii'g 3

1 c4 tt::lf6 2 tt::l c 3 e6 3 d4 ..tb4 4 e3 0-0 5 ..td3 d5 6 tt::lf3 tt::l c 6 7 0-0 dxc4 8 ..txc4 ..td6 9 tt::l b 5? ! ..te7? 1 (9 . . . e 5!?; 9 . . . 'ii'e 7 ! ? ) 1 0 h3 ( 1 0 'ii'c2 ! ? , preparing l:td 1 ) 1 0 . . . a6 11 tt::lc3 ..td6 1 2 e4 e5 1 3 ..te3 exd4?1 1 4 tt::lx d4? ( 1 4 ..txd4 ! ! ) 1 4 . . . ..td7?! ( 1 4 . . . tt::lx d4 1 5 ..txd4

32 l:te3

32 tt::l x d7 .Uh 1 + ! . 32 . . .

l:td8 !

The last black piece joins the attack (33 .. J:td h8 is threatened ) . Wh ite resigned . Wh ite lost q u ickly, without making any obvious positional mistakes . There was a simple reason : Gavrikov had not developed the skill of prophylactic th inking, which Yusu pov possessed . For this reason , i n this game they proved , i n boxing term inology, to be 'in different weight categories'. Prophylaxis is one of my favou rite themes . I could expand on it at length , but it is time to stop. In conclusion I should like to d ispel the impression , which you may have acq u i red , that I am altogether against the serious study of opening theory. Remember the examples we have examined, and the difficulties which one of the sides sometimes encou ntered due to being insufficiently well-prepared i n the opening, and you will realise that I am by no means appealing you to g ive up studying the open ing. And in genera l , every player has the right to study that aspect of chess which most attracts him. I wanted only to warn against concentrating on opening theory alone, and to show what fascinating and useful discoveries can be made, by immers­ ing yourself in the amazing world of chess. ' Here morn ing stole up on Scheherazade, and she ceased her permitted d iscourse . '

b5! 1 6 e5 c5 ! ! ) 1 5 .l:.e1

fle7?

1 6 ..tg5!

tt::l x d4

1 6 . . . ..te5 was better, althoug h after 1 7 lLlf3 White would still have retained the advantage. 17 lLld5 !

T h e only way! T h e prim itive 1 7 'ii'xd4? would have allowed Black to simpl ify the position by 1 7 . . . 'i!ke5 . 17 . . .

'ii'e 5

1 8 f4

tt::l f 3+

The only move .

1 9 'ii'xf3?

ctJ

In the Footsteps of one Game

Tai manov makes the natu ral move, retaining an advantage, taking i nto acco u nt the variation 19 . . . 'it'd4+ 20 'ith 1 'iixc4 2 1 i.. xf6 (it is probable that 2 1 l2Jxf6+ gxf6 22 i.. xf6 also works). The captu re with the pawn also came seriously i nto consideratio n , so as not to allow the black q ueen to go to d4. I n good positions one usually does not want to take u n necessary risks , especially since no elementary win is appa rent after 1 9 gxf3 'iix b2 . For example, 20 .l:!.b 1 'iia 3 2 1 i.. xf6?! 'iic 5+ or 20 l:te2 'iia 3 2 1 i.. xf6?! gxf6 22 l2Jxf6+ 'ith8 23 l2Jxd7 �g8 + . I n t h e conversion o f a n advantage, a t some point you have to exert you rself, calculate variations accu rately, and fi nd a concrete way to the goa l . Wh ite would have won by 1 9 gxf3! �xb2 20 .l:!.e2 (20 .l:!. b 1 'ii'a 3 2 1 tt:Jxf6+ i s also not bad) 2 0 . . . 'ifa3 (if 20 . . . i.. c 5+ 21 'ith 1 'it'd4 , then at the least 22 i.. xf6 gxf6 23 'iix d4 i.. x d4 24 l:!.d 1 c5 25 .l:!.xd4 cxd4 26 tt:Jxf6+ and 27 l2Jxd7 is possible) 2 1 l2Jxf6+ gxf6 22 i.. xf6 . After 22 . . . i.. x h3, the reply which concerned Taimanov, there follows either 23 l:th2! 'i¥e3+ 24 'it h 1 'ii'xf4 25 e5! , or 23 'it'd5 ! h6 (23 . . . i.. c 5+ 24 'ith2; 23 . . . i.. xf4 24 'ifh5) 24 'i!Vh5 'iic 5+ 25 'i¥xc5 i.. x c5+ 26 '.th2 i.. e 6 27 l:tg2+ 'ith7 28 ng?+ 'it>h8 29 l:!.g5+ or 29 l:txf7+ 'itg8 30 .l:!.xf8+ 'itxf8 3 1 i.. xe6 . Excessive laziness or caution ca n in fact turn out merely to be new obstructions on the way to the winning of a game. 19 . . .

'i!Vd4+

20 'ith1

l2Jg4 ! ?

Fischer was hoping that t h e wea kening of the enemy king provoked by this move wou ld subsequently come in u sefu l for h i m . 2 0 . . . l2Jxd5 2 1 i.. x d5 was unpromising . 21 hxg4

'iixc4

22 b3

'i¥ b5

The q ueen has no secure shelter: if 22 . . . 'i¥d4 , then 23 .l:tad 1 is strong . 23 a4

'i¥a5

247

24l:!.ed 1 !

Th reatening 25 b4 ! . 24 . . .

i.. c 6

25 e5

Noth ing is g iven by 25 l2Je7+ i.. x e7 26 i.. x e7 .l:!.fe8 27 b4 'ii b 6 28 i.. c5 on account of 28 . . . i.. xe4 . 25 . . .

i.. b4

26 'ir'e4

Wh ite could have won a pawn by 26 i.e? i.. x e 7 27 ttJxe 7 + 'ith8 28 l2Jxc6 bxc6 29 'ii'x c6 . But with all the heavy pieces on the board it would not have been easy to convert it, in view of the insecu re position of his king (this is where the effect of 20 . . . l2Jg4 is felt). Tai manov wisely avoids the temptatio n . H a v i n g a stable positional advantage, i t is impo rtant not to 'sell too cheaply' . 26 . . .

i.. x d5

27 .l:txd5

'i!Vb6

28 f5 !

An accu rate move order. While attacking on the kingside, Wh ite at the same time creates a th reat on the q ueenside: 29 a5 'ii'c6 30 .l:!.c1 . Wea ker was 28 a5 'it'c6 29 f5 , after which apart from 29 . . . b5!? 30 .l:tc1 'ii' b 7 there is also the clever stroke 29 . . . i.. d 2 ! , suggested b y the eleven-year-old candidate

248

�

In the Footsteps of one Game

master Sasha Ryazantsev (30 �xd2 l1ad8, and Black regains the piece). 28 . . .

�c3

29 .l:tc1

�b2

By threatening the b3-pawn , Fischer tries to divert the opponent from his attack. 30 l:tb1

�c3

34 fxg7

cxd5

35 gxf8ft'+

'it>xf8

Also possible was 35 . . . .l:txf8 36 'i!Vxd5 'i!i'xb3 37 �f6 (37 �h6? 'i¥g3 , and 38 �xf8? 'iie 1 + 39 'it>h2 �xe5+ is bad for Wh ite) 37 . . . 'ife3! with a n u nclear position (but, of cou rse, not 37 . . . 'i¥g3? 38 'iid 2 ! ) . 36 'iVxh7

If 36 'ii'x d5 there fol lows 36 . . . l1xe5 37 'ii'd 2 'ii'e 6. 36 . . .

�xeS

37 l:tf4!

'i!Ve6

37 . . . �g7 came i nto consideration . 38 l:!.f1

31 l:tc1 ?

31 b4! would have won , for example 3 1 .. J1ae8 32 l:tc5! f6 33 �e3 �xe5 34 11xe5. 31 . . .

�b2

32 .l:tc4?!

By playing 32 l1c2 ! , White would have forced 32 . . . �xb3, after which there would follow 33 l:txc7 , retaining a n attack. The move in the game allows Black a n important tempo to bring up his reserves . 32 . . .

l1ae8!

There beg ins a tactical skirmish, i n which an important role is assigned to the bishop on b2 , attacking the e5-pawn . Both 33 . . . c6 and 33 .. .f6 are threatened . 33 f6?

33 �e3 'i!i'xb3 34 .l:txc7 was objectively stronger, since now Black could have seized the in itiative, by playing 33 . . . l:te6! with the idea of 34 fxg7 l:tfe8 . 33 . . .

c6

Black has successfu lly conducted a d ifficult defence and only a l ittle more was requ i red i n order to com pletely neutra l ise the oppo­ nent's i n itiative. He should have moved his bishop along the a 1 -h8 d iagonal ; only not to g7, in view of 39 �c1 ! (39 �d2? 'iixg4) 39 . . . 'it'e7 40 g5 'it'b4 ( intending 4 1 . . J1e 1 ) 4 1 �b2! d4 42 'iif5 'ili'xb3 43 g6 f6 44 �xd4 , but to c3 or d4, when in the event of 39 Si.c1 there is the reply 39 . . . 'it>e7 ! . Fatigued b y such a g ruelling game, Fischer makes a m istake, which could have proved decisive. 38 . . . 39 axb5?

b5?

ttJ

In the Footsteps of one Game

Strangely enou g h , this natural exchange (especially i n time-trouble - j u st before the fortieth move ! ) prevents Wh ite from con­ cl uding his attack victoriously. He could have won by 39 .l:te 1 ! , creating the i rresistible threat of 40 �h6+ �e7 41 �g7. 39 . . .

axb5

40 �d2

Now if 40 .Ue 1 there is the su itable reply 40 .. J�a8 ! ! - the threat of the exchange 4 1 .. J:ta 1 neutralises the wh ite rook. For example, 41 �h6+ �e7 42 �g7 .l:!.a 1 ! 43 .l:txa 1 (better is 43 �h4+ f6 44 l:txa 1 �xa 1 45 g5 with eq ual ity) 43 . . . �xa 1 44 �xa 1 �e 1 + 45 'it>h2 �xa 1 , and the q ueen ending is i n Black's favou r. 40 . . .

�e7!

41 j_ b4+

'it>d8

42 .Uxf7

42 �a5+ �c8 43 l:.xf7 .l:th8 44 .l:tc7+ �b8 45 .Ub7+ etc. would also have led to a d raw. 42 . . .

llh8

43 l:.f8+

.U.xf8

44 �xf8

'iif6 !

