Anatoly Karpov & Gik - Chess Kaleidoscope

Chess Kaleidoscope

A.KARPOV & Y.GIK PERGAMOM PRESS

Chess Kaleidoscope In Chess Kaleidoscope, World Champion Analoly Karpov has collaborated with Mathematician and Chess Journalist Yevgeny Qik to produce an instructive and entertaining book, which should satisfy the demands both of chess enthusiasts and of serious players. The book is in three parts, the first of which tells of matches For the World Championship, of decisive games, of combinations, problems and studies on the chess board, and of the unusual properties of the board and pieces. Part two deals with a currently fashionable topic. The Computer at the Ches1 Board , and in the third part the World Champion annotates fifteen of his best games, many of which are from recent events dnd may be unfamiliar to the Western reader.

PERGAMON RUSSIAM CHESS SERIES

0 08 026896 X

PERGAMON RUSSIAN CHESS SERIES General Editor Kenneth P. Neat Executive Editor Martin J. Richardson

AVERBAKH, Y. Chess Endings: Essential Knowledge BOTVINNIK, M. M. Achieving the Aim Anatoly Karpov: His Road to the World Championship Selected Games 1967-70 ESTRIN, Y. & PANOV, V. N. Comprehensive Chess Openings KARPOV, A. & ROSHAL, A. Anatoly Karpov: Chess is My Life LIVSHITZ, A. Test Your Chess IQ, Books 1 & 2 NEISHTADT, Y. Catastrophe in the Opening POLUGAYEVSKY, L. Grandmaster Preparation SUETIN, A. S. Modern Chess Opening Theory TAL, M., CHEPIZHNY, V. & ROSHAL, A. Montreal 1979: Tournament of Stars

Chess Kaleidoscope By

A. KARPOV and Y. GIK Translated by KENNETH P. NEAT

PERGAMON PRESS OXFORD • NEW YORK • TORONTO • SYDNEY • PARIS • FRANKFURT

UJC.

Pergamon Press Ltd.. Headington Hill Hall, Oxford OX3 OBW, England

U.S.A.

Pergamon Press Inc., Maxwell House. Fairview Park, Elmsford, New York 10S23, U.S.A. .

CANADA

Pwgamon Press Canada Lid., Suite 104, ISO Consumers Road, Willowdale, Ontario M y 1P9, Canada

AUSTRALIA

Pergamon Press (Aust.) Pty. Ltd., P.O. Box 544, Polls Point, N.S.W. 2011, Australia

FRANCE

Pergamon Press SARL, 24 rue des Ecoles, 75240 Paris, Cedex 05, France

FEDERAL REPUBLIC OF GERMANY

Pergamon Press GmbH, 6242 Kronberg-Taunus, Hammerweg 6, Federal Republic of Germany Copyright© 1981 Pergamon Press Ltd. All Righls Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic tape, mechanical, photocopying, recording or otherwise, uijfhout permission in writing from the publishers. Firs! edition 1981 Britilh Library Cataloguing in Publication Data Karpov, Anatolii Chess kaleidoscope. - (Pergamon Russian chess series) 1. Chess I. Title II. Gik, Y. III. Shakhmaty kaleidoscope. English 794.1 G VI445 ISBN 0-08-026897-8 (Hardcover) ISBN 0-08-026896-X (Flexicover) Library of Congress Catalog Card no: 81-82527

Printed in Great Britain by A. Wntatan & Co. Ltd,, Exeter

.

Foreword

The game of chess is inexhaustible, and no one definitely knows what it is — science, art or sport, A professional interest in this ancient but eternally youthful game is displayed by people of various professions: psychologists and doctors, mathematicians and cyberneticists, as well as historians and teachers. And in their leisure time, millions of people simply like to play chess, from school children to academicians, and from wise phi losophers to light-minded drummers in pop groups. Of course, the sport element In chess is also very great, and competitive success is of considerable significance. The recognition and popularity of a player, and ultimately his very livelihood, depend on the results which he scores in tournaments. This is why many players strive to perfect their mastery. For practical players in many countries of the world a great deal of chess literatur-e is published — opening monographs, games collections, and books on the endgame. But chess would be dead, were it not for the many thousands of admirers and supporters who have a disinterested love of chess, who know everything about i t , and at the same time do not have any great ambitions. The authors have written the present book in the hope of satisfying to some extent the demands of both categories of chess player — both enthusiasts and serious players. The book acquaints the reader with the most varied aspects of chess, both as a game and an art, and therefore we have decided to call it Chess Kaleidoscope. The book is in three parts. The f i r s t part, 'Seven Chess Letters', tells of matches for the World Championship and decisive games, of combinations, problems and studies on the chess board, and of the unusual properties of the chess board and pieces. The second part of the book, 'The Computer at the Chess Board', is wholly devoted to a currently fashionable lopic — to chess playing by electronic computers. Here are described competitions between machines, including three World Computer Chess Championships. Games by computers are given, played both against other computers, and against humans. We describe the successes of the f i r s t Computer World Champion, the Soviet program Kaissa, in the analysis of chess endings (one of the creators of Kaissa, mathematician and chess master A. Bit man, assisted in the writing of this part of the book).

vi

Foreword

The t h i r d part of the book, 'Fifteen of the World Champion's Best Games', is purely chess. Each game is annotated by the World Champion, and is accompanied by a brief introduction, e x p l a i n i n g why that p a r t i c u l a r game is dear lo the author, and why it was included in the select Ion. How did the idea of w r i t i n g this book arise? Its authors f i r s t became acquainted in 1968 in (he Moscow University Chess Championship, when the game between the f i n a l - y e a r student Y. Gik and the f i r s t - y e a r student in the same University f a c u l t y A. Karpov played an Important role. The latter came out on top, and also won the t i t l e of Moscow University Champion, while the former finished In second place (the Karnov-Gik game opens the t h i r d part of the book). Shortly a f t e r wards, one of the authors moved from Moscow to L e n i n g r a d , and t r a n s ferred !o the Economics Faculty of Leningrad University (he finished with a distinction there several years l a t e r ) . In 1975 Anatoiy Karpov became chess champion of the world, and three years later succeeded in r e t a i n i n g this t i t l e in the match in Baguio. During the intervening years the second author has defended his mathematics dissertation, published more than t h i r t y scientific works, and, in passing, written two chess books. Now World Champion Anatoiy Karpov once again lives in Moscow; he is edi t o r - i n - c h j e f of the Chess Review 64, while Yevgeny Gik, who is a member of the USSR Union of Journalists, uses his connections to write from time to time for this magazine. Since 1980 the authors have been w r i t i n g a chess column i n the popular science magazine Kvant* ( i n this way the World Champion has succeeded in r e t u r n i n g to mathematics!}. The idea of this book In fact came to the authors when w r i t i n g their 'chess letters' for Kvant. This, In p a r t i c u l a r , explains why the chapters of the f i r s t part of the book are called ' l e t t e r s ' . Although d u r i n g the work on the book the ' l e t t e r s ' from Kvant were at hand, in the 'kaleidoscope', as should be the case, their contents have been completely mixed up. The result, we hope, is a r i c h , diverse and unusual book, In which every reader w i l l f i n d for himself something of interest.

A. Karpov, Y . Gik

• K v a n t = Quantum

(K.P.N.)

Contents

PART ONE - SEVEN CHESS LETTERS

-

.

First Letler — The Geometry of the Chess Board Second Letter — Symmetry in Chess Third Letter — An Ancient Mate Fourth Letter — Forty Chess Quartets Fifth Letter- — Trios and Duets on the Chess Board Sixth Letter — Two Decisive Games Seventh Letter - Before the 30th Match for the World Championship

3 15 24 33 54 66 7B

99

PART TWO - THE COMPUTER AT THE CHESS BOARD Why program computers to play chess? Competitions with the Participation of Computers The Computer Analyzes the Endgame PART THREE No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 No. 9

101 103 115

FIFTEEN OF THE WORLD CHAMPION'S BEST GAMES

123

Karpou-Gik, Moscow University Championship, 1968 Karpov-Hort, Moscow 1971 Karpov-Polugayevsky, Q u a r t e r - F i n a l Candidates' Match, Moscow 1974 Karpov-Spassky, Semi-Final Candidates' Match, Leningrad \9Tt Karpov-Korchnoi, Final Candidates' Match, Moscow 1974 Portisch-Karpou, European Team Championship, Moscow 1977 Tlmman-Karpov, Montreal 1979 Karpov-HCibner, Bad Lauterberg 1980 Kar-pov-Tal, Bugojno 1980

125 129

vii

133 137 lft H '* * H9

1

Contents No. No. No. No. No.

10 11

12 13 14

No. 15

Gl i g o r i c - K a r p o v , Bugojno 1980 K a r p o v - R i b l i, Amsterdam 1980 L a r s e n - K a r p o v , Amsterdam 1980 K a r p o v - 5 p a s s k y , T i l b u r g 1980 K a r p o v - H o r t , Malta Olympiad 1980 L j u b o j e v i c - K a r p o v , L i n a r e s 1981

151 154 157 159 162 166

Part One Seven Chess Letters

First Letter THE GEOMETRY OF THE CHESS BOARD

In order to play well al chess i l is nol essential to be a good mathematician. The continuous calculation of variations, which a player has to conduct during a game, is of a different nature from the work of the calculating mathematician. Me vert he less, the game of chess contains certain mathematical features, with which we are going to acquaint you. We will begin with the fact lhat the chess board possesses unusual geometric properties, which are of great importance in the theory of the endgame. Let us begin with the following famous study. R. Reti, 1921

White to play and draw. Surprising though it may seem, White succeeds in catching the enemy pawn. Of course, if the king heads for it by the direct route (1 Kh7 h4 2 Kh6 h3, etc.) the pawn safely promotes to a queen. But White chooses a more cunning route: I Kg7! h4 2 Kf6! Kb6 (on 2. . .h3 there follows 3 Ke7 h2 4 c? Kb7 5 Kd7, and the pawns queen simultaneously) 3 Ke5! Now 3...h3 & Kd6 h2 5 c7 again leads to a draw, but no better is 3...Kxc6 4 Kf4 h3 5 Kg3 hZ 6 Kxh2 - the king catches the pawn on the threshold of queening. As we see, White is saved here by a purely geometric idea, which consists of the fact that the shortest distance on the chess board is not necessarily measured in a straight line. In the given example the

4

Seven Chess Letters

path of the white k i n g from he to hZ occupies six moves, both in a straight line, and in a z i g - i a g ; but in the second instance Black is forced to lose two extra temp i, which enables his 'unstoppable' pawn lo be halted. When this pawn study was published, it caused a genuine sensation in the chess world. The geometric idea on which it i ? based has subsequently been improved many times, but for p u r i t y of form it is Impossible to surpass the o r i g i n a l . Seven years later, Retl gave his 'discovery' an even more paradoxical form.

R. Reti,

White to play A passed 5 Kd6, 1...f5

by

single pawns! with a 2 Kxg7

1928

and

draw.

white pawn draws against three opposing connected 1 Kg6 Kb6 2 Kxg7 h5 ( 2 . . . f S 3 Kf6 f4 h Ke5 f3 draw) 3 Kxf6 hA 4 Ke5, with the familiar f i n i s h ; f4 3 Kf6 13 !3...Kb6 4 Ke5) 4 Ke7(e6). Draw.

We will consider one further a 'geometric' method.

I.

endgame

Maizelis,

White to play

study,

in

which

White

wins

1921

and w i n .

The a7 pawn cannot be defended, and Black's one chance is to answer the inevitable Kxa7 with . . . K c 7 , not allowing the enemy k i n g out of c a p t i v i t y . The path of the white k i n g to the a7 pawn takes five steps, and there are 30 different ways of capturing the pawn In this number of moves, but only one of them wins: 1 Ke6t Kc3 2 Kd3! The white k i n g , as chess players say, 'shoulder charges' his black opponent. Now the latter cannot go to d4, and loses a decisive tempo: 2..,Kd3 3 Kc6 Kd4 * Kb7 Kc5 5 Kxa7 Kc6 6 Kb8, etc. Other ways

1st Letter — The Geometry of the Chess Board

5

do not work, for example 1 Ke6 Kc3 2 Kd6 Kd4 3 Kc6 Ke5! 4 Kb7 Kd6 5 Kxa7 Kc7, with a draw. A tragic incident, associated w i t h . . . the geometry of the chess board, occurred in 1951 iri the match for the World Championship between Botvinnik and Bronstein, as will be described in the seventh 'letter' . Let us now examine (wo very simple geometric endings, which every chess player should know.

rules

for pawn

The Rule of the Square In the following position the while king is not participating in the play, and everything depends upon whether or not its black opponent can catch the h3 pawn.

Inexperienced players usually reason as follows: the pawn goes here, the king — there, the pawn — here, the king — there, and so on. In doing so they frequently get confused (especial ly if there are other pawns on the board}, and in the end miscalculate. However, the outcome of the game can easily be evaluated using the 'rule of the square'. It is sufficient to check whether or not the king can, on Its move, enter the 'square' of the pawn — In the given Instance that depicted in the diagram. For convenience one need mentally draw only one line — the diagonal of the square (h3-c8). Thus, in our position Black lo move can draw (he enters the square), but if It is the opponent's move Black loses. The well-known chess master and composer N. Grigoriev was involved in an amusing incident in the days of his youth (1913). Once in a chess club he was playing a game against an old man. The youth was very agitated, since he assumed that he was up against some eminent old player. But things went quite well, and in the end the following position was reached.

With a trembl ing hand the old man took the b-pawn — 1 ...Kxb2,

6

Seven Chess Letters

but then the other pawn rushed forward — 2 a4. The enemy king hastened after it — 2...Ka3 3 a5 Ka4. Al this point the young player began to have doubts; what if the old man should know some magical secret, and was not pursuing the pawn in vain? There was nothing to lose, and the race continued — 4 a6 Ka5 5 a7 Ka6. The black king has kept r i g h t behind the pawn. Here, with a cry of ' q u e « n ! ' , White made the last move with his pawn — 6 aB=CH. The old man's face was covered in g r i e f . "Ah, too l a t e ! ! " , he sighed d e s p a i r i n g l y . The old man had believed in his good fortune, bul the sudden appearance of the queen shattered al I his hopes. . . . See whal a sad tragedy can result from not knowing the rule of the square! Who knows, perhaps this episode with the pursuit of the pawn played its p a r t , but, at any event, Nikolai Grigoriev subsequently became one of the greatest specialists in the world in the f i e l d of pawn endings. Here is a s t r i k i n g ex amp le; to an international competition for studies in France, held in 1936, he sent ten of his pawn studies, and a l l ten won prizes!

The Rule of the

Triangle

In the following position Black to move loses he allows the white king in at b6 and loses his only

White to play

immediately, pawn.

since

and w i n .

But It is now White's move, and he has to give his opponent the move, in other words, gain a tempo. After 1 Kd5 KcB nothing is achieved by 2 KdG Kd8 3 c7+ Kc8 4 Kc6 stalemate!, while 2 Kc5 Kc7 leads to the i n i t i a l position. The goal is achieved by means of ' t r i angulation1. For the given example this t r i a n g l e (c4-d4-d5) is depicted in the diagram. After 1 Kd5 Kc8 2 Kd4 Kb8 3 Kc4! Kc8 4 Kd5 the necessary tempo has been won! Now on 4 . . . K d 8 White wins by 5 Kd6 Kc8 6 c7, and on 4...Kc7 by 5 Kc5.

Geometric

Opposition

Opposition plays a basic role in the p l a y i n g of pawn endings. Geometric opposition means that the white and black kings are situated on the same line, and are separated by ar\ odd number of squares. If there is one square, the opposition Is eel led close, if there ere three or f i v e , it is called d i s t a n t . With the kings on the same f i l e , rank or diagonal, the opposition is called vertical, horizontal or diagonal, respectively. Thus, to gain the opposition, one has to use arithmetic — calculate the number of squares between the k i n g s . If the pawns are f i x e d , the side possessing the opposition (with the opponent

1st Letter — The Geometry of the Chess Board

7

to move) attains his goal — wins, or gains a draw. In every case, a decisive role is played by manoeuvres of the kings (and pawns, if they are not blocking one another).

White to play

and draw.

.,

In order to gain a draw, White must take the opposition. But in the event of close opposition he is hindered dy his own pawn: 1 Kfi? Kd2 2 Kf2 Kd3!, and the opposition is lost - 3 Kg3 Ke3 4 Kg2 Ke2 5 Kg3 Kfl 6 Kh3 Kf2 7 Kg4 Kg2, and wins. The game is saved only by the distant opposition: 1 KM ! Kc2 (after 1...g4 2 Kg2 Kd2 3 f x g i eU 4 g5 the pawns promote simultaneously) 2 Kg2 Kd2 3 Kh2! Ke2 4 Kg2 Ke3 5 Kg3, with a draw. Now an

amusing

example,

which

White to play

also

relates

to

the

opposition.

and w i n .

The reader may be confused by the task, since after all White has an extra rook. But here there is one additional condition — the rook is allowed to move, only if i t . . . gives mate! After 1 Kg2! White takes the opposition (the kings are separated by an odd number of squares — f i v e ) . If now the black king keeps on the g - f i l e , White retains the opposition — 1...Kg7 2 Kg3! (the distance Is again an odd number, three squares) 2...Kg6 3 Kg4! (one square). Thus, Black is forced to move off the g - f i l e - 3...Kh6 4 Kf5! Up t i l l now the white king could not stand in front of the rook, since his black opponent would have immediately escaped to freedom across the f - f l le. But now this possibility has appeared, and White carries out an outflanking manoeuvre. 4 . . .Kg7 (alas, after 4 . . .Kh6 the rook is allowed to come Into play - 5 Rhi mate) 5 Kg5! (again the opposition

has been won) 5 . . . K h 7 6 Kf6! KgB 7 Kg6! KhB 8 Rf8 mate ( 6 . . . K h 8 7 Kf7 Kh7 8 Rhi mate). After 1 Kh2? KhS! 2 Kg3 Kg7 3 Kh4 Kh6 the b l a c k k i n g cannot be p i n n e d a g a i n s t the w a l l .

8

Seven Chess Letters The Theory of Corresponding Squares

On making the acquaintance of the concept of opposition, we saw that in pawn endings, in order to win (or d r a w ) , the one k i n g , for various positions of his opponent, must occupy corresponding squares (the opposition), and thai if He cannot do t h i s , th« goal is not achieved. This correspondence often proves to be more complex than that examined above, and therefore the opposition is essentially a specific instance of the concept of 'corresponding squares', a r i s i n g In the analysis of positions with a blocked pawn structure. in their study, various methods are used: 'Bianchett i 's c r i t i c a l distances', 'Ebersr's co-ordinate system', and others. The theory of such endings is cal led the theory of corresponding squares. The analysis of each concrete position can be regarded as the solution of a subtle mathematical problem, although there is no single algorithm in existence. Uet us investigate the following f a i r l y simple example (While to move).

Black's position is not easy — the white k i n g is threatening to break into his camp either via f3, or via b7. Black must endeavour to prevent both these intrusions. Let us consider various positions of the white k i n g , and determine the corresponding squares for Black. Let us begin with the c r i t i c a l intrusion squares — f3 and a6. If the white k i n g stands at f3, 6lack must not allow it (o reach g4, i . e . his k i n g must stand at g5 (from h5 he w i l l be too f a r away from the 0 side). Thus on the corresponding squares f3 and g5 we w r i t e the number 1 . If the white king has a r r i v e d at a6, Black must meet it at b8 (number 2 ) , Suppose that the white k i n g stands at e2. Since it is threatening to occupy square 1 in one move, and square 2 in four, the black k i n g must be at f6 (at e2 and f6 we place the number 3 ) . The white king can go to a6 from a5 and bS, and this means that these squares correspond to c8 (number 4 ) . The white k i n g can reach the squares with number 4 from b4 or c4, and they correspond to d8 (number 5 ) . From d3 the king can go to squares 3 and 5, i . e . it corresponds to square e7 (number 6 ) . S i m i l a r l y , from c3 the king can occupy squares 5 and 6, and it corresponds to e8 (number 7 } . From d2 the king has moves to squares 3, 6 and 7, and the corresponding square is f7 (number 8 ) . The remaining squares do not play any role. Thus the corresponding squares are not mutually equivalent — two p a i r s of white squares each correspond to one black square, and this decides matters: 1 Kf3 KgS 2 Ke2 Kf6 3 Kd3 (3 Kd2 Kf7) 3 . . ,Ke7 A Kc4 (4 Kc3 Ke8) 4,..Kd8. Up t i l l now Black has found the necessary reply, but after 5 Kb4! (the white king stays on a square with the number 5) Black loses the correspondence and with it Ihe game: on 5-..Ke7(e6) there follows G Kb5, and on 5...Kc8 - 6 Kc3 Kd8 7 Kc4I, and on one sector of the board Ihe white king breaks through to the black pawns. In the

given

example

Black

has

little

opportunily

for

manoeuvre.

1st Letter — The Geometry of the Chess Board

9

and therefore White could even have permitted himself an inaccuracy. For example, if he had gone S Kb5 (instead of 5 Kb4), then after 5 . . . Kc8 he could have repaired h i 5 mistake by 6 Ka5! etc. But Often in such positions one inexact move can have irreparable consequences.

White to play and w i n .

'

The analysis of this position is considerably more complicated. White intends to break through either v i a d6, or via f4, and the black king must hinder both these plans. Thus if the while king reaches c5, Black must meet it at e7 (with his king at d7 he would not have time to defend the g4 pawn after Kc5-d4-e3-f4), i.e. e7 corresponds to the square c5. With the white king at f4, Black's must be at h5, i.e. h5 corresponds lo f4. If the white king has reached d4, Black's at this point must occupy f7, so as to answer Kc5 with ...Ke7, and Ke3 with . . .Kg6. From c4 White can go both Kc5 and Kd4, and in this case Black's king must be at fB, so as to stand at e7 (after Kc5) or at f7 (after Kd4). From d3 the moves Kc4, Kd4 and Ke3 are possible, and therefore the square corresponding to d3 is g7. By successively going round all the most important squares at the disposal of the white k i n g , and seeking the coresponding squares for Its black colleague, we obtain the f o l lowing picture, where, as above, corresponding squares are indicated by the same number.

Now the solution can be found almost automatically. White should be guided by the following r u l e : to place his king on a squar-e which at the given instant corresponds to the square of the black k i n g , or on a square, the corresponding square to which is Inaccessible to t black king within one move. Since the corresponding square to bl is g7, and that to b2 is h7, and to a2 — h8, the only move to win is 1 Ka1-a2!! After 1 Kb1? Kg7! or 1 Kb2? Kh7! Black gains a draw. Since the subsequent play Is simpler, we give only the main variation (if Black plays otherwise, he loses more q u i c k l y ) . 1 Ka2!I Kh7 2 Kb2! Kg7 3 Kb3! Kg8 4 Kc3! Kf8 5 Kc4! Kf7 6 Kd4! By th« staircase depicted i n the d i a g r a m , the white k i n g has completed its

10

Seven Chess Letters

ascent onto the highest step, and now Black is defenceless! Obviously, by means of our table of numbers it is also easy to evaluate the position for other i n i t i a l piacings of the kings, but only for the given pawn structure. Up t i l l now we have been t a l k i n g only about pawn endings. However, various mathematical elements are also contained in other endings. Consider, for example, one form of the ending 'rook and pawn against rook 1 .

Draw How should this position be evaluated? White has an extra pawn, but It turns out that he cannot win. For the evaluation of such positions, in which Black's k i n g is cut off from (he white pawn, but his rook is attacking the pawn from the front, there exists a simple arithmetic r u l e . It Is called the ' r u l e of f i v e ' , and goes as follows. If the number of the rank occupied by the pawn and the number of files separating it from the weaker side's king in total give a number less than or equal to f i v e , the position is drawn; but If this number is greater than f i v e , the stronger side wins. Thus the position in the above pawn stands on the t h i r d rank, and 3 + 2 = 5! Play can go as follows: Rc8+ 4 Kb5 Rd8 5 Rdi Kf6 6 d4 with a simple draw — the black king front of the pawn.

diagram is drawn. Indeed, the the king is cut off by two f i l e s , 1 Kc3 RcB+ 2 Kb4 RdB 3 Kc4 (6 Kc6 KeS) 6...Ke7 7 Kc6 Rc8+, succeeds in occupying a square in

White wins. In this position, which differs from the previous one in that the pawn and k i n g have been moved up one rank, White now wins — the sum in question is greater than f i v e : 4 + 2 = 6! This is how the win is achieved: 1 KcA Rc8+ 2 Kb5 RdB 3 Kc5 RcB+ 4 Kb6 RdB 5 Rdi Kf6 6 Kc7! RdS 7 Kc6 Rdfl 8 d5, and the pawn attains the eighth rank without d i f f i c u l t y .

1

1st Letter" — The Geometry of

the Chess Board

11

White wins. And t h i s position d i f f e r s from the i n i t i a l one by the black k i n g and white rook being moved one f i l e to the r i g h t . The black k i n g is cut off from the pawn by three f i l e s , and again 3 + 3 = 6! White decides the game by 1 Kc3 RcB+ 2 Kdfi RdS+ 3 Ke4 Re8+ 4 Kf5 Rf8+ 5 Ke6 Rd8 6 Rdl Rd4 7 Ke5 RdB 8 d4, etc. We have studied three positions i n which White's e x t r a pawn was a central one. However, the given r u l e is also perfectly applicable to positions with a bishop's or k n i g h t ' s pawn.

Draw According to the ' r u l e of f i v e ' , t h i s is a draw. 1 Kb3 RbB+ 2 Ka4 Rc8 3 Kb4 RbB+ 4 Ka5 Rc8 5 Rci Ke3 6 c4 Kd2, or 1 Re6 Kf5 2 Rb6 KeS 3 Kb3 Kd5, w i t h a simple d r a w . The result does not change if t h i s last position \s shifted either- one f i le to the r i g h t , or to the left. It has to be said that the ' r u l e of f i v e 1 can prove useful only to •a human. But for a computer p l a y i n g chess, it is no use at a l l . As w i l l be described in the second p a r t of the book, a computer has made an exhaustive study of a l l endings of the type 'rook and pawn against rook 1 , and is able to make a f a u l t l e s s e v a l u a t i o n of any one of them. We w i l l now describe an amusing instance, which bears some r e l a t i o n to our theme. It happened 15 years ago in a group of students, who were good mathematicians (but not very good chess p l a y e r s ! ) . One of the authors of t h i s book, then s t l l I a student (at that time the second author was studying at school, and could h a r d l y have guessed that w i t h i n ten years he would become the strongest player on this p l a n e t ) , for amusement offered his colleagues the following problem of a so-called reverse mate.

12

Seven Chess Letters

White to play

and force Black

to give mate.

In this position Black does not want to mate the white k i n g , but White forces him to do t h i s . Since Ihe task was an unusual one, I decided f i r s t of a l l . . . to show my colleagues how to give mate, and so I took the white pieces: 1 Bg2 Bd5 2 Qd8 Bb7 3 Be4 Bc6 4 0c8 Bb7 5 Bd5 Bc6 6 Qh8 Bb7 7 Bb6, and Black is forced to play 7...Bxc6 mate. The students said that it was a l l clear to them, and we turned the board round. But after 1 Bg2 Be4 2 0d8 Bd5 3 Qe8 Bc6 4 Qf8 Bb7 5 Qh8 Bf3 6 QdB Bd5 7 Be4 Bb7 8 Qc8 Bc6 9 Bd5 Bb7 they were disappointed to notice that the i n i t i a l position had again been reached. This was repealed several times — I , p l a y i n g White, achieved my goal, while my partners with the same colour could in no way force Black to mate them. In the end we made a bet, and my colleagues had independently to disclose the secret of the mysterious queen and bishop manoeuvres. The students llyashenko and Petri (now well-known mathematicians) compiled a table of corresponding squares between the white pieces and Ihe black bishop, and soon (after a l l , they were good mathematicians!} the exact law was establ ished! It [urns out that, however the white queen moves about along the eighth r a n k , nothing w i l l force Black, who chooses the corresponding squares for his bishop, to mate the white k i n g ! Black must merely observe the following r u l e . With the queen at c8 he must keep between the bishops a distance of one square along the diagonal (as in the i n i t i a l position}, with the queen at d8 — two squares, with the queen at e8 — three squares, with the queen at f8 — four squares, and f i n a l l y , with the queen at h8 the black bishop must stand next to White's. Essentially we have here as many as f i v e amazing oppositions! Here is a possible course of events: 1 Bg2 Be4! (with the queen at c8 — a distance of one square) 2 Bf3 Bd5 3 Qe8 Bb7! (queen at e8 — three squares) 4 QdB Bc6 (two squares) 5 Be4 Bb7 6 Bg2 (not possible is 6 Qf8 Bxe4!, and this is not mate, since there is the move 7 Qf3; in general, the queen can stand to the r i g h t of its bishop only at h8) 6...Bd5 7 Qf8 Bb7! (four squares) 8 Qe8 (again the bishop cannot move due to its capture) 8...Bc6 9 Qh8 Bf3! (next to the white bishop) 10 Qc8 Be4 11 QdS Bd5 etc. Thus my colleagues, having shown that the probfem could not be solved, almost won their bet. Almost, because White has at his disposal another p l a n , involving the retreal of his queen from the eighth r a n k . After 1 Qh8 Bc6 2 Q a i l Bb7 3 Bc6 Black, so as not to give mate, is forced to move his bishop from bB. After this White captures f i r s t one bishop — 4 Bxb7+, within a few moves the second, then the a7 pawn, then he gives up his own Bishop (he no longer needs i t ) , and queens his pawn. Finally, using the two queens he constructs the following position.

1st Letter -

The Geometry of (he Chess Board

13

There now follows Qa2-g2+!, and Black is forced to give mate -h3xg2 mate! This example illustrates an important distinction (in the type of thinking) between mathematics and chess: the solving of a mathematical problem demands exact and deep analysis, concentrated in a narrow field (the student mathematicians accomplished it successfully), but in the game of chess Ihe number of variations is immensurably large, to take them all into account is impossible, and solutions (moves) are sought only approximately. An exhaustive analysis of the position, as we see, can be refuted in a completely unexpected and illogical way!

In [his position, by Berger, the white queen again has to battle ogainsl two bishops, but this time with a more usual aim — that of mating the black k i n g . An analysis of the correspondence between the queen and the bishops shows that it can be achieved only by very subtle play, and not earlier than the )3th move (against accurate defence by Black): 1 Qb8! (not allowing the black-squared bishop to escape; other continuations lead to a draw) 1...BC4 2 Qe5 Ba6 3 Qe1 Bb5 4 Qc1 Bfi 5 Qf4 Ba6 6 QgA Bb7 7 Qdi Be4 B Kg3 Bf3 9 Qc1 Bh5 10 Qai Bg4 11 QhB+ Bn3 12 Qxh3+ Bh2 13 Qxh2 mate. We have taken this problem from a book by the well-known Polish mathematician and popu I ar izer of science, G. Steinhausa, Mathematical Kaleidoscope. It thus links two 'kaleidoscopes' — mathematical and chess 1 We have examined with you only positions, the analysis of which resembles to some extent the solving of a mathematical problem. We have not considered problems and studies in which the solution is Burely chess one, but contains various geometric motifs: the opening and crossing of lines (blocking, interference), the graphical ly precise co-ordination of pieces (systematic movement) or their geometric movement on the board ( i n a triangle, square, rhombus or c i r c l e ) . For long time the study of geometric ideas has attracted the attention of chess composers, and in almost every col lection of problems or studie;, one can find a section devoted to 'chess geometry'. We should add that geometric ideas also occur often in practice. Take, for instance,

Seven Chess Letters (f6 f7+!

-

-

cf

-

the

Second Letter SYMMETRY IN CHESS

Symmetry, as a general principle of harmony in molecules, crystals, and l i v i n g nature, is of deep significance. The study of Us manifestations and regularities plays an important role in mathematics, physics, chemistry and biology. "Symmetry, however broad or- narrow our understanding of this word, Is an idea by means of which man has for centuries attempted to explain and create order, beauty and perfection", thus wrote the wel Iknown German mathematician Herman Weil. In our everyday life we constantly encounter various motifs of symmetry. Ornaments, mosaics and decorative designs delight our eyes by the symmetrical arrangement of their patterns. In many engravings by the oustanding Dutch a r t i s t Escher (including some with a chess theme!), symmetry predominates. Even the school geometry course is based now on the concept of symmetry (which is regarded as a specific case of spatial transforma t i o n ) , and, it has to be said, for many parents the school syllabus has become very d i f f i c u l t . . . . Various symmetry motifs are also encountered on the chess b o a r d . On the one hand, one can talk about natural symmetry, i.e. that arising during the course of a game, and, on the other hand, symmetry used in chess problems and studies. Our account is not a fundamental study of the given question, especially since it does not give any recipes for p l a y i n g or for the solving of problems. The aim has been merely to acquaint the reader with some unusual chess games and positions, the basic feature of which is symmetry. It w i l l be remembered that Symmetries of various types occur, the most common of which are a x i a l and c e n t r a l . In axial symmetry on the chess board, the axis is the line separating the left and r i g h t sides of the board (the boundary between Ihe d - and e - f i l e s ) , or the lower and upper parts (the boundary between the fourth and f i f t h r a n k s ) . If, say, a white knight stands at c2, and a black knight at c7, we say that these knights are arranged symmetrically (evidently, with a x i a l symmetry the corresponding squares are of opposite colours). With central symmetry on the board, the centre of symmetry Is Ihe poinl at which the four central squares touch — d4, dS, e4 and e5. In this case knights at c2 and f7 are symmetrical (the colours of the square? coincide). Subsequently we will not soecify each time the type of 15

16

Seven Chess Letters

symmetry in question, as this w i l l be clear from the context. We w i l l begin w i t h the fact that the i n i t i a l arrangement of the chess pieces possesses symmetry. In s h a i r a n j , also symmetrical were the ancient opening tabiya fIhe positions from which play began), for example:

Now imagine that a game begins, and Black copies White's moves exactly, wishing to retain for as long as possible the symmetry on the board. Of course, this method of play does not lead to any good. Exploiting his opponent's 'adherence to p r i n c i p l e s ' , White can mate the enemy king in only four moves, In one of two ways: 1 c4 c5 2 Qa4 Qa5 3 Qc6 Qc3 4 OxcS mate; 1 d4 d5 2 Qd3 Qd6 3 0h3 Qh6 4 Qxc8 mate. There is the following amusing story. Someone appeared at a chess club and announced that he had found a sure way of not losing as Black. "How?", they asked him. "Very simple", the guest replied, "by repeating the opponent's moves!" The naive inventor was challenged to a game by Sam Loyd, who mated him in four moves. Thus, with symmetrical play the white queen can mate the black k i n g as early as the fourth move. It takes a few more moves for a rook, bishop, knight and pawn to mate: 1 h4 h5 2 g i g5 3 Bg2 Bg? 4 Nh3 Nh6 5 hxg5 hxg4 & gxh6 gxh3 7 hxg7 hxg2 8 Rxh8 mate; 1 el e5 2 KeZ Ke7 3 Kf3 Kf6 4 Kg3 Kg6 5 Be2 Be7 6 Bf3 Bf6 7 d3 d6 8 Bh5 mate; 1 g3 g6 2 Nc3 Nc6 3 e3 e6 4 Nge2 Nge7 5 Ne4 Ne5 6 Nf6 mate; I g4 gS 2 h4 h5 3 Nf3 Nf6 4 Ne5 Ne4 5 hxg5 hxg4 6 g6 g3 7 gxf7 mate. F i n a l l y , on the ninth move the white king can announce mate: 1 d3 d6 2 Kd2 Kd7 3 Kc3 Kc6 4 Kb3 Kb6 5 Ka3 Ka6 6 Be3 Be6 7 Bb6 8b3 8 axb3 axb6 9 Kb4 mate. In his book 13 Children of Kaissa, I . Kreichik called one of his ' c h i l d r e n ' "When two do one and the same". In this humoresque he gives three symmetrical games with an identical finish — the black king gets mated. Here is one of them, in which, in Kreichik's opinion, the Queen's Gambit for Black is refuted: 1 d4 d5 2 Nf3 Nf6 3 c4 c5 4 Bg5 Bg4 5 e3 e6 6 Nc3 Nc6 7 BeZ Be7 8 0-0 0-0 9 Bxf6 Bxf3 10 Bxg7 Bxg2 11 BxfB Bxfi 12 Bxe7 Bxe2 13 Bxd8 Bxdi 14 cxd5 cxd4 15 dxc6 dxc3 16 cxb7 cxb2 17 bxa8=Q bxa1=Q 18 Bf6 mate. In two other games the Spanish and I t a l i a n Games are refuted in similar fashion. In the four-move miniatures given above, White's play was based exclusively on his opponent's stuDborness. For example, Loyd, taking account of this factor, was not a f r a i d to place his own queen en prise. At the same time, in the games suggested by Kreichik, the play proceeds quite sensibly. But symmetrical games can be found not only in humorous stories; they also occur in serious tournaments. One of the most notable oxatnptos of this type Is a game played at the beginning of this

century.

2nd Leiler — Symmetry

in Chess

17

Rot I e v i -EI y ashov Four Knights' Game 1 e 4 e5 7 Bxc6 Bxt3 U Bxf6 Bxf3

2 Nf3 Nf6 3 Nc3 Nc6 4 Bb5 Bb4 5 O-CT O-O f> d3 d6 8 Bxb7 Bxb2 9 Bxa8 Bxal 10 Bg5 Bg4 11 Qxal QxaB 13 Bxg7 Bxg2 U Bxf8 Bxf1 15 Qxfl Qxf8 16 0g2+ Qg?.

Here, evidently a f r a i d of seriously disturbing the symmetry, the players agreed a draw. I I is interesting that, several years later, the position after Black's 12th move was repeated in a game Traxler-Samanek. On this occasion Black did not succeed in demonstrating that the opponent's moves could be copied indefinitely. Instead of 13 Bxg7 there followed 13 Bxe5! Bxe4 14 Bxg7 Bxg2 15 Bxf8, and after 15...Bxfl 16 Qg7 mate Black was unable to repeat his opponent's move. In the following game, which was played in our time, the symmetry was maintained for a f u 11 19 moves (apart from one 'hiccup' between ihe 7th and 9th moves) — a record In its own way! Stolyar-Shukshta English Opening I c4 c5 2 g3 g6 3 Bg2 Bg7 4 Nc3 Nc6 5 a3 a6 6 Rbi Rb8 7 b4 cxb4 8 axb4 b5 9 cxb5 axb5 10 Nh3 Nh6 11 0-0 0-0 12 d4 dS 13 Bxh6 Bxh3 H Bxg7 BKg2 15 Bxf8 Bxfl 16 Bxe7 Bxe2 17 BxdB Bxdi 18 Bc7 Bc2 19 Bb2 Rb7.

20 Be5 Nxe5. Unfortunately, the symmetrical 20.. .Be4 21 Nxe4 Nxe5 f a i l s to 22 Nf6+ and 23 dxe5. After 21 dxe5 d4 22 Rxc2 Rc7 23 Kf1 3 5 24 Ke2 dxc3 25 Kd3 Rc4 26 Rxc3 R«b4 27 Rc7 the players agreed a draw. The reader may have gained the erroneous Impression that, by copying While's move, Black can at best draw. But as the following short game shows, by copying exactly his opponent's moves, Black ha^. the chance as early as the eighth move... to mate the white k i n g !

Seven Chess Letters

18 1 e4 <

2 Ke2 Ke7

3 Ke3 Ke6

4 Qf3 Qf6

5 Ne2 Ne7

6 b3 b6

7 Ba3 Ba6.

8 Nd4+, and Black is forced to give mate:

8.. .exd4 male!

Thus by copying White's moves, Black can both be mated, and can male the enemy k i n g . But, clearly, in both the one case and the other, the final position can in no way be symmetric. l l is curious, however, that with stalemate the final arrangement of [he pieces can perfectly well be symmetric, I.e. the stalemate is mutual! As an example, we give the following record game, which lasts only 19 moves. In i t , it is true, it is sometimes Black, and sometimes White who repeals moves, but the main thing is that, in the symmetrical final position, neither the one side nor the other has any moves. 1 e4 d5 2 e5 d4 3 c3 f6 4 Qf3 Kf7 S Qxb7 Qd5 6 Kdi Qxo.2 7 Kc2 Qxfi B Qxc8 Qxgi 9 Qxb8 RxbB 10 Rxg1 Rb3 11 Rg6 Ra3 12 Rh6 9xh6 13 bxa3 Kg7 14 Kb2 d3 15 e6 a5 16 h4 aU 17 h5"c5 18 f4 c4 19 f5. Stalen.ate to White and Black {the position on the board is centrally symmetric).

In the given example both kings are symmetrically stalemated, but five pieces on each side are exchanged. The following game is unique, since, f i r s t l y , only one knight on each side disappears from the board, secondly, both kings are symmetrically stalemated, and t h i r d l y , and this is the most amazing, all of White's and Black's moves are symmetric ! 1 Nf3 Nc6 2 Nc3 Nf6 3 Nb5 Ng4 4 h3 a6 5 Na7 Nh2 6 Nxh2 Nxa7 (the f i r s t and last exchanges) 7 g4 b5 8 Bg2 Bb7 9 eh d5 10 Ke2 Kd7! 11 Qg1 Qb8! (at the start of play the i n i t i a l positions of Ihe kings and queens were clearly not centrally symmetric, but now complete order has been establ ished on the board) 12 b4 g5 13 Bb2

Bg7 U Rfi RcB 15 Bd4 Be5 16 f3 c6 17 Bf2 Bc7 18 Bel Bd6 19 Kf2 Kc7 20 a4 h5 21 s5 h4 22 c* f5 23 c5 f* 24 e5 d4 25 e6 d3. Stalemate to both kings!

2nd Letter — Symmetry

in Chess

We recall here an old chess anecdote, which shows thai copying moves can sometimes prove highly beneficial. Once during an inter-national tournament, a stranger approached Emanuel Lasker with a proposal that he should play by correspondence against his talented son. The request was supported by a substantial material reward, and Lasker agreed. In the event of him winning he would receive 500 dollars, while if the lad should by some miracle win — which, of course, the father was in no way expecting, then the grandmaster would have to fork out 1500 d o l l a r s . In the end the World Champion succeeded in winning, although only with enormous d i f f i c u l t y . H turned out that Lasker had been playing correspondence with Capablanca, who had agreed the same conditions with the stranger, while the son was merely a hoax. When Lasker, playing White, made his f i r s t move, it was repeated by the trickster in the game with Capablanca. His reply, in turn, was reproduced in the f i r s t game for Black, and Lasker's reply — in the second game for White, and so on. Clearly, by such symmetrical strategy the 'father 1 guaranteed himself a drawn result in the match with the grandmasters, and hence, a net p r o f i t to the tune of one thousand dollars. Alekhine figures in other versions of this anecdole. He, allegedly, agreed to play for a big stake with two unknown persons, who had devised a similar intrigue with the repetition of moves, forcing Alekhine to play against himself. The situation appeared hopeless, but the World Champion thought of a way to punish the swindlers. In one of the games he deliberately made a bad blunder. The partners decided not to repeat on the second board the obviously losing move of the grandmaster, and attempted to win both games. But this renouncing of the safe tactics had fatal consequences — Aiekhine succeeded in mating both kings! Uf) t i l l now we have been talking mainly about one player copying the moves of the other, sitting opposite. But what about repeating the moves or even entire variations which have occurred in the games of quite different players? Of course, if these moves are good, no one w i l l decline to copy them. But it can happen that the deliberate repetition of unsuccessful variations can have a certain sense. We w i l l describe an amusing instance which occurred in the game Taimanov -Si mag in twenty years ago in the 1961 USSR Championship. Black chose the King's Indian Defence, which he had previously never employed. A few days earlier a fresh issue of Shakhmaty v SSSR had appeared. In which was given a detailed analysis of the Talmanov-GIIgoric game from an inlernational tournament in Argentina. It has to be assumed that Simagin knew this game, which had been convincingly won by White, and nevertheless he copied Giigoric's moves exactly. This put the Leningrad grandmaster on his guard. He realized lhat his opponent would hardly be suicidal enough to repeat

20

Seven Chess Letters

the game to the end, and this meant that he had certainly prepared an opening surprise. Si mag in made his moves {belonging to Gl igoric I) easily, whereas Taimanov chose his moves with agonizing care. The whole paradox was that Si magi n, who was p l a y i n g the variation for the f i r s t time, replied instantly, whereas Taimanov, who was an expert on i t , played slowly, constantly checking and controlling himself. By the twentieth move Taimanov's nerves were stretched to the limit, but the game was s t i l l an extact repetition of his game with GHgorlc. His patience had dried up. And suddenly something completely unexpected happened — Simagin sank into deep though!, and d i d n ' t reply for nearly an hour! The secret was revealed — Taimanov's opponent had not prepared any improvement, It was pure b l u f f ! The grandmaster had counted on disturbing his opponent with the threat of employing an innovation, and of forcing him to deviate from the familiar p a t h . But in this case there would have been no danger for Black! Although in the battle of character White had held out, his opponent's psychological attack bore its f r u i t s — when the war of nerves came to an end, Taimanov's strength was on the wane, and he promptly made a mistake, and then a second, completely handing his opponent the i n i t i a t i v e . But the curious thing is that Simagin declined to exploit his winning chances, since he too was exhausted by the psychological duel. On concluding peace, both players were happy that the game had ended in a draw, that in general it had ended... . The symmetry described up t i l l now has mainly related to the i n i t i a l position, the start of play, and unusual games. But if one talks about chess composition, here symmetry plays a special role. Thanks to its original ideas, problems and studies acquire an a d d i tional aesthetic value. Consider [he following classic miniature, which is known throughout the w o r l d . R. Bfanchelti,

White to p l a y

1925

-••••-.

and w i n .

After the introductory 1 Bb2! e l l the pieces on the board are a r r a n g e d along the long d i a g o n a l . Black is threatened with the loss of his rook, a n d , depending on where i t moves, two elegant symmetric variations arise. 1 . . . Rf8 < 1 — R f 7 2 Rh3+ Kg8 2 Rh8 mate) 2 Rc7+ KgB 3 Rg7+ Kh8 4 Ka2! (but not 4 Kb1 because of A...Rf1+ 5 Ka2 Ra1+ 6 Kb3 Ra3+ 7 Kc2 Rc3+ 8 Sxc3 stalemate), and White wins the rook. Simila r l y , 1 . . . Rh6 2 Rg3+ Kh7 3 Rg7+ Kh8 4 Kbi ! U Ka2 Ra6+ e t c . ) , w i t h the same f i n i s h . The p o i n t

of a problem

or

study

is

often

provided

by

a

slight

2nd Letter — Symmetry

in Chess

21

disturbance of the symmetry in the i n i t i a l position, which is uti I ized significantly in the solution. In chess composition such a device beers the name of asymmetry. Here is one classic example on this theme. V. Pauli, 1920

Mate in 4 moves. At f i r s t sight the position appears completely symmetric (relative to the d - f i l e ) , but t h i s is not quite so — the h - f i l e slightly breaks the ideal symmetry, and this is reflected in the solution. 1 Bg7! Ba5(c7) 2 Bf8 BdB 3 Bd6!, with inevitable mate - 4 Nc3 or 4 Ne3 if one of the black knights moves; 4 Nb6 or 4 Nf6 if the black bishop moves. At the same time (and this is important for the problem! ) the symmetric continuation 1 Ba7 does not work, since the black bishop makes use of the h - f i l e : l...Bh4! 2 BbS Bf2! 3 Bd6 Bd4, and there is no mate. Lot us now examine a miniature study on the same theme, T. Dawson, 1924

White to play

and win.

White has two logical continuations — 1 Nd4+ and 1 Nf4+. The a - f i l e , which introduces an asymmetry into the position, appears to play no role. In other words, if the knight check at d4 wins, the goal should also be achieved by the symmetric check at f4. But only one of these moves leads to a win. 1 NdA+! K«3 2 Nxf3 Kxf3 3 Kf8! d2 4 eB=Q <J1=Q 5 Qh5+, and it is all over. If Black does not queen his pawn, then it is well 1 known that the ending 'queen against central pawn Is easily won. The move 1 Nf4+ leads to a draw - 1...Ke3 2 Nxd3 Kxd3 3 Kd8 f2 4 e8=Q Kd2! The difference is that the ending 'queen against bishop's pawn' is drawn (provided, of course, that the white king does not stand close to the pawn): 5 QbS Kei 6 Qb1+ Ke2 7 Qe4+ Kf1 8 Ke7

22

Seven Chess Letters

Kgl 9 Qg4+ KM 10 Gf3+ Kgl 11 Qg3+ Khl ! 12 Qxf2 stalemate. It is curious that, if the a - f i l e is detached from the board, there is no win in either v a r i a t i o n , because the d-pawn is transformed from a central pawn into a bishop's pawn! The number of different examples, i l l u s t r a t i n g the idjas of symmetry and asymmetry in chess composition, can readily be increased ( c f . , for example, problems 22 and 23 In the fourth ' l e t t e r ' ) . The following incident took place a few years ago during the Championship of the country in Leningrad. One of the authors of the book (at the time the other was, as usual, playing in an international tournament) was s i t t i n g with David Bronstein in the auditorium and conversing q u i e t l y . ( I f , of course, one could call it a conversation — Bronstein was sharing with me his o r i g i n a l and clever ideas regarding chess in the t h i r d millenium, and I admiringly kept s i l e n t ) . The round was only just s t a r t i n g , and the demonstration boards had not yet drawn our attention. The grandmaster took out his pocket set and arranged on the board two kings and six pawns.

"How do you evaluate this position?", he asked intriguingly. "Most probably a draw, whoever it is to move", I thought. But then what was the trick in the position? "Most probably, a win for the side whose move it is?" But this reply too seemed inconvincing, since why in this symmetric position, in which none of the six pawns has moved, and the kings have not left the back rank, should the turn to move play a decisive role? "That means, whoever begins, loses?" But this conclusion, for the same reasons, appeared even more strange. Of course, the most unusual and unexpected things happen on the chess board. We constantly encounter positions of zugzwang and mutual zugzwang. It is d i f f i c u l t to surprise us with a n y t h i n g . But the position suggested by Bronstein, it was a special case. It was engagingly simple, and, in addition, was notable for a kind of compiete, in a certain sense, ideal symmetry. How many different associations rushed through my head, while my gaze was f i x e d on this landscape with its two kings and six pawns! There was plenty to think about — the mysterious and inexhaustible nature of chess, its severe beauty, paradoxical ity and depth! ( I f you feel the desire to analyze also tackle a second position,

the given

position,

you

should

2nd Letter — Symmetry in Chess

23

which is a k i n d of antipode of the f i r s t — the pawns have made a l l of f o u r steps, but the e v a l u a t i o n s of the positions — w i t h the same t u r n to move — are o p p o s i t e ) . I d i d not succeed in coming to any c o n c l u s i o n . Finally I raised my head a n d looked i m p l o r i n g l y at the g r a n d m a s t e r . But Brans t e i n gave me a c r a f t y smile, and instead of dispel I Ing my doubts, merely whisper-so1 q u i e t l y ; "You should w r i t e about I t " , and Imperceptibly left the h a l l . So the g r a n d m a s t e r ' s w i s h had to be f u l f i l l e d , and t h i s account had to be w r i t t e n about symmetry on the chess b o a r d . P.S. If we should have succeeded in i n t r i g u i n g some of our readers w i t h these last two p o s i t i o n s , it would be u n j u s t on our p a r t to leave them without a s o l u t i o n . . . In the f i r s t p o s i t i o n , the side t h a t moves f i r s t , w i n s . Here are the m a i n v a r i a t i o n s : 1 Ke2 ( w i t h the aim of stopping the b l a c k pawns) l . . . K d 7 (or I . . . h 5 2 Kf3 g5 3 a4 h4 4 Kg4 f5+ 5 Kh3 Kd7 6 cA Kc6 7 a5 w i t h a d e c i s i v e a d v a n t a g e f o r White) 2 Kf3 Kc6 3 a i h5 4 c4 fS (each side is now t h r e a t e n i n g to advance h i s r o o k ' s pawn) 5 Kg3 Kb6 6 b4 g5 (or 6 . . . 9 6 ) 7 a5+ Ka6 8 cS h4+ ( 8 . . . K b S 9 Kg2! or 8...Kb7 9 b5) 9 Kh3 f 4 ( 9 . . . K b 5 10 Kh2 g4 11 Kg2 f 4 12 Kgi ) 10 c6 f3 11 b5+ Ka7 12 b6+ Kb8 13 a6 g4+ 14 Kh2 g3+ 15 K g ! h3 16 a7+ Ka8 17 c7, and White is the f i r s t to queen a p a w n . During the course of the s o l u t i o n several symmetric positions have a g a i n been reached, and only on the 19th move ( ! ) a f t e r 1 7 . . . h 2 + - IB KM f2 19 c8-Q+ is Black u n a b l e to copy White's move. In the second p o s i t i o n the side which b e g i n s , loses. After 1 cS fh 2 b5 Kb8! White is in zugzwang on both w i n g s , for e x a m p l e : 3 a6 Ka7 4 c6 Kb6, and the white pawns are b l o c k a d e d , whereas the black pawns break t h r o u g h : 5 KhZ f3 6 Kg1 h 3 ! 7 Kf2 h 2 . The r e p l y 1 . . . f 4 also follows after 1 KhZ or 1 b 5 . On 1 Kgl there can follow 1 . . . K a 6 2 Kg2 f i 3 c5 Kb5 4 Kgi g d , when Black w i n s .

Third Letter AN ANCIENT MATE

Slowly rotating our 'kaleidoscope', you can observe in it the various colours of the chess spectrum: topics from chess history, psychology, composition and geometry. Considerable attention has also been devoted to pure chess. In the last ' l e t t e r ' of the book, lo say nothing of its t h i r d p a r t , w i l l be given games (or extracts from them) belonging — without exception — to World Champions! Although the works created by well-known grandmasters are interesting and Instructive from various points of view, this by no means signifies that they contain multi-move combinations with cascades of sacrifices and mate to the enemy k i n g in the centre of the board. Sometimes events develop less v i o l e n t l y . Nevertheless, it has to be admitted that tt is to the s t r i k i n g and spectacular combinations which develop on the 64 squares that chess is chiefly indebted for its p o p u l a r i t y . The a r t i s t i c and aesthetic value of combinations, their emotional effect on chess players, f i r s t and foremost raises the game to the level of a genuine a r t . And hence it is not s u r p r i s i n g that one of our ' l e t t e r s ' we have devoted entirely to combinations. The number of clever, spectacular and deep combinations on the chess board is extremely great, and it would be impossible to describe them a l l at once. Our choice has f a l l e n on one of the oldest combinations, known by the name of 'smothered mate'. This mate, which is given by a knight, when a l l the enemy k i n g ' s escape squares are occupied by his own pieces and pawns, was described back in the 15th century in the treatise by Juan Lucena — the f i r s t printed guide to chess. Here is this famous position.

2A

3rd Letter - An Ancient Male

25

White sacrifices his queen and announces a smothered male in five moves: 1 Qe6+ Kh8 2 Nf7+ KgB 3 Nh6»+ Kh8 4 QgB+M Rxg8 5 Nf7 mate. The king perishes encircled by his own troops, and the deciding blow is struck by the crafty knight, for which there are no impregnable fortresses. Note that in Lucena's position there is a d u a l : 3 Nd8+ Kh8 and 4 Qe8+, but this is, of course, a t r i f l e — five centuries ago the word ' d u a l ' was not yet even k n o w n . . . . The following position is by Phil I ip Stamma, and belongs already 10 the mid-IBth century.

1 Be4+- Rb7 2 0b8+! Rxb8 3 Rxa7+! Bxa7 4 Nc7 mate, and the black king is in the trap. On this occasion White sacrifices not only his queen, but also a rook. Of course, for the 250 years separating these two positions, the progress, to be frank, is not very great — in those distant times chess art developed at a slow rate. It is curious that it was only a l the end of the last century that the orientalist and chesa historian H.J.R. Murray established the Arabic o r i g i n of the smothered mate. In his History of Chess he cites the Persian mansubat of the 9th century, in which a white knight mates the black k i n g , hemmed in by his own pieces. Does this ancient combination occur In tournament games? Yes, strangely enough, rather often. The record belongs to grandmaster Paul Keres, who once announced a smothered mate... as early as the sixth move! What is amusing is that this curious instance occurred not in a simultaneous display, but in an encounter with a master. Keres-Arlamowski Caro-Kann Defence 1 eft c6 2 Nc3 d5 3 Nf3 dxe4 4 Nxe4 kd7. It is d i f f i c u l t believe that this game w i l l last only two more incomplete moves.

to

5 Qe2 Ngf6?! (the exclamation mark is for the creation of a unique 'helpmate') 6 Nd6 mate.

26

5even Chess Letters

A classic example of a smothered mate belongs to the genius of combinations from the 19th century, Paul Morphy. .. Morphy-Brain

All is ready for the final scene; it only remains to divert the black queen away from the square from which White will give mate. 1 e5! Qg5

2

h4! Qg4

3 Qa3+ Kg8

4 Ne7+ Kf8

5 Ng6++

Kg8

6 Qf8+ Rxf8 7 Ne7 mate. The concluding position deserves a diagram (as, however, does every position of this type).

Here Is another American player.

episode

from

one of

the

games of

the

great

Morphy-N.N.

In this case things proceed wi rhout any preparatory operations: 1 Nc5+ Kb8 2 Nd7+ KcB 3 Nb6++ Kb8 4 Qc8+ Rxc8 5 Nd7 mate. A reflection of this unfading combination can sometimes also be seen In the games of the outstanding players of our time. The following position was reached In a game from one of the 1976 Interzonal Tournaments.

3rd Lelter - An Ancieni Mate

27

Tal-Portisch

\

The Ex-World Champion has been cleverly lands the opponent a spectacular and decisive resigned immediately, since on 1...Re7 there 3 Ngf7 mate, while Black has no other defence mate by 2 Qg8+ RxgB 3 Nf7 mate.

attacking, and he now blow — 1 Nh6! Black follows 2 Qxe7 Bxe7 against the smothered

One of the authors of this book once attempted to give a smothered mate in a tournament (simultaneous displays do not count, of course; in them this ancient combination is eternally young). The game in question was that against Vladimir Tukmakov in the 1973 USSR Championship. Black's queen occupied b6, his knight — g4, and there followed 1...Nf2+. Unfortunately, White d i d not go in for the variation 2 Kgl Nh3++ 3 Khl Qg! + 4 Rxgi NfZ mate, but gave up the exchange by 2 Rxf2> and after wild complications the game ended In a draw. The other author has been more fortunate. In the University Championship he once succeeded In announcing a smothered mate to the President of the Moscow University Chess Club. BairamovGik

The white knight, heading from f2 to g3, was in transit on the edge of the board. But to complete its journey it did not have time — l . . . R c 2 ! 2 Bxc2 (or 2 Qe3 Ne2+ 3 Kf2 Ng4+) 2...Ne2 mate! Here Black managed without a queen sacrifice, but used a different combinational device — the p i n . In the following episode too (from a game played in 1931) the queen Is given a modest role — that of forcing Black to surround his own king with pieces. Alekh i ne-Lugovsky (see next diagram) Kg8

I

1 Ne6-H Mxe6 (on the capture by the bishop there fol lows 2 Qe7+ 3 Qe8 mate) 2 Qe7+ Kg8 3 Qe8+ Nf8 4 Ne7 mate. A smothered mate is always beautiful, especially if it is accom-

28

Seven Chess Letters

ponied by ihe sacrifice of Ihe queen. However, it Is not essential that the queen should be given up at g8. This operation can also be carried oul on a neighbouring square, as occurred, for example, in a game played in 1967 al an internalional tournament in Copenhagen. Taimanov-Jakob sen

;

1 0g7+! Nxg7 2 Nh6 mare. As we see, variations on the theme of smothered mate {or else the mate itself!) occur in events of Ihe highest rank, and not only in games between inexperienced players. It 's amusing that in 1970 at the World Chess Olympiad in Siegen this ancient combination was carried out in two games. Here are ihe conclusions of these miniatures. Unzicker-Sarapu

Black has sacrificed a piece, but is placing his hopes on the p i n . But White succeeds in reinforcing the Knight: 1 Bf4! Nxf4 2 Qxf7+ Kh8 3 Qg8+ RxgB 4 Nf7 mate. Augous t i -Uzman Ruy Lopez Nc6

The game is so short that we tan give it in f u l l : 1 e4 e5 2 Nf3 3 BbS a6 4 Ba4 Nf6 5 0-0 b5 6 Bb3 Bb7 7 Rei Bc5 8 c3

3rd Letter - An Ancient Mate

29

exd4 10 cxd4? (10 h3 should have been played) 1O...Nxd4! Qh4! 12 Nf3 Qxf2+ 13 Kh1 Q91+ 14 Rxgi Nf2 mate. Thus, p r a c t i c a l players are always ready at a convenient moment to give up their queen and announce mate w i t h a k n i g h t . B^t what of chess composers? 11 stands to reason that they c e r t a i n l y could not f a i l to consider such a f a s c i n a t i n g theme in t h e i r work. We give a few which the ancient smothered mate combination is expressed eK ample5 in in paradoxical form. Up t i l l now the decisive role has always belonged to ihe k n i g h t , although we couldn 1 1 have managed without the queen. The idea of replacing the queen w i t h a weaker piece — a bishop, could have been devised only by such an inventive composer as Loyd.

S. Loyd, 1B58

Mate in 6 moves. 1 Bh2! (with the threat of 2 Bg1 + ) 1 . . .Ka7 2 b6+! KaB (2...Xa6 3 Bgi Rc6
1932

Mate in 12 moves. In this fantastic picture Black has a ful I set of pieces, but this does not save h i m : 1 Nf4+ Kc5 2 Ne6+ Kd5 3 Nxc7+ Kc5 A Nxa6+ KdS 5 Nc7+ Kc5 6 Net* Kd5 7 Nf4+ Kc5 8 Ke4 d5+ 9 Ke5 Bf6+ 10 Ke6 NdB+ 11 Kd7 and 12 Nxd3 mate.

30

Seven Chess Letters O. Blathy,

White to play

1922

and w i n .

Another chess monster devised by Blalhy. Again, as on a parade, Black is represented by his entire army, though it is true that on this occasion his forces are less mobile. But where is the white knight? It is some time before it appears on the board. 1 Kxe1 Qa1 2 h3! (the pawn is in no h u r r y ; the subtlety is revealed later! 2...Qa2 (Black has plenty of pieces, but only his queen is able to move) 3 h4 Qai 4 h5 QaZ 5 h6 Qai 6 h7 Q.a2 7 h8=N! (at last) 7...Qa1 8 Nf7 Qa2 9 Ng5 Qa1 10 Ne6 Qa2 11 Nxc5 Qai 12 Nb7 Qa2 13 Nd6 Qai 14 Nxc4 Qa2 15 Na5 Qa1 16 Nxb3 mate! ( i f on the second move White had immediately advanced his pawn two squares, now, after any attack by the newly-Oorn knight on the b3 rook, the black queen would have defended i t , and the result would have been a d r a w ) . We now return

to the world of realistic

chess.

E. Pogosyants, 1979

White to play and w i n . The position could well have arisen in maslers. This study is interesting for the fact in a classic smothered mate, at the moment White black king would appear to have no intention of The mating construction arises in the course play.

a game between two [hat, although it ends has no queen, and the going into the corner. of subtle preliminary

1 c7 0-0! 2 Nc6! Now the threat is 3 Nd8, and Black cannot play either 2,..RaB 3 NbB, or 2...Rc8 3 Me7+ Kf7 k NxcG e3 5 Ne7 e2 6 c8=0 e1=0 7 Qc4+! Kf6 (7...Kf8 B Qg8 mate) 8 Qf4 mate. Thus the white pawn queens, but the black pawn also succeeds in promoting simultaneously. 2...e3 3 Nd8 e2 4 c8eQ e1=Q. Only now do we see the familiar- p i c t u r e . 5 Qc4+ Kh8 6 Nf7+ Kg8 7 Nh6++ Kh8 8 Qg8+ Rxgfl 9 Nf7 mate.

31

3rd Letter — An Ancient Mate

The following study is perhaps the most famous on the given theme. I " ' n e initial position, where the pieces of both sides are scattered across the board, i t can hardly be conceived thai lying in store for the black king is a smothered mate. The attack begins with a quiet queen move, which allows Black to give check. A. Seletsky,

1933

White to play and w i n . 1 Qg5! Ke6+ (the pawn cannot be captured by 1...Bxd7 because of 2 Nf4!, wilh the irresistible threat of 3 Bh5+, while at the same lime 2 dS-0 was threatened) 2 K g l ! Kxd7 (2...Bxd7 3 Bg4+ Kd6 4 Qc5+). Black has re-established material equality, but now the mating mechanism goes into operation. 3 Nc5+ KcB [other k i n g moves lead to the loss of the queen, for example 3...Kd6 4 Qg3+ Kd5 5 Bc4+! Kxc4 6 Qb3+ Kxc5 7 Qa3+ and White has a theoretically won position) 4 Ba6+ KbB 5 Qg3+ Ka8 6 Bb7+! Bxb7 7 Nd7!! QdB. The one way of defending the squares b6 and b9 against the mating threats. Now comes a spectacular blow, a f i t t i n g culrnination to this splendid work of a r t . 8 Qb8+1! Qxb8 9 Nb6 mate. And in conclusion, a 'historical' problem, in which it is demanded lhal White should win by a smothered mate (the existence of which, at f i r s t signt, appears impossible). K. Jaenisch, 1&49

Smothered mate in 10 moves. The great

Russian

chess

master

arid composer of

the

last

century

Carl Jaenisch called his problem 'The iron cage of Tamerlan' . This is how the black king falls into captivity: 1 f3+ gxf3 2 exd3+ cxd3 3 Bf5+ exf5 4 Re6+ dxeG 5 Rd4+ cxd4 6 a8»B+ Qd5 7 Bxd5+ exdS B Nf6+ gxf6 9 Qe5+ fxe5 10 Ng5 mate. The king Is in the cage!

32

Seven Chess Letters

Thus, by means of the ancient smolhcred male combination, did Jaenisch succeed in depicting Ihe cruelty and despotism of the Central Asian conqueror of the 14th century, Timur Tamerlan.

,-l.r..

Fourth Letter FORTY CHESS QUARTETS

The fewer the pieces on the board, the more eager we are to solve a problem or s t u d y . Especially popular among chess enthusiasts are miniatures — in which the number of pieces does not exceed seven. The number of miniatures created by chess composers is quite immense, and there are many thousands of compositions w i t h f i v e pieces or less. The minimum material necessary lo create a worthy work of chess art is four pieces (three-piece compositions w i l l be the subject of the following 'letter'). Problems and studies with this number of pieces we w i l l call quartets. Our aim is to acquaint you with a l l forms of quartets, and to represent each of them by some single p o s i t i o n . In the f i r s t instance we try to give a study or an orthodox problem ( i . e . a problem with a normal task —White to play and give mate in so many moves). But for c e r t a i n quartets we were unable to f i n d either a study or such a problem, and in this case we have offered a helpmate problem. In a 'helpmate' it is s t i l l White who gives mate, but Black does not hinder him, b u t , on the c o n t r a r y , does e v e r y t h i n g to assist in the mating of his own k i n g . It has to be said that by no means a l l the sets of f o u r pieces are of equal v a l u e , but we have adhered to our p r i n c i p l e , and i l l u s trate each of them with exactly one quartet ( t h e i r o v e r a l l number f a r exceeds a thousand). Let us f i r s t ascertain how many sets of four pieces there are altogether. Since two places must be reserved for the k i n g s , there are two different cases: (1) White and Black both have one piece, apart from the k i n g ; (2) White has a k i n g and two pieces, while Black has a lone k i n g , [n the f i r s t case, combining one of the f i v e white pieces (queen, rook, bishop, knight and pawn) w i t h one of the f i v e black pieces, we o b t a i n 25 possible combinations. Here sets which d i f f e r by the colour of the pieces are regarded as d i f f e r e n t . Indeed, 'white rook against black p a w n ' ( w i t h White to move) Is by no means the same as ' w h i t e pawn against black r o o k 1 . In the second case the number of sets is fewer by ten — 15. Here each p a i r of pieces (they are both white) obviously gives only one set. Thus 40 different quartets have found t h e i r way into our ' l e t t e r ' . If both kings are accompanied by one of their own pieces, both 33

34

Seven Chess Letters

problems and studies have been d e v i s e d , but if Black has only his k i n g it is u n d e r s t a n d a b l e that only problems can be expected. Of course, each reader can compile hfs own l i s t of quartets on the chess b o a r d , and it is q u i t e possible that it w i l l be no less i n t e r e s t i n g than the one suggested b y u s .

ONE WHITE AND ONE BLACK PIECE Pawn against pawn The most popular, of course, is the 'geometric' study examined by us in the f i r s t ' letter' together with other" pawn The given material is of a purely endgame nature, and it possible to give a whole series of beautiful studies. However, enough, clever problems with two pawns are also known, and one of them. No. 1

by Ret i, quartets. would be strangely No. 1 is

E. Pogosyants, 1970

Mate in 5 moves. t a7 Kh2 ( 1 . . . f 5 2 aB=Q+ and 3 Og2 mate) 2 a8=Rt! Kh3 3 Ra4! f5 4 Rf4! This is the point — with a queen at fit it would now be stalemate. 4...KhZ 5 Rh4 mate. Pawn againsl knigtit What can depend on one white pawn, opposing a black knight? Whether or not the knight can stop it w i l l , after a l l , be decided by the knight itself. But to advance at the right time can also have no small significance, as in No. 2. No. 2

F. Prokop, 1925

4th Letter — 40 Chess Quartets White to play

3S

and w i n .

Sooner or later the white k i n g must make way for the pawn, but excessive g a l l a n t r y would be inappropriate here, since after any k i n g move the knight catches the pawn; ! Ka8? Ne3! 2 b6 Nc4 3 b7 Nb6+ 1 Ka7? N f i ! 2 66 Ne6 3 b7 NdB 4 b8=Q Nc * ; 1 Ka6? a nd 4...Nd7; Nf4 2 b6 Ne6 3 67 Nc5->; f i n a l l y , 1 Kc8? He3 2 b6 Nc4 3 b7 Nd6+. Thus great restraint is demanded of the k i n g . 1 b6! Now the square thai White w i l l choose for his leader depends on the reaction of the k n i g h t . 1...Ne3 2 Ka6! Nd5 3 b7 Nc7+ 4 Ka5, or 1...W4 2 Kc8! Nd5 3 b7 Nb6+ KdB, and in both v a r i ations the pawn queens.

Pawn against

bishop

In problem No. 3, 1 fxg8=Q? leads to stalemate. The solution is 1 fxg8=R! (once again a rook appears on the board instead of a queen) l...Ka2 2 Rg3! Kai 3 Ra3 mate. No. 3

0.

Oehler,

1923

Mate in 3 moves. The point of the problem is that after 1 fB=Q? Ka2 black bishop comes to the aid of its k i n g — 2 . . . B b 3 ! , and mate.

Pawn against

over art.

2 Qb4 the there is no

rook

One of the few ratios of forces where one does not have to ponder the choice. Study No. 4 is part of the golden treasury of chess No. 4

G. Barbier,

F. Saavedra,

White to play

and w i n .

1895

36

Seven Chess Letters

Along w i t h R e t i ' s pawn s t u d y , t h i s i s the most famous quartet on the chess b o a r d . Its solution is f a m i l i a r lo a l l lovers of the game. 1 c7 Rd6+ (after 1...Rd2 2 c8=0 Rb2+ 3 Ka5 ihe white k i n g moves down on the a - and b - f i l e s ) 2 Kb5 (2 Kb7 Rd7 and the rook captures the pawn; 2 Kc5 s i m i l a r l y f a i l s to win after 2 . . .Rdi and 3...Rc1+) 2...Rd5+ 3 Kb4 ( p l a y i n g ihe k i n g to the c - f i l t a g a i n leads to a draw) 3...Rd4+ 4 Kb3 (4 Kc3 Rdl 5 Kc2 d i f f e r s from the main v a r i a t i o n only by a transposition of moves) 4 . . .Rd3+ 5 KcZ! Now the rook cannot attack the white k i n g from Ihe r e a r , and a l l would appear to be over. However, the most f a s c i n a t i n g events are s t i l l to come. 5...RdA!l An amazing resource for B l a c k ; on 6 c8=Q he s a c r i f i c e s his rook by 6 . . .Rc4+! 7 Qxc4, which leads to stalemate. But B l a c k ' s hopes are not destined to be r e a l i z e d . 6 c8=R!! Ra4 (the threat was 7 Ra8 mate) 7 Kb3! A Spectacular concluding s t r o k e ; Black loses h i s rook or is mated in one move. The four pieces have played out a genuine chess spectacle! The study in question is one of the earliest on the theme of 'under-promotion' . In the whole of chess l i t e r a t u r e you w i l l not f i n d such a clever and r i ;h f i n a l e w i t h such l i m i t e d m a t e r i a l . The history of how t h i s study o r i g i n a t e d is as f o l l o w s . In a game Fenton-Potter, p l a y e d in England in 1875, a position s i m i l a r to ours was reached; the white k i n g stood at c6, and his pawn at b6, while Black '£ k i n g was at h3 and his rook at a5. Here White agreed to a d r a w , since he had not noticed that after 1 b? Ra6+ 2 Kc5 his k i n g could move down the b o a r d , and the pawn would queen. Shortly a f t e r , Z u k e r t o r l , one of the strongest p l a y e r s of the last c e n t u r y and S t e i n i t ? ' s opponent In the f i r s t o f f i c i a l match for the World Champions h i p , published in his chess column the position from t h i s game, and showed the simple w i n n i n g method for White. Twenty years l a t e r , on l e a r n i n g of the death of the English master Potter, the French language professor Georges Barbier remembered this ancient game, and, on analysing it, discovered In a similar position (cf. diagram) the unexpected stalemate Idea (6 c8=Q Rc4+1). He published the position f a m i l i a r to us, but with a d i f f e r e n t task —White to p l a y , a n d . . . Black to d r a w . The newspaper which p u b l i s h e d the study dedicated to Potter came to the notice of Ihe p r i e s t Saavedra. While s o l v i n g Barbier 1 s p o s i t i o n , this hitherto unknown chess p l a y e r conceived an Idea which was f a n t a s t i c for that time — why not promote the pawn to a rook, thus a v o i d i n g the stalemate — 6 cS=R! On v i s i t i n g the Glasgow chess c l u b , Saavedra acquainted its members with his sensational discovery, thanks io which he soon became famous throughout the chess w o r l d . Thus one single move gave Saavedra the r i g h t to i m m o r t a l i t y ! An amazing instance in chess h i s t o r y ! Although In p o p u l a r p u b l i c a t i o n s t h i s unique quartet w i t h white pawn against black rook is u s u a l l y called 5aavedra's position, it would be f a i r e r to r e g a r d the study as a c o l l e c t i v e c r e a t i o n , belonging to two authors — Sarbier and Saavedra.

Pawn a g a i n s t

queen

With this material one cannot devise either a study or a normal proRlem, and for the f i r s t time we are forced to have recourse to the helpmate genre — No. 5.

4th Letter — 40 Chess Quartets No. 5

37

E. Holladay, 1972

Helpmate in 4 moves. The black queen parlicipates in the confinement of her k i n g : 1 Kf3 (remember that in helpmate problems it is Black who begins, and in the solution his moves are given first) 1...Kf6 2 Kg4 e7 3 Kh5 e8=R! 4 Qg4! Rh8 mate. Knight against pawn With such a balance of material, of course, White does not usually entertain any hopes of winning. An exception is provided by instances where the black pawn restricts its own king, and as a result it can be ma led. We, however, give example No. 6, in which, by means of intricate leaps, the white knight has to catch the enemy pawn, so as to save the position. No. 6

N. Grigoriev, 1932

White to p l a y and 1 Nb4! 3 Ne5, w i t h Kg4, and it 6 N f i , with e f f o r t s come

draw.

h5 2 Nc6! (2 Nd5+? Kf3 and wins} 2...Ke4 ( 2 . . . h 4 a simple d r a w ) 3 Na5! (3 Nd8? h4 4 Ne6 Kf5 5 Nd4+ is a l l over) 3...h4 4 Nc4! Kf3 ( 4 . . . h 3 5 Nd2+ and a draw) 5 Ne5+ ( a f t e r 5 Nd2+ Ke2 6 He*. h3 a l l White's to n o t h i n g ) 5...Kg3 6 Nc4! h3 7 Ne3. Draw.

Knight against piece If on the board there Is a white knight and a black knight, rook or bishop, one cannot compose either an orthodox problem, or a study. For helpmate manoeuvres, however, there are rich possibilities. We give one quartet for each black piece (Nos. 7-9).

38

Seven Chess Letters No. 7

Z. Mach, 193?

f:

Helpmate in 4 moves. Here is the solution to 3 Kat Kb3 4 Nb1 Nc2 mate.

this

No. 8

'square':

1 Na3 KaA

2 Kb2 NbA

D. Bebiz, 1967

Helpmate in 7 moves. The black king is adjacent io the corner square hB, but It is mated in the opposite corner of the board: 1 Kf6 Kci 2 Ke5 KbZ 3 KdA Ka3 4 Kc3 Ka4 5 Kt>2 Kb* 6 Kal Ka3 7 Bbi Nb3 mate. No. 9

M. Myllyniemi, 1965

Helpmate in 5 moves. The solution unexpectedly begins with castling! 2 Kb8 Nc4 3 Ka8 Kc7 4 Rd7+ Kca 5 Ra7 Nb6 mate.

1 O-O-CH! Kc6

Alas, with a white knight against a black queen it is impossible io devise even a helpmate — such a mating position simply does not exist. We give position No. 10 merely so as to avoid creating a

39

4th Letter — 40 Chess Quarteis vacuum (according to our plan, each r e p r . e 5 e n t e d , if only by one quartet).

set of four

pieces

must be

No. 10 Conclusion of a study

White to play and draw. - An i I lustration of the fact that in chess everything is possible — hopes of saving the game should not be abandoned even when one has only a lone knight against a queen: 1 Ng3+ Kh2 2 Nfl + , with perpetual check. Of course, this is not a genuine work of a r t , but merely an amusing epilogue to some study with many pieces in the initial position. Bishop against pawn Although simple, example No. I I

Is not without

interest.

No. 11 E. Pogosyants, 1979

White to play and draw. 1 Bb3! The only possibility of averting the promotion of the pawn (1 8fa5 Kc2!, and Black wins). Now two symmetric variations arise: 1...Ka1 2 Bc2, or l...Kc1 2 Ba2, and the pawn is stopped. Bishop against knight or bishop Here too, as with a white knight against a black minor piece, we have to restrict ourselves to helpmates — Nos. 12 and 13.

40

Seven Chess t e l l e r s No. 12

l Hi

E. Albert, 1963

SIB

BH

I• • 1

MO

Helpmate in 5 moves The kings go round in a circle, exchanging places, after which the black leader ends up in a mating net: 1 Kb6 Kb8 2 Nc6+ Kc8 3 Ka7 Ba6 4 Ka8 Kc7 5 Na7 Bb7 mate. No. 13

£ . Pogosyants,

1980

•"

r

~

'

" ";/

Helpmate i n 5 moves. Clearly, male is possible only w i t h opposite-coloured b i s h o p s . 1 KeA Ba8! ( a w a i t i n g move) 2 Kf3 Kd5 2 Kg2 Ke4 ( d i a g o n a l p u r s u i t , w h i c h in studies is encountered only w i t h a l a r g e number of pieces) h K M Kf3 5 Bh2 Kf2 mate. Bishop a g a i n s t Study No. 14 is a f u r t h e r No. 14

rook

illustration

on the theme of symmetry.

E. Pogosyants, 1978

White to p l a y a n d d r a w .

4th Letter - 40 Chess Quartets

*'

The symmetry is s l i g h t l y broken (the a - f i l e d i s t u r b s i t ) , and ttiis i S of decisive s i g n i f i c a n c e . 1 Bb3! The symmetrical I Bh3 loses to 1 ...Rh4 2 Bg2 Ra4! (this is where the i n i t i a l asymmetry manifests i t s e l f l 3 K f l Rai mate, or 3 Kd1 Ra1 + 4 Kc2 Ra2+. 1...Rb4 2 Bc2t Rf4. On 2 . . , B h 4 3 Kdi there follows 3...Bh1 mate, but after 3 Kf 1 ! RM + ft Kg2 Black does not succeed in winning the bishop — to the r i g h t of the e - f i l e there Is one less free f i l e than A draw also results from 2,..Rc4 3 Kdl Rhfc 4 Kcl Rh1 + l 0 the left. 5 KbZ. 3 Bb3! Noi 3 Bbl Rd4 4 Bc2 Rd2! 5 Bb3 Rb2!, and wins. 3...Rf2 * K d l ! (after the r e t u r n ol the bishop lo i t s i n i t i a l square, 4*Be6, Black wins by 4 . . . R f 6 5 Bb3 Rb6 6 Bc2 Ra6 7 K f l Ral+ and 8...Ra2) 4 . . , K d 3 5 Kei Ke3 6 K d l , and the draw is i n e v i t a b l e . Bishop against queen Position No. 15, like No. 10, t a n only serve as a good f i n i s h to a study. No. 15 Conclusion of a study

White to play

and draw.

t Bg3! Qxg3 stalemate. Note that only two quartets — knight or bishop against queen — do not allow either a study or a problem to be composed on the chess board. Any other balance of forces can serve as the basis for the creation of a chess work of a r t .

Rook againsl pawn Here there is r i c h material f o r studies with White to play a r | d win. I t was not easy to choose one of them, and so we have settled on problem No. 16. No. 16

K. Alheim, 1966

(see next

diagram)

Mate In 4 moves. > Ra7! Kh4

(1...Kh2

2 Ra5! «6 3 Re5! Kh3 f i r s t quartet, but here rook.

2 Rxe7

4 Rh5 male. the pawn has

Kg!

3 Rh7 Kfl

4 Rhi mate)

The problem resembles the very already been transformed into a

42

Seven Cliess

(Diagram for problem No.