Wh ite's extra pawn does not play any sign ificant role. The game should end i n a draw.

249

45 .. .'iff1 + 46 �g 1 �d4 suggests itself, forcing 4 7 �h2 with roughly equal chances. F ischer was probably trying for more, but he overlooked his opponent's simple reply. 46 �g 1 !

'ii'f4

Black is forced to allow the exchange of q ueens and go i nto an u n pleasant bishop end i n g . True , the d raw has not yet been thrown away. 47 'ii'e 7+

�c8

48 'ii'f8+

'ii'xf8

49 �xf8

..tg3!

50 �1

d3!

51 � b4

�d7

52 �e1

�f4

53 j_cJ

..tg3

54 g5

�e6?!

An i naccu racy. 54 . . . e7 55 g6 �f8 was simpler. 55 g6

�e7

The position of the king on e7 g ives Wh ite a tempo, necessary to free his king from i mpriso n ment. 56 ..te1

..tf4

57 � h4+

�8

58 g 3 !

�d6!

45 �c5

45 . . .

d4?

The natural move 58 . . . ..te3? would have lost in view of 59 �f6 followed by �e 1 and �f6-e5-f4 .

250

�

In the Footsteps of one Game

59 'iiif2 �c5+ 60 'iitf3 'it>g7 6 1 �g5 'it>xg6 62 �f4 'it>h5! 63 'it>e4 (63 g4+ 'it> h4) 63 ... 'it>g4 64 'it>xd3 'iitf3 65 �c7 �f2 66 �d6 �e1 67 'it>d4 'it>g4 68 'it>c5 b4 69 'it>b5 'it>f5 70 'it>c4 'it>e6 71 �c7 'iitf5 72 'it>d3 Wg4 73 �d6 �c3 74 'it>c4 i.e1 75 i.xb4 i.xg3 76 i.c3 �d6 77 'it>d5 i.e7 78 i.d4 i.b4 79 'it>c4 �as 80 i.c3 i.d8 8 1 b4

If we now make the moves 81 . . . 'it>f5 82 'it>d5 , we obtai n the position i n which David Janowski resigned to Jose Raul Capablanca (at the New York Tou rnament of 1 9 1 6 ). The black bishop is controlling a sq uare wh ich the pawn must cross. Wh ite will try to evict the enemy bishop, by using intercep­ tion - a highly important method in such situations. Only the black king can prevent the interception . An analysis of this type of ending was made in the mid-1 9th centu ry by the Italian player Luigi Centu ri n i . He establ ished the main idea of the defence - the black king should

Yuri Averbakh showed that J anowski was wrong to resign against Capablanca - he could have reached Centu rini's position : 1 . . . 'it>f4 ! ! 2 il.. d 4 (2 �e5+ 'it>e3 3 b5 'it>d3 4 'it>c6 'it> c4) 2 . . . 'it>f3! 3 b5 'it>e2 4 Wc6 'it>d3 5 �b6 �g5 6 il.. c 7 (after 6 'it>b7 'it>c4 7 'it>a6 B lack again takes h i s king to the rear of Wh ite's - 7 . . . '1t>b3! 8 i.f2 �d8 9 i.e1 'it>a4! with a d raw) 7 . . . i.e3 8 'it>d5! (8 i.d6 'lt>c4) 8 . . . i.d2 ! ! (bad is 8 . . . 'it>c3 9 �d6 �b6 1 0 'it>c6 o r 9 . . . 'it>b3 1 0 i.c5 'it>a4 1 1 'it> c6) 9 .id8 (9 b6 i.a5) 9 . . . i.e3 1 0 i.e? i.b6! 1 2 c.t>c6 �a5 1 3 i.d6 'it>c4 . Even i n his youth (at the time he was just seventeen ) Robert Fischer seriously studied chess as a whole, and not only opening theory. He was familiar with Averbakh's analysis and therefore he gai ned a draw without d ifficulty.

be positioned to the rear of the white king. Here is one of his positions, showi ng

how Black should construct the defence :

(see diagram) 1 il.. c7 is not possible, while after 1 i.e3 iL.a5 2 �b6 i.d2 3 �c7 i.e3 i nterception is not feasible, since the c5-square is control­ led by the black king.

Draw.

81 . . .

'it>f4 !

82 b5

'it>e4

83 �d4

i.c7

84 'it>c5

'it>d3!

85 'it>c6

'it>c4

86 i.b6

i.g3

87 i.a7

i.c7

ltJ

In the Footsteps of one Game

G u l ko - Dvoretsky

2 1 l:td2

43rd U S S R Championsh i p , Yerevan 1 975

22 l:te1

Ragozin Defence

Threatening 23 l1e5 .

1 d4 lt:Jf6 2 c4 e6 3 lt:Jc3 � b4 4 lt:Jf3 d5 5 e3

22 . . .

0-0 6 �d3 lt:Jc6 7 0-0 a6 8 h3 dxc4 9 ..ixc4 ..id6 1 0 e4 e5 1 1 �e3 exd4?! 1 2 ..ixd4 ! ! lt:Jxd4 1 3 'ii'x d4 b 5 1 4 e5? ( 1 4 ..ib3! c 5 1 5 'ii'e 3 c4 1 6 ..ic2 ! b4 1 7 e5! ) 1 4 . . . bxc4 1 5 exd6

23 'it>h2

What should Black play now? Having got away with a slight fright in the open ing, i n my joy I i mmediately com m itted a sign ificant i naccu racy. Since in any case the q ueenside pawns will remain broke n , I should at least have retai ned the more active c4-pawn , which fixes Wh ite's weak­ ness on b2. After 1 5 . . . �e6! the two sides' chances are roughly eq u a l . U nfortu nately, I conducted t h e su bseq uent stage of the game su perficially and my opponent g rad ually outplayed me. 15 . . .

'ii'x d6? !

1 6 'ii'xc4

..ie6

1 7 'it'e2

.Ufe8

1 7 . . . lt:Jd5 came i nto consideration . 1 8 .Ufd 1

�c5

Wh ite would also have stood better after 1 8 . . . �b6 1 9 iVc2 . 1 9 �d2

�h5

20 �f4

c5

25 1

h6

'ii'g 6

Black's position is clearly worse ; the wh ite pieces are more active (he has to reckon with .Ue5 and �d6 ) and his q ueenside pawns a re weak. In such cases it is important to fi nd a n idea which may d ivert the opponent from the natural idea of fu rther strengthening his position , and to pose h i m s o m e problems. 23 . . .

'ii'f5 !

I thought that, in an ending with rooks on the board , the bishop might prove stronger than the wh ite knight and that to some extent this would compensate for the weakness of my pawns. In the midd legame, by contrast, knig hts usually coord inate better with the q ueen . I n add ition B lack sets his opponent a posi­ tional trap - he tempts him i nto playing 24 'ii'd 6?! (with the th reat of 25 .i:!.e5), against which he had prepared 24 . . . .i:!.ad8! 25 �xd8 .l:i.xd8 26 .i:!.xd8+ 'it>h7. I n the resu lting posi­ tion I have rea l counter-chances, associated with . . . �h5 and . . . g7-g5-g4. I ncidentally, it was i n connection with this plan that on the previous move Black l u red the king to h2. 24 'it'xf5!

�xf5

252

�

In the Footsteps of one Game

25 .l:!.e5!

l:lxe5

26 tt::lx e5

.:tea

27 tt::l c4

Here I had to th ink for a long time. The position is very hard to defend . For example, if 27 . . . i.e6 the simple 28 tt::le 3 followed by 29 l:td6 is strong. It is u nfavou rable to play 27 . . . l:te 1 28 l:!.d6 i.e6 29 tt::l e 3 or 28 . . . tt::le4 29 tt::lxe4 i.xe4 30 f3 ! i.b7 (30 . . . i. b 1 3 1 .lba6) 3 1 tt::la 5 :e7 3 2 b3. 27 . . . tt::l e4 !? looks better. Wh ite has a choice: a) 28 tt::lxe4 i.xe4 (28 . . . .l:r.xe4? 29 tt::l d 6), and now none of the following conti nuations is convincing: 29 tt::ld 6 l::te 6, or 29 l::t e 2 lte6 with the idea of . . . i.d5 , or, finally, 29 l:.d6 i.b 1 ! 30 l:.xa6 i.d 3 . b) 28 l:.e2 ! ? �8 (weaker is 28 . . . tt::l f6 2 9 l:.xe8+ tt::l xe8 3 0 tt::la 4 i.e6 3 1 b 3 ! ? i.xc4 32 bxc4 tt::l d 6 33 tt::lx c5 a5 34 �g3 tt::lxc4 35 �f4 ) 29 tt::la4 i.d7 ! 30 tt::l c b6 (the simple 30 tt::la b6 leaves Wh ite with somewhat the better position ) 30 . . . i.b5 31 .l:.c2 i.d3 32 .l:tc1 tt::lxf2 33 tt::l xc5 . The endgame looks dangerous for Black, but 33 . . . l:te2 would appear to maintain the balance . 1 decided to employ m y favou rite defensive method - to tempt the opponent i nto the win of a pawn , i n order i n return to activate my pieces to the maxi m u m .

27 . . .

�f8 ! ?

28 .l:.d6

The cool-headed 28 tt::l e 3 would have been more u npleasant for me. 28 . . .

tt::l e 41

29 tt::lx e4

l:txe4

30 tt::le 3

i.e&

31 .l:!.xa6

.l:td4

Black had a i med for this position . H i s rook is now active (the th reat is 32 . . . .l:.d2 , winning a pawn ), and h i s bishop is stronger than the knight, restricting its mobil ity (it was for such a situation that I was hoping , when I exchanged the queens). The chances of a d raw are q u ite rea l . 1 w a s expecting t h e natural move 3 2 b3, parrying the threat of 32 . . . .l:.d2 . After 32 . . . �e7 Wh ite has to reckon with . . . f7-f5-f4 (espe­ cially in the event of �g3 ). If 33 l:.c6 there is 33 . . . .l:!.d2 (the a2- and b3-pawns a re vul ner­ able), while if 33 .l:!.a5 - 33 . . . �d6.

In such cases another well-known psycho­ logical effect often operates in Black's favou r. The opponent does not real ise that the pawn was sacrificed for definite positional compensation ; he th i n ks that he simply won it. Reckoning that the goal is a l ready close, and that the remainder is a matter of tech nique, he freq uently relaxes and beg ins playing carelessly, which it is usually possi­ ble to exploit.

CLJ

In the Footsteps of one Game

But I was u n l u cky - G u l ko did not make th is psychological mistake . He thought for a long time (leaving h i m self with j u st 1 0 m i nutes for 8 moves) and found an excellent practical chance. 32 .l::i.b 6!

Wh ite does not want to place his pawns on light sq uares and he plans a2-a3, reta i n i n g t h e possibil ity o f also advancing th is pawn fu rther. In add itio n , the move in the game involves a cunning trap, wh ich I, alas, failed to spot. Of cou rse, 32 . . . ..ixa2?? 33 .:I.b8+ e4 58 'it>e2 h 5 a l so does not lose) 58 b4 'itc6 followed by . . . 'it>b5 and . . . �xb4 .

Black's position is strateg ically hopeless. If 50 .. .'�g8, then 5 1 %:te4 is strong . I decided to provoke my opponent i nto a combination with the win of a pawn , since I saw that the resulting bishop ending might prove d ifficult to win . 50 . . .

'ii'f5! ?

51 'ii'xf8+!

�xf8

52 l:txf6

'ii'xf6

53 �xf6+

'it>g8

I n the event of 54 g4? Black succeeds in advantageously exchanging a pai r of pawns on the kingside: 54 . . . 'it>f7 55 �c3 c4 ! 56 'it>g2 'it>g6 57 �3 h 5 . Then the worst that Black is th reatened with is the loss of his c-pawn, after which there a rises a d rawn ending, a l ready fam i l i a r to u s from the Ta imanov­ Fischer game. But it is not apparent how Wh ite can ach ieve even this, for example, 58 gxh5+ 'it>xh5 59 'it>e4 �h6! 60 'it>d4 i.c1 61 'it>xc4 �xb2 62 �xb2 'it>h4. G u l ko assessed the position excellently and made the winning move . 54 'it>g 1 ! 55 �c3

'it>e6

56 �

'it>f5

If 56 . . . c4 , then 57 'it>e3 'it>d5 58 �4 is decisive . 57 'it>e3 !

Again Wh ite is carefu l . If 57 �3? there would have followed 57 . . . c4! 58 g4+ 'it>g6 (or 59 . . . 'it>g5 ) and . . . h 7-h5 . 57 . . .

h5

57 . . . c4 58 'it>d4 �4 59 'it>xc4 . 58 b3

Only now has the time come to fix the black pawn . 58 . . .

Which moves for Wh ite suggest themselves in the first i nstance? Probably 54 b3, fixing

�e7

59 'it>d3

h4

60 �e1 !

'it>e5

61 'it>c4

'it>d6

62 b4!

Black resigned .

ctJ

255

PART V Artur Yusupov

Games by Pupils of the School n th is chapter we will once again retu rn to problems which players face when preparing for a game and d u ring the playing of its i n itial stage, by analysing the typical mistakes revealed i n the analysis of games by our pupils. The n u m ber i n brackets ind icates the age of the pupil .

I the

Sitni k (8) - Stepanavichus

Tal l i n n 1 989

action in the centre. A just retribution for violating the principles of development would have been the variation g iven by Dvoretsky: 1 4 . . . tt:Jxe5! 1 5 'ii'e4 tZ'lc5! 1 6 'i!Vxa8 .lib? . The q ueen is trapped and the advantage passes to Black. 1 5 fxg6

tt:Jxe5

1 6 gxf7+

'it>h8

1 7 'it'g8+!

.U.xg8

18 fxg8'it' mate Gaponenko ( 1 4) - Repkova

E u ropean G i rls Championship 1 99 1

French Defence 1 e4

e6

2 d4

d5

3 tZ'lc3

tt:Jf6

4 .JigS

.lie7

5 e5

tZ'lfd7

6 h4

How should Wh ite conti nue? He is actively placed , but the development of his pieces is not yet complete . Of course, he should bring his knight i nto play by 14 tZ'lc3 ! . S ubse­ quently the knight will go to d5 or e4 . I n stead of this, the young player decided to attack the opponent's king i m med iately.

The C hatard-Aiekhine Attack. Wh ite sacri­ fices a pawn for the sake of gaining time and opening the h-file. B lack usually decl ines the gambit and aims to carry out the standard central cou nter . . . c7-c5. If 6 . . . c5 is played i mmed iately, Black has to reckon with the knight sortie to b5 after the exchange of the da rk-sq uare bishops, and so the simplest is to fi rst play 6 . . . a6. 6 . . .

h6?!

7 .lie3

c5

Now Wh ite ach ieves his a i m . Unprepared

8 'i!Vg4

�f8

wing attacks should be met by energetic

9 .l:!.h3?

1 4 f5?

tt:Jc5?

256

�

Games by Pupils of the School

The same mistake as in the previous game you should not launch an a ttack without completing your development.

-

Correct was 9 t'bf3 t'bc6 1 0 0-0-0 with the better chances for Wh ite. 9 . . .

t'bc6

1 0 l::t g 3

g6

to captu re on f3 with the q ueen , and for the moment the knight ca nnot be developed on the best square f3 . H owever, the position arising after 1 5 ifxf3 'ii'x h4 1ooks anxious for Black ( Dvoretsky). 14 t'bxf3

11 .l:!.f3?

1 1 t'bf3 was better.

11 . . .

t'bdxe5?

Black senses that a cou nterblow i n the centre should refute her opponent's unpre­ pared attack, but the concrete method she chooses is unfortunate. I n her annotations lnna Gaponenko gave the correct way: 11 . . . cxd4! 1 2 'it'xg6 t'bdxe5 1 3 il.xh6+ �e8 1 4 1Wg7 t'bxf3+ 1 5 t'bxf3 �f6 , and Black wins. 12 dxe5

t'bxe5

1 3 'ii'f4

t'bxf3+?!

It is well known that 'mistakes do not come si ngly' . 1 3 . . . il.d6! was essential. In the event of 14 'ii'a4 the simplest is 1 4 . . . il.d7 1 5 'ii'a 3 t'bxf3+ 16 t'bxf3 'ii'e 7 with advantage to Black. Far stronger is 14 0-0-0 ! (development fi rst and foremost!), not fearing 1 4 . . . t'bd3+? 1 5 �xd3 �xf4 1 6 �xf4, when the minor pieces are clearly superior to the enemy queen. 14 . . . t'bxf3 has to be played . Wh ite is forced

It is a mazing how q u ickly the situation has changed . Black has exchanged both of her a l ready developed knig hts , and now the su periority i n the placing of the wh ite pieces is very obvious. 14 . . .

d4?

1 4 . . . 'it>g7 was better. After the move in the game Gaponenko builds up a decisive attack on the king. 15 tt:Je5

�f6

1 6 t'be4!

dxe3

1 7 .l:i.d1

exf2+

1 8 �e2 !

1 8 'it>xf2 'i:Vxd 1 1 9 'i:Vxf6 ifd4+ 20 Wg3 l:th7 21 t'bxg6+ 'it>g8 22 t'be7+ �f8 leads to a d raw. 18 . . .

'i!le7

1 9 t'bxf6

'it> g7

20 t'be4

f6

21 'i:Vg3 !

g5

If 21 . . . fxe5 22 1Wxe5+ 'it>g8 Wh ite was i ntending 23 l::td 3! followed by 24 t'bf6+ 'it>f7 25 l::tf3 .

ltJ

Games by Pupils of the School

22 lb g4

.l:.f8

23 hxg5

f5

24 gxh6+

�h7

25 lDf6+

.l:.xf6

26 lDxf6+

'ifxf6

27 l:.d8 !

'ifxb2

If 27 . . .f4 (hoping for 28 'ifg8+ �xh6 29 .l::lx c8?? 'ii'e 5+ ) , the simplest is 28 'ifd3+ 'it'f5 (28 . . . �xh6 29 'ii' h 3+) 29 "ii'xf5+ exf5 30 �xf2 b6 31 i.e2 (31 i.a6) 31 . . . �xh6 (31 . . . i.b 7 32 l::t d 7+) 32 i.f3 l:.b8 33 i.c6 (Gaponenko). 28 'ii'c 7+!

�xh6

29 'ilfh2+

�g5

257

f8-bishop, and i n add ition he has exchanged his more central d-pawn. As a consequence the i n itiative passes to h i s opponent. 6 dxc5

'ifa5+

6 . . . lba6?! is weaker on account of 7 'ii'd 4, while if 6 . . . lbbd7 there is 7 b4 . 7 'ii'd 2

'ii'x c5

Th reatening 8 . . . lbe4 . 8 i.xf6

gxf6

9 g3

30 .U.g8+

B lack resig ned . Mugerman - Maka riev ( 1 4)

Moscow 1 989 1 d4

Trompowsky Attack lbf6

2 i.g5

e6

3 lDd2

c5

4 lbe4? !

T h e main objectives at t h e start o f a g a m e are t h e rapid development o f t h e pieces and the struggle for the centre . Therefore you should refrain from making repea ted moves with one and the same piece

(unless, of cou rse, such a manoeuvre brings some substantial gain or is forced ) . The danger of violati ng this ru le is i l l u strated by the fol lowi ng short variation: 4 dxc5?! i.xc5 5 lbe4?? lbxe4 6 i.xd8 i.xf2 mate . It was better to reinforce the centre by 4 e3. 4 . . .

d5

4 . . . cxd4? ! 5 'ii'x d4 i.e? 6 lbd6+ favou rs White , since he is able to h i nder the opponent's development. 5 lbxc5

i.xc5

As a result of his d u bious operation Wh ite has spent two tempi on the exchange of the

9 . . .

lbc6

A normal developing move . Its only d raw­ back is that after the natura l reply 1 0 i.g2 the thematic advance . . . e6-e5 is h i ndered . Black should have fought more actively for the centre. Concrete analysis shows that the i m med iate 9 . . . e5! 1 0 i.g2 i.e6 was possible: a ) 1 1 c4? 'ii' xc4 12 .:tc1 'ii'x a2 1 3 i.xd5 i.xd5 1 4 llc8+ �d7 1 5 l:txh8 'it'b 1 + , and Wh ite remains a piece down ; b) 1 1 0-0-0?! lbc6 1 2 i.xd5? 0-0-0 1 3 i.xe6+ ( 1 3 e4 lbb4) 1 3 . . .fxe6 1 4 "ii'e 1 .:.Xd 1 + 1 5 'ii'xd 1 lld8 ( 1 5 . . . "ii'xf2 is also good ) 1 6 'ii'e 1 'ii'd 5 ( 1 6 . . . lbb4!? 1 7 c3 lbxa2+ 1 8 � b 1 'ifb5 ! with the threat of 1 9 . . . l:.d 1 + ! 20 'ii'x d 1 lbxc3 + , while the knight is invul nerable i n view of the fork 1 9 . . . 'ii'd 5+) 1 7 lDf3 'i!i'xa2 1 8 lDd2 ( 1 8 c3 lba5 ! ) 1 8 . . . a5!? with a dangerous i n itiative on the q ueenside

258

�

Games by Pupils of the School

(the th reat is . . . a5-a4-a3); c) 11 l:1d 1 lbc6 12 i.xd5 i.xd5 ! ( 1 2 . . . 0-0-0 1 3 c4 lbb4 1 4 e4 4Jxa2 is unconvinci ng) 1 3 'ii'x d5 'ii'b 4+ 1 4 c3 (Wh ite is also worse after 14 'ii'd 2 'ii'x b2) 1 4 . . . 'ii'x b2 1 5 'ii'd 7+ 'itf8 1 6 'ii'd 2?! ( 1 6 'jj'f5 ! 'ii'x c3+ 1 7 'itf1 maintains the balance) 16 ... 'ii'x d2+ 17 l:.xd2 'ite7 . Even in t h e endgame Black's better devel­ opment g ives him the advantage. 10 i.g2

i.d7

1 1 lDf3

1 1 e3!? followed by 4Je2 came i nto consid­ eration . 11 . . .