Letters

16;

Rook against

see previous

page)

knight

The chess h i s t o r i a n van der Linde has discovered examples of the battle between rook and knight in an ancient Arabic manuscript, dated 1140! As is reported in the second part of the book, positions with rook against knight can be f a u l t l e s s l y evaluated by a computer. ,.....
Vi. 5chinkman,

1905

Mate in 5 moves. 1 Rb7! All the pieces are arranged on one d i a g o n a l . Now, depending on the move of the black k n i g h t , there are two symmetric mates: t...Nd4+ 2 Kb6! Ne6 3 Rh7, 4 Rh8 and 5 RxN mate; 1...Ke5+ 2 Kc7T Nc4 3 Rb1, 4 Rai and 5 RxN mate.

Rook against

bishop

We again settle on a problem — No. 18. No.

18

K.

Fabel,

1933

Mate in 3 moves.

4th Letter — 40 Chess Quartets "file:

The black king is cut off on the rank, 1 Kc2!, 2 Ra6+ and 3 RxB mate.

but

43 gets meted on

the

Rook against rook In the helpmate No. 19 the black rook, before being removed from the board, manoeuvres subtly, allowing the white king to approach directly the centre of events. No. 19

E. Albert, 1963

Helpmate in 4 moves. 1 Rh4! Kd3

2 Rh5! Ke4

3 Rh3! KfS

4 Kh5 Rxh3 mate.

Rook against queen The ancient position No. 20 can be regarded as a study, Is to be found in a l l books on the endgame. No. 20

although

J. Berger, 1889

White to play and draw, 1 Rh7+ Kg2 2 Rg7+ Kh3 (the f - f i l e is p r o h i b i t e d to the k i n g , In View of Rf7) 3 Rh7+ Kg4 4 Rg7+ KhS 5 Rh7+ Kg5 6 Rg7+ Kh6 | 7 Rh7+ Kg6 B Rh6+! Kxh6 stalemate. In the i n i t i a l p o s i t i o n the black queen can occupy any square on the f - f i l e apart from f5, and the i result w i l l be the same — a d r a w . Queen against pawn Endings of this type have been exhaustively studied in theory and

44 are of a text-book

Seven Chess Letters nature,

bul

No. 21

interesting

problems

occur

—No. 21.

R. Lermet, 1914

Mate in k m o v e s .

• '.

The white king, having pinned its opposite number to the edge of the board, unexpectedly allows him to break free, but, of course, not for long. 1 Kd5! Kb5(b6> 2 Qb8+ Ka4 (a5) 3 Kc4 (the king retraces its steps) 3...Ka5(a4) 4 Qb4 mate. Queen against knight In the struggle against a minor piece, the queen usually accomplishes a clever manoeuvre, depriving the enemy king of the support of Its piece — No, 22. No.

22

R. Garraukh, 1923

• '

Mate in 3 moves. 1 O a l ! Ke8(cB) 2 Qg7(a7), a n d mate next move. Once a g a i n we see the theme of asymmetry - a f t e r 1 Qgl? Nf6! ( l . . . N f 8 ? 2 Qg7) the w h i t e queen has no square on the left analogous to h 8 . Queen a g a i n s t

bishop

In the next problem, No. 23, as i n the p r e v i o u s one, a l l the pieces a r e concentrated on one f i l e , b u t on t h i s occasion the queen must remain on i t .

4th Letter - 40 Chess Quartets No. 23

R. Lermet, 1923

Mate in 2 moves. Here, in contrast, asymmetrical continuations do noi work: Qb27 Bb4!, 1 Qh2? Bh4!, and there is no mate. After the correct Qe5! the symmetry of the pieces is maintained, and two symmetric cho-mates arise: 1...Kd8 (Bb4, Bc3) 2 Qb8 mate; 1...KfB (BhA, Bg3) Qh8 mate. Queen against rook In problem No. 24 ihe black king is on the back rank, while the ok too does not occupy the most favourable position, but nevertheless ore than 10 moves are required to male the black k i n g . No. 24

V. Shpekman, 1976

. •

Mate in 11 moves. 1 Ke6. Now Black has two possibilities (castling is ruled out, since either the king or the rook has just moved). On l . . . K f 8 , mate quickly results: 2 Qf4+ Kg7 3 Qf?+ Kh6 4 Kf6 Rf8 5 QxfB* Kh5 6 Obi! Kh6 7 Qh4 mate. The main variation arises after l...Rh7! White has to make a series of exact moves with his queen: 2 Qg5 Kf8 3 Qd8+ Kg7 A Qe7* Kg8 5 QeB+ Kg7 6 Qf74 Kh6 7 Qf6+ Kh5 8 Kf5 Rf7 9 Qxf7+ Kh4 10 Qb3! KhS I I Qh3 mate. Queen against queen We w i l l conclude our examination of number of pieces with the helpmate No. 25.

quartets

with

an

equal

5even Chess Letters No. 25

R. Lermet,

1937

Helpmate in 3 moves. 1 Qh7! Kg3!

2 Kg7 Kf4+

3 Kh6 Qg5 mate.

TWO WHITE PIECES Such an advantage plenty of problems, and

for White is no good for studies, but there are making the selection was by no means easy.

Queen and pawn Perhaps chess b o a r d .

the

most

widespread

No. 26

Y.

combination

Vladimirov,

In

quartets

on

the

1976

Mate in 4 moves. This four-mover illustrates the co-ordination theme of two queens (the second w i l l soon appear) on squares of opposite colour. 1 QgB! Now a series of interesting v a r i a t i o n s arises: 1 . . .Kd4 2 d8=0+ Kc5 (2...Kc3 3 Qa2!) 3 Qb3T; t...Kd2 2 d8=0+ Kci 3 Qg2!; l...Kf3 2 d8«Q Ke4 (2...KH 3 Qd4+) 3 Og4+ etc. It goes without saying that each time White gives mate an the fourth move.

Queen and

knight

Three-movers with this material can and Ho. 27 is the most famous of them.

have

very

clever

solutions,

4th Letter — 40 Chess Quartets No. 27

W. Schinkman,

47

1885

_•$•_• Mate in 3 moves. 1 Na8M This 1...Kd6 2 Kd4 Kc6

paradoxical f i r s t 3 Qd5 mate.

move creates a strong impression.

Queen and bishop A classic combination for three-move quartets. The following problem, together with two other quartets also depicting the Russian letler ' r ' (G), was published on Space Exploration Day — 12th A p r i l 1964. This set of problems was dedicated by the author to Yury Gagarin — the f i r s t man to conquer space. No. 23

E. Pogosyants, 1964

Mate in 3 moves. 1 Qh3! KgS

2 Qf3! K96 (2...Kh4

3 Bf6 mate]

3 Qg4 mate.

Queen and rook White's advantage is mate is not always easy.

too

No. 29

great,

but

to f i n d

the shortest

path

to

W. Schinkman, 1885

(see next diagram)

.. -

r

Mate in 4 moves.

Kd4

See how the noose 2 Qe6 Kc3 (2...Kc5

is tightened 3 Rh4 KbS

around the black k i n g : 4 Rh5 mate) 3 Qd7! Kc4

1 Qf7! 4 Rc2

Seven Chess Letters

male; Ke3j mate)

1...Ke5 2 Kb4 (this move is also decisive after" l . . . K d 3 or 1 . . . 2...Kd6 (on 2...Kd4 or 2...Ke4 there follows 3 Re2 and k Qc4 3 Rh5 Kc6 * Rh6 male.

Two queens On a normal board with two queens, four moves are more than enough to mate the black k i n g . It might be supposed that, if the dimensions of the board are increased, the number of moves necessary for mating w i l l also increase. But strangely enough, whalever the dimensions of the board, mate can always be given not later than the fourth move! On the f i r s t move one of Ihe queens gives check along a file. After the black king moves to one of the adjacent f i l e s , the other queen (with the help of the f i r s t ) squeezes it inio two f i l e s . We then have approximately the position shown in No. 30 (we assume here that the normal chess board represents a fragment of a board of a r b i t r a r y dimensions). No. 30

An innumerable multitude of problems

On any board two white queens give mate not later than the fourth move Now any move by the king is answered by an appropriate h o r i zontal check, for example 2. ..Ke4 3 Qc4+, or 2. ..Kd5 3 Qf5+, and mate next move: In the f i r s t case 3 . . .Ke3(e5) 4 Off4 mate, in the second — 3...Kd4(d6t 4 0cc5 mate. The black king could s i m i l a r l y be ' c a u g h t ' along a r a n k . Obviously, the board can have any dimensions, and can even be i n f i n i t e . At the same time, the i n i t i a l p l a c i n g of the white queens and both kings is unimportant (the white king Is altogether not necessary).

I

4th Letter — 40 Chess Quartets

49

Rook a n d pawn In problem No. 31 the iwo k i n g s does honour only to the white leader. No. 31

go r o u n d

G. Glass,

in a c i r c l e ,

but

this

1903

Male in 4 moves. 1 K f l ! Kd2

2 Kf2 Kdi

3 Ke3 Kel

4 Rd

mate.

Rook and k n i g h t In No. 32 mate cannot be given without the support so White must b r i n g h i s up to face the opponent. No. 32

M. Ebert,

of

the

king,

1975

Mate in 6 moves. 1 Ra2! Kc1

2 Nd2 Kdl

3 Nb3! Kel

4 Kg2 Kdi

5 Kf3 Ke1

6

mate. Rook and bishop As iri the q u a r t e t w i t h queen and b i s h o p , we have chosen problem in w h i c h the bishop is a ' s u p e r f l u o u s ' piece — N o . 33. No. 33

K. Law,

(see next

1928

diagram)

Mate in 3 moves.

a

Seven Chess Letters

50

1 Bh2! Kxh2

2 Ra3 Khl

3 Rh3 mate.

Two rooks •

In order of his rook s.

to achieve

his

No. 34

goal

in

No. 34, White must

^ v

give up one

W. Schinkman, 1877

Mate in 3 moves.

mate;

1 P h i ! ! Kxhl 2 Kg3 Kg1 3 Re! mate; 1...KfZ 2 Kh3 Kf3 3 Bfi mate.

1...Kf3

2 Rh2 Kf4

3 Rf2

Bishop and pawn. The solution of No. 35 is not d i f f i c u l t , but it contains simultaneously two problem motifs — a sacrifice and an under-promotion. No. 35

A. McKenzie, 1691

"*_•„•_ Mate in 3 moves.

4th Letter — 40 Chess Quartets

51

I Ba7! Kxa7 2 cB=R! Ka6 3 Ra8 mate. We should mention one f u r t h e r curious d e t a i l . Since in the I n i t i a l ' p o s i t i o n it is While's move, Black must have been the lost to p l a y . But this could only have been a move with his king from a7 to aB. This means that before this the black k i n g was In check by the bishop at b8. Bui the bishop Itself could not have got there, I .e. it could only have appeared as a result of a promotion. Thus in this modest problem there are as though two promotions: one (lo a bishop) in its 'past' (relro-play), and the other (to a rook) in Its ' f u t u r e ' (the solution i t s e l f | .

Bishop and

knight

In order to give mate with bishop and knight one needs, so to speak, special p r e p a r a t i o n . Instances are known where a player in a tournament game d i d not manage it in the allotted time (the rules allow f i f t y moves for the mate). It is amusing that, in the 47th USSR Championship (Premier League), international master Anikayev tested the mating technique of grandmaster Balashov for some twenty moves in this ending, but when his k i n g was slowly but surely forced towards a corner of the same colour as the bishop (an essential c o n d i t i o n ! ) , he curtai led his resistance. Problem No. 36 is of a certain practical interest (as, however, is any problem of this t y p e ) . Imagine that you have played this ending f l i p p a n t l y , and at the given moment have already used 45 moves oul of ihe allotted f i f t y . . . No. 36

S. Loyd,

1857

-

Mate in 5 moves 1 8d6 Kb1 2 Kb3 (White's reverse order) Z,..Kal 3 Ba3! Kbi

first two moves can be 4 Nc3+ Ka1 5 Bb2 male.

made

in

Two bishops This combination of pieces, as they say, leaves something to be desired. However, although It is easy to give mate (this task is w i t h i n the powers even of a beginner), to f i n d the shortest path to the goal ' * not always easy — No. 37. No. 37

V.

(see next

Pauli,

1919

diagram)

Mate in 7 moves.

52

Seven Ches5 Letters

1 Bf3! Kti3 2 Kf2 Kh4 3 Se2 ( b e g i n n i n g to d r i v e the k i n g into the comer) 3...Kh3 4 Bg5 Kh2 5 Bfl KM 6 Bg2+ Kh2 7 Bf4 mate. Two knights

.

Of course, a mate is possible only if the b l a c k k i n g p l a y s White's hands. A corresponding helpmate is shown in No. 38. Ho. 3B

-

•

•

*

%

into

£ . A l b e r t , 1964

Helpmate in h 1 Kb5! Nd7 2 Ka4t { B l a c k marks tim»», but does not prevent the w h i t e k i n g from f u l f i l l i n g i t s role) 2...KC5 3 Kt>3 Kb6 A Ka4 Nc5 mate.

Knight and pawn No. 39

P- Terngren, 1928

Mate in 3 moves. 1 fB=R! (1 f8=Q stalemate!)

1...Kh6

2 Rf7 KhS

3 Rh7 mate.

4th Letter - 40 Chess Quartets

53

Two pawns Problems with two pawns are normally rather interesting, since [hoy always contain unusual promotions. We conclude our parade of quartets with No. 40, in which essentially four problems are represented. In each case one of the pawns promotes to a queen, while the other promotes to four different pieces — queen, knight, bishop and rook! No. 40

11

(a) jb) 3 moves. (c) moves. 1 (d) moves. 1

' "

0.

Dehler,

1925

Mate in 2 moves.

-

'

As in the diagram. 1 c8=Q and 2 b8=Q mate, Move the white k i n g to b3. (V. Shpekman, 1964). Mate In 1 bB=Q Ka5 2 c8=N! Ka6 3 Qb6 mate. Move the b7 pawn to a7. ( L . Kubbel, 1940). Mate In 3 a 8 - B ! , 2 cB~Q and 3 Qb7 mate. Move the c7 pawn to g7. (H. Schtaudte, 1965K Mate in 3 bB=R! Ka7 2 g8=Q Ka6 3 Oo2 mate.

Thus, there have appeared before you 40 quartets on the chess board. As we have seen, between some of them lies a whole c e n t u r y . . . And even so this modest material has been far from exhausted by chess composers.

Fifth Letter TRIOS AND DUETS ON THE CHESS BOARD

In the preceding ' l e t t e r ' we saw that, for the composing of an interesting problem or study, four nieces are quite sufficient. But what about with less material? Is it possible to think up anything deserving attention with three or two pieces on the board? It turns out that, even in this 'chess m i c r o - w o r l d ' , a number of interesting works of the most varied genres have been created, although It is true that there is a predominance of unusual or joke problems. In this ' l e t t e r ' we w i l l meet a series of such e x h i b i t s . If we are t a l k i n g about t r a d i t i o n a l , orthodox problems, they can turn out in one of three ways: against a lone black k i n g the leader of the white forces w i l l be accompanied by a pawn, rook or queen.

No.

1

K. Tonipltnson, 1845

Mate in 2 moves. The ancient two-movers Nos. 1 and 2 relate to the stalemate theme — in both, the solution is to promote the pawn to a rook (but not a queen): I n N o , 1 - 1 c 6 = R ! , In N o . 2 - 1 g6-*R!, a n d 2 RaS m a t e .

No.

2

E . C o o k , 1868

(see next

diagram)

54

5th L e t t e r — Trios and Duets on the Chess Board

55

Mate in 2 moves. The later problem No. 3 is an improved version of No. 1 . Here the appearance of the rook is delayed by one move — 1 Kc6! Ka7 2 c8=R! Ka6 3 Ra8 mate. No. 3

H. Meyer,

1920

Male in 3 moves. The idea of promoling a pawn to a rook w i t h such l i m i t e d material can also take on a t r i c k form — No. 4. No. 4

E. Pogosyants,

1980

Mate In 3 moves. The subtle point Is that White f i r s t of a l l . . . folds the board in two! Now there follows 1 Kc2! Ka3 (the squares a5 and b5 are on the other half of the b o a r d l ) 2 c4=R! (the f o u r t h r a n k has become the last} 2...Ka2 3 RaA mate.

56

Seven Chess Letters No. 5

D. Hoiman, 1971

Mate in 3 moves. In No. 5 the introduction is rather crude — 1 c8=Q, but on the other hand the subsequent manoeuvre on the next move by the new-born queen is the only solution: 1...Ka4 2 Qh3! Ka5 3 Qa3 mate. Are ther-e to be found among trios any genuine studies? It can, of course, only be a question of the material 'King and pawn against king' . Positions of this 'ype are normal ly considered theoretical, but certain of them are customarily regarded as studies, as for example Nos. 6 and 7. No. 6

J. Dr-tina,

White to play

1908

and w i n .

No. 6. 1 Kc2! Ke7 2 Kb3! {a roundabout manoeuvre with the aim of taking the opposition; a draw results from 2 Kd3? Kd7 3 Kd4 Kd6!) 2...Kd6 3 Kb* Kc6 4 Kc4, and the pawn queens. If the pawn is moved from c3 to b4, the deciding king route is lengthened by one move; 1 Kc2 Ke7 2 Kb3 Kd6 3 Ka4! Kc6 4 Ka5! etc.

No. 7

A. Hilderbrand,

1954

5th Letter — Trios and Duets on the Chess Board

57

White to play and win. No. 7. 1 Ka& Kc6 2 Ka5 Kb7 3 Kb5 Ka7 4 Kc6 Ka6 5 b5+ Ka7 6 Kc7 Ka8 7 Kb6 Kb8 8 Ka6t Ka8 9 b6 Kbfl 10 b7 Kc7 11 Ka7 and 12 bfl=Q. It is interesting that, in the symmetric position after Black's 7th move, the manoeuvre of the white king lo «6 is the only solution, since a draw results from the symmetric 8 Kc6? Ka7 9 06+ Ka8 10 Kc7 stalemate! No. 8

A. Mandler, 1969

White to play and draw. In study No. 8 after the transfer of the kings to the other wing — I Kb2! Kb6 2 Kc2! Kc6 3 Kd2! Kd6 4 Ke2 Ke6 5 Kf2 Kf6 6 Kg2! Black loses command of the corresponding squares. Of course, the solution is very simple, but on the other hand we have a unique example of White being represented In a ;study by his king alone. Helpmate trios can prove more fascinating than positions with a normal task, since in them accurate play is demanded not only of the two white pieces, but also of the black k i n g . No. 9

E. Bon, 1961

White to p l a y . Helpmate in 7 moves. In No. 9 hurry — I d3! a tempo creates * d6 Kc5 5 d7 .have been able the new queen)

the pawn must, of course, advance, but it is not in a With such limited material, this popular idea of losing a strong impression. 1 . . .Kf2 2 d4 Ke3 3 d5 Kd4 Kb6 (after the hasty 1 d4 the black king would not now to cross the square b6, which would be controlled by 6 dB=Q+ Ka7 7 Qa5 mate.

Two f u r t h e r t r i o s w i t h the p a r t i c i p a t i o n of a white pawn w i l examined a l i t t l e l a t e r , b u t now i t i s the t u r n of the r o o k .

ft

be

58

Seven Chess Letters Ha. 10

L.

Tsagler,

1973

Mate in 12 moves. Everyone can mate a bare k i n g w i t h a rook, but to do this in the shortest possible way — No. 10, one has to rack one's b r a i n s . 1 Ka3! Kc3 2 Ka4 Kc4 3 R d l ! Kc5 4 Kb3 Kb5 5 Rci Ka5 6 Kc4 Kb6 7 Kb4 Ka6 8 Kc5 Kb7 9 Kb5 Ka7 10 Kc6 Ka8 I I Kc7 Ka7 12 Ra1 mate. This serious problem resembles the 'trio with the immobile rook1, examined by us in the f i r s t ' l e t t e r ' . No. 11

E. Pogosyants,

1980

Mate in 2 moves. While in No. 4 the board was folded in h a l f , in joke-problem No. 11 an even more unexpected m a n i p u l a t i o n is c a r r i e d out on i t . in order to mate the b l a c k k i n g in such short time, you must f i r s t . . . saw off the last r a n k from the b o a r d (or cover i t , if you begrudge the board). Now the k i n g has nowhere to g o : 1 Kb5! Ka7 2 Rc7 m a l e .

No. 12

M. V u k o v i c ,

1962

Black to p l a y . Helpmate in 3 moves.

5th Letter — Trios and Duels on the Chess Board

59

If in position No. 12 [he black king were to 'resist', even twelve moves would not suffice. His assistance speeds up (he solution considerably: 1 Ke7t (Black begins) 1...Kg7 2 KeB! Kf6 3 Kf8 Rh8 mate. The normal mating of a lone king by a queen Is too primitive a task, and therefore we i lluatrate this trio with the helpmate No. 13, No. 13

R. Forstor, 1966

Helpmate in 4 moves. 1 Kb4 Qh2! 2 Kc3 Kg2! (a hidden, in the literal sense, possibility of admit ling the black king to the place of its downfall ) 3 Kd2 Kf3+ 4 Kei Qe2 mate. No. 14

A. Gerbstman, V. Slobodenyuk, 1976

White takes back his last move and gives mate in \ move. (a) as in the diagram; (b) the king is moved from c7 to c8.

i t . ' . . ' .>•

A whole series of ultra-miniatures are associated with taking a move back. In the twin-problems Mo. H White takes back the move b6xa7, instead of which there fol lows 1 b7 mate, and, as can readily be checked, in the first case the piece el a7 could only be a black knight, and in the second case — only a black bishop. Replacing the pawn at a7 In (b) by a white knight, we obtain a third ' t r i p l e t ' Instead of Nb5xa7 the solution Is 1 Nc? mate.

60

Seven Chess Letters No. 15

While takes back his last

K.

Fabel, 1949

move and gives mate in 2 moves.

In joke-problem No. 15 try guessing that White's last move w a s . . . K-side c a s t l i n g . The most s u r p r i s i n g thing is that the king alone p a r t i c i p a l e d in i t , moving from e1 to gi . The point is that While was p l a y i n g a game with ihe odds of two rooks, anct the rook at h3 is the result of a promotion! Instead of castling K-side he now castles oh the Q-side, as a result of which his king goes from el not to g i , but to c1 (the rook at a 1 , given as odds, l i k e the hi rook participates in the castling purely symbolical l y ) . After 1 0-0-0 the black king is forced into the corner, 1...Ka1, where it is mated by 2 Ra3.

No.

16

E. Pogosyants,

1980

. . I , !•-•.•

White takes back

his

The joke-problem No. Pushkin's novel in verse. Lensky and Olga:

last

move and gives mate in 1 move.

16 can be regarded as an i l l u s t r a t i o n to Remember the well-known 'game' between

They meet, from others far retreating. Above a game of chess they bow, Now heads on elbows resting, now In deepest meditation s i t t i n g ; A pawn of Lensky's takes a rt>ok, His own, d i d he but care to look.* The present position arose precisely after the absenl-minded Lensky had broken the rules of the game and with his g4 pawn captured his own rook at h5. Now he begs Olga to excuse him, takes back *The translator is indebted to Mr Henry Jones of Hiding berland, for this translation from Yevgeny Oniegm.

Mill,

Northum-

5th Letter — Trios and Duets on the Chess Board

61

the move gixhS and announces check and mate to the bemused g i r l by t Rh5-h8! This may be a joke-problem, but an interesting fact is that it does not permit any false rumours. For example, the assumption that Lensky had captured his queen on h5, and not his rook, would have meant that 01 ga too was p l a y i n g absent-mindedly — otherwise there is no explanation as to why the black king moved into check. Pushkin, however, has nothing to say on t h i s !

No.

17

J. Sunyer,

1923

White and Black take back their last move, then Black makes a move and White gives mate in 1 move. The clever problem No. 17 broadens the possibilities of the two sides to such an extent that the kings alone suffice, and without any jokes! With Its last move the white king from g6 captured a black rook at h5 (Kg6xRh5), which, in t u r n , had from h8 just taken a white queen at h5 (Bh8xQh5). In the restored position Black castles K-slde, . . .0-0!, and is immediately mated by 1 Qh5-h7. Thus, we have played the f i r s t duet on the chess board! Very similar to it is another problem with two kings — No. IB. No. 18

F. Vaird,

1910

White and Black take back their last move, then Black makes a move and White gives mate in 1 move. Here too White's last move was to capture e black rook Kg<>xRf5, which prior to this took the white queen - RfixQfS. Instead of this Black plays Rf1-h1, and is mated by 1 Qf5-f2. Amusing trios and duets can be played at give-away chess, where taking is obligatory ( i f there are several possibil i ties, the choice is a r b i t r a r y ) , and the player who wins is the one who is the f i r s t to

62 give away

Seven Chess Letters all

his

pieces,

including

No.

19

K.

Ihe

king

Fabel,

White to play

and

(or

stalemate

them).

1955

lose,

" '

Problem No. 19 shows that in give-away chess the presence of only two pieces ( i n the g i v e n instance pawns) is more than adequate to devise subtle and o r i g i n a l s i t u a t i o n s . 1 a3! The f a m i l i a r idea of losing a tempo, b u t . . . in a completely d i f f e r e n t genre! I...h5 2 aA h4 3 a5 h3 4 a6 h2 5 a7 h1=R! If Black promotes to a queen or bishop, then after- any promotion of the white nawn he w i l l be forced to take it immediately. On 5 . . . M " N there follows 6 a8=Q, and on a move by the k n i g h t - 7 Qh1 . I n t e r e s t i n g v a r i a t i o n s a r i s e after the appearance of a black k i n g — 5 - . . h 1 = K. Bad now for White is 6 aB=Q or a8=B because of 6 . . . K g 2 ! ; 6 aB=K leads to a draw (the k i n g s cannot approach each other, and w i l l wander Oerpetually around the board) ; on 6 a8=N there follows 6...Kg2 7 Nb6 Kf3 8 Nd7 Ke3 9 Nb6 Kf4 10 Nc8 Ke4 11 Na7 Ke5, and the k n i g h t is ' c a u g h t ' - - on the oexl move it w i l l have to take the king. The solution is 6 aB=R! Kg2 7 Ra4 Kf2 8 Rd4 Kg2 9 Re4 Kh2 10 Rf4 Kh1 11 Rf3 Kg2 12 Rf2+ Kxf2, and the goal is achieved. 6 aB=8!! The only move. White promotes to an even weaker piece than h i s opponent. Now, wherever the rook moves, there follows 7 B h l ! , a n d t h i s game of g i v e - a w a y chess ends i n White's f a v o u r .

No. 20

E.

Matt

Pogosyants,

In \

1963

move.

Our f i n g l d u e l , No. 20, also demands of the reader a c e r t a i n sense of humour. White has already managed to make h a l f a move — he has removed from the board his h? pawn, and now be completes the move, p l a c i n g on h8 a queen! And t h i s means that White's complete move is 1 h7-h8=Q mate.

5th Letter

— Trios and Duels on the Chess Board

63

Thus, we have played on the chess board forty quartets end twenty trios and duets. But the most d i f f i c u l t chess puzzles, strangely enough, contain only one piece — a solo on the chess board! We have In mind problems i n v o l v i n g the journey of a piece over all the SQuares of the b o a r d . About the k n i g h t ' s tour alone, numerous books and scientific researches have been w r i t t e n . It w i l l be recalled that the 'problem of the k n i g h t move1 was studied by the great mathematician Euler. In the present book the detailed study of 'chess geography' problems w i l l be left to one side. After a l l , chess Is inexhaustible, and, as is known, it is impossible to encompass the unbounded!

Sixth Letter TWO DECISIVE GAMES

Decisive chess games, games on which depends the fale of a major i nternational tournament, or sometimes even ihe fate of a player, games which determine the winner of ihe recurrent stage of the battle for the World Championship, or even the possessor" of the chess crown, have always provoked a special, exceptional interest among the numerous lovers of the ancient game. In a decisive game, where 'the slake is greater [nan life i t s e l f , victory is often determined not only by the talent of a grandmaster. Character, will-power, the a b i l i t y to collect oneself at the c r i t i c a l point of the struggle — these qualities play no less a role than chess mastery. Chess history knows of a number of examples, when a defeat in a decisive game had such an effect on a player thai he was never again able to climb to the heights which he had once been able to conquer. Bui a defeat in a decisive game, for a l l the tension of the situation, cannot break a true f i g h t e r . The a b i l i t y of a great master to r i d himself of the obsessive thought, that his many years of effort were ruined by one hasty move, the a b i l i t y to shake off the weight of disappointment, provoke enormous respect and admiration in his personality. Chess lovers will well remember Boris Spassky's two vexing defeats in his decisive games in the last rounds of two USSR Championships — in 1958 against M. Tal and in 1961 against L. Stein. Each time defeal closed the door for the talented grandmaster to the Interzonal Tournament, and hence put off for at least three years any hope of a match for the World Championship. But these defeats, in the opinion of Spassky himself, mer-ely hardened him, and prepared him for further great battles. By defeating Tigran Petrosian in 1969, Spassky became the tenth chess champion of the world. One of the most famous decisive games is Ihe concluding, tenth clash In the match for the world championship between Emanuel Lasker and Carl Schlechter, which took place in 1910. The name of the great Lasker is known to everyone. Today even some chess masters know of the great Austrian grandmaster Schlechter only by hearsay. And yet, if Schlechter had won this game against Lasker (and he was very close to his goal), he would have become World Champion, and would have been ranked alongside such giants as Steinitz, Lasker, Capablanca and Alekhine (the reader w i l l f i n d this game in the seventh ' l e t t e r ' of I,

6*

6th L e t t e r — Two Decisive Games

65

•the b o o k ) . Three h i s t o r i c a l ly d e c i s i v e games belong to grandmaster David [ B r o n s t e i n . One of Ihem was p l a y e d In the f i r s t C a n d i d a t e s ' Tournament, [•Which was held in 19S0 in Budapest. Before ihe f i n a l round the Soviet (•grandmaster Isaac Boleslavsky was l e a d i n g Bronstein by half a p o i n t . t B o l e s l a v s k y was confident lhat h i s young r i v a l would not succeed, to order, as they s a y , In d e f e a t i n g the h i g h l y experienced Paul Keres, and in the f i n a l r o u n d he q u i c k l y agreed a d r a w w i t h the Swedish grandmaster Gideon S t a h l b e r g . Soon he had cause to regret hi5 haste. Bronstein accomplished a competitive feat — he overcame his dangerous opponent in b r i l l i a n t s t y l e , and caught up w i t h B o l e s l a v s k y . The next decisive game was p l a y e d by him in the match designed to determine Ifie opponent of World Champion M i k h a i l B o t v i n n i k . The twelve s t i p u l ated games between Bronstein and Boleslavsky d i d not g i v e a w i n n e r , and the question a g a i n remained open. Two a d d i t i o n a l games were stipulated. The f i r s t of these ended in a d r a w . In the decisive duel Bronstein once a g a i n d i s p l a y e d h i s exceptional c h a r a c t e r , won i t , and gained the r i g h t to p l a y B o t v i n n i k . In t h i s match for the World Champ i o n s h i p , B r o n s t e i n ' s t h i r d decisive game was p l a y e d . To the end of the marathon there remained two games. B o t v i n n i k , who was t r a i l i n g b y a p o i n t , h a d White f o r the last time, a n d i t was s u f f i c i e n t f o r Bronstein to hold out in the 23rd game, when he would almost c e r t a i n l y have become the second Soviet World Champion. But B o t v i n n i k ' s w i l l proved to be the s t r o n g e r . In a tough b a t t l e of nerves he succeeded In e x t r a c t i n g a w i n in t h i s d e c i s i v e encounter. A short draw in the f i n a l , 24th game of the match enabled him to r e t a i n the t i t i e of World Champion... . It stands to reason that the story of d e c i s i v e chess duels deserves a separate book, and perhaps some time it w i l l De w r i t t e n . In our ' l e t t e r ' we suggest that you make the a c q u a i n t a n c e of two d e c i s i v e games, in w h i c h the a u t h o r s of t h i s book p a r t i c i p a t e d . T r u e , in the f i r s t game L u l i k o v - T a i m a n o v , one of the a u t h o r s p a r t i c i p a t e d only in the c a p a c i t y o f . . . grandmaster Mark T a i m a n o v ' s second. While the i n c l u s i o n in the book of the d e c i s i v e c l a s h from the World Championship Match in Baguio does not demand any e x p l a n a t i o n , the f i r s t game may appear to have been chosen r a n d o m l y . But t h i s Is not q u i t e so. The game was p l a y e d in the f i n a l round of the 37th USSR Championship, held in Moscow in 1969, which was also a Zonal Tournament. Victory in it took grandmaster Taimanov into the Interzonal Tournament, and 1 from there ' b y i n e r t i a — i n t o the C a n d i d a t e s ' Matches. The FischerTaimanov q u a r t e r r f i n a i match became an h i s t o r i c one. By w i n n i n g by the u n i q u e score of 6-0, the American grandmaster began h i s s w i f t offensive in the b a t t l e for the w o r l d c r o w n . The Dane Bent Larsen was overcome by the same enigmatic score. The next to suffer at F i s c h e r ' s hands was the Ex-World Champion T i g r a n P e t r o s i a n . By w i n n i n g the match in 1972 a g a i n s t Boris Spassky, Robert Fischer became the 11 th Chess k i n g in h i s t o r y . Three years later he decl ined to defend his t i t l e in a match w i t h Anatoly K a r p o v , and the Soviet grandmaster was declared the 12th Champion of the W o r l d . By w i n n i n g three years 'ater a match a g a i n s t V i k t o r Korchnoi, Karpov r e t a i n e d his t i t l e . As you see, a l l the events o c c u r r i n g In the chess world are in some way Connected one w i t h a n o t h e r !

66

Seven Chess Letters

From a grandmaster's

second:

Final round of the 37th USSR Championship, 1969 Lutikov-Taimanov Sicilian Defence The hall was shaking and buzzing with excitement. The controller cal led in vain for silence. The hearts of grandmaster Taimanov's supporters suffered an unbearable s t r a i n . One of his staunches! friends, the writer Leonid Zorin, who had not missed a single evening at the Championship, admitted after the game: " I don't know how I stayed al ive". It stands to reason that, before the start of ihe round, the tournament situation was taken into account. Four players were to go through to the Interzonal Tournament, and grandmasters Polugayevsky, Smyslov and Geller had already reserved three of the vacant places. There remained one, for which ihe contenders were Mark Taimanov and Leonid Stein, who were level on points. The most probable outcome of the Polugayevsky-Stein game seemed to be a draw. The prospect of a match with Leonid Stein was not especially appealing, and so it was decided to make every effort to avoid It. The opening most appropriate to this aim was the sharp Paulsen Variation (known in the West as the Taimanov Variation! — Translator), the favourite weapon of my 'charge' in the Sicilian Defence. As expected, on his first move Lutikov advanced his e-pawn two squares. I M c5 7 Bd3 b5

2 Nf3 Nc6

The decisive game had begun,

3 dA cxd4

A NxdA e6

5 Nc3 0c7

G Be3 06

And so, everything was going according lo plan — the Pautsen Variation h$s been (alayed. But at this point, as I now recall, grandmaster Bronstein said to his neighbours in the press box: "Lutikov is a great expert in this variation; he has scored a number of beautiful wins with i t " . 1 cannot say that these words did much to reassure me, to say nothing of Zorin. The only consoi ing factor was that in this opening Taimanov too had won a number of spectacular encounters. Instead of the advance of the b-pawn, 7.,.Nf6 is more often met. But this had been played several times against Lutikov, and in particular, we were farnillar with his game against Tal, which had continued 7 . . . Nf6 8 Qd2, etc. Taimanov therefore prefers a more unusual continuation. 8 Nxc6 Qxc6

9 Bd4

The idea of this move is to tie I ay somewhat the development of Slack's K-side. Bronstein once played simi lariy against Taimanov {with the bishop at e2) > when Black contrived to blunder away a pawn: 9 . . . f 6 10 0-0 Bc5 11 8*b5! axb5 12 Qh5+. On this occasion he was on the alert. 9 . . . Bb7

10 Qe2 Ne7

A flexible move; subsequently the knight will have a choice between the squares c6 and g6. In the event of 1O...Nf6 11 Bxf6 gxf6 the advance of the white knight to dS has to be reckoned w i t h . II

f4

The f i r s t

of

a series

of

moves which

impose on White

special

6th Letter — Two Decisive Games obi i gat ions. However, H 0-0 by 11. . . N f S ! . 11 . . .

b4

11 0-0-0 can be well

answered by

67

1 T . . . b4,

and

12 Nb1

12 Ndl would seem to leave White more possibilities. 12 . . .

Ng6

13 QfZ

What more, would it seem, can one expect of one move — the queen defends the f4 p a w n , strengthen? White's control of the g l - a 7 d i a g o n a l , and creates t h r e a t s on the K-side. And even so, B l a c k ' s c o u n t e r - p l a y now comes d i s t i n c t l y to l i g h t . 13 . . .

Bd6!

One of those b r i l l i a n t , p u r e l y T a i m a n c v - N k e moves. On 14 Bxg7 there follows 1 4 . . . N x f 4 15 Bxha Nxd3+ 16 cxd3 0c1 + . Since 14 g3 is too r i s k y , and 14 e5 is u n p l e a s a n t l y answered by 14, . -Be7 w i t h the threat of . . . B h 4 , White's next move is e s s e n t i a l l y f o r c e d . 14 Be3 0-0

15 Nd2 Rac8

16 h4

It is c u r i o u s t h a t , at t h i s p o i n t , it seemed to almost a l l the spectators, even to the chess masters, that While's threats were extremely dangerous, whereas f o r Taimanov, as he admitted after the game, it was at t h i s very point that he sensed that he could go through to the Interzonal Tournament without any a d d i t i o n a l event. White takes upon himself o b l i g a t i o n s w h i c h are too g r e a t , and for Which it w i l l be d i f f i c u l t f o r him to p a y . 16

Qc7

16.. .e5 17 g3 exfft IS gxffi looked t e m p t i n g , but the move p l a y e d 'S the most f l e x i b l e . Since 17 f5 is now impossible due to 17. . .Sg3, White is p r a c t i c a l l y forced to p l a y e i t h e r 17 g3, or J-? e5, a f t e r which

68

I

Seven Chess Letters

the strength of the bishop a! b? grows 17 e5 Bc5 20 0-0-0 is In great danger. Z0 . . . Nf 5

18 h5 Bxe3 unpleasantly

consider-ably,

19 Qxe3 Ne? met by

20 Nc4

2O...Nd5,

when

the white king

is

21 Qd2

On 21 Bxf5 Taimanov had prepared not 21...Qxc4, as we in the h a l l assumed, but 2 1 . . ,exf5 22 Nd6 Qxc2 23 Nxc8 (23 Nxb7 Qxg2) 23...Rxc8, s a c r i f i c i n g the exchange, but gaining for it more lhan sufficient compensation. He was no doubt r i g h t — large-scale strategy is required of the Sicilian p l a y e r . 21 . . . Bd5

Black,

22 Ne3 Nxe3

23 Qxe3 Qc5

The exchange of queens would have but 24 Qd2 w a j less r i s k y . 24

h6

24 Qg3

led lo a marked advantage for

25 Rh4

The c r i t i c a l point. After- 2S...Kh8 26 Rg4 Rg8 followed by . . .Qd4 Black would have gained a v i r t u a l l y overwhelming positional advantage. But Taimanov impulsively checks with his queen, assuming that it is always good to d r i v e the king into the centre. But things turn out to be much more complicated, 25 . . . Qgit After this move gains in b e a u t y . . . .

the game loses in strategic completeness, Some totally s t a r t l i n g events now set i n .

but

26 Kd2 The game enters a phase of boundless complications, the f i n a l consequences of which were impossible to foresee. It was subsequently stated that 26 Ke2 would have won for White, ana given Stein, who, es expected, had already drawn his game, a place in the Interzonal Tournament. But analysis showed that this place would at best have been acquired as a result of an additional match. After 26.. .Rxc2+ 27 Bxc2 Bc6+ 28 Kd2 128 Kf3 Qxal 29 RgA Qf1 + , 3O...Qe2+ and 3 1 . . . Qxc2, and White has only one check at g7) 28...Gd4+ 29 Kci Qg1 + Black is assured of perpetual check (30 Bdl Rc8). A fascinating battle also develops in the event of 2 6 . . .Qd4 27 f5 Rxc2+ 28 Bxc2 Qxb2 29 f6 Qxc2+ 30 Kfl g5 31 hx g 6 f x g 6 32 Rxh6 Be4.

6th Letter — Two Decisive Games

26

69

QdA

The capture of the rook was not very pleasant, since after 27 Rg4 White gives mate, while after 26...Qxg2+ 27 Qxg2 Bxg2 2B Rgi Bd5 29 Rhg4 the black king Is again in an unenviable position. Now the attack 27 Rg4 can be parried, thanks to 27...Be4! 28 Rxg7+ KhB, when against the various threats (and in particular ...RgB —along the same f i l e where White has been attacking!) there la no defence. 27 f5

With his queen attacked, and the further advance of threatened, Black appears to be on the edge of the abyss.

27

the f-pawn

Rxc2+!