4Je5?

l lya Makariev repeats his opponent's m is­ take: he wastes time in the open i n g . He should have contin ued his development: 1 1 . . . e5! 1 2 l:.d 1 i.e6 1 3 0-0 0-0-0 would have g iven h i m control of the centre and led to a promising position . 1 2 0-0

4Jxf3+

The outcome is that Black has wasted two tempi on the u n necessary exchange of knights. However, if 1 2 . . . 4Jc4 there is the unpleasant reply 1 3 'ii' h 6!?, suggested by Dvoretsky. 1 3 exf3 ! ?

If 1 3 i.xf3 Makariev was planning 1 3 . . . 'it e7!? - the obvious 1 3 ... 0-0-0 did not appeal to him because of 1 4 c3 followed by b2-b4 , a2-a4 and so o n . 13. . .

0-0-0

1 3 . . . e5?! 1 4 f4 ! i.e6 1 5 fxe5 fxe5 1 6 'ii'g 5. 1 4 f4

i.c6?!

Another mistake, similar to 9 . . . 4Jc6 - Black misses an opportunity for activity i n the centre by 14 . . . d4! and only then . . . i.c6.

(see diagram) 1 5 l:tfe1 ? !

Retu rning t h e favou r. Wh ite fig hts against . . . e6-e5, but allows a more dangerous advance. 1 5 c3 ! , h i ndering the thematic . . . d5-d4, was correct.

- position after 14 .tc6? ! . . .

It is dangerous to reply 1 5 . . . e5? 1 6 l:tfe 1 l:.he8 ( 1 6 . . . 'ifd6? 1 7 fxe5 fxe5 1 8 'ii'e 3; 1 6 . . . d4? 1 7 i.xc6 bxc6 1 8 fxe5 fxe5 1 9 'ii'e2 or 1 9 cxd4 with advantage to Wh ite) 1 7 fxe5 fxe5 1 8 'ii'g 5 f6 ( if 1 8 . . . 'ii'd 6 or 1 8 . . . 'ii'e7 there follows 1 9 'ii'f5+ ) 1 9 'ii'xf6 l:r.f8 in view of 20 'ii'x e5 'ii'xf2+ 2 1 'ith 1 , and the constant th reat of a bishop check on h3 severely restricts Black's possibilities. If instead 1 5 . . . d4? ! , then after 1 6 i.xc6 bxc6 1 7 cxd4 Wh ite has the advantage. More tenacious is 1 6 . . . dxc3 1 7 i.xb7+ 'itxb7 with good d rawing chances. Black should probably prefer the prophylac­ tic 1 5 . . . 'itb8 ! ? , preparing both . . . d 5-d4 and . . . e6-e5. 15 . . .

d41

1 6 i.xc6

'ii'x c6

1 7 .l:.ad 1

:ds

1 8 'ifd3

'itb8 ! ?

1 9 l:.e4

llhd8

Black has achieved a good game. Gasymov - Zviagi ntsev ( 1 3)

Len i n g rad 1 990

Queen 's Pawn Opening 1 d4

d5

2 lDf3

4Jf6

3 g3

i.f5 ! ?

ctJ

Games by Pupils of the School

4 c4

e6

5 'i¥b3

The q ueen is a very i mportant piece and its placing has a g reat i nfl uence on the charac­ ter of the su bsequent play. The queen 's position should not be determined too early. Very often, as happened in the

present game, by attacking the q ueen the opponent gains time for the development of his forces. I n chess there a re no ru les which a pply i n every case. T h e correctness o f 'absol ute' truths must each time be checked with the concrete featu res of the position . Thus the early q ueen move to b3 is a fai rly standard reaction to the development of B lack's lig ht­ square bishop in the Queen's Gambit. Even so, it is better for young players first to master general rules and only then seek exceptions to them . 5 . . .

tt:lc6!

6 .ltd2?!

6 c5 is preferable. Now Black gains an opportu nity to exploit the advanced position of the quee n . 6 . . .

dxc4!

7 �xc4

If 7 �xb7, then 7 . . . .lte4 ! 8 'fib5 .l:tb8.

259

hopes to disru pt the coord i nation of the opponent's forces . However, this is achieved at the cost of a loss of time: Black manoeuvres i n the opening with his a l ready developed pieces, com m itting the same fu ndamental mistake as in the previous example. His lead i n development (Wh ite has lost time on q ueen moves, and his bishop on d2 is not too well placed ) should have been transformed i nto a more stable advantage - superior pawn structu re . The correct cou rse was suggested by Peter Svidler: 7 . . . .lte4 ! (here this repeat move with the bishop is justified by the fact that Wh ite in turn is forced to spend time on the defence of the d4-pawn) 8 .ltc3 .ltxf3 . There can follow 9 exf3 'it'd5 1 0 'ii'x d5 tt:lxd5 1 1 .ltb5 'it'd? 1 2 tt:ld2 a6 1 3 .ltxc6+ 'it>xc6 1 4 tt:lc4 .ltd6 with the better endgame for Black. 8 .ltg2

tt:lb6

9 'ifc3? !

Here t h e q ueen deprives its m i n o r pieces of the convenient c3-sq uare . 9 'ii' b 3 was better, not fearing 9 . . . .lte4 1 0 0-0! .ltxf3?! 1 1 'ii'xf3 'ifxd4 1 2 .ltc3 with excellent compen­ sation for the sacrificed pawn . 9 . . .

.lte4?!

I would have preferred to complete my development with 9 . . . .ltb4 1 0 'it'b3 a5 or 9 . . . .lte7 1 0 0-0 0-0 followed by . . . .ltf6 . 1 0 0-0 !

.lte7

1 O . . . .ltxf3?! 1 1 'ii'xf3 'ir'xd4 1 2 �c3 . 1 1 'ir'e3 !

f5

1 1 . . . .ltg6! ? . 1 2 .ltc3

0-0

1 3 'ili'c1

Wh ite prepa res 1 4 tt:lbd2. 13 . . .

7 . . .

tt:ld7?!

By attacking the q ueen Vad i m Zviagi ntsev

tt:lc4

By end lessly reg rouping with his already developed pieces, Black g radually loses the i n itiative. 1 3 . . . a5 or 1 3 . . . .ltf6 came into consideration . 1 4 b3

tt:ld6

260

�

Games by Pupils of the School

1 5 i.. b2 lDe7?!

1 6 l:td1

16 . . . 'ife8 17 lDbd2 ( 1 7 lDc3 i.. xf3 and 18 . . . lDxd4) 1 7 . . . 'ifh5 was more i nteresting. The move in the game allows the knight to come out to the more active c3-square . 1 7 lDc3

i.. c6? !

I n his commentary Zviagintsev recommends 17 . . . lDd5!?, and if 1 8 lDxe4 ( 1 8 lDa4 ! ? ) 1 8 . . . fxe4 1 9 lDe5, then 1 9 . . . �g5 ! with chances for both sides. 18 'ife3

'ilt'c8

1 9 i.h3 ! ?

lDe4

Black is also worse after 1 9 . . . lDd5 20 lDxd5 �xd5 2 1 .U.ac1 . 20 lDxe4

i.. xe4

8 . . .

21 lDg5

The advantage has passed to Wh ite - j u st retribution for Black's slow and aimless manoeuvring.

N i konovich - Ba klan (1 2)

Alushta 1 990

Queen 's Pawn Opening 1 d4

lDf6

2 lDf3

c5

3 dxc5? 1

The exchange of the more central d-pawn for the c-pawn is normally u nfavourable. White made a similar mistake i n the Mugerman-Maka riev game, exam i ned ear­ l ier. 3 d5 or 3 e3 is better 3. . .

e6

4 g3

i.. x c5

5 i.. g2

lDc6

6 0-0 6. . .

0-0

7 c4

b6

lDxe5 ! !

F a r weaker is the restra i ned 8 . . . i.. b 7? 9 i.f4 with equal ity. After sacrificing the exchange, Volodya Baklan beg i n s a n attack on the opponent's king. Wh ite's lig ht-square bishop is cut off from the kings ide and from its main forces. 9 �xa8

d5!

Of cou rse, not 9 . . . i.a6 1 0 i.. g 2 lDxc4 11 lDc3 . Black aims not to regain the material, but for rapid development and a n attack. 1 0 cxd5

Often the best, and sometimes the only defence agai nst a gambit is the timely returning of the extra material, with the aim of consolidating one's forces or simpl ifying the position. Thus here Wh ite could have tried to buy off his opponent with two pawns by 1 0 i.f4 lDxc4 1 1 i.. c6 lDxb2 1 2 'ii' b 3, althoug h even i n th is case Black has the advantage. Besides, instead of 1 1 . . . lDxb2 there is 1 1 . . .e5!? 1 2 i.. g 5 i.. h 3 . 10 . . .

6 c4? ! 'ii'a 5+ ! .

8 lDe5?

A serious mistake , which we have already encountered in the earlier games. By wast­ ing time in the ope n i n g , Wh ite q u ickly ends u p i n a d ifficult position.

�a6 1 ?

T h e simple 1 O . . . exd 5 ! ? is also not bad, including the bishop i n the attack along the c8-h3 diagona l . For exa mple: 1 1 i.. f4 lDg6 ( 1 1 . . . i.. h 3? is not in the spirit of the position:

4:J

Games by Pupils of the School

1 2 .Jtxe5 .Jtxf1 1 3 'it'xf1 'it'xa8 1 4 �xf6 gxf6 1 5 tt:Jd2, and the chances a re only with Wh ite ; however, Black's play can be im­ proved with 1 3 . . . tt:Jg4 ! ) 12 .Jtg5 �h3 1 3 �c6 .Jtxf1 1 4 �xf1 'it'd6 1 5 �b5 ( 1 5 �xf6 'it'xf6 1 6 �xd5 'ti'xb2 ) 1 5 . . . tt:Je4 and wins. Or 1 2 �c6 tt:Jxf4 1 3 gxf4 'i!Vd6, and Wh ite's position is u nenviable. The only way of establishing some kind of defence was by 1 1 tt:Jc3! �h3 1 2 �xd5 �xf1 1 3 'it>xf1 .

261

1 4 'it>g2

1 4 tt:Jd2 'ii' h 3 1 5 tt:Jf3 tt:Jfg4 with u navoidable mate . .JiL eS !

14 . . .

The point of Black's idea . 1 5 .l:!. h 1

'ifh3+

1 6 'it> g 1

tt:Jfg4

1 7 e3

d4!

Wh ite resigned . He was severely pun ished for neglecting the development of his pieces.

1 1 .Jtc6

1 1 .JiLf4 lt.Jg6 ! ? 1 2 .Jtc6 tt:Jxf4 1 3 gxf4 'it'd6 with an attack. 11 . . .

exd5

1 2 �a4

b5 ! ?

A thematic move , b u t objectively not the strongest. Black should probably have pre­ ferred 1 2 . . . �c8 ! 1 3 'it>g2 (if 1 3 .JiLf4 , then 1 3 .. .'ifh3! is decisive) 1 3 . . . 'ili'f5 . 1 3 �c2?

Zviagi ntsev ( 1 5) - Feigin

CIS J u nior Championship, J u rmala 1 992

Benoni Defence 1 d4

e6

2 c4

c5

3 d5

exd5

4 cxd5

d6

5 tt:Jc3

g6

6 e4

�g7

7 tt:Jge2

tt:Je7

A rare conti n u atio n , the point of which is to make an early attack on the centre by . . . f7f5 . The inclusion of the moves 7 . . . a6?! 8 a4 weakens this idea : 8 . . . tt:Je7 9 tt:Jg3! 0-0 1 0 .Jte2 f5 1 1 exf5 tt:Jxf5 1 2 tt:Jxf5 �xf5 1 3 0-0 , and Black has problems with the develop­ ment of his knight - 1 3 . . . tt:Jd7? is not pos­ sible i n view of 1 4 g4! �xc3 1 5 bxc3 .Jte4 1 6 f3 (analysis by Zviagintsev). But now after 8 tt:Jg3 0-0 9 .Jte2 f5 1 0 exf5 tt:Jxf5 1 1 tt:Jxf5 �xf5 1 2 0-0 Black can play 1 2 . . . tt:Ja6 ! . 8 �g5! ?

Wh ite fails to see his opponent's cunning idea . As Baklan pointed out, the only defence was 1 3 �f4 ! .l:i.e8 ( 1 3 . . . tt:Jg6 14 .Jtc2 tt:Jxf4 1 5 gxf4 'it'd? 1 6 f5 ! ) 1 4 �c2 b4 1 5 lt.Jd2 tt:Jeg4 1 6 �d3 'ii' b 6 with a n u nclear game. 13 . ..

'it'd7 !

The q ueen penetrates t o h 3 , from where it creates i rresistible threats .

Wh ite provokes a weakening of the kingside pawns. 8 . . .

h6

9 .Jte3

f5 ! ?

If 9 . . . 0-0?! there would have followed 1 0 ii'd2 with gain of tempo. Black plans to exchange on e4 and then h a rass the bishop on e3 with . . . tt:Jf5 .

262

�

Games by Pupils of the School

..ib5+ ! , but after 1 3 . . .� ! ? things a re not so clear. 1 3 . . . �8 is wea ker on account of 1 4 hxg4 tt:'lxg4 1 5 tt:'le6+! o r 1 4 . . . ..ixg4 1 5 .ie2 with a dangerous i n itiative for Wh ite. 1 2 . . . it'a5 ! was also possible: 1 3 'ii' b 3!? ( 1 3 it'c2 ..ixc3+! 1 4 bxc3 tt:'le5) 1 3 . . . ..ixc3+! ( 1 3 . . . tt:'le5 1 4 ..ib5+ with a n attack for Wh ite) 1 4 bxc3 tt:'le5 1 5 ..ib5+ �d8! with double­ edged play (Zviagi ntsev) . If 1 3 tt:'le6! ? , then 1 3 . . . ..ixc3+ 1 4 bxc3 tt:'le5. 1 3 tt:'le6

If 1 3 . . . ..ie5? Zviagi ntsev was plann ing 1 4 tt:'lb5 � 1 5 a4 ! . Black should have played 1 3 . . . ..ixc3+ 1 4 bxc3 tt:'le5.

1 0 g4! !

An excellent decision! I n the King's I ndian Defence Wh ite sometimes advances his g­ pawn in order to gain control of the f5square and restrict the mobil ity of a knight on e7, but usually this does not i nvolve giving up a pawn . In the event of the acceptance of the pawn sacrifice, Wh ite gains compensation in the form of his strong and mobile pawn centre . 10 . . .

fxg4

Weaker is 1 0 . . . fxe4?! 1 1 tt:'lg3 with advan­ tage. 11 h3

By inviting the exchange of the g-pawn , Zviagintsev wants also to exchange the lig ht-sq uare bishops. Then he will endeav­ our to invade on the weakened e6-squ a re with a knight. 11 . . .

'it'xb2?

The move i n the game is a poor one. Black violates one of the simplest rules: don 't go 'pawn-grabbing ' in the opening. Now Wh ite wins by force , by exploiting the poor position of the enemy q ueen . 1 4 tt:'lxg7+

�f7

1 5 ..i d 2 !

� xg7

1 6 llb1

�a3

1 7 l:tb3

'ii'a 5

1 8 tt:'l b5

it'd B

1 9 .Jtc3+

tt:'le5

Wh ite has i ncluded h is pieces in the attack with gain of tem po and prepared a conclud­ ing combinative stroke i n the centre .

tt:'l d 7 ! ?

Black would play into his opponent's hands by 11 . . . gxh3 12 ..ixh3 or 11 . . . a6 12 hxg4 ..ixg4 1 3 .ll. h 3, with excellent compensation for the sacrificed pawn . 1 2 tt:'lf4

1 2 hxg4! ? tt:'le5 1 3 tt:'lf4 came i nto considera­ tion. 12 . . .

'ii' b 6?

The start of a fau lty manoeuvre. If 1 2 . . . tt:'le5 Black was probably concerned about 1 3

20 tt:'lxd6!

"ili'xd6

ltJ

Games by Pupils of the School

21 'ii'a 1

'it>f6

22 hxg4

g5

23 f4!

gxf4

24 .ltxe5+!

'iix e5

25 .l:txh6+!

.l:!.xh6

26 g5+

Black resigned . Kram n i k - Zviagi ntsev ( 1 4)

Leningrad 1 990

Philidor Defence 1 e4

d6

2 d4

lbf6

3 lbc3

e5

4 lbf3

Wh ite can also consider tra nsposing i nto a superior endgame by 4 dxe5 dxe5 5 'ili'xd8+ 'it>xd8 6 lbf3 .lid6 7 .ltc4 . The move i n the game leads to a variation of the P h i l idor Defence. 4. . .

lbbd7

5 .ltc4

.lte7

6 0-0

0-0

9 . . .

263

h6

Zviagi ntsev did not play 9 . . . b6? ! , because he was afraid of 1 0 d5. For example, 1 0 . . . .lib 7 11 dxc6 .ltxc6 1 2 .ltb5! (in the event of the i mmed iate 1 2 .ltg5 B lack has the equalising 1 2 . . . .ltxe4 ! ) 1 2 . . . .ltb7 1 3 i.g5 with advantage to Wh ite. I n the game Kuch u kh idze-Zviagintsev, played earlier in the same tou rnament, i n reply t o 9 . . . h6 Wh ite chose 1 0 a5!?. After 1 O . . . l:.b8 nothing is g iven by 1 1 d5 b5 1 2 axb6 lbxb6 , but serious consideration should h ave been g iven to the prophylactic 1 1 .lta2 ! , and if 1 1 . . . b5 1 2 axb6 axb6 1 3 d5. I n stead of this there followed 11 .lte3, and Black successfu lly solved the problem of his q ueenside by 11 ... b5! 12 axb6 axb6 1 3 d5 b5 14 Ua7?! lib? 1 5 l:txb7 .ltxb7 16 .ltb3 b4! ? 1 7 lba2 cxd5 1 8 exd5 'iia 5 . N atural ly, Volodya Kra m n i k had prepared for the game and he tried to improve Wh ite's play. 1 0 i.e3 ! ?

7 l:.e1

7 'ii'e 2. 7 . . .

c6

8 a4

'iic 7

The standard reaction 8 . . . a 5 ! ? , recom­ mended by modern theory, restricts the opponent's possibilities on the q ueenside. After this Wh ite has tried 9 h 3 exd4 1 0 lbxd4 ( 1 0 'ii'x d4 lbc5 1 1 .ltg5 lbe6 1 2 .ltxe6 .ltxe6 1 3 ltad 1 l:te8 1 4 .ltxf6 .ltxf6 1 5 'ii'x d6 'ii' b 6! with a n u nclear game, Ti moshchenko­ Pianinc, Polanica Zd roj 1 979) 1 O . . . lbc5 1 1 .lif4 'ii' b 6 1 2 lbb3 .lte6 1 3 .ltxe6 lbxe6 1 4 .lte3 'iic7 1 5 lbd4 lbxd4 1 6 .ltxd4 l:!.ad8 , and the chances a re equal ( lvkov-Pian i nc, Am­ sterdam 1 974). 9 h3

Preparing t h e development o f t h e b i s h o p a t e3. An alternative is 9 a 5 ! ? .

10 . . .

l:t e8?!