Making such a move is pleasant not only for the p l a y e r d u r i n g the game, but also f o r the commentator in the course of h i s a n a l y s i s . 28 Kxc2 b3+! The b-pawn must 30 Rbi is i n s u f f i c i e n t .

be brought

into

play;

28...RcB+

29 Kdi

Qxb2

29 Kdi After maKing this move, White loses control of events. The dangerous pawn should have been taken — after 29 axb3 Bxb3+ 30 Kd2 (30 Kxb3 f a i l s to 3O...Rb8+ 31 Kc2 Qxb2+ 32 Kd1 Qxa1+ 33 Ke2 Rb2+ 34 Ke3 Q c H , and now either 35 Ke4 Rb4+ 36 Kf3 Qd1+ 37 Ke3 Qxd3+! 33 Kxd3 Rb3+, or 35 Kf3 Qd1+ 36 Kf4 Rb4+ 37 Be4 Qfl+ 38 Qf3 Rxe4+! 39 Kxe4 Qc4+, in each case with a won ending for Black) 3O...Qxb2+ 31 Ke3 Qxal 32 f6 Q g H 33 Kd2 Qd1+ Black c e r t a i n l y has a perpetual check, but a p p a r e n t l y nothing more. This would probably have been the n a t u r a l result f o r such a dramatic game. 29 . . . Q g H

30 Gel Qxg2

31 Qfi

Black is p l a y i n g , as If n o t h i n g had happened, fl cook down! 31 Rf4 Taimanov was intending to continue q u i e t l y with 31...Rc8. 31 . . . Bf3+ 36 Kg3 fig21

32 Kei Qxb2

33 Rbi OxeS+

(see next A move of great

aesthetic

force!

34 Kf2 bxa2

On

35 Ret Qf6

diagram) It

displays

the p u r e l y

problem-

Seven Chess Letters

like idea of over-loading k i n g , nor the queen.

— the bishop

can be captured

neither

by

the

37 Qg1 exfS! Again a totally unexpected, quiet move. One of masters asked in p e r p l e x i t y ; "Why separate the p a w n s ? . . . " 38 Qd4 Qg5+

the

grand-

39 Kh2

The f i r s t point of the move 37...exf5 — here 39 Rg4 is impossible. And now its second point is revealed. After the idea of over-loading comes that of interference. Sheer geometry... . 39 . . . Be4!

••• .i1

•->.-•>"

The f i n a l chord of this musical poem, played, as one of the admiring spectators remarked, in the style of an heroic symphony. The co-ordination of the white pieces is totally destroyed, and Black threatens ...Qxhi>+ and ...Qg2 mate. 40 Rhxe4 fxeft

41 0xe4

IT

'

'

In this position Taimanov wrote on his score sheet the move fit... Qxh5+, and went behind the scenes. With shaking hands he l i t up the f i r s t cigarette in his l i f e . On the face of Mikhail Tal, standing beside him, was depicted envy, which could not be erased even by the words addressed to him by Taimanov: "Misha, today — I am your p u p i l ! " Paul Keres, made wise by experience, came up to him. "Mark — today you played like Liszt", was how he addressed his old colleague, and a well-known p i a n i s t . Thus, the game was adjourned. In prospect was some night-time analysis, although it is true that it d i d not pr-omise to be d i f f i c u l t . After the white k i n g moves out of check there would have followed . ..g7-g6 and . . . R b 8 . Obviously this seemed to grandmaster Lutikov to be sufficiently clear, and he nobly decided to same Taimanov any

6th Letter — Two Decisive Games further

71

anxiety.

41 . . . Qxh5+

White resigns

A f a i r y t a l e game! It was a worthy culmination to the Championship of l h e country! The ha 11 thundered with applause, and on the stairs friends embraced Mark Taimanov, who that day was the happiest person a l i v e . Some ten years later, in the course of working on the present book, one of its authors, the one who had been Taimanov's second, met the grandmaster for 'an evening of pleasant reminiscences'. "Mark Yevgenyevich, your decisive encounter with Lutikov was called by j o u r n a l i s t s at the time the 'game of your l i f e ' . Can you now confirm the correctness of this epithet?" "Of course, ten years ago these words — 'the game of my l i f e 1 — were merely a t i t l e . That is not, of course, how a player t h i n k s . But what is most s u r p r i s i n g is that now, judging events in retrospect, I see that i t was this game, l i k e no other, which s i g n i f i c a n t l y influenced my subsequent l i f e , and, therefore, f u l l y j u s t i f i e d its name. In the end, victory in i t raised me to the heights of my competitive achievements, and at the same time brought me the deepest disappointments. If events are analysed f a t a l i s t r e a l l y , it can be said that this game subsequently led me to a series of dramatic changes in my l i f e , and not only in the chess aspect of i t . Everything here was interlaced — chess, my piano p l a y i n g , my personal l i f e . . . . In my everyday concerns I have not returned so often to this 'game of my l i f e ' , but now you have reminded me, and you see what a great deal it has s t i r r e d op". "Do you think that the psychological tension, which a g r a n d master inevitably experiences in a decisive game, is an aid to chess creativity?" "Yes, I think so. During that distant evening, at any rate, I experienced an unusual enthusiasm, a state rather of s p i r i t u a l i t y , of Illumination. It showed itself both in that, by g i v i n g the impulsive queen check at g i , I switched the play into an almost incontrol I able state, and in that I then made v i r t u a l l y fantastic efforts to turn the i r r a t i o n a l battle onto the desired l i n e s . . . . Yes, it was a truly euphoric state, when fantasy works at the limit of its p o s s i b i l i t i e s , and you have the feeling that you are hovering over events. It is perhaps 3t such moments that discoveries are made. Chess history knows of decisive games which were not distinguished for their especially r i c h content. But the encounter with Lutikov also proved highly interesting in the purely chess sense. It is no accident that it was remembered 'or a long time by a l l the spectators who were present in the hall that day. 1 ts content, its aesthetic element are perfectly equivalent, so lo speak, to its competitive v a l u e . " "The 'game of your l i f e ' led you in the end to a match with Fischer, the result of which was p i t i f u l . . . " . "Yes, this encounter brought me much disappointment, but even so, I do not regret that it took place. Robert Fischer is an outstanding phenomenon in chess history, and I consider myself fortunate to have Played with him an entire match, even though the result was so b a d . I " the creative sense the games were interesting, and I am not ashamed of them." "How do you explain Fischer's exit from the chess world?" "On this point I have a completely firm o p i n i o n . For the majority of the eleventh World Champion's predecessors, the game of chess was not only not the only thing in their lives, but sometimes not even the

72

Seven Chess Letters

main thing. But for Fischer, chess was, without a doubt, the meaning of his life, and therefore all his criteria were of an exclusively chess nature. Fischer, 11 could be said, was closelv linked to chess; it was his atmosphere, the only one he could breathe. This is why the title of World Champion meant more for Fischer than the recognition of his competitive merits alone. The supreme title as though assigned to him the role of a chess Messiah on Earth. If chess is the main, indeed the only value in life, and he is the chess k i n g , then in i t he must be a kind of infallible absolute. This, apparently, is what Fischer thought. And indeed, to judge by his actions following his victory over Spassky, the new Champion took upon himself numerous different obligations. The American grandmaster decided that he had no right to make incorrect pronouncements about chess, and he withdrew all the books written by him, very good books, incidentally, thinking that they were inappropriate to his new status. He considered that he did not have the right to play badly in a tournament, or even to lose a single game. In short, Fischer took upon himself such a burden that he simply bent under Us weight. On sensing, most probably, that he could not be the person he should be in his own imagination, Fischer began avoiding chess. And the more time that passed, the stronger that this psychological complex developed in him. The result was that the passionate and incomparable love forchess, characteristic of Fischer in his earlier years, gave way to a sense of fear, and a fear hot of some specific player, but of the very chess pieces and the chess board. For the average chess player It is, of course, difficult to understand why the chess king, at the height of his powers, should abandon his kingdom. But an exceptional personality may have a different outlook, which, although abnormal, also deserves respect... . "

From the World Champion:

-

'

'

I am in total agreement with the description given of my predecessor by grandmaster Mark Taimanov. Fischer is undoubtedly an exceptional person and an outstanding chess player. Six years ago I made intensive preparations for a match with him, realizing that I was faced with a very difficult struggle in which Fischer's chances wei-e objectively better. Nevertheless, as any player would, I hoped for success. Alas, Fischer conceded his title to me voluntarily. Not once in my life have I sat down at the board against the American grandmaster-, and I can only envy Taimanov, Petrosian and 5passky, who have each played a match with him. I must confess that, when the FIDE President Max Euwe invested me with the Champion's wreath, I experienced a twofold sensation. On the one hand, and 1 will not conceal the fact, there was joy, but at the same time there appeared an irresistible urge to demonstrate to the chess world that I had been declared the strongest chess player by rights. And there began for me a long-range duel with my predecessor. Perhaps the reader will be interested in certain comparisons... . Fischer in his career did not lose a single match; as yet, neither have I . Fischer gained the title of World Champion by his victor-y over Boris Spassky, and after eleven games of this match he was leading by 7-A (I have disregarded the loss resulting from his failure to appear for the second game). Two years later I won a Candidates' Semi-Final Match against Spassky by the score of . . . 7-4! (And the overall score in games with the 10th World Champion, where (here has been a definite result, is currently 9-1 in my favour).

73

6th Letter — Two Decisive Games

Fischer won two matches by the score of 6-0. Here it is d i f f i c u l t for me to compete with him — I am not such a maximalist. However, j n the European Team Championship in 1977, playing on the top board, concluded my appearance with a score of 5-0! Judging from Fischer's book My GO Memorable Games, he won a >tal of 25 tournaments (including those where he shared f i r s t pi ace, ut excluding team events). At the age of 29 Fischer gave up chess. Now, when this book is being written, I am 29 years old, and in 1979 the Tournament of Stars in Montreal I won my 25th tournament. Bt Such is the arithmetic. Can rny duel with Fischer be regarded as ^finished? It is not for me to j u d g e . . . . It siands to reason that the most Important and crucial event in •which I have yet participated was my as yet only match for the World Championship, the match in Baguia with Viktor Korchnoi. I think that e l l the reversals of fortune associated with this match will s t i l l be fresh in the memories of chess enthusiasts, and there is no point here in dwelling on them in d e t a i l . The match had already been in progress for more than two months, and things were going f a i r l y well. After the 27th game the score in wins became 5-2 (the winner was the f i r s t to score six wins), and the match appeared to be coming to an end. 1 had no doubt at all that 1 would win, and this over-confidence clearly weakened me. In addition, the enormous, purely physical fatigue was beginning to tell. And then, when there remained only one step to overall victory, -the unforeseen occurred. Korchnoi, who was possibly reconciled to ftJefeat, began playing unrestrainedly and by desperate efforts levelled the score —5-5! Within four games 1 had lost that which I had gained in 27! There was plenty to be distressed over. But nevertheless I succeeded in casting off the burden of my vexing oversights, ^endeavoured to rest well, and geared myself up for a decisive battle In he 32nd game. In this game I played calmly and confidently, and, on taining an overwhelming position, said to myself: "Just don't be loo , remembering that a desire to r e a l i i e an advantage as quickly jas possible had let me down more than once in the match. Victory in the 32nd, deciding game enabled me to retain the t i t l e of World •Champion.

Last game of the Match for the World Championship Baguio, 1978 A. Karpov - V. Korchnoi Pirc-Uf imlsev Defence e4 d6

2 d4 Nf6

3 Nc3 g6

4 Nf3 Bg7

5 Be2 0-0

6 0-0 c5

The choice of this or that opening in the match was made mainly for psychological reasons. The main events normal ly developed in the subsequent stages, and it was they that exerted a decisive influence on the outcome of the majority of the games. The present game is a good confirmation of t h i s , Korchnoi Is true to his favourite tactics — that of changing as often as possible the chess pattern. In the 18th game, where the Pirc-Uflmtsev Defence was also played, the discussion developed along 'he lines of the main v a r i a t i o n : 6...Bg4 7 8e3 Nc6. Then I had employed an innovation; 8 Qd3!? On this occasion, evidently so as to avoid such surprises, Korchnoi makes a sharp change of course. In Principle the entire variation after 7 dnc5 dxc5 is assessed by theory a s slightly favourable for White. Bui I could assume that the c h a l lenger had prepared some forcing improvement in Black's p l a y , and So,

L

74

Seven Chess Letters

quite n a t u r a l l y ,

I avoided the exchanging continuation.

7 d5 Na6 Black pins his hopes on flank operations. It is d i f f i c u l t to say which of the two plans - this oi- the one Involving ihe undermining move ...e7-e6 — should be preferred. 8 BH White deploys breakthrough e4-e5. a..,Nc7

his

9 a* b6

forces

so

10 ftel

as

Bb7

to

prepare

the

natural

central

11 Bc4

The prophylactic 11 h3, with the idea of retaining the bishop on the h2-bB diagonal, could have proved to be a serious delay. After 11.,.Qd7! 12 Bc4 Rad9! 13 Qd3 e5 very sharp play develops. 11 . . A waste of valuable time. Instead of standing stil I, it was essential to play 11.,.Qd7, aiming after 12 e5 dxe5 13 Nxe5 Qf5 to provoke tactical complications at any cost. For example, 1ft Nxg6 fxg6 15 Bxc7 Ng4. Evidently on 11...Qd7 White should reply 12 Qd3, maintaining si ightly the better changes, but on the whole a very tense struggle wouild have arisen. It seems to me that, on this occasion,

Korchnoi 's predilection 12 BgS Nhf6 Now

all

for

solid

positions

did

him a

bad service.

13 Qd3 directions

it

Is

much

more

difficult

for

Black

to

Obtain 13 . . . a6

14 Rad1 Rb8

IS h3

At this Dolnt I decided not to take a r i s k . In p r i n c i p l e , the consistent continuation, which would have increased White's spatial advantage and increased his Bressijre, was 15 e5!? For example, 15...dxe5 16 Nxe5 b5?! 1? axb5 axE>5 18 Bxb5! NcxdS 19 Nd7 N«c3 20 bxc3 Nxd? 21 Bxe7, ana Sl«ck loses material. l S , . . N c i d 5 can also be well met by the simple 19 N*d5 Bxd5 (19...0>td5 20 Qg3 Qa2 21 Bcu QxO2 Zl Nxf7) 20 C4 Ba8 (2O...Be6 21 Nc6 QxiJ3 22 Nxe7+1 21 Nd7. In a diffei-enl situation I would have definitely played that way, bul in this game there was no way that I coulti r i s k White's entire initiative.

•1

6th

75

Letter — Two Decisive Games

15 . . . Nd7 Black cannot test his opponent's patience for long, and sensibly prevents the threatened breakthrough in the centre. in addition, he can now hope for counter-play of the type 16,,.b5 17 axb5 Nb6, or 16..-Ne5 17 Nxe5 dxe5 IB d6 exd6 19 Qxd6 Qxd6 20 Rxd6 b5.

16 Qe3 Occupying break. 16

Ba8

another

important

17 Bh6 b5

Without the fianchettoed Q-side resembles an infantry bombardment.

line,

and

18 Bxg7 Kxg7 bishop attack

preventing

19 Bf1 Nf6

the

...b6-b5

20 axb5 axb5

at g7, the pawn offensive on the without the appropriate a r t i l l e r y

21 Ne2 Bb7 Black's chief misfortune lies in his poor communications. If he were able to play ...e7-e5, the switching of his forces from one side of the board to the other would be much easier. But 21 . , .e5 (or . . .e6) leads after 22 dxe6 Nxe6 23 Ng3 to a significant weakening of his pawn formation. Therefore Black's reserves have lo cross from one wing to the other along the back rank, i . e . l i t e r a l l y 'on a l l f o u r s ' , and it is by no means all the chess pieces that are accustomed to this and able to do i t . i v , » . . •''•--.. ^ 22 Ng3 RaB

23 c3 Ra4

24 Bd3 QaB

„ .... .

Black disregards his opponent's threats. The queen retreats to Ihe edge of the board at the very point when the storm clouds are gathering over his king position. Korchnoi obviously under-esli mated the danger of White's threats.

25 e5! A d a g g e r b l o w , r e i n f o r c e d by a p p r o p r i a t e a r g u m e n t s : 2 5 . . -Nfxd5 26 NhS+ (or 26 Nf5+) 26.,.gxh5 (26...«hB 27 Qh6 Rg8 28 Ng5) 27 Qg5+ KhS 28 Qf5. ... . 25

dxe5

26 Gxe5

26 0xc5 would have been a mistake • he white queen is in Immediate danger.

because of

26.. .Ncxd5,

when

Seven Chess Letlers

76

26 . . . Ncxd5 The only temporarily.

27 Bxb5 Ra7

way of maintaining

the material

balance,

even if only

28 Nh4

After this manoeuvre there are threats impending not only over the black k i n g . In many cases the prosaic c3-cft is no less unpleasant. Thus on 2B...Bc6 there can follows 29 Bxc6 Qxc6 30 c4 Nb4 31 Rd6 exd6 32 Nh5+ gxhS 33 Qg5+ Kh8 3ft Qxf6+ KgS 35 Nf5, while 28...Qb9 29 cii Qxe5 30 Rxe5 leads to a hopeless ending for Black. 28

BcB

29 Be2!

I think you will agree that there is something attractive in this bishop move. Henceforward a l l the white pieces will act with co-ordination and precision. After the game I was told that It was this move which brought such long-awaited oeace to my colleagues' troubled sou Is. 29 . . . Be6 30 c4 Nbi 34 Rd2 Nc6 35 Qh6!

31 QxcS QbB

32 Bfi RcS

33 Qg5 Kh8

It turns out that White has not only an extra pawn, but a very strong i n i t i a t i v e into the bargain. Black is torn apart in the direct sense of the words; he has to parry the attack on his k i n g , and keep a careful watch on the Q-side pawns, 35 . . . RgS 36 Nf3 Qf8 37 Qe3 Kg7? The decisive mistake, after which Black's resistance becomes hopeless. He could still have prolonged the game by 37...Rb7, preventing the advance of the white pawns. 38 NgS Bd7

39 t>4 QaB

6th Letter — Two Decisive Games Black's

time-trouble agony draws to a close.

77 Already nothing can

help 40 b5 Na5

41 b6

The saying 'All roads lead to Rome' is p a r t i c u l a r l y appropriate to this p o s i t i o n . Korchnoi sealed 41 . . . Rb7, but the following day his second announced his resignation of the game and the match.

Seventh Letter BEFORE THE 30TH MATCH FOR THE WORLD CHAMPIONSHIP

In this year, as you know, the current cycle in the battle for the World Championship is completed. Interest in chess in Ihis year increases exceptionally, and in the summer the attention of a l l chess enthusiasts will be focused on the Match for the World Championship. Since the publication of our book is timed to coincide with the start of this ' j u b i l e e ' match, we have decided that one of its ' letters' should definitely be devoted to the history of competitions for the title of the strongest chess player in the world. History knows of numerous 'uncrowned kings' — the strongest players of their time, the unofficial World Champions —Greco, Phllidor, La Bourdonnais, Staunton, Ancierssen and Morphy. In the I8BO5 the chess world decided that it was time thai they had a ' r e a l ' k i n g , and in 1886, afler his victory over Zukertort, the f i r s t official World Champion was declared to be Wilhelm Steinitz. All the matches for- the world crown are given by us in a table (see next page). Here you will find the names of the participants in each match (the winner, or, in the event of a drawn result, the World Champion retaining his t i t l e , is given f i r s t ) , the year when it was held, the number of wins, draws and defeats and the overall store, and the venue of the match. The f i r s t seventeen matches were not held under any system — the World Champion played when he wanled, with whom he wanted (and not always with his strongest r i v a l ! ) and under the conditions which he liked best. The very f i r s t duel was due to conclude when one side had gained ten wins, with 3 draw to be recorded if the score stood at 9-9, but during the course of the match the contestants agreed in this event to extend the marathon up to a further eight wins. Other matches were held either up to a definite number of wins, or for a majority of the points. Thus the battle between Capablonca and Alekhlne was up to six victories (with a score of 5-5 the World Champion was to retain his t i t l e ) , whereas in all four of Alekhine's subsequent World Championship Matches, t h i r t y games were planned, and 15j points was enough to secure victory. The most enigmatic duel is that between Lasker and Schlechter. There is even the hypothesis (though it has not been proved!) that this wasn't a match for the World Championship el a l l . In certain sources it is indicated that to gain the Champion's title it was s u f f i 78

7th L e t t e r — Before Ihe 30th Match f o r i h e World Championship

79

THIRTY MATCHES FOR THE WORLD CHAMPIONSHIP | . Steini tz-Zukertort 2. 3. 4. 5. 6. 7. 8.

USA Havana New York Havana USA Moscow USA Dusseldorf,

Steinit7-Chigorin Steinitz-Gunsberg Stelnitz-Chigortn Lasker-Steinlt2 Lasker-Stelni tz Lasker-Marshall Lasker-Tarrasch

Mu n i c h 9. 10. 11. 12. 13. 14-

Lasker-Janowski Lasker-Schlechter Lasker-Janoivski Capablanca-Lasker Alekhine-Capablanca Alekhine-Bogoljubov

1909 1910 1910 1921 1927 1929

+ 7 - I =2 + 1 - 1 =8

+ 4 - 0 =10 (9-5) + 6 - 3 =25 (18i-15^) +11 - 5 =9 (15|-gi)

Paris Vienna, Berl in Berl i n Havana Buenos Aires Germany,

15. 16. 17. 18.

Alekhine-Bogoljubov Euwe-Alekhine Alekhine-Euwe Match-Tournament of five grandmasters 1) Botvinnik 2} Smyslov 3-4) Keres 3-4) Reshevsky 5) Euwe

1934 1935 1937

+ 8 - 3 =15 (15i-1 + 9 - 8 =13 (15*-l4) + 10 -4 -11 (1Si-9i)

Holland Germany Holland Holland

1948

Botvinnlk-Bronstein Botvinnik-Smyslov Smyslov-Botvinnik Botvinnik-Smyslov Tal-Bolvinnik 8otvinnik-Tal Petroslan-Botvinnik Petrosian-Spassky Spassky-Petr-osian Fischer-Spassky Karpov-Korchnoi Karpov-Korchnoi

1951 1954 1957 19SQ 1960 1961 1963 1966 1969 1972 1978 1981

The Hague, 14 ( + 10 -2 =8) Moscow 11 (2-3 against Botvinnik) 10T (1-4 against Botvinnik) IOJ (1^-3-j against Botvinnik) 4 (1^-35 against Botvinnik) + 5 - 5 = 1 4 • •"• 12-12) Moscow Moscow + 7 - 7 =10 (12-12) Moscow + 6 - 3 13 (12^-9Jj) + 7 -5 =11 ilzl-10) Moscow + 6 -2 =13 |12j-Bi> Moscow Moscow + 10 - 5 =6 (13-8) Moscow + 5 -2 =15 ( 1 2 i - E . . + 4 -3 =17 (125-11?) Moscow + 6 - 4 =13 ( 1 2 i - 1 0 | ) Moscow R e y yj kjavik i ^ + 7 - 3 =11 + 6 - 5 =21 ( 1 6 j - 1 5 i ) Baguio

19. 20. 21. 22. 23. 24. 25. 26. 27. 26. 29. 30.

(8-2) (5-5)

+ 8 -0 =3 (9i-1sI

cient f o r Schlechter to g a i n 5^ points out of 10. But we nevertheless adhere to the generally-accepted point of view, which is t h a t , f i r s t l y . the chess crown was being played f o r i n the match, and t h a t , secondly, the c h a l l e n g e r h a d to w i n by a margin of two p o i n t s . After the death of Alekhine In 1946, the chess w o r l d was, f o r t h e f i r s t time, without a champion. For this reason, under number IB i n the lable there appears a match-tournament, i n which the strongest grandmasters of that time determined their new leader. A l l subsequent matches have been held under the aegis of the I n t e r n a t i o n a l Chess Federation (F IDE), a n d the World Champion has been d e p r i v e d of the P o s s i b i l i t y of himself choosing h i s opponents. From t h i s point onwards a s t r i c t system of e l i m i n a t i o n events every three years determines a new chal lenger. Since then, f o r almost t h i r t y y e a r s , matches were held on 'he best of 24 games. At the same time, i n the f i r s t four cycles t h e World Champion h a d the r i g h t to a / f ^ t u r n match, and f o r t h i s reason

80

Seven Chess Letters

two 'additional ' matches were played. In the table the year 5975 is 'missing 1 , since it was in this year that Fischer declined to defend his titleOf course, the Karpov-Korchnoi match of 1974, although In the end it gave the name of the new World Champion, is not included in our table — the winner of it could not have supposed that within a year the chess crown would be passed on to him without a struggle. In the 1978 match, as in the present one, the system employed has been different from the earlier one — play now proceeds up to six wins without limit on the number of games. In contrast to the AlekhineCapablanca duel, with the score at 5-S the Champion does not retain his t i t l e , and the battle continues. In the event of the loss of his crown, he has the right to a return match. Certain chess enthusiasts assume that this right is considerably more important for the World Champion than a reserve of one point in 'normal time'. However, as we will see, this is not altogether so. If the match in Baguio had been held under the old system, as Alekhine and Capablanca Rlayed half a century ago, it would all have been over without any trouble a who's five games e a r l i e r ! Indeed, after the World Champion gained his f i f t h victory in the 27th game, and the score became 5-2, the battle would in fact have come to an end! batches for the World Championship are undoubtedly highly important chess events. In the course of their preparations, and during the match itself, the participants make us of all the latest achievements of chess science and theory. On the other hand, the matches themselves exert an enormous influence on the subsequent development of chess a r t , and, one can say, chess phi losophy. The I i terature devoted to the majority of the matches is both numerous ana thorough, and it is 1 obvious that in one ' letter of our book it is impossible to dwell on all the reversals of fortune in the battle for tne world crown. We have therefore decided to illustrate each Match for the World Championship with just one extract from the play (with brief annotations, which do not pretend to be complete). This will be either a spectacular combination, or an unexpected tactical blow, or some dramatic episode from the match. It has to be saifl that the realization of this aim proved to be not at all easy. Sometimes, in choosing one single example out of the twenty to t h i r t y games in a match, our eyes became dazzled... . We should mention that, altogether in matches for the World Championship, more than 600 games have been played, and even a cur-sory examination of them took us a mass of time (although it also afforded us considerable pleasure). And so, we invite the reader to make a short history of competitions for the World Championship.

1.

excursion

into . •

the

Steinitz-Zukertort

Steini\z was unofficially recognized as the strongest player in the world in 1866 after his victory over Anderssen. Now, twenty years later, after winning a match against his outstanding contemporary Zukertort, 5teinit7 was declared the f i r s l official World Champion in history. After five games the future king was losing 1-4, but he then succeeded in changing the course of events, and gained nine wins with one one defeat. We give the conclusion of the penultimate game of the match, highly energetically conducted by the f i r s t chess k i n g .

7th Letter — Before the 30th Match for

the World Championship

Zukeriort-Steini tz 20 . . . Nd4! (threatening ...Nc2 or ...Ne2) 21 exd4 Qxd4+ 22 Khi ^3 Black has sacrificed a knight, but his pawns are irresistible. 23 Hc3 Bf6! 24 Ndbi d2! 25 Qc2 Bb3 26 Qxf5 d1=Q 27 Nxdi Bxdl" 28 Nc3 e2 29 Raxdi Qxc3 White resigns.

•'

2.

Steinilz-ChigorTn

'

'

'

•

While Stein i tz is r i g h t l y considered the founder of the positional school of ptay, ChigoHn was an acknowledged master of the combinational style. Thus the match between these two leading I ights of the past was in the nature of a crucial creative debate. In the entire history of the battle for the world crown there has not been such a 'bloody' match — only the last game In it ended peaceful l y . The approach to chess worked out by Steinltz triumphed, and the Champ iorj retained his t i t l e .

Stei ni tz-Chigorin In the opening of the fourth game Black has committed several Positional errors, and 5teinitz skilfully exploits them: 9 d5! exd5 10 a 3 ! Nd4 (fatal is 1O...Bxc3 !1 Bxc3 and Bxg.7, or 10. . .Bd6 11 Nxd5 n c5 12 Be3 Qa5+ 13 b4) 11 Bd3 0-0-0 12 axb4 Nxf3+ 13 Qxf3! Qxa1 + 14 Ke2 Qxb2 15 Rbl Qa3 16 Nb5 Qa6 17 Qxf7 Qb6 IS Rc1 Nh6 19 Qxg7 dxe4 20 Qxc7+ Qxc7 21 Rxc7+ Kb8 22 Bxe4 Resigns.

3.

Steinltz-Gunsberg

At the end of the 1880s Gonsberg won a series of major tournaments, played a drawn match with Chigorin. This gave him the grounds "°r throwing down the gauntlet to the Champion. Although Steinitz did n ° t achieve a big advantage in points, he retained his t i t l e f a i r l y confidently.

L

82

Seven Chess Letters Here is how the seventh game of

the match concluded.

Slei ni tz-Gunsberg 22 Rxf6! gxf6 (on 22...Qxf6 White wins by 23 d7 Red8 24 QxaB! RxaS 25 Rc8+ QdS 36 Rxa8 ClxaS 27 Be7) 23 d7 RgB 24 dxe5 RgS {24...fxe5 25 Bb2l 25 Qxa8! QxaS 26 Rc8+ Rg8 27 Rxa8 Rxa8 28 e6 Resigns. 4.

Steinitz-Chigorin

The second match between the World Champion and the great Russian player was much tenser than the f i r s t . After 23 games Steinitz was one point ahead, but in the event of Chigorln winning the next game the score would have become 9-9, whereupon, under the conditions of the match, play would have continued up to a further three wins, and everything would have been to play for. But a tragic incident occurred, the equal of which is unknown In the history of Chess.

Chi gor in-Steinitz White is a piece up, and after 32 Rxb7 (fol lowed by 32. . .Bti5 33 Rb3 Bf7 34 Nf4!, or 3 2 . . . f 4 33 Rxf4+ Bf5 34 Rfl Rxd5 35 Ng7!, or 32...h3 33 Bf4 Rf2 34 Bg5+L Ke5 35 Rel+ Rde2 36 Rxe2+) he would have won this decisive game. But White played the inconceivable 32 Bb4??, and after 32-..RxhZ+ the match concluded (33 Kg! Rdg2 mate).

5.

Lasker-Steinitz

Emanuel Lasker, the founder of the psychological approach to the chess struggle, was 32 years younger than his great predecessor, and this age difference proved too great in their match. On losing the match, Steinitz proclaimed 'three cheers' in honour of the second World Champion.

7th Letter -

Before (he 30th Match for

the World Championship

83

Lasker-Stein itz This is a position from the seventh game of the match. From the opening Black gained an enormous advantage, and in spite of his opponent's desperate efforts to confuse matters, the situation remained unchanged. Had he now advanced his Q-side pawns (35 Qh6 is not dangerous in view of 35...Rg3) Steinitz could have won and taken Che lead. But his nerves gave way, with his next move he made a decisive mistake, and he soon resigned the game — the f i r s t of a series of f i v e defeats. 35 . . . gxf5? 35 Qh5+ Ke7 36 Rg8 Kd6 37 Rxf5 Qe6 38 RxeB Qxefl 39 Rxf6+ Kc5 40 Qh6 Re7 41 Qh2 Qd7 (41...Qd8 42 Qf2+ and 43 Rf6) 42 Qg1 + d4 43 Gg5+ Qd5 44 Rf5 Qxf5 45 Qxf5+ Kd6 46 Qf6+ Resigns.

6.

Lasker-Steinltz

The f i r s t chess k i n g was quite unable to reconcile himself to ihe fact that he had been removed from the throne, and he chal lenged Lasker to a return match. In a l l the subsequent history of battles for the World Championship, there has not been an instance where a player of over 60 has taken p a r t . Steinitz suffered a crushing defeat. An interesting fact is that this match was held in Moscow, whi le the next event of such high rank took place in our capital only more than half a century later. In the second s a m e mating combination.

from

the

match

Lasker

carried

out

a

famous

Lasker-Steini tz 31 h.4 h5 (the capture of the bishop at f4 leads to a quick mate after Nf6+) 32 Bg5 Bd8 33 g4! hxg4 34 h5 Nf8 35 Nec5+! dxc5 36 Nxc5+ Kd6 (after 36...Kc7, 37 Bxe7 Bxe7 38 Rxe7+ Kb6 39 Rxg7 is sufficient to win, but now follows e forced mate) 37 Bf4+ Kd5 38 Re5+ Kc4 (38...Kd6 39 F)f5 mate; 38...Kxd4 39 R1e4 mate) 29 Rc1+ Kxd4 f39...Kb4 40 Bd2 mate) 40 Nb3+ Kd3 41 Re3 male. The concluding

84

Seven Chess Letters

mating formation is most spectacular. 7.

Lasker-MarshaM

Ten years a f t e r his second v i c t o r y over 5 t e i n l t 7 , the World Champion p l a y e d a new match f o r the World C h a m p i o n s h i p . His a d v a n t a g e was u n d i s p u t e d , ana he r e t a i n e d his t i t l e without d i f f i c u l t y . Here is the conclusion of the t h i r d game of the m a t c h .

Marshal I-Lasker

' -

'•'•

•'••'

3 8 . . .Nf3! 38 g x f 3 Qxh3+ 39 K g ! Qg3+ 40 K M RfS 41 Qd8+? (correct was 41 Qh5 Rh4+ itZ Q x h i l 41...Kh7 42 Bfi? (the f i n a l m i s take; there were s t i l l d r a w i n g chances a f t e r 42 Rc2 Rh4+ 43 Qxh4 Qxh4+ 44 Kg2) 42...Rf5I Now mate is i n e v i t a b l e (43 Qe8 Qh4* 44 Kg2 RgS mate), a n d White resigned.

B.

Losker-Tarrasch

For many years grandmaster Tarrasch was a creative opponent of Lasker. He sought absolute truth in chess, whereas Lasker would aim lo find the key to each of his opponents. He was also able easily to disarm Tarrasch... . It should be said that the match between these two outstanding players provoked unprecedented interest for that time. There was an elegant conclusion to the second game of the match.

Tarrasch-Lasker 40 . . . f 3 ! (4O...Bf2 41 Rxe6 Rxe6 42 Qd7+ would have led to perpetual check) 41 gxf3 Bg5 (now 42 Bxe6 Rxe6 43 Q67+ f a i l s to draw due to 43...Re7) White resigns.

7th Letter — Before Ihe 30th Match for 9.

the World Championship

85

Lasker-Janowski

In 1909 Janowski drew a demonstration match of four games with Lasker. This success provided the grounds for challenging Lasker to a match for the World Championship. However, the match brought Janowski nothing but disillusionment.

Janowski-Lasker White's position (from the second game of the match! appears solid, but as a result of a subtle knight manoeuvre Black's advantage becomes decisive: 24...Ng7! 25 c3 (despite its unaesthetic appearance, 25 d5 was more tenacious) 25...Ne6 26 Bfi (the threat was ...Ng5} 26...f5 27 R4g2 Rf6! 28 Bd3 g5! 29 Rht (Janowski could have immortalized this position in the event of 29 exf5 Qxh3+!! 30 Kxh3 Rh6+ 31 Kg4 Rh4 mate) 29...g4! 30 Be2 Ng5 (not wishing lo lose his queen after 30...Qxh3 31 Kgl ) 31 fxg4 f3 32 Rg3 fxe2 White resigns.

10.

Lasker-Schlechter

Initially it was planned that this match would consist of t h i r t y games, but due to lack of funds it had lo be restricted to ten. Apparently, taking account of the shori distance, Lasker set the condition that, for overall v i c t o r y , the challenger needed an advantage of two points. Before the last game Schlechter was leading by one point, and he was also close to victory in the concluding encounter... . As promised, we give (with brief comments) the tenth, decisive game of this match. Lasker-Schlechter Slav Defence 1 d4 dS 2 c* c6 3 Nf3 Nf6 4 e3 g6 (this opening set-up now bears the name of the Schlechter Defence) 5 Nc3 Bg7 6 Bd3 0-0 7 Qc2 Na6 8 a3 dxc4 9 Bxc4 b5 10 Bd3 b4 I I Na4 bxa3 12 bxa3 (not 12 0.xc6 Nb4 13 Qxa8 Nxd3+ 14 Ke2 Nxc1 + 15 R h x d Ba6+, or 14 Kd2 Nxf2, with a winning position for Black) 12...Bb7 13 Rb1 Oc7 14 Ne5 Nh5 (White has played the opening q u i e t l y , and has a clear advantage; Schlechter tries to complicate the game) 15 g4 (Lasker f a l l s In with his opponent's wishes; the simple 15 0-0 would have retained for him = persistent advantage) 15...Bxe5 16 gxh5 Bg7 17 hxg6 hxg6 18 Qc4 Bc8 19 Rgi (with the threat of Rxg6) 19...0a5+ 20 Bd2 Qd5 21 Rcl Bb7 22 Oc2. A draw in the game would have retained for Lasker his Champion's t i t l e , although he would have lost the matchl "e is obliged to play f ° r a win on prestige grounds, otherwise it is d i f f i c u l t to explain why tie avoids the exchange of queens, after which

86

Seven Chess Letters

he certainly did not risk losing. 22-..Qh5 23 Bxg6? (provoked by anxiety, this opening of the f - f i l e Is In Black's favour; later the World Champion gave this v a r i a t i o n : 23 Rbi Qxh2 24 Rf 1 Qc? 25 6x96! fxg6 26 Qb3+, with a clear advantage to White) 23...Qxh2 24 Rfl fxg6 25 Ob3+ Rf7 26 Qxb7 RafB! (in is reply was presumably overlooked by White; the capture of the knight is decisively met by ...Rxf2) 27 Qb3 Kh8 28 f4 g5! 29 Qd3 gxf4 30 exH QhA+ 31 Ke2 Qh2+ 32 Rf2 Qh5+ 33 Rf3 Nc7 34 Rxc6 Nb5! 35 Rc4.

Lasker-Schlechter

"

"

'•-•(••'

'• '••

•"

The c r i t i c a l position of this historic game. After 35...Rd8 36 Be3 e5! 37 d5 Nd6!, or 37 Rc5 Nxd4+ 38 Bxd4 (38 0xd4 Qxf3+| 38...Rxf4 Black would have retained every chance of winning;. Schlechter unexpectedly sacrifices the exchange, losing his last chance to become the t h i r d Champion of the World. 35...Rxf4? 36 Bxf4 Rxf4 37 Rc&+ BfB 38 Kf2! Qh2+ (after 3 6 . . . Qh4+ 39 Kg2 Qg4+ 40 Rg3 Qxc8 41 Qg6 the black k i n g Is mated) 39 Kei Qhl-v? (Black no longer had a w i n , but now he also misses a draw, which could have been attained by 39...Oh4+ 40 Kd2 Qh2+ 41 Ke3 Rxf3+ 42 Kxf3 Qh3+ 43 Ke2 Qxc8) 40 Rfl Qh4+ 41 Kd2 Rxfi A2 Qxfi Qxd4+ 43 Qd3 0f2+ 44 Kdi Nd6 45 Rc5. The remainder of the game involves White r e a l i z i n g h i s advantage of the exchange. Lasker copes successfully with this problem, and levels the score in the match. 45...Bh6 46 Rd5 Kg8 47 Nc5 0g1+ 48 Kc2 Qcl+ 49 Kb3 Bg7 50 Ne6 Ob2+ 51 Ka4 Kf7 52 Nxg7 Qxg7 S3 Qb3! Ke8 54 Qb8+ Kf7 55 Qxa7 Qg4+ 56 Qd4 Qd7+ 57 Kb3 Qb7+ 58 Ka2 Qc6 59 0d3 Ke6 60 Rg5 Kd7 61 Re5 Qg2+ 62 Re2 Qg4 63 Rd2 Qa4 64 Qf5+ Kc7 65 Qc2+ Qxc2 66 Rxc2+ Kb6 67 Re2 Nc8 68 Kb3 Kc6 69 Rc2+ Kb? 70 Kb4 Na7 71 Kc5 Resigns.

11.

Lasker-JanowskI

The defeat in his first attempt did not convince Janowski of the f u t i l i t y of his hopes, and he challenged Lasker to a new match. On this occasion the World Champion's victory was even more impressive. The following position, taken from the f i f t h game of the match, Is sometimes cited as an Illustration of Lasker's psychological method. He often chose continuations which were r i s k y , and objectively not the best, but in doing so took account of the fact that they would not be to the l i k i n g of that particular opponent.

7th Lelter - Before the 30th Match for

(he World Championship

87

11

L ask er-Ja n o w sk 1 Black has a marked advantage, and after the fine queen sacrifice 17.. .Qxc3+! 18 Nxc3 Nxd4! he would have obtained for it sufficient compensation, and, in addition, a very dangerous attack. But Janowski displays indecision, and the game concludes quite differently. 17...Bh4+? 18 g3 Qe4 19 0-0 Bf6 20 Rxf6! Now Black's position is already c r i t i c a l . 2O...gxf6 21 Bf3 Qe5 22 Nxa7+ Kc7 23 Naxc6 bxc6 24 Rxc6+ KbB 25 Rb6+ Kc8 26 Qc14 Kd7 27 Nxe6 fxe6 28 Rb7+ Ke8 29 Bc6+ Resigns.

12.