A routine move, vacati ng the f8-square for the knight or bishop. But now White suc­ ceeds in setting u p a bind on the q ueenside. O n encou ntering a su rprise i n the opening, Zviagi ntsev began playing rather passively and he conceded the i n itiative to his more experienced opponent.

264

�

Games by Pupils of the School

1 0 . . . lt:Jxe4 1 1 lt:Jxe4 d5 did not work account of 12 i. xd5 cxd5 1 3 lZ'l c3. Black should have fought against his opponent's plans, by choosing either 1 0 . . . a 5 ! ?, or 1 0 . . . b6!? (with the idea of . . . a7-a6, . . . i. b7 and . . . b6-b5). If

after 1 0 . b6 White replies 1 1 lZ'lh4, then

20 axb7

i.xb7

2 1 i.xb7

If 21 'ir'a4 there again follows 21 . . . ii.e7 ! . 21 . . .

'ir'xb7

22 !tb1

. .

1 1 . . . lt:J xe4? does not work becau se of 1 2

lt:Jg6! or 1 2 lt:Jxe4 i.xh4 1 3 ifg4 , but 11 . . . exd4 and 1 2 . . . lt:Je5 is possible. And if 1 1 d5, then 1 1 . . . i.b7 1 2 dxc6 i.xc6 1 3 lZ'ld2 a6!? 14 ife2 ifb7 with chances for both sides. 11 a5! White is planning 1 2 d5. 11 . . .

i.f8?!

11 ... lt:Jf8 12 d5 lZ'lg6 is more natura l . 1 2 d5!

lt:Jc5

12 . . . cxd5 1 3 i.xd5 lt:Jc5 is also u nattractive. 1 3 lZ'ld 2 If desired , it is also possible to spoil the opponent's pawn structure by 1 3 .Jtxc5 dxc5 14 dxc6 bxc6 . 13 . . .

i.d7

14 b4!

cxd5

The only defence. 1 5 lZ'lxd5

lt:Jxd5

1 6 ii.xd5

lt:Je6

1 6 . . . lt:Ja4? 1 7 "i¥f3! and 1 6 . . . lt:Ja6? 1 7 'ir'b 1 ! ? were both bad for Black. 17 c4

lZ'lf4 ! ?

1 8 i.xf4

exf4

1 9 a6 Wh ite's in itiative on the q ueenside and i n the centre h a s l e d to the creation o f pawn weaknesses in the opponent's position . However, if the game is opened u p , B lack will acq u i re certai n counter-chances . In the event of 1 9 h4 (with the th reat of 20 'iff3) there is the reply 1 9 . . . i.e7, activating the 'bad' bishop, but 1 9 ii'f3 ! ? or 1 9 lZ'lf3!? came into consideration . 19 . . .

i.c6

22 . . .

g5?

S iegbert Tarrasch rightly commented that if one piece stands badly, the game also stands badly. The principles of development apply not only i n the open i n g . B lack should have activated his bishop, but without wea kening his castled position in the proc­ ess. The logical move was 22 . . . ..ll. e 7. Zviag intsev's suggestion 22 . . . a5!? 23 bxa5 'ir'c7 24 'it'g4 11xa5 25 it'xf4 .l:!.a2 was also i nteresting. 23 h4!

I n this way Wh ite gains a n obvious advan­ tage. 23 . . .

�e6

24 1Vh5

Weaker was 24 hxg5 hxg5 25 lZ'lf3 l:!.xe4 26 lt:Jxg5 .U.xe 1 + 27 "i:Vxe 1 'ir'e7 with counter­ chances (Zviagintsev). 24 . . .

�g6

25 hxg5

hxg5

26 e5! I n the event of 26 lZ'lf3 Black would have

Games by Pupils of the School

gai ned cou nterplay by 26 . . . l:.ea! 27 ltJxg5 ..ih6. 26 . . .

l:. e 8

27 exd6

l:.xe1 +

28 .l:.xe 1

'ii'x b4?

This loses i m mediately. Black should have eliminated the more dangerous d-pawn : 2a . . . l:.xd6. However, even in this case Wh ite would have retained a g reat advantage by conti n u i ng 29 b5! (weaker is 29 lDf3 'ili'xb4 30 Vxg5+ l:.g6 31 'ii'xf4 a5). For example: 29 .. Jbd2?! 30 'it'xg5+ �h7 3 1 .l:.ea 'ii' b 6! 32 .:Xfa 'il'xf2+ 33 �h2 'ii'g 3+ (otherwise things end i n mate) 34 'il'xg3 fxg3+ 35 �xg3 �g7 36 :tea with a won rook endgame (variation suggested by Mark Dvoretsky). 'ii'd 6

29 d7!

29 . . . 'iii'x d2 30 l:td 1 . 30 ltJe4 !

Black resigned in view of 30 . . . 'il'xd7 3 1 'it'xg6+! fxg6 32 lDf6+. Romanishin - Aiexandrov ( 1 7)

Pula 1 990

English Opening A game with a g randmaster, especially one of such high calibre, is not only a serious test, but also a n excellent opportun ity to learn some real chess. 1 d4

ltJf6

2 c4

e6

3 lL!f3

d5

4 g3

dxc4

5 ..ig2

ltJc6

6 'il'a4

6 0-0 is another attempt. 6. . .

..i b4+

7 ..id2

ltJd5

8 ..ixb4 a 'ii' b 5 .i.xd2+ 9 lL!bxd2 c3 1 0 bxc3 ltJxc3 1 1 'ii'd 3 has also occu rred , when White has compensation for the sacrificed pawn .

8

.

.

.

265

ltJ dx b4

9 0-0

For a long time the move 9 a3 suffered a crisis because of the game lvanch u k­ Korchnoi (Til b u rg 1 9a9), in which after 9 . . . b5 1 0 Wxb5 ltJc2+ 1 1 �d2 ltJxa 1 1 2 'il'xc6+ ..id7 1 3 'il'xc4 Victor Korchnoi employed the novelty 1 3 . . . c5 ! and gained a spectacular w i n : 14 'iii'a 2 WaS+ 1 5 b4 cxb4 1 6 'il'xa 1 .:tea 1 7 lDe5 ..ib5 1 a �e3? l:c2 (Black has an obvious adva ntage) 1 9 ..if3 0-0 20 a4 f6 21 lL!d3 ..ic4 22 ltJd2? Wg5+ (22 . . . ..ixd 3 ! ) 23 lL!f4 e5 24 ltJxc4 exf4+ 25 gxf4 Wf5 26 ltJd6 'il'e6+ Wh ite resigned. The attem pt by Wh ite to i mprove with 1 a lDd3 i n the game Polovod i n-S . I vanov (St. Petersburg 1 992) led to the same inauspi­ cious result: 1 a . . . ..ixd3 1 9 'it>xd3 0-0 20 'ii'b 2 bxa3 2 1 'ii'x a3?! (2 1 lDxa3 was better, and if 21 . . . l:tba 22 lL!c4 ! ) 2 1 . . . 'ii' b 5+ 22 �d2 l:tfda 23 e3?! Wf5 ! 24 �e2 l:tc2+ 25 lDd2 l::!. x d2+ 26 �xd2 'ii'xf2+ 27 �d3 Wxg2 2a l:.b 1 e5 29 l:lb2 .l::!.x d4+ Wh ite resigned. S u bseq uently it transpired that Wh ite does better to defend with 1 a d5 or 1 a a4. I n add itio n , instead of 1 5 b4 i t makes sense to play 1 5 lL!c3 ! ? cxd4 1 6 ltJxd4 , as in the game Bareev-Adams, Dortm und 2000, which ended i n a d raw. 9 . . .

l:.b8

1 0 ltJc3

.i.d7?!

266

�

Games by Pupils of the School

For the moment Black is a pawn u p , but he is behind in development and his pieces a re uncoordinated . 1 O . . . a6 1 1 tbe5 0-0 ( 1 1 . . . 'i!i'xd4 1 2 tbxc6 favou rs Wh ite) 1 2 tbxc6 tbxc6 1 3 �xc6 bxc6 1 4 'it'xc4 �xb2 has occu rred several times, but after 1 5 l:tab 1 �b6 1 6 'ii'c5 Wh ite retains some i n itiative. I n the game Krasenkow-Med nis (Palma d e Mallo rca 1 987) there followed 1 6 . . .f6 1 7 a4 Ue8 1 8 a5 nxb 1 1 9 .Uxb 1 'ii'd 6 20 �xd6 cxd6 21 .l:b6, and now Black should have conti nued 2 1 . . . 'it>f7 ! 22 lhc6 'it>e7 with equalising chances. Later Uwe Bonsch managed to neutral ise Oleg Roma nishin's in itiative (Berlin 1 990) by 16 . . . h6 1 7 a4 a5 18 �fd 1 �a6 1 9 e3 'it'g5 20 .l:xb6 'it'xc5 2 1 dxc5 cxb6 22 cxb6 l:tb8 23 l:tb1 �d3 24 l:tb2 'it>f8 . Instead of 1 7 a4 Alexa nder Khal ifman tried 1 7 .tf.fd 1 against Sergey Ivanov (St. Petersburg 1 996), and also failed to gain an advantage: 1 7 . . . �xb 1 18 .Uxb 1 'i¥d6! 1 9 tbe4 'ii'd 5 20 'i!Vxd5 cxd5 2 1 tbc5 �e8 22 .Ub8 'it>f8 .

( Dvoretsky), and the q ueen is trapped . The d i rect 1 4 . . . �a8 {hoping for 1 5 'it'b7? tba5) is less good on account of 1 5 'it'xa8 ! 'ifxa8 1 6 exd5 'it'b7 1 7 tbc3; 1 4 exd5 axb5 1 5 'ii'd 1 exd5 , and it is White who has to try and eq ual ise; 14 tbc3 tbxc3 1 5 bxc3 tbxd4 1 6 'it'xa6 (bad is 1 6 'ii'x c4? tbxf3+ 1 7 �xf3 �b5) 1 6 . . . �b5 1 7 1li'a7 tbe2+ 1 8 'it>h 1 tbxc3 with cha nces for both sides ( Dvoretsky). 13 . . .

.Uxb2

1 3 . . . tbxc3 1 4 bxc3 tbxd4 1 5 'i¥xc4 l2Jxf3+ 1 6 i.xf3 Si.b5 1 7 'iic 5 leads to a n adva ntage for Wh ite . 1 4 'it'xc4

tba5

1 5 'it'd3

.l::t b3

1 6 �fc 1

The move in the game is an i nteresting novelty. Alexey Alexandrov is prepared to retu rn the pawn for the sake of very promising counterplay. 1 1 a3

b5!

12 lbxb5

tbd5!

As shown by Roma n i s h i n , 1 2 . . . a6 1 3 tbc3 tbxd4 was weaker because of 1 4 'it'a5! lbb3 1 5 'ii'e 5 f6 1 6 'ife4 ( 1 6 'ifh5+!? g6 1 7 'ii h 6) 16 . . . tbxa 1 17 axb4 lbb3 1 8 'it'xc4 , and White has a won positio n . 1 3 tbc3 !

After 1 3 e4? ! tbce7 1 4 exd5 �xb5 1 5 'ilt'xa7 exd5 Wh ite's position is preferable. Black's play can be improved by 14 . . . exd5! (instead of 1 4 . . . �xb5) 1 5 'ii'x a7 .l:xb5 1 6 tbe5 0-0 with approxi mate equal ity. Incidentally, in­ stead of the knight retreat Alexa ndrov had prepared 13 . . . a6! ? , when the following vari­ ations are possible: 14 'ifxa6? lbb6! with the threat of 1 5 . . . i.c8

16 . . .

c5!

Usually one should not delay castl ing, but situations sometimes occur when other factors prove more important than simple development. In Black's position there i s a serious defect - the backwa rd c-pawn , which may prove weak. For the moment the activity of the pieces compensates for this d rawback, but if Wh ite should succeed in disentangling himself, he will gain a n obvi­ ous advantage. Therefore Alexandrov hur­ ries to create cou nterplay in the centre and on the q ueenside.

Games by Pupils of the School

267

Let us see what would have happened after the routine 1 6 . . . 0-0 . I n reply 1 7 tt::\d 2?! is a mistake in view of 1 7 . . . .ib5 1 8 'iff3 .U.xc3 ! 1 9 l:txc3 i.xe2 ! 20 'ifxe2 tt::\x c3 2 1 'ifa6 c5 or 2 1 . . . 'ii'x d4 . 1 7 tt::\e 5! is fa r more dangerous. After q u iet conti nuations Black's position is clearly worse : 1 7 . . . c6 1 8 e4 tt::\f6 1 9 'ii'd 1 l:tb8 20 .l:!.ab 1 or 1 7 .. .f6 1 8 tt::\x d7 'i!Vxd7 1 9 i.xd5 exd5 20 'ii'a 6 ( Roma n i s h i n ) . Black is also not saved by 1 7 . . . i.b5 1 8 'ifd2 tt::\x c3 1 9 l:!.xc3 f6 - now both 20 tt::\c6 .ixc6 (20 . . . tt::\x c6 ! ? 2 1 l:!.xb3 tt::\ x d4 22 l:t b 1 ! i.xe2 23 l:!.b4 ! tt::\f3+ 24 i.xf3 'ifxd2 25 l:txd2 i.xf3 26 �d7) 2 1 i.xc6 .U.xc3 22 ll¥xc3 tt::\xc6 23 'ii'xc6 'ii'x d4 24 'ifxe6+ '.t>h8 25 .l:!.c1 c5 26 'ii'c6 ! and 20 tt::\f3 ! ? i.xe2? ! 2 1 l:!.xb3 tt::\x b3 22 �e2 tt::\ x a 1 23 �e6+ ! Wh8 24 'ifa2 a re inauspicious for Black, according to analysis by Dvoretsky and Alexandrov. 1 7 'ii'd 2

After 1 7 dxc5 i.b5 bad is 1 8 'i!fd4 tt::\x c3 1 9 "i!Vxg7 tt::\ x e2+ 20 '.t>h 1 .l::i. f8 , and Black remains a piece u p . Wh ite would have to sacrifice his q ueen : 1 8 'iix b5+ (or 1 8 tt::\x b5 .l::i.x d3 1 9 tt::\d 6+ '.t>e7 20 exd3) 1 8 .. Jixb5 1 9 tt::\x b5 tt::\ b 3 20 tt::\d 6+ 'it>e 7 2 1 tt::\ e 5 - with u nclear conseq uences. The most accu rate contin u ation was prob­ ably 1 7 tt::\e 5! i.b5 ( 1 7 . . . tt::\x c3 1 8 �xc3 i.b5? does not work because of 1 9 'ii'f3) 1 8 'ifd2 tt::\ x c3 1 9 .l:!.xc3, when there appears to be noth ing better tha n 1 9 . . . c4 , tra nsposing i nto a favou rable position for Wh ite , wh ich occu rred i n the game. 17 . . .

c4

1 8 tt::\e 5

tt::\ x c3

1 9 .l:!.xc3

(see diagram) 19 . . .

i.b5

Up to this point both contesta nts had played splendidly and the young player had suc­ cessfu lly stood up to his experienced oppo­ nent. But here Black's play bega n to be

- position after 19 l:txc3 -

affected by lack of time for thought. The move i n the game makes things easier for Wh ite. 1 9 . . . l:!.b8? was incorrect: 20 tt::\xc4 tt::\ b 3 2 1 tt::\d 6+ '.t>e 7 (2 1 . . . '.t>f8 2 2 'iff4) 2 2 .l:!.xb3 .Uxb3 23 'ii'g 5+ f6 24 'ii'c5 '.t>f8 25 ll¥xa7 with advantage to Wh ite (Roma n i s h i n ) . T h a t w h i c h is good as an exception should not be made i nto a ru le! I t was now time to castle 1 9 . . . 0-0 ! ? , when 20 tt::\xc4 'ii'c7 2 1 .l::i.a c1 i s bad because of 2 1 . . . l:!.xc3! 2 2 'i!Vxc3 l1c8 . Alas, as Dvoretsky later established , Wh ite would nevertheless have retai ned his extra pawn , by conti n u i ng 2 1 �cc1 ! ! (instead of 21 �ac1 ?) 21 . . . tt::\xc4 22 �c2 . I n view of the fact that a carefu l analysis of all Black's alternative possibil ities, begin­ ning from the 1 2th move, has not enabled an improvement i n h i s play to be fou n d , one is forced to conclude that Alexandrov's clever opening idea is objectively not altogether correct. 20 a4!

f6

Both players saw that i n the event of 20 . . . i.a6 2 1 .l:!.xb3 Wh ite would gain a powerfu l attack: a ) 21 . . . cxb3 22 tt::\ c6 tt::\ xc6 23 i.xc6+ rtJe7 (23 . . .'it>f8 24 'ii' b 4+ ) 24 �g5+ f6 25 'it'c5+; b ) 2 1 . . . tt::\x b3 22 i.c6+! (weaker is 22 'i!Vb4 li¥xd4) 22 . . . 'it>f8 23 ll¥b4+ '.t>g8 24 .l::i. d 1 h 5

268

�

Games by Pupils of the School

(24 . . . 'ii'f8 25 'iic 3) 25 d5! (th reatening 26 lt:Jxf7 ! ) . 21 axb5

A subtle decisio n . If 22 lhb3 lt:Jxb3 23 ii.c6+, then Black obtains some cou nterplay in the endgame: 23 . . . 'it>f8 24 'i:Vb4+ 'i:Ve7 25 "ifxe7+ 'it>xe? 26 l:txa7+ 'it>f6 27 dxe5+ 'it>xe5, and 28 llxg7?? c3! is bad for Wh ite (Dvoretsky).

Bazh i n

Modern Benoni 1 d4

lt:Jf6

2 c4

e6

3 lt:Jc3

c5

4 d5

exd5

5 cxd5

d6

6 e4

g6

lt:Jxc6

7 f4

ii.g7

8 e5

dxe5

l:txc3

8 . . . lt:Jfd 7 ! ? .

Forced . 23 bxc6

-

U S S R J u n ior Tea m C h a mpionship 1 990

fxe5

22 ii.c6+ !

22 . . .

Bog u slavsky ( 1 5)

23 . . .'it'xd4 would have led to a hopeless end ing: 24 'i:Vxd4 exd4 25 l:.xc4 .

9 fxe5

lt:Jfd7

1 0 e6

fxe6

24 "it'xc3

'it'xd4

1 1 dxe6

'i!Ve7

25 'ii'a 3!

'it>f7

1 2 lt:Jd5!

'i!Vxe6+

1 3 'ii'e 2

'ifxe2+

1 4 ..txe2

..te5?!

25 . . . llf8 would not have hel ped : 26 c7 "ifxf2+ 27 'it>h 1 'it'd? 28 .Ud 1 + ( Romanishin ). If 25 . . . c3 there follows 26 c7! (less good is 26 .l:!.c1 c2 27 'it'b3 0-0 28 e3 'ii'd 2 29 'ii'xc2 'ii'xc2 30 .l:!.xc2 .l:!.c8 , and Black retains saving chances) 26 . . . 'it>f7 (26 . . . 'ii'c4 27 'it'a4+! 'it'xa4 28 c8'ii+ ) 27 .l:tc1 llc8 28 .l:txc3 and wins. 26 e3

�d3

In the event of 26 . . . 'i:Ve4 the simple 27 l�Vxa?+ 'it>f6 28 c7 is good . If 26 . . . 'it'b6 (hoping for 27 'ii'd 6? 'it>f6 ! ) , then , as shown by Romanish i n , Wh ite wins the endgame: 27 'i:Vxa7+ 'i:Vxa7 28 .l::i.x a7+ 'it,Jf6 29 l:ta4 e4 30 llxc4 .l:tc8 (30 . . . 'it>e5 3 1 .l:td4 and 32 lld7) 31 'it>g2 'it>e5 32 'it>h3 (or 32 c? ) 32 . . . '1t>d5 33 .l:!.d4+ . 27 'it'xa7+

'it>f6

28 'it'b7

c3

28 . . . 'i'b3 was more tenacious. 29 .l:!.a7

.Ug8

30 'it'f7 +

'it>g5

3 1 h4+

'it> g4

32 'lt>g2

'ii'e2

33 .l:!.a4+

Black resigned .

Such very sharp opening variations demand a n accu rate knowledge of theory. It is extremely d ifficult to play them, simply on the basis of common sense - the very first i naccu racy may prove fata l . To avoid conced ing t h e i n itiative t o his opponent, Black should h ave sacrificed a whole rook: 1 4 . . . 0-0 ! ! 1 5 lt:Jc7 lt:Jc6 1 6 lt:Jxa8 lt:Jb4 . But is it conceivable to take such a decision at the board , without preparatory analysis at home? 15 lt:Jf3

lt:Jf6

1 5 . . . ii.d6? 1 6 ii.h6 ! . 1 6 ii.c4

lt:Jxd5

1 7 ii.xd5

ii.f6

1 8 0-0

lt:Jc6

(see diagram) 1 9 ii.g5!