Capablanca-Lasker

Emanuel Lasker was Champion of the World for 27 years! In our stormy times, when there are so many wishing to ascend to the chess throne, it is d i f f i c u l t to imagine that anyone w i l l even approach this record. Capablanca challenged Lasker to a match as far back as 1911. The unbeaten champion, possibly sensing that the lime had come to concede Ihe crown, for ten years avoided meeting the great Cuban, and then even announced his abdication. But the chess world was thirsting for a spectacle. Although subsequently Lasker had several outstanding results, this last match went badly for him. Without playing on to the stipulated ten wins, he resigned the match on account of illness. Capablanca became the t h i r d Champion of ihe World. Evidence of Lasker's poor form is provided by the conclusion of the f i f t h game, in which he suffered his f i r s t defeat.

Cepablanca-Lasker Having successfully conducted a d i f f i c u l t defence, Black coulO now have drawn easily by playing ...Ke6 or- . . . K f 6 . But there followed *5...Kf8??, and after 46 Qba*-! Lasker resigned (A6...Kg7 47 QhB+, or 46...Ke7 kl Qe5+).

88

Seven Chess Letters 13.

Alekhine-Capablanca

The encounter between the two chess genii Aiekhine and Capablanca is to this day regarded as the most outstanding event in the history of chess. In the struggle for the world crown there has never been a match more prolonged (34 games!), or more tense (32 exhausting Queen's Gambits! ). But, most important, this duel saw the meeting of two chess giants, at the height of their powers. The i n v i n c i b l e Jose Raul Capablanca with his phenomenal technique, which enabled him to express the hypothesis aboul 'the drawing death of chess', and the great master of combination Alexander Alekhine, who with his irrepressible fantasy refuted this hypothesis! By gaining the six necessary victories in the match, Aiekhine not only became the fourth chess k i n g , but also demonstrated to the world the inexhaustible nature of chess. We give

the

conclusion

to

the

21st

game

of

this

•:

Capablanca-Alekhlne

(-••

historic

* n

•

match.

>:.

<*••

\

26.. .Bb2! The start of an unusual and s u r p r i s i n g combination. 11 turns out that the white rook has no good square. 27 Rei (27 Rbi Na3! 28 Qxb2 Nxbl 29 Qxbi Qb3! 30 Ofi bxa4 31 h3 a3, etc., or 27 Rdl bxa4! 28 Clxa4 Nb6 29 Rxd5 Nxa4 30 Rdl Nc3 31 Rel Rc4, with a winning position) 27...Rd8 28 axbS axb5 29 h3 e5 30 Rb1 e4 31 Nd4 (no better is 31 Nh2 Qd3! 32 RxbZ Qxb3 33 Rxb3 Rd1+ 34 Nf1 NdZ 35 Ra3 Nxfi - v a r i a t i o n by Lasker, or 31 Nei Qd2 32 Qc2 Oxc2 33 Nxc2 Rd2 34 Nei Na3 - v a r i a t i o n by Aiekhine) 31 . . .Bxd4 32 Rd1 Nxe3! A spectacular concluding blow. White resigns, in view of the v a r i a t i o n 33 Qxd5 Rxd5 3ft Rxd4 Rxd4 35 fxe3 Rxb4.

14. Alekhine-Bogoljubov Alekhine's f i r s t match as Champion was against Ewfim Bogoljubov. At that time Aiekhine had no equals in the chess world, a fact which was confirmed in the present encounter. The eighth game of the match ended in a pure mate.

Bogo I j ubo v -A I ek h i ne

7th Letler 26...Ng3+! Bh2 mate.

Before the 30th Match for the World Championship

27 hxg3 hxg3+

15.

28 Nh3 Bxh3

29 gxh3 Rxh3+

89

30 Kg2

Alekhine-Bogoljubov

The repeat match between the same opponents differed l i t t l e from the previous one. Alekhine's advantage was again undisputed, and he retained his t i tie. We give an episode from the sixteenth game.

Alekhine-Bogol jubov

•

' :•

With his last move 29.. .Rh8-g8 Black decided to drive away the knight (correct was 29...Ng5 with a complicated game), but the knight remains a! its post... . 30 e6!! (a beautiful combination, which refutes 8lack's manoeuvre) 3O...Rdxg7 31 Nxg7 Rxg7 32 Rxd5! (a further spectacular blow, on which the entire combination is based) 32...cxdS 33 Rf8+ Kc7 34 Rf7+ Kd6 lales, after the exchange of rooks the white pawn queens) 35 Rxg7 Kxe6 36 Rg6+ Ke5 37 Kg2 b5 3S a5! d4 39 Rxa6 b4 40 Kf3 c3 41 bxc3 bxc3 42 Re6+! (a highly prosaic finish) 42...Kxe6 43 Kxe4 Resigns.

16.

Euwe-Aiekhine

In his meeting with the Dutch player, Alekhine clearly underestimated his opponent, and as a result lost his crown for two years. Max Euwe became the f i f t h World Champion in history. The phase of the match between the 20th and 26th games went well for him — four wins and three draws! Here is how the f i r s t of these wins was achieved.

Euwe-Aiekhine

I

18 Ng5! fxgS

(!8...Bf5

19 Qb3-<- Kh8

20 Bxe5 Qxe5

21 Nf7+, and

90

Seven Chess Letters

so that the game should not conclude with the combination described by us in the third ' letter 1 , Black is forced to give up the exchange) 19 Bxe5 Bf6 20 Bxb8 Bxc3 21 Bd6 Rf7 22 bxc3 Rfd7 23 Rbl Rxd6 24 Rxb7 R8d7 25 Rxd7 Bxd7 26 Be4 c5 27 c4 Bxa4 28 Bd5+ Kf8 29 Rai Ra6 30 Ra2 Ke7 31 fU g x f i 32 gxf4 Kf6 33 e4 g5 3A f5 h5 35 h4! g xh4 36 KhZ Kg5 37 Kh3 Ra5 38 Bb7 Kf6 39 Bd5 Kg5 40 Bb7 Kf6 41 Bc6 Resigns. 17.

Alekhine-Euwe

The return match, which took place two years later, ended in a convincing victory for the great Russian player. Alekhine played in his best style and easily regained his crown. The sixth game d i d not last long. Alekhine-Euwe Slav Defence 1 d4 d5 2 c4 c6 3 Nc3 dxe4 4 e4 e5 S Bxc4 exd4. This v a r i ation does not promise White a great deal, but Alekhine was obviously pinning his hopes on the following stunning continuation: 6 Nf3!?

Alfikhine-Euwe 5uch positions a^e normally encountered in simultaneous displays... Subsequently, in a book on this match, Botvinnik showed that the knight sacrifice was incorrect: 6.. .dxc3 7 Bxf7+ Ke7 8 Qb3 cxb2! 9 Bxb2 Ob6! 10 Ba3+ (10 Bxg8 Rxg8 11 0xg8 Qbft+ 12 Nd2 Qxb2) 10. ..c5 11 BxgB RxgS 12 Bxc5+ (12 Q«g8 Qa5->- 13 Nd£ Qxa3) 1 2 . . . Qxc5 13 0-0 Qh5! (13...Rh6? 14 Raci Qb6 15 RKc8 Qxb3 16 axb3 36 17 Rfd1 and 18 Rdd8) U GxgS Be6 IS QhB Nc6, and Black has a won position. Of course, to examine a l l these variations at the board was not easy, and Euwe decided not to accept the knight sacrifice. 6...b5? 7 Nxb5 Ba6 (evidently the Dutch grandmaster had overlooked that on 7...cxb5 there follows B BdS) 8 Qb3 Qe7 9 0-0 BxbS 10 Bxb5 Nf6 11 Bc4 Nbd7 12 Nxd4. This White has an extra pawn and a positional advantage; Black's further resistance Is hopeless. 12.. Rb8 13 0c2 Qc5 14 Nf5 Ne5 15 Bf4 Nh5 16 8xf7+ Kxf7 17 Qxc5 Bxc5 IB Bxe5 Rb5 19 Bd6 Bb6 20 b4 Rd9 21 Radi c5 22 6xc5 Bxc5 23 Rd5 Resigns. 18.

The match-tournament of five grandmasters

In the mid-forties Mikhail hine's most worthy opponents.

Botvinnik was considered one of AlekIn fact, there had already been Orelim-

?th Letter — Before Ihe 30th Match for

the World Championshio

91

(nary negotiations between them about a match for the World ChampionshipHowever, Alekhine's dealh deprived the chess world of a most Interesting encounter. Now Botvinnik had to demonstrate that the discussion about a match had been no accident. By convincingly g i n n i n g , by a margin of three points, the match-tournament of the five strongest grandmasters, he became the f i r s t Soviet, and sixth overall, Champion of the Wortd. The following game took place in the tenth round, when a l l was not yet clear1 — in the event of a win for Keres he would have caught up with Botvinnik and shared the leadership with him. But by gaining a spectacular victory, Botvinnik hurled a long way back one of his main r i v a l s .

Botvinnik-Keres 21 Rxg7+!

22 Nh5+ Kg6

The black

king

finds

itself

in a mating net.

2I...Kxg7

23 Qe3 Resigns. 19.

Botvinnik-Bronstein

Having shown that he was Ihe strongest player in the world, Botvinnik gave up chess for three years, and devoted himself entirely to science — in that time he completed work on his Doctor's dissertation and defended it immediately a f t e r the match. This ' b e t r a y a l ' of chess could have cost him d e a r l y . i n the f i r s t match held under the control of FIDE, David Bronstein played splendidly, was in no way inferior to nis opponent, and with only a I i ttle bit of luck would have ascended the chess throne. An episode which occurred i n the sixth game of the match again illustrates the unusual geometry of the chess b o a r d . . . .

once

Bronstei n-Botvinnik In this position White can e a s i l y draw by 57 Ne6+ and 58 Nd&. To on the safe side, Bronstein decided to b r i n g up his king towards

92

Seven Chess Letters

the dangerous pawn, and went 57 Kc2. Of course, the grandmaster saw the possibility of the black king appearing at f2, but considered only the direct route Kf4-f3-f2, t h i n k i n g that here too he would have lime to play Ne6 and Nd4+ with a draw. How dumbfounded he must have been when the enemy king did indeed set off for the square f2, but not by the direct route (as we established in the f i r s t ' l e t t e r ' , the shortest distance on the board is not necessarily measured in a straight l i n e ! ) . After S7...Kg3!! White had to resign, since it turned out that the e3 pawn could not be slopped: on 58 Ne6 there follows 58. • .e2, and the white knight moves to d4 without check (59 Kd2 K f 2 ! ) .

20.

Botvirtnik-Smyslov

This was the third and, as yet, the last match for the World Championship to end in a draw. In the f i r s t half of i t strange things happened, and the score changed 'sinusoidally' . First Botvinnik won three games out of four, with one draw. Two further draws, and the score became 4 ^ - 1 ^ in his favour. But in the next five meetings the World Champion g a i n e d . . . half a point, and Smyslov shot into the lead — 6-5. A win in the 12th game, the conclusion of which we give below, enabled Botvinnik to level the scores, after which he never again fell behind. The second half of the match, like the f i r s t , ended in a draw, and the World Champion retained his t i t l e .

Botv inn ik-5myslov The black knight has just moved from c5, and after the capture by the white pawn on f6 — 30 exf6 and the zwischenzug 3O...Ne4 Smyslov was apparently feel ing highly optimistic. Indeed, after the retreat of the queen — 31 Qg2 there fol lows 31 . . . Nxf6, and White's position is lost, since all his pawns are hopelessly weak. 31 f7+! An unpleasant sur-prise. White wins thanks to geometric motifs. The pawn cannot be taken by the king because of Qxg7+ (intersection of the seventh rank and the g-f i l e ! ) , while on 3 1 . . -Rxf7, as in fact occurred in the game, there followed 32 Qd8+- Kh7 33 Bxd5 f interseciion of the d-file and the a2-gS diagonal!) 33.. .Nf2+- 34 Kg2 Qf6 35 Qwf6 Rxf6 36 Kxf2 Rxf5+ 37 B13 Rf4 38 Rgft, and Black resigned. •>

21.

Smyfilov-Botvinnik

The 1950s were marked by the r i v a l r y between Mikhail Botvinnik and Vastly Stnyslov. While in the f i r s t match the challenger had to be satisfied with an honourable draw, in the next cycle he succeeded in ascending to the summit. After five games Botvinnik was ahead. After

7th Letter -

Before the 30th Match for

the World Championship

93

ing the score in the sixth encounter, the f i n i s h of which we glue below, Smyslov never again fell behind, but merely increased his advantage. As a result he became the seventh World Champion in the history of chess.

Smyslov-Botvinnik 23 Rxd5! A spectacular continuation, which immediately decides the outcome of the game. 23...exd5 (23...Rxd5 24 Nxc7 Rc5+ 25 k b i , with the deadly threat of Na6) 24 Nxc7 RdcB 25 BxcS Rxc8 26 Nxd5 Rxc6+ 27 Kd2 Ke6 28 Nc3 Resigns.

22.

Botvinnik-Smyslov

Botvlnnlk prepared splendidly for- the return match, and confidently regained his crown. The match began wilh three successive wins for Botvlnnik, and by the fifteenth game the score could have been 10-5. Botvlnnik '3 position was much superior, and any sensible move would have retained his advantage. The grandmaster sank into thought, so as to work out the winning plan right to the end. One can imagine Botvinnik's surprise, when the controller came up to the board and announced that Black had lost on time. The only instance of this type in the entire history of matches for the World Championship!

Botvl nn ik-Smyslov This position was reached In the eighteenth game of the match. The mating r i n g round the black king Is about to close. However, Wtitte converted his advantage into a win only ...50 moves later (what's more, 'on the way' he himself could have been mated!). Nevertheless, ihe f i r s t impression of the position is not erroneous ™hite had at his disposal a striking combination, which, alas, remained behind the scenes: 23 Nd4!! cxd4 (no better is 23...Nxd4 2 * Bd5+! Rxd5 25 Re7 Rf7 26 Re8+) 24 Bd5+! Rxd5 (Z4...KhB 25 Re7) 25 R E 8 ! , and mate is inevitable.

hit

96

Sever Chess Letters 23.

Tal-Botvinnik

In the late 1950s and early sixties, Mikhail Tal, with his Inconceivable combinations, caused trepidation among even the most steadfast of grandmasters. He surmounted the path from master to World Champion in three years! By winning the match against Botvlnnik, the 23*-year-old Tal became the youngest chess king in history, In combinational storms he was clearly superior to his mighty opponent. A tactical blow brought him victory in the 17th, the decisive game of the match, after which i I became clear that the chess world would soon receive a new Champion, the eighth.

Tal-Botvinnik Black has a marked advantage, which he could have maintained by 39...Ka8. But time trouble stepped In: 39.. .Qd5? 40 Rxa6-t! Kb8 (the accentance of the rook sacrifice lands to mate) 41 Qa4, and Black was forced to resign, since there is no defence against the mating threats. Ik.

Botvinnik-Tal

.

p •

.

.

,. .•

B o l v i n n i k a g a i n d i s p l a y e d his a b i l i t y to f i n d the ' A c h i l l e s h e e l ' in h i s oponents' p l a y . After deeply a n a l y z i n g the reasons for h i s f a i l u r e in the p r e v i o u s match, he s u r p r i s i n g l y e a s i l y regained the chess c r o w n . Here is the f i n a l scene of the c o n c l u d i n g , 21st encounter.

Botvi nnik-Tal 28 Ne4! Nd7 (28.. .Nxe'. 29 Ba4+!> 29 Nxd&f 31 Nxe4 Bd7 32 Rf7 Kc7 33 d6+ Resigns. 25.

Kd8

30 RxfB+ NxfB

Petrosian-Botvinnik

At the height of his powers, i.e.

in the mid 1960s, Tigrsn Petrosian

7th Letter — Before the 30th Match for

the World Championship

95

was a great master of defence, an almost invincible player. Journalists conferred upon him the title of 'the iron t i g e r ' . In the f i r s t game he was unable to control his nerves, and suffered a defeat. But in the subsequent games Petrosian only once had to stop the clocks, while he gained f i v e wins over his celebrated opponent. Me became the ninth Champion of the World. This was the last match played by Mikhail Botvlnnik, the Patriarch of Soviet chess. The right lo a return match had been abolished, and Botvinnlk did not wish to begin 'from 1 scratch'. Here is an 'excerpt from the 18th game.

Botvinnik-Petrosian Black elegantly realizes his positional advantage: 52 Kxc3 (52 Kci can be well answered by 52...Nxg4! 53

51. ..c3+! hxg4 h3)

52...Rc7« 53 Kd2 Nec4+ 54 Kdi Na3! 55 Rb2 Ndc4 56 Ra2 axbA 57 axb5 NxbS 56 Ra6 Nc3+ 59 Kef Nxd5 60 Ba4 Rec8 61 Net Nf4 (now 62 Rh2 Is decisively met by 62...Re7 63 Nc2 Nd3+ 64 Kbi Nc5) White resigns. 26.

Petrosian-Spassky

In his fourth cycle of battling for the world crown, Spassky succeeded, at last, in reaching a meeting with ihe World Champion. However, his hour had not yet come. Petrosian played more strongly and more subtly, did not once fall behind on points, and f a i r l y confidently retained his t i t l e . In the tenth game the World Champion carried out a most striking combination, which has gone into a l l ihe books on chess tactics.

Petrosian-Soassky Well known is Petrosian's predilection for sacrificing the exchange for the initiative. On this occasion, as you can see, both rooks have been given up for minor pieces. 27 Nxdfi Qg5+ 28 KM Raa7 29 Bxf7+ Rxf7. Now White could have

96

Seven Chess Letters

regained the second exchange by 30 Nxf7, remaining a pawn up, but this would have prolonged the game. 30 QhB+!! This would seem to be the longest move In, matches for the World Championship! White sacrifices his queen — an unusual occurrence for an event of such high rank. Black immediately resigned, since after 30.. .KxhB 31 Nxf7+ and 32 Nxg5 he comes out a piece down.

27.

Spassky-Petrosian

In. 1954, as a seventeen-year-old youth, Boris Spassky first p a r t i cipated in a zonal elimination tournament for the World Championship. In this cycle, 'without stopping', he went through into the Candidates' Tournament. But then came two dramatic failures (about which we have already talked in the previous ' l e t t e r ' ) . The fourth attempt proved to be 'more successful', but Spassky's dream was s t i l l not realized. And now, at last, the fifteen-year titanic struggle for the world crown brought the grandmaster complete success. By defeating Petrosian In a match, Boris Spassky became the tenth Champion of the World. There was a spectacular finish

to the nineteenth game of the match.

Spassky-Petrosi an 21 e5! (vacating a square for the knight at c3) 21...dxe5 22 Ne4! Nh5 (both knights are invulnerable: 22...exd4 23 Nxf6 Re7 24 Qg6, or 22.. . Nxe4 23 RstfB+, and !n both cases the black king is mated on the following move! 23 Qg6! exd4 ( 2 3 . . . N U 24 Rxf4 exf4 25 Nf3! Ob6 26 RgS! Qd8 27 Ne5, and White wins) 24 Ng5!, and Black resigned, since after 24...hxg5 25 dxh5+ Kg8 26 Qf7i- Kh8 27 Rf3 mate is inevitable.

28.

Fischer-Spassky

In the early '970s the American grandmaster Robert Fischer startled the world with his fantastic victories, and the chess world considered the b i r t h of the eleventh Champion to be a f i t t i n g occurrence. Unfortunately, after the match with Spassky, bitter disappointment awaited supporters of the ancient game — the new king abandoned chess... Fischer's superiority was appreciable, but to be f a i r it must be mentioned that the American made a number of psychological attacks on his opponent, and not only at Ihe chess b o a r d . . . . This was clearly reflected in Spassky's p l a y . Here is an episode from the f i f t h game of the match (by winning i t , Fischer only levelled the scores!).

7th Letter — Before the 30th Match f o r the World Championship

97

Spassky-Fischer 26...Nf* 27 Oc2? The i n i l i a t i v e is w i t h B l a c k , but after 27 Obi White's position would s t i l I have been p e r f e c t l y t e n a b l e . Spassky makes a bad b l u n d e r , a l l o w i n g Black to conclude the game w i t h an elegant stroke - 2 7 . . . B x a 4 ! White r e s i g n e d , since a f t e r 28 Qxa4 (28 0b1 Bxdi 29 Oxdi Qxe4) 28...Qxe4 he cannot avoid mate. 29.

Karpov-Korchnoi

The decisive game of this 'letter1. The f o l l o w i n g fragment

match is annotated in the previous is taken from the e i g h t h game.

Karpov-Korchnoi White e l e g a n t l y concluded h i s attack on the hostile k i n g , 26 Rd7! Rb8 ( a f t e r 2 6 . . . B x d 7 the game ends w i t h a p r o b l e m - l i k e mate: 27 0xf7+ Rxf7 28 Rxf7) 27 Nxf7 Bxd7 (the zwischenzug 2 7 . . . B g 4 is met by the ' q u i e t ' move 28 Qf4) 28 Ndfl+I, and on any move by the k i n g there follows 29 Qf8 mate. Black r e s i g n s . (This was the second instance in this match when the c h a l l e n g e r admitted defeat one move before mate).

30.

Karpov-Korchnoi

98 The author's of the diagram (the only one coming, t h i r t i e t h match World Champion... .

Seven Chess Letters book hope that the position depicted In this we can as yet reproduce! ) Will prove in the for the worlcj crown to be favourable for the

I Part Two The Computer at the Chess Board

Part Two THE COMPUTER AT THE CHESS BOARD

The topic 'the computer and chess' has for many years been Irovoking great interest both among chess players, and among mathematicians. But while in the 1950s computers were making in chess Only their f i r s t steps, during the last decade they have made a rapid Intrusion into a l l the fields in which man in some way or another Dmes into contact with chess. This part of the book is devoted to an account of the chess chievements of the computer. Our attention w i l l be mainly focused on following two questions: (1) the playing by the computer of normal chess; (2) the analysis by the computer of endings.

Why program computers to play

chess?

In economics, in control problems and in long-range planning, le choosing of the optimal solution is normally a very difficult Sroblem. The point is that the modern engineer, planner or economist Has in his work to take commercial or economic decisions in a limited I*me, in a complex, changing situation, depending on a large number factors, which are not subject to a definite evaluation or to mechani c a l control. The point of modern automizalion consists of transferring lo the omputer functions such as the perception of the situation, and the I b l l l t y to compare and evaluate different situations, and lo draw jical conclusions. In order that the computer should be able to »lve these problems, it is necessary f i r s t of a l l to formalize them, |hon to devise effective algorithms enabl ing a solution to be obtained a real istic time, and, f i n a l l y , to put these algorithms into practice the form of a computer program. It is here that we are helped b y . . . chess. In chess it is easy formulate a f i n a l goal and many intermediate goals, and at the same mc it is practical ly impossible to give an exact recipe for their attainment. The choosing of a move in a chess game — this is the a " i n g of a decision in the complex situation described above. This is y scientists from many countries, working on the problem of a r t i -ial intel ligence, have chosen chess as a model for their researches. 101

102

The Computer at the Chess Board

We do not have space here to go i n t o the d e t a i l s of how a computer is programmed to p l a y c h e s s . * Instead we w i l l examine a number of games p l a y e d by computers w h i c h w i l l r-eveal some of the d i f f i c u l t i e s i n v o l v e d In w r i t i n g a successful chess p l a y i n g programme.

h.

.•.•I -i

. . -

.

.

•

!

*For further detai Is of the history of computer programming the reader is referred to "The Machine Plays Chess?" by Alex Bell (Pergamon Press, 1976) and "Chess and Computers" by David Levy (Batsford, 1976).

Competitions with the Participation of Computers

COMPUTER AGAINST COMPUTER

t

,,j.

The f i r s t international meeting in history of computers at the chess board took place In 1967. The Soviet program Kaissa played a telegraph match of four games against the American program created at Stanford University, and won it by the score of 3-1 ( + 2 -0 = 2). The game given below discloses f a i r l y clearly both the strong and the weak sides In the play of both programs.

Kaissa — Stanford University Four Knights' Game 1 e4 e5

2 Nf3 Nc6

3 Nc3 BcS

.

b Nxe5!

In 1966 chess programs were not yet provided with opening l i b r a r i e s , and began ' c r e a t i n g ' from the very f i r s t moves. The move 4 Nxe5 came as a surprise to the creators themselves of Kaissa, since it values the r i g h t to c a s t l e . Nevertheless, the positional advantages sained in r e t u r n tipped the scales in favour of Ihis capture (in the other pan was the move 4 Bc4). At the same time, the computer indicated the best v a r i a t i o n f o r both sides: 4...Bxf2+ 5 Kxf2 Nxe5 6 d4. 4 . . . Nxe5

5 d4 Bd6

6 dxeS Bxe5

7 f4 Bxc3+

8 bxc3 Nf6

As in a game between people, plans can change proceeds: in Its preliminary calculations, Kaissa was play 9 QcJ4 here, but now it sees new p o s s i b i t i t l e s . 9

. . . Ne4

9 e5

as the game intending to

10 Qd3

After the game it was tested t h a i , had it been c a l c u l a t i n g six half-moves ahead, Kaissa would have made the strongest move 10 Qd5!. The point Is t h a t , in the v a r i a t i o n 10 Qa5 Nxc3 11 Qc4 Qh4+ 12 g3 Black is obliged to make a s i x t h half-move, after which White wins the knight. But with a calculation of f i v e half-moves, which Kaissa was 103

104

The Computer ai the Chess Board

doing in this game, after 12 g3 a position with an extra pawn Black is reached, and therefore the move 10 Qd5 was rejected. 1O...Nc5

for

11 Qd5 Ne6

The stronger move 11 . . ,d6 must certainly have been examined by the American program, and the fact that it did not make it indicates the weak nature of its evaluation function. 12 f5 Ng5? Kaissa saw that this loses a piece after 13 h4, and r i g h t l y considered the strongest reply to be 12...c6. Clearly, the move 13 h4 did not enter the American program's f i e l d of vision, i.e. its authors had not succeeded in limiting the search in the correct way. 13 h4 f6

14 hxg5 fxgS

15 Rxh7!

This simple tactical blow calculation to one half-move. 15 . . . Rf8

16 Rxg7 c6

would

17 Qd6

have =*

been

found

even

with

a

-<••••

On making this move, the computer announced that it could 'see' male in al I variations, except the one beginning w i t h 17.. .Qf6 18 exf6 Kd8. 17 . . .

Rxf5

(Black prefers a swifter

end)

16 Rg8+ Rf8

19 Qxf8

mate. The chess match between two computers was widely reported in the press, and this gave a powerful stimulus to the development of chess programming in a whole series of European countries, and especially in Ihe USA. Since 1970, under the aegis of the Association of Computing Technology, North American computer championships have regularly been held in the USA. In the f i r s t such tournament (New York, 1970), which attracted six participants, the winner was the program Chess 3.0 from the North-Western University. The same program Chess (from now on we w i l l drop the designations of its modifications — 3.0, 4.0, etc.) was also victorious in the fol lowing three championships. Subsequently Chess has twice had to concede the championship — in 1974 to the Canadian program Ribbit, and in 1978 to its compatriot — the program Belle. It should be said that the short tournament distance (3-4 rounds on the Swiss system) often reduces the question of f i r s t place to the result of the one game between the r i v a l s . A more trustworthy evaluation of a computer's p l a y i n g strength is given by its r a t i n g ( i t s coefficient on the Elo system), c a l c u l a t e d \ In exactly the same way as for human players. In regaining the title 1 'of champion in 1979, Chess increased its r a t i n g to 2099, whereas Belle was able to achieve a level of only 1982. In 1974 in Stockholm the f i r s t world computer chess championship took place. By that time in Europe the operating programs numbered about ten, and in the USA — more than 50. In this situation the staging of a world championship was most opportune. This contest essentially summed up the i n i t i a l development period of chess programming, and provided a review of the latest achievements in this f i e l d . Thirteen computers from eight countries took part in the battle

L

Competi lions with the Participation of Computers

105

for Ihe title of champion. The representation was as follows: from the USA —four programs, from England — three, and one each from Austria, Canada, Hungary, Norway, Switzerland and USSR. The tournament committee worked out a set of rules to take Into 'account the specific nature of this unusual event. For example, a certain lime was allotted for the elimination of technical faults In the computer, which might arise during a game, for the correcting of a wrongly-entered move, and so on. In other respects, normal tournament rules were In operation. The time control was established at 2 hours for* 40 moves, irrespective of the computer speed. According to the experts, there were two favourites — the American Chess and the Soviet Kaissa. However, in the second round Chess unexpectedly lost to the program Chaos, coming under a crushing attack right from the opening. Here is an episode from this game, with Chess playing Black.

Here there followed: 16 Nxe6! fxe6 17 Qxe6+ Be7 18 Rei Qd8 19 Bf4 KfB (20 Bc7! was threatened) 20 Rad1 Ra7 21 R d NgS 22 Rcdi o5 23 Bd6 Bxd6 24 Qxd&t Me? 25 Nc5 Bf5 26 g4 QeB 2? Ba4!, and White easily converted its advantage into a w i n . After this defeat, Chess won both its remaining games, but could not catch Kaissa, which had defeated a l l its opponents. The Soviet program normally gained Its successes in a sharp struggle, and felt confident in tactical complications. Here is how it Concluded a tense encounter from the first round with the Austrian program Fran?.

Black has just played 3O...RgB-e8, pinning the white bishop. The reply 31 Oc6! had been overlooked by Franz. Now 31 .. .Rxe5 loses to 32 RdB+ Ka7 33 Ra8 mate. At the same time White has created a multitude of threats: Bxc7+, Qb6+ and Rd7, which Black is unable to Parry. The game concluded 3 1 . . .Qg6 32 0xc7+ Ka8 33 Rd7 Qf5 34 Oc6 mate. The final results of the event were: Kaissa A/4; Chess, Chaos (both USA) and Ribblt (Canada) 3. At the closing ceremony Kaissa was swarded a memorial gold medal as the f i r s t world computer champion.

106

The Computer at the Chess Board

Three years later, in the Canadian town of Toronto, a second w o r l d championship was h e l d . The number of p a r t i c i p a n t s had grown to 16, a " d the o v e r a l l p l a y i n g strength of the electronic p l a y e r s had also grown. On t h i s occasion Kaissa lost i t s t i t l e , and shared 2nd place w i t h the American program Duchess, while the new w o r l d champion became the program Chess, which won a l l four games and f i n i s h e d a point ahead of K a i s s a . The tournament began w i t h a s e n s a t i o n : i n the f i r s t round Kaissa lost to Duchess i n a game which f o r several d a y s e x c i t e d the minds of fans and programmers, and which appeared in many chess p u b l i c a t i o n s . We g i v e here this game, which i l l u s t r a t e s a number of i n t e r e s t i n g points a r i s i n g i n the c r e a t i o n of a chess p r o g r a m .

Duchess (USA) - Kaissa (USSR) Centre Counter Game 1 e4 d5 2 exd5 7 Nc3 0-0 8 Be3 Bg4

Nf6

3 CJ4 Nxd5

4 Nf3 g6

5 Be2 Bg7

At the s t a r t of the game both programs make opening d i r e c t o r i e s . Now independent p l a y b e g i n s . 9 c5 Nd5

10 0-0 e6

from

their

.

1O...Nc6 Is b e t t e r . By the move i n the game the viewpoint of i t s e v a l u a t i o n f u n c t i o n ) a strong g i v e n p o s i t i o n , however, this is not so i m p o r t a n t , had c a l c u l a t e d v a r i a t i o n s a l i t t l e deeper, it would 11 Qb3 b6

moves

6 c4 Nb6

12 Nxd5 exdS

13 Bg5 Qd7

Kaissa creates (from point at d5. I n the and i f the computer have ' r e a l i z e d ' t h i s ,

lA h3 BfS

15 Qc3!

A subtle move, by which While prevents the development of [he k n i g h t at bBOn 1S...Nc6 there now follows 16 cxbfi cxb6 17 Bb5, and wins. In the event of 15 Racl the move l 5 . . . N c 6 becomes possible, since on 16 cxb6 there is the reply 16...Na5. ,•-. ••• 15 . . . Re8

16 Rfel Be4

Black intends by development problems. 17 Nd2 Qf5

\7.. .Qf5

fol lowed

by

18.. .Nd7

solve

its

18 Be3 Qe6

The threat was 19 f 3 . Concrete, calculating play and for the moment both programs ar-e up to the mark. 19 Nxe4 dxe4 Evidently,

to

20 cxb6 cxb6

22. . . f 5

did

not

23 Qc6 Nf6 24 Be2 Radfl 28 Racl Bf6 29 Qb3

21 Reel Nd7 appeal

to

25 Qa4 Be7

is in progress,

2Z Bg4 Qd5

Black

because

26 6b5 QfS

^ of

^

23 Bdl •

27 Rc2 Nd&

Black has successfully regrouped its pieces, and its knighi occupies an excellent position in the tent re, but what to do n e x t '

Compelitions with the Par! icipation of Computers

107

A human in this position would set aboul restricting the opponent's possibi I ities, by playing 29.. .h5, 30.. .Kg7, etc. If White plays passively, there Is the possible plan of advancing ...g6-g5-g4 and opening the h-file. In the event of the exchange of the b5 bishop for the knight, the d4 pawn comes under siege. As yet the construction of promising (and at the same lime, correct) plans is inaccessible to a computer. 29

a5?

A move which loses the game due to the existence for White of a latent threat. In order to discover i t , a calculation of 9 half-moves was required. Had it been calculating to that depth, the computer would most probably have played 29...h5!

30 g4! Qe6 3O...Qf3 is b a d because of 31 RcS. 31 Rc6 a4 Black now sees lhat it loses a piece i n the v a r i a t i o n 3 1 . . . R d 6 32 Rc8+ Kg7 33 g 5 . The move 3 1 . . . a 4 lengthens the v a r i a t i o n by two half-moves, and the computer t h i n k s that it is losing only a p a w n . 32 Qxa4! Rd6

33 Rxd6 Qxd6

34 Qa8+!

34 . . . ReB?! Unexpectedly Kaissa gives up a whole rook. The commentators were perplexed, and explained In confusion to the spectators that chess programs were s t i l l a long way from perfection, and that anything could be expected of them. Great was the general amazement when Kaissa explained its 'blunder' by the following v a r i a t i o n : 34...Kg7 35 QfB+!! KxfB 36 Bh6+ and 37 Rc8+ with inevitable mate! Not one of the players present at the championship found this spectacular queen

108

The Computer al

the Chess Board

sacrifice. I t is not known whether or not Duchess would have but from p r a c t i c a l considerations 34...Kg7 should have been since p l a y i n g on a rook down is absolutely hopeless, whereas not be every program (and not every master!) that would f i n d I f , in reply to 34. . .Kg7, While had been intending to win a 35 g5, he himself would have lost to 3S...Nxe3 36 gxf6+ Qxf6 Qg5+ and . ..Qxb5, with a decisive advantage lo Black. As we see, in this game Kaissa f e l l v i c t i m sharp-sightedness, but essentially (his is simply p a r t of the program authors, since the p r a c t i c a l tioned above can easily be programmed. 35 Qxe8+ Kg7

seen i t , chosen, it would 35 Qf8*. piece by 37 fxe3

to i t s own tactical an omission on the considerations men-

36 g5 BdB

The conclusion of the game moves Black conceded defeat.

is

not

of

any

interest — w i t h i n

a few

The program Chess, which took f i r s t place, played a l l its games evenly and s t r o n g l y . Usually by the middlegame it had already gained a serious advantage. Here is a t y p i c a l example.

Chess-Duchess E x p l o i t i n g the more a c t i v e p l a c i n g of i t s pieces, White develops a Strong i n i t i a t i v e in the centre and on the K-side. 23 Ng5! Rc8 24 Nf6 Rdc7 25 Ne6 Rf7 36 Nxf8 RfxfB 27 Nxh7! Kxh7 28 Bxd6 RfdS 29 Bf4 Rd4 30 BgS. Having won a pawn a n d , thanks to the two a c t i v e bishops, retained its positional advanlage, While easily converted its advantage into a w i n . In the t h i r d world championship, which took place in\ September 1980 i n A u s t r i a , 18 programs from s i x countries p a r t i c i p a t e d X In nib welcoming speech at the opening of the championship, the President of FIDE, the Icelandic grandmaster Fridrik Olafsson, remarked on the great interest caused by such events in the chess w o r l d , and promised a l l possible help and support on the p a r t of FIDE for the Association of Chess Programming. By t r a d i t i o n , the championship was held as a four-round Swiss tournament. First, with the identical result of i\ p o i n t s , were two American programs, Belle and Chaos. An a d d i t i o n a l game, played between them d i r e c t l y In the USA (only the score of the game was transmitted to A u s t r i a ) brought victory and the t i t l e of champion to the program Belle. Both ex-champions, Chess end Kaissa, made a rather modest r e s u l t , scoring 2 j and 2 points respectively. The success of Belle >•• explained to a large extent by its considerable technical s u p e r i o r i t y over i ts opponents. The champion used a special I y developed chps^

Competitions

with

the Participation of Computers

109

computer, in which the extracting of moves, the moving of the pieces and the evaluation were implemented not by programming, but schema t i c a l l y , i.e. they were prepared machine commands. Thanks to this, Belle succeeded in examining some 20,000 moves in a second, and in the middlegame made calculations to 7-8 half-moves. We give the additional game, which determined the world champion, in which a fierce attack runs into a cool defence.

Belle-Chaos Alekhine' 5 Defence 1 e4 Nf6 2 e 5 Nd5 3 d4 d6 4 Nf3 dxe5 5 Nxe5 g6 6 g3 Bf5 7 c4 Nb4 8 Qa4+ N4c6 9 d5 Bc2 10 Gb5 Qd6 11 Nxc6 Nxc6 12 Nc3 Bg7 13 Qxb7 0-0 14 Qxc6 Qb4 15 Kd2 Be4 16 Rg! RfbB 17 Bh3 8h6+ 18 f4 QaS 19 Re1 f5 20 Qe6+ Kf8 21 b3 Bg? 22 Bb2 Bd4 23 g4 Rb6 24 Qd7 Rd6 25 Qa4 Qb6 26 Ba3 Bxc3+ 27 Kxc3 RddS 28 Radi Qf2 29 gxf5 Qc2+ 30 Kd4 gxf5 31 Qc6 Gf2+32 Ke5 Kg8 33 Rgl+ Kh8 34 Bxe7 Qg2 35 Qf6+ Kg8 36 Bxg2 Rxd5+ 37 Ke6 h6 38 Qxh6 Re5+ 39 fxe5 Rf8 40 Bf3 mate. The standard of computer play is gradually rising. Here, for example, is an absolutely human-like game, played by two American programs in the 1979 championship of the country in Detroit.

Belle-Chess Modern Benoni Defence

i-

1 d4 Nf6 2 c4 c5 3 d5 e6 4 Nc3 exdS 5 cx<)5 d6 6 e4 g6 7 Nf3 Bg7 8 Be2 0-0 9 0-0 Re8 10 Nd2 Na6 11 f3 Nc7 12 a4 b6 13 Nc4 Ba6 14 Bg5 h6 15 Bh4 g5 16 Bf2 Nh5 17 Ne3 BcB IB Qc2 Nf« 19 Bc4 Bd7 20 Rfdl Qf6 21 Bg3 Nh5 22 Bel Nf4 23 Khi a6 24 Bg3 b5 25 axb5 axb5 26 RxaB RxaB 27 Bf1 b4 28 Ne2 b3 29 Qbi Nh5 30 Bf2 Nf4 31 Nc4 Nxe2 32 Bxe2 Bb5 33 Bg3 Ra4 34 Qci Bf8 35 Rd2 QdB 36 0f1 h5 37 Kgl h4 38 Bf2 Bg7 39 Ne3 Bxe2 40 Qxe2 Ra1 + 41 Rdl Ra2 42 Qd3 Rxb2 « Nc4 Rc2

44 e5T Bxe5 49 Bb6 h3! 50

5 Nxe5 dxe5 46 Qxb3 Re2 47 Kfi c4! 48 Qb7 Ra2 Qf6! 51 Qd8+ Qxd8 52 Bxd8 Rxg2 53 Rei?

White had many ways to win - d5-d6, Be7, Bc7 or Ba5. By temporarily giving up material, it would have forced the advance of its passed pawn. Instead of this, it attempts not to fal I behind the opponent in the pursuit of pawns, and as a result lets slip the w i n . l i is here that the basic weakness of the computer tells — the Inability at the necessary moment to calculate a long but hardly-branching

110

The Computer gt the Chess Board

variation.

53 . . . 5B d6 Rxh2 Draw.

c3 54 Rxe5 c2 55 ReB+ Kg? 56 Bx£5 Rxg5 57 RcB Rg2 59 d7 Rd2 60 Kgi ftxd7 6i Rxc2 Rd3 62 Rf2 Kf6 63 Kh2.

MAN AGAINST COMPUTER Can the computer r e a l i s t i c a l l y compete with man at the chess Board? This question was posed back at the dawn of chess programitiing. In the f i r s t instance the question must be put more precisely. If one has in mind the analysis of certain types of endgame, or the solving of chess problems, the answer is undoubtedly yes. Bui if one considers the computer in single combat with man in a normal tournament game, here its successes have bee.n much more modest. Nevertheless, for some time chess players have been watching with interest and caution how the family of chess computers has been developing and gathering strength, and how i l is mounting its offensive on the position of the human p l a y e r . The only question is, to what level the computer can rise, and how soon it w i l l begin p l a y i n g on equal terms with masters. In 1968 international master David Levy made a bet that within 10 years no computer would be able to beat him in a match. During this lime he has twice played a match with the program Chess, and h once with Greenblatt s program from MIT. Incidentally, the latler program is a veteran among chess robots — it is already about two decades o l d , A curious fact Is that it Is directed specially towards playing with humans, and as a matter of p r i n c i p l e does not Join battle with its own k i n d . It is true that l^evy defeated it f a i r l y easily by 2-0, but in the last match with Chess he had to work pretty hard to overcome the robot's resistance. This match was the best of 6 games. The computer, according to the condition, had to score 3^ points. The f i r s t game was advantageous for Chess, but ended in a draw. In the second and t h i r d games Chess lost, but in the fourth the human w^s at last defeated — a highly significant fact. And although Levy won the f i f t h game, it would appear that the f i r s t warning bell has sounded —a l i t t l e more a n d . . . . At any rate, as yet Levy has not made a new bet. We give the f i r s t game of this match between man and computer. D. Levy •- Chess King's Indian Attack 1 g3 d5 2 Bg2 e5 3 d3 Nf6 h Nf3 Nc6 5 0-0 -Bd7 7 Bb2 Qe7 8 a3 e4 9 Ne1 0-0 10 d4 Bd6 11 e3 Ng& 12 h.3.

6

Bc5

Competitions with the Participation of Computers

in

Black's attack could have been refuted by 12 c4! The move played by Levy provokes Black into making a piece sacrifice. Obviously he assumed that the computer would not b r i n g itself to give up the knight. Now White's position becomes c r i t i c a l . T2 . . . Nxe3! 13 fxe3 Qg5 Qxf2+ 17 Qxf2 Bxf2+ 18 Kxf2 f5 22 Nc3 RhS 23 Kh2 RfB Z4 Ndi 27 Ng2 Rf3 28 cxd5 Rhxh3+.

1* g * Qxe3+ 15 Rf2 Bg3 16 Qe2 19 gxf5 Ne7 20 cA Rxf5* 21 Kgi c6 Ng6 25 Hc1 Bxh3! 26 Bxh3 Rfi

The simplest was 28...Rfxh3+ 29 Kgl allowing the opponent to open a line for his the threats. The subsequent play of Chess computer's weak endgame technique allows the

RhH 30 Kf2 Rxd5, not rook, and maintaining a l l is most uncertain. The master to save the game.

29 Kg1 cxd5 30 Rc8+ Nf8 31 Bc3 Rd3 32 Nde3 Rdxe3 33 Nxe3 Rxe3 34 Bb4 Rf3 35 Rd8 h6 36 Rxd5 Rxb3 37 Rd8 Rf3 38 Ra8 g5 39 d5 h5 40 d6 Kg7 41 Rxa7 Rf7 42 Ra5 Kf6 43 Bc3+ Kg6 44 Re5 Rf3 45 Bb4 Rf4 46 Re7 Rf7 47 Rxe4 Rd7 48 Re7 h4 49 Kg2 g4 50 Kh2 b6 51 Kg2 Rd8 52 a4 Nd7 53 aS! Nf6 54 axb6 Nd5 55 b7 Nxe7 56 dxe7 RhS 57 Bd6 Kf6 58 b8=0. Rxb3 59 Bxb8 Kxe7. Draw. One can cite a number of examples of successful play by computers against professional p l a y e r s , but in hastening to report on the computers' successes, reporters often omit important details from their accounts. Thus, grandmaster Walter Browne did indeed lose a game to the program Chess, b u t . . . in a simultaneous d i s p l a y . Another holder of the highest chess t i t l e — the Englishman Michael Stean —suffered a defeat against the same Chess, b u t . . . In a five-minute b l i t z game, and so o n . The shorter the time allotted to the game, the easier it is for the computer against a human. If a tournament for the world championship were held with a time control of 30 seconds per game, it is (Possible that already now a computer would become champion — a man would simply physically be unable to make the moves. But such play fs of l i t t l e interest. As regards serious chess, as yet computers are ( t i l l a long way from grandmasters. But at the same time, grandmasters. . . are close to computers. Ex-World Champion Mikhai I Botvinnik has already been engaged in [chess programming for a number of years. Another Ex-World Champion, [the American Bobby Fischer, according to official but as yet unconf i r m e d reports (with Fischer there are always such d i f f i c u l t i e s ! ! , is llso making an active study of how chess programs are created. At any rate (and this is now an indisputable f a c t ) , he has [ p l a y e d a short match against the program of Greenblatt mentioned [ e a r l i e r — that same one which avoids other computers. Well, the ^opponents are worthy of one another, since, after al I, for many years now Fischer has not played against people. To Fischer's credit, it must be said that his chess seclusion Iwould not appear to have diminished his s t r e n g t h . He conducted the i(natch In his best style, and defeated the computer by the score of 3-0. W think that the readers w i l l f i n d it interesting to look at one of the games from this encounter.

112

The Computer at the Chess Board Compu ter-F i scher Sici I ian Defence

I e4c5 2 Nf3 g6 3 d4 Bg7 4 Nc3 cxd4 7 Nxc6 bxc6 8 e5 NgS 9 f4 f6 10 exf6

5 Nxd& Nc6

6 Be3 Nf6

Machines, like people, play the opening q u i c k l y . Modern computers can 'remember' thousands of opening variations, with which they are equipped by their programmers. But there is a I irntt to what can be stored. 5ooner or later the machine has to think for itself, and i t is then that its true strength is revealed. Strictly speaking, this has always been the problem in chess programming: to find a good move in an original situation. White's f i r s t 'independent' move in this game is clearly not the beat; the theoretical continuation 10 Bd4 maintains for him a slight come Quickly into p l a y . 10 . . . Nxf6

advantage,

whereas

now the black

pieces

11 Bc4

The second poor move in succession. II

. . . dS

12 Be2

The computer evidently thought that the backward pawn at e7 would compromise the opponent's position. Other justifications for the move 11 BcA would force one to suggest that the machine's search depth was too short, or that i t d i d not consider the reply H . , . d 5 , which is improbable. 12 . . . Rb8

13 b3 Ng4

14 Bd4 e5!

•

s-.-. •

•-

After 14. . .Ne3 White could have sacrificed his queen by 15 Bxg7! Nxd5 IS BxhS Nxc3 17 Bxc3, achieving a position which would be d i f f i c u l t to breach. The move made by Fischer is clearly stronger. 15 fxe5 0-0! Keeping the white king in the centre. IS BxgA Qxh4+

17 g3 Qxg4

19 Kd2 was more 19 . . . RxfU

tenacious.

18 Qxg4 Bxgft

;-• "-

- \'-

19 Rf 1 .

Now White loses by force.

<• '-

20 Kxfi

:•!.••!? 20 . . . c5! 21 Bf2 6xe5 25 Bxc3 Bxc3 26 Rfi Bf5

22 Be1 Rf8+

23 Kg2 Rf3

24 h3 Rxc3

Competitions with the Participation of Computers

113

Chess players resign in such positions, but the machine, demons t r a t i n g its ' i r o n ' determination, fought on to the last move. We think that in this position the readers would be able to win against any • opponent, and so this hopeless ending for White can be o m i t l e d . , . In recent times computers have become so omnipresent that they are constantly mentioned in reports of International events. For example, at the international tournament In Bad Kissingen (West Germany, 1980), for advertisement purposes the organizers dec idea io hold a simultaneous display by four grandmaster against 100 computers. Strictly speaking, each player took on 25 computers, but for greater respectabi 1 i ty it was announced that one hundred machines would simu Itaneously join battle with the grandmasters. What did the average p a r t i c i p a n t in the display look like? It comprised three p a r t s : its own mini-computer, a chess board and a keyboard. As yet its cost is rather high — about a t h i r d of the average monthly wage. In a d d i t i o n , it requires a considerable amount of e t e c t r i c i t y , the consumption of which depends on the working level. Thus, in the adjoining halls were displayed computers, which anyone could set to an appropriate p l a y i n g strength, by a l l o t t i n g half a minute io a move, or a minute, or more. The v i s i t o r s , not wishing to lose, gave their 'opponents 1 as l i t t l e time as possible. During the display the grandmasters frequently experimented, so as to discover the possibilities of the machines. For example, one of Ihe authors of this book, who numbered among those g i v i n g the d i s p l a y , won four identical games, move for move, while Boris Spassky sacrificed so much that he had to make considerable efforts so as not to f a l l behind his colleagues, and to score a 100% r e s u l t . In general, chess mini-computers are becoming more and more common. While inferior in strength to their senior colleagues — the big computers, by v i r t u e of their a v a i l a b i l i t y they can serve as excellent t r a i n i n g devices for a wide circle of chess enthusiasts. A big computer, on the other hand, can give a simultaneous display against average strength amateurs, and what's more, highly successfully. One of the most impressive enterprises of this type was held at the end of 1977 in Paris. The American program Chess was opposed by ten players — in the main prominent c u l t u r a l and public figures. Among them were the composer Guy Bear, the f i l m director Roget Vadim, the playwright Fernando Arrabal, and others. Also included in the p a r t i c i p a n t s was 17-year-old Manuel Apitsella — the French Junior Champion. The display, which was in the nature of a demonstration of the achievements of modern technology, was held in a hall which was linked directly by s a t e l l i t e to the computer in the USA. All the games were reproduced on demonstration boards and on television, and comments on them made by French masters. The machine won seven games, drew one and lost two. In half of the games it took Black. If account is taken of the fact that the chess r a t i n g of its opponents varied from somewhere between f i r s t and t h i r d category, the result can be considered excellent. cipant

Here is how the in the d i s p l a y .

computer's

game

went

with

the

youngest

parti-

Chess-Apitsella Nimzow i tsch Defence 1 e4 Nc6 2 d4 e6 3 Nf3 d5 4 eS Bd7 7 0-0 f6 8 exf6 gxf6 9 Bh6 Kf7 10 QdZ Bd6

5 Nc3 Bb4 6 Bd3 Nge7 11 Nb5 Ng6 12 Nxd6+

114

The Computer at the Chess Board

cxd6 13 h4 e5 14 h5 Nge7 15 dxe5 dxe5 16 c4 Rc8 17 cxdS 18 Bxh7 Rxh7 19 Qxd5+ Be6 20 Qxd8 RxdB 21 Be3 RxhS 22 g3 Bh3 23 Rlci Bg4 24 Nh4 Rg8 25 f3 Bxf3 26 Nxf3 R*g3+ 27 Kf2 Rhh3 28 Ngi Rxe3 29 Nxh3 Rxh3 30 Rhl Rd3 31 Rn7* Ke6 32 Rxb7 e4 33 Rcl Rd6 3ft Ke3 a5 35 Kxe4 Rd4+ 36 Ke3 Rd6 37 Rc5 f5 38 Rh7 Ne5 39 Rh8 U+ 40 Ke4 Ng6 41 Rh6 Resigns. After the display, the well-known French mathematician Francois le Uonnais remarked that the last 10 years had seen marked progress in the creation of chess programs. However, It was not yet sufficient to solve the main problem — to disclose the secret of human t h i n k i n g . "The most important and most interesting", the scientist said, "is not the game itself, however splendid ana highly intellectual it may be. The main thing is the methods and algorithms which are necessary for automating the game of chess, since they can also be extended to other fields of human a c t i v i t y " . Significant in this sense is the interest shown in chess programming by major computing firms. Thus the firm CDC grants unlimited free use of its best machine Cyber-176 to the group at North-Western University that created the program Chess, demanding for this merely permission to include this program in a software pack. Many firms allow free use of their computers for participation in tournaments of chess programs. Recently the Dutch computing form Uolmak put up a prize of 50,000 dollars for the creators of a chess program which would succeed in defeating E*-World Champion Max Euwe in a match of four games. The offer remains in force until 1st January 1984, In announcing such a competition, the directors of the firm stated that they were aiming, f i r s t and foremost, to stimulate the creation of programs which would play significantly better than those currently in existence, and also that they hoped that the match would assist the raising of the general interest in chess.

The Ccmputer Analyzes the Endgame

Already when the f i r s t chess programs were being created, it was [observed that the handling of the endgame was the weakest aspect of a (Computer's p l a y . In the endgame the forming of plans several moves in advance becomes of chief Importance. What's more, for a human player it is ', easier to find the correct plan In the endgame than in the middlegame, 1 since the opponent has fewer opportunities for preventing Its implementation, and also the number of actual plans in the endgame is significantly less. Therefore, in the f i n a l stage of the game a player t can see the concluding positions of very long variations, and can (often be certain that a chain of moves w i l l lead to them. That is [•"how things are in f a i r l y complex endings, where a player is guided by Intuition and experience — precisely that which as yet lends itself [ b a d l y to formal ization, and in which computers are lacking. However, in the playing of certain endings with few pieces, the t'lnachine is now superior to man, and moreover, is in a posi tion to know fthe 'absolute truth ' . . . . I n 1968 the t r a d i t i o n a l match between Moscow and Leningrad ftook place in the USSR c a p i t a l . With the score standing at 39j-395 [(the match was held on 40 boards in iwo rounds) there remained one tunfinished game, which would decide the result of the match. The j Leningrad player was a pawn up, and in the event of him being ^successful his team would win. The resumption of the game lasted a • long time, and the Leningrad team were already in danger of missing \ their t r a i n , so the game was sent for adjudication in the following (-position, with the Leningrad player p l a y i n g Black.

115

116

The Computer at the Chess Board

The game was adjudicated by an authoritative grandmaster commission. However-, the whole trouble was that the ending 'queen and k n i g h t ' s pawn against queen' had been studied by players for many years, but to establish precisely which positions were won, and which drawn, they had nol yet been able. As for the given position, the Jury in i l s perplexity judged it to be drawn, which provoked a natural objection on the part of the Leningrad team. The matter ended with the return v i s i t of the Muscovites to Leningrad not taking place, and a t r a d i t i o n of many years' standing was b r o k e n . . . . It is clear that, i f the computer had ' i n v e s t i g a t e d ' endings of this type, no misunderstanding would have occurred. For the analysis of such queen endings it was decided to enlist the services of Kaissa*. Here it is appropriate to describe the general idea which lies at the basis of an algorithm for the analysis of chess endings. F i r s t of a l l it is assumed that, in the analysis of this or that class of ending, the evaluations of a l l positions of so-called secondary endings, i.e. those obtained from the ones being studied by a change of material — a capture or the promotion of a pawn, are already known. We w i l l consider a class of endings in which White is t r y i n g to win, and Black is h a t t l i n g for a draw. All positions of this class naturally divide into two groups — those with White to move, which are designated by W, and those with Black to move — B. Out of the group W let us single out those positions in which White has a move leading immediately to a winning secondary ending (an elementarily won position). Let us denote the aggregate of these positions by Wo and call it the zero rank (a win in zero moves). Let us remove Wo from the group W. For the moment the remaining positions form a class of 'unsorted Whites', which we designate UW. In analogous fashion, we remove from group B those positions in which Black in one move can go into a drawn {or won for him) secondary ending — Bo, end we obtain a class LIB — 'unsorted B l a c k s ' , After the procedure described, a basic multi-step ranking algorithm is put into practice. Let us consider the f i r s t step of the algorithm. We w i l I pick out from UB those Dositions from which a l l moves by Black lead to Wo. Obviously, these w i l l be positions which are lost for- Black in one move. We will call this grouo of positions She f i r s t black rank and w i l l denote them by RBI. Note that RBI consists of positions which have no moves to UW. It is this feature which is u t i l i z e d in the actual construction of RB1 . Let us now select from UW positions from which at least one move leads to RBI. As a result we obtain a group of positions RW1, which are won for White in one move. We w i l l remove RBI from UB and RW1 from UW. Everything is ready for the second step of the algorithm. We continue in similar fashion: we construct RB2 as a group of positions having no moves to UW, and RW2 having no moves to LIB and so on. The ranking process concludes when the next group RB or RW to lie formed turns out to be empty. Positions remaining in the groups UB and UW are definitely drawn. As the ranked positions are obtained, they can be removed onto, for example, magnetic tape, and used for play, or else printed out. For the implementation of the described algorithm in practice, two conditions must be f u l f i l l e d . Firstly, the number of different positions w i l h the given material must not be too large. Taking *5ince a l l the programs which w i l l be mentioned below were created by the same group of mathematicians as created Kaissa (under the leadership of V. Arlazarov), for convenience we w i l l also call them by the name of the muse of chess.

The Computer Analyzes the Endgame

117

account of the resources of the modern computer, il can be said that the analysis of five-piece endings ( i n c l u d i n g Ihe kings) Is at the limit of its p o s s i b i l i t i e s . Secondly, the machine must be able to evaluate, in the sense of a definite r e s u l t , any position of a secondary ending. It should be noted that. In the programming of five-piece endings It has proved sensible to remember only the black ranks, they being less dense. Nevertheless, in the ending 'rook and pawn against rook 1 about sixty million positions had to be remembered. Returning to our queen and pawn ending, we can report that at present, although Kaissa has constantly been diverted by more important matters, it has studied the class of positions with a k n i g h t ' s pawn on the penultimate rank. Now about each such position it can be definitely stated whether or not it is won for the stronger side, and if it is won, then in how many moves. In the game from the aforementioned match, the pawn, as we saw, stood on the sixth r a n k , and this means that the machine has only one more step to make before being able to evaluate i t . There is thus the hope that the MoscowLeningrad matches may soon be resumed... . An interesting fact is that, when analysing the queen ending, Kaissa discovered two won positions, in which with best play for both sides (he balance of forces can be changed only in 59 moves! Here is one of these positions (in which it is Black to move).

Subtle manoeuvring tenacious defence, leads

of the white k i n g w i t h i n 53 moves

and queen, despite Black's to fhe following position.

Here the black queen is forced to occupy a passive position — 54...Q9B, and after 55 Qb6+ Ka3 56 Qb7 Ka4 57 Kc3 Ka5 58 Qb4+ Ka6 59 Qc4+ White f i n a l l y exchanges queens and promotes his pawn. The reader will no doubt be famll iar with the rule by which a game ends in a draw if both sides have made at least 50 moves, in the course of which no piece has been taken, and no pawn moved. In the USSR Chess Code there is the following interpretation of this r u l e : "For positions w i t h k i n g and two knights against k i n g and pawn the number of 50 moves is increased to 75. It can be increased for other specific

118

The Computer a l the Chess Board

positions, only on condition that this number and these positions are precisely mentioned In the regulations for a tournament or match". The positions discovered by Kaissa show that the number of 50 moves should also be increased in the code for the ending ' k i n g , queen and pawn against king and queen' . This is the f i r s t instance In history when the computer has ' interfered' in the chess code! (The ending 'king and two knights against king and pawn' wes studied many years ago, and without the help of the computer). There is another unusual Instance concerning the ending in question, when the computer f i r s t rendered practical assistance to a grandmaster. This occurred in 1975 in the US5B Zonal Tournament at Vilnius. The game Grigorian-Br-onstein was adjourned in a queen ending with an extra pawn for Black. The grandmaster knew of Kaissa's success, and he turned to it for 'consultation'. Not long before the start of the adjournment session, Bronstein received a letter with an analysis of the position. True. Grigorian played inaccurately right at the start of the adjournment, so that Kaissa's 'prompting' was not required. In practice, rook endings occur much more frequently than queen endings. One of the most common forms of this ending — 'rook and pawn against rook' has also been given to Kaissa for analysis. Spending 60 hours of machine time, it coped b r i l l i a n t l y with its task, and can now evaluate any position of this type, irrespective of the position of the pawn. At the same time, the machine has established a number of interesting facts. For example, It has discovered a position which can be won in not less than 60 moves (as before, by won we have in mind a transition into a won secondary ending). Here is one such position (Black to move),

Incidentally, in the given position White succeeds in moving his pawn only on the 32nd move after lengthy manoeuvring, in the course of which there are several instances where there is only one correct move, and one which is difficult for a human player to f i n d . For the rapid evaluation of endgame positions in practice, it is useful to imagine a drawing or winning sone for the placing of one piece, while the positions of the remaining pieces are f i x e d . Consider the following diagram, on which the positions of four of the five pieces are fixed. The f i f t h piece, the black king, can stand on any permissible square of the board. (See diagram at top of page 119) In this position It is Black to move. For some positions of his king he loses, for others there is no win for White, The results of the analysis carried out by Kaissa are shown directly on the diagram. If a square is blank, the position is drawn, but if a certain number (n) is written in i t , White wins in n moves. We see that, with the black

The Computer Analyzes

ihe Endgame

119

king cut off beyond Ihe f - f l l e , the one saving square for if we disregard g l and gZ, on which the k i n g simply captures rook and Black w i n s ) .

is g4 (if the white

This diagram represents a real puzzle. Imagine lhat you are p l a y i n g White, it is now the opponent's move, and you are granted the r i g h t to place your king on any square of the board. Which one should be chosen so as to gain a win? Amazingly, there is only one square: White wins only with his king at e8! An interesting fact is [hat grandmaster Averbakh, one of the greatest specialists in the f i e l d of the endgame, after p l a y i n g for several hours with Kaissa at 'rook endings', admitted its complete superiority. He remarked that such a s p a r r i n g - p a r t n e r would be highly useful to any grandmaster for improving bis technique in the endgame. Of the four-piece endings, the most interesting knight. On its study Kaissa spent only 15 minutes!

[.
is

rook

against

In this position it is Black to move, and with correct defence on his part his knight is trapped only on the 27th move! We give the main variation of the solution; I...Ne2+ 2 Kd2 (After 2 Kc2 White can no longer win) 2...Nd4 3 Kc3. And now 3 K63 would be a mistake. However, right up to the point when the knight is caught, White has to make a number of

120

The Computer at

the Chess Board

precise moves. It is d i f f i c u l t to imagine that they could a l l be found by a player d u r i n g a game! 3...NbS+ 4 Kc4 Nd6+ 5 Kc5 Nb7+ 6 Kb6 Nd6 7 Rf4! (the rook moves more r a r e l y than the k i n g , but Its movements are more subtle) 7.,.Kb3 8 KcS Nb7+ 9 Kc6 Nd6+ 10 KbS Ne6 11 RT3+ Kc2 12 Kc4 Kd2 13 Rf5 Kc2 14 Rf2+ Kdi 15 Kd3 Nc5* 16 Kd4 Nb3+ 17 Kc3 Kel 16 Rb2! Nc5 19 Kd4 Ne6+ 20 Ke3 Kdi 21 Rb6 Ng5 (after 21..,Nc5 22 Kd4 Nd7 23 Rd6 the knight is caught more q u i c k l y l 22 Rc6! Nf7 23 Rc7 Ne5 24 Ke4! Ng4 25 Rg7! Nf6+ 26 Ke5 Nh5 27 Rg5, and the k n i g h t is caught. We should mention lhat the a n a l y s i s of the ending 'rook against k n i g h t 1 has a h i s t o r y . Back in 1970 a German mathematician included a study of It in his d i s s e r t a t i o n , devoted to combinatorial methods. Eight years later a complete study was made of t h i s ending, independently of each other, by Kaissa and an American computer. It is i n t e r e s t i n g that the positions w i t h a record number of moves, found by each of the machines, coincided almost e x a c t l y . Only, in the diagram g i v e n above the American computer ' p l a c e d ' the k n i g h t at e2 (instead of g1), which shortens the solution by one half-move. Evidently the machine p r i n t - o u t contained positions w i t h While to move. Simpler for a human p l a y e r is the ending 'rook against b i s h o p ' . Here there are p r a c t i c a l l y no positions where the e v a l u a t i o n is in doubt. Bui here too those positions are of interest, where the win is. achieved i n the maximum number of moves. Here is one of the records established by Kaissa. White to move wins only on the 16th move.

Thus, Kaissa can a l r e a d y hegin producing its own theoretical publications! But what are the f u t u r e prospects for the computer in the a n a l y s i s of chess endings? The r a n k i n g method described above can be used to study several other types of ending, which are important for theory. Firstly, the work i n v o l v i n g the complete a n a l y s i s of the ending 'queen and pawn against queen' (for any position of the pawn) is a w a i t i n g completion. Here i t cannot be r u l e d out that positions w i l l be found in which the stronger side requires more than 100 moves to w i n . Secondly, the results of the a n a l y s i s of 'rook and bishop against rook 1 would be highly Interesting. It is t r a d i t i o n a l l y supposed that the majority of endings of this type are d r a w n . However, an exact a n a l y s i s of the positions is made d i f f i c u l t due to the large number of v a r i a t i o n s , and the absence of r e l i a b l e c r i t e r i a for e v a l u a t i n g the r e s u l t i n g positions. And it cannot be r u l e d out that the current opinion w i l l have to be changed... . I t would also be useful to use the computer to analyse c e r t a i n other endings which occur r a r e l y in p r a c t i c e , but which nevertheless remain problematic. These could include queen against two minor pieces (the most interesting combination of which is k n i g h t and bishop), and two minor pieces against a knight. Many questions associated

The Computer Analyzes the Endgame

121

with these sorts of positions have been accumulated by chess composers, and the computer could render them qualified assistance. Unfortunately, an Increase in the number of pieces on the board seriously complicates the problem. At the same time it Is to be hoped that the analysis of six-piece endings will be accessible to the next generation of computers.

Part Three Fifteen of the World Champions Best Games

Part Three FIFTEEN OF THE WORLD CHAMPION'S BEST GAMES

From 1960, when one of the authors of this book succeeded in naming the first category norm, to the present year, 1961, in which for -)e second time he will be defending his title of World Champion, I .e. Dver a period of more than twenty years, he has played about a thousand tournament games. Choosing fifteen of the best games out of such a number was no easy matter. I think that the chess content of the fifteen games chosen speaks for itself, but, so as to explain their Competitive Importance, each game is preceded by & short introduction. 3f course, the most decisive encounter in my chess career to date was the last game of the match for the World Championship in Baguio. [This game is given separately, in the f i r s t part of the book. Bearing in mind that the book is aimed at a broad range of Chess enthusiasts, in my comments on the games I have not tried to Jive an exhaustive analysis of variations, but have placed the Bmphasis on general evaluations of the resulting positions.

Moscow University Championship, 1968 Karpov-Gik Sici I ian Defence A. Karpov: This was my f i r s t tournament in the c a p i t a l , and in fcepder to 'conquer' Moscow it was essential to win i t . The next time 1 e to Moscow was three years later, when I was already a grandjltiaster, and in the very strong Alekhine Memorial Tournament I shared [ f i r s t place. 1 think that I f i n a l l y succeeded in 'subjugating' Moscow a ffurther three year-s later, when I won two Candidates' Matches in the [Capital - the quarter-final and the f i n a l . . . But I decided to take up permanent residence in Moscow only 10 years later! And who would lhave thought that my opponent in this game would later turn out to be |Tny co-author! Y. Gik: In !968 my opponent was a young master, and merely a f t l r s t - y e a r student in the Department of Mathematics and Mechanics. By ' t h a t time I had already finished my course in this department, and therefore I raied my chances in our game as h i g h e r . . . Nevertheless

125

15 of the World Champion's Best Games

126

the studeni defeated the graduate, and as a result became Champion of Moscow University, one point ahead of me. My opponent could not have guessed that he was playing against his future co-author, but I loo was not to know that within seven years my conqueror would become the strongest player on this planet! I eft c5 2 Nf3 d6 3 dft cxdft 4 Nxdft Nf6 5 Nc3 g6 6 Be3 Bg7 7 f3 0-0 B Bc4 Nc6 9 Qd2 QaS 10 0-0-0 Bd7 A.K.: This is the so-called Dragon Variation (from the formation of black pawns on the e- to h-filest. The plans for the two sides are wei I known — White attacks on the K-side, while Black aims for counter-play on the Q-side. Although the history of this variation is one of the most fascinating and mysterious in chess theory — how many times has it been completely refuted, only to be revived anew — it nevertheless has to be admitted that the expression 'dragon torment' was not devised for nothing: in this variation Black more often than not manages to get mated just before he reaches the enemy k i n g . Y.G.: I think that in the last few years the World Champion has played some fifteen games with White against the Dragon Variation, the majority of them with grandmasters, and has not conceded a single draw! II

h4

,

.

.

.

.

.

A.K.; The magazine Shakhmaty v SSSR 1968 No. 7 published the game BIkhovsky-Gik, which Black won. Y.G.: Wei I yes, since I had defeated our chief youth trainer, I thought that I would be able to cope somehow or other with one of his 'fledgl ings' . A.K.: Perhaps Black would have managed to do this, if I had played the same as BIkhovsky — 11 g4. But I had prepared for the game, and chose the more dangerous plan of advancing my rook's pawn. 11

Ne5

12 Bb3

Rfc8

Y.G.: Whenever moving the k i n g ' s rook to c8 is refuted, the analysts promptly suggest that Black play his queen's rook to c8. When this manoeuvre is also refuted, they once again recommend moving the rook from the f - f i l e to c8. This is how the theory of the Dragon Variation is created. A.K.: Incidentally, I first refuted ...Rac8, and in rather spectacular fashion It has to be said, In my first match with Korchnoi (cf. Game No. 5 ) . 13 h5 Nxh5

14 Bh6

S "*" S "."" ";""'§1**

it

nut

t

\zwSm m mm

> 13

Game No. 1

127

A.K.: At thai time this position was very fashionable, and an animated discussion was revolving around the variation with 14...Nd3+. I n i t i a l l y it was thought that this check made the bishop move to h6 Impossible, but some six months before the University Championship, at the Student Olympiad in Ybbs, the German player Duebal I showed that In this case White retains the advantage: 15 Kbl Nxb2 (15...Bxd4 16 Nd5!) 16 Kxb2 Bxh6 17 Qxh6, and now neither 1 7 . . . Qxc3+ nor 17...Rxc3 gives Black sufficient counter-play. In the present game 1 wanted to test my preparations, but my opponent deviated from my analysis. Y.G.:

It

would have been better

14 . . . Bxh6

if

I hadn't

done so.

15 Qxh6 Rxc3

A.K.: The standard exchange sacrifice in this v a r i a t i o n . Black removes once and for all the threat of the knight moving to d5, and hopes to get at the white king as soon as possible. s . '"

16 bxc3 Qxc3

Y.G.: It was d i f f i c u l t to imagine that this capture would be the decisive mistake. I knew that )6...Nf6 leads to a sharp game, but assumed that it wouldn't do any harm to take the pawn. 17 Ne2! A.K.; The start of a lengthy, forcing manoeuvre. copes splendidly with the task of d r i v i n g away the queen, same time joins the attack on the K-side.

17

The knight and at the

Qc5

Y.G.: In a certain theoretical a r t i c l e 16...Qxc3 was considered weak because of the reply 17 K b l . However, it then examined only [•17.. .Nci and 17...Nf6, whereas I was pinning my hopes on 1 7 . . . a 5 ! . The modest knight retreat came as a surprise to me. I saw to my regret that after 17...Nd3+ 18 Rxd3 Qa1+ 19 Kd2 Qxhi 20 g4 Ng3 21 Qxhi Nxhi 22 Ke3! and 23 Rd! my knight would be successfully [.Caught. My queen therefore had to retreat. IB g4 Nf6

g5 Nh5

20 Rxh5!

A.K.: There Is no time to lose. 20 Ng3, which I considered for long time, had to be rejected because of a spectacular reply which I noticed only at the last minute — 20.. .Bg4!, and the white queen is shut out of the game.

20

9xh5

21 Rhl Qe3+

22 Kb1 !

128

15 of the World Champion's Best Games

A.K.: The D-ragon V a r i a t i o n is renowned f o r the fact t h a i the s l i g h t e s t inaccuracy can spoil m a t t e r s . Thus, f o r example, 22 Kb2 would have g i v e n Black at least a d r a w : 22...Nd3+ 23 cxd3 (23 Kbi even loses after 2 3 . . . Q x f 3 ! ) 23...Qxe2+ 24 Kal Qxd3, and Black is assured of perpetual check. Qxf3

22

A.K.: The f a t e of the k n i g h t is u n i m p o r t a n t — it is the l i f e of the black k i n g t h a t is a l s t a k e : 22.--Qx.e2 23 Qxh5 e6 24 Qxh7+ Kf8 25 QhB+ Ke7 26 Qf6+ Ke8 27 RhS male. Y.G.: 1 saw that I would have lost a f t e r 2 2 . . . e 6 23 Qxh5 Qxf3 (23...Ng6 24 Qxh7+ KfS 25 Ng3 and 26 Nf5| 24 Qxh7+ Kf8 25 Nd4, b u l a f t e r my last move I f e l l contented.

23 Rxh5 e6 A.K.: Black cannot defend h7, since 23. . .Qxe4 loses lo 24 g 6 ! Qxg6 25 Rg5. It is also d i f f i c u l t f o r him to save the game after 2 3 . . . N g 6 , for example 24 Qxh7+ KfS 25 Rh6 e6 26 Rxg6 f x g 6 27 Qxd7 0xe2 28 Qxd6* Kg7 29 Qe7+ Kh8 30 Qf6+ Kh7 31 Qf7+ KhS 32 Qxg6.

24

g 6 !

.

,: ..,.,.

•

... -.

A.K. ; This b r a v e pawn s a c r i f i c e s i t s e l f to break down B l a c k ' s defences. The hasty 24 Qxh7+ would have al lowed the k i n g to escape: 2 4 . . . K f 8 , and now 25 QhB+ Ke7 26 Qxa8 f a i l s lo 2 6 . . . Q x h 5 , w h i l e 25 Nd4 is also not p o s s i b l e , in view of 25...Qd1+ 26 Kb2 Qxd4+. 2ft . . .

Nxg6

V.G.: Capturing with the pawn d i d not work — 24...fxg6 25 Qxh7+ Kf8 26 Qh8+ Ke7 27 Rh7+ Nf7 28 QxaB, but after the c a p t u r e w i t h the k n i g h l I a n t i c i p a l e d a q u i c k w i n — White's attack is r e p e l l e d , and in a d d i t i o n he is two pawns d o w n . 25 Qxh7+ Kf8

26 Rf5H

Game No. 2

129

Y.G.: For me t h i s move came I ike 3 bolt from the b l u e ! This elegant geometrical idea immediately decides the game. Two I ines, the aZ-gS d i a g o n a l and the f - f i l e , intersect a l the c r i t i c a l point f 7 . White threatens 0xf7 mate, and h i s queen is supported along the f i l e by the r o o k , and ( i n the event of 2 6 . . . e x f 5 ) along the diagonal by the b i s h o p . Black has n o t h i n g better than to p a r t w i t h his queen. Now I r e a l i z e d that I had been wrong to under-estimate the mathematical c a p a b i l I ties of my opponent. The p u r e l y geometrical s o l u i i o n to the position no doubt occurred to the student, t h a n k s to an intensive study of a n a l y t i c a l geometry! 26 . . . Qxb3+

exf5

27

28 Nf4!

A.K.: Another s t r i k i n g blow (perhaps bra!?). White e x p l o i t s the undefended state t o t a l l y destroys the black k i n g ' s s h e l t e r . 28

Bd8

29 Qh6+

A.K.: The final pawn with check. 20 . . . Ke8

from the f i e l d of a l g e of the rook at a8, and

slight

30 Nxg6 fxg6

finesse — White wants to capture the g6 31 Qxg6+ K*7

32 Qg5+!

A.K.; Accuracy Is required right to the end: Black could s t i l l have put up a defence, whereas defence against the rapid advance of the f-pawn. 32

Kefl

33 exf5 RcB

34 QgB+ Ke7

r

after 32 exf5 Rf8 now there is no

35 Qg7+ Resigns

Y.G.: A rare instance when the loser of a game also remained Content. Wfth the game, of course, but not the result of the tournament! In an article published in Shakhmatnaya Moskva (Moscow Chess) and devoted to Ihe University Championship, I complained that during the tournament the new champion had constantly been absent from Moscow, thus placing the other participants in an unfavourable position — they had to play earlier, thus disclosing their cards to the main rival. it is understandable that the USSR Team Championship in Riga, In which the mathematics student made the absolute best score (10 out of 11) was more important, because it was one of the elimination stages for the World Junior Championship. "But is not the Championship of the largest educational establishment in the world also of some importance!" I wrote somewhat reproachfully. Had I only known that within a few years. . . A.K.:

It

is

pleasant,

for

all

that,

to

remember

No. 2 Alekhine Memorial Tournament, Moscow 1971 Karpov-Hort Sic! I Ian Defence

one's

' •

youth!

'

This was my f i r s t really strong tournament, and it was after it that they began talking In the chess world about a new contender for the world crown. And, as it later turned out, not for nothing — in the very next cycle I became World Champion. The present game was the turning point in the Memorial Tournament. Up to the meeting with the

130

15 of

the World Champion's Best Games

Czech grandmaster- I had played a whole series of draws, winning only one game. This second victory gave me a creative and competitive Impulse. After it three more grandmasters were defeated, and r i g h t at the f i n i s h 1 succeeded in catching Leonid Stein. In the chess sense the present game is noteworthy for the unusual play of the k i n g ' s rook. 11 is a strange picture: this unwieldy piece l i t e r a l l y rushes aboul the board, its movements seemingly devoid of any sense, but with each move Black's position grows worse and worse... 2 Nf3 d6 The enduring the knight and to tactical grounds.

3 d4

Nxd4 Nf6

5 Nc3 e6

6 g4

weapon devised by Keres. The desire to drive back b u i l d up an attack on the K-side is supported on

6 . . . Nc6 6...h6

,

is more of a hindrance to White's p l a n .

7 g5 Nd7

•. , ;

• ,'t'

• ;._

8 Be3

A sharp game with good prospects for White results after 8 Ndb5 Nb6 9 Bf4 UeS 10 QhS g6 11 Qh3, bul a detailed examination of opening variations is not p a r t of our task.

8

a6

9 f*

Nowadays 9 Rgl is often preferred her«. '.i.

9 . . . Be7 On 9 . . . h 6 my opponent was most probably a f r a i d of the move which I was indeed intending to play — 10 Nxe6 ( s t r i c t l y speaking, there is no a l t e r n a t i v e , since 10 gxh6 and 10 g6 are simply bad because of 10.. .Qh4+). Boundless camp I ications could then have arisen; 1O...fxe6 11 Qh5+ Ke7 12 Bh3 Qe8 13 Qh&, w i t h consequences d i f f i c u l t to assess. It was just this that my tournament position demanded. 10 Rgl

Nxd4

Normally Black is in no hurry to make this exchange, but here it is d i f f i c u l t for him to f i n d an a l t e r n a t i v e . Thus 1 0 . . .0c7 looks too slow, it is not easy to decide on 10...0-0, straight into the attack, and 10...Nc5 11 Nxc6 bxc6 12 BxcS gives White a clear advantage.

12 Qd2

Qxd4 Bad for 16 Nd5.

Black,

of course,

13 BxfA Ne5 was

13...0b6

14 Rg3 Qxb2

15 Rb1 Qa3

U Be2 Bee Black cannot play actively ending is clearly advantageous to 17 Radi Be6 18 Kci . 15 Nd5

with 1ft.. .Qa5, since White: 15 Nd5 Qxd2+

the resulting 16 Kxd2 Bd8

Game No. 2 Immediately! Otherwise position (15 0-0-0 Qa5! ). 15 . . . Bxd5

131

the opposing queen takes up an active

16 exdS

Normally one tries to occupy the blockading square with a piece, and indeed, 16 Gxd5 would also have given me an advantage, since Black's d6 would have required constant defence. But ihen my e4 pawn would in some instances also have needed to be guarded, and this could have restricted my while-squared bishop. But now this bishop has freedom of movement, especially since its black opposite number has already left the board. 16

Ng6

B l a c k ' s p o s i t i o n is m a r k e d l y I n f e r i o r , and so Hort t r i e s lo f i n d a t a c t i c a l s o l u t i o n to h i s problems. He would have been condemned to p a s s i v e defence a f t e r c a s t l i n g short, as well as long ( a f t e r ' 16...QC7).

17 Be3 h6t? It is difficult to know what kind of mark to attach to this move, so great Is the significance, and not merely from the chess viewpoint, which is contained in i t . It shows fighting spirit, and the desire (o give the game a (actical turn. It also shows an accurate assessment of the position, with a clear understanding of its drawbacks. In [t'ithort, it is a move which is risky for both sides! 19 Kd1

18

The white king is not embarrassed by the loss of castling. At the same time, right to the end of the game Black's king has such a possibility open to it, but does not manage to utilize it. 19

gxh6

2O...Qf6 mobi li ty would regrouping of 1 In turn was

20 Bxh6 Bf6

probably did not appeal to Black, since his bishop's then be severely reduced. Hort had planned an ideal his forces, but in chess there are Iwo sides playing, and able to take the necessary counter-measures.

21 c3 Be5

It would appear that Black has achieved his aim — he threatens 22...Qh4, while after 22 Bg5 Qb6 2ft 8e3 0c7 White has achieved very l i t t l e . But White finds a very strong rejoinder.

15 of the World Champion's Best Games

132

22 Rg4! White is not concerned about his h2 pawn. What is most important at present is not to allow the enemy queen to take up an active position on the K-side, where, after a l l , White's pieces are s l i g h t l y 'hanging'. 22 . . . QfG

. . ' . . : • - .

Comparatively best was 2 2 . . . B x h 2 , r e - e s t a b l i s h i n g material e q u a l ity. But one can also understand the Czech grandmaster, who wanted to complete his development. . > . . . . . 23 h4!

.

.

.

Now 23.. .Nxhft? f a i l s to 24 Bg7, and in general it w i l l soon be d i f f i c u l t to win back this pawn, which u n t i l recently was so helpless, but which now g r a d u a l l y acquires formidable s t r e n g t h . 23. . .0-0-0 also f a i l s , to 24 BgS. 23 . . . Qf5

-

Black defends against hopes to castle Q-side.

the

threat

of

. 24

h5,

. and

. simultaneously

24 Rb4! An excellent square for the rook! 2 4 . . . 0 - 0 - 0 is s t i l l not possible, i n view of 25 Bg4, and meanwhile the b7 pawn is attacked. 24 . . . Bf6

25 h5 Ne7

,

. . . . . -.-.,- -- ..........

Of course, this is a less respectable square than e5, but here pseudo-activity on B l a c k ' s p a r t could have cost him a piece (after 25...Ne5 26 Rf4}. In passing, it should be noted that a l l this time Black has had no opportunity to s a c r i f i c e the exchange by . . .Rxh626 RfA Qe5

Game No. 3

133

27 Rf3! The u n w i e l d y rook d i s p l a y s m i r a c l e s of m a n o e u v r a b i l i t y , l l creates one threat a f t e r a n o t h e r , and operates e f f i c i e n t l y not only in a t t a c k , but also i n defence. Thus Slack achieves nothing now by 2 7 . . .Qxh5 28 Rxf6 QM+ 29 B f i (the rook defends both bishops) 29...Ng8 30 Qel+, and White w i n s . 27 . . . Nxd5

28 Rd3 Rxh6

There is n o t h i n g b e t t e r — o n 2B...Ne7 there follows 29 Bf4. 29 Rxd5 29 0xh6

would

be a mistake,

i n view of 2 9 . . .Bg5 and 3 0 . . .Ne3+.

29 . . . Qe4 Black appears to have everything in order,

but...

30 Rd3! That rook again! A player is accustomed to working hard, but I think it would be agreed that, rook, this applies mainly to the endgame.

all his pieces as regards the

30 . . . QM + Black

is forced

30 Kc2 Qxal

to go in

31 Qxh6 Be5

for

this

far

from equivalent

exchange.

32 QgS

Preventing Q-side castling, and creating irresistible threats against Black's stranded k i n g . in this hopeless position Hort overstepped the time l i m i t . This game was voted the best in the Alekhine Memorial Tournament, and the Yugoslav Infarmatoir called it one of the best creative achievements of the 1971 chess year. I must confess that 1 too am <jery fond of i t . . . .

No. 3 Quarter-Final Candidates' Match Moscow 1974 Karpov-Polugayevsky Sici I ian Defence This game is memorable mainly for the episode which occurred after Black's 16th move (I am saying nothing, of course, about the fact that victory in it brought me closer to a place in the Candidates' Seml-Finals). Usually in the Sicilian Defence, after Ihe exchange of the black bishop for a white knight at d5, there follows exd5. In the given situation this standard capture did not promise White anything in Particular, but in a normal game I would nevertheless have chosen i t . However, probably in view of the enormous tension and the fact that the finish of the match was In sight, I experienced a feeling of inspiration and excitement. The result was the move 17 Qxd5, which involved the sacrifice of two pawns. Many of the players present erroneously

15 of Ihe World Champion's Besl Games

134

decided that this was a p r e p a r e d v a r i a t i o n . The one to s u f f e r most in t h i s respect was P o l u g a y e v s k y himself — f o r a long time he looked f a r a forced loss, a n d f a i led to f i n d i t , but v a l u a b l e time h a d been wasted. Thus the unexpected c a p t u r e w i t h the queen on d5 h a d a decisive p s y c h o l o g i c a l effect on my opponent. 1 M c5 2 Nf3 d6 3 d& cxd4 4 Nxd4 Nfe 5 Nc3 a6 7 Nb3 Be? 8 0-0 Be6 9 U Qc7 10 a4 Nbd7 11 K h l 0-0

6 Be2 ,.5

This v a r i a t i o n of the S i c i l i a n Defence occurred each time t h a i Polugayevsky h a d B l a c k . D u r i n g o u r p r e p a r a t i o n s f o r the m a l c h , m y t r a i n e r , grandmaster Furman, a n d I had reckoned w i i h this p o s s i b i l i t y , and h a d o u t l i n e d the most expedient deployment of White's f o r c e s . The usual c o n t i n u a t i o n here is 12 f 5 , a n d i n 1971 I h a d managed to win in this v a r i a t i o n against Bronstein. I also p l a y e d t h i s l a t e r , i n the L e n i n g r a d Interzonal against B y r n e , though t h i s time w i t h o u t success. 12

Be3

. , • : : •

This idea is not new, but it acquired a second wind at high level after the game Geller-lukov, 1973. White maintains the tension in the centre, for the moment not disclosing his intentions. 12 . . . exf4

13 Rxf4 Ne5

14 a5

*'

m

'

In the fourth game of the malch I played the weaker 14 Nd4, and after 14...Rad6 Black secured a good game. 14 . . . Nfd7

15 Rf1 Bf6

16 Nd5 Bxd5

17 Qxd51?

Here there is something I should like to explain. Certain commentators sometimes accuse me of being a dry, rational, careful player. It is true that I have a practical approach, and that my game is based to a great extent on technique. I attempt to play 'correct' chess, and never take risks as, say, Larsen does. As White, like everyone else, I aim for an advantage from the very start, but as Black I try first of all to equalize. But when there is a choice of moves, ! by no means choose the simplest, but rather the most expedient. If there are several equally good continuations, my choice depends to a great extent on my opponents. Against Tal, for Instance, I prefer to 90 in for simple positions, which are not to his taste, whereas against Petrosian I try to make the play more complicated. But Ff I see only one correct path, then, whoever is playing against me, I go only along that one path. Let us return to the game. The sacrifice of two pawns, which I offered to Polugayevsky, was pure improvisation, as I have already mentioned. Even today 1 find this a ' t e r r i b l e ' decision on my part.

Game No. 3

135

To be f a i r , it should be mentioned that after the quiet 17 exd5 Nc4 18 Bxc4 Qxc& not a trace remains of While's opening advantage.

T'-

17

0xc2

Polugayeusky spent more than an hour wondering whether or not to take the pawn. But in the event of Black declining the sacrifice, he would have come under heavy positional pressure. 18 Nd4 0xb2 As they say, " I n for a penny, in for a pound!" The attempt by Black to limit himself to a small gain would have led to d position with material equality and a marked positional advantage for White, for example: 18...Qc5 19 NfS! Qxd5 20 exd5, and the black d-pawn falls. 19 Rab1 Qc3 19...Qa3 is weaker — the black queen would have been out of play on the edge of the board. 20 NfS The aim of this move is to defend the bishop and not allow the block queen to c5, which could have occurred after 20 Rb3 0c5 %\ NfS Qxd5 22 exd5 NcS, when Black can defend. 20

§

Qc2!

Polugayevsky manoeuvres excellently. Any delay would have allowed White to co-ordinate his pieces, and he was already threatening 21 Bd4 Qd2 22 Rb2, or 21 Rfcl Qa3 22 Rai Ub2 23 Ra2 Qb4 24 8d2!, trapping the queen. 21 Rbel An inaccuracy. After 21 Rfe1 White would advantage, whereas now Black finds a defence. 21

NcS

22 Nxd6 Ncd3

This is the point. white rooks would now be would have had only two Be7 27 R f i , or 24...Ne5 for White.

25 Qxb7

23 Bxd3 Nxd3

have

gained

a

big

24 Rdi Nb4

If on my 21st move I had played Rfel, the standing at b1 and dl . Consequently, Black possibilities: 2A...Nf2+ 25 Bxf2 Qxf2 26 e5 25 Nxb7, in each case with a clear i n i t i a t i v e

136

15 of

the World Champion's Best Games

Having been f i r s t two pawns down, and then one, perhaps I was subconsciously aiming to eliminate this 'material deficiency' . Meanwhile, 25 Qh5 would have threatened 26 Rxf6 gxf6 27 Qg4+ Kh8 28 Bh6, which is not easily p a r r i e d (25. ..g6 26 Qc5).

25

RabB

26 Qa? Qc6

The decisive mistake. Also bad was 26...Nc6 27 Qc7 28 Qxc2 Nxc2 29 Bb6, with advantage to White. But 26...Qe2 would have given Black some counter-play, although even here after 27 Bb6 the advantage is with White. 27 Bf4 Later 1 learned t h a t , in the press centre at t h i s p o i n t , Furman had shown that White could have won by the exchange sacrifice 27 Rxf6! gxf6 28 Bh6, threatening 29 Qe3; on 28...Nd3 ihere follows 29 BxfB, and on 28...Qc2 - 29 Rcl Qd3 30 Qc5!. He was no doubt r i g h t , but t h i s is not the only w i n n i n g p a t h . True, i t is s a i d t h a t , when I p l a y e d 27 Bf4 w i t h the threat of 28 e5, my t r a i n e r remarked contentedly " T h a t ' s also g o o d " . Indeed, the threat of e4-e5 r e s t r i c t s Black, and his k n i g h t f i n d s itself out of p l a y . 27

Ra8

Defending against 28 Qf2 Rad8

28 e5.

29 Qg3!

The threat is stronger than Its execution! often is Indeed the case. White does not h u r r y f i r s t increases the pressure. 29 . . . Qc3

attack

30 Rf3 Qc2

31 Rdfi

Bd4

Again p r e v e n t i n g the b r e a k t h r o u g h can no longer be h a l t e d .

Strangely enough, this to advance e4-e5, but

•"

by

the e - p a w n .

f

•"• i

However,

the

32 Bh6 White d i r e c t s

his

fire

simultaneously

against

two squares — f? and

97. 32 . . . Nc6 In the hope of 33 Nf5! Qb2

c o m p l i c a t i n g matters after

33 Nxf7 Q c 4 l .

Game No. 4

137

There would have been an interesting finish 34 Bxg7 Bxg3 35 Rxf3, when Black cannot avoid mate.

after

33.. .Be5

34 Bel! White vacates h6 for his knight with gain of tempo. longer any defence. 34 . . . Ob5

35 Nh6+ Kh8

Or 36...Kg8

37 Nh6+ Kh8

37 Rxf7 Bf6

38 Qf2 Kg8

There is no

36 Nxf7* Rxf7 38 Rxf8+. 39 Rxf6 gxf6

40 Qxf6 Resigns.

No. 4 Semi-Final Candidates' Match Leningrad 1974 Karpov-Spassky Queen's Gambit

- . -' • .*-.-. ' •

The concluding game of the match, victory in which took me into e Final of the Candidates' Event. Of particular interest in the game "is the theoretical duel. Prior to the match I had never opened with the queen's pawn against a strong opponent, and 1 d4 had been prepared as a surprise for Spassky. In the present game I managed to win bril lianlly in the so-called Makagonov-Bondarevsky Variation, It should be mentioned that for many years the late grandmaster Bondarevsky was Spassky's trainer, and It Is no accident that the given variation has given faithful service to the Ex-World Champion. d4 Nf6

2

e6

3 Nf3

I decided against 3 Nc3, so as to avoid Ihe Nimzo-lndian Defence, which worked well for Spassky in the f i f t h game of the match. 3 . . . d5 4 Nc3 Be7 ,9 Bxf6 Bxf6 10 cxdS exd5

5 Bg5 h6 11 0-0

6

0-0

7 e3 b6

8 Be2 Bb7

So far all according to theory, although II Rci is also to be considered, so as to answer 11...Nd7 with (2 NbS c6 13 Nd6. 11

Qd6

11.. .Nd7 is more oflen played, when White achieves nothing by 12 Qb3 c6 13 e4 dxe4 14 NxeA c5. But if he doesn't play e3-e4

136

15 of the World Champion's Best Games

immediately, there follows on the d4 pawn.

12...ReS

and

...Nf8-e6,

with a f u r t h e r

attack

12 Rd a6 It

is essential

to prevent

Nb5,

and

after

12...c6

the bishop at b7

is blocked In, allowing the e-pawn to advance. 13 a3 Nd7

14 b4 b5

In ihis v a r i a t i o n it is p r a c t i c a l l y impossible for Black to be the f i r s t to play a c t i v e l y . His task is to entrench himself and to be f u l l y prepared to meet any action by the opponent. By b2-b4 White has weakened his c4 square, and this gives Black c o u n t e r - p l a y .

15 Nei An inaccuracy. After 15 Nd2! followed have allowed the freeing advance . . . a 5 . 15 . . . c6!

by

Nb3-c5

While

need

nol

16 Nd3 Nb6?

Black f a i l s to u t i l i z e the opportunity o f f e r e d . By p l a y i n g 16...a5 he would have equalized easily, opening the a - f l l e and eliminating p r a c t i c a l l y all his weaknesses. Spassky presumably thought that he would always have time to move the a-pawn, but after White's next move his pieces prove to be very awkwardly placed. 17 a4! Bd8 It is d i f f i c u l t to condemn Black for this move. Faced with the necessity of conducting a passive defence, the Ex-World Champion decides to threaten mate, in the hope of e x p l o i t i n g the weakening of the opponent's castled position. In doing so, however, he disturbs the co-ordinal ion of his pieces: the rook at aB Is now doomed to remain where it is and defend the weak pawn. 17.. . RadS was essential, allowing the painless retreat of the bishop lo c8. With his rook at dS and bishop at f6, the central breakthrough would have held no great

terrors for Black, and the opening presented any particular danger. IS Nc5 Bc8

19 a5 Bc7

20 g3 Nc4

of

the

a-file

21 eA Bh3

would

not

have

22 Rei dxe4

This creates a new weakness in B l a c k ' s position — his c6 pawn. In a d d i t i o n , i t activates the white k n i g h t s , which gain the splendid t r a n s i t square e4. Spas5ky endeavours at a l l costs to prevent the blocking of the position. Although after e4-e5 Black loses any possi-

Game No. 4

139

b i l ity of active c o u n t e r - p l a y , the path to White's goal is also greatly lengthened. In the event of 22...Rfe6 23 8f3 dxe4? 24 N3xe4 Qg6 (24...Qe7 loses a pawn after 25 Nc3) 25 Bh5, Black cannot play 25. . .0h7 because of 26 B*f7+! Kxf7 27 0h5+. Comparatively best is 23...Be6, when, u n f o r t u n a t e l y , the combination 24 Nxa6 Rxa6 25 exd5 cxd5 26 Nxb5 Qc6 27 Nxc7 Qxc7 28 Bxd5 f a i l s to 25...Bd7. 23 N3xe4 Qg6

24 Bh5!

By 2 2 . . . d x e 4 Black o f f e r e d a pawn s a c r i f i c e , but White n a t u r a l l y r e j e c t s t h i s Greek g i f t . After 24 Bxc4? bxc4 25 Rxc4 f 5 ! a n d 2 6 . . . f 4 the opponent would have g a i n e d a very s t r o n g a t t a c k .

24 . . . Qh7 If

I

24...Qf5,

then

25 Rc3, and Black

has no good defence to g3-g4.

25 Qf3 f5? This loses by force, since White gives up h i s bishop to penetrate with h i s rook onto the seventh r a n k , and acquires splendid squares for his queen and k n i g h t s . After 25...Qf5 it would s t i l l not have been easy for White to realize h i s a d v a n t a g e . Here is how events might have developed: 26 Nc3 8d6! (26...Rfe8 27 Qxc6, or 26...Rfd8 27 Re7 Qxf3 28 Bxf3 Bd6 29 Rb7) 27 QxfS Bxf5 2Q Bf3, or 26 0xf5 Bxf5 27 Bf3 Rfc8! 28 Be2! Nd6 29 Bd3, w i t h advantage to White. 26 Nc3 g6

27 Qxc6 gxh5

28 Nd5

f A picturesque position! The black bishop is attacked, impossible to defend i t . In addition, White threatens 29 Re7. 28 . . . fh,

f

and

it

is

29 Re7 Of5

After 2O...fxg3 White wins most simply by 30 Nf6+ Rxf6 31 Qxa8+ Rf8 32 Qxf8+ Kxf8 33 Rxh7 gxh2+ 34 KM BU 35 Rei. Not 30 Rxh7??, of course, in view of 3O...gxf2+ 31 Kh1 fl=Q+ 32 Rxfl Rxfl mate.

KfB

30 Rxc7 Rae8 35 Nf4 Resigns.

31 Qxh6 Rf7

32 Rxf7 Kxf7

33 Qxf4 Re2

34 Qc7*

140

15 of the World Champion's Best Games No. 5 Candidates' Match Moscow 1974 Karpov-Korchno 1 S i c i l i a n Defence

Final

There Is no need for- me to t a l k about the competitive importance of this game. A victory at the very s t a r ! of [he match, and in such crushing s t y l e , could not fail to influence the e n t i r e subsequent struggle. The reader can see for himself that the game turned out to be u n u s u a l l y spectacular. 1 e4 c5

2 Nf3 d6

3 d4 cxd4

4 NxdA Nf6

5 Nc3 g6

At that time Korchnoi alone of the top players used to play the Dragon V a r i a t i o n , and in p a r t i c u l a r , this opening twice occurred in his 1971 Candidates' Match with Gel l e r . | n a t u r a l l y did not r u l e out the p o s s i b i l i t y of t h i s v a r i a t i o n o c c u r r i n g in our match.

Ne5

6 Be3 Bg7 12 0-0-0

7 f3 Nc6

8 Qd2 0-0

9 Bc4 6d7

10 t>4 Rca

II

Bb3

The immediate 12 h5 Is also p o s s i b l e . 12 . . . NcA

13 BxcA Rxc4

14 h5 Nxh5

15 g4 Nf6

16 NdeZ!

The logical basis for the retreat of the k n i g h t from the centre is roughly as fol lows. The c3 square is the most important point in White's s e l - u p , against which Black concentrates his a t t a c k . The standard exchange sacrifice . . .Rxc3 often occurs on Ihis square, when the opponent gains a strong a t t a c k . It is a curious fact that, after the doubling of the white pawns on the c - f i l e , B l a c k ' s position is so r i c h in p o s s i b i l i t i e s that even in the endgame he can maintain the balance, as was splendidly demonstrated in his games by the outstanding Soviet grandmaster Leonid Stein. Thus ttie basic idea of the k n i g h t move is to reinforce c3. i n a d d i t i o n , from e2 the k n i g h t can easily be t r a n s f e r r e d for a direct attack on the hostile k i n g . As the reader w i l l already have noticed, both p l a y e r s , without w o r r y i n g about loss of m a t e r i a l , are mounting attacks on opposite wings — as normally happens in positions w i t h c a s t l i n g on opposite sides. All these general considerations should not obscure the concr-ete threat created by White along the d - f i l e — 17 e5 and 18 g5. ; , .«

15 . . . Qa5

17 Bh6

A standard p l o y . In order to develop his attack successfully, White is obliged to exchange the bishop at g7, which Is the sole defender of the black squares around Its k i n g . Besides, this bishop, I ike a long-range gun, can often put White's Q-side under fire. Grandmaster Vladimir Si mag i n , a p l a y e r who was r i c h l y endowed w i t h creative f a n t a s y , even devised in such positions the thematic exchange sacrifice . . . B h 8 , merely so as to preserve his f a v o u r i t e piece from exchange.

17 .. 1?...Rfc8

18 Bxg7 Kxg7

IB Qxh6 RfcB

19 Bd3!

19 Qh6+ Kg3 transposes.

Game No. 5

141

Up to this point both players had been moving almost instantly, but here I played a move prepared beforehand, causing Korchnoi to spend a long time deep in thought. And indeed, there was plenty f o r him to think about. . . . It had been established that the theoretical continuation 19 Rd5 does not b r i n g White any real gains. For example: !9...Qd8 20 g5 Nh5 21 Ng3 Qf8! 22 Qxf8+ Rxf8! (earlier it was thought lhat Black had to take with the k i n g , but then White's position Is a l i t t l e better) 23 NnhS gxh5 24 Rxh5 f51. and things are slightly more pleasant now for Black. Or 20 e5 d*e5 21 g5 Nh5 22 Ng3 Qfg 23 Rxh5 (23 Nxh5 Qxh6 24 gxh6 Bc6! ) 23...gxh5 24 Rxd7 Qxh6 25 gxh6, and the outside passed pawn on the h-file gives Black good counter-chances. After 19 Rd3! the knight at c3 is overprotected, and in a number of variations the knight at e2 is freed for the attack. If, without this move, White had attempted to advance immediately — 19 g5 Nh5 20 Ng3, the unpleasant counter-blow 2O...Rxc3 would have awaited him.

19

R4c5

When during our preparations lor the match we analysed 19 Rd3, we came to the conclusion that the best reply to it was 19...R8c5. After 36 minutes of thought my opponent decided that he had to safeguard himself againsl the constantly threatening thrusls of the e- and g-pawns. I think nevertheless that Slack's best practical chance was 19...Qd8, recommended later by BoWinnik. But now, after spending 18 minutes in search of a refutation of the rook move, I found a fine forcing combination.

20 g5 The knights at c3 and f6 defend their kings, and therefore it is they that are subject to the greatest danger (the removal of the black knight from f6 will almost immediately lead to the intrusion of the , white knight at d5) . 20 . . . Rxg5 Of course, ^knight, remains

Rd5!

not 21 Nd5 Rxd5!, ' a ! ive' .

21 . . . Rxd5

when

Black's

chief

defender,

22 Nxd5 Re8

Here 22. . .Qd8 no longer works: 23 Nef4 0f8 24 Nxf6+ 25 0xh7 mate, while if 22...Nh5, then 23 Nxe7+ Kh8 24 Nxc8. 23 Nef4 Bc6

his

exf6

IS of Ihe World Champion's Best Games

142

The d5 square has to be a t t a c k e d , as otherwise there follows Nxf6+ followed by Nd5, a n d m . P S . On 23...Be6 I h a d p r e p a r e d 24 Nxe6 fxe6 25 Nxf6+ exf6 26 Qxh7+ Kf8 27 Qxb7 Qg5+ 28 Kbi Re7 29 0b8+ ReB 30 Qx a 7 ( b u t d e f i n i t e l y not 30 Rh8+?? K g 7 ! , when it is B l a c k , who threatens 31...Qg1 mate, who wins) 3O...Re7 31 OtiB+ Ree 32 0xd6+ - a r a r e a n d d i s t i n c t i v e type of ' w i n d m i l l ' .

24 e5! B l o c k i n g that possibilities, but in f a i Is to win by the the p o i n t ! ) 26 0xg5 24

same f i f t h r a n k . There is a d a z z l i n g a r r a y of fact this is the only d e c i s i v e c o n t i n u a t i o n . White s t r a i g h t f o r w a r d 24 Nxf6+ exf6 25 Nh5 Qg5+ ( t h i s is fxg5 27 Nf6+ Kg7 28 Nxe8* BxeB.

BxdS

After 2 4 . . . d x e 5

25 Nxf6+ exf6

26 Nh5, mate is I n e v i t a b l e .

25 exf6 exf6 In such positions the most important t h i n g is — s e l f - c o n t r o l ! One w a n t s , of course, to make a b r i l l i a n t move as q u i c k l y as p o s s i b l e , but here it was not yet too late to lose the game: 26 Mh5 would have been met by the sobering Z6...Re1 + . 26 Qxh7+ Kf8 If 2 7 . . . K e 7 ,

27 Qh8+ Resigns then 28 Nxd5+ Qxd5 29 Re1 + .

No. 6 European Team Championship Moscow 1977 Port i sch-Kar pov K i n g ' s I n d i a n Attack A most u n u s u a l game, i n that I succeeded in l i t e r a l l y c r u s h i n g my opponent in t w e n t y - t h r e e moves. I wonder i f a n y t h i n g s i m i l a r had happened to him before!? At one time I h a d a minus score against P o r t i s c h , then I e l i m i n a t e d i t . But i n the chess world the H u n g a r i a n grandmaster was c u s t o m a r i l y r e g a r d e d as a d i f f i c u l t opponent f o r me. The present game w i l l perhaps f i n a l l y r e f u t e t h i s o p i n i o n . . . . 1 Nf3 Nf6

2 g3 b6

It would appear

3 Bg2 Bb7

4 O-O e6

5 d J d5

6 Nbd2 Nbd7

that this move had not been played before.

An

Game No. 6

143

entertaining game was played in the First League of the 1976 USSR Championship: 6...Be7 7 e4 dxe4 8 dxe4 Nxe4? 9 Neb, and Black resigned, since on 9...Nd6 there follows 10 Bxb7 Nxb7 11 Qf3, with a double attack on f7 and b7 (Kochiev-I vanov). 7 R*1 Bc5 The combination of this bishop manoeuvre with the previous knight move is a completely new idea. As a result Black has immediately solved all his opening problems. 8 c4 The innovation puts Portisch out of his stride, and he avoids the traditional advance of the e-pawn. An equal game would have resulted from 8 eh dex4 9 Ng5 e3 10 fxe3 Bxg2 11 Kxg2 Nd5 12 Nde4, whereas now it becomes clear that White has made a pointless move with his rook to el . 8

0-0

9 cxd5

10 Nb3

It is understandable that the bishop at c5 should be playing on White's nerves, but at b3 the knight will be badly placed.

10 . . . Bb4! Weaker is 1O...Bd6

11 Nbd4 Re8

12 Nb5.

11 Bd2 a5 Biack gains some advantage by aiming for more complicated p l a y .

11...6xd2

12 Qxd2 c5, but

I was

12 Nbd4 Re8 There is no point in immediately d r i v i n g away the knight by 12.•• c5. The rook move is a l l the same useful, and White has nothing better than to 'force' the advance of the c-pawn. 13 Rci A risky the centre.

c5

14 Nf5 Nf8

decision,

15 . . . Ne4 This move is

since

15 d4? Black

Is

better

prepared

for

a

clash

in

16 dxc57 associated with an oversight

in the resulting compl'-

15 of the World Champion's Best Games

144

cations. Perhaps Portisch ever, White was already in 16 a3, then 16...Bxd2 17 formed from an active piece

16

Nxd2

simply over-looked Black's 17th move? Howconsiderable d i f f i c u l t i e s . If, for instance, Nxd2 Qf6, and the knight at f5 is transinto a highly vulnerable one.

17 Nxd2 Qg5!

Suddenly bolh knights are ' h a n g i n g ' . 18 Ne3 loses immediately to 18...Rxe3 19 fxe3 0xe3+ and 10...Bxd2, so White has no choice. 18 Nd6 Bxd2 Here, on the other hand, White has a wide choice, b u t . . . a cheerless one: 19 ha Qh6 20 NxeB (after 20 NfS 0f6 White loses a piece) 20... Rxe8 21 c6 Ba6, and White's position is d i f f i c u l t . More interesting is 20 Nxb7, and if 2O...Bxe1, then 21 Q*el Rxe2 22 Qxe2 Qxcl+ 23 Kh2 (the king has an escape square — an important contrast to that which occurs in the game) 23...bxc5 24 Bx65, and White gains some counterplay. The whole point, however, is that after 19 h4 Qh6 20 Nxb7 Black takes the other rook — 2Q...Bxc1 21 Qxcl Rxe2, with a quick win. While can also move another pawn — 19 c6, when there follows 19...Ba6 20 f4 (20 NxeB Rxe8) 2O...Qe7 (better- than the immediate 20. ..Be3+, when 21 Khi Ge7 22 BxdS is possible), and on 21 NxeB Black has the zwischenzug 21...Qe3+! 19 Nxb7 Bxel

20 Qxel Rxe2

21 0xe2 Qxc1+

After 23 c6 Black wins by 23...RC8 23

Rc8

22 Qfi Qd2!

23 cxb6

24 Qb5 Qc1+ and 2S...Qxc6.

White resigns

He either loses his queen, or Is mated after 24 Qb5 Rc2!

No. 7 'Tournament of S t a r s ' , Montreal 1979 Timman-Ksrpov E n g l i s h Opening T h i s game is noteworthy f o r the f a c l t h a t i n i t t h e Dutch g r a n d master f e l l into a t r a p , p r e p a r e d . . . f o r K o r c h n o i . It should be said that f o r the match i n Baguio we p r e p a r e d a number of opening s u r p r i s e s , and by no means a l l of them were u s e d . Several ideas a r e s t i l l a w a i t i n g t h e i r t u r n , and it is q u i t e impossible to tell who w i l l be

US

Game No. 7

their 'undeserving' victims. Incidentally, in Ihe same Montreal tournament, Portisch too was unlucky, Playing the French Defence with Black against l a I, he also ran into a v a r i a t i o n prepared for the World Championship Match.

1 c
3 Nf3 Nc6

A e3

In Baguio the continuation 4 g3 Bb4 was Intensively Timman chooses a different v a r i a t i o n of the English opening.

tested,

so

4 . . . Be7 4...Bb4 is very popular here, and now 5 0c2 0-0 6 Nd5 Re8 lat the 1978 Amsterdam Tournament, Timman himself, with Black against Adorjan, continued 6...a5 7 a3 Bc5 8 Ng5 g6 9 Nxf6+ Qxf6 10 Ne4 Qe7 11 Be2 d6 12 b3 Ba7 13 0-0 f5 14 Nc3 f i . 15 Nd5 Qg5 16 Kh1 6h3 17 gxh3 f3 18 Bxf3 Rxf3 19 Qe4 Raf8, but after 20 Qg4 White retained a slight advantage) 7 Qf5. It is in this direction that theoretical searchings have developed in recent times. Thus Ronianishin, as' Black against Kasparov in the 1978 USSR Championship (Premier League), employed here an interesting, but hardly correct pawn s a c r i fice: 7...Be7?! 8 NxeS Nb4 9 Nxb4 Bxb4 10 a3 d5?! (White also retains a certain advantage after 10...d6 11 Qc2 Bxd2+ 12 Bxd2 dxe5) 11 Qc2 Rxe5 12 axb4 Bf5 13 d3 0e7 14 Bd2 d4 15 Be2 dxe3 16 fxe3 Bg4 17 Bf3!, and in the end the compi ications proved favourable for White. 5 d * exd4

6 NxdA 0-0

7 Nxc6

Also possible is 7 Be2, delaying the exchange on c6. After 7. , . Re8 8 0-0 Bf8 9 b3 d6 10 Bb2 Bd7 11 Rci Nxd4 12 Qxd4 c6 13 Rfdi (Stean-Garcia, Lone Pine 1978} White retained a si ight opening advantage. 7 . . . bxc6

8 Be2 d5

9 0-0 Bd6

Al I the indications are that Black stands wel I, and on the f l a n k s . It is important for him merely actions of his pieces. 10 b3 Qe7

both in the centre to co-ordinate the

Bb2 dxc4!

This idea was conceived during our preparations for the World B6hs
Black

Capturing with the acquires all kinds

bishop weakens White's control of g4, and of attacking possibilities, such as 12...Oe5

15 of the World Champion's Best Games

146

13 g3 Bh3, 12...Ng4 13 g3 Nxh2, or even 1 2 . . . B x h 2 + ! ? queen's rook comes into p l a y w i t h g a i n of tempo.

But

now

the

12 . . . R68! White has to be on his guard: double attacks are In the air. For example, 13 Bb1? Rxb2 14 Rxb2 Qe5, etc. 13 Qd Ng4 Black's altack develops easily and spontaneously. White has noi even a trace of an advantage left; on the contrary, he has to follow with anxiety the development of events on the K-side. 14 g3 After 14 Bxg4 Bxgi 15...Rb4!

IS Rei Slack can increase the pressure by

14 . . . Re8 Of course, I could have played ! 4 . . .Nxh2 15 Kxh2 Qh4+ Immediately, forcing a draw, but I already had every justification for playing for a win. After 14...Re8 the combination with the knight sacrifice at h2 will not be so inoffensive, since the rook can be brought up with decisive effecl onto the sixth rank (15...Nxh2 16 Kxh2 Qh4+ 17 Kg2 Qh3+ ie Kgi 6xg3 19 fxg3 Qxg3+ 20 KM Re6, e t c . ] . 15 Nd1 I think that White is also In trouble after recommended by certain commentators. For example, 16 Be2 Bh3 17 Rdi is less clear) 16 Bxg4 (bad is Ne5, etc.) 16...Bxa4 17 f3 Bh3 18 Rf2 Qg6, with for Black.

15

IS Bf3, which was 15.. .Of6 (15.. .Me5 16 Bg2 Qh6 17 h3 a strong initiative

Nxh2!

A sacrifice, By which Black Increases his advantage. Tinman foresaw this, of course, but was pinning his hopes on the strength of his answering intermediate move. But before beginning the combination I had taken accurate account of all possible tactical nuances. 16 c5 Qxg3+

Of course, not 16 Kxh2 Qh4+ 17 Kg2 Qh3+ 18 Kgi Bxg3 19 fx9 3 20 Khl Re4! <20...Re6 21 Bf6) 21 Rf4 Bh3, and it is all over.

A

H?

Game No. 8 16 . . . N x f i !

17 cxd6 Nxg3!

l l was this blow, completing ihe destruction of White's K-side, thai Ihe Dulch grandmaster overlooked in his calculations. Since 18 dxe7 Nxe2+ 19 Kf! Nxci Is completely hopeless for White, he has no choice. 18 fxg3 Qxd6

19 Kf2

Equally cheerless is 19 Kg2 Qh6 19 . . . Qh6

20 Bd4

Or 20 Qc3 Re6, b r i n g i n g 20 . . . Qh2+

up new reserves

into the attack,

21 Kel Qxg3+

White's days are numbered. b i g material advantage. 22 Kd2 Qg2 27 Kc3

20 g4 Gg5 (or 2O...Rb4).

23 Nb2 Ba6

Black

has a powerful

24 Nd3 Bxd3

initiative

and

25 Kxd3 RbdB

26 Bfi Qe4+

white

This

:S 27

cS!

Clearing the final quickest way to w i n .

28 Bxc5 Qc6 29

approaches

Rb8+

to

the

30 Ka3 Re5 31

king.

is

the

Qb6

White resigns.

No. 8 Bad Lauterberg 1980 K a rpo v -H u b n er Sicil ian Defence This is, I t h i n k , one of the best tournament games of 1980. It very r a r e l y happens lhat both players play practically faultlessly, especially when the game is abounding in tactical nuances. Here a sharp skirmish concluded in a forced win for While. 1 e4 c5 Hiibner

2 Nf3 d6 first

invites

me

to

play

the

Najdorf

Variation,

and

then

1*8

15 of the World Champion's

switches

9 KM

Best Games

to Ihe Scheveningen.

3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 a6 6 Be2 e6 7 0-0 Be7 Nc6 10 Be3 Bd7 11 Qei Nxd4 12 BxO4 Be6 13 Qg3 b5

S U 0-0 U a3 96

An attempt to take the pressure off g7 as soon as possible. this well known position Black usually plays . . . g 6 a l i t t l e later. 15 Bf3 Qd7

16 Radi Qb7

In

17 f5

The most vigorous way of exploiting the early advance of Black's g-pawn. The simple variation 17.. .Nxe4 18 Nxe4 Bxe4 19 f6 Bd8 20 Qh4 Bxf3 21 Qh6 Bxg2+ 22 Kgi demonstrates the immunity of White's e-pawn. 17 . . . e5

18 Be3 b4

19 axb4 Qxb4

20 Bg5 Qxb2

21 Rd3

Creating the vei led threat of (rapping the queen by 22 Rbi 23 Bdi . Bui in the meantime White gives ur> a second pawn.

Gxc2

21 . . . Qxc2

22 Bdi The a l t e r n a t i v e , also a p p a r e n t l y good, was joint analysis with my opponent after the game discover any decisive c o n t i n u a t i o n . 22 . . . Qb2

23 Bxf6 Bxf6

22 R f d 1 , but in a we were unable to

24 Rxd6 Bb5

On 2 4 . . .Bh4 there would have fol lowed not the inexact 26 f6 in view of 26...Qci> 27...Kh8.

25 Qxh4 Qxc3 26 Qh6, but 27 Rg1 (or 27 Rf3 Qxe4)

25 Rf2 Qct After the game HUbner s a i d that 25...Qa1 would have been better. This is indeed so, had White repeated what occurred in the game. But I had i n mind 26 fxg6, and on 2 6 . . . B g 7 27 gxf?+. 26 Rxf6 The preceding t a c t i c a l play has led to mutual nevertheless sharperied my t a c t i c a l v i s i o n .

time t r o u b l e ,

which

149

Game No.

White has an amusing win here after 26...Rad8: 27 fxg6 Rxdl + 28 Nxdi Qxd1 + 29 Rfi Bxfi 30 gxf7++ KhB 31 Qxe5 Bxg2++ 32 Kxg2 Qg4+ 33 Kf2 Qg7 34 h 4 ! , and this pawn, on reaching h6, decides the outcome. 27...hxg6 28 h3 Rd3 29 Qxg6+ (t w i l l not say that this is the only solution, but it is c e r t a i n l y the most spectacular) 29...fxg6 30 Rxf8+ Kh7 31 R2f7+ Kh6 32 Rh8+ Kg5 33 h* mate. 26 . . . Rac8

27 Re2 Qai

28 fxg6

hxg6

On 28...Rxc3 the f a m i l i a r mechanism again operates: 29 gxf7++ KhS 30 Gxe5 Qxdl + 31 Rf1 mate. In this case an 'optical i l l u s i o n ' can arise: 3O...Re3 (3O...Ra3 31 R b 2 ! l , and Black appears to divert the queen by himself threatening mate. But normal vision is restored immediately: 31 Qxa1 ! Re1 + 32 Rf1 mate. 29 Rd6 Rc7 After 29...Rfd8 mates by 33 Rc8+ Kh7 30 OxeS Rfc8

30 Qxe5 34 RhB.

31 Qd5 Kg7

Rxd6

31 Qxd6 Rxc3

32 Qd4+ Kh7

32 Qd4 Ra3 White

33 Nxb5 Resigns

No. 9 Bugojno 1980 Karpov-Tal Slav Defence In this game, which seems to me to be a very complete one, I employed an innovation in the so-called Meran V a r i a t i o n . True, I learned of this from grandmaster Gligoric, who said that the move employed by me had never been played before, and that he had been keeping it secret for more than a year. What is s u r p r i s i n g is that I made this move (the move in question is c a s t l i n g ) after only a few minutes' thought, and that I had never analysed the position which Tal went in f o r . I think that the reason why I q u i c k l y found the correct move was that I looked at the position from a fresh viewpoint. The continuations recommencJed in theoretical guides never even occurred to me! I should also mention that in this game, after a whole series of draws, I for the f i r s t lime gained a win over Mikhail T a l . The Ex-World Champion and I are on f r i e n d l y terms, but I had no choice — only a win gave me hopes of catching Larsen, who was confidently leading the tournament. e6

2 Nc3 d5

c6

Nf6

5 Nf3 Nbd7

6 Bd3

ISO

15 of the World Champion's

7 Bxc4 bS 13 0-0

8 Bd3 a6

This is was p l a y e d .

9 e4 cS

10 dS c4

the afore-mentioned

13 . . . Bb7

Best Games

11 dxe6 t x d 3

innovation.

Usually

12 exd7+ Qxd7

13 6g5 or

13 e5

14 Rei Bb4

H . . . 0 - 0 - 0 is r i 5 k y in would have followed 15 BgS.

view

of

15 Bf4,

while

on

U,..RdB

there

\ 15 Ne5 Qe6 Other moves also f a i l to <3ive complete e q u a l i t y , for 15...Qe7 16 Nxd3 Bxc3 17 bxc3 0-0-0 18 f 3 , or 15...Qd4 0-0-0 17 Nxb4 Qxb4 18 Qc2. 16 Nxd3 Bxc3

example, 16 Nxd3

••-•"'•

17 Nf4I

Black

This intermediate move is avoids the chief dangers.

the

17 . . . Qd7 Unsatisfactory faMs to 20 Bb2.

whole

point.

.. is

17...Qe5

After

...

18 bxc3 0-0

.

;.

19 Nd3,

17 bxc3 O-O-O

>; . when

19...Qxc3

18 bxc3 Nxe« Afier 16. , .0-0-0 (this possibility has constantly to be taken inio account) 19 Oxd7+ Rxd7 20 f3, or 18...Gxd1 19 Rxdi Nxe4 20 Ba3 White retains an advantage. 19 Q*d7+ Kxd7

20 B«3 Rhe8

21 Red1+ Kc7

22 f3

The knight Is diverted from the defence of the important squares, and White, exploiting the fact that the opponent's queen's rook Is not yet in play, succeeds In exposing the position of the black k i n g and mounting an attack on i t . 22 . . . Nf6 23 BdfrH Kb6 24 C4 RecB 27 axb5 Rd7 28 Rd4 Rcd8 29 Rodi RcB

Rxd4

29. . ,g5 allows 31 Bxd8+ Kc5

25 cxb5 axb5

White to gain a decisive 32 Rxdfi Kxd4 33 Bxf6+.

advantage

26 a* RedS

by

30 Be7l

Game No. 10

151

30 Be5 Re7 Afier 3O...Rxd4 31 Bxdfc+ Kxb5 no chance of saving Ihe game. 31 Rd6+ Kxb5 Kc4

32 Bxf6 gxf6

33 Rd7 Black has

32 Rbl+ Kc4

The noose around the black king Is tightening: 32...Kc5 33 Nd3+ 34 Rb4 mate, or 32...Ka5 33 Rd2 Rc4 34 Ra2+ Ra4 35 Bc3+. 33 Rd4+

33

mate)

Kc5

Mate also follows after 33...Kc3 35 Rb2+ Kci 36 Ne2 mate.

34 Rd3+ Kc2 (34...Kc4

35 Rc3

34 Nd3* Resigns Mate cannot be avoided: 34...Kc6 35 Rcl+ Kb6 37 Ra1 + Ba6 38 Bd4+ Ka8 39 Rxa6+ and 40 Rxa7 mate.

36 Rb4+ Ka7

No. 10 Bugojno 1980 Gl igoric-Karpov Nimzo- Indian Defence In my notes to the previous game I mentioned that in my game with Tal I made a desperate attempt to catch Larsen in this international tournament. Four rounds from the f i n i s h this seemed impossible, since the gap was one and a half points. But hy gaining three successive wins I drew level with the Danish grandmaster. This f i n i s h i n g spurt so affected my r i v a l that he faltered, and in the last round missed a win in a won ending. This now gave me chances of sole f i r s t place, for which I had to defeat Gligoric with Black. This victory did not come easi l y . In an inferior position the Yugoslav player set me enough! traps to suffice for several games. By f i n a l l y breaking the resistance of the experienced grandmaster, I gave myself a nice present (the last round took place on my b i r t h d a y ! ) . d4 Nf6

3 Nc3 Bb&

4 e3 c5

5 Nge2 cxd4

In the f i f t h game of the World Championship Match in Baguto I preferred 5 . . . d 5 . After 6 a3 Bxc3+ 7 Nxc3 cxd4 8 exd4 dxc4 9 Bxc4 Nc6 10 Be3 0-0 11 0-0 b6 12 Qd3 White retained a minimal positional advantage.

152

15 of the World Champion's Best Games

6 exd4 0-0 Nxd2

The t h i r d game of the same match went 6 . . .d5 9 Qxd2 a5. 7 a3 Be7

8 d5 exd5

7 c5 Ne4

8 Bd2

9 cxd5 Re8

After 9 . . . d 6 10 Nd4 While has time lo play Be2 a n d 0 - 0 , successfully removing h i s k i n g from the centre a n d r e t a i n i n g an a d v a n tage i n space.

10 d6!? An i n t e r e s t i n g , a l t h o u g h r i s k y move. The pawn w i l l undoubtedly be lost, but in this time White hopes to g a i n a lead i n development. The normal c o n t i n u a t i o n is \0 g3 Bc5, w i t h an u n c l e a r game. 10 . . . Bf8 11 g3 Re6 15 Nf4 Re8 16 b4

12 Bg2 Rxd6

13 Qc2 Nc6

U

0-0 Re6

Wilh his four^move rook manoeuvre Black has won the 'thematic' pawn, but has fallen somewhat behind in development. It is not so easy, however, to exploit this factor, for example: 16 NfdS Nxd5 17 Nxti5 d6 18 Bf4 Bg4, etc. 16 . . . d6

18 Bb2 Bd7

18 Ncd5

A complicated game results from 18 b5 Na5 19 Qd3 Rc8, but 18 NeA was to be considered. After 18...Nxe4 19 Bxeft h6 20 Bd5 White retains the necessary compensation for the pawn, so Black would probably have had to play 18...Ng4. 18 . . . NxdS

19 Bxd5 Rc8

20 Qb3 Re7

21 Radi

On 21 Rfel Black has the defence 21...Be8, when his plans include exchanging on el and transferring the second rook from c8 to e7.

Game No. 10

153

21 . . . b5! S t a b i l i z i n g the position on the Q-side. On 21...Bg4, with the idea of consolidating by 22...Od7, White replies 22 f 3 . 22 Sg2 Re8 Here on 22...Bg4 there could have followed not only 23 f3 Bf5 24 Rd5, but also 23 Rd2 with the threat of 24 h3, in both cases with sufficient chances for White. 23 Bd5 Re7

24 Bg2 Ne5!

A f u r t h e r retreat by repetition of the p o s i t i o n . ini tiative_ 25 RxdG Nc4

the rook would have led to a three-fold By r e t u r n i n g the pawn Black assumes the

26 Rddi

The sharp 26 Rd5 was an interesting a l t e r n a t i v e . 26 . . . Qe8 3! h4 h6 32 Bf3?

27 Bd4 Bc6

An unfortunate decision. merely a s l i g h t advantage.

28 6c5 Re5 After

29 BxfB Kxf8

32 Kh2

Bd?

30 NdS Kg8

33 Nf4 Black

has

32 . . . Bd7 An inaccuracy in r e p l y . White's erroneous bishop move could have been exploited by the spectacular manoeuvre 32,..Qe6 33 Nf4 (33 Qb3 Nb2) 33...Nd2L!, for example: 34*Nxe6 Nxf3+ 35 Qxf3 ( k i n g moves are met by 35...Nd4+) 35,..Bxf3 36 Rd8* Rxd8 37 Nxd8 Re8!, p i c k i n g up the k n i g h t . 33 Nf4 BfS

34 Nd3 Re?

35 Bg2 Be6

36 Qbl Nxa3

37 Qa1 Nc2

B l a c k ' s advantage is now d e f i n i t e . 38 Qb2 Bf5

39 Kh2

After 39 NcS Nei 39 . . . Rc4

40 Bd5 Bh3 the white rook is amusingly

40 Bd5 Rd4

41 Nf4 Rxb4

42 8b3

Better defensive chances were offered by 42 Qc3, w i t h tion of p l a y i n g 43 Bb3 and 44 Nd5. A2 . . . Rxf4

43 g x f i Re2

trapped.

the

inten-

44 Kg3!

On 44 Bd5 Black has not only 44...Ne3 Bxg&, but also the more vigorous 44...Qe31.

45 Qxe2 Ng4t

46 Qxg4

44 . . . Bg6 Not the careless 44...Qc6? 45 RdB+ Kh7 47 B98+ Kg6 48 h5+! Kxh5 49 Qxe2+. 45 BxcZ

46 Bxf7 Ne3, in view of

15*

15 of the World Champion's Best Games The t h r e a t was 45...Ne3 45 . . . Bxc2

46 Rd6 f6

48 Kf3 0h5+

49 Kg3 is u n p l e a s a n t l y

46 . . . Qg4 53 Kei Bxf5+ Kh7

46 Qxe2 NfS+. 47 Qd4 Qg6+

49 R g i Qxh4+

48 Kh2 met by 4 9 . . . B e 4 .

SO Kg2 Re?

51 f 5 Be4+

52 K H Qh3+

53 . . . Qxf5 is i n a c c u r a t e : 54 Rd8+ Kh7 55 Rh8+! Kxh8 57 0xe7, a n d unexpectedly it is White who w i n s .

55 Qd8+

54 Kd2 Kh7 55 Qf4 b4 56 Rg3 Qh5 57 Qf3 0 x f 3 58 Rxf3 Bc8 59 Rf4 Rb7 60 Kc2 a5 61 Kb2 a4 62 Ra6 Bd7 White r e s i g n s -••<-•"!£'!.

No.

11

.*:;•!

Amsterdam 1980 Karpov-Ribli English Opening

•'• .;._

..

•-

» ,;•

•. .- ' 7 ' :

•

The International lournament in Amsterdam was a double-round event. In the first of these Ribli defeated me, and, what's more, in the endgame. Since I do not especially like to remain ' i n debt', in the second cycle I aimed for revenge. I succeeded in this, and also in the endgame! In the competitive sense the game was doubly important, since my pursuer, grandmaster" Timman, was only half a point behind. The game, in my opinion, was highly interesting In all its phases. 1 c4 c5

2 Nf3 Nf6

3 Nc3 Nc6

4 g3 dS S cxd5 Nxd5

6 Bg2 g6

When before the game I was considering the possible opening schemes, this variation was the one I least expected of the Hungarian grandmaster. Usually he employs King's Indian set-ups, Of" else the GrunfeiO Defence. 7

0-0 Bg7

. i , • » , . ! • . ....*'•.-{'

Her-e I remembered that five years earlier I had analysis of this position, but had not found anything equality.

8 Nxd5 Qxd5

' • . . ] ' .

9 d3 0-0

10 Be3 Bd7!

a

mtm

, . • - , . • .

. • JJ-*#, •'-.

done much more than . v "

•-'•

tit B tiW

• H • B 1 £G9 BJ • "p

This modest bishop move is the whole point. Earlier Black used to be sfraid of the various possible moves by the white knight, and

J

Game No. 11

155

played 1O...Qd6, which gave White a definite lead in development. But then it became clear that Slack need not waste time on retreating his queen. On 11 Nd2 It takes up a good post at h5, while after 11 Ng5 Black has the perfectly sound 11 . . .QeS, with an attack on the b2 pawn. 11 NdA Nevertheless, it is worth d r i v i n g away the black queen — as long as it remains on d5, it ties down the rook at ai . 11 . . . Qd6

12 Nxc6

12 Nb5 is parried by the same 12...Qe5. 12 . . . Bxc6

13 Bxc6 Qxc6

U Rc1 Qe6

The c r i t i c a l position of the entire v a r i a t i o n . White is obliged to capture on c5, otherwise by 15. ..b6 Black stabilizes the position and obtains a spatial advantage. 15 Bxc5 Bxb2 16 Rbi , with the i n i t i a t i v e , looks tempting. At f i r s t it might seem that White also gains an advantage after 15...Qxa2 16 Bxe7 Re8 17 Ba3, but 17...Bxb2 turns out to be possible, when a draw Is imminent. Capturing with the rook is therefore the only move. 15 Rxc5 Qxa2 Not 15...Bxb2

16 Rb5 Bf6, when 17 Qa4 is highly

unpleasant,

16 RbS! An innovation, thought up by me during ihe game. After the standard 16 b4 White gets into difficulties — the a1-h8 diagonal is completely opened, as weli as the second rank. I cannot say that the new rook move gives White a big advantage, but the main thing is that Black is now faced with problems which have to be solved at the board. 16 . . . b6 The b2 pawn Is Immune - )6...Bxb2 17 Qc2. The other possib i l i t y was 16..,Qs6, but then 17 Qb3 nevertheless forces . . . b 6 . 17 Qatt The idea of this manoeuvre is to tie the black rook at a8 to the defence of the a-pawn. If Black were able to transfer his rook from a8 to his second rank, he would immediately gain a slight advantage in the ending.

156

15 of the World Champion's Best Games 17 . . . Qxal?! "i

17...Qe6 was preferable. Black does not sense tne dangers awaiting hirn in the ending. After the exchange of queens the best that he can hope for is a rook ending wilh four pawns (e, f. g and h) against five (d, e, f, g and h ) . 18 Rxa1 Rfb8 Hxe2

After 18...Rfc8 White gains an advantage by 21 Rbxb6 followed by winning the a - p a « n .

19 Ra6 Rc2

20 b3

have no reasonable

moves.

19 Ra6 Kf8 Practically 20 Rb4 Be5

forced

—the other pieces

21 RbaA

On 21 b3 Black has time to defend by 21...Bc7

22 Rba4 Rb7.

21 . . . b5 Black should have played 21...Bxb2 22 Bxb6 Rb7 when the resulting ending with five pawns against four easy to w i n . 22 Ra2 Rb7

23 Bxa7 Rc8, is not at a l l

23 b3

By p l a y i n g 23 Bxa7 Bxb2 24 Rxb2 Rxa7 25 Rxs7 Rxa? 26 Rxb5 White could have forced a rook ending with an extra pawn, but I reckoned lhat White's positional pressure offered better chances. 23 . . . Bb8

24 Bc5 Ke8

25 d4 Kd7

26 e4

A crucial decision, demanding deep calculation. 26 d5 would have forced Black into a d i f f i c u l t rook ending; 26..,Bd(j 27 Bxd6 exd6, but at that moment 1 recalled the chess saying which runs " a l l rook endings are d r a w n . . . " . 26 . . . e6

27 b4 KcB

.

•• •

,

•.

• .. v, ..

28 d5 There is no time to lose, since after 28 Kg2 Rd7 39 R6a5 Bd6! 30 Rxb5 Bxc5 31 Rxc5+ Kd8! White's king is not in time to defend his far-advanced pawns. 28 . . . exd5

29 exd5 Rd?

Game No. 12

157

Black has no tlrne to move h i s bishop out of 30 Re2, w i t h the threat of a b r e a k t h r o u g h on eB. 30 d6 Rd8

the cage — 29...Be5

31 Kg2

The o p p o n e n t ' s win the game.

pieces

are stalemated,

but only

the white k i n g can

31 . . . Kd7 I n t e n d i n g to t r a n s f e r the rook to e6, which was Impossible immediately: 31...Re8 32 d7+! Kxd7 33 Rf6, and w i n s . In the event of passive defence, 31...Kt)7, White decides matters by a b r e a k t h r o u g h on the K-side: 32 Kf3 Re8 33 Kf4 h6 34 h4 followed by g3-g4, h4-hS and g4-g5. 32 Re2! The position is not so clear after 32 R2a5 Ke6 33 Rxb5, and now not 33...Bxd6, which loses to 34 Rb7, but f i r s t 33...Rd7 and then ...Bxd6.

32 . . . KcB Bad, of reaches dS.

course,

33 Re7 Rd7

is

32...Re8

33 Rxe8 KxeB,

when

the white

king

34 Ra2!

This wins Immediately, since Black cannot defend against the threats on the c - f i l e . 34.. .Bxd6 loses a piece after 35 Rxd7 Kxd7 36 Rd2. No better is 34...KdB 35 8b6+! (also possible is 35 Rae2 Rxe7 36 Rxe7, but not 36 dxe7+?> 35...Kc8 36 Rc2+. Instead of White's last move, 34 Rc6+ suggests itself, but after 34,..Kb7 35 Rxd7+ (35 Rc7+ Rxc7 36 dxc7 Bxc7 37 8d6 Rc8) 35... Kxc6 36 Rd8 a5 Black can put up a stubborn resistance. 34 . . . a5

35 Rc2 Resigns

No. 12 Amsterdam 1980 Larsen-Karpov Queen's Indian Defence In the competitive sense this game from the last round d i d not mean a great deal — only in the event of me losing and Timman w i n n i n g would the Dutch grandmaster catch me. But to win against Larson is always pleasant, the more so with Black and in 23 moves. The last round began earl ier ihan u s u a l , at eleven a . m . This was an unpleasant surprise f o r the p l a y e r s , who had become accustomed to going to bed late and r i s i n g around midday, Larsen suffered most of a l l , as he normally goes to sleep early In the morning, and gets up in time for l u n c h . . . . 1 Nf3 Nf6 2 g3 b6 3 Bg2 Bb7 4 0-0 e6 7 d4 Ne4 8 Qc2 Nxc3 9 Qxc3 c5 10 Rdi d6

5 c4 Be7

6 Nc3 0-0

158

15 of the World Champion's Best Games

A well-known variation of the Queen's Indian been highly popular in recent years. Nevertheless, Idea. 11 Be3 Nd7

Defence, which has Larsen finds a new

12 Raci

White f i r s t removes his rook for the long diagonal, d4-d5 to have the possibility of advancing his b-pawn, 12 . . . BcB

13 d5 exd5

U

so

as

after

cxd5

The p o s i t i o n is h i g h l y interesting, a n d f u l l of d i f f e r e n t possibilities. Thus Black has the i n t e r e s t i n g p l a n of a O-side o f f e n s i v e , b e g i n n i n g with 1 4 . . . a 5 , which also prevents b2-b4. Then 15 a4 c a n be met by 1 5 . . . B s 6 and 16. . . b 5 . Black also has a p r o m i s i n g game a f t e r U...Nf6 15 Mel Ng4 followed by . . . B f 6 . I d e c i d e d , however, to make a useful w a i t i n g move, o c c u p y i n g the semi-open e - f i l e w i t h my r o o k .

U

Be8 15

8f6

At f i r s t sight it appears that Black can seize the i n i t i a t i v e by 15...Nf6 16 Nei (16 Nh4 is weaker, in view o f 16...Ng4 17 Nf5 Bf6) 16...Ng4 17 Bfft Bf6, and the rook, which was developed on the previous move, comes into p l a y . But White has a different way of defending his central pawn: 16 Qb3!, simultaneously removing his queen from the c-file and ihe a1-h8 diagonal. Now 16...c4 Bxd5 f a i l s to 18 Rxd5 Nxd5 19 RxcB and 20 Qxd5, while after 17 Nd4 Black experiences d i f f i c u l t i e s . 16 Qd2 cxb4

17 Rxc8

Forced, since otherwise one of 18 Bxcl Rxe2, or 18 Rxcl Bxd5. 17 . . . 0xc8

17 Rxc4 16...cxb4

18 QxbA Qc2

the pawns

is

lost:

17 Qxb4 Rxci

Game No. 13

159

White has succeeded in achieving the positional gains he was aiming for. In turn, Black has a certain activity for his pieces, and, if he can hold out, a promising ending, thanks to his Q,-side pawn majority. 19 Rci 19 Qd2 was rather more s o l i d . Black could then have exchanged queens either immediately — 19...Qxd2 20 NxdZ! (weaker is 10 Rxd2 Rc8, with advantage to Black) 20. . .NcS, or after tlrst luring out the bishop: 19...Rc8 20 Bh3 Qxd2 21 Nxd2 RdB, which, I think, gives him the better chances. 19 . . . 0xe2

20 Qxd6?

It would seem that Larsen overlooked Black's reply, which places him in a critical position. 20 Qa4 was interesting, when Black appears to be all right in the variation 20.. .Rxe3 21 Qxd7 Re7, except that White has the murderous 22 Qxb7!. Black would have had to reply 20. ..RdB 21 Rc7 Ba6!, with rich possibilities. 20

Nf8!

Black seems to move his knight to a passive position, but in fact this is the best square for it — now there are no problems over the defence of his back rank, and although the knight takes no part in the play, his remaining pieces are extremely active. 21 Nd*? This loses Immediately, bul it was not easy to suggest anything worthwhi le for White. Under attack are his bishop at e3, and his pawns at a2 and d5. 21

Bxd4

22 Bxd4 Qd2

White's lack of co-ordination is his r u i n . both pieces, but it turns out that his k i n g ' s weakened.

He manages to defend position is hopelessly

23 Of 4 Re1 + White resigns, in view of 24 Rxel Qxe1+

25 Bfi Ba6.

No. 13 Tilburg 1980 Karpov-Spassky Sicilian Defence \n this game I chose the Keres Attack, one of my favourite lines In the Sicilian Defence. There was an interesting moment on the 13th move, when, rather to my surprise, Spassky took the h4 pawn with his bishop. Formerly, as far as I recall, the Ex-World Champion would not have even looked at such a pawn, but times change, and a passion for 'pawn-grabbing' can arise even in such a player as Spassky. As a result the black king was caught in the centre, after which events developed in rather interesting fashion.

160

15 of the World Champion's Best Games 1 e4 cS

2 Nf3 e6

3 d4 cxd4

4 Nxd4 Nf6

5 Nc3 d6

6 g4

Thus, a g a i n the Keres V a r i a t i o n , which w i l l a l r e a d y be f a m i l i a r from p r e v i o u s games of mine. White immediately begins a c t i v e p l a y on the K-side, before f u l l y m o b i l i z i n g h i s f o r c e s . I n r e p l y to these f l a n k operations Black prepares the customary c o u n t e r - b l o w in the c e n t r e .

6 . . . h6

7 h4

The f i r s t time 1 have played this. Normally I continue 7 g5 hxgS 8 Bxg5, but in some recent games with ihe Swedish grandmaster Andersson, a great expert on this v a r i a t i o n , I had been unable to achieve complete success, and so I decided to try another continuation. 7 . . . Nc6 The alternative immediately attacking the d8-h4 diagonal.

is 7...Be7, so as to have the possibility of the advanced h-pawn, and of gaining control of •

. : -

,

•

'

.

•:

.

r

8 Rgl d5!

The c r i t i c a l continuation, if Black is aiming for an active battle for the centre. Otherwise White, by advancing his g-pawn on his next move, would have driven back Ihe knight from f 6 . •• 9 Bb5 Bd7

10 exd5 Nxd5

11 NxdS exd5

12 Be3

•••

r-

Development f i r s t and foremost! After 12 g5 hxg5 13 hxg5 Rh4 Black brings his rook Into play with great effect, whi le the cautious 12 h5 is not even worth considering. White also had another tempting possibility: 12 Qe2+, which after 12...Q.e? 13 Be3 Nxd4 14 Bxd7+ Kxd7 15 Bxd4 leads to a certain advantage for White, but after 1 2 . . . Be7 13 Nf5 Bxf5 14 gxf5 KfB the position to me seems rather unclear, 12 . . . Be7

13 Qd2

White could also have attempted to exploit the open position of the black king by 13 Qe2, although he is then subjected to danger along the undefended a5-e1 diagonal, and also in some cases along the e-flle. Black would have had a choice between 13...Oa5+ 14 c3 N*d4 15 Bxd7t Kxd7 16 Bxd4 Rhe8 17 Kfl Bf6 IB Qf3 Re6, and 13...0-0 (threatening to capture on d4) 14 0-0-0 Qa5, which after 15 Bxc6 bxc6 16 Kb1 leads to an advantage for White (as does 13...a6 14 Bxc6 bxc6 15 0-0-0).

13 . . . Bxh4

Game No. 13

161

This is the episode which I recalled in my introduction lo the game. Black captures one of the pawns threatening his position, in the hope of castling K-side, which at present is not possible — 13...0-0 U Nf5!, and if 14...Bf6, then 15 Nxh6+ gxh6 16 95, with a decisive attack. 14 0-0-0 Bf6 Bad, of course,

Is U . . . N x d 4

15 Bxd7+ Qxd7

16 Bxd4.

15 Nf5 This move c a n be c r i t i c i z e d on c e r t a i n counts: White allows h i s opponent to exchange one of h i s most dangerous a t t a c k i n g pieces, simultaneously d o u b l i n g h i s pawns on the f - f i le, which r u l e s out the g4-g5 b r e a k t h r o u g h . But in r e t u r n White also g a i n s c e r t a i n a d v a n t a g e s : the d5 pawn is immediately a t t a c k e d , a n d the g - f i l e is opened, which is r a t h e r unpleasant f o r B l a c k , whose k i n g cannot escape to the f l a n k by c a s t l i n g . White would have l i k e d to have thrown f o r w a r d h i s f - p a w n by 15 f 4 , so as to then advance also h i s g - p a w n , but I was u n a b l e to f i n d a d e f i n i t e a d v a n t a g e a f t e r 1 5 . . .Qa5 (Black can also b a t t l e f o r e q u a l i t y by 1 5 . . . N x d 4 16 Bxd7+ Qxd7 17 Bxd4 0-0-0 18 Bxa7 Qa4 19 Qf2 d4 20 g5 Qxa7) 16 Qxa5 (16 Qe2 0 - 0 - 0 ! ) 16...Nxa5 17 Bxd7* Kxd?, and here I considered o n l y the f o r c i n g 18 g5 hxg5 19 fxgS Be5 20 Nf3, w h i c h is r e f u t e d by 2O...Bxb2+, a n d a fork at c 4 . Immediately a f t e r the game I discovered that i n t h i s v a r i a t i o n While would have g a i n e d excellent chances by 18 N f 5 ! . 15 . . . Bxf5 15. . . 0 - 0 a g a i n loses to 16 Nxh6+ gxh6 16 g x f 5 a6

17 g 5 ! .

17 Bxc6+

It is a l w a y s r e g r e t t a b l e to g i v e up such a bishop, but White has no time to r e t r e a t i t , since then the d-pawn would advance, a n d Black would safely manoeuvre h i s k i n g to f 8 , a n d from there, i f necessary, to g 8 . 17

bxc6

IB

Rbfl

18 Bc5

Perfectly reasonable. Black strives for counter-play, before White has time for the projected cZ-c4. He could also have played 18...0d7, but this did not appeal to Spassky, evidently because of 19 Qd3, when

162

15 of the World Champion's Best Games

the black queen is the simple 20 Bd4.

in

a passive

position,

and 19. . .Rb8 can be met by

19 b4 The quieter alternative was 19 b3, to answer 20 Bd4. But after the bold advance of the b-pawn, p r a c t i c a l l y forced to sacrifice the exchange. 19 . . . Rb5

20 Rge1 + Kd7

21 ch Rxc5

19.. .Rb5 Black is

with soon

22 bxc5 Bg5?

A mistake caused by a miscalculation. Black should have played either 22...Qc7, or 22...Db8. During the game I considered my position to be better, and in both cases was intending to play 23 f4, completely blocking out the bishoo arid forcing the opponent to advance his d pawn. But in analysis it was found that after 22. . .Qb8 23 f4 Black has the excellent reply 23...Kc8, when White cannot capture twice on d5, and 24...RdB is threatened. Therefore in reply to 22...QbB White should have continued 23 cxd5! Bg5 24 Re3, when Black has to seek salvation in the variation 24...Bxe3 25 fxe3 Qe5 26 dxc6++ Kxc6 27 Qd7+ KxcS 28 Qa7+ Kb5. .ti.

23 f4 Qf6

•>:•

24 cxdS!

This is the whole p o i n t ! The bishop cannot be taken — 24 fxgS? 0a1+ 25 Kc2 Qxa2+ 26 Kci Qa1+ 27 Kc2 Qa4+, when both 28 Kci RbS and 2B Kd3 0xc4+ 29 Ke3 hxg5 are bad for White. 24 . . . Qa1+ 25 KcZ Qxa2+ 26 Kd3 0xd2+ 27 BxdZ Bxf4 2B Ra2 cxd5 29 Rxa6 h5 30 Kd4 h4 31 Kxd5 Rb8 32 f 6 gxf6 33 Rxf6 Bg3 34 Rxf7+ KdB 35 Rf8-t Resigns •

No. 14 Malta Olympiad 19B0 Karpov-Hort Caro-Kann Defence

r."

•> • : • * » «

.il-V.. . "

e> «; . <* •.•r^(^

Victory in the last Olympiad d i d not come easily to our team. The Hungarians immediately went ahead, whereas we faltered at the s t a r t , and spent the whole tournament t r y i n g to catch u p . At f i r s t I loo was not in very good form. Before the present game I had played f i v e successive draws, and i t seemed that there would be no end to this series. Nevertheless, the game with Hort broke this sequence, and even proved to be the turnming p o i n t . After my win over the Czech grandmaster, things went well — in the last s i x rounds I scored S5 points. I «4c6

2 d4 dS

3 exdS

The Panov Attack occurs r a r e l y in my tournament practice, but when p r e p a r i n g for this game I remembered the game Balashov-Hort from the recent international tournament in Buenos Aires. Balsshov succeeded in g a i n i n g an advantage, and I decided to test Hort s readiness to repeat the v a r i a t i o n .

3 . . . cxd5

4 c4 e6

Game No. 14

163

In [he latter game the Czech grandmaster continued 4 . . .Nf6, and then developed his queen's bishop. 5 Nf3 Nf6

6 Nc3 Be7

6...Bb4 would have transposed into a variation of the NlmzoIndian Defence, whereas now we have a position typical of the Queen's Gambi t.

7 cxd5 Nxd5

8 Bd3 Nc6 9 0-0 0-0

10 Rei Nf6

11 a3 Qd6

One of the possible ways of developing. Temporarily delaying the development of his queen's bishop, Black aims for immediate pressure on the isolated pawn. The move appears rather provocative, in view of the vulnerable position of the black queen in the centre of the board, but knight moves fail to give White any immediate gain, and he has to be content with a more modest continuation. 12 Be3 Rd8

13 Qc2

As far as I am aware, a new p l a n . White takes control of the b1-h7 diagonal, and prevents the flank development of the c8 bishop, since 13...b6 f a i l s to 14 Ne4 Nxe4 15 Bxe4, with a double attack. Of course, the queen w i l l not stay long on the c-f i le, but time is gained for bringing the queen's rook into p l a y . 13

Bd7

14 Radl

White could have hindered the regrouping of the opponent's pieces by 14 Nb5, driving the queen to b8. But I reckoned that at c8 the rook would not have any particular prospects, and that the queen would voluntarily vacate d6, 14 . . . Rac8

15 Bg5!

Forcing Black to determine the position of his K-side pawns. 15...g6 looks rather unpleasant after 16 Qe2 followed by the transfer of the bishop to a2 and the d4-d5 breakthrough (the absence of the black bishop from the aS-M diagonal becomes apparent!). Hort decides to advance the other pawn, but this merely helps White's bishop to switch to a very strong position. 15

h6

16 Bh4

It is perfectly possible that, when considering this position beforehand, Black had intended playing 16.. .Nh5. After the exchange of bishops his position is perfectly satisfactory, but White has the

164

15 of the World Champion's Best Games

p o s s i b i l i t y of 17 dS! exd5 16 . . .

18 Bh7+ Kha

19 Rxc5,

winning.

0b8

The only move, since a f t e r ?6...Be8 17 Bg3 Qd7 extremely d i f f i c u l t for Black to u n t a n g l e his pieces. 17 Qe2 BeS

16 Bb5 it

is

18 Bbi

At last the l o n g - r a n g e queen/bishop b a t t e r y goes Into o p e r a t i o n ! It is a l l the more dangerous f o r Black h a v i n g advanced his h - p a w n , as this weakens the g6 square a n d creates the conditions f o r v a r i o u s tactical blows. 18 . . .

Nd5

19 Qd3 g6

20 Bg3 Bd6

21 Nxd5 exd5

22 Ne5!

As a r e s u l t of the exchange on d5, a symmetrical pawn s t r u c t u r e has a r i s e n . However, a l l B l a c k ' s pieces are in s t r a n g e p o s i t i o n s , a n d , most i m p o r t a n t , h i s queen is too f a r away from the K-side. 22 . . . Qc7

23 Qe3

Here I could have won a pawn by force a f t e r 23 Ng4, f o r example 23...Bxg3 24 hxg3 h5? 25 Nf6+ Kg7 26 Nxh5+. The pawn s i m i l a r l y cannot be defended by 24...K3-7, because of 25 Qd2 g5 26 Qd3. But Black can p l a y the more solid 2 4 . . . Q 6 6 25 Nxh6+ Kg? 26 Ng4 6d7, and a l t h o u g h White has an e x t r a pawn and a c l e a r a d v a n t a g e , there are numerous technical d i f f i c u l t i e s to overcome before he can w i n , 23 . . . Qb6 2 3 . . .h5 26 NxdB.

is

unsatisfactory,

24 Qxh6 Nxe5

25 dxe5 Bf8

in

view

of

26 Qg5 Qe6

24 Bh4

Be7

25 Nxc6 Bxh4

27 Ba2?!

On a c h i e v i n g a won p o s i t i o n , I began p l a y i n g r a t h e r superficially. I should have c a r r i e d out a simple f i n a l r e g r o u p i n g , b e g i n n i n g w i t h 27 Bf4 followed by the advance of the h - p a w n , instead of which 1 began e n g a g i n g in unnecessary s u b t l e t i e s , w i t h the r e s u l t that the bishop remained on g 3 . As a consequence, the game had to be won for the second t i m e . . . . 27 . . . Bb5

28 hi

Rd7

29 Qe3

Another in a series of moves l a c k i n g in specific c a l c u l a t i o n . White forces Black to advance h i s b - p a w n , b l o c k i n g the s i x t h r a n k for manoeuvres w i t h h i s queen. On the other h a n d , the queen was possibly not p a r t i c u l a r l y aiming to go to a6 or b6, end subsequently White loses the p o s s i b i l i t y of d e v e l o p i n g h i s queen or bishop on the g1-a7 d i a g o n a l w i t h g a i n of l i m e . 29 . . . b6

30 Gf4 Bc4

Whi le W h i t e ' s e x t r a pawns on the K-side s t a n d s t i l I, Black's passed pawn begins to become a c t i v e . My bishop is forced to r e t r e a t to a p a s s i v e p o s i t i o n , since a f t e r 31 Bxc4 Rxc4 the opponent's pieces acQuire c o n s i d e r a b l e freedom on the open c - f i l e and the white squares

Game No. 14

165

on the Q-side. 31 Bbi Bb3

32 Rd2

33 Bd3 Rdc7

34 Kh2 Rci

By forcing the exchange of a pair of rooks, Slack either seizes the f i r s ! rank, or activates his bishop by switching tt to the b1-h7 diagonal. Incidentally, Black is not averse lo the exchange of all four rooks. 35 Rde2 Rxel

36 Rxel Bc2

37 Ba6

Beginning from the twenty-seventh move, I played this game as though in a trance. It is perfectly obvious thai it is more advantageous for Black to place his rook ahead of his queen on the c-f i le, so why force him to do this?! 37 Be2 shoutd have been played immediately. 37 42 Qd4

..

Rc5

38 Be2 Bf5

39 Rdi

Qc6

40 Rd2 Rc2

41 Bf3 Be6

White has somehow succeeded In mounting some pressure on the d pawn, but he has wasted a great deal of time, and his bishop is s t i l l asleep at g3. Black now decides to exchange the rooks, which undoubtedly eases the position of the d5 pawn. White in turn no longer has to worry about the open c - f i l e , and can prepare the advance of his K-side pawns. The a l t e r n a t i v e was 4 2 . . . R c 5 . 42 . . . Rxd2

43 Qxd2 Qc4

44 Be2 Qa2!?

An interesting decision! I analysed only moves by the queen to e4, or along the c - f i l e . In the f i r s t instance t was Intending to continue 44...Qe4 45 f 3 , and now 45...Qb1 46 Bd3, or 45...Qa4 46 Bf2. In the second case White gains the p o s s i b i l i t y of advancing his h-pawn, since the opponent's queen has no i n t r u s i o n squares. The move chosen by Black forces White to take specific decisions, i n v o l v i n g the e v a l u a t i o n of the different possible types of endgame. 45 Bf4 Bc5

46 Be3

Not f e a r i n g the exchange 46...Bxe3 47 fxe3, when the position of the d5 pawn is f i x e d f o r good. White gains use of h i s d4 square, transfers h i s k i n g to f 2 , and begins the advance of h i s K-side pawns.

46

d4

47 BgS

White has only very s l i g h t 48 Be3 0xd2 49 Bxd2 Bxf2.

winning

chances

after

47

Bxd4 Qd5

166

15 of 47 . . . Qd5

the World Champion's Best Games

48 Bf6 Bf8

49 Bd3 Bg?

50 0f4! Kh7

There is an interesting finish after 5 0 . . .Bxf6 51 exffi Bf5 S2 Qb8+ Kh7 53 Bc4! Qxc4 54 Qf6, end mate is i n e v i t a b l e . This is not the only v a r i a t i o n , and the prosaic SZ Bxf5 gxf5 53 Qh6 Qd6+ 54 g3 Qf8 55 Qg5+- is also decisive, but i t i l l u s t r a t e s the t a c t i c a l p o s s i b i l i t i e s of the p o s i t i o n . 51 f3 b5

52 Kg3

I t is useful to b r i n g the k i n g into the square of the event of a bishop or pawn ending being reached. 52 . . . Qd7

53 Bxg7 Kxg7

54 Qf6+ Kg8

57 h6!

Qf8

d-pawn,

in

55 h4

It is the t u r n of this pawn to have 55..,Bc4 f a i l s to 56 Qd6! Qxd6 57 exd6 Kf8 60.g4 Kd7 61 h5, and w i n s . 55 . . . Qe8 56 h5 Bc4 60 Qf5+ Kh8 61 Qc8+ Resigns

the

the decisive w o r d . Now 58 Bxc4 b x c i 59 Kf2 Ke8

58 Bxg6 Qxh6

No. 15 Linares 19S1 Ljubojevlc-Karpov Caro-Kann Defence

59 Bxf7+ Kh7

..

A grandmaster's lot is never a bed only of roses. However many successes a p l a y e r may have g a i n e d , there comes a time when h i s form deserts him, and he achieves a result which is much worse than expected. One such unhappy event in my chess career was the 1980 i n t e r n a t i o n a l tournament in A r g e n t i n a , in which I did not even f i n i s h in the f i r s t three. The tournament in Spain was my next i n t e r n a t i o n a l appearance (not counting the Malta O l y m p i a d ] , and so I very much wanted to r e h a b i l i t a t e myself after Buenos Air-es. The young American grandmaster Christiansen provided very s t i f f competition, and in order to remain ahead I had to win a number of games. 1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 7 h4 h.6 8 h5 Bh7 9 Bd3 Bxd3 10 Qxd3 e6 11 Bf4

6 Nf3 Nd7

The aim of this check is to ' r e c a l l ' the bishop to d2. Some grandmasters are also happy to play this position as Black w i t h the white bishop at f4, an example being Larsen in games in 1980 against Tal, Timman and myself. 12 Bd2 0c7 13 0-0-0 Ngf6 17 Qc4 Bd6 18 Qa4 Kb8 19 Ne5

H

N«4 0-0-0

Oh 19 c4 Black can play 19...Qe7, knight at e4 and his queen at f 6 .

with

15 g3 Nc5

the

Idea

16 Nxc5 Bxc5

of

placing

his

19 . . . Nd5

occurs

Possibly more accurate i n the game.

was

t9...Nd7,

forcing

the

variation

which

Game No. J5

167

20 f4 20 Nc4 would also have been answered by 20 . . . Nb6

21 Qb3 Bxe5

2O...Nb6.

22 dxe5 Nd5

23 c47 An unnecessary weakening of the d3 and d4 squares. White could have retained a minimal opening advantage by 23 Qf3, e . g . 23...Ne7 24 g4, with the threat of 25 fS. In this case I was intending 2 3 . . . Rd7, when the sharp v a r i a t i o n 24 Qg4 Rhd8 25 Qxg7 Qb6 26 0g4 Ne3 27 Bxe3 Qxe3+ 26 Kb! Qxg3 29 Rxd7 Qxg4 30 Rxd8+ Kc7 31 Rhd1 gives White a s l i g h t l y favourable e n d i n g , but 28...Rxd1+ 29 Rxdl Rxd1+ 30 Qxdi Kc8! gives equal chances. 23 28 Qd3?1

He7

24 Be3 c5

25 Rxd6+ Rxd8

26 Rd1 R x d H

27 Qxdf

b6

A significant inaccuracy. 28 g4 Qb7 maintains the balance, although there is the interesting possibil ity of 2B.. .Oc6 29 Qd6+ Qxd6 30 exd6 Ng8 31 g5 Kb7 ( 3 1 . . . f 6 32 g6 Kb7 33 b i cxb4 34 f5 eS 35 c5, w i t h the betler chances for White) 32 b4! Kc6 (32...cxb4 is dangerous - 33 Bd4 f6 34 g6 Kc6 35 c5 bxc5 36 Bxc5}, and now an unclear position resulls from 33 bxc5 b*c5 34 Bxc5 g 6 ! .

28 . . . g6! 29 33 Bf2 Qd7 34 a* Kc7

30

a3

h5

31

32

b3

Qe4

Nf5

35 Kc2

If 35 Qa8, then 35...Qd3, nol f e a r i n g ghosts - after 36 Qa7+ Kc8 37 Qa7+ Kd7 38 0b7+ Ke8 39 Qb8+ Kf7 40 Qb7+ Ne7 the black k i n g successfully avoids perpetual check. 35 . . . OdS

36 Kci

Black

to

looks

g5

have

37 fxg5 Qxg5+ a

good

38 Kc2 Ne7

alternative

in

39 Clr.7 Kd7

39.. .Qg6+

40

168

15 of the World Champion's Best Games

Nxg6 41 Be3 Kd7 (41 ...Nxe5 manoeuvring the knight to d4.

«

Bf4)

42

Bg5

Nxe5

followed

by

40 Qe4 Qf5 41 Qd3+ Kc6 43 CUfS c*xf5 43 Be3 Ng6 44 e6 Kd6 45 Bg5 Kxe6 46 Kd3 f 4 ! 47 gxf4 h4 48 Ke3 h3 49 Kf3 Kf5 50 Kg3 Nxf4 51 Bd8 N«2+ 52 Kxh3 Nd4 53 Bxb6 53 Kg2 also f a l l s to save the game: after S3...Ke4 5A Kf1 Kd3 55 Kel Nxb3 56 Kdi Kxc4 57 Kc2 Nd4+ 58 Kbg Ne6 59 Bxb6 Kb4 60 Ba7 Kxa4 Black remains with two extra pawns. 63 . . . 58 Bf&+ Kc2

Nxb3 54 BdB Ke4 55 Kg4 Kd4 56 Kf4 Kxc4 59 Be5 c4 60 Ke3 c3 61 Bf6 Nc5 62 Ke2

57 Ke4 Kc3

The aitempt to reach the a5 pawn by 62 Kd4 is easily parried after 62...Nxa4 63 Kc4 Nb6+! (but not 63...Nb2+? 64 Kb5 a4 65 Kb4 with a draw) 64 Kb5 a4 65 Kb4 Nd5+. Also inadequate is 62 Bd8 Kd1 63 Bxa5 c2 64 Bd2 Nb3. 62

...

White

Kb3 resigns.

• --' • ' . • • • • •

" .

••-

•

••• •