The strategy chosen by Maxim Bog uslavsky, typical of such positions, is rather i nstruc­ tive . Exchange the opponent's already developed pieces - then your lead in development will become especially ap­ preciable.

ttJ

Games by Pupils of the School

-

position after 18 ... ttlc6-

19 . . .

�xb2?

This ' pawn-g rabbing' when beh i n d i n devel­ opment is severely pun ished . 1 9 . . . �xg5 20 lt:JxgS �fS was essenti a l . 20 �xc6+ 1

bxc6

2 1 l:tae 1 +

d7

22 lt:Je5+ 1 ?

The same idea! However, the enemy bishop would also have been excha nged i n the variation 22 .l:te7+! d6 23 .l:td 1 + .i.d4+ 24 tt:Jxd4 cxd4 25 l:lxd4+ cs 26 �e3 bs 27 .l::r.e 5+ cS 28 .l:.d3! with a q u ick win ( i n dicated by Dvoretsky). 22 . . .

�xe5

23 l:l.xe5

when there a re opposite-colour bishops a material advantage has no particular sign ifi­ cance - it is far more i m portant to have an attack. The black king is in deadly danger. If 23 . . . c4 ! ? Boguslavsky was i ntending 24 .l:.f7+ d6 25 �f4 �a6 (25 . . . �g4 26 h 3 g5 27 lhg5+ e6 28 l:tc7 �f5 29 .l:txc6+ dS 30 .l:tf6 ) 26 .l:.aS+ e6 27 .l:tc7 .i.bS (27 . . . �c8 2 8 .l:.e5+ f6 2 9 l:xc6+ f7 3 0 �gS �f5 31 :c7+ g8 32 �h6, and Wh ite wins) 28 a4 , overlooking the reply 28 . . . a6! . In­ stead of 26 .l:ta5+ stronger is 26 :tgS+ e6 27 l:tc7 .l:.hf8 (27 . . . .l:.hc8 28 .l:tg5+ d6 29 .l:taS+) 28 .l:txc6+ d7 29 .l:tc7+ e6 30 g3 or 30 �g3, reta i n i ng the advantage, but even so, with 22 .l:te7+ ! Wh ite would have achieved more . 23 . . .

c7

24 l:te7+

�d7

24 . . . b6 would have led to mate: 25 l:tb 1 + as 26 �d2+ a4 27 lle4+ a3 28 �c1 + xa2 29 .l:tb2+. .l:tad8

25 .l:td1

2 5 . . . .l:thd8 26 .l:txh7 . 26 �f4+

b6

26 . . . c8 27 .l:.b 1 . 27 l:[ b 1 +

a5

28 �d2+

a4

29 l:le4+

c4

30 l:txc4+

a3

31 �c1 +

xa2

32 .l:tb2+

a3

3 3 .l:tb7+

B lack resigned . Svidler ( 1 5) - Arkh ipov

Gausdal 1 99 1

French Defence 1 e4

In the endgame Black is two pawns u p . But

269

e6

2 d4

d5

3 lt:Jd2

lt:Jf6

4 e5

lt:Jfd7

270

�

Games by Pupils of the School

5 c3

c5

6 Ji.d3

lbc6

7 lbe2

cxd4

16 . . .

Ji.e7

'ii' b 6

1 7 'i!lg4

g6!

8 cxd4

1 8 f4!

9 0-0

A problematic pawn sacrifice. As compensa­ tion White gains an endu ring i nitiative, thanks to his lead in development. 9 . . .

lbxd4

1 0 lbxd4

'ii'x d4

1 1 lbf3

'ii'b 6

1 2 'ii'a4

'i!Vb4

The manoeuvre of the wh ite q ueen to g4 should be prevented . 1 3 'i!Vc2

'i!Vc5

1 4 'i!Ve2

The theoretical equal ity.

of 1 7 Ji.d2! 'ii'c 7 1 8 lbb5 'ii' b 8 1 9 Ji.f4 f6 20 l:tac1 ! Ji.d6 21 llxc8 + ! .

1 4 Ji.xh7 b6 leads to

Not 1 8 Ji.xg6? lbxe5. 18 . . .

lbc5? !

I n Svidler's opin ion , 1 8 . . . h 5 ! ? 1 9 'ii'g 3 Ji.c5 came i nto consideration , but not 1 9 . . . lbc5?! 20 Ji.xg6! fxg6 (20 . . . l:r.g8 21 Ji.xf7 + ! �xf7 22 'ii' h 3) 2 1 'ifxg6+ �d8 22 f5! exf5 23 Ji.g5! .U.e8 24 b4! (or 24 'ii'd 6+ Ji.d7 25 b4! ) , when 24 . . . 'ii'x b4 25 'ifd6+ Ji.d7 26 lbe6+ lbxe6 27 Ji.xe7+ llxe7 28 'ii' x b4 is bad for Black. 19 Ji.xg61

Destroying the pawn screen ; i n ful l accord­ ance with the demands of the position, Wh ite beg ins a n attack on the king. 19 . . .

fxg6

1 9 . . . l:tg8 20 Ji.xf7+ �xf7 21 hopeless.

'it'h5+ is

20 'ifxg6+

�d8

2 1 f5

exf5

If 2 1 . . .'ilc7 , then 22 lbb5 ! . 2 2 e6

'it'a6

23 .Uad 1

Here Wh ite had a serious alternative: 23 .l:tac1 !? lbxe6 24 lbxf5 (but not 24 .l:txc8+? .l:txc8 25 lbxe6+ �d7, and it is B lack who wi ns) 24 . . . lle8 25 lbxh6!? with a strong attack. 23 . . . 14 . . .

h6?!

It is more logical for Black to continue his development: 14 . . . i.e7 1 5 Ji.e3 'ii'a 5, al­ though after 16 'ifc2 ! ? White retains q u ite good compensation for the sacrificed pawn . 1 5 Ji.e3

'ii'a5

1 6 lbd4!

The direct consequence of the loss of time on the move of the rook's pawn. Wh ite prepares f2-f4 . 1 6 . . . lbxe5? is bad because

Ji.xe6

23 . . . lbxe6 24 lbxf5 was worse . 24 lbxf5 !

.U.g8!

Accord ing to analysis by Svidler, other conti n uations lose: a) 24 . . . .l:tc8 25 lbxe7 �xe7 26 Ji.xc5+ .U.xc5 27 'it'g7 + ; b) 24 . . . lbe4 25 lbxe7 �xe7 26 'it'g7+ �d6 27 Ji.f4 + ; c) 2 4 . . . Ji.xf5 25 'it'xf5 ! 'it'e6 26 .l:txd 5 + . 25 'it'h7

Ji.f8 !

ctJ

Games by Pupils of the School

Bad is 25 . . . .l:te8 26 liJxe7 lhe7 27 'iii' h 8+ .l:.e8 28 'iff6+ 'it'd? 29 .llx c5, while if 25 ... .l::t x g2+ White decides matters with 26 'itxg2 'ii'e 2+ 27 'ith 1 .llxf5 28 'iii'xf5 'iii'x e3 29 'iii'x d5+ 'itc7 30 I:tde 1 'ii'g 5 3 1 I:txe7+ 'iii'x e7 32 11f7 (Svidler). H owever, 25 . . . .lld 6 ! ? came into consideration . 26 liJxh6

.ll x h6

27 'ii'x h6

27 1

28 . . . 'itd7 , then 29 'ifh7+ 'itc6 30 l:tc1 + 'itb5 31 llxa8 llxa8 32 a4+ ! is decisive . 29 l::t x a8

lba8

29 . . . l::1 x g2+? 30 'ith 1 ! is pointless. 29 . . . 'ife2 ! ? would have posed more problems. Then only a d raw results from 30 'ifh7+ .llf7 31 'it'h4+ 'it'd? 32 'ii' h 3+ 'ite7 33 llc1 ! ? (33 .l::.f 1 l:!.xa8 34 'ifh4+ 'ite6) 33 . . . l:txa8 ! (33 . . J�xg2+? 34 'ith 1 ! ) 34 llc7+ 'itf6 35 'ifh6+ .ll g 6 36 'ii'g 7+ 'itf5 37 'it'd?+ 'ite5 38 'ii'e 7+ (Svidler). However, stronger is 30 'ii' h 4+! liJf6 (30 . . . 'itd7 3 1 'ii' h 7+) 31 .ll c 5+ 'itf7 ( 3 1 . . . 'itd7 32 'ii'a 4+ 'itc7 33 'iii'f4+ 'itc6 34 'ii'd 6+ 'itb5 35 a4+ 'itc4 36 l:!c1 + 'itb3 37 .l:.xg8 ) 32 l:tf8 + ! ! l:!.xf8 33 l:lf1 , and Wh ite wins. 30 'ii' h 4+!

'itd7

If 30 . . . 'itd6, then 31 'ii'x e4 ! . Now the captu re of the knight is less convincing i n view of 3 1 . . . 'ii'e 2 , and so White plays for mate.

27 . . .

liJe4?

The middlegame is in progress, but Black has not yet completed his development. He would l i ke to include h i s rook in the play, and therefore 27 . . . .l:tc8 looks logica l . But then there follows 28 .llx c5 ! llxc5 29 .llf8+ , for example, 29 . . . 'itc7 30 .llx g8 .llx g8 3 1 'ii'g 7+ 'itb6 32 'iii'x g8 'iie 2 (hoping for 33 l::1f 1 'ii'e 3+ 34 'ith 1 l:tc 1 ) 33 'ii'g 6+ and 34 'ii'd 3 , and Wh ite retains an advantage sufficient for a wi n . Svidler suggests 27 . . . 'ii'e 2 ! 28 g3 liJe4 2 9 'ifxe6 'ifxe3+ 30 'itg2 'ife2+ 3 1 'itg 1 'ii'e 3+ with perpetual check. If White does not want a d raw, he can try 29 .llf4! ? . After Black's mistake t h e attack becomes irresistible. 28 .l:tf8+

'ite7

28 . . . .l::t xf8? 29 'ii'xf8+ 'it'd? 30 'ii'x a8 'ife2 does not work in view of 3 1 'ii'x b 7 